润肤剂和保湿剂治疗湿疹
Referencias
References to studies included in this review
References to studies excluded from this review
References to ongoing studies
Additional references
References to other published versions of this review
Characteristics of studies
Characteristics of included studies [ordered by study ID]
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled study Setting Multicentre (10) in USA Date of study Unspecified. Duration of intervention: 50 days | |
| Participants | N = 218 (127 female, 91 male) Mean age = 44.5 years (18 to 84 years) Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 218 (MAS063DP group = 145, vehicle group = 73) Withdrawals/losses to follow‐up 32 overall (14.6%); MAS063DP group = 16/145 (11%), vehicle group = 16/73 (21.9%)
Baseline data Duration of the current episode of atopic dermatitis ranging from 5 days to 58 years. The most common atopic dermatitis pattern was constant or frequent (80% of participants) and with no seasonal course (66%) Mean EASI: MAS063DP group 5.76 (SD 4.68), vehicle group 6.03 (SD 6.62) Mean itch (VAS): MAS063DP group 7.51 (SD 1.6), vehicle group 7.05 (SD 1.62) | |
| Interventions | Intervention
Comparator
Topical corticosteroids, immunomodulators, antihistamines, non‐steroidal anti‐inflammatory drugs and phototherapy were excluded during a washout period prior to treatment initiation and during the study. | |
| Outcomes | Assessments (5): at baseline, days 8, 22, 36, and 50 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared. However, the product under research is produced by Sinclair Pharma Ltd, Godalming, UK | |
| Declaration of interest | Almost all investigators were investigators or consultants (or both) of Sinclair Pharma Ltd, as well as of Graceway Pharmaceuticals, LCC and received compensations | |
| Notes | MAS063DP (Atopiclair, Sinclair Pharma Ltd, Godalming, UK) contains glycyrrhetinic acid, which is believed to contribute to the cream's antipruritic and anti‐inflammatory properties. It inhibits the enzyme 11β‐hydroxysteroid dehydrogenase (11β‐HSD), responsible for the metabolism of cortisol into cortisone. Glycyrrhetinic acid, has a chemical structure related to cortisone and has been shown to potentiate the action of hydrocortisone on skin. Further ingredients include: aqua, ethylhexyl palmitate, Vitellaria paradoxa (formerly called Butyrospermum parkii), pentylene glycol, arachidyl alcohol, behenyl alcohol, arachidyl glucoside, glyceryl stearate, PEG‐100 stearate, butylene glycol, glycyrrhetinic acid, capryloyl glycine, bisabolol, tocopher acetate, carbomer, ethylhexylglycerin, piroctone olamine, sodium hydroxide, allantoin, DMDM hydantoin, Vitis vinifera, sodium hyaluronate, disodium EDTA, ascorbyl tetraisopalmitate, propyl gallate, telmesteine We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 328) and (page 237 of 2006 publication): "Patients were randomized (2:1)" "Patients were assigned a study number according to their entry into the study and following a computer generated randomization code". Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: central allocation via pharmaceutical company, sequentially numbered tubes of identical appearance. Comment: central allocation, de‐identified tubes. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 327): "double‐blind..." Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "tubes of drug or placebo/vehicle looking identical" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 327): "double‐blind..." Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. After email communication: "tubes of drug or placebo/vehicle looking identical" Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | High risk | 32 (14.6%) overall, MAS063DP (Atopiclair) group = 16/145 (11%), vehicle group = 16/73 (21.9%). Per‐protocol analysis. Comment: the total number of dropouts was unbalanced between the groups, which, combined with a per‐protocol analysis represents a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appears to be free from other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Department of Medical Sciences, University Hospital, Uppsala, and ACO Hud AB, Stockholm, Sweden Date of study February to April, not clear which year. Duration of intervention: 30 days | |
| Participants | N = 50 (38 female, 12 male) Age = 18 to 55 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 50 (5% urea (Canoderm) group = 25, 4% urea (Fenuril) group = 25) Withdrawals/losses to follow‐up 2/50 (4%)
Baseline data | |
| Interventions | Intervention
Comparator
Allowed to continue use of topical corticosteroids | |
| Outcomes | Assessments (3): baseline, days 15 and 31 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, however, 1 of the products under research is produced by ACO Hud AB, Sweden | |
| Declaration of interest | None declared. Dr Lodén was an employee of ACO Hud AB, the manufacturer of the principal intervention | |
| Notes | Fenuril (Pharmacia AB, Sweden) contained 4% urea and 4% NaCl as water‐binding substances in an oil‐in‐water emulsion, pH about 5. Other ingredients were liquid paraffin, PEG‐ 5‐glyceryl stearate, cetyl alcohol, stearyl alcohol, stearic acid, trometamol, methyl para‐hydroxybenzoate, propyl para‐hydroxybenzoate, hydrochloric acid and water As the study was 17 years old we did not contact the investigators for data. The data all needed to be estimated from box‐and whisker plots, which made them difficult to use (Table 4). The adverse events were addressed in a separate publication (Lodén 1999 added in 'References to studies' under the primary publication Andersson 1999) | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 165): "randomized double‐blind study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 165): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 165): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. |
| Incomplete outcome data (attrition bias) | Low risk | 2/50 (4%) withdrew for "reasons not related to treatment" in Fenuril group. Per‐protocol analysis. Comment: low number of losses to follow‐up, and despite use of per‐protocol analysis, considered to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Department of Dermatology, University of Lübeck, Lübeck, Germany Date of study Unspecified. Duration of intervention: 1 week, followed by 3 weeks' treatment with 'moisturiser only' | |
| Participants | N = 20 (16 female, 4 male) Age = 12 to 65 years, median age = 26.2 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 20, to either forearm (within‐participant) Withdrawals/losses to follow‐up 2/20 (10%)
Baseline data Mean local SCORAD: O/W formulation group 7.0, hydrocortisone group 6.5 Mean itch severity (VAS): O/W formulation group 3.5, hydrocortisone group 3.75 Mean TEWL (g/m²/h): O/W formulation group 16.47, hydrocortisone group 16.28 | |
| Interventions | Intervention
Comparator
After 1 week both arms were treated with the O/W (oil in water) formulation and no other treatment | |
| Outcomes | Assessments (3): baseline, days 7 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 9): "The study was funded by Beiersdorf AG" | |
| Declaration of interest | Quote (page 9): "I, Angelova‐Fischer has been investigator for and received honoraria as a speaker from Beiersdorf AG" [sic] | |
| Notes | Licochalcone A is an extract from Glycyrrhiza inflata that has anti‐inflammatory properties. After 1 week both arms were treated with a moisturiser, therefore we have only included data from the first week We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 10): "were randomized to receive" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "random number table" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: this remains unclear. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 10): "investigator‐blinded" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each forearm, to permit a clear judgement, and participants were not blinded. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 10): "investigator‐blinded". Outcomes were investigator‐assessed as well as participant‐assessed. Comment: there was uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/20 (10%), 2 excluded in the second and fourth study weeks because of worsening of the skin condition outside the test area that required topical treatment. Comment: as we only included data from the first week we judged this as being at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free from other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled study Setting European Institute of Dermatology, Milan, Italy Date of study Unspecified. Duration of intervention: 22 days, with follow‐up of 2 weeks | |
| Participants | N = 30 (14 female, 16 male) Age = 13 to 43 years, median age = 22.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 30 (MAS063D group = 15, vehicle group = 15) Withdrawals/losses to follow‐up All participants finished the study protocol Baseline data Eczema severity and location of the lesions, disease duration and current episode were comparable | |
| Interventions | Intervention
Comparator
Participants receiving systemic medication (antihistamines, corticosteroids, NSAIDs, or other topical and systemic investigational drugs) were maintained on their medications at a constant dose throughout the study. Participants receiving topical medications (e.g. topical antihistamines, corticosteroids, NSAIDs) were taken off the medication in a washout period of 7 days, so that no participants were using these medications 7 days before the study or during it. Washout period for phototherapy and tranquillisers was 4 weeks and 5 days respectively. | |
| Outcomes | Assessments (5): baseline, days 8, 15, 22 and 36 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 35): "The study was supported by a grant from Sinclair Pharmaceuticals, Godalming, Surrey, UK" | |
| Declaration of interest | None declared | |
| Notes | Participants receiving topical medications (e.g. topical antihistamines, corticosteroids, NSAIDs) were taken off the medication in a washout period. See in Notes section of Characteristics of included studies of Abramovits 2008 for details on Atopiclair | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 32): "Randomisation was carried out in blocks of six patients using a random number generator in Microsoft Excel". Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 32): "Patients were randomised to receive MAS063D or vehicle‐only control, according to their order of entry into the study. MAS063D and control were presented in identical, blindly, pre‐labelled containers. Each container was labelled with patient's study number and patients, observers and all trial personnel were blinded to study code". Comment: central allocation, de‐identified drug containers. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 32): "The jars containing MAS063D or control were presented blindly, labelled with identical directions for use..." Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 32): "The jars containing MAS063D or control were presented blindly, labelled with identical directions for use..." Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, controlled, within‐participant study Setting Multicentre (3), baby clinics in Oslo, Norway Date of study 2008 to 2011; duration of intervention: 4 weeks | |
| Participants | N = 9 (6 female, 3 male) Mean age = 18.5 months Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 9, however contralateral eczema spots were randomised (18 spots left/right) (within‐participant) Withdrawals/losses to follow‐up 3/9 participants (33.3%)
Baseline data Mean SCORAD: 35 (range 22‐45) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (5): baseline, weeks 1, 2, 3 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | Quote (page 6): "The authors declare that they have no competing interests" | |
| Notes | None of our outcomes were addressed (Table 4). Apobase creme contains: aqua, paraffinum liquidum, petrolatum, cetearyl alcohol, ceteareth‐20, ceteareth‐12, sodium gluconate, caprylyl glycol, phenoxyethanol, and has a total lipid content of 30% We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (pages 2‐3): "Another physician, who did not see the child, was responsible for the randomization". Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: " Physician 1 prepared 12 letters, 6 right and 6 left, with treatment information. These were then folded, so the information was hidden. Thereafter one and one was randomly drawn and included into envelopes numbered 1‐12. Finally these envelopes were sealed. This procedure was performed physically outside the clinical department where physician 2 works. Physician 1 was the only one with access to the randomization code." Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 3): "The child was given a randomization number and the mothers were then informed on which side to apply the fresh expressed human milk and emollient, and on which side to apply emollient alone". Comment: there was insufficient information to permit a clear judgement. After email communication: "Physician 2 received one sealed, numbered envelope, for each new included patient, from physician 1". Comment: allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 2): "physician‐blinded" Comment: The report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "Physician 2 delivered the sealed envelope with the correct patient number, to the mother. The mother opened the sealed envelope outside the sight of physician 2. If she had any questions she could talk to physician 1 by telephone, without physician 2 listening. This happened once". Comment: the report provided sufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received. However, participants were not blinded and so we judged this study as being at an unclear risk of bias. |
| Blinding of outcome assessment (detection bias) | Low risk | Outcomes were investigator‐assessed. Blinding of the outcomes assessors was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | High risk | 3/9 (33.3%). Per‐protocol analysis. Comment: high number of losses to follow‐up, combined with per‐protocol analysis meant we considered this study to be at a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (NCT02381028), and the prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appears to be free from other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐controlled study Setting Multicentre (39), 6 countries in Europe Date of study Recruitment January 1998 to July 1999. Duration of intervention: 4 weeks for stabilisation phase, and 16 weeks maintenance phase. We only included the data from the maintenance phase. | |
| Participants | N = 376 (205 female, 171 male) Mean age = 28.8 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 376 (fluticasone cream once daily group = 95, fluticasone cream twice daily group = 91, fluticasone ointment once daily group = 100, fluticasone ointment twice daily group = 90) Withdrawals/losses to follow‐up Stabilisation phase: 83/376 overall (22%); fluticasone cream once daily group = 19/95, fluticasone cream twice daily group = 15/91, fluticasone ointment once daily group = 23/100, fluticasone ointment twice daily group = 26/90, reasons reported (2/83, however, entered into the maintenance phase) Maintenance phase:
Baseline data Mean extent of atopic dermatitis (%): fluticasone cream once daily group 28.8% (SD 19), fluticasone cream twice daily group 17.7% (SD 16.2), fluticasone ointment once daily group 17.5% (SD 14.6), fluticasone ointment twice daily group 18.4% (SD 16.1) Median TIS score: fluticasone cream once daily group 5.0 (range 4‐6), fluticasone cream twice daily group 5.0 (range 4‐9), fluticasone ointment once daily group 5.0 (range 4‐7), fluticasone ointment twice daily group 5.0 (range 4‐7) | |
| Interventions | Initially the flare was stabilised with fluticasone propionate cream 0.05% or fluticasone propionate ointment 0.005%, once or twice daily, for 4 weeks. After that the maintenance phase continued as follows: Intervention
Comparator 1
Intervention 2
Comparator 2
| |
| Outcomes | Assessments (7): baseline, weeks 2 and 4 (end of stabilisation phase), weeks 2, 6, 10 and 16 (end maintenance phase) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 6): "Funding: Glaxo Wellcome (now GlaxoSmithKline) R & D, United Kingdom" | |
| Declaration of interest | Quote (page 6): "Competing interests: CP is employed full time by GlaxoSmith‐Kline" | |
| Notes | The moisturiser was a cetomacrogol‐based cream We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 2): "Investigators at each centre allocated patients to treatment groups in equal numbers according to a computer generated randomisation code. The block size for the study was eight, and each recruiting centre received 16 treatment allocation numbers". Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 2): "Investigators at each centre allocated patients to treatment groups in equal numbers according to a computer generated randomisation code. The block size for the study was eight, and each recruiting centre received 16 treatment allocation numbers". Comment: form of central allocation, probably done. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 1 and 2): "double‐blind" and "Patients who achieved remission (see assessments) then entered a maintenance phase and, using the same formulation as in the stabilisation phase, applied fluticasone propionate or its placebo base on two successive evenings per week for up to 16 weeks". Comment: the report did not provide sufficient detail about the specific measures, used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "blinding was achieved using matching tubes containing either active medication or vehicle alone". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 1 and 2): "double‐blind" and "Patients who achieved remission (see assessments) then entered a maintenance phase and, using the same formulation as in the stabilisation phase, applied fluticasone propionate or its placebo base on two successive evenings per week for up to 16 weeks". Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (participants/healthcare providers) during the study, insufficient information to permit a clear judgement. After email communication: "blinding was achieved using matching tubes containing either active medication or vehicle alone". Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | Loss to follow‐up in stabilisation phase 83/376 (22%), reasons reported, balanced between groups, intention‐to‐treat and per‐protocol analysis. Discontinued in maintenance phase 27/295 (9.2%); fluticasone cream twice weekly plus moisturiser = 5/70, vehicle cream twice weekly plus moisturiser = 7/84, fluticasone ointment twice weekly plus moisturiser = 6/68, vehicle ointment twice weekly plus moisturiser 9/73); Reasons: lost to follow‐up (11), withdrew consent (3), protocol violation (7), adverse events (4), other (2). Intention‐to treat analysis. "We conducted all analyses on an intention to treat basis (all subjects were included in the analysis if they were randomised and applied the study medication at least once)." Without further information regarding the maintenance phase (the number initially randomised, does not match the number in the maintenance phase). |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre, Montreal Quebec, Canada Date of study March 2007 to October 2007. Duration of intervention: 42 days | |
| Participants | N = 100 (71 female, 29 male) Mean age = 36.1 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 100 (urea 5% moisturiser group = 50, urea 10% lotion group = 50) Withdrawals/losses to follow‐up 12/100 overall (12%); 6 in each group discontinued treatment
Baseline data Mean SCORAD: 5% moisturiser group 20.35 (SD 5.81), 10% lotion group 21.4 (SD 4.96) | |
| Interventions | Intervention
Comparator
Washout period: topical immuno modulators and phototherapy (2 weeks), PUVA and systemic treatment (4 weeks) Use of topical corticosteroids permissible if frequency of application constant for 2 weeks prior to start of study and throughout | |
| Outcomes | Assessments (3): day 3; day 21 phone call to verify compliance and tolerability; day 42 follow‐up visit, efficacy and safety Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared, but 2 investigators are employed by La Roche‐Posay Laboratoire Pharmaceutique, Asnières Cedex, France | |
| Notes | Iso‐Urea, La Roche‐Posay Laboratoire Pharmaceutique, Asnières Cedex, France. Ingredients: aqua ⁄/water, Vitellaria paradoxa (formerly called Butyrospermum parkii; shea butter), glycerol, cyclohexasiloxane, urea, paraffinum liquidum ⁄ mineral oil, sodium lactate, cetearyl alcohol, PEG‐100 stearate, glyceryl stearate, propylene glycol, glycine, tocopherol, stearic acid, myristic acid, palmitic acid, bisabolol, triethanolamine, dimethicone, dimethiconol, disodium EDTA, hydroxyethyl piperazine ethane sulfonic acid, xanthan gum, acrylates ⁄ C10‐30 alkyl acrylate crosspolymer, citric acid, chlorhexidine digluconate, phenoxyethanol, methylparaben, propylparaben, fragrance 10% urea lotion contains: aqua, urea, sodium lactate, paraffinum liquidum, octyldodecanol, caprylic ⁄ capric triglyceride, isopropyl palmitate, glycerol, PEG‐7 hydrogenated castor oil, benzyl alcohol, methoxy PEG‐22 ⁄ docecyl glycol copolymer, PEG‐45 ⁄ dodecyl glycol copolymer, dimethicone, magnesium sulfate, lactic acid, ozokerite, PEG‐2 hydrogenated castor oil, sorbitan isostearate, hydrogenated castor oil We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 17): "Subjects were randomized (1:1)" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The allocation sequence was generated by an independent statistician" "Individuals were randomised by a computer‐generated list, which was maintained centrally " Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Individuals were randomised by a computer‐generated list, which was maintained centrally. Treatments were assigned to consecutive patients in a sequential order". Comment: form of central allocation, probably done. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 17 ): "double‐blind..." Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "Two copies of the randomisation list were prepared: one was used by for the labeling of the tubes performed by the sponsor (tubes were designated for each patient ‐ All tubes were equal in weight, and similar in appearance) and the other one was kept by the sponsor until the end of the study". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 17 ): "double‐blind..." Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. After email communication: blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | 12/100 (12%); 6 in each group discontinued treatment. Reasons reported. Per‐protocol analysis (88/100) Missing data for 4 participants on: SCORAD, cosmetic acceptability and quality of life questionnaire. Comment: balanced and moderate number of dropouts at follow‐up, combined with the per‐protocol analysis, poses an unclear risk of bias for this domain. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled study Setting Multicentre (7), Dermatology and Pediatric departments, children's hospitals in USA Date of study November 2005 to May 2006. Duration of intervention: 43 days | |
| Participants | N = 142 (74 female, 68 male) Mean age = 5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 142 (MAS063DP group = 72, vehicle group = 70) Withdrawals/losses to follow‐up 36/142 overall (25.4%); MAS063DP group = 19/72 (26.3%), vehicle group = 17/70 (24.2%)
Baseline data IGA of atopic dermatitis mild: MAS063DP group (37/72), vehicle group (37/70) IGA of atopic dermatitis moderate: MAS063DP group (35/72), vehicle group (33/70) Itch (VAS score) mean: MAS063DP group 6.2 (SD 11.7), vehicle group 6.7 (SD 1.7) | |
| Interventions | Intervention
Comparator
If clinically indicated, a low potency rescue topical steroid was prescribed for participants by the study investigator | |
| Outcomes | Assessments (6): baseline, days 3, 8, 22, 29, 43 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 854): "This work was sponsored by Sinclair Pharmaceuticals Ltd (Surrey, UK)" | |
| Declaration of interest | None declared | |
| Notes | Quote (page 855): "Patients and caregivers agreed to refrain from using other topical and systemic medications (including phototherapy) during the wash‐out and study periods. A washout period of 7 or 14 days was used for patients on topical and systemic medications, respectively, including topical and systemic corticosteroids, topical calcineurin inhibitors, antihistamines,and phototherapy". See Notes of Characteristics of included studies of Abramovits 2008 for details of Atopiclair We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 855): "Eligible subjects were randomized to receive either MAS063DP or vehicle in a 1:1 ratio according to a computer generated code" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 855): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 17 ): "double‐blind..." Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Unclear risk | 36/142 (25.4%); MAS063DP group = 19/72 (26.3%), vehicle group 17/70 (24.2%), reasons reported. Both per‐protocol and intention‐to‐treat analysis. Comment: the percentage of drop‐outs exceeded 20%, however, the investigators provided both a per‐protocol and an intention‐to‐treat analysis (last observation carried forward). We judged this as being at an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, placebo‐controlled, within‐participant study Setting BioSkin Institut für dermatologische Forschung und Enwicklung, Hamburg, Germany Date of study Unspecified. Duration of intervention: 4 weeks, followed by a 5‐month open phase for dermal tolerance | |
| Participants | N = 41 (31 female, 10 male) Mean age = 31 years Inclusion criteria of the trial
Exclusion criteria of the trial
RandomisedN = 41, to either forearm (within‐participant) Baseline dataCorneometry mean units: urea 10% lotion side 40.4 (SD 6.6), vehicle side 40.0 (SD 8.0) | |
| Interventions | Intervention
Comparator
No other treatments and moisturising shower and bath products were allowed. | |
| Outcomes | Assessments (6): baseline, days 8, 15, 22, 29 and 32
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, however, 1 of the products under research is from Beiersdorf AG, Hamburg, Germany | |
| Declaration of interest | None declared, however 2 of the investigators are employees of Beiersdorf AG, Hamburg, Germany, the manufacturer of the principal intervention | |
| Notes | As the study was 19 years old we did not contact the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 35): “randomisiert zugeordnet” (translation: assigned randomly) Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 34): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each forearm, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 34): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. |
| Incomplete outcome data (attrition bias) | Low risk | 3/41 (7.3%) withdrew for worsening of the skin condition (1) and reasons not related to treatment (2). Per‐protocol analysis Comment: low number of losses to follow‐up, and although per‐protocol analysis, considered as being at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled study Setting Multicentre in France (4 centres), Estonia (5 centres), Lithuania (5 centres), Poland (9 centres), and Romania (7) Date of study November 2011 to May 2012. Duration of the intervention 28 days with a follow‐up (open‐label) period of 46 days | |
| Participants | N = 251 (122 female, 127 male and 2 gender unknown) Mean age = 4 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 251 (V0034CR01B moisturiser group = 125, vehicle group= 126) Withdrawals/losses to follow‐up 10/251 overall (4%); V0034CR01B group = 5/125, vehicle group = 5/126
Baseline data Mean objective SCORAD: V0034CR01B group 11.7 (SD 2.1), vehicle group 11.2 (SD 2.1) Mean itch (VAS): V0034CR01B group 2.8 (SD 2.2), vehicle group 2.5 (SD 1.9) Mean corneometry units: V0034CR01B group 29.1 (SD 10.3), vehicle group 29.4 (SD 11.3) | |
| Interventions | Intervention
Comparator
In case of atopic dermatitis flares, the test treatment had to be applied in the morning and a moderately potent corticosteroid 0.1% desonide in the evening until complete resolution of inflammatory skin lesions | |
| Outcomes | Assessments (5): days 1, 7, 14, 21 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1456): "This study was funded by Pierre Fabre" | |
| Declaration of interest | Quote (page 1456): "F.B. received consulting fees and fees for participation in review activities from Pierre Fabre. M.S.A. and A.D. are employees of Pierre Fabre. G.T. received consulting fees from Pierre Fabre. H.R., A.K. and M.B. declare no conflicts of interest related to this article" | |
| Notes | We only included the double‐blind period of 28 days. Ingredients of V0034CR01B moisturiser: glycerol 15%, liquid and soft paraffin 10%, glycerol monostearate, stearic acid, polydimethylcyclosiloxane, silicone oil, macrogol 600, trolamine, propyl parahydroxybenzoate and purified water We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1457): "During the double‐blind period, patients were randomized 1:1 using a computer‐generated and ‐validated list to be treated for 28 days..." Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 1457): "The randomization list was generated by the Clinical Pharmacology Department using proprietary software and with parallel groups and a block size of four, and the list was validated by the Biometry Department. Treatments were provided in identical, sequentially numbered containers and were assigned to patients by investigators on the basis of the sequence number". Comment: central allocation, de‐identified drug containers. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 1457): "Treatments were provided in identical, sequentially numbered containers and were assigned to patients by investigators on the basis of the sequence number. Thus, patients, investigators and pharmacists were blinded to which treatment (emollient or vehicle) was supplied". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Outcomes were investigator‐assessed as well as participant‐assessed. Comment: blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 10/251 (4%): V0034CR01B group = 5, vehicle group = 5. Intention‐to‐treat analysis. Comment: the low total number of drop‐outs, balanced between the groups combined with a intention‐to‐treat analysis, meant we considered this study to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (EudraCT number 2011‐003295‐37), and the prespecified outcomes and those mentioned in the methods section did not all appear to have been reported. VAS data on pruritus are missing and very limited data on SCORAD and objective SCORAD are available. Comment: initially we judged this as being at a high risk of bias. After email communication: we received all necessary data. Comment: subsequently, we judged this as being at a low risk of bias. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, placebo‐controlled, within‐participant study Setting Department of Dermatology, Friedrich Schiller University, Jena, Germany Date of study November 2004 to March 2006. Duration of intervention: 4 weeks | |
| Participants | N = 24 (8 female, 16 male) Mean age = 23 years (range 15‐49 years) Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 24, to either forearm (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data The eczema on both forearms was of comparable severity Mean adaptation of SCORAD: glycerol 20% side 2.9 (SD 1.3), vehicle side 3.2 (SD 1.3) Mean TEWL (g/m²/h): glycerol 20% side 18.4 (SD 8.2), vehicle side 26.6 (SD 17.3) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (7): baseline, weeks 1, 2, 3, 4, 5 and 6 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 44): "This study was supported by Spirig AG, Egerkingen, Switzerland" | |
| Declaration of interest | None declared, but 1 of the investigators was an employee of Spirig AG | |
| Notes | Assessment were performed 12 hours after last application and after a washout period of 2 weeks. It remains unclear if this is a product being developed by this company. The glycerol cream contained 20% glycerol (200 mg/g) and the following ingredients: aqua, cetearyl alcohol, isopropyl myristate, paraffinum liquidum, PEG‐40 hydrogenated castor oil, glyceryl behenate, glyceryl dibehenate, tribehenin, citric acid, sodium citrate, methylparaben, propylparaben. The composition of the placebo was identical but without glycerol We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 40): "According to a randomization list, the right or left forearm was selected..." Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The randomization list was generated from the sponsor of the study not being part of the study during the trial". Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: initially there was insufficient information to permit a clear judgement. After email communication: "we received just A and B labelled tubes and allocated them according to the randomization list to each of the arms". Comment: central allocation, de‐identified drug containers. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 39): "...double‐blind..." .."The composition of the placebo was identical without glycerol" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "The study was fully double‐blinded until the statistical analysis was finalized. Both creams had the identical consistence and filled in identical tubes by the sponsor of the study". Comment: the email communication provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 39): "...double‐blind..." Outcomes were investigator‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (participants/healthcare providers) during the study. Insufficient information to permit a clear judgement. After email communication: "The study was fully double‐blinded until the statistical analysis was finalized. Both creams had the identical consistence and filled in identical tubes by the sponsor of the study". Blinding of the outcomes assessors, key personnel, was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | Quote (page 41): trial registered on European Clinical Trials Database (EudraCT (2004‐004443‐22)), protocol not accessible. The prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Unclear risk | Baseline differences in TEWL and erythema in favour of active treatment group. |
| Study characteristics | ||
| Methods | Randomised, controlled, within‐participant study Setting The University of Sheffield, South Yorkshire, UK Date of study Unspecified. Duration of intervention: 2 weeks | |
| Participants | N = 38 (29 female, 9 male) Mean age = 32 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 50, to either forearm (within‐participant) Withdrawals/losses to follow‐up 7/38, reasons unrelated to the trial Baseline data Mean TEWL (g/m²/h): Aqueous cream BP 12.868 (SD 3.253), Oilatum Junior Bath additive 12.819 (SD 3.446) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and week 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 45): "This study was funded by a research grant from Stiefel, a GSK company" | |
| Declaration of interest | None declared | |
| Notes | Poster abstract. Little information was provided. We received responses to our request for study details (Table 2; Table 4). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 44): "randomized comparison" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "with a randomization list" Comment: probably done. |
| Allocation concealment (selection bias) | High risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: initially, there was insufficient information to permit a clear judgement. After email communication: "no concealment" Comment: subsequently, we judged this as at a high risk of bias. |
| Blinding of participants and personnel (performance bias) | High risk | Nothing reported regarding blinding. Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each forearm, to permit a clear judgement. After email communication: "no blinding" Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Outcomes were investigator‐assessed. Insufficient information to permit a clear judgement. After email communication: "no blinding" Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Unclear risk | Nothing reported, little information provided. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled study Setting Multicentre in Spain, 12 dermatologists in private practices Date of study August to October 2007. Duration of intervention: 3 weeks | |
| Participants | N = 80 (44 female, 36 male) Mean age = 2.3 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 80 (STELATOPIA moisturiser = 40, hydrocortisone butyric‐propionate cream group = 40) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean SCORAD: STELATOPIA moisturiser 36.86 (SD 12.01), hydrocortisone butyric‐propionate cream group 37.19 (SD 15.28) | |
| Interventions | Intervention
Comparator
All children also received a body hygiene product in its marketed form (STELATOPIA milky bath oil, Mustela; Laboratoires Expanscience), to be used at least once every other day | |
| Outcomes | Assessments (3): baseline, weeks 1 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 364): "This study was supported by Laboratoires Expanscience, R&D Center, France. Dr C. de Belilovsky was supported and paid by Laboratoires Expanscience for managing the project" | |
| Declaration of interest | Quote (page 364): "The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper". Several authors were employees of Laboratoires Expanscience, the manufacturer of the primary intervention. | |
| Notes | STELATOPIA moisturiser, Mustela; Laboratoires Expanscience, France contains: 2% sunflower oleodistillate, essential fatty acids, bio‐ceramides, β‐sitosterol and a complex of emulsifying sugars We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (360): "were randomly assigned by the dermatologists" and "following a chronologic order of inclusion on a randomized attribution list" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: initially, there was insufficient information to permit a clear judgement. After email communication: allocation was controlled by the Clinical Reseach Organisation. Comment: form of central allocation. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (360): "single‐blind" and "observation blinded" Comment: the report alone did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "Anonymous sealed identical packaging provided by the Clinical Reseach Organisation" Comment: the report provided sufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received, however, participants were not blinded. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (360): "single‐blind" and "observation blinded" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: blinding of the outcomes assessors, was ensured, but participants and parents were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled study Setting Dermatology Clinic, High Point, North Carolina, USA Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 60 (gender not reported) Age range = 5 to 80 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 60 (fluocinonide cream 0.05% + cleansing bar group = 20, fluocinonide cream 0.05% + MVE liquid cleanser group = 20, fluocinonide cream 0.05% + MVE liquid cleanser + MVE moisturising cream = 20) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean Global Disease Severity: fluocinonide cream 0.05% + cleansing bar group = 3.0, fluocinonide cream 0.05% + MVE liquid cleanser group = 3.1, fluocinonide cream 0.05% + MVE liquid cleanser + MVE moisturising cream = 3.0 | |
| Interventions | Intervention
Comparator 1
Comparator 2
Participants underwent a 4‐week oral and topical eczema treatment washout period before study entry | |
| Outcomes | Assessments (4): baseline, weeks 1, 2, and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 87): "This study was supported by an unrestricted educational grant from Coria Laboratories, Ltd" | |
| Declaration of interest | Quote (page 87): "The author reports no conflict of interest" | |
| Notes | MVE = multilamellar vesicular emulsion. Quote (page 87): "In MVE, there are concentric layers of oil‐in‐water emulsions, which are referred to as vesicles. The vesicles are unfolded when placed in contact with the skin surface to release ceramides; cholesterol; free fatty acids; phytosphingosine; and other moisturising ingredients, such as dimethicone, glycerol, and hyaluronic acid, onto the skin surface" We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 88): "Subjects were randomized to 1 of 3 balanced treatment groups of 20 subjects each" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Participants were not blinded, unclear if physicians were blinded. Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 89): "Blinded investigator assessments" Outcomes were (mainly) investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of healthcare providers during the study, and participants and parents were not blinded. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigated‐blinded, 'other moisturiser'‐controlled, within‐participant study Setting Dermatology Clinic, High Point, North Carolina, USA Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 60 (gender not reported) Age > 18 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 60 to either forearm or leg (within‐participant) Withdrawals/losses to follow‐up 1/60 (1.7%)
Baseline data The groups were properly balanced at baseline between arm and leg target sites and eczema severity, as no statistically significant differences between groups were present at baseline. Mean investigator‐assessed disease severity (6‐point Likert scale): Albolene side 2.6, MimyX side 2.6 | |
| Interventions | Intervention
Comparator
In people with moderate eczema, treatment was combined with 0.1% triamcinolone acetonide | |
| Outcomes | Assessments (4): baseline, weeks 1, 2, and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 40): "This study was funded through an unrestricted grant from DSE Healthcare Solutions, which manufactures Albolene, one of the products evaluated in this research" | |
| Declaration of interest | None declared | |
| Notes | Albolene, Clarion Brands Inc contains mineral oils, petrolatum, paraffin, ceresin and betacarotene. Mimyx, Stiefel, Coral Gables, FL, contains lipids (triglycerides, phospholipids), squalene, phytosterole, and N‐palmitoylethanolamide (has anti‐inflammatory effect) We mailed the investigator numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 41): "were asked to use the OTC moisturizer twice daily to the randomized right or left target limb and an Rx device to the other randomized target limb" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 41): "The study could only be investigator blinded, as it was not possible to make the two study moisturizers appear identical, since the products were studied as currently marketed". Participants were not blinded. Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each body part, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 41): "The study could only be investigator blinded" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of healthcare providers during the study, and participants and parents were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 1/60 (1.7%), reason reported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled, within‐participant study Setting Dermatology Clinic, High Point, North Carolina, USA Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 20 (all female) Mean age was not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 20 to either body site (within‐participant) Withdrawals/losses to follow‐up 2/20 (10%)
Baseline data Mean overall severity score 2 = mild, 3 = moderate: hyaluronic acid‐based emollient foam 2.9, ceramide‐containing emulsion cream 2.9 | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | EpiCeram contains a blend of ceramides, cholesterol and free fatty acids. Hyalotopic contains water, glycerol, ethylhexyl palmitate, cetearyl alcohol, Theobroma grandiflorum seed butter, petrolatum, dimethicone, parabens, tocopherol acetate and sodium hyalorunate We mailed the investigator numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 186): "This single‐centered, double‐blinded, randomized, split body study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 186): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 186): "Blinded investigator ratings" and "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 2/20 (10%), worsening atopic dermatitis at all sites (1), lost to follow‐up (1). Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, considered as at a low risk of bias. |
| Selective reporting (reporting bias) | High risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared not to be all reported. Results on participant assessments of target site skin appearance for redness, peeling, dryness, stinging/burning, and overall skin irritation were missing. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Outpatient clinic of Dermatology Clinical Trials Center, Mount Sinai School of Medicine, New York, USA Date of study September 2009 to March 2010. Duration of intervention: 4 weeks | |
| Participants | N = 20 (gender not reported) Age range = 14 to 50 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 20 to either body site (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported. Baseline data PGA mild: Eletone N = 2, pimecrolimus N = 2 PGA moderate: Eletone N = 15, pimecrolimus N = 15 Mean PGA (0‐5): Eletone side 3.08, pimecrolimus side 3.03 Mean participants' self‐assessment (0‐3): Eletone side 2.32, pimecrolimus side 2.33 | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 741): "Funding Sources: Fernadale Laboratories" | |
| Declaration of interest | Quote (page 471): "The authors have no relevant conflict of interest to disclose" | |
| Notes | Eletone contains a high lipid content that utilizes a specialized hydrolipid technology, a reverse‐phase formulation of 70% lipids dispersed in 30% outer phase of water We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 737): "medication was randomized in a 1:1 ratio" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 737): "investigator blinded" and "subsequently dispensed by an unblinded study coordinator to ensure investigator blindness" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 737): "investigator‐blinded". Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Dermatology Outpatient Department of the Jose R Reyes Memorial Medical Center, Manila, Philippines Date of study March 2011 to June 2012. Duration of intervention: 8 weeks | |
| Participants | N = 117 (50 female, 47 male and 20 gender unknown) Mean age = 4.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 117 (virgin coconut oil group = 59, mineral oil group = 58) Withdrawals/losses to follow‐up 16/117 overall (13.7%); virgin coconut oil group = 5/59 (8.5%), mineral oil group = 11/58 (19.0%)
Reasons for withdrawal
Baseline data Mean SCORAD index: virgin coconut oil group 13.28, mineral oil group 12.29 Mean TEWL (g/m²/h): virgin coconut oil group 26.68, mineral oil group 24.12 Mean skin capacitance (corneometry units): virgin coconut oil group 32.00, mineral oil group 31.31 | |
| Interventions | Intervention
Comparator
All parents were asked to give the participating children a bath once daily with warm water for 5–10 minutes and to apply the assigned oil immediately after bathing and at night. | |
| Outcomes | Assessments (4): baseline, weeks 2, 4 and 8 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | Quote (page 100): "Conflicts of interest: None" | |
| Notes | Parents were advised not to give or apply any other medication (excluding multivitamins) or moisturiser during the study period. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 102): "The study statistician generated a list of random numbers using the table of random numbers" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 102): "uniform opaque plastic bottles" and "The bottles were coded (A or B) by the pharmacist" and "An assigned resident, who was blinded to the codes, allocated the treatments randomly using the list and dispensed the packaged bottles accordingly. The codes were not disclosed to the investigators until the end of the study". Comment: central allocation, de‐identified uniform bottles. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 102): "Both VCO and mineral oil were obtained from local companies. They were repackaged into uniform opaque plastic bottles with a small opening to mask the color and scent of both oils. There are no other apparent differences between the oils as both are clear, colorless, and of similar viscosity. The bottles were coded (A or B) by the pharmacist". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 102): "Both VCO and mineral oil were obtained from local companies. They were repackaged into uniform opaque plastic bottles with a small opening to mask the color and scent of both oils. There are no other apparent differences between the oils as both are clear, colorless, and of similar viscosity. The bottles were coded (A or B) by the pharmacist". Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | 16/117 (13.7%), virgin coconut oil group = 5/59 (8.5%), mineral oil group = 11/58 (19.0%), reasons reported. Intention‐to‐treat analysis. Comment: the total number of dropouts were unbalanced between the groups and although combined with a intention‐to‐treat analysis, represents an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free from other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, 'other moisturiser'‐controlled within‐participant study Setting 2 Dermatology departments Sahlgrenska University Hospital, Göteborg and University Hospital MAS, Malmö, Sweden Date of study October to March unclear which years. Duration of intervention: 2 weeks | |
| Participants | N = 56 (39 female, 17 male) Mean age = 46 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 56 to either leg (within‐participant) Withdrawals/losses to follow‐up 1/56 (1.8%) before start of treatment Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and week 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 307): "This study was supported by Schering‐Plough AB, Sweden and writing of the article was supported by TFS Trial Form Support, Sweden" | |
| Declaration of interest | Quote (page 307): "Conflict of interest: Petter Olsson is an employee of Schering‐Plough AB, Sweden" | |
| Notes | The investigator was not able to provide us with more information on study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 305‐6): "according to randomization 1:1 ratio" and "The subjects were randomized into two groups (1:1)" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 305): "Single‐blind" and "The allocation of treatment was known to the patients but not to the investigator" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each leg, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 305): "Single‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 1/56 (1.8%) before start of treatment. Intention‐to‐treat analyses. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, 'other moisturiser'‐controlled study Setting Dermatology Department, Hospital do Desterro, Lisbon, Portugal Date of study Unspecified. Duration of intervention: 12 weeks | |
| Participants | N = 23 (7 female, 16 male) Mean age = 9.7 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 23 (unclear how many to each group) Withdrawals/losses to follow‐up Nothing reported Baseline data Mean xerosis scores: Nioleol group 26.6 (SD 22.5), Atopic group 23.3 (SD 17.5), Uriage group 30.0 (SD 28.2), Atoderm group 33.3 (SD 27.3) Mean pruritus scores: Nioleol group 14 (SD 11.4), Atopic group 13.5 (SD 12.1), Uriage group 11.7 (SD 9.8), Atoderm group 14.6 (SD 10.5) Mean cutaneous hydration units: Nioleol group 109 (SD 15.6), Atopic group 107 (SD 12.8), Uriage group 108 (SD 21.3), Atoderm group 99 (SD 16.8) Mean TEWL (g/m²/h): Nioleol group 9.25 (SD 3.5), Atopic group 9.05 (SD 4.3), Uriage group 9.05 (SD 2.4), Atoderm group 8.15 (SD 3.0) | |
| Interventions | Intervention
Comparator 1
Comparator 2
Comparator 3
| |
| Outcomes | Assessments (8): baseline, weeks 1, 2, 3 4, 6, 8 and 12 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Atoderm, Bioderma, Laboratoire Dermatologique contains: aqua, mineral oil (paraffinum liquidum), glycerol, cetearyl isonanoate, glyceryl stearate, PEG‐100 stearate, myreth‐3 myristate, steareth‐21, tocopheryl acetate, mannitol, xylitol, rhamnose, fructo‐oligosaccharides, laminaria ochroleuca extract, cyclopentasiloxane, cyclohexasiloxane, triethanolamine, cetyl alcohol, palmitic acid, stearic acid, acrylates/C10‐30 alkyl acrylate crosspolymer, caprylic/capric trygliceride, disoidum EDTA, phenoxyethanol, chlorphenesin, sodium hydroxide As the study was 18 years old we have not contacted the investigators for missing trial data. As it is unclear how many were randomised to each group, this study is included in Table 4 | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 139): "They were randomised into four groups" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Unclear risk | Nothing reported, unclear how many were randomised to each group. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Outpatient clinic of Dermatology Clinical Trials Center, Mount Sinai School of Medicine, New York, USA Date of study March 2010 to October 2010. Duration of intervention: 4 weeks | |
| Participants | N = 30 (gender not reported) Age range = 4 to 69 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 30 to either body site (within‐participant) Withdrawals/losses to follow‐up 2/30 (6.7%) reasons unreported Baseline data Mean IGA: Hylatopic side 2.62 (SE 0.14), pimecrolimus side 2.62 (SE 0.14) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 671): "Study support was provided by Onset Dermatologics" | |
| Declaration of interest | Quote (page 671): "The authors have no relevant financial conflicts of interest to disclose" | |
| Notes | Hyalotopic contains water, glycerol, ethyl‐hexyl palmitate, cetearyl alcohol, Theobroma grandiflorum seed butter, petrolatum, dimethicone, parabens, tocopherol acetate and sodium hyaluronate We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 669): "were screened and randomized in a 1:1 ratio" Quote (page 668): "A randomized list was created to determine which side the subject applied..." Comment: probably done |
| Allocation concealment (selection bias) | Unclear risk | Quote (page 668): "The list was only available to the un‐blinded study coordinator" The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 668): "investigator‐blinded" "subsequently dispensed by an unblinded study coordinator to ensure investigator blindness". Participants were not blinded. Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 668): "investigator‐blinded". Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/30 (6.7%); reasons unreported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (NCT01202149), and the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two randomised, double‐blind, 'other moisturiser'‐controlled, within‐participant studies Setting Two Dermatology clinics of the Central Hospitals Västerás and Sundsvall, Sweden Date of study During winter 1972‐73. Duration of interventions: 4 weeks | |
| Participants | N = 2 studies of 30 (gender not reported) Mean age not reported Inclusion criteria of the trials
Exclusion criteria of the trials
Randomised N = 2 studies of 30, to either side of the body (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
No other treatments were permitted | |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Quote (page 442): "Aquacare® is a moisturising emulsion containing 10% urea, multi sterols, phospholipids and fatty diols. Calmurid® contains 10% urea, betaine, and a high content lactic acid" As the study was 41 years old we have not contacted the investigators for data. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 443): "after being randomly marked left and right" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | Quote (page 443): "Tubes containing 90 gm of Aquacare HP Cream and Calmurid Cream were packed into identical cartons after being randomly marked left and right". Unclear who packed the tubes and if allocation was indeed concealed to the investigators. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 443): "These were dispensed in a double‐blind fashion" and "Tubes containing 90 gm of Aquacare HP Cream and Calmurid Cream were packed into identical cartons after being randomly marked left and right". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 443): "These were dispensed in a double‐blind fashion" and "Tubes containing 90 gm of Aquacare HP Cream and Calmurid Cream were packed into identical cartons after being randomly marked left and right". Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. Data presented on all randomised participants. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, active‐controlled study Setting Hospital in China Date of study Unspecified. Duration of intervention: 2 weeks | |
| Participants | N = 196 (108 female, 88 male) Mean age = 4 months Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 196 (BoPao + 10% urea ointment = 101, BoPao cream only = 95) Withdrawals/losses to follow‐up Nothing reported Baseline data Disease duration 14 days in BoPao + 10% urea cream and 10 days in Bopao only group | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and week 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | We were unable to retrieve the correct email address for Dr Gao | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 294): translation: "randomly divided" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | Nothing reported, there seems to have been an intention‐to‐treat analysis. Comment: we judged this as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, placebo‐controlled study Setting Investigational site in Poland Date of study March 2012 to June 2013. Duration of intervention: 6 months | |
| Participants | N = 130 (56 female, 67 male and 7 gender unknown) Mean age = 5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 130 (65 to either group) Withdrawals/losses to follow‐up 7/165 overall (4.2%); SBT complex = 3/65 (4.6%), moisturiser base 4/65 (6.2%)
Baseline data Mean SCORAD: SBT complex 26.5 (SE 0.9), moisturiser base 26.4 (SE 1.0) Mean PO SCORAD: SBT complex 32.1 (SE 1.9), moisturiser base 32.4 (SE 1.7) | |
| Interventions | Intervention
Comparator
Quote (page 108): "Alternately or adjunctively to their conventional topical AD treatment (topical corticosteroids or calcineurin inhibitors such as pimecrolimus or tacrolimus). AD treatments were prescribed at selection visit, prior to entering the study. Subjects were not allowed to alter their prescribed AD medication regimen and to continue concomitantly their current moisturiser or moisturising products" | |
| Outcomes | Assessments (4): baseline, days 56, 112 and 168 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 112): "The study was granted by Laboratoire Bioderma, France" | |
| Declaration of interest | Quote (page 112): "All authors are employees of Laboratoire Bioderma, Lyon, France" | |
| Notes | Quote (page 108): "SBT complex contains vitamin B3 palmitoyl ethanolamide (PEA) an anti‐pruritus agent, sucro‐esters, β‐sitosterol an anti‐inflammatory agent and zinc known for its antibacterial action against Staphylococcus aureus" We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 108): "Subjects were randomized at the investigational site in a 1:1 way" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The allocation sequence was generated under SAS v9.2 statistical software" Comment: probably done |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "The allocation sequence was provided to the study center before the beginning of the study. This list assigned each of the product code to a subject number. The complete list was available for the investigator from the beginning of the study. The participants didn't see the allocation sequence. It was impossible for them to guess which product they or the following subjects received". Comment: central allocation. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 165): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "The tested products were packaged identically in white plastic bottle with pomp dispenser [SIC], with the exact same labeling. The color and texture of both topical products were the same: white emulsion. The only difference between both product were the identification code BIXXXVX. Neither the investigator nor the caregivers knew the nature of the products used (tested product or placebo). The investigator and CRO staff was just following the randomization list to allocate the product without knowing if it was the tested product or the placebo. The nature of the products was only known by the Sponsor" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement After email communication: blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 7/165 (4.2%); SBT complex = 3/65 (4.6%), moisturiser base 4/65 (6.2%). Per‐protocol analysis Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐controlled study Setting Dermatology clinic, Karlsruhe, Germany Date of study Unspecified. Duration of intervention: 2 weeks | |
| Participants | N = 69 (39 female, 24 male and 6 gender unknown) Mean age = 27 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 69 (water‐in‐oil emulsion group = 31, water‐in‐oil emulsion plus hydrocortisone group = 32) Withdrawals/losses to follow‐up 6/69 (8.7%) unclear from which group
Baseline data Mean roughness according to participants: 3.97 (SD 2.12) water‐in‐oil emulsion group, 4.31 (SD 1.57) hydrocortisone group Mean itch according to participants: 4.65 (SD 2.51) water‐in‐oil emulsion group, 4.72 (SD 2.46) hydrocortisone group Mean redness according to physicians: 2.23 (SD 0.80) water‐in‐oil emulsion group, 1,97 (SD 0.82) hydrocortisone group Mean roughness according to physicians: 3.42 (SD 0.62) water‐in‐oil emulsion group, 3.13 (SD 0.71) hydrocortisone group Mean TEWL (g/m²/h): 24 water‐in‐oil emulsion group, 20 hydrocortisone group | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 1 and 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Excipial U lipo lotion, Spirig Pharma, Erkingen, Germany contains: urea 4%, lipids 36% We have only included the first week of the trial. As the study was 20 years old we have not contacted the investigators for data. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 554): "randomized double‐blind study design" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 165): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 6/69 (8.7%) unclear from which group. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two randomised, double‐blind, vehicle‐controlled, within‐participant studies Setting Dermatology Clinc, Karlsruhe, Germany Date of study Unspecified. Duration of intervention: 4 weeks with 1 week follow up | |
| Participants | Study 1: N = 20 (14 female, 6 male); Study 2: N = 20 (18 female, 2 male) Mean age study 1 = 25.1 years, mean age study 2 = 22.9 years Inclusion criteria of the trials
Exclusion criteria of the trial
Randomised Study 1: N = 20 to either forearm (within‐participant); Study 2: N = 20 to either forearm (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up in study 1 1/20 lost to follow‐up in Study 2 due to acute exacerbation of eczema Baseline data Study 1 mean TEWL (g/m²/h): 1.38, corneometry 4.8 Study 2 mean TEWL (g/m²/h): 1.50, corneometry 1.9 | |
| Interventions | Study 1 Intervention
Comparator
Study 2 Intervention
Comparator
| |
| Outcomes | Assessments (6): baseline, weeks 1, 2, 3, 4 and 5 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Ingredients Study 1: evening primrose oil 30.0 ml, glycerol monostearate 4.0 ml, cetyl alcohol 6.0 ml, polyoxyethylene glycerol monostearate 7.0 ml, glycerol 15.0 ml, propylene glycol 10.0 ml, purified water to 150.0 ml Ingredients Study 2: evening primrose oil 20%, aqua, paraffinum liquidum, caprylic/capric triglyceride, urea, myristyl lactate, dimethicone, methoxy PEG‐22/dodecyl, glycol copolymer, sodium lactate, PEG‐7 hydrogenated castor oil, sorbitan isostearate, PEG‐2 hydrogenated castor oil, ozokerite, hydrogenated castor oil, chlorhexidine, triclosan, lactic acid, perfume As the study was 17 years old, we have not contacted the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 636): "a vehicle‐controlled, randomised, double‐blind trial" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 165): "double‐blind" and "taken from containers labelled "RIGHT" and "LEFT". The vehicle was similar in both active treatment and placebo arms. Comment: the report provided sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each forearm, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Outcomes were investigator‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up in Study 1, but 1/20 lost to follow‐up in Study 2 due to acute exacerbation of eczema. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, controlled (inactive) study Setting Department of Dermatology Hôpital Purpan Toulouse, France Date of study Unspecified. Duration of intervention: 2 months | |
| Participants | 76 children (gender not reported) Mean age = 4 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 76 (active treatment group = 37, non‐treated group = 39) Withdrawals/losses to follow‐up Baseline data Mean SCORAD: active treatment group 25.96 (SD 7.67), non‐treated group 23.3 (SD 7.63) | |
| Interventions | Both groups received a cleansing bar (A‐Derma, Pierre Fabre Laboratories) Intervention
Comparator
Use of strong/moderate topical corticosteroids allowed. No other moisturiser permitted | |
| Outcomes | Assessments (3): baseline, days 28 and 56 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared. However, the product under research comes from the Pierre Fabre Research Institute, the manufacturer of the drug under investigation | |
| Declaration of interest | None declared, but 1 of the investigators was an employee of Pierre Fabre Research Institute, the manufacturer of the drug under investigation | |
| Notes | Washout period of 1 week before study started Exomega milk contains: evening primrose oil, Rhealba oat extract, chlorphenizine, phenoxyethanol, butyl hydroxy toluene, glycols, paraffin jelly, paraffin oil, shea butter We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 79): "randomized..." Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The randomization was generated by the means of a program created with SAS® software" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Sequence generation, has been done centrally by the pharmaceutical company" Comment: probably done. |
| Blinding of participants and personnel (performance bias) | High risk | Quote (page 79 ): "It was packaged in unidentifiable 200 ml bottles" After email communication: "The investigators were not blind because they could see the products when they were handed over to the patients" Comment: no similar placebo intervention in control group. The outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 79): "It was packaged in unidentifiable 200 ml bottles" Comment: no similar placebo intervention in control group. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. After email communication: "2 subjects were losses in Exomega group between D0 and D28" Comment: we judged this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two‐phase study, the maintenance phase of which was randomised, double‐blind, and placebo‐controlled Setting Multicentre (N = 13) in the Netherlands (N = 12) and Belgium (N = 1) Date of study Unspecified. Duration of first phase 4 weeks (not randomised), then a 16‐week randomised phase | |
| Participants | N = 90 (52 female, 38 male) Mean age = 5.7 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 90, in maintenance phase N = 75 (fluticasone group = 39, placebo group = 36) Withdrawals/losses to follow‐up 3/90 in first phase; serious adverse event (1), withdrawal of consent (1), lost to follow‐up (1) 12/87 (13.8%) in maintenance phase; target TIS > 1 (9), consent withdrawn (1), target TIS ≤ 1 at the original lesion but target TIS > 1 at another lesion (1) and exacerbation of atopic dermatitis (1) between the end of the acute phase and the start of the maintenance phase Baseline data at maintenance phase Mean TIS score: for both groups 0 Mean objective SCORAD: fluticasone group 3.6, placebo group 7.0 | |
| Interventions | Acute phase; all children received twice daily fluticasone propionate 0.005% ointment over 4 weeks on all originally affected sites, even if no visible signs of atopic dermatitis were detectable, and on all newly occurring lesions. Standard moisturiser was supplied for use as required. Intervention
Comparator
During maintenance phase a moisturiser was applied twice daily. On days when study medication was applied, moisturiser was applied only in the morning | |
| Outcomes | Assessments (7): baseline, weeks 4, 6, 8, 12, 16 and 20 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 66): "This study was conducted with financial support from GlaxoSmithKline, London, UK" | |
| Declaration of interest | None declared, but the lead author is an employee of GlaxoSmithKline, Zeist, The Netherlands, the manufacturer of fluticasone propionate 0.005% ointment | |
| Notes | We included only the randomised maintenance phase We mailed investigators numerous times to clarify study details, as there was inconsistency in the data reported regarding objective SCORAD, in the text and the corresponding data in the table, which was also not in agreement with a comment made by the investigators "overall a statistically significant difference was observed between treatment groups (P = 0.021)". However, we received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 60): "Randomization was achieved by a computer‐generated scheme and performed by the statistician" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 60): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 60): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement |
| Incomplete outcome data (attrition bias) | Unclear risk | 12/87 (13.8%) in maintenance phase, reasons reported. Per‐protocol analysis. Comment: balanced and moderate number of drop‐outs combined with the per‐protocol analysis poses an unclear risk of bias for this domain. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, controlled (inactive) study Setting Multicentre, 41 dermatologists in France Date of study January 2004 to end July 2004. Duration of intervention: 6 weeks | |
| Participants | N = 173 (69 female, 93 male and 11 gender unknown) Mean age = 6 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised Withdrawals/losses to follow‐up
Baseline data | |
| Interventions | Both groups received a "hygiene product ‐ pain dermatologique" Intervention
Comparator
Inflammatory lesions in both groups were treated with topical corticosteroids of high [micronized desonide 0.1% cream (Locatop, Laboratoires P. Fabre)] or moderate potency [desonide 0.1% cream (Locapred,Laboratoires P. Fabre)], or both, according to the investigators' regular practice | |
| Outcomes | Assessments (3): baseline, days 21 and 42 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 66): "This work was supported by the Institut de Recherche Pierre Fabre, France" | |
| Declaration of interest | None declared, but 1 of the investigators was an employee of Pierre Fabre Research Institute, the manufacturer of the drug under investigation | |
| Notes | Outcomes assessment parent proxy reports Exomega lotion, Laboratoires Pierre Fabre, France, mainly contains water, petrolatum, shea butter, evening primrose oil, glycerol, paraffin oil, niacinamide, butylene glycol, benzoic acid, carbomer and also specific active Rhealba oat extracts that had previously demonstrated potential beneficial effects on skin inflammation We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 62): "randomly allocated..." "according to a randomization list on the basis of their order of inclusion" After email communication: "the list was provided from the promotors of the study." "A randomisation list was established by Pierre Fabre Biometric Department. The randomization was generated by the means of a program created with SAS® software" "randomization was perform in blocks of 4" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. After email communication: "The study products were labelled with the subject number according the randomization, and each center have to follow and attribute the subject number in the chronological order of arrival at the randomization visit. The entry order of the subjects in the study and the following of the randomization has been checked and validated during the blind review of the study." Comment: central allocation, judged as at a low risk of bias. |
| Blinding of participants and personnel (performance bias) | High risk | Quote (page 62 ): "This emollient was packed in unidentifiable 400‐ml bottles" After email communication: "We prepared the study product RV2478B and the hygiene product as the following; emollient group received hygiene product and RV2478B milk...non‐treated group hygiene product alone" Comment: no corresponding intervention or placebo in control group, uncertainty that knowledge of the allocated intervention was adequately prevented during the study. The outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No corresponding intervention or placebo in control group. Outcomes investigator‐assessed as well as by participants' carers. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | High risk | 7/91 in moisturiser group and 4/82 in the no moisturiser group, were excluded from the analysis reasons reported (9/173 no follow‐up data, 2/173 late exclusions). Incomplete outcome data: 103/173 IDQOL and 102/173 DFI questionnaires completed Comment: we judged this as being at a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled, within‐participant study Setting Section of Dermatology and Venereology, Department of Medicine, Karolinska Institutet at Huddinge University Hospital, Stockholm, Sweden Date of study January to May 1998. Duration of intervention: 2 weeks | |
| Participants | N = 22 (16 female, 6 male) Mean age = 32 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 22 to either forearm (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean TEWL (mg/m²/h): NaCl 40 mg/g and urea 40 mg/g in an oil‐in‐water emulsion side 7.5 (SD 3), urea side 6 (SD 2.5) Mean corneometry units: NaCl 40 mg/g and urea 40 mg/g in an oil‐in‐water emulsion side 54 (SD 4), urea side 51 (SD 5) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 1 and 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 33): "This study was supported by grants from the Swedish Council for Work Life Research, the Swedish Society of Medicine, Karolinska Institutet, Edvard Welander Foundation, Finsen Foundation, Åke Wiberg Foundation, Tore Nilson Foundation for Medical Research and Pharmacia Upjohn Sweden AB" | |
| Declaration of interest | None declared | |
| Notes | With the exception of the NaCl and urea, the other ingredients in the emulsions were identical, namely liquid paraffin, PEG‐5 glyceryl stearate, cetyl alcohol, stearyl alcohol, stearic acid, trometamol, methylparaben, propylparaben, hydrochloric acid and water. Both creams were prepared by Pharmacia & Upjohn Sweden AB, Stockholm, Sweden As the study was 15 years old we have not contacted the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 28): "The study was comparative, double‐blind and randomised" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 28): "The study was comparative, double‐blind and randomised" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each forearm, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, controlled (no treatment), within‐participant study Setting Department of Medicine, Section of Dermatology and Venereology, Karolinska Institutet at Huddinge University Hospital, Stockholm, Sweden Date of study Unspecified. Duration of the intervention 3 weeks | |
| Participants | N = 24 (20 female, 4 male) Mean age = 35 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 24 to either forearm (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
During the study, the participants were allowed to wash in the usual manner, but not to use any other skin care products on their arms | |
| Outcomes | Assessments (3): baseline, days 10 and 21 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | Quote (page 272): "None declared" | |
| Notes | The study also included 20 healthy participants Proderm, Ponsus Pharma AB, Stockholm, Sweden contains: dimethicone (< 1%), glycerol (1% to 5%), stearic acid (5% to10%), propylene glycol (5% to 10%), polyvinyl pyrolidone (1% to 5%), polysorbate 20 (1% to 5%), triethanolamine (1% to 5%), methyl‐, butyl‐, ethyl‐ and propylparabens, and purified water (approximately 80%) We failed to contact any of the investigators of this study. No baseline data nor end value data were reported. The data were reported in box‐and‐whisker plots, and were not interpretable (Table 4). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 273): "a randomized forearm" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | No blinding reported. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, placebo‐controlled, cross‐over study Setting Multicentre (5), Dermatological clinics in Fukuoka Prefecture of Japan Date of study March to June 2006. Duration of intervention: 4 weeks (cross‐over after 2 weeks) | |
| Participants | N = 42 (26 female, 13 male and 3 gender unknown) Mean age = 17.9 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 42 (camellia oil first group = 30, purified water first group = 9, 3 unknown) Withdrawals/losses to follow‐up 3/42 (7.1%), reasons not reported Baseline data Mild eczema: camellia oil first group 17 participants, purified water group 4 participants Moderate eczema: camellia oil first group 13 participants, purified water group 5 participants | |
| Interventions | Intervention
Comparator
Any treatments including topical ointments (e.g. corticosteroids) and oral medications (e.g. antihistamine, corticosteroids) were allowed without changing during the trial | |
| Outcomes | Assessments (2): baseline and week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | We only included the first 2 weeks, before cross‐over. Except for disease severity, no separate data were provided at the end of 2 weeks, only after 4 weeks | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 218): "randomized controlled study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 214): "double‐blind" (translated) Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 214): "double‐blind" (translated) Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information provided to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 3/42 (7.1%), reasons not reported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although a per‐protocol analysis we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Multicentre (unclear how many centres) in US Date of study Unspecified. Duration of intervention: 3 weeks | |
| Participants | N = 80 (51 female, 29 males) Mean age = 24.4 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 80 to either body side (within‐participant) Withdrawals/losses to follow‐up 2/80 (2.5%)
Baseline data "The overall severity (mean total sign and symptom) scores were similar for both sides" | |
| Interventions | Intervention
Comparator
In order to standardise the cleansing regimen, all participants were provided with a non‐medicated cleansing bar (Cetaphil Gentle Cleansing bar, Galderma Laboratories, Inc) | |
| Outcomes | Assessments (3): baseline, weeks 1 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 232): "This study was financially supported by Galderma Laboratories Inc, Fort Worth, Texas" | |
| Declaration of interest | None declared, but 1 of the investigators was an employee at Galderma Laboratories the manufacturer of the desonide lotion (DesOwen), the moisturiser and the cleansing bar | |
| Notes | The moisturiser (Cetaphil Moisturising cream, Galderma Laboratories, Inc, Forth Worth, Texas) possesses humectant, emollient and occlusive properties and contains: aqua, glycerol, petrolatum, dicaprylyl ether, dimethicone, glyceryl stearate, cetyl alcohol, Prunus amygdalus (synonym: Prunus dulcis) (sweet almond) oil, PEG‐30 stearate, tocopheryl acetate, acrylates/C10‐30 alkyl acrylate crosspolymer, dimethiconol, benzyl alcohol, phenoxyethanol, glyceryl acrylate/acrylic acid copolymer, propylene glycol, disodium EDTA, sodium hydroxide As the study was 18 years old we did not contact the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 228): "Treatments were randomly assigned by a third party" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 227): "investigator‐blinded", however, participants were not blinded. Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 227): "investigator‐blinded". Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/80 (2.5%), reasons reported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Three‐phase study in which the maintenance phase was randomised, double‐blind, and had an active control Setting Multicentre (16) in USA and Canada Date of study Unspecified. Duration of intervention: Stabilisation phase was up to 4 weeks and maintenance phase was 20 weeks | |
| Participants | N = 372 entered the stabilisation phase (216 female, 156 male), and 348 entered the maintenance phase (more than half the population was female) Mean age = 16.8 years (7.1 years for children and 35.9 years for adults) Inclusion criteria of the trial
For entrance in maintenance phase
Exclusion criteria of the trial
Randomised N = 348 (maintenance phase; fluticasone propionate group = 229, vehicle group = 119) Withdrawals/losses to follow‐up Lost to follow‐up: 44/348 overall (12.6%); fluticasone propionate group = 32/229 (14%), vehicle group = 12/119 (10%) Relapse: 110/348 overall (31.6%); fluticasone propionate group = 27/229 (11.8%), vehicle group = 83/119 (69.7%) Baseline data Moderate eczema 63%, mean Rajka and Langeland Severity Grading score was 7 for all participants, approximately 3/4 of the population suffered continuously from atopic dermatitis without remission | |
| Interventions | All participants entered an open‐label stabilisation phase of up to 4 weeks' duration on twice daily fluticasone propionate 0.05% cream and used a moisturiser at least once a day Maintenance phase Intervention
Comparator
It is unclear which moisturiser was used | |
| Outcomes | Assessments Maintenance phase(7): 'baseline', weeks 2, 4, 8, 12, 16 and 20, then 'follow‐up phase' Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 536): "This study (FPC40002) was conducted with a grant from Glaxo Wellcome Inc., Research Triangle Park, NC, U.S.A." | |
| Declaration of interest | None declared, but 1 of the investigators was an employee of Glaxo Wellcome Inc, the manufacturer of the fluticasone propionate cream | |
| Notes | For participants who completed the maintenance phase on intermittent fluticasone propionate 0.05% cream without a relapse, the intermittent fluticasone propionate 0.05% cream dosing/moisturiser regimen was extended for a further 24 weeks in a subsequent follow‐up phase As the study was 14 years old we have not contacted the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 147): "Subjects who entered the second part of the study were randomized (2: 1 within each age stratum)" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 147): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information provided to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Unclear risk | 44/348 (12.6%) were lost to follow‐up; fluticasone propionate group = 32/229 (14%), vehicle group = 12/119 (10%). Intention‐to‐treat analysis. Comment: balanced but moderate number of drop‐outs (although an intention‐to‐treat analysis), poses an unclear risk of bias for this domain. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, controlled, cross‐over study Setting Out‐patient Dermatology Clinic at Queen Elizabeth Hospital for Children, London, UK Date of study Unspecified. Duration of intervention: two 4‐week periods (cross‐over study) with 2‐week separation | |
| Participants | N = 30 (17 female, 9 male and 4 gender unknown) Mean age = 4.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 30 (unclear how many to each arm) Withdrawals/losses to follow‐up 4/30 (13.3%)
Baseline data Therapeutic control was adequate for 10 patients and 'marginal with exacerbations in 16' | |
| Interventions | Intervention
Comparator
Use of other bath additives was not allowed. Emulsifying ointment or aqueous cream were used as a soap substitute in all cases throughout the study period. Pre‐study topical therapy was continued unaltered during the study | |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 (then cross‐over phase) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, but under acknowledgements "Also Dennis Joseph from Stiefel for his contribution" | |
| Declaration of interest | None declared | |
| Notes | Oilatum Plus includes triclosan and benzalkonium chloride. We have only included the first 4 weeks before the cross‐over part As the study was 21 years old we have not contacted the investigators for data Inconsistency of reporting of number of participants; stated on 2nd page that 32 patients (88%) experienced at least 3 exacerbations. Number randomised to each arm was unclear and there are inconsistencies in data reporting (Table 4). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 42): "The order in which the two preparations were used was determined by a computer generated random code" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 42): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement |
| Incomplete outcome data (attrition bias) | Unclear risk | 4/30, reasons reported, unclear how many participants randomised to each arm, and inconsistent reporting of number of participants throughout study. Intention‐to treat‐analysis. Comment: we judged this as being at an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two randomised, investigator‐blinded, 'other moisturiser'‐controlled, studies Setting Red Cross Children's War Memorial Hospital, Cape Town, South Africa Date of study February 2013 to July 2013. Duration of intervention: 3 months | |
| Participants | N = 120: Study 1 N = 40 (17 female, 20 male and 3 gender unknown); Study 2 N = 80 (37 female, 36 male and 7 gender unknown) Mean age = 5.6 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 120: Study 1 N = 40 (aqueous cream group = 20, baby oil group = 20), Study 2 N = 80 (cetomacrogol group = 20, emulsifying ointment group = 20, glycerol/petroleum group = 20, petroleum group = 20) Withdrawals/losses to follow‐up Study 1: 3/40 (7.5%); 3 lost to follow‐up in baby oil group Study 2: 7/80 (8.8%); cetomacrogol group (3), emulsifying ointment group (0), glycerol/petroleum group (1), petroleum group (3) Baseline data Study 1: mean SCORAD: aqueous cream group 33.11 (SD 16.82), baby oil group 23.02 (SD 11.79) Study 1: mean POEM: aqueous cream group 11.50 (SD 7.97), baby oil group 9.00 (SD 5.06) Study 1: mean NESS: aqueous cream group 9.50 (SD 3.25), baby oil group 10.20 (SD 1.96) Study 1: mean IDQOL aqueous cream group 7.00 (SD 4.86), baby oil group 6.45 (SD 4.78) Study 2: mean SCORAD: cetomacrogol group 24.78 (SD 16.90), emulsifying ointment group 28.40 (SD 16.16), glycerol/petroleum group 21.52 (SD 15.24), petroleum group 26.77 (SD 12.09) Study 2 mean POEM: cetomacrogol group 9.47 (SD 6.02), emulsifying ointment group 9.10 (SD 4.56), glycerol/petroleum group 9.75 (SD 7.23), petroleum group 11.29 (SD 7.06) Study 2 mean NESS: cetomacrogol group 9.35 (SD 2.40), emulsifying ointment group 9.85 (SD 2.83), glycerin/petroleum group 9.25 (SD 3.64), petroleum group 10.00 (SD 3.35) Study 2 mean IDQOL: cetomacrogol group 7.23 (SD 4.48), emulsifying ointment group 7.25 (SD 5.84), glycerin/petroleum group 7.35 (SD 6.75), petroleum group 6.76 (SD 3.72) | |
| Interventions | Study 1 Intervention
Comparator
All the participants in study 1 used emulsifying ointment as a moisturiser Study 2 Intervention
Comparator 1
Comparator 2
Comparator 3
All participants in Study 2 washed with baby oil and applied this as soap instead of aqueous cream. All the participants in both studies continued to use clinic‐prescribed topical steroids during the study period, and brought back all tubes so that the amount used during the previous month could be recorded at every visit. | |
| Outcomes | Assessments (4): baseline, months 1, 2 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 738): "The study was investigator initiated, and all emollients were requested from and donated by Sekpharma (SA), a generic medicine supply company" | |
| Declaration of interest | None declared | |
| Notes | Frequency of use during day or week were not reported. There were some inconsistencies between text and figures. No end data were reported for Study 2, just stated that all scores tended to decline. We mailed investigators numerous times to clarify study details, but received no response (Table 2; Table 4). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 781): "Patients were randomised using an automatic online enrolment system in a 1:1 ratio for study 1 or a 1:1:1:1 ratio for study 2" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 781): "single (assessor)‐blind trial" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 781): "single (assessor)‐blind trial" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | Study 1: 3/40 (7.5%): 3 lost to follow‐up in baby oil group. Study 2: 7/80 (8.8%): cetomacrogol group (3), emulsifying ointment group (0), glycerol/petroleum group (1), petroleum group (3). Per‐protocol analysis |
| Selective reporting (reporting bias) | Unclear risk | The protocol for the study was available (NCT02084472) and the prespecified outcomes and those mentioned in the Methods section appear to have been reported. However, for Study 2 very few data were provided ("all mean scores tended to decline over time", without providing the data). |
| Other bias | Unclear risk | There was baseline imbalance in SCORAD for study 1, but not for the other disease severity scores. Comment: we judged this as at an unclear risk of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐controlled study Setting Multicentre (4), dermatology outpatient clinics in Rotterdam, the Netherlands Date of study February 2009 to February 2012. Duration of intervention: 4 weeks | |
| Participants | N = 39 (14 female, 25 male) Mean age = 3.4 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 39 (corticosteroid group = 19, placebo group = 20) Withdrawals/losses to follow‐up 4/39 (9.8%); all in placebo group
Baseline data Mean objective SCORAD was 35 to 40 or higher (indicating severe eczema) on 2 measuring points before start of treatment | |
| Interventions | Intervention
Comparator
For the face, mometasone furoate 0.1% ointment was diluted 1:19, although pimecrolimus 1% cream was used in some participants. For the body, a 1:3 dilution was used Participants were not allowed to use other moisturisers or topical agents | |
| Outcomes | Assessments (5): baseline, days 1, 4, 7, 14 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1076): "Nonrestricted funds were received from Schering‐Plough BV, Fagron, Astellas, Molnlycke, Aardbeesie Project (www.aardbeesie.nl), and Foundation for Pediatric Dermatology Rotterdam (www.pediatric‐dermatology.com)" | |
| Declaration of interest | Quote (page 1076): "Conflicts of interest: None declared" | |
| Notes | We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1077): "Computerized randomization was performed by our statistician" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 1076): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 1076): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement |
| Incomplete outcome data (attrition bias) | Low risk | 4/39 (9.8%); all in placebo group, reasons reported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this study to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (EudraCT Number: 2007‐005232‐81) and the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Skin Center, Faculty of Medicine, Srinakharinwirot University, Bangkok, Thailand Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 31 (11 female, 18 male and 2 gender unknown) Mean age = 4.3 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 31 to either body site (within‐participant) Withdrawals/losses to follow‐up 2/31 (6.5%) lost to follow‐up Baseline data Mean SCORAD: spinosa kernel oil cream side 25.6 (SD 7.9), hydrocortisone cream side 25.7 (SD 7.5) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (4): baseline, weeks 2, 4 and 8 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 824): "The authors wish to thank Hoe Pharmaceutical Sdn Bhd for their support of the experimental S [spinosa kernel oil] cream for this research" | |
| Declaration of interest | Quote (page 824): "Potential conflicts of interest: None" | |
| Notes | The spinosa kernel oil cream consisted of linoleic acid (omega‐6), oleic acid, palmitic acid, stearic acid, polyphenols, tocopherols, phenolic acid, and squalene. Cream base was comprised of butylene glycol, mineral oil, ethylhexyl stearate, tocopheryl acetate, sodium polyacrylate | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 821): "All patients who met eligible criteria were randomized. The randomization schedule was prepared by a third party" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 821): "randomized assessment blind controlled trial" and "The cassettes of both agents were similar in shape and color, however they were labeled as 'left' or 'right'." On the hydrocortisone side two different tubes or cassettes were used compared to one tube or cassette on the S [spinosa kernel oil] cream side. The trialists state "assessment blind", but did not report on the blinding of participants. Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 821): "randomized assessment blind controlled trial" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: uncertainty about the effectiveness of blinding of healthcare providers during the study and participants and parents seem not to be blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/31 (6.5%) lost to follow‐up. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this trial to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant pilot study Setting Indiana University Medical Center, Indianapolis, USA Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 6 (gender not reported) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 6 to either body side (within‐participant) Withdrawals/losses to follow‐up Nothing reported Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page AB67): "Commercial support: Poster production sponsored by Ferndale Laboratories, Inc" | |
| Declaration of interest | None declared | |
| Notes | Poster abstract. Little information was provided, PI was not able to provide more study data (Table 2), therefore the study is moved to Table 4 | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page AB67): "were randomized" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 1482): "investigator‐blinded study" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed and participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Unclear risk | Nothing reported, little information provided. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, 'other moisturiser'‐controlled, within‐participant study Setting Icahn School of Medicine at Mount Sinai, New York, USA Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 10 (6 female, 4 male) Mean age = 31 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 10 to either body side (within‐participant) Withdrawals/losses to follow‐up None reported Baseline data Mean IGA scores: 2 (5 participants), 3 (5 participants) | |
| Interventions | Intervention
Comparator
Frequency of application not reported | |
| Outcomes | Assessments (2): baseline and week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1484): "This study is funded by Puracap Pharmaceutical, LLC. " | |
| Declaration of interest | Quote (page 1484): "Dr. Kircik has served as an advisor, investigator, and consultant for Puracap Pharmaceutical, LLC" | |
| Notes | EpiCeram contains a blend of ceramides, cholesterol and free fatty acids Most data had to be estimated from graphs, principal investigator was not able to provide us with more exact study data (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1482): "Subjects were randomized at a 1:1 ratio" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "pre scheduled list by the CRO" [Clinical research organisation] Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 1482): "investigator‐blind study" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 1482): "investigator‐blind study" Outcomes were investigator‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, vehicle‐controlled study Setting Multicentre (3), outpatient centres in Cologne, Kiel and Heilbronn, Germany Date of study October 2003 and January 2005. Duration of intervention: 4 weeks | |
| Participants | N = 99 (39 female, 58 male and 2 gender unknown) Mean age = 3.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 99 (pale sulfonated shale oil (PSSO) = 51, vehicle = 48) Withdrawals/losses to follow‐up 2/99 (2%), 1 participant in each group did not appear for any follow‐up visits 7/99 (11.7%) "were excluded from per‐protocol analysis"
Baseline data Mild eczema: PSSO group 35 participants, vehicle group 37 participants Moderate eczema: PSSO group 15 participants, vehicle group 10 participants Mean EASI: PSSO group 13.4 (SD 3.7), vehicle group 13.0 (SD 3.1) | |
| Interventions | Intervention
Comparator
Accompanying medications of any kind (systemic or topical) were not permitted. The use of skin care products on areas of unaffected skin was allowed as usual | |
| Outcomes | Assessments (4): baseline, weeks 1, 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1176): "The preparation of this article was supported by an educational grant to Dr Schöllmann." | |
| Declaration of interest | Quote (page 1176): "Prof. Korting collaborates with Ichthyol‐Gesellschaft, Hamburg, in the development of topical drugs for skin diseases. Dr Cholcha and Dr Wolff are employed by the company that supported this multicentre study but do not have any personal financial interest in the research described in the manuscript" | |
| Notes | Quote (page 1177): "The verum product was an oil‐in‐water cream with sodium bituminosulfonate 4%, pale (trade name: Ichthosin cream; manufacturer: Ichthyol‐Gesellschaft, Hamburg, Germany). The vehicle was a correspondingly coloured cream without an active ingredient, containing propylene glycol, glycerol monostearate, cetyl alcohol, medium chain triglycerides, macrogol‐1000‐glycerol monostearate, white vaseline, purified water, and – for adjustment to the verum regarding colour – additionally spirit caramel and quinoline yellow." The lead author died in 2012 and the other authors could not be contacted | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1177‐ 8): "The randomization occurred in blocks of 2 with a randomizing ratio of 1:1" and "Block randomization was chosen and calculated with the validated program RanCode (version 3.6, IDV Gauting, Germany)" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/99 (2%), 1 from each group did not appear for any follow‐up visit and were excluded from intention‐to treat analysis. Although 7 others appeared to have violated the protocol, these seem to have been included in the analyses. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled, within‐participant study Setting Multicentre (4) in France and Italy Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 46 (21 female, 25 male) Age range 6 to 12 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 46 to either body side (within‐participant) Withdrawals/losses to follow‐up 2/46 (4.3%), exacerbation (1), personal reasons (1) Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | The emulsion contained almond oil, olive oil, gamma oryzanol and alpha tocopherol As the study was 20 years old we did not contact the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 720): "une etude clinique randomisée" (translation: randomised clinical study) Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 720): "double‐aveugle" (translation: double‐blind): pre‐coded tubes and the investigator was not aware of content. Products were identical except for the lactic acid. Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received on each body side, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 720): "double‐aveugle" (translation: double‐blind): pre‐coded tubes and the investigator was not aware of content. Products were identical except for the lactic acid. Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 2/46 (4.3%), exacerbation (1), personal reasons (1). Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | High risk | The following predefined outcomes were not addressed; pruritus, and efficacy according to the parents and only P value provided for erythema,papules and excoriations. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, controlled, within‐participant study Setting Multicentre (6) USA Date of study Unspecified. Duration of intervention: 12 weeks | |
| Participants | N = 74 (42 female, 32 male) Mean age = 26 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 74 to either body side (within‐participant) Withdrawals/losses to follow‐up 8/74 (10.8%)
Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
Rescue medication (triamcinolone cream 0.1%) was provided in the event of flare | |
| Outcomes | Assessments (7): baseline, weeks 2, 4, 6, 8, 10 and 12 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared but product under investigation was from Stiefel Labs Inc | |
| Declaration of interest | None declared | |
| Notes | MimyX cream contains purified water, olive oil, glycerol, pentylene glycol, palm glycerides, vegetable oil, hydrogenated lecithin, squalane, betaine, palmitamide MEA, hydroxyethyl cellulose, sodium carbomer, carbomer, xanthan gum Poster abstract. Little information was provided. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 45): "Subjects were randomized" and "randomly assigned to one of two groups" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "each subject was assigned a number in numerical sequence of entry. This number was used to enter a computer‐generated assignment to each of the two groups" Comment: probably done |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: Central allocation by sponsor and "The study products were dispensed in opaque bags" Comment: central allocation, de‐identified study products. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 45): "investigator blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. After email communication: "investigators were blinded and the containers were not labeled with what it contained" and "The study products were dispensed in opaque bags" Comment: blinding of investigators was ensured, but participants were not blinded. We judged this as an unclear risk of bias. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 45): "investigator blind" Outcomes were investigator‐assessed. Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study. Insufficient information to permit a clear judgement. After email communication: "investigators were blinded and the containers were not labeled with what it contained" and "The study products were dispensed in opaque bags" Blinding of the outcomes assessors was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 8/74 (10.8%), reasons reported. Intention‐to‐treat analysis. |
| Selective reporting (reporting bias) | High risk | Erythema, pruritus, and papulation/induration/oedema assessment and BSA assessment were predefined outcomes that were not addressed in the Results. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre (4) in Stockholm (2), Uppsala, Linköping, Sweden Date of study Specified February to April but unclear which year. Duration of intervention: 30 days | |
| Participants | N = 110 (93 female, 16 male and 1 gender unknown) Mean age = 34 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 110 (unclear how many randomised to each arm) Withdrawals/losses to follow‐up 1/110 (< 1 %) dropped out, unclear from which group Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and day 31 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, however, study was probably sponsored by ACO Hud AB, Sweden (see Andersson 1999; Lodén 2002). | |
| Declaration of interest | None declared. Dr Lodén was an employee of ACO Hud AB. | |
| Notes | Unclear how many participants were randomised to each arm, and also how frequently the treatments were applied. The data were estimated from box and whisker plots, which made them difficult to use (Table 4). As the study was 15 years old, we did not contact the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 201): "randomized, double‐blind" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 201): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 1/110 (< 1 %) dropped out, unclear from which group. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐ and placebo‐controlled study Setting Multicentre in Stockholm, Uppsala, Linköping, Göteborg, Umeå and Malmö, Sweden Date of study February‐April, unclear which year. Duration of intervention: 30 days | |
| Participants | N = 197 (151 female, 46 male) Mean age = 33 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 197 (glycerol group = 68, urea group = 63, placebo group = 66) Withdrawals/losses to follow‐up Not reported Baseline data Not reported | |
| Interventions | Intervention
Comparator 1
Comparator 2
The participants were allowed to continue use of topical corticosteroids | |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 47): "The study was sponsored by Aco Hud AB, Stockholm, Sweden" | |
| Declaration of interest | None declared. Dr Lodén was an employee of ACO Hud AB. | |
| Notes | The glycerol cream contained 20% glycerol, aqua, petrolatum, cream, canola, mineral oil, cetearyl alcohol, glyceryl stearate, dimethicone, PEG‐100 stearate, glyceryl polymethacrylate, cholesterol, propylene glycol, methylparaben and propylparaben. In the placebo cream, glycerol was replaced with water. The urea cream contained 4% urea and 4% NaCl as water‐binding substances in an oil‐in‐water emulsion, pH about 5. Other ingredients were paraffin liquidum, PEG‐5‐glyceryl‐stearate, cetyl alcohol, stearyl alcohol, stearic acid, trometamol, methylparaben, propylparaben, hydrochloric acid and water As the study was 14 years old we did not contact the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 45): "were randomized into three groups" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 45): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 45): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | No drop‐outs reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, 'other moisturiser'‐controlled study Setting Multicentre (5), dermatology departments in Pavia, Bargamo, Bari, Sassudo and Milan, Italy Date of study January to May 2013. Duration of intervention: 6 weeks | |
| Participants | N = 107 (48 female, 59 male) Mean age = 6 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 107 (pro‐AMP group = 72, placebo group = 35) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean eczema severity score: pro‐AMP group 6.1 (SD 2.5), placebo group 5.3 (SD 3) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 3 and 6 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 274): "This was a non‐profit study" and "Isdin Srl has kindly provided the products used in the trial" | |
| Declaration of interest | Quote (page 274): "MM is an employee of Isdin Srl. He was involved in the study design and in the manuscript preparation" and "The other authors have declared no conflict of interest" | |
| Notes | Nutratopic Pro‐AMP cream contains 2.5% rhamnosoft (Biosaccharide GUM‐2), ceramides (ceramide 3), and 2% ILE (Isdin Barcelona, Spain) We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 272): "Computer‐generated number randomization list was used" "with imbalanced treatment allocation 2:1" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Concealment was obtained using sealed envelopes" Comment: allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 272): "assessor blinded" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement and parents and participants seem not to be blinded. After email communication: "Physicians performing the first visit with treatment allocation were not the physicians performing the follow up visits. The latter were not aware of the type of treatment". Comment: the report provided sufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received. However, participants were not blinded and so we judged this as being at an unclear risk of bias. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 272): "assessor blinded" Outcomes were investigator‐assessed as well as participant assessed. Comment: blinding of the outcomes assessors, was ensured, and it was unlikely that the blinding could have been broken, however, participants and parents seem not to have been blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (NTR4084) and the prespecified outcomes and those mentioned in the methods section appear to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Wake Forest University Health Sciences Clinical Studies Center, North Carolina, USA Date of study Not reported. Duration of intervention: 3 weeks | |
| Participants | N = 39 (22 female, 17 male) Mean age = 7.9 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 39 (BRC‐Gly group = 13, BRC‐Cer group = 13, OTC‐Pet group = 13) Withdrawals/losses to follow‐up 1/39 (2.5%) of OTC‐Pet group due to disease flaring Baseline data Mean EASI: BRC‐Gly group 4.05 (SD 3.68), BRC‐Cer group 4.23 (SD 2.34), OTC‐Pet group 5.30 (SD 3.70) | |
| Interventions | Intervention
Comparator 1
Comparator 2
| |
| Outcomes | Assessments (3): baseline, days 7 and 21 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Text in the submitted version that was deleted in the published version read: "This study was supported by Beiersdorf, Inc". | |
| Declaration of interest | Text in the submitted version that was deleted in the published version read: "Dr Fleischer has received research, speaking and/or consulting support from Astellas, Centocor, Amgen, Galderma, Stiefel, Medicis and Intendis. Dr Weber is an employee of Beiersdorf Inc". | |
| Notes | Oral antihistamines were allowed if not initiated or discontinued during study. Compliance was monitored with a diary. See Notes of Characteristics of included studies of Abramovits 2008 for details on Atopiclair and of Draelos 2011 for details on EpiCeram. Aquaphor Healing Ointment contains petrolatum, cera microcristallina, panthenol, glycerol, bisabolol. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 531 ): "were randomized 1:1:1..." Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "This was an investigator‐initiated protocol. The clinical studies coordinator, who did no assessments, followed a randomization chart obtained online for trichotomization into 3 groups. On this chart the coordinator followed the next randomization" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "No person involved in any clinical assessment had access to the randomization scheme. There were no sealed envelopes" Comment: reasonable to assume allocation adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 535): "Each of the moisturizers was distributed in a plain white jar so that subjects and investigators were blinded..." |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 535): "Each of the moisturizers was distributed in a plain white jar so that subjects and investigators were blinded...". After email communication:"No person involved in any clinical assessment had access to the randomization scheme" Comment: Outcomes were investigator‐assessed as well as participant assessed. Blinding of participants and key study personnel was ensured, and it is unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 1/39 (2.5%) of OTC‐Pet group, reason reported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (NCT01093469), and the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, active‐controlled study Setting Multi‐centre (20 paediatricians) in France Date of study March to June 2003. Duration of intervention: 3 weeks | |
| Participants | N = 90 children (41 female, 45 male and 4 gender unknown) Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised Withdrawals/losses to follow‐up 4/90 (4.4%): unclear from which groups, lost to follow‐up (1), prematurely stopped the study for their own convenience (not for side effects) (2), forms incorrectly filled out or incomplete (1) Baseline data | |
| Interventions | Intervention (Group A)
Comparator 1 (Group B)
Comparator 2 (Group C)
Comparator 3 (Group D)
Comparator 4 (Group E)
| |
| Outcomes | Assessments (3): baseline, days 7 and 21 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared, but 3 investigators were employees of Laboratoires Expanscience, Epernon, France, manufacturer of the product under research | |
| Notes | STELATOPIA moisturiser, Mustela; Laboratoires Expanscience, France contains: 2% sunflower oleodistillate, essential fatty acids, bio‐ceramides, β‐sitosterol and a complex of emulsifying sugars We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 607): "randomly assigned...". "Each study center received one test kit containing five treatments. The five treatment options studies were successively allocated to patients according to chronological order of entry in the study. Patients were randomized based on chronological order of entry in the study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "there was randomized attribution list, the order of attribution to different subjects was pre‐determined in a balanced fashion, following the attribution list" Comment: probably done |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: Central allocation, identical packages "On this pack, was only printed the number of chronological attribution to follow. The doctor did not know what was inserted in this pack (what protocol of emollient and steroid)" Comment: reasonable to assume allocation adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. After email communication: "One was lost to follow‐up. Two stopped prematurely the study for their own convenience (not for side effect). One file was not correctly filled and incomplete and was excluded from the analysis. So, we studied 86 children." Per‐protocol analysis. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Open randomised, 'other moisturiser'‐controlled study Setting Institute of Preventive Pediatrics and Rehabilitation Scientific Center of Children's Health RAMS, Moscow, Russia Date of study Unspecified. Duration of intervention: 1 year | |
| Participants | N = 44 (24 female, 20 male) Age range: 6 months to 12 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 44 (Locobase Repair group = 22, Atoderm group = 22) Withdrawals/losses to follow‐up 4/44 (9%); Locobase Repair group due to respiratory infection (1), Atoderm group due to protocol violation (1), worsening skin (2) Baseline data Median SCORAD: Locobase Repair group 38.3, Atoderm group 36.6 Mean daily consumption of topical corticosteroid: Locobase Repair group 2.8 g, Atoderm group 3.0 g Mean CDLQI: Locobase Repair group 15.83, Atoderm group 15.35 | |
| Interventions | Intervention
Comparator
As a basic therapy the participants received 0.1% hydrocortisone 17‐butyrate cream (Locoid) 1 to 3 times a day as needed. Systemic therapy in both groups was of antihistamines (cetirizine, levocetirizine) in dosages for age | |
| Outcomes | Assessments (4): baseline, months 1, 6 and 12 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Locobase Repair contains petrolatum, water, paraffin, liquid paraffin, glycerol, sorbitan oleate, carnauba wax, cholesterol, ceramides 3, oleic acid, palmitic acid, carbomer, tromethamine. Atoderm contains vaseline‐glycerol complex in dispersed state, sodium salt of EDTA, vitamin E, phenoxyethanol, parabens We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (translated from Russian): "All the children were randomly assigned into 2 groups of 22 people each" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "a computer generated randomisation list was used" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "sealed envelops for allocation, provided by research department" Comment: reasonable to assume allocation adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 4/44 (9%), reasons reported, per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre (2) in USA Date of study Unspecified. Duration of intervention: 8 weeks | |
| Participants | N = 50 (gender not reported) Mean age = 29.8 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 50 (oatmeal group = 25, occlusive vehicle group = 25) Withdrawals/losses to follow‐up 3/50 (6%): all in occlusive vehicle group Baseline data IGA score: oatmeal group 2.58, occlusive vehicle group 2.58 Mean EASI: oatmeal group 6.55 (SD 5.10), occlusive vehicle 8.87 (SD 6.10) Mean DLQI: oatmeal group 5.96 (SD 5.80), occlusive vehicle 7.46 (SD 4.29) | |
| Interventions | Intervention
Comparator
Participants were allowed to use their normal topical medications for their atopic dermatitis | |
| Outcomes | Assessments (3): baseline weeks 2, 4 and 8 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page AB67) "Commercial support: 100% sponsored by Johnson and Johnson Consumer Products Worldwide" | |
| Declaration of interest | None declared, but several investigators were employees of Johnson and Johnson Consumer Products Worldwide, the manufacturer of the drug under investigation | |
| Notes | Poster abstract with limited details provided. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page AB67): "multicenter, double‐blinded, randomized clinical study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The randomization was generated by SAS by a statistician. The method used to assign patients to treatment was a block design. Subjects were assigned in a 1:1 ratio to the two treatments, with a block size of 4." Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "The treatment codes (M and C) were kept in a sealed envelope by our clinical supplies department. The tubes were randomized labeled prior to sending to the investigator ‐ so subjects were dispensed treatment according to enrollment/randomization order. Both treatment creams were in the same 6 oz. beige tube, same cap color. Both creams were basically the same color and unfragranced". Comment: central allocation, de‐identified tubes. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page AB67): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "Both treatment creams were in the same 6 oz. beige tube, same cap color. Both creams were basically the same color and unfragranced" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page AB67): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. After email communication: "Both treatment creams were in the same 6 oz. beige tube, same cap color. Both creams were basically the same color and unfragranced" Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 3/50 (6%), all in the occlusive vehicle group. Intention‐to‐treat analysis performed. Comment: we considered this trial to be at a low risk of bias because the total number of dropouts was low, and intention‐to‐treat analysis was performed. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. We received additional information from the principal investigator. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, 'other moisturiser'‐controlled, within‐participant study Setting Multicentre, Korea Date of study Unspecified | |
| Participants | N = 31, 5 participants had atopic dermatitis (gender not reported) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 31 to either side of the face (within‐participant), 5 participants had atopic dermatitis Withdrawals/losses to follow‐up Nothing reported Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and week 2 Outcomes of the trial (as reported)
| |
| Funding source | None declared | |
| Declaration of interest | None declared, but 3 authors were employees of R&I Center, Cosmax Co. Ltd, the manufacturer of the drug under investigation | |
| Notes | Three PPARα activators used (Euryale ferox, Euphorbia lathyris, Rosa multiflora) Only 5 participants met our inclusion criteria. No individual patient data (Table 4) | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page S95): "were randomly applied on half of the face" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | Nothing reported regarding blinding. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Nothing reported regarding blinding. Outcomes were investigator‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Unclear risk | Little information provided. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Department of Dermatology and Cutaneous Biology Research Institute, Yonsei University College of Medicine, Seoul, Korea Date of study Unspecified. Duration of intervention: 6 weeks | |
| Participants | N = 40 (17 female, 23 male) Mean age = 12.95 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 40 (ceramide containing moisturiser group = 20, control moisturiser group = 20) Withdrawals/losses to follow‐up 8/40 overall (20%): ceramide containing moisturiser group (5), control moisturiser group (3)
Baseline data Mean EASI: ceramide containing moisturiser group 10.45 (SE 1.17), control moisturiser group 9.51 (SE 1.02) Mean TEWL (g/m²/h): ceramide containing moisturiser group 45 (SE 5), control moisturiser group 42 (SE 4) Mean corneometry units: ceramide containing moisturiser group 18 (SE 2), control moisturiser group 24 (SE 3) | |
| Interventions | Intervention
Comparator
Corticosteroid cream also applied (Zemaderm) twice daily for 6 weeks | |
| Outcomes | Assessments (3): baseline, weeks 3 and 6 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Atobarrier cream, ceramide‐containing moisturiser, APDDR‐0801. APDDR‐0801 is the research name of in the Atobarrier cream We received responses to our request for study details (Table 2) | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 227): "randomized, controlled, double‐blinded" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "block randomization method" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "central randomization independent of researcher who's doing enrollment and also used coded container" Comment: central allocation, de‐identified drug containers. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 227): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 227): "double‐blind" Outcomes were investigator‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | High risk | 8/40 (20%); ceramide containing moisturiser group (5), control moisturiser group (3). Per‐protocol analysis. Comment: the total number of dropouts, combined with a per‐protocol analysis represented a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre (USA) Date of study Unspecified. Duration of intervention: 3 weeks | |
| Participants | N = 49 (gender not reported) Age = 2 to 15 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 49 (colloidal oatmeal group = 25, prescription barrier repair cream group = 24) Withdrawals/losses to follow‐up 7/49 (14.3%), unclear from which group(s) Baseline data Mean EASI: colloidal oatmeal group 3.56 (SD 1.8), prescription barrier repair cream group 3.15 (SD 2.5) Mean VAS itch: colloidal oatmeal group 1.4 (SD 2.3), prescription barrier repair cream group 1.4 (SD 3.3) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (4): baseline, weeks 1, 2 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page AB73): "This study was sponsored by Johnson & Johnson Consumer Companies, Inc." | |
| Declaration of interest | None declared, but 4 investigators were employees of Johnson & Johnson Consumer Companies, Inc, the manufacturer of the drug under investigation | |
| Notes | See Notes section of Characteristics of included studies of Draelos 2011 for details on EpiCeram. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page AB73): "randomly assigned" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "Eligible subjects were randomly assigned to one of the two investigational products in a 1:1 ratio. The randomization scheme was generated by the Sponsor using the random number generating procedure in SAS (Statistical Analysis System) package" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Eligible subjects were randomized in a sequential order. That is, the smallest random code in the randomization list was picked up and assigned to the next eligible subject. Thus, both subjects and investigator couldn't foresee the upcoming treatment assignment. J&J Clinical supply team completely overwrapped the products with plain white labels with study information printed on the white label. That would include randomization number, study site, PI phone number and instructions for use. We did not repackage the products from their original container" Comment: central allocation, de‐identified tubes. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Nothing reported regarding blinding. Comment: the outcome was likely to be influenced by the lack of blinding. After email communication: "double‐blind, products were overwrapped and randomization number and instructions for use were included on the overwrapped label" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Nothing reported regarding blinding. Outcomes were investigator‐assessed and participant‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. After email communication: "double‐blind, products were overwrapped and randomization number and instructions for use were included on the overwrapped label". Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | Little information provided. Comment: there was insufficient information to permit a clear judgement. After email communication: 7/49 (14.3%), unclear from which group. Intention‐to treat analysis Comment: we judged this as being at an unclear risk of bias. |
| Selective reporting (reporting bias) | Unclear risk | Comment: there was insufficient information to permit a clear judgement. After email communication: "The study was part of protocol (NCT01326910), there was an additional outcome provided in the protocol "Investigator's Global Atopic Dermatitis Assessment" Comment: as this was a poster presentation, and the sub‐study has not yet been published in full, we judged this as being at an unclear risk of bias. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled, within‐participant study Setting Department of Dermatology, Chungnam National University School of Medicine, Daejeon, Korea Date of study June to November 2011. Duration of intervention: 4 weeks | |
| Participants | N = 30 (20 female, 10 male) Mean age = 14.2 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 30 to either side of the body (within‐participant) Withdrawals/losses to follow‐up 2/30 (6.7%) lost to follow‐up Baseline data Mean IGA: treatment side 2.75 (SD 0.70), control side 2.71 (SD 0.66) Mean VAS: treatment side 6.53 (SD 1.77), control side 6.50 (SD 1.75) Mean TEWL (g/m²/h): treatment side 37.10 (SD 9.30), control side 34.30 (SD 8.40) Mean skin capacitance: treatment side 27.40 (SD 10.80), control side 28.90 (SD 10.00) | |
| Interventions | Intervention
Comparator
Continued use of antihistamines and topical steroids was permitted | |
| Outcomes | Assessments (4): baseline, weeks 1, 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 155): "This study was supported by a grant from the Korea Ministry of Health and Welfare (A091121)" | |
| Declaration of interest | None declared | |
| Notes | We mailed investigators numerous times for more precise study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 151): "was randomly assigned" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 150): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each part of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 2/30 (6.7%) lost to follow‐up. Per‐protocol analysis. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐ and vehicle‐controlled study Setting Two dermatological sites, Department of Specialistic and Experimental Clinical Medicine, Division of Dermatology, University of Bologna, Bologna, San Gallicano Dermatological Institute, Rome, Italy Date of study February 2005 to April 2006. Duration of intervention: 43 days | |
| Participants | N = 60 (30 female, 30 male) Mean age = 5.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 60 (MAS063DP (Atopiclair) group = 20, MAS060 (Atopiclair light) group = 20, vehicle group = 20) Withdrawals/losses to follow‐up 14/60 overall (23.3%): MAS063DP group = 1/20 (5%), MAS060 group = 7/20 (35%), vehicle group = 6/20 (30%)
Baseline data Number of participants with:
| |
| Interventions | Intervention 1
Comparator 1
Comparator 2
Participants had to observe a 7‐ or 14‐day washout period if appropriate, and had to refrain from using other medications for the treatment of atopic dermatitis during the study. The use of other topical or systemic medications for the treatment of atopic dermatitis, and phototherapy were not permitted during the washout and study period. If atopic dermatitis was poorly controlled to the degree that additional therapy/rescue regimen was needed, then either hydrocortisone butyrate 0.1% or desonide 0.5% cream were allowed twice daily for 1 week. The use for rescue medication was an end‐point of the study | |
| Outcomes | Assessments (7): baseline, days 1, 8, 15, 22, 29 and 43 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 625): "This study was supported by Sinclair Pharmaceuticals Ltd, Godalming Business Centre, Woolsack way, Godalming, GU7 1XW Surrey, UK" | |
| Declaration of interest | None declared | |
| Notes | MAS063DP = Atopiclair. See Notes section of Characteristics of included studies of Abramovits 2008 for details on Atopiclair. MAS060 = Atopiclair light, a water‐in‐oil formulation containing the same key ingredients of MAS063DP, but at lower concentration and with no preservatives Participants were also provided with a specific non medicated, fragrance‐free cleanser (Cetaphil; Galderma, Alby‐Sur‐Cheran, France) for the entire study duration | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 620): "The randomization list was generated by computer" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: sponsor provided numbered identical boxes containing the different products. Comment: central allocation, de‐identified boxes. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 620): "the tubes were labelled in a double‐blind manner" and "MAS063DP, MAS060 and vehicle were similar in consistency and in colour". Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 620): "Investigators, patients and caregivers remained blinded" and "MAS063DP, MAS060 and vehicle were similar in consistency and in colour". Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | 14/60 (23.3%): MAS063DP group = 1/20 (5%), MAS060 group = 7/20 (35%), vehicle group = 6/20 (30%). Intention‐to‐treat analysis based on 57/60. Comment: although 'modified' intention‐to‐treat analysis, the high number of losses to follow‐up posed an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, open, controlled study Setting Multicentre (3) in Italy and Romania Date of study Unspecified. Duration of intervention: 28 days | |
| Participants | N = 55 (gender not reported) Mean age = 2.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 55 (emollient balm group = 28, control group = 27) Withdrawals/losses to follow‐up 1/55 (1.8%), from control group Baseline data Mean SCORAD: emollient balm group 11.7 (SD 3.1), control group 10.2 (SD 3.3) Mean pruritus intensity: emollient balm group 1.67 (SD 1.20), control group 1.77 (SD 1.52) Mean TEWL (g/m²/h): emollient balm group 35.46 (SD 20.90), control group 21.55 (SD 9.43) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, days 15 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 170 of co‐publication Bianchi 2016, under primary reference): "The study was funded and conducted by Pierre Fabre Dermo‐Cosmetique" | |
| Declaration of interest | Quote (page 170 of co‐publication Bianchi 2016, under primary reference): "P. Bianchi, J. Theunis, C. Casas, C. Villeneuve, C. Phulpin, A. Bacquey, D. Redoules, V. Mengeaud, and A‐M Schmitt are employees of Pierre Fabre Dermo‐ Cosmetique. The principal investigator, Dr. Patrizi, signed a contract with Pierre Fabre Dermo‐Cosmetique for this study" | |
| Notes | The study product (Avene Xeracalm balm; Pierre Fabre Dermo Cosmetique, Boulogne, France) was an oil‐in‐water emulsion. The associated hygiene product (emollient cleansing gel; Trixera; Pierre Fabre Dermo, Boulogne, France) was indicated for dry and atopic skin and was used for bathing and then rinsed off once a day. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 170 of co‐publication Bianchi 2016): "were randomized to receive" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The randomization list was generated by the Pierre Fabre Biometrie Department. Our statistical department used a specific software to generate the randomization list" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "The randomization list was not given to the investigator. Only sponsor had access to this list and could see it" Comment: central allocation, allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | Quote (page AB62): "open, randomized" Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page AB62): "open, randomized" Outcomes were investigator‐assessed and participant‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 1/55 (1.8%), from control group. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Three randomised, double‐blind, vehicle‐controlled, within‐participant studies, only 1 of which was in atopic dermatitis Setting Department of Physiology, Institute of Biomedicine, University of Turku, Turku, Finland Date of study October 2008 to May 2009. Duration of intervention: 28 days | |
| Participants | N = 14 (9 female, 5 male) Mean age = 28 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 14 to either forearm (within‐participant) Withdrawals/losses to follow‐up 2/14 (14.3%): withdrew consent (1), and worsening eczema (1) Baseline data Mild atopic dermatitis: 10/13 Moderate atopic dermatitis: 3/13 | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 420): "The 3 studies were funded by BioCis Pharma Ltd" | |
| Declaration of interest | Quote (page 420): "The authors LP, LL, and JKL were employees of BioCis Pharma Ltd at the time the studies were conducted. LL and JKL are also shareholders and patent inventors for BioCis Pharma Ltd" | |
| Notes | 5% (w/w) cis‐UCA emulsion cream (BioCis Pharma, Turku, Finland) and the same vehicle emulsion cream base (Orion, Espoo, Finland). The cream base contained aqua, decyl oleate, cetearyl alcohol, glycerol, sodium cetearyl sulphate, and methyl paraben. Both products were pH 6.5. Data needed to be estimated from figures (TEWL), otherwise no precise data were provided other than the statement that "there were no significant differences." We mailed investigators numerous times for more precise study details, but received no response (Table 2; Table 4). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 415): Randomisation was performed by computer‐generated (SAS System, Cary, NC, USA) lists by a randomisation expert with no clinical involvement in the trials" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 415): "the investigators, and the study site personnel were blinded for the identity of the treatments" and "The products were packed in identical tubes labelled for either arm for each subject number" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 415) "The products were packed in identical tubes labelled for either arm for each subject number" Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Unclear risk | 2/14 (14.3%): withdrew consent (1), and worsening eczema (1). Per‐protocol analysis Comment: moderate number of dropouts at follow‐up combined with the per‐protocol analysis posed an unclear risk of bias for this domain. |
| Selective reporting (reporting bias) | Unclear risk | The protocol for the study was available (Eudra‐2008‐005075‐10). Not all the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. Data on skin swelling, formation of papules, formation of vesicles or bullae, and scaling were missing. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two‐phase randomised, double‐blind, active‐controlled study Setting Multicentre (20) in Germany, Italy and Spain Date of study August 2005 to January 2006. Duration of intervention: acute treatment phase of up to 4 weeks and a 16‐week maintenance phase | |
| Participants | N = 249 in acute treatment phase, and 221 in maintenance phase (142 female, 79 male) Mean age = 30.9 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 221 (moisturiser group = 109, MPA group = 112) Withdrawals/losses to follow‐up 8/221 overall (3.6%): moisturiser group = 3, MPA group = 5
Baseline data at start maintenance phase Mean EASI: moisturiser group 1.4, MPA group 1.6 Mean itching: moisturiser group 8.7, MPA group 10.1 | |
| Interventions | First phase open‐label methylprednisolone aceponate (MPA) cream once a day as well as open‐label moisturiser once a day for a maximum of 4 weeks Second, maintenance phase: Intervention
Comparator
| |
| Outcomes | Assessments (5): baseline, weeks 2, 6, 10 and 16 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 806): "This study was sponsored by Intendis GmbH, Berlin" | |
| Declaration of interest | None declared, but 1 of the authors was an employee at Intendis GmbH, Berlin, manufacturer of Advabase | |
| Notes | Advabase, Intendis GmbH, Berlin, Germany contains aqua, decyl oleate, glyceryl stearate, caprylic/capric/stearic triglyceride, glycerol, cetearyl alcohol, hydrogenated coco‐glycerides, benzyl alcohol, disodium EDTA, BHT We had email communication with the corresponding author, but he could not help us with missing trial details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 802): "Randomization at the end of the AP [acute treatment phase] was carried out in blocks according to the patients' arrival at the study centre and aimed to achieve a 1:1 randomization ratio overall and within each centre" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 801): "double‐blind" and "MP medication was packed in identical tubes to ensure blinding" In view of the difference in number of tubes (1 or 2) in the interventions groups it remains unclear if blinding was effective. Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 801): "double‐blind" and "MP medication was packed in identical tubes to ensure blinding". In view of the difference in number of tubes (1 or 2) in the interventions groups it remains unclear if blinding was effective. Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 8/221 (3.6%); moisturiser group = 3, MPA group = 5. Both intention‐to‐treat analysis and per‐protocol analysis. Comment: low number of losses to follow‐up, and combined with both intention‐to‐treat as well as per‐protocol analysis meant we considered this trial as being at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Department of Dermatology, University and Ospedale Maggiore IRCCS of Milan, Milan, and S Gallicano Dermatological Institute, Rome, Italy Unspecified. Duration of intervention: 1 month | |
| Participants | N = 70 (30 healthy adults: 8 female, 22 male, and 40 adults with atopic dermatitis: 10 female, 30 male) Age range = 27 to 45 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 70 unclear how many randomised to each treatment arm Withdrawals/losses to follow‐up Not reported Baseline data Not reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline, week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | Cream containing urea (Laceran, Beiersdorf AG) and Essex Base Cream containing white petrolatum, paraffin, cresol chloride and polyethylene glycol (Schering‐Plough) Unclear how many were randomised to each treatment arm, no separate data for healthy and atopic adults (Table 4) As the study was 20 years old we did not contact the investigators for data. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 171): "The patients and controls were assigned randomly to one or the other treatment" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 171): "Patients, normal controls and the physicians (AB and CC) who examined them were blind to the treatments" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 171): "Patients, normal controls and the physicians (AB and CC) who examined them were blind to the treatments" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Unclear risk | Nothing reported, little information provided. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Unclear risk | Comment: there was insufficient information to permit a clear judgement. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Klinik für Dermatologie und Venereologie der Universität Lubeck, Germany Date of study Unspecified. Duration of intervention: 2 weeks | |
| Participants | N = 54 (20 healthy participants, 34 participants with atopic dermatitis, gender not reported) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 54 unclear how many to each treatment arm Withdrawals/losses to follow‐up 5/54 (9.3%) reasons unreported Baseline data Nothing reported | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (3): baseline, weeks 1 and 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, but Dr Puschmann was employed by HERMAL, the manufacturer of the moisturiser under investigation, therefore the trial was likely to have been funded by HERMAL Kurt Herrmann GmbH & Co, Reinbek | |
| Declaration of interest | None declared, but Dr Puschmann was employed by HERMAL Kurt Herrmann GmbH & Co, Reinbek, the manufacturer of the moisturiser under investigation | |
| Notes | Commercial cream contains polidocanol 3% and urea 5%, water, octyl dodecanol, polymethylphenylsiloxane, stearin palm, dimethicone, glycerol, paraffin, hexadecyl palmitate, polysorbate 40, carbomer, benzyl alcohol, and trometamol. The products differ chiefly in the number of carbomers the molecules contain. Unclear how many were randomised to each treatment arm, no separate data for healthy participants and atopic participants (Table 4) As the study was 13 years old we did not contact the investigators for data. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 78): "einer doppelblinden, randomisierten" (translation: double‐blind, randomised) Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 78): "doppelblinden" (translation: double‐blind) Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 78): "doppelblinden" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Unclear risk | 5/54 dropouts (9.3%): reasons unreported, unclear from which treatment arm(s) they came, little information provided. Per‐protocol analysis. Comment: there was insufficient information to permit a clear judgement. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, participant‐blinded, controlled, within‐participant study Setting Multicentre, Sacramento, California, USA Date of study January 2014 to May 2015. Duration of the intervention 10 minutes | |
| Participants | N = 20 of which 10 had atopic dermatitis and 10 were healthy subjects (9 female, 11 male) Mean age = 26.3 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 20 to either forearm (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data of 10 participants with atopic dermatitis Severity of eczema: mild (n = 5), moderate (n = 3), severe (n = 2) Mean TEWL (g/m²/h): bleach side 12.58 (SD 8.74), water side 11.89 (SD 8.64) Mean corneometry units: bleach side 23.66 (SD 11.38), water side 24.47 (SD 10.44) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (4): baseline, 15 minutes, 30 and 60 minutes Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page AB76): "Commercial support: None identified" | |
| Declaration of interest | None declared | |
| Notes | Poster abstract. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page AB76): "randomized controlled study" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "The allocation sequence was generated a priori with binary randomization generator." Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication (and in Shi 2016): "The randomization was performed prior to recruitment by the study coordinator, and stored in sealed envelopes that were not opened until the subject was recruited by the investigators." Comment: allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | According to protocol (NCT02594969 available at clinicaltrials.gov): "participant‐blinded" and "Participants remained blinded to the water or dilute hypochlorite immersion". Investigators were not blinded. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Outcomes were investigator‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up. |
| Selective reporting (reporting bias) | Low risk | The protocol was available (NCT02594969 available at clinicaltrials.gov), and the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐ and vehicle‐controlled, within‐participant study Setting Multicentre (25) in Japan Date of study Unspecified. Duration of intervention: 4 weeks with 2 months of follow‐up | |
| Participants | N = 552 (gender not reported) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 552 to either side of the body (within‐participant) Withdrawals/losses to follow‐up 61/552 (11%), unclear from which groups, due to:
Baseline data of per‐protocol population Ichtyosis vulgaris (137), senile xerosis (170), atopic skin (184) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (5): baseline, weeks 1, 2, 3 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | The data were confusingly reported in this study and did not lend themselves to further analysis (Table 4). As the study was 39 years old we did not contact the investigators for data. | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 104): "doubleblind randomised control trial" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 104): "doubleblind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 104): "doubleblind" Outcomes were investigator‐assessed as well as participant‐assessed. |
| Incomplete outcome data (attrition bias) | Unclear risk | 61/552 (11%), unclear from which groups. Per‐protocol analysis. Comment: we judged this to be at an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, controlled, within‐participant study (study D) Setting Multi‐centre (3) in USA Date of study | |
| Participants | N = 127 (gender not reported) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 127 to either side of the body (within‐participant) Withdrawals/losses to follow‐up 4/127 (3%), reasons unreported Baseline data | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (5): days 0, 7, 14, 21 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 748): "The studies were supported by Galderma R & D" | |
| Declaration of interest | Quote (page 748): "Dr Simpson is a consultant for Galderma. Dr Dutronc is an employee of Galderma" | |
| Notes | This reference includes data on 4 studies, of which only study D matched our inclusion criteria Cetaphil RestoraDerm contains aqua, glycerol, caprylic/capric triglyceride, Helianthus annus (sunflower) seed oil, pentylene glycol, Vitellaria paradoxa (formerly called Butyrospermum parkii; shea butter), sorbitol, cyclopentasiloxane, cetearyl alcohol, behenyl alcohol, glyceryl stearate, tocopheryl acetate, hydroxypalmitoyl sphinganine, niacinamide, allantoin, panthenol, arginine, disodium ethylene dicocamide PEG‐15 disulfate, glyceryl stearate citrate, sodium PCA, ceteareth‐20, sodium polyacrylate, caprylyl glycol, citric acid, dimethiconol, disodium EDTA, sodium hyaluronate, cetyl alcohol We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 745): "...randomized..." Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "This was a left/right comparison study, comparing treatment with Cetaphil cream on one side of the body versus no treatment on the contralateral side. The allocation of treatments (Cetaphil cream or no treatment) was randomized and the allocation sequence was generated by an independent statistician" Comment: probably done |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "The study medications were dispensed by someone other than the investigator designated as Study Drug Dispenser. The randomization list was secured in a locked cabinet and in an electronic file with restricted access to only the designated personnel directly responsible for labelling and handling the study medications, until the study database was locked and ready to be unblinded." Comment: reasonable to assume allocation adequately concealed. We judged this as at a low risk of bias. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 745): "evaluator‐blinded" Comment: participants were not blinded. The report provided insufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 745): "evaluator‐blinded" Comment: uncertainty about the effectiveness of blinding of outcomes assessors (healthcare providers) during the study, and participants were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 4/127 (3%), reasons unreported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, controlled (inactive), within‐participant study Setting Pro‐DERM Institute in Germany Date of study December 2010 to February 2011. Duration of intervention: 27 days | |
| Participants | N = 20 (16 female, 4 male) Mean age = 40.9 years Inclusion criteria of the trial
Exclusion criteria of the trial1
Randomised N = 20 to either leg (within‐participant) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean dryness: CRM‐treated area 2.05 (SD 0.63), untreated area 2.07 (SD 0.63) Mean skin hydration (corneometry units): CRM‐treated area 17.77 (SD 5.24), untreated area 18.25 (SD 5.91) Mean TEWL (g/m²/h): CRM‐treated area 5.15 (SD 1.53), untreated area 15.32 (SD 1.73) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and day 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared but the manufacturer of the drug under investigation is Galderma, R&D, Sophia Antipolis, France | |
| Declaration of interest | Quote (page 125): "Dr. Simpson is a consultant for Galderma, and Dr. Böhling and Mr. Bielfeldt received investigator fees for this research study. Ms. Bosc and Mr. Kerrouche are employees of Galderma" | |
| Notes | Cetaphil Restoraderm Body Moisturiser (CRM – Galderma S.A.) contains filaggrin breakdown products (components of NMF), ceramide precursor, and niacinamide, fatty acids, humectants, filmogenic substances, emollients, and shea butter. CRM contains no fragrances and has a pH of approximately 5.5 We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 122): "statistician generated a randomization list" and "randomly assigned to the right or left lower leg (for each subject, one 4x4 cm area treated with CRM and one symmetric untreated control area) with a block size of four subjects" Comment: probably done |
| Allocation concealment (selection bias) | Low risk | Quote (page 122): "a statistician generated a randomization list and each bottle of product was identified by a randomization number and was labeled with the side to be treated...The randomization list was kept under restricted access until the study database was locked and ready to be unblinded for statistical analyses" Comment: central allocation, de‐identified bottles. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 122): "investigator‐blinded"; participants were not blinded. Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received on each leg, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 122): "investigator‐blinded"; participants were not blinded. Outcomes were (mainly) investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled study Setting Multicentre (5) in USA Date of study December 2006 to February 2007. Duration of intervention: 4 weeks | |
| Participants | N = 121 (73 female, 48 male) Mean age = 7.1 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 121 (barrier repair group = 59, fluticasone group = 62) Withdrawals/losses to follow‐up 12/121 overall (9.9%); barrier repair group = 6/59, fluticasone group 3/62
Baseline data Mean SCORAD: barrier repair group 37.2, fluticasone group 33.8 Mean self‐assessment: barrier repair group 0.1 (SD 0.40), fluticasone group 0.11 (SD 0.46) | |
| Interventions | Intervention
Comparator
Cetaphil lotion was applied to clinically uninvolved areas twice daily | |
| Outcomes | Assessments (3): baseline, days 14 and 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1110): "company‐sponsored study (Ceragenix Corporation, Dever, CO) | |
| Declaration of interest | Quote (page 1110): "Neither of the authors is a consultant or shareholder at Ceragenix" | |
| Notes | See Notes section of Characteristics of included studies of Draelos 2011 for details on EpiCeram. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1107): "were assigned randomly at entry" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "A central randomization schedule was produced providing random allocation to the two treatment groups. A block size of 4 will be used in the generation of the randomization code" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "central allocation by sponsor" Comment: allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 1106): "investigator‐blinded". Participants were not blinded, quote (page 1107) "the subjects could not be blinded" Comment: the report did not provide sufficient detail about the specific measures used to blind study personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "The person dispensing the study medication and performing ongoing study medication accountability will NOT perform any of the efficacy assessments required by the protocol" Comment: the report provided sufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received, however, participants were not blinded. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 1106): "investigator‐blinded" Outcomes were investigator‐assessed as well as participant‐assessed. Comment: blinding of the outcomes assessors, was ensured but participants and parents were not blinded. The outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 12/121 (9.9%), reasons reported. Intention‐to‐treat analysis. Comment: low number of losses to follow‐up, combined with intention‐to‐treat analysis, meant we considered this domain to be at a low risk of bias. |
| Selective reporting (reporting bias) | Unclear risk | The protocol for the study was available (NCT00616538) and the prespecified outcome "Patient/family self assessments of improvement" and "IGA" mentioned in the methods section was inadequately reported ("demonstrated highly significant improvement") Comment: we judged this as being at an unclear risk of bias. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Two‐phase, randomised, open‐label, active‐controlled study Setting Multicentre (8) in Japan Date of study Unspecified. Induction phase 1‐4 weeks, maintenance phase > 4 weeks | |
| Participants | N = 70 (37 female, 33 male) Mean age = 31 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 70 (induction phase: moisturiser group = 35, tacrolimus group = 35; maintenance phase moisturiser group = 21, tacrolimus group = 23) Withdrawals/losses to follow‐up Induction phase 2/70 (2.9%): 1 in each group and in 24 the VAS‐itch score did not reduce by more than 20 and did not make it to maintenance phase, 1 refused to continue Maintenance phase 1/43 (2.3%): in moisturiser group Baseline data Mean itch (VAS): moisturiser group 19.3 (SD 16.7), tacrolimus group 28.1 (SD 15.4) | |
| Interventions | Phase 1: induction phase, all participants received topical tacrolimus (0.03% < 16 years old and 0.1% otherwise). Moisturisers twice daily in addition to usual topical corticosteroid treatment (maximum use, 10 g/week). Participants who showed a reduced VAS‐itch score by > 20 points were considered to show relief from pruritus, and only these induction therapy responders proceeded into maintenance treatment Phase 2: maintenance phase Intervention
Comparator
| |
| Outcomes | Assessments (2): baseline and day 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 150): "This work was supported by research grants from the Ministry of Health, Labour and Welfare, Japan" | |
| Declaration of interest | None declared | |
| Notes | We only include data on phase 2, the maintenance phase of the study We mailed investigators numerous times for more precise study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 146): "random allocated" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | Quote (page 144): "open‐label" Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 144): "open‐label" Outcomes were investigator‐assessed as well as participant assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Unclear risk | We only include data on maintenance phase, 1/43 dropped out in the moisturiser group. Without further information regarding the maintenance phase (the number initially randomised, do not match the number in the maintenance phase). |
| Selective reporting (reporting bias) | High risk | Atopic dermatitis severity score (SCORAD), was a predefined outcome in the Methods section, but was not reported for the maintenance phase. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled study Setting National Skin Centre, Singapore Date of study Unspecified. Duration of intervention: 41 days | |
| Participants | N = 60 (21 female, 39 male) Mean age = 18 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 60 (study cream group = 30, vehicle group = 30) Withdrawals/losses to follow‐up No losses to follow‐up Baseline data Mild atopic dermatitis (number of participants): study cream group (7), vehicle group (7) Moderate atopic dermatitis (number of participants): study cream group (23), vehicle group (23) Mean SCORAD: study cream group 28.85, vehicle group 29.51 | |
| Interventions | For ethical reasons and recruitment effectiveness, all participants were provided with 0.025% betamethasone valerate cream, for the first 27 days. Following this, participants who still had persistent eczema were allowed to continue use of the steroid if necessary, which most did. Intervention
Comparator
Participants were provided with emulsifying ointment as cleansers throughout the study period and were instructed not to use any systemic or topical antibiotics, antibacterial soap or antibacterial shampoo until the end of study | |
| Outcomes | Assessments (4): baseline, days 14, 27 and 41 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page e109): "This study was sponsored by Hygieia Healthcare Ltd" | |
| Declaration of interest | None declared | |
| Notes | All participants underwent a washout period of 1 week during which no topical corticosteroid was allowed. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page e110): "Patients were assigned to one of two treatment groups according to a computer‐generated randomization list, stratified according to disease severity" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Pharmaceutical company prepared identical labelled bottles" Comment: central allocation, de‐identified bottles. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page e109 and e110): "double‐blind" and "All study personnel and physicians remained blinded to allocation of treatment until completion of data analysis. Both the study cream and vehicle were odourless, of the same appearance and consistency, and provided in identical bottles" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page e109 and e110): "double‐blind" and "All study personnel and physicians remained blinded to allocation of treatment until completion of data analysis. Both the study cream and vehicle were odourless, of the same appearance and consistency, and provided in identical bottles" Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐ and placebo‐controlled study Setting Klinikum Mannheim gGmbH, Universitätsklinikum, Fakultät für Klinische Medizin Mannheim der Universität Heidelberg, Germany Date of study Unspecified. Duration of intervention: 4 weeks | |
| Participants | N = 58 (40 female, 18 male) Mean age = 31.1 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 58 (sea buckthorn oil (Hippophae rhamnoides oil) 10 % group = 17, sea buckthorn oil 20% group = 17, placebo group = 19, 5 unclear) Withdrawals/losses to follow‐up 5/58 (8.6%), reasons unreported and unclear from which groups Baseline data Nothing reported | |
| Interventions | Intervention
Comparator 1
Comparator 2
Application frequency unclear | |
| Outcomes | Assessments (5): baseline, weeks 1, 2, 3 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared, but 1 of the authors was an employee of Weleda AG, Schwabisch Gmund, the manufacturer of the product under investigation | |
| Notes | As the study was 16 years old we have not contacted the investigators for data | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 285): "randomized" and on page 287 "nach dem Zufallsprinzip" (translation: at random) Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 285): "double blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 5/58 (8.6%) lost to follow‐up, reasons unreported and unclear from which groups. Per‐protocol analysis Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this domain to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre Rome, Italy Date of study Unspecified. Duration of intervention: 2 weeks | |
| Participants | N = 117 (46 female, 71 male) Mean age = 4.3 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 117 (furfuryl palmitate enriched moisturiser group = 57, moisturiser group = 60) Withdrawals/losses to follow‐up 29/117 overall (24.8%); furfuryl palmitate enriched moisturiser group 18/57, moisturiser group 11/60
Baseline data Mean SCORAD: furfuryl palmitate enriched moisturiser group 28.1 (SD 10.6), moisturiser group 25.6 (SD 10.1) | |
| Interventions | Intervention
Comparator
Treatment with systemic or topical corticosteroids, topical immunomodulators, and topical or oral antihistamines was not allowed. | |
| Outcomes | Assessments (3): baseline, weeks 1 and 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, however, quote (page 207): "Both products were provided by the same manufacturer (ICIM International Srl, Milan, Italy)" | |
| Declaration of interest | None declared | |
| Notes | Moisturiser containing various antioxidant molecules (superoxide dismutase, 18‐ß‐glycyrrhetinic acid, vitamin E, and α‐bisabolol). Furfuryl palmitate enriched formulation was same moisturiser with addition of furfuryl palmitate We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 206): "Randomization was performed using a computer‐generated list" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "Each tube had a letter "A" or "B", and we assigned the cream preparation to the patients on the basis of computer randomization list made directly by the Producer and sent to the each center before starting the trial. Obviously the investigators did not know if "A" or "B" was the active or not" Comment: central allocation, de‐identified tubes; allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 207): "The study was double blind, with the key code disclosed only after statistical evaluations. The 2 topical products had an identical color, smell, and consistency and were indistinguishable to both parents and examining pediatricians" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 207): "The study was double blind, with the key code disclosed only after statistical evaluations. The 2 topical products had an identical color, smell, and consistency and were indistinguishable to both parents and examining pediatricians" Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | High risk | 29/117 (24.8%) lost to follow‐up; furfuryl palmitate‐enriched moisturiser group 18/57, moisturiser group 11/60, reasons reported. Per‐protocol analysis. Comment: the total number of dropouts, combined with a per‐protocol analysis represents a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, investigator‐blinded, active‐controlled, within‐participant study Setting Skin Center, Faculty of Medicine, Srinakharinwirot University, Bangkok, Thailand Date of study Unspecified. Study duration 6 weeks | |
| Participants | N = 30 (12 female, 14 male and 4 gender unknown) Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 30 to either body side (within‐participant) Withdrawals/losses to follow‐up 4/30 (13%): bacterial folliculitis (1) incomplete follow‐up (3) Baseline data Mean SCORAD: licochalcone A side 28.28, hydrocortisone acetate 1% side 28.52 | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (6): efficacy for the first 4 weeks; 2 weeks of relapse‐rate Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared. However, the study design is identical ‐ with a similar comparison with licochalcone ‐ to Angelova‐Fischer 2014 and Wanakul 2013, which were sponsored by Beiersdorf. | |
| Declaration of interest | Quote (page 660): "No conflict of interest" | |
| Notes | Licochalcone in ceramide and linoleic acid lipid base formulation (Eucerin Soothing lotion 12% omega, Beiersdorf, Germany). Licochalcone A is an extract from Glycyrrhiza inflata (Chinese liquorice) that has anti‐inflammatory property We received several responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 661): "Treatment was randomly assigned by a third party" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | Quote (page 661): "The trial codes were broken only after data were analyzed" The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 660‐661): "investigator‐blinded and "The tubes were similar in shape and colour" Comment: the report provided sufficient detail about the measures used to blind study personnel from knowledge of which intervention a participant received on each side of the body, however, participants were not blinded. |
| Blinding of outcome assessment (detection bias) | High risk | Quote (page 660): "investigator‐blinded" Outcomes were investigator and participant‐assessed. Comment: we judged this domain to be at a high risk of bias. |
| Incomplete outcome data (attrition bias) | Unclear risk | 4/30 (13%) lost to follow‐up; bacterial folliculitis (1) incomplete follow‐up (3) low number, and per‐protocol analysis. Comment: moderate number of losses to follow‐up, combined with per‐protocol analysis, so we considered domain to be at an unclear risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Two outpatient dermatology clinics Makati City, Philippines Date of study Unspecified. Study duration 4 weeks | |
| Participants | N = 52 (25 female, 27 male) Mean age = 32 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 52 (virgin coconut oil group = 26, virgin olive oil group = 26) Withdrawals/losses to follow‐up No dropouts or protocol violators Baseline data Mean objective SCORAD: virgin coconut oil group 39.2 (SD 6.4), virgin olive oil group 36.6 (SD 6.3) Colonisation with Stapylococcus aureus: virgin coconut oil group 20/26, virgin olive oil group 12/26 | |
| Interventions | Intervention
Comparator
No other moisturisers, creams, or oil‐based products were allowed | |
| Outcomes | Assessments (2): baseline and week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | ||
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 309): "the randomization key, and the codes were carried out by the pharmacist of Skin Sciences Laboratory, Inc., and was disclosed to the investigators only at the end of the study" and "underwent simple concealed random allocation (by drawing rolled pieces of paper labeled "A" or "B") to control or treatment arms by the two dermatology residents, both of whom were blind to the codes" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 309): "the randomization key, and the codes were carried out by the pharmacist of Skin Sciences Laboratory, Inc., and was disclosed to the investigators only at the end of the study" and "repackaged in uniform medicinal opaque plastic bottles with a small opening to mask the color and scent of both oils" Comment: central allocation, de‐identified bottles; allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 208‐209): "double‐blind" and "repackaged in uniform medicinal opaque plastic bottles with a small opening to mask the color and scent of both oils" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement of low risk. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 208‐209): "double‐blind" and "repackaged in uniform medicinal opaque plastic bottles with a small opening to mask the color and scent of both oils" Outcomes were investigator‐assessed. |
| Incomplete outcome data (attrition bias) | Low risk | Quote (page 311 ): "There were no dropouts or protocol violators" |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the outcomes mentioned in the Methods section appeared to have been reported. |
| Other bias | Unclear risk | There was baseline imbalance regarding colonisation with S aureus in favour of the virgin oil group (77% versus 46%). |
| Study characteristics | ||
| Methods | Randomised, double‐blind, active‐controlled, within‐participant study Setting Multicentre (4) in Bangkok, Thailand Date of study March 2010 to December 2011. Duration of intervention: 4 weeks followed by 4 weeks open label licochalcone A | |
| Participants | N = 55 (28 female, 27 male) Mean age = 3.1 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 55 to either side of the body (within‐participant) Withdrawals/losses to follow‐up 3/55 (5.5%), reasons unreported Baseline data Mean SCORAD: licochalcone side 26 (SD 8.8), hydrocortisone 1% side 26 (SD 9.8) Mean TEWL (g/m²/h): licochalcone side 29 (SD 12.1), hydrocortisone 1% side 27 (SD 14.0) | |
| Interventions | Intervention
Comparator
After 4 weeks both sides were treated with licochalcone A lotion. We only include the first 4 randomised weeks. | |
| Outcomes | Assessments (3): baseline, weeks 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 1140): "The present study was supported by Beiersdorf (Thailand) Co. Ltd. The sponsor had no influence on study design, data collection, and data analyses" | |
| Declaration of interest | None declared | |
| Notes | Licochalcone in ceramide and linoleic acid lipid base formulation (Eucerin Soothing lotion 12% omega, Beiersdorf, Germany). Licochalcone A is an extract of Glycyrrhiza inflata (Chinese liquorice) that has anti‐inflammatory property | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1136): "The assignment of applying the test substance was performed by block randomization by the principal investigator to all four sites before the present study was started" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | Quote (page 1136): "These code numbers were assigned sequentially so that the patients were enrolled in each site" and "supplied in similar containers" Comment: although this suggests central allocation with delivering de‐identified containers to the 4 study sites, the principal investigator generated the randomisation sequence and it remains unclear whether the allocation sequence was adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 1136): "During the present study, investigators and patients were blinded regarding the use of either moisturizer containing Lic A lotion or 1% hydrocortisone lotion, which were supplied in similar containers" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement of low risk. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 1136): "During the present study, investigators and patients were blinded regarding the use of either moisturizer containing Lic A lotion or 1% hydrocortisone lotion, which were supplied in similar containers" Outcomes were investigator‐assessed as well as participant‐assessed. Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 3/55 (5.5%) lost to follow‐up, reasons unreported. Per‐protocol analysis Comment: low number of losses to follow‐up, and although per‐protocol analysis, we considered this domain to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the outcomes mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, controlled (inactive), study Setting Single dermatology centre, Colorado Springs, CO, USA Date of study Unspecified. Duration of intervention: 6 months (study consisted of 3 phases, we only included the second, i.e. the maintenance phase) | |
| Participants | N = 45 (21 female, 24 male) Mean age = 3.5 years (7 months to 11 years 5 months) Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 45 (moisturiser group = 21, control group = 24) Withdrawals/losses to follow‐up 2/45 overall (4.4%); moisturiser group 1/21, control group 1/24
Baseline data "The eczema history questionnaire completed at baseline showed that 78% of subjects had flared 3 to 4 times in the previous 12 months, while 18% had flared 1 to 2 times during this time period" | |
| Interventions | Washout phase of 2 weeks, then maintenance phase Intervention
Comparator
In case of acute therapy for active atopic dermatitis lesions Eucerin Eczema Relief Instant Therapy was provided and participants entered the Treatment phase for 4 weeks | |
| Outcomes | Assessments (4): baseline, weeks 2, 6 and until flare Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared but 3 of the authors were employees of Beiersdorf, therefore it is most likely that the study was funded by Beiersdorf | |
| Declaration of interest | Quote (page 485): "Teresa M. Weber PhD is an employee of Beiersdorf Inc, the manufacturer of Eucerin Eczema Relief Body Creme and Eucerin Eczema Relief Flare Treatment. Alexander Filbry PhD and Frank Rippke MD are employees of BeiersdorfAG. Frank Samaria MD and Michael J. Babcock MD have no conflicts of interest to declare" | |
| Notes | Eucerin Eczema Relief Body Creme, Seiersdorf, Wilton, CT contains 1% colloidal oatmeal, licochalcone A (anti‐inflammatory property), ceramide 3, an epidermal barrier Iipid, aqua, glycerol, castor‐oil‐plant (Ricinus communis) seed oil, mineral oil, cetyl alcohol, glyceryl sterate, caprylic/capric triglyceride, octyldodecanol, cetyl palmitate, PEG‐40 stearate, Chinese liquorice (Glycyrrhiza inflata) root extract, 2‐hexanediol, phenoxyethanol, piroctone olamine, caprylyl glycol, ethylhexylglycerin, benzyl alcohol, citric acid We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 479): "randomly assigned" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "In order to enroll subjects across the inclusion age range and to assure approximately equal age distribution in the two treatment cells (moisturizer vs. no moisturizer), subjects were randomized into 3 age groups, 3 months up to 12 months, >12 months to 24 months, and >24 months to 12 years. Therefore the study coordinator created 3 randomization lists (1 per age group) with subject number codes" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "treatment cell allocation and product distribution was handled separately, in a different part of the clinical site from the examination rooms, away from the investigator. The investigator was not involved in cell allocation, nor dispensing the treatment supplies" Comment: form of central allocation. Allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | 2/45 (4.4%): moisturiser group 1/21, control group 1/24. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the outcomes mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, vehicle‐controlled, within‐participant study Setting ProDERM Institut für Angewandte Dermatologische Forschung, Schenefeld, Germany Date of study Unspecified. Duration of the intervention 4 weeks | |
| Participants | N = 80 (58 female, 22 male) Mean age not reported Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 80 to either body side (within‐participant) Withdrawals/losses to follow‐up 3/80 (3.8%), reasons unreported Baseline data 65% experience acute episode of atopic dermatitis within last year, 72.5% suffered from moderate to severe itch Mean sum score of erythema, dryness, induration/papules: urea cream side 4.05, vehicle side 4.11 Mean corneometry: urea cream side 48.6 (SD 9.1), vehicle side 48.2 (SD 10.0) | |
| Interventions | Intervention
Comparator
During the study no systemic treatments with corticosteroids, antihistamines, NSAIDs, cytostatic drugs or immunotherapy, nor topical treatments were permitted | |
| Outcomes | Assessments (4): baseline, weeks 1, 2 and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared, but 1 of the investigators was an employee of Beiersdorf, Hamburg, Germany, the manufacturer of the drug under investigation | |
| Notes | Vehicle contains: benzyl alcohol, paraffin, sodium lactate, magnesium stearate, ceresin, glyceryl diisostearate, isopropyl palmitate, wool alcohols, magnesium sulphate, purified water | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Unclear risk | Quote (page 26): "randomised" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | Unclear risk | Quote (page 165): "double‐blind" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received on each side of the body, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Unclear risk | Quote (page 165): "double‐blind" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement. |
| Incomplete outcome data (attrition bias) | Low risk | 3/80 (3.4%), reasons unreported. Per‐protocol analysis. Comment: low number of losses to follow‐up, and although per‐protocol analysis considered as at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the prespecified outcomes and those mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | Comment: the study appeared to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, controlled (inactive) study Setting Multicentre (5), dermatology outpatient clinics in Sweden Date of study Unspecified. Duration of intervention: 3 weeks for treatment phase, and 6 months maintenance phase. We only include the data from the maintenance phase | |
| Participants | N = 44 (32 female, 12 male) Mean age = 33.5 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised for maintenance phase N = 44 (moisturiser group = 22, no moisturiser = 22) Withdrawals/losses to follow‐up No losses to follow‐up reported Baseline data Mean TEWL (g/m²/h): moisturiser group 11.9, no moisturiser 14.1 | |
| Interventions | Treatment phase: 3 weeks betamethasone valerate 0.01% cream (Betnoderm or Betnovat) (55) Maintenance phase: Intervention
Comparator
Concomitant treatment of other body areas with other topical preparations (e.g. cosmetics and corticosteroids) was allowed throughout the study | |
| Outcomes | Assessments (2): baseline and week 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared, but the drug under investigation is manufactured by ACO HUD NORDIC AB, Upplands Väsby, Sweden | |
| Declaration of interest | Quote (page 1267): "None declared", but the first author was an employee of ACO HUD NORDIC AB, Upplands Väsby, Sweden, the manufacturer of the drug under investigation | |
| Notes | Betnoderm ACO HUD AB, Upplands Väsby, Sweden or Betnovat GlaxoSmithKline, Solna, Sweden Canoderm cream 5%, ACO HUD NORDIC AB, Upplands Väsby, Sweden; oil‐in‐water emulsion containing fractionated coconut oil, emulsifying wax, hydrogenated canola oil, propylene glycol, carbomer, dimethicone, hard paraffin, glycerol polymetacrylate, propyl‐ and methyl parahydroxybenzoate, sodium lactate solution, lactic acid, glyceryl stearate, polyoxyethylene stearate, purified water The co‐publication, Hjalte 2010, under the primary reference, contains economic outcomes We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 1268): "randomized" Comment: insufficient detail was reported about the method used to generate the allocation sequence to allow a clear assessment of whether it would produce comparable groups. After email communication: "Randomisation was performed by the Sponsor and there was one randomisation list for each part of the study. The randomization was computer aided (Excel). The investigator kept the randomisation lists in a sealed envelope only to be opened in case of a SAE. When patients were included in the first part of the study, patients received a chronological identification number and received one of the interventional drugs labelled with the chronological identification number. If patients were eligible for the part two of the study, then they were to receive either Canoderm or no treatment, also according to the predetermined randomization list. In this part of the study, blocked randomization (balanced within blocks of 20) were used to ensure equal number of Canoderm treatment and no treatment" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. After email communication: "The investigator kept the randomisation lists in a sealed envelope only to be opened in case of a SAE..." (see above) and "The Sponsor dispensed the intervention labelled with appropriate identification number" Comment: central allocation; allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | Low risk | No losses to follow‐up reported. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the outcomes mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, active‐controlled study Setting Department of Dermatology, Guangxi Minzu Hospital, Nanning, China Date of study February 2012 to July 2013. Duration of intervention: 3 weeks with 2 weeks of follow‐up | |
| Participants | N = 125 (62 female, 63 male) Mean age = 47 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 125 (pilot group = 63, control group = 62) Withdrawals/losses to follow‐up 20/125 overall (16%): pilot group 3/63, control group 17/62; reasons unreported Baseline data Mean disease duration: pilot group 18.5 months (SD 35.7), control group 20.3 months (SD 34.6) Mean EASI: pilot group 18.9 (SD 0.86), control group 17.94 (SD 0.71) | |
| Interventions | Intervention
Comparator
| |
| Outcomes | Assessments (6): baseline, weeks 1, 2, 3, 5 and 8 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | None declared | |
| Declaration of interest | None declared | |
| Notes | We mailed investigators numerous times to clarify study details, but received no response (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Translated "Allocated by using a random number table" Comment: probably done. |
| Allocation concealment (selection bias) | Unclear risk | The method used to conceal the allocation sequence, that is to determine whether intervention allocations could have been foreseen in advance of, or during, enrolment, was not reported. Comment: there was insufficient information to permit a clear judgement. |
| Blinding of participants and personnel (performance bias) | High risk | No reporting of measures used to blind study participants and personnel from knowledge of which intervention a participant received. Comment: the outcome was likely to be influenced by the lack of blinding. |
| Blinding of outcome assessment (detection bias) | High risk | No blinding reported. Outcomes were investigator‐assessed as well as participant‐assessed. Comment: the outcome measurement was likely to be influenced by the lack of blinding. |
| Incomplete outcome data (attrition bias) | High risk | 20/125 (16%): pilot group 3/63, control group 17/62, reasons unreported. Per‐protocol analysis. Comment: moderate dropout rate, but very unbalanced, which, combined with per‐protocol analyses, poses a high risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was not available, but the outcomes mentioned in the Methods section appeared to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
| Study characteristics | ||
| Methods | Randomised, double‐blind, 'other moisturiser'‐controlled study Setting Multicentre (15) in Finland, Noway and Sweden Date of study September 2011 to September 2012. After the stabilisation phase, participants entered the randomised maintenance phase of 180 days. We only include the data from the maintenance phase | |
| Participants | N = 172 (101 female, 71 male) Mean age = 28 years Inclusion criteria of the trial
Exclusion criteria of the trial
Randomised N = 172 (barrier‐strengthening moisturiser = 87, reference cream = 85) Withdrawals/losses to follow‐up 7/172 overall (4.1%): barrier‐strengthening moisturiser 3/87 (3.4%), reference cream 4/85 (4.7%)
Baseline data stabilisation phase Mean SCORAD: barrier‐strengthening moisturiser 35.62 (SD 12.24), reference cream 37.68 (SD 13.56) Baseline data maintenance phase: Mean SCORAD: barrier‐strengthening moisturiser 6.14 (SD 6.38), reference cream 5.97 (SD 5.84) | |
| Interventions | Stabilisation phase: once‐daily topical mometasone furoate cream 0.1% (Elocon, Merck Sharp & Dohme B.V, Netherlands) on the trunk and extremities and/or hydrocortisone acetate cream 1% (Hyderm ACO Hud Nordic, Upplands Väsby, Sweden) on the face, groin and armpits. In addition, participants used a medicinal moisturiser containing 20% glycerol (Miniderm, ACO Hud Nordic, Upplands Väsby, Sweden) Maintenance phase Intervention
Comparator
| |
| Outcomes | Assessments (4): baseline, end of stabilisation phase (visit 2), relapse visit (visit 3), day 180 (visit 4) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review | |
| Funding source | Quote (page 591): "This study was sponsored by ACO Hud Nordic AB, an Omega Pharma Company and partly funded by the Knowledge Foundation, Sweden. The sponsor was responsible for the study design, co‐ordination and collection of data provided by the investigators. The authors were responsible for data analysis, manuscript preparation and the decision to submit the manuscript for publication" | |
| Declaration of interest | Quote (page 591):"U.Å., K.W. and P.S. are employees of ACO Hud Nordic AB. M.G. was an employee at ACO Hud Nordic AB at the time when the study was conducted. Å.S. participated as principal investigator, but did not receive any personal compensation. S.R. has acted as an expert and/or given lectures for ACO Hud Nordic AB, Dignity Sciences, and Astellas Pharma Europe. M.B. has acted as an expert and/or given lectures for ACO Hud Nordic AB. M.L. has acted as a paid consultant to ACO Hud Nordic AB and has received funding for research carried out in this work.”L.K., none declared. L.R., none declared. T.L., none declared" | |
| Notes | Canoderm (ACO Hud Nordic, Sweden), an oil‐in‐water emulsion containing 5% urea, fractionated coconut oil, polysorbate 60, hydrogenated canola oil, propylene glycol, carbomer, dimethicone, hard paraffin, glyceryl polymethacrylate, propyl‐ and methyl parahydroxybenzoate, sodium lactate, lactic acid, glyceryl stearate, polyoxyethylene stearate, cetostearyl alcohol and purified water. The reference cream was Miniderm without glycerol (ACO Hud Nordic, Upplands Väsby, Sweden) and contains white soft paraffin, hydrogenated canola oil, light liquid paraffin, glyceryl stearate, polyoxyethylene stearate, cetostearyl alcohol, hard paraffin, dimethicone, cholesterol, propyl‐ and methyl parahydroxybenzoate and purified water. We received responses to our request for study details (Table 2). | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote (page 588): "Randomization to test cream and reference cream groups in 1:1, was performed according to a randomization list, with a block size of 4, and stratified for country with one randomization list for each country" and "The randomization was prepared by an independent statistician using a validated SAS® program" Comment: probably done. |
| Allocation concealment (selection bias) | Low risk | Quote (page 558): "The patients were provided with a randomization number" and "All study personnel at the clinics and the sponsor staff remained blinded during the maintenance part of the study" Comment: there was insufficient information to permit a clear judgement. After email communication: "The clinics received sealed envelopes containing individual treatment codes. The Investigator kept the envelopes in a secure, limited‐access location to prevent inadvertent breaking of the blind" Comment: central allocation; allocation appears to have been adequately concealed. |
| Blinding of participants and personnel (performance bias) | Low risk | Quote (page 557‐8): "double‐blind" and "All study personnel at the clinics and the sponsor staff remained blinded during the maintenance part of the study" Comment: the report did not provide sufficient detail about the specific measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. After email communication: "The test cream and the reference cream used in the maintenance phase were produced by Bioglan, Malmö, Sweden. The creams were packaged in identical plastic pump jars of 500 g and in tubes of 100 g" and "Before the products were distributed to the patients, they were also be labelled with the randomisation number" Comment: the report provided sufficient detail about the measures used to blind study participants and personnel from knowledge of which intervention a participant received, to permit a clear judgement. |
| Blinding of outcome assessment (detection bias) | Low risk | Quote (page 557‐8): "double‐blind" and "All study personnel at the clinics and the sponsor staff remained blinded during the maintenance part of the study" Outcomes were investigator‐assessed as well as participant‐assessed. Insufficient information to permit a clear judgement After email communication:(see domain above) Blinding of the outcomes assessors, key personnel, and participants was ensured, and it was unlikely that the blinding could have been broken. |
| Incomplete outcome data (attrition bias) | Low risk | 7/172 (4.1%); barrier‐strengthening moisturiser 3/87 (3.4%), reference cream 4/85 (4.7%), reasons reported. Intention‐to‐treat analysis. Comment: low number of losses to follow‐up, combined with intention‐to‐treat analysis, so considered to be at a low risk of bias. |
| Selective reporting (reporting bias) | Low risk | The protocol for the study was available (EudraCT Number: 2011‐001068‐23) and the prespecified outcomes and those mentioned in the Methods section appear to have been reported. |
| Other bias | Low risk | The study appears to be free of other forms of bias. |
AD: atopic dermatitis
ADSI: Atopic Dermatitis Severity Index
BHT: butylated hydroxytoluene
BSA: body surface area
CDLQI: Children’s Dermatology Life Quality Index
cis‐UCA: cis urocanic acid
DASI: dry skin area and severity index
DFI: Dermatitis Family Impact
DLQI: dermatology quality of life index
EASI: eczema area and severity index
ECG: electrocardiograph
EDTA: ethylenediaminetetraacetic acid
GCS: Global Change Scale
GIS: Global Impression Scale
HI: hydration index
HPA: hypothalamic‐pituitary‐adrenal
IDQOL: the Infant’s Dermatitis Quality of Life index
IGA: investigator Global Assessment
ILE: L‐isoleucine
MEA: monoethanolamine
MPA: methylprednisolone aceponate
MVE: multilamellar vesicular emulsion
NESS: Nottingham Atopic Eczema Severity Score
NSAIDs: non‐steroidal anti‐inflammatories
OTC: over‐the‐counter
PPARα: peroxisome proliferator‐activated receptor alpha
PCA: pyrrolidone carboxylic acid
PEG: polyethylene glycol
PGA: Physician Global Assessment
PUVA: photochemotherapy, a type of ultraviolet radiation treatment
POEM: Patient‐Oriented Eczema Measure
QoL: quality of life
SCORAD: scoring atopic dermatitis
TEWL: transepidermal water loss
TIS: three item severity score
UVA, UVB: types of ultraviolet light
VAS: visual analogue scale
VCO: virgin coconut oil
VOO: virgin olive oil
Characteristics of excluded studies [ordered by study ID]
| Study | Reason for exclusion |
| After reading full text, appeared to be a report on BATHE trial, BEEP study and CLOTHES trial and no primary research | |
| Two studies, one RCT in healthy volunteers, second study observational in atopic dermatitis | |
| Report on two studies, one was a CCT in atopic dermatitis | |
| CCT: email communication with investigators confirmed participants were allocated by alternation. "We included patients consecutively and every second patient was assigned to one group, while the others to the second treatment group" | |
| After translation, appeared to be a CCT |
Abbreviations
BATHE: Bath Additives in the Treatment of cHildhood Eczema (see Santer 2015)
BEEP: Barrier Enhancement for Eczema Prevention
CCT: controlled clinical trial
CLOTHES: Clothing for the relief of Eczema Symptoms
RCT: randomised controlled trial
Characteristics of ongoing studies [ordered by study ID]
| Study name | A randomised, double blind, placebo controlled comparative trial of ceramide cream and ceramide cleanser in the management of moderate eczema in adults |
| Methods | Randomised, double‐blind, placebo‐controlled study, Braeside Vic, Australia |
| Participants | N = 100 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (2): baseline, day 28 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | Not yet recruiting (website accessed 08‐02‐2016) |
| Contact information | Dr Kerryn Greive ([email protected]) |
| Notes |
| Study name | A randomised double blinded placebo controlled study investigating Atopis for the improvement of skin health in subjects with eczema |
| Methods | Randomised, double‐blind, placebo‐controlled study in St Rangiora, Christchurch, New Zealand |
| Participants | N = 60 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (3): baseline, weeks 2, and 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 04‐09‐2015, follow‐up complete (website accessed 08‐02‐2016) |
| Contact information | Dr Iona Weir ([email protected]) and Dr Simon Carson ([email protected]) |
| Notes | No study results posted |
| Study name | Pilot study comparing two moisturisers, Emulsifying ointment B.P. and Epaderm topical ointment in children with eczema |
| Methods | Randomised, double‐blind, controlled within‐participant study in Wellington, New Zealand |
| Participants | N = 60 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (5): baseline, weeks 2, 4, 6 and 8 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | Recruiting (website accessed 08‐02‐2016) |
| Contact information | |
| Notes |
| Study name | Effect and efficacy of a linoleic acid‐ceramide containing moisturizer as an adjunctive therapy in the treatment of atopic dermatitis: a randomized controlled trial |
| Methods | Randomised controlled trial in Shanghai |
| Participants | N = 10 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2003, completed (website accessed 11‐02‐2016) |
| Contact information | Xia Li ([email protected]) and Jie Zheng (jie‐[email protected]) |
| Notes | No study results posted. Similar study in chronic eczema is ongoing ChiCTR‐IOR‐15007140 with 29 participants in each treatment arm. However, unclear if this is atopic eczema |
| Study name | Evaluation of the impact of the moisturizing milk RV2478B on the use of corticoids and on quality of life, in a population of children with atopic dermatitis |
| Methods | Open, randomised controlled study in Italy |
| Participants | N = 210 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2004, completed (website accessed 09‐02‐2016) |
| Contact information | Not provided, sponsor is Pierre Fabre Médicament |
| Notes | Very little information provided; no study results posted |
| Study name | Long‐term management of atopic dermatitis with the moisturiser V0034 CR. A randomised, placebo‐controlled, parallel‐groups, double‐blind study in infants and children |
| Methods | Randomised controlled trial in Estonia, Finland, Germany, Latvia |
| Participants | N = 300 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (5): baseline, days 7, 14, 21 and 28 Outcomes of the trial (as reported) To evaluate the overall benefit of a regular treatment by the moisturiser V0034CR in the management of atopic dermatitis: reduction of corticosteroids consumption, reduction of flares
＊Denotes outcomes prespecified for this review |
| Starting date | 3‐11‐2005, completed (website accessed 09‐02‐2016) |
| Contact information | Not provided, sponsor is Pierre Fabre Médicament, however, see www.fda.gov/iceci/enforcementactions/warningletters/ucm211391.htm |
| Notes | Not clear from the website if desonide cream was used in both arms. No study results posted |
| Study name | A 6 week, multi centre study to evaluate the efficacy and safety of E45 complete moisturiser therapy and E45 itch relief cream in the treatment of mild to moderate atopic eczema compared to a patient's normal moisturiser regime defined as the frequent use of moisturisers (not containing lauromacrogols) listed in the BNF (Harbour Study) |
| Methods | Randomised controlled trial in UK |
| Participants | N = 180 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
|
| Outcomes | Assessments (3): baseline, days 7 and 21 Outcomes of the trial (as reported) To compare the efficacy of E45 complete moisturiser therapy against an unchanged moisturiser regime along with the role of E45 anti itch cream in controlling itch. This will be evaluated through validated quality of life questionnaires, IGA and diary cards.
＊Denotes outcomes prespecified for this review |
| Starting date | 27‐4‐2006, ongoing (website accessed 16‐2‐2016) |
| Contact information | Reckitt Benckiser Healthcare International, no further contact details available |
| Notes |
| Study name | A multicenter, randomized, double‐blind clinical study to examine the efficacy and safety of Zarzenda in comparison to Elidel in the management of mild to moderate atopic dermatitis in children and adolescents |
| Methods | Randomised, double‐blind, active‐controlled study, multicentre in Germany, Spain and Italy |
| Participants | N = 80 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (unspecified): baseline and at each visit (information unclear) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | Completed in Germany and Italy, but still ongoing in Spain (website accessed 09‐02‐2016) |
| Contact information | Not provided, sponsor is INTENDIS GmbH |
| Notes | Zardenda is another name for Atopiclair. No study results posted |
| Study name | Impact of the V0034CR 01B moisturiser on atopic dermatitis symptoms in children. A randomised, placebo‐controlled, parallel‐groups, double‐blind study |
| Methods | Randomised controlled trial in Estonia, France, Germany, Italy, Latvia, Lithuania |
| Participants | N= 550 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention (not clear from the study protocol on the website if desonide cream was used in both arms)
Comparator
|
| Outcomes | Assessments (unspecified): baseline and at each visit (information unclear) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | October 2008, completed in all countries except in France (ongoing) (website accessed 09‐02‐2016) |
| Contact information | Not provided, sponsor is Pierre Fabre Médicament |
| Notes | No study results posted |
| Study name | Evaluation of a developmental ‘long lasting’ moisturiser in subjects with dry skin |
| Methods | Randomised, double‐blind, controlled, within‐participant study in UK |
| Participants | N = 40 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments: unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2008, completed (website accessed 09‐02‐2016) |
| Contact information | Not provided, sponsor is Dermal Laboratories Ltd |
| Notes | No study results posted |
| Study name | Is treatment with moisturisers always beneficial? Comparative biophysical and molecular biological studies of the skin barrier function in patients with atopic dermatitis, ichthyosis vulgaris and X‐linked ichthyosis |
| Methods | Randomised, investigator‐blinded, controlled study in Sweden |
| Participants | N = 60 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
|
| Outcomes | Assessments (2): baseline, week 4 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2009, ongoing (website accessed 09‐02‐2016) |
| Contact information | Department of Medical Sciences, Dermatology and Venerology, Uppsala University, Uppsala Sweden, no further details provided |
| Notes |
| Study name | Arm immersion test to compare the skin effects of routine bathing with and without the use of an emollient bath additive |
| Methods | Randomised, investigator‐blinded, controlled within‐participant study in UK |
| Participants | N = 24 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2010, not recruiting (website accessed 12‐02‐2016) |
| Contact information | Sponsor Dermal Laboratories Ltd UK |
| Notes |
| Study name | Moisturisers in the management of atopic dermatitis in children: prevention of flares |
| Methods | Randomised, open, controlled study in Lithuania and Estonia |
| Participants | N = 409 Inclusion criteria of the trial Boys or girls, presenting with the following criteria (V1):
After treatment of the current flare, the following are required for randomisation (V2):
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
|
| Outcomes | Assessments (4): baseline, days 28, 56 and 84 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2013, completed (website accessed 09‐02‐2016) |
| Contact information | Pierre Fabre Médicament, Carine Fabre (carine.fabre@pierre‐fabre.com) |
| Notes | V0034CR = Dexeryl. A flare will be defined as the following: measurable increased extent or intensity of lesions in < 2 weeks under continued treatment (in this case: moisturiser or no moisturiser), corresponding to a significant increase (> 25%) in medical score (SCORAD or last PO‐SCORAD) or to the introduction of a new line of therapy (topical corticosteroid). No study results posted |
| Study name | Randomised, double‐blind, bilateral comparison of two moisturisers in patients with dry skin |
| Methods | Randomised, double‐blind, controlled within‐participant study in the UK |
| Participants | N = 20 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (16): baseline, days 1 until 5 (3 times a day) Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2014, completed (website accessed 09‐02‐2016) |
| Contact information | Dermal Laboratories Limited, Amanda Wigens, [email protected] |
| Notes | No study results posted |
| Study name | Investigating whether topical application of a non toxic molecule improves epidermal pH and barrier function in atopic dermatitis |
| Methods | Randomised, vehicle‐controlled study in UK |
| Participants | Not specified |
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (7): baseline and then weekly up to 6 weeks Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 01‐09‐2008, completed (website accessed 8‐2‐2016) |
| Contact information | Dr Neil Gibbs ([email protected]) |
| Notes | No study results posted |
| Study name | Randomized controlled trial to evaluate clinical efficacy of once‐ or twice‐daily skin care with moisturiser for prevention of atopic dermatitis exacerbation |
| Methods | Randomised, investigator‐blinded, controlled study in Japan |
| Participants | N = 150 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2011, recruiting (website accessed 12‐02‐2016 |
| Contact information | Yukihiro Ohya, National Center for Child Health and Development, Okura, Setagayaku, Tokyo, Japan |
| Notes | No further details regarding skin care regimen being tested |
| Study name | Study to evaluate the efficacy of conventional topical therapy for atopic dermatitis based on filaggrin‐gene mutations |
| Methods | Randomised, double‐blind controlled study in Japan |
| Participants | N = 30 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
Treatment includes steroid ointment and tacrolimus ointment |
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | 2012, still recruiting (website accessed 12‐02‐2016) |
| Contact information | Osamu Mizuno ([email protected]‐net.ne.jp) |
| Notes | Applying split‐plot design, multiple factors randomised. 4 sites at each cubital fossa and popliteal fossa are used as independent sites. First randomisation factor is presence or absence of filaggrin gene mutation. Second factor is combination of the usage or non‐usage of moisturiser and treatment. Treatment includes steroid ointment and tacrolimus ointment |
| Study name | The effect of moisturizer on both sweating function and sensitive skin in patients with atopic dermatitis, a randomized comparative study |
| Methods | Randomised, double‐blinded, placebo‐controlled study in Osaka, Japan |
| Participants | N = 25 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | June 2015 |
| Contact information | Hiroyuki Murota (h‐[email protected]‐u.ac.jp) |
| Notes | No study results posted |
| Study name | Control of therapy with Elidel vs placebo in patients with atopic dermatitis using bioengineering methods |
| Methods | Randomised, double‐blind, controlled study in Dresden, Germany |
| Participants | N = 24 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments: unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | April 2005, completed (website accessed 10‐02‐2016) |
| Contact information | Roland Aschoff, MD Department of Dermatology. Medical Faculty, Technical University Dresden, Germany |
| Notes | No study results posted |
| Study name | A multi‐centre, parallel, randomised study of the skin tolerance of betamethasone creams on atopic eczema and the influence of moisturiser treatment on the recurrence of eczema |
| Methods | Randomised, double‐blind, controlled study in Sweden |
| Participants | N = 55 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
|
| Outcomes | Assessments (3): baseline, week 3 and month 6 Outcomes of the trial (as reported) To study the:
＊Denotes outcomes prespecified for this review |
| Starting date | January 2004, completed (website accessed 11‐02‐2016) |
| Contact information | Principal investigator Berit Berne Hudkliniken, Akademiska sjukhuset, Uppsala, Sweden |
| Notes | No study results posted |
| Study name | A randomised, investigator‐blind, six‐week, parallel group, multicenter pilot study to compare the safety and efficacy of EpiCeram skin barrier emulsion and desonide cream 0.05% in the twice daily treatment of paediatric subjects with moderate atopic dermatitis |
| Methods | Randomised, investigator‐blinded, controlled study in the USA |
| Participants | N = 100 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (2): baseline and week 6 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | March 2009, completed (website accessed 10‐02‐2016) |
| Contact information | Promius Pharma LCC, Joanne Fraser ([email protected]) |
| Notes | Few data are available for one outcome |
| Study name | Moisturisers in the management of atopic dermatitis in children: Prevention of flares |
| Methods | Randomised controlled open‐label trial in Estonia, France, Lithuania, Poland, Romania |
| Participants | N = 347 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
|
| Outcomes | Assessments: baseline and further time points unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | February 2013, completed (website accessed 10‐2‐2016) |
| Contact information | Pierre Fabre Medicament. No further details available |
| Notes | No study results posted |
| Study name | Efficacy of KAM2904 face cream and KAM3008 body lotion treatment in children with atopic dermatitis (AD) |
| Methods | Randomised, double‐blind, controlled trial in Spain |
| Participants | N = 156 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (4): baseline, week 2, 4 and 6 Outcomes of the trial (as reported)
|
| Starting date | February 2013, completed (website accessed 10‐02‐2016) |
| Contact information | Miri Sani ([email protected]) |
| Notes | No study results posted |
| Study name | A randomized, open‐label, comparative study to evaluate an intermittent dosing regimen of fluticasone propionate 0.05% cream (twice per week) in reducing the risk of relapse when added to regular daily moisturization using PHYSIOGEL lotion in paediatric subjects with stabilized atopic dermatitis |
| Methods | Open, 2‐phase, randomised controlled study in China |
| Participants | N = 107 Inclusion criteria of the trial
Exclusion criteria of the trial (in the acute phase)
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments: baseline and further time points unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | December 2013, completed (website accessed 10‐02‐2016) |
| Contact information | GlaxoSmithKline. No further details available |
| Notes | No study results posted |
| Study name | The effect of bathing and moisturizers on skin hydration in atopic dermatitis: an in vivo study |
| Methods | Randomised, investigator‐blinded, controlled, cross‐over study in Thailand |
| Participants | N = 22 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
Comparator 3
The same 4 regimens were repeated with soaking an arm with mild cleanser All regimens were completed in all participants with 2 visits (7 days between visits) |
| Outcomes | Assessments: at baseline and every 30 minutes for 120 minutes after each regimen Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | January 2013, completed (website accessed 11‐02‐2016) |
| Contact information | Arisa Kaewkes, MSc dermatology Skin center, Faculty of Medicine, Srinakharinwirot University |
| Notes | No study results posted |
| Study name | Skin effects of a topical amino acid moisturizing cream and desonide in atopic dermatitis |
| Methods | Randomised, double‐blind, controlled study in Boston, USA |
| Participants | N = 42 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (4): baseline, week 1, 2 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | November 2014, still recruiting (website accessed 10‐2‐2016) |
| Contact information | Barbara A Green, RPh ([email protected]) and Justine K Gostomski, MS ([email protected]) |
| Notes |
| Study name | An explorative clinical trial to evaluate an intra patient comparison design of topical agents in adults with mild to moderate atopic dermatitis |
| Methods | Randomised, double‐blind, controlled study in Israel and Canada |
| Participants | N = 30 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
Comparator 3
|
| Outcomes | Assessments (2): baseline and week 2 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | February 2015, completed (website accessed 11‐02‐2016) |
| Contact information | Principal investigator Emma Guttmann, MD, Icahn Scool of Medicine |
| Notes | No study results posted |
| Study name | Effect of Cetaphil Restoraderm moisturizer on very dry skin in children with a controlled atopic dermatitis:a randomised, parallel group study |
| Methods | Randomised, investigator‐blinded controlled study in China and the Phillipines |
| Participants | N = 120 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
|
| Outcomes | Assessments (2): baseline and week 12 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | May 2015, still recruiting (website accessed 11‐2‐2016) |
| Contact information | Philippe Martel ([email protected]) and Stéphanie Leclerc ([email protected]) |
| Notes |
| Study name | Efficacy of a skin barrier repair cream (Dermalex Eczema) in atopic dermatitis patients ‐ EDA |
| Methods | Randomised, double‐blind, controlled, within‐participant study in the Netherlands |
| Participants | N = 100 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
|
| Outcomes | Assessments unclear Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | May 2014, recruiting (website accessed 12‐02‐2016) |
| Contact information | SA Koppes ([email protected]) |
| Notes |
| Study name | Choice Of Moisturiser in Eczema Treatment (COMET) |
| Methods | Randomised, investigator‐blinded, controlled study Setting University of Bristol, UK Date of study June 2014 to July 2015. Duration 3 months |
| Participants | N = 160 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator 1
Comparator 2
Comparator 3
Children will attend appointments for their eczema as normal and use other medications (for example, topical corticosteroids) as normally directed. Co‐prescribing of other leave on moisturisers and bath additives will be discouraged, but allowed |
| Outcomes | Assessments (4): baseline, months 1, 2 and 3 Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | June 2014. Completed (website accessed 8‐2‐2016) |
| Contact information | Dr Victoria J Wilson ([email protected]) |
| Notes | Feasability study. No study results posted |
| Study name | Bath additives for the treatment of childhood eczema |
| Methods | Open, randomised controlled study Setting Multicentre general practitioner practices in England and Wales, UK Date of study Unspecified. Duration 1 year |
| Participants | N = 423 Inclusion criteria of the trial
Exclusion criteria of the trial
|
| Interventions | Intervention
Comparator
Standard care includes advice to use leave‐on moisturisers regularly, plus intermittent topical corticosteroids if required |
| Outcomes | Assessments (25): baseline and then weekly up to 16 weeks, thereafter monthly Outcomes of the trial (as reported)
＊Denotes outcomes prespecified for this review |
| Starting date | The recruitment was expected to start in November 2014 (website accessed 8‐2‐2016) |
| Contact information | Kate Martinson, Trial Manager ([email protected]) |
| Notes | Trial registration number: ISRCTN84102309 |
Abbreviations
ADSI: Atopic Dermatitis Severity Index
BNF: British National Formulary
BSA: body surface area
CDLQI: Children’s Dermatology Life Quality Index
COPD: chronic obstructive pulmonary disease
DELP: Brandname of the gel under research
DFI: Dermatitis Family Impact questionnaire
EASI: eczema area and severity index
IDQOL: Infant's Dermatitis Quality of Life Index
IGA: Investigator Global Assessment
NSAIDs: non‐steroidal anti‐inflammatories
UVA and UVB: types of ultraviolet light
POEM: Patient‐Oriented Eczema Measure
PO SCORAD: Patient Oriented SCORAD
PSGA: Physician Static Global Assessment
SCORAD: scoring atopic dermatitis
SGA: Subjective Global Assessment
TADSI: Target Lesion Atopic Dermatitis Severity Index
TEWL: transepidermal water loss
TIS: three item severity score
UVA, UVB: types of ultraviolet light
VAS: visual analogue scale
Data and analyses
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Change from baseline in SCORAD Show forest plot | 3 | 276 | Mean Difference (IV, Random, 95% CI) | ‐2.42 [‐4.55, ‐0.28] |
| Analysis 1.1  Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 1: Change from baseline in SCORAD | ||||
| 1.2 Number of participants experiencing a flare Show forest plot | 2 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.40 [0.23, 0.70] |
| Analysis 1.2  Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 2: Number of participants experiencing a flare | ||||
| 1.3 Rate of flare Show forest plot | 2 | 87 | Hazard Ratio (IV, Random, 95% CI) | 3.74 [1.86, 7.50] |
| Analysis 1.3  Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 3: Rate of flare | ||||
| 1.4 Amount of topical steroids used Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| Analysis 1.4  Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 4: Amount of topical steroids used | ||||
| 1.4.1 Amount of topical corticosteroids first 3‐4 weeks | 2 | 222 | Mean Difference (IV, Random, 95% CI) | ‐8.25 [‐17.22, 0.72] |
| 1.4.2 Amount of topical corticosteroids used last 3‐4 weeks | 1 | 74 | Mean Difference (IV, Random, 95% CI) | 0.50 [‐4.70, 5.70] |
| 1.4.3 Total amount of topical corticosteroids used in 6 to 8 weeks | 2 | 222 | Mean Difference (IV, Random, 95% CI) | ‐9.30 [‐15.33, ‐3.27] |
| 1.5 Change from baseline in quality of life Show forest plot | 2 | 177 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.15 [‐0.55, 0.24] |
| Analysis 1.5  Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 5: Change from baseline in quality of life | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Number of participants who experienced good improvement to total resolution Show forest plot | 3 | 390 | Risk Ratio (M‐H, Random, 95% CI) | 4.51 [2.19, 9.29] |
| Analysis 2.1  Comparison 2: Atopiclair versus vehicle, Outcome 1: Number of participants who experienced good improvement to total resolution | ||||
| 2.1.1 Number of participants who experienced good improvement to total resolution (low risk of bias) | 1 | 30 | Risk Ratio (M‐H, Random, 95% CI) | 4.00 [1.01, 15.81] |
| 2.1.2 Number of participants who experienced good improvement to total resolution (unclear risk of bias) | 1 | 142 | Risk Ratio (M‐H, Random, 95% CI) | 8.06 [3.95, 16.42] |
| 2.1.3 Number of participants who experienced good improvement to total resolution (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 3.02 [2.00, 4.56] |
| 2.2 Change from baseline in itch measured on a VAS Show forest plot | 4 | 396 | Mean Difference (IV, Random, 95% CI) | ‐2.65 [‐4.21, ‐1.09] |
| Analysis 2.2  Comparison 2: Atopiclair versus vehicle, Outcome 2: Change from baseline in itch measured on a VAS | ||||
| 2.2.1 Change from baseline in itch measured on a VAS (low risk of bias) | 1 | 30 | Mean Difference (IV, Random, 95% CI) | ‐0.80 [‐1.20, ‐0.40] |
| 2.2.2 Change from baseline in itch measured on a VAS (unclear risk of bias) | 2 | 180 | Mean Difference (IV, Random, 95% CI) | ‐3.10 [‐4.47, ‐1.73] |
| 2.2.3 Change from baseline in itch measured on a VAS (high risk of bias) | 1 | 186 | Mean Difference (IV, Random, 95% CI) | ‐3.70 [‐4.66, ‐2.74] |
| 2.3 Number of participants reporting an adverse event Show forest plot | 4 | 430 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.79, 1.33] |
| Analysis 2.3  Comparison 2: Atopiclair versus vehicle, Outcome 3: Number of participants reporting an adverse event | ||||
| 2.4 Change from baseline in EASI Show forest plot | 4 | 426 | Mean Difference (IV, Random, 95% CI) | ‐4.00 [‐5.42, ‐2.57] |
| Analysis 2.4  Comparison 2: Atopiclair versus vehicle, Outcome 4: Change from baseline in EASI | ||||
| 2.4.1 Change from baseline in EASI (low risk of bias) | 1 | 30 | Mean Difference (IV, Random, 95% CI) | ‐3.30 [‐5.67, ‐0.93] |
| 2.4.2 Change from baseline in EASI (unclear risk of bias) | 2 | 180 | Mean Difference (IV, Random, 95% CI) | ‐4.42 [‐7.73, ‐1.10] |
| 2.4.3 Change from baseline in EASI (high risk of bias) | 1 | 216 | Mean Difference (IV, Random, 95% CI) | ‐3.62 [‐5.06, ‐2.18] |
| 2.5 Number of participants experiencing a flare Show forest plot | 3 | 397 | Risk Ratio (M‐H, Random, 95% CI) | 0.18 [0.11, 0.31] |
| Analysis 2.5  Comparison 2: Atopiclair versus vehicle, Outcome 5: Number of participants experiencing a flare | ||||
| 2.5.1 Number of participants experiencing a flare (unclear risk of bias) | 2 | 181 | Risk Ratio (M‐H, Random, 95% CI) | 0.26 [0.12, 0.57] |
| 2.5.2 Number of participants experiencing a flare (high risk of bias) | 1 | 216 | Risk Ratio (M‐H, Random, 95% CI) | 0.14 [0.07, 0.28] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 3.1 Change from baseline in skin capacitance Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | 1.23 [‐7.39, 9.86] | |
| Analysis 3.1  Comparison 3: Urea‐containing moisturiser versus vehicle, Outcome 1: Change from baseline in skin capacitance | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 4.1 Number of participants reporting an adverse event Show forest plot | 2 | 385 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.68, 1.19] |
| Analysis 4.1  Comparison 4: Glycerin cream versus placebo cream, Outcome 1: Number of participants reporting an adverse event | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 5.1 Change in disease severity as assessed by the investigators (SCORAD and EASI) Show forest plot | 3 | 272 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.23 [‐0.66, 0.21] |
| Analysis 5.1  Comparison 5: Oat‐containing cream versus vehicle or no treatment, Outcome 1: Change in disease severity as assessed by the investigators (SCORAD and EASI) | ||||
| 5.1.1 Change in disease severity as assessed by the investigator (EASI) (low risk of bias) | 1 | 50 | Std. Mean Difference (IV, Random, 95% CI) | 0.01 [‐0.55, 0.56] |
| 5.1.2 Change in disease severity as assessed by the investigators (SCORAD) (high risk of bias) | 2 | 222 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.33 [‐0.98, 0.32] |
| 5.2 Change from baseline in quality of life Show forest plot | 3 | 226 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.09 [‐0.37, 0.19] |
| Analysis 5.2  Comparison 5: Oat‐containing cream versus vehicle or no treatment, Outcome 2: Change from baseline in quality of life | ||||
| 5.2.1 Change from baseline in quality of life (low risk of bias) | 1 | 50 | Std. Mean Difference (IV, Random, 95% CI) | 0.10 [‐0.46, 0.65] |
| 5.2.2 Change from baseline in quality of life (high risk of bias) | 2 | 176 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.16 [‐0.55, 0.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 6.1 Number of participants who experienced improvement Show forest plot | 5 | 572 | Risk Ratio (M‐H, Random, 95% CI) | 2.46 [1.16, 5.23] |
| Analysis 6.1  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 1: Number of participants who experienced improvement | ||||
| 6.1.1 Number of participants who experienced improvement (low risk of bias) | 2 | 80 | Risk Ratio (M‐H, Random, 95% CI) | 1.86 [0.41, 8.31] |
| 6.1.2 Number of participants who experienced improvement (unclear risk of bias) | 2 | 274 | Risk Ratio (M‐H, Random, 95% CI) | 3.11 [0.25, 38.71] |
| 6.1.3 Number of participants who experienced improvement (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 3.02 [2.00, 4.56] |
| 6.2 Change from baseline in itch Show forest plot | 7 | 749 | Std. Mean Difference (IV, Random, 95% CI) | ‐1.10 [‐1.83, ‐0.38] |
| Analysis 6.2  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 2: Change from baseline in itch | ||||
| 6.2.1 Change from baseline in itch (low risk of bias) | 3 | 329 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.36 [‐1.16, 0.43] |
| 6.2.2 Change from baseline in itch (unclear risk of bias) | 2 | 180 | Std. Mean Difference (IV, Random, 95% CI) | ‐2.29 [‐2.67, ‐1.91] |
| 6.2.3 Change from baseline in itch (high risk of bias) | 2 | 240 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.95 [‐1.75, ‐0.16] |
| 6.3 Number of participants who expressed treatment satisfaction Show forest plot | 3 | 298 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [0.77, 2.36] |
| Analysis 6.3  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 3: Number of participants who expressed treatment satisfaction | ||||
| 6.3.1 Number of participants who expressed treatment satisfaction (low risk of bias) | 2 | 80 | Risk Ratio (M‐H, Random, 95% CI) | 1.04 [0.77, 1.42] |
| 6.3.2 Number of participants who expressed treatment satisfaction (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 2.14 [1.58, 2.89] |
| 6.4 Number of participants reporting an adverse event Show forest plot | 10 | 1275 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.82, 1.30] |
| Analysis 6.4  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 4: Number of participants reporting an adverse event | ||||
| 6.4.1 Number of participants reporting an adverse event (low risk of bias) | 4 | 471 | Risk Ratio (M‐H, Random, 95% CI) | 0.83 [0.58, 1.18] |
| 6.4.2 Number of participants reporting an adverse events (unclear risk of bias) | 3 | 314 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.82, 1.42] |
| 6.4.3 Number of participants reporting an adverse events (high risk of bias) | 3 | 490 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [0.37, 4.96] |
| 6.5 Change in disease severity as assessed by the investigators Show forest plot | 12 | 1281 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.65 [‐0.89, ‐0.41] |
| Analysis 6.5  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 5: Change in disease severity as assessed by the investigators | ||||
| 6.5.1 Change in disease severity as assessed by the investigators (low risk of bias) | 5 | 512 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.45 [‐0.74, ‐0.15] |
| 6.5.2 Change in disease severity as assessed by the investigators (unclear risk of bias) | 2 | 180 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.93 [‐1.29, ‐0.57] |
| 6.5.3 Change in disease severity as assessed by the investigators (high risk of bias) | 5 | 589 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.77 [‐1.23, ‐0.30] |
| 6.6 Number of participants experiencing a flare Show forest plot | 6 | 607 | Risk Ratio (M‐H, Random, 95% CI) | 0.33 [0.17, 0.62] |
| Analysis 6.6  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 6: Number of participants experiencing a flare | ||||
| 6.6.1 Number of participants experiencing a flare (low risk of bias) | 1 | 123 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.48, 1.34] |
| 6.6.2 Number of participants experiencing a flare (unclear risk of bias) | 2 | 181 | Risk Ratio (M‐H, Random, 95% CI) | 0.26 [0.12, 0.57] |
| 6.6.3 Number of participants experiencing a flare (high risk of bias) | 3 | 303 | Risk Ratio (M‐H, Random, 95% CI) | 0.27 [0.12, 0.59] |
| 6.7 Change from baseline in quality of life Show forest plot | 3 | 300 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.39 [‐0.90, 0.12] |
| Analysis 6.7  Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 7: Change from baseline in quality of life | ||||
| 6.7.1 Change from baseline in quality of life (low risk of bias) | 1 | 123 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.81 [‐1.18, ‐0.44] |
| 6.7.2 Change from baseline in quality of life (high risk of bias) | 2 | 177 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.15 [‐0.55, 0.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 7.1 Change from baseline in TEWL Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | ‐0.34 [‐1.44, 0.76] | |
| Analysis 7.1  Comparison 7: Evening primrose oil versus placebo oil, Outcome 1: Change from baseline in TEWL | ||||
| 7.2 Change from baseline in skin hydration Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | 0.34 [‐2.54, 3.21] | |
| Analysis 7.2  Comparison 7: Evening primrose oil versus placebo oil, Outcome 2: Change from baseline in skin hydration | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 8.1 Change from baseline in itch (VAS) Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | ‐0.48 [‐1.46, 0.50] | |
| Analysis 8.1  Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 1: Change from baseline in itch (VAS) | ||||
| 8.1.1 Change from baseline in itch (VAS) (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | 0.00 [‐0.47, 0.47] | |
| 8.1.2 Change from baseline in itch (VAS) (high risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐1.00 [‐1.61, ‐0.39] | |
| 8.2 Change from baseline in SCORAD Show forest plot | 3 | Mean Difference (IV, Random, 95% CI) | 0.08 [‐1.96, 2.13] | |
| Analysis 8.2  Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 2: Change from baseline in SCORAD | ||||
| 8.2.1 Change from baseline in SCORAD (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐2.00 [‐3.47, ‐0.53] | |
| 8.2.2 Change from baseline in SCORAD (high risk of bias) | 2 | Mean Difference (IV, Random, 95% CI) | 1.12 [‐1.38, 3.61] | |
| 8.3 Change from baseline in TEWL Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | ‐0.50 [‐5.88, 4.87] | |
| Analysis 8.3  Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 3: Change from baseline in TEWL | ||||
| 8.3.1 Change from baseline in TEWL (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐3.00 [‐4.71, ‐1.29] | |
| 8.3.2 Change from baseline in TEWL (high risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | 2.51 [‐1.21, 6.23] | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 9.1 Rate of flare Show forest plot | 1 | Hazard Ratio (IV, Random, 95% CI) | Totals not selected | |
| Analysis 9.1  Comparison 9: Advabase versus MPA cream twice weekly and emollient, Outcome 1: Rate of flare | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 10.1 Number of participants reporting an adverse event Show forest plot | 3 | 718 | Risk Ratio (M‐H, Random, 95% CI) | 0.51 [0.22, 1.14] |
| Analysis 10.1  Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 1: Number of participants reporting an adverse event | ||||
| 10.2 Number of participants experiencing a flare Show forest plot | 3 | 718 | Risk Ratio (M‐H, Random, 95% CI) | 2.17 [1.51, 3.11] |
| Analysis 10.2  Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 2: Number of participants experiencing a flare | ||||
| 10.3 Rate of flare Show forest plot | 3 | 723 | Hazard Ratio (IV, Random, 95% CI) | 3.69 [1.80, 7.55] |
| Analysis 10.3  Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 3: Rate of flare | ||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 11.1 Change in disease severity as assessed by the investigators Show forest plot | 2 | 192 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.87 [‐1.17, ‐0.57] |
| Analysis 11.1  Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 1: Change in disease severity as assessed by the investigators | ||||
| 11.2 Change in quality of life IDQOL Show forest plot | 1 | 67 | Mean Difference (IV, Random, 95% CI) | ‐1.31 [‐2.70, 0.09] |
| Analysis 11.2  Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 2: Change in quality of life IDQOL | ||||
| 11.3 Change in quality of life DFI Show forest plot | 1 | 67 | Mean Difference (IV, Random, 95% CI) | ‐1.03 [‐2.47, 0.42] |
| Analysis 11.3  Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 3: Change in quality of life DFI | ||||
Study flow diagram
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies
Risk of bias summary: review authors' judgements about each risk of bias item for each included study
Funnel plot of comparison 5: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), outcome: 5.5 Change in disease severity as assessed by the investigators
Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 1: Change from baseline in SCORAD
Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 2: Number of participants experiencing a flare
Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 3: Rate of flare
Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 4: Amount of topical steroids used
Comparison 1: Moisturisers versus no treatment (i.e. no moisturiser), Outcome 5: Change from baseline in quality of life
Comparison 2: Atopiclair versus vehicle, Outcome 1: Number of participants who experienced good improvement to total resolution
Comparison 2: Atopiclair versus vehicle, Outcome 2: Change from baseline in itch measured on a VAS
Comparison 2: Atopiclair versus vehicle, Outcome 3: Number of participants reporting an adverse event
Comparison 2: Atopiclair versus vehicle, Outcome 4: Change from baseline in EASI
Comparison 2: Atopiclair versus vehicle, Outcome 5: Number of participants experiencing a flare
Comparison 3: Urea‐containing moisturiser versus vehicle, Outcome 1: Change from baseline in skin capacitance
Comparison 4: Glycerin cream versus placebo cream, Outcome 1: Number of participants reporting an adverse event
Comparison 5: Oat‐containing cream versus vehicle or no treatment, Outcome 1: Change in disease severity as assessed by the investigators (SCORAD and EASI)
Comparison 5: Oat‐containing cream versus vehicle or no treatment, Outcome 2: Change from baseline in quality of life
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 1: Number of participants who experienced improvement
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 2: Change from baseline in itch
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 3: Number of participants who expressed treatment satisfaction
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 4: Number of participants reporting an adverse event
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 5: Change in disease severity as assessed by the investigators
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 6: Number of participants experiencing a flare
Comparison 6: All moisturisers versus vehicle, placebo or no treatment (no moisturiser), Outcome 7: Change from baseline in quality of life
Comparison 7: Evening primrose oil versus placebo oil, Outcome 1: Change from baseline in TEWL
Comparison 7: Evening primrose oil versus placebo oil, Outcome 2: Change from baseline in skin hydration
Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 1: Change from baseline in itch (VAS)
Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 2: Change from baseline in SCORAD
Comparison 8: Licochalcone versus hydrocortisone acetate (HCA) 1%, Outcome 3: Change from baseline in TEWL
Comparison 9: Advabase versus MPA cream twice weekly and emollient, Outcome 1: Rate of flare
Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 1: Number of participants reporting an adverse event
Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 2: Number of participants experiencing a flare
Comparison 10: Vehicle + daily moisturiser versus fluticasone propionate (FP) + daily moisturiser, Outcome 3: Rate of flare
Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 1: Change in disease severity as assessed by the investigators
Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 2: Change in quality of life IDQOL
Comparison 11: Active treatment in combination with a moisturiser versus active treatment only, Outcome 3: Change in quality of life DFI
| Moisturisers versus no moisturiser for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with no moisturiser | Risk with moisturisers | |||||
| Change from baseline in disease severity according to participants ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Participant satisfaction ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Number of participants reporting an adverse event | Study population | RR15.34 | 173 | ⊕⊕⊝⊝ | 8/91 versus 0/82 reported an adverse event. Peto OR 7.26 (95% CI 1.76 to 29.92). 3 adverse events were reported to be mild, 3 moderate, and 2 were severe leading to treatment discontinuation. No adverse events were reported in the study of Simpson 2013 (within‐participant). | |
| 1 per 100 (0.5/82)a | 9 per 100 (1 to100) | |||||
| Change from baseline in disease severity as assessed by the investigators | The mean change from baseline in disease severity as assessed by the investigators ranged from ‐2.4 to ‐19.5 | The mean change from baseline in disease severity as assessed by the investigators in the intervention group was 2.42 lower (4.55 lower to 0.28 lower) | ‐ | 276 | ⊕⊕⊝⊝ | Reductions from baseline in Giordano‐Labadie 2006 and Grimalt 2007 met MID (= 8.7 Schram 2012) in both treatment arms. There was greater severity of disease in these studies than in Patrizi 2014. A MD of ‐2.42, although statistically significant, is not clinically important. |
| Number of participants who experienced a flare | Study population | RR 0.40 | 87 | ⊕⊕⊝⊝ | There were fewer flares in the moisturiser groups. The rate of flare in the control group was 3.74 times the rate in the moisturiser group (hazard ratio (HR) 3.74, 95% CI 1.86 to 7.50; P = 0.0002). | |
| 67 per 100 | 27 per 100 | |||||
| Amount of corticosteroids used | The mean amount of corticosteroids used ranged from 22.73 g to 62.1 g | The mean amount of corticosteroids used in the intervention group was 9.30 g less (15.30 g less to 3.27 g less) | ‐ | 222 | ⊕⊕⊝⊝ | P = 0.003. There was a statistically significant difference showing that the use of moisturisers decreased the use of topical corticosteroids to achieve similar reductions in SCORAD. |
| Change from baseline in health‐related quality of life | ‐ | The mean change from baseline in health‐related quality of life in the intervention group calculated as the SMD was 0.15 lower (0.55 lower to 0.24 higher) | ‐ | 177 | ⊕⊕⊝⊝ | There was no statistically significant difference in change from baseline of quality of life between the 2 treatment arms. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aWe had to put a value other than 0 in GRADEproGDT to calculate the risk with no moisturiser in relation to the RR, and we chose 0.5 (after discussion with the GRADE working Group). GRADEproGDT then calculates the risk with moisturiser. 2Downgraded one level due to high risk of bias because of performance and detection bias. 3Downgraded one level due to serious imprecision (wide confidence interval, low occurrence of events). 4Giordano‐Labadie 2006, Grimalt 2007, Patrizi 2014. 5Downgraded one level for serious inconsistency (I² = 68%), caused by Grimalt 2007. 7Downgraded one level for serious imprecision (small sample size). 8Giordano‐Labadie 2006, Grimalt 2007. 9Downgraded one level for serious inconsistency (I² = 68%). In the study of Giordano‐Labadie 2006, far more topical corticosteroids were used and the difference between the two arms was much larger. 10Downgraded one level for serious imprecision (wide confidence interval). | ||||||
| Atopiclair versus vehicle for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with vehicle | Risk with Atopiclair | |||||
| Change from baseline in disease severity according to participants (number of participants who considered their skin to have improved) | Study population | RR 4.51 | 390 | ⊕⊕⊕⊝ | Participants considered Atopiclair more effective than its vehicle. NNTB = 2, 95% CI 1 to 2 | |
| 17 per 100 | 77 per 100 | |||||
| Participant satisfaction | Study population | Not estimable | 248 | ⊕⊕⊕⊝ | Abramovits 2008: 119/145 (Atopiclair) vs 28/73 (vehicle) wished to continue (RR 2.14, 95% CI 1.58 to 2.89; P < 0.00001; NNTB = 2; 95% CI 2 to 3). Belloni 2005: 5/15 vs 0/15 would use again (Peto OR 10.18, 95% CI 1.54 to 67.23; P = 0.02) | |
| Not pooled | Not pooled | |||||
| Number of participants reporting an adverse event | Study population | RR 1.03 | 430 | ⊕⊕⊕⊝ | The number of participants reporting adverse events was not statistically different between the 2 groups. | |
| 33 per 100 | 34 per 100 | |||||
| Change from baseline in disease severity according to the investigators | The mean change from baseline in disease severity according to the investigators ranged from ‐1.7 to 0.84 | The mean change from baseline in disease severity according to the investigators in the intervention group was 4 lower (5.42 lower to 2.57 lower) | ‐ | 426 | ⊕⊕⊕⊝ | Although there is a statistically significant difference in favour of Atopiclair, the difference between the treatment group is not clinically important (MID EASI is 6.6 (Schram 2012)). |
| Number of participants who experienced a flare | Study population | RR 0.18 | 397 | ⊕⊕⊕⊝ | Participants in the Atopiclair group experienced fewer flares than the vehicle group (NNTB 3, 95% CI 3 to 5). | |
| 35 per 100 | 6 per 100 | |||||
| Change in use of topical active treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in health‐related quality of life ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Abramovits 2008, Belloni 2005, Boguniewicz 2008. 2Downgraded one level for serious imprecision (wide confidence interval). 3Abramovits 2008, Belloni 2005. 4Downgraded one level for serious indirectness, as outcomes did not exactly match participant satisfaction. 5Abramovits 2008, Belloni 2005, Boguniewicz 2008, Patrizi 2008. 6Downgraded one level for serious imprecision (small sample size and CI includes no difference (1) and appreciable harm (1.25)). 7Downgraded one level for serious inconsistency (I² = 51%), caused by Boguniewicz 2008, which showed a larger effect size. 8Abramovits 2008, Boguniewicz 2008, Patrizi 2008. 9Downgraded one level for risk of bias (Abramovits 2008: high risk for attrition bias, Boguniewicz 2008: unclear risk of bias for allocation concealment blinding and incomplete outcome data, and Patrizi 2008: at unclear risk of bias due to incomplete outcome data). | ||||||
| Urea‐containing moisturiser versus vehicle, placebo or no moisturiser for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with vehicle, placebo or no moisturiser | Risk with urea‐ containing moisturiser | |||||
| Change from baseline in disease severity according to the participants (number of participants who considered their skin to have improved) | Study population | RR 1.28 | 129 | ⊕⊕⊝⊝ | P = 0.0009. NNTB = 5 (95% CI 3 to 18). Participants considered that urea‐containing moisturiser provided more improvement than placebo cream without urea. In Wilhelm 1998 (n = 77, within‐participant design), 61% considered that the side treated with urea cream showed moderate to very good improvement, and 48.1% felt the vehicle‐treated side showed moderate to very good improvement. | |
| 70 per 100 | 89 per 100 | |||||
| Participant satisfaction Assessed with: Likert scale Follow‐up: mean 4 weeks | ‐ | 38 (1 RCT) 4, 5 | ⊕⊕⊝⊝ | Smell, spreadability, penetration into the skin, and skin feel were assessed. None of these features were assessed as being better on the urea‐treated side than on the vehicle‐treated side. For details, see comparison 2b under Effects of interventions. | ||
| Number of participants reporting an adverse event | Study population | RR 1.65 | 129 | ⊕⊕⊕⊝ | P = 0.005; NNTH = 4, 95% CI 2 to 11.There were fewer adverse events in the group treated with placebo cream. | |
| 39 per 100 | 65 per 100 | |||||
| Change from baseline in disease severity according to the investigators (number of participants who improved according to the investigators) | Study population | RR 1.40 | 129 | ⊕⊕⊕⊝ | The assessments of the investigators were in line with the assessments of the participants. P = 0.001; NNTB = 4, 95% CI 3 to 9. The within‐participant study of Wilhelm 1998 demonstrated a mean of the paired differences of ‐0.57 (95% CI ‐1.14 to 0.0) in favour of urea moisturiser (lower score being better), and is more or less in line with the parallel‐design study of Lodén 2002. | |
| 64 per 100 | 89 per 100 | |||||
| Number of participants who experienced a flare | Study population | RR 0.47 | 44 | ⊕⊕⊝⊝ | P = 0.03; NNTB = 3, 95% CI 2 to 11 The rate of flare in the group that did not use a moisturiser was 3.2 times the rate in the group treated with urea cream (HR 3.2, 95% CI 1.3 to 7.8; P < 0.01). | |
| 68 per 100 | 32 per 100 | |||||
| Change in use of topical active treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in health‐related quality of life ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 2Downgraded one level for serious indirectness, the study of Lodén 2002 had a parallel study design and the study of Wilhelm 1998 had a within‐participant design. 3Downgraded one level for serious imprecision (small sample size). 5Within‐participant design. 6Downgraded two levels for very serious imprecision (very small sample size). 8Downgraded one level for risk of bias as Wirén 2009 was assessed as at high risk of bias as the study was not blinded. | ||||||
| Glycerin/glycerol‐containing moisturiser versus vehicle or placebo for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with vehicle or placebo | Risk with glycerol‐containing moisturiser | |||||
| Change from baseline in disease severity as assessed by the participants | Study population | RR 1.22 | 134 | ⊕⊕⊕⊝ | Participants considered glycerol‐containing moisturiser more effective for improving dry skin than placebo cream (P = 0.03; NNTB = 6, 95% CI 3 to 60) | |
| 70 per 100 | 85 per 100 | |||||
| Participant satisfaction ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Number of participants reporting an adverse event | Study population | RR 0.90 | 385 | ⊕⊕⊕⊝ | The adverse events were mild to moderate and consisted of smarting, erythema, pruritus, or burning. | |
| 35 per 100 | 32 per 100 | |||||
| Change from baseline in disease severity as assessed by the investigators | The mean change from baseline in disease severity as assessed by the investigators was ‐3.1 | The mean change from baseline in disease severity as assessed by the investigators in the intervention group was 2.2 lower (3.44 lower to 0.96 lower) | ‐ | 249 | ⊕⊕⊕⊕ | P = 0.0005, but does not meet the MID (which is 8.2 for objective SCORAD (Schram 2012)). The study of Breternitz 2008 had a within‐participant design and confirmed these data. The mean of the paired differences was ‐1.10, CI 95% ‐1.63 to ‐0.57. In Lodén 2002, in the glycerol group 58/68 showed improvement in 'dryness' of the skin versus 42/66 in the vehicle group (RR 1.34, 95% CI 1.09 to 1.65; P = 0.0006, NNTB 5, 95% CI 3 to 14) |
| Number of participants who experienced a flare ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change in use of topical active treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in health‐related quality of life ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 2Downgraded one level for serious imprecision (lower bound of CI approaches 1). 3Lodén 2002 and Boralevi 2014. 4Downgraded one level for serious imprecision (small sample size and CI includes appreciable benefit (0.75) and no difference (1)). | ||||||
| Oat‐containing moisturiser versus vehicle or no moisturiser | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with vehicle or no moisturiser | Risk with oat‐containing moisturiser | |||||
| Change from baseline in disease severity as assessed by the participants (number of participants who considered their skin to have improved) | Study population | RR 1.11 | 50 | ⊕⊕⊝⊝ | P = 0.45. Participants did not consider that the oat‐containing moisturiser was more effective than the control treatment (occlusive vehicle). | |
| 76 per 100 | 84 per 100 | |||||
| Participant satisfaction | Study population | RR 1.06 | 50 | ⊕⊝⊝⊝ | P = 0.76. Participants were not more satisfied with oat‐containing moisturiser than with the occlusive vehicle. | |
| 68 per 100 | 72 per 100 | |||||
| Number of participants reporting an adverse event | Study population | RR 15.34 | 173 | ⊕⊕⊝⊝ | 8/91 versus 0/82 reported an adverse event. | |
| 1 per 100 | 9 per 100 | |||||
| Change from baseline in disease severity as assessed by the investigators | ‐ | The mean change from baseline in disease severity in the intervention group calculated as the SMD was 0.23 lower (0.66 lower to 0.21 higher) | ‐ | 272 | ⊕⊕⊝⊝ | P = 0.30. There was no statistically significant difference according to the investigators between the 2 treatment arms. |
| Number of participants who experienced a flare | Study population | RR 0.31 | 43 | ⊕⊕⊝⊝ | P = 0.01; NNTB = 2, 95% CI 1 to 5. The HR for rate of flare was 4.74 (95% CI 1.57 to 14.34; P = 0.006) in favour of the oat‐containing cream. | |
| 65 per 100 | 20 per 100 | |||||
| Total amount of topical corticosteroids used | The mean total amount of topical corticosteroids used ranged from 22.73 g to 62.1 g | The mean total amount of topical corticosteroids used in the intervention group was 9.3 g lower (15.3 g less to 3.27 g less) | ‐ | 222 | ⊕⊕⊝⊝ | P = 0.003. There is a statistically significant difference showing that the use of moisturisers decreased the use of topical corticosteroids to achieve similar reductions in disease severity. |
| Change from baseline in health‐related quality of life | ‐ | The mean change from baseline in health‐related quality of life in the intervention group calculated as the SMD was 0.09 lower (0.37 lower to 0.19 higher) | ‐ | 226 | ⊕⊕⊝⊝ | There was no statistically significant difference in change from baseline in quality of life between the 2 treatment arms. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| aWe had to put a value other than 0 in GRADEproGDT to calculate the risk with no moisturiser in relation to the RR, and we chose 0.5 (after discussion with the GRADE working Group). GRADEproGDT then calculates the risk with moisturiser. 2Downgraded two levels level for very serious imprecision (small sample size and CI includes no effect (1) and appreciable benefit (1.25)). 3Downgraded one level for serious indirectness as the outcome was more about soothing and calming of the skin and not really participant satisfaction. 4Downgraded two levels for very serious imprecision as the CI includes both no effect, and benefit of both oat‐containing cream as well as of the vehicle. 5Downgraded one level for risk of bias because of performance and detection bias. 6Downgraded one level for serious imprecision (wide confidence interval, low occurrence of events). 8Giordano‐Labadie 2006, Grimalt 2007, Nebus 2009. 9Downgraded one level for serious inconsistency (I² = 65%), caused by Giordano‐Labadie 2006, which was the study showing a favourable result for the oat‐containing creams whilst the other studies showed no difference between the treatment arms. 10Downgraded one level for serious imprecision; the CI creates uncertainty with the effect, ranging from moderate effect to small harmful effect. 11Weber 2015. 12Downgraded one level for serious imprecision (small sample size). 13Giordano‐Labadie 2006 and Grimalt 2007. 14Downgraded one level for serious inconsistency (I² = 68%). In the study of Giordano‐Labadie 2006, far more topical corticosteroids were used and the difference between the two arms was much larger. 15Downgraded one level for serious imprecision (wide confidence interval). 16Downgraded one level for serious risk of bias because of performance, detection, and attrition bias. 17Downgraded one level for serious imprecision (the CI creates uncertainty with the effect, ranging from small effect to small harmful effect). | ||||||
| All moisturisers compared to vehicle, placebo or no moisturiser for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with vehicle, placebo or no moisturiser | Risk with all moisturisers | |||||
| Change from baseline in disease severity as assessed by the participants (number of participants who considered their skin to have improved) | Study population | RR 2.46 | 572 | ⊕⊕⊝⊝ | Participants considered the use of a moisturiser to be more effective than vehicle/placebo or no moisturiser. P = 0.02, NNTB = 2, 95% CI 2 to 3 | |
| 37 per 100 | 91 per 100 | |||||
| Participant satisfaction | Study population | RR 1.35 | 298 | ⊕⊕⊝⊝ | P = 0.29. According to the participants, there was no difference between the 2 treatment arms for this outcome. Results are supported by the within‐participant study (Bohnsack 1997). | |
| 48 per 100 | 65 per 100 | |||||
| Number of participants reporting an adverse event | Study population | RR 1.03 | 1275 | ⊕⊕⊕⊝ | There was no statistically significant difference in number of participants experiencing an adverse event. | |
| 23 per 100 | 24 per 100 | |||||
| Change from baseline in disease severity as assessed by the investigators | ‐ | The mean change from baseline in disease severity as assessed by the investigators in the intervention group calculated as the SMD was 1.04 lower (1.57 lower to 0.51 lower) | ‐ | 1281 | ⊕⊕⊕⊕ | P < 0.0001 The investigators considered the use of moisturisers to be more beneficial than the vehicle, placebo, or no moisturiser. However, clinical impact was unclear. |
| Number of participants who experienced a flare | Study population | RR 0.33 | 607 | ⊕⊕⊕⊝ | P = 0.006; NNTB = 4, 95% CI 3 to 5. The rate of flare in the control group was 3.74 times the rate in the moisturiser group based on Weber 2015 and Wirén 2009 (HR 3.74, 95% CI 1.86 to 7.50; P = 0.0002 in favour of moisturiser). | |
| 41 per 100 | 13 per 100 | |||||
| Total amount of topical corticosteroids used | The mean amount of corticosteroids used ranged from 22.73 g to 62.1 g | The mean amount of corticosteroids used in the intervention group was 9.30 g less (15.30 g less to 3.27 g less) | ‐ | 222 | ⊕⊕⊝⊝ | P = 0.003. There was a statistically significant difference showing that the use of moisturisers decreased the use of topical corticosteroids to achieve similar reductions in eczema severity. |
| Change from baseline in health‐related quality of life | ‐ | The mean change from baseline in health‐related quality of life in the intervention group calculated as the SMD was 0.39 lower (0.9 lower to 0.12 higher) | ‐ | 300 | ⊕⊕⊝⊝ | The effect on quality of life ranges from a moderate effect on quality of life in favour of moisturisers to no difference between the groups. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Abramovits 2008, Belloni 2005, Boguniewicz 2008, Lodén 2002, Nebus 2009. 2Downgraded one level for inconsistency (I² = 95%), which was in part caused studies by studies at risk for attrition bias (Abramovits 2008 and Boguniewicz 2008). 3Downgraded one level for serious imprecision (wide confidence interval). 4Abramovits 2008, Belloni 2005, Nebus 2009. 5Downgraded one level for serious inconsistency (I² = 83%). All heterogeneity was removed when a study at high risk of bias was excluded (Abramovits 2008); we did not double count for risk of bias. 6Downgraded one level for serious imprecision (CI interval includes no effect (1) and appreciable benefit (1.25)). 7Abramovits 2008, Belloni 2005, Boguniewicz 2008, Boralevi 2014, Gayraud 2015, Grimalt 2007, Korting 2010, Lodén 2002, Patrizi 2008, Tan 2010. 8Downgraded one level for imprecision (CI interval included no difference (1) and appreciable harm (1.25)). 9Abramovits 2008, Belloni 2005, Boguniewicz 2008, Boralevi 2014, Gayraud 2015, Giordano‐Labadie 2006, Grimalt 2007, Korting 2010, Nebus 2009, Patrizi 2008, Patrizi 2014, Tan 2010. 10We did not downgrade for inconsistency as all sensitivity analyses show a clear positive effect of moisturisers. 11Abramovits 2008, Boguniewicz 2008, Gayraud 2015, Patrizi 2008, Weber 2015, Wirén 2009. 12Downgraded one level for serious inconsistency (I² = 73%), which was caused by the studies at unclear to high risk of bias showing better results. 13Giordano‐Labadie 2006, Grimalt 2007. 14Downgraded one level for serious inconsistency (I² = 68%). In the study of Giordano‐Labadie 2006, far more topical corticosteroids were used and the difference between the two arms was much larger. 15Gayraud 2015, Giordano‐Labadie 2006, Grimalt 2007. 16We did not downgrade for risk of bias, as, although there was attrition bias in Grimalt 2007, it did not impact the overall result, and even reduced the direction of effect. 17Downgraded one level for serious inconsistency (I² = 79%), it might have no effect at all, signal around 0. 18Downgraded one level for serious imprecision (CI includes moderate effect in favour of moisturisers as well as no difference). | ||||||
| Licochalcone‐containing moisturiser versus hydrocortisone acetate1% cream for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with hydrocortisone acetate 1% cream | Risk with licochalcone‐containing moisturiser | |||||
| Change from baseline in disease severity according to participants ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Number of participants who rated treatment satisfaction as good to excellent | ‐ | ‐ | ‐ | 30 | ⊕⊕⊝⊝ | On both treatment sides, 22/30 participants rated their satisfaction good to excellent with no difference between either side. |
| Number of participants reporting an adverse event | ‐ | ‐ | ‐ | 18 | ⊕⊕⊝⊝ | Both Udompataikul 2011 and Wanakul 2013 reported no adverse events on any side during the study. Side effects in Angelova‐Fischer 2014 (within‐participant study) were skin tightness, itch, and scaling on both sides. 9 side effects were reported on each forearm (n = 18). |
| Change from baseline in disease severityas assessed by the investigators | The mean disease severity as assessed by the investigators ranged from ‐3.50 to ‐21.29 | The mean disease severity as assessed by the investigators in the intervention group was 0.08 higher (1.96 lower to 2.13 higher) | ‐ | 96 | ⊕⊕⊝⊝ | There was no statistically significant difference between the 2 treatments, which is in accordance with the data for participant satisfaction. |
| Number of participants who experienced a flare | ‐ | ‐ | ‐ | 30 | ⊕⊕⊝⊝ | 3/30 experienced a flare on the side treated with licochalcone and 6/30 on the contralateral side treated with hydrocortisone acetate 1%. |
| Change in use of active topical treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in quality of life ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Within‐participant study Udompataikul 2011. 2We did not downgrade for detection bias as the participants were not blinded, but they considered treatments equally satisfactory. 3Downgraded two levels for serious imprecision (very small sample size). 4Within‐participant study Angelova‐Fischer 2014. 5Not downgraded for risk of bias (participants in Angelova‐Fischer 2014 and Udompataikul 2011 were not blinded) as there was no difference between the both treatment arms regarding reporting adverse events. 6Downgraded two levels for very serious imprecision (very small sample size). 7We did not downgrade for detection bias as investigators were blinded. 8Downgraded two levels for very serious inconsistency (I² = 85%); it could benefit both treatments. We therefore did not downgrade further for imprecision. Differences in study duration, and, in Angelova‐Fischer 2014, only forearms were treated. 9Downgraded one level for serious imprecision (small sample size and as we downgraded for risk of bias, we only downgraded once for imprecision for this outcome). 10Downgraded one level for risk of bias (no blinding of participants). | ||||||
| Vehicle treatment + daily moisturiser compared to fluticasone propionate twice weekly + daily moisturiser for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with fluticasone propionate twice weekly + moisturiser | Risk with vehicle treatment + moisturiser | |||||
| Change from baseline in disease severity as assessed by the participants (number of participants reporting good to excellent result) | Study population | RR 0.45 | 348 | ⊕⊕⊕⊕ | NNTB = 3, 95% CI 2 to 3, in favour of fluticasone propionate twice weekly + daily moisturiser | |
| 71 per 100 | 32 per 100 | |||||
| Participant satisfaction ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Number of participants reporting an adverse event | Study population | RR 0.51 | 718 | ⊕⊕⊝⊝ | Although there was a trend favouring the vehicle treatment + daily moisturiser, the 2 comparisons of Berth‐Jones 2003 implied that they might be equally safe (no adverse events in either group). | |
| 22 per 100 | 11 per 100 | |||||
| Change from baseline in disease severityas assessed by the investigators | 75 (1 RCT)5 | ⊕⊕⊕⊝ | There were reporting inconsistencies in the paper between the data table and text regarding the increase in SCORAD in the twice‐weekly fluticasone propionate + daily moisturiser group. These were reported as 7.1 in the table and as 3.8 in the text. In the vehicle + daily moisturiser group, the increase was 12.2 in both table and text. | |||
| Number of participants who experienced a flare | Study population | RR 2.17 | 718 | ⊕⊕⊕⊝ | NNTB = 3, 95% CI 2 to 3. Twice‐weekly fluticasone propionate combined with moisturiser resulted in fewer flares than moisturiser alone. HR of rate of flare 3.69, 95% CI 1.80 to 7.55 in favour of fluticasone propionate twice weekly + daily moisturiser | |
| 28 per 100 | 61 per 100 | |||||
| Change in use of topical active treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in quality of life ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 2Berth‐Jones 2003 (two comparisons), Glazenburg 2009, Hanifin 2002. 3Downgraded one level for serious inconsistency (I² = 67%); as there were no adverse events in both comparisons in Berth‐Jones 2003 in both treatment arms, they could be equally safe. 4Downgraded one level for serious imprecision (CI includes appreciable benefit and no difference). 5Glazenburg 2009 (See 'Comments'). 6Downgraded one level for serious imprecision (small sample size). 7Downgraded one level for serious inconsistency (I² = 72%). | ||||||
| Topical active treatment in combination with moisturiser compared to topical active treatment alone for eczema | ||||||
| Patient or population: people with eczema | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Quality of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with active treatment alone | Risk with active treatment in combination with moisturiser | |||||
| Change from baseline in disease severity according to participants ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Participant satisfaction | ‐ | ‐ | 201 | ⊕⊕⊝⊝ | Hanifin 1998: 96% of 78 preferred the combination treatment and just 4% the active treatment 'only'. Simpson 2011: 84.3% to 96.7% of 123 felt that the addition of the RestoraDerm to the routine use of their topical steroids "reduces inflammation, relieves dry and itchy skin, provides long lasting hydration, leaves skin protected and maintains healthy skin". | |
| Number of participants reporting an adverse event Follow‐up: mean 3 weeks | Study population | RR 0.39 | 125 | ⊕⊝⊝⊝ | Draelos 2008: no adverse events. Hanifin 1998 (within‐participant): 10 participants reported burning and stinging on the side treated with desonide 0.05% combined with moisturiser versus 11 on the other side treated with only desonide 0.05%. | |
| 16 per 100 | 6 per 100 | |||||
| Change from baseline in disease severity as assessed by the investigators | ‐ | The mean change from baseline in disease severity as assessed by the investigators in the intervention group calculated as the SMD0.87 lower (1.17 lower to 0.57 lower) | ‐ | 192 | ⊕⊕⊕⊝ | According to the assessments of the investigators, adding a moisturiser to topical active treatment is more effective than topical active treatment alone. |
| Number of participants who experienced a flare | Study population | RR 0.43 | 105 | ⊕⊕⊝⊝ | Adding a moisturiser to active treatment reduced the number of flares (NNTB = 6, 95% CI 3 to 57). | |
| 31 per 100 | 13 per 100 | |||||
| Change in amount of use topical active treatment ‐ not measured | ‐ | ‐ | ‐ | ‐ | ‐ | This outcome was not assessed in any of the studies. |
| Change from baseline in health‐related quality of life | The mean change from baseline in health‐related quality of life ranged from ‐2.07 to ‐3.17 | The mean change from baseline in health‐related quality of life in the intervention group was 1.31 lower (2.7 lower to 0.09 higher) | ‐ | 67 | ⊕⊕⊝⊝ | The study duration of 3 weeks was short; there was no difference in changes from baseline in quality of life between the 2 treatment groups. Results of DFI confirmed this (MD ‐1.03, 95% CI ‐2.47 to 0.42) |
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1Within‐participant design. 3Downgraded one level for serious risk of detection bias (no blinding of participants). 4Downgraded one level for serious indirectness as in both studies satisfaction was not really assessed. 5Wu 2014. 6Downgraded one level for risk of bias. Hanifin 1998 and Wu 2014 were assessed as being at a high risk of bias. 7Downgraded one level for serious indirectness for different reporting on adverse events including outcome definitions. 8Downgraded one level for serious imprecision (small sample size, too few adverse events, and wide CI). 9Msika 2008 (2 comparisons), Wu 2014. 10Downgraded one level for risk of bias (no blinding of outcome assessors). 11Downgraded one level for risk of bias (attrition bias (17/62) in the control group). 12Downgraded one level for serious imprecision (small sample size). 13Msika 2008 (2 comparisons). 14Downgraded one level for risk of bias (no blinding of participants). | ||||||
| Term | Definition |
|---|---|
| Adverse events | Unwanted side effects of using medication |
| Allergic contact dermatitis | A form of eczema after contact with a substance (an allergen) that produces (elicits) an immune‐mediated response in the skin |
| Allergic rhinitis | 'Hay fever': inflammation of the nose caused by allergens such as house dust mite, animals, pollen. Symptoms include sneezing, itchiness in the nose, watery eyes, runny or blocked nose |
| Ameliorate | Improve, to make something (such as a problem) better |
| Atopy | The individual's genetic predisposition to develop allergic reactions such as eczema, allergic rhinitis and asthma. Atopy often involves production of IgE antibodies against allergens such as, for example, house dust mite, animals, grass and tree pollen, and food proteins. |
| Bacteria | Also referred to as germs, bacteria are tiny micro‐organisms that are invisible to the eye. They are found everywhere and can be harmful, e.g. causing infections, or helpful, e.g. aiding digestion of food |
| Ceramides | Lipid (fatty) molecules found in the lipid bilayer of the intercellular matrix (see `Intercellular lipid matrix' below) |
| Colonisation | The point at which an Infection begins, when an organism successfully enters the body, grows and multiplies |
| Control | The alternative treatment, placebo, or absence of treatment against which the intervention of interest in the review is compared |
| Corneodesmosomes | Any of a class of proteins that hold corneocytes (cells in the epidermis, or outer layers of the skin) together; their degradation leads to desquamation (see 'Desquamation' below) |
| DASI | Dry skin area and severity index: a tool used to evaluate dryness and severity of dry skin (Serup 1995) |
| Desquamation | Skin peeling |
| Dizygotic | Non‐identical twins, i.e. twins formed from two different eggs fertilised by separate sperm cells, are referred to as dizygotic. |
| DLQI | Dermatology Life Quality Index: an assessment tool to evaluate the impact of eczema and its treatment on quality of life (Finlay 1994) |
| EASI | Eczema Area and Severity Index: a tool used to measure the extent (area) and severity of eczema (Hanifin 2001) |
| Emollients | The terms `emollients' and `moisturisers' are often used interchangeably. But, since 'emollient' sometimes refers to a specific ingredient that soothes the skin, it is more appropriate to use the term 'moisturiser'. Emollients are included within ointments, creams, lotions, gels, bath oils and sprays, and are used to keep the skin soft and supple and reduce scaling. Application to the skin reduces water loss by covering it with a protective film. They can be used frequently and might ease itching |
| Epidermis | The outermost layers of cells in the skin which consist mainly of keratinocytes that mature to become corneocytes |
| Exacerbation | Periods of worsening the symptoms and signs of eczema |
| Excoriation | Abrasion, scratched skin |
| Extensor | The opposite site of a flexure point, i.e. the outer side of, for example elbow, knee or wrist |
| Filaggrin | An epidermal barrier protein |
| Flare | Periods of worsening of eczema symptoms and signs, or escalation in use of medication (Thomas 2015) |
| Flexural dermatitis | Eczema at the flexure points (inner sides) of elbow, knees, wrists, groin and armpits |
| Gene | Part of DNA that encodes a protein involved in body function |
| Genome‐wide linkage study | An established tool to map inherited diseases |
| Humectant | Substance or product that is 'water loving' and draws water towards it |
| Hygroscopic | Absorbing water |
| Hypersensitivity | An exaggerated immune response toward an allergen (for example pollen, house dust mite, but also for contact allergens such as nickel and fragrances) |
| IgE (immunoglobulin E) | A class of antibody that is important in defence against parasitic disease, and plays a key role in the disease process of allergic diseases. People with eczema often have an increased level of IgE in their blood |
| Immune response | The process through which the body identifies and defends itself against bacteria, viruses and other harmful agents |
| Intercellular lipid matrix | Space surrounding corneocytes with stacked layers of lipids |
| Lesion | A region or area of damaged skin |
| Lesional | Concerning lesions, or accompanied by lesions |
| Lichenification | Skin thickening |
| Moisturisers | Ointments, creams, lotions, gels, bath oils and sprays that are used to keep the skin soft and supple and reduce scaling. Application to the skin reduces water loss and covers it with a protective film. Moisturisers can be used frequently and might ease itching. |
| Monozygotic | Identical twins, i.e. twins who develop from a single fertilised egg (zygote) that splits to form two identical embryos, are referred to as monozygotic (in contrast to dizygotic twins (see above)). |
| Objective | Something observed and verified by physician or investigator by visible physical signs or laboratory tests (i.e. based on facts, not emotions or feelings) |
| Objective‐ SCORAD | Objective ‐ SCORing Atopic Dermatitis is a clinical scoring system, that uses the SCORAD system and excludes subjective symptoms, which cannot be measured accurately, such as daytime itching (pruritus) and sleep loss (Kunz 1997) |
| Occlusive | Describes an agent or process that seals something off |
| Over‐the‐counter (OTC) medicines | Medicines that can be bought without a prescription |
| Papulovesicular | Relating to an eruption of papules (clearly defined (circumscribed), solid elevations of skin with no visible fluid) and vesicles (small fluid‐filled sacs on the skin) |
| Pathogenesis | Origin of disease and how it develops |
| Pathognomonic biomarker | A specific indicator for a disease |
| Photochemotherapy (PUVA) | PUVA is a combination treatment of a drug (psoralen) with ultraviolet A (UVA) light. The psoralen makes the skin temporarily more sensitive to the ultraviolet light |
| Phototherapy | Treatment with ultraviolet light (UVB or UVA) |
| Placebo | A 'dummy' or fake medicine that has no expected benefit. In this review placebo means, in accordance with the terminology used by the investigators, a moisturiser without the ingredient considered to be the most beneficial, and so, of a different composition than the moisturiser being studied. Use of placebo treatments allows patients and staff to be blinded, as the placebo and active treatments appear the same, so it is impossible to tell which has been used. |
| POEM | The Patient Oriented Eczema Measure is a self‐assessment tool for monitoring eczema severity, based on signs and symptoms (Charman 2004) |
| PO‐SCORAD | The Patient‐Oriented SCORing Atopic Dermatitis (PO‐SCORAD) index is a self‐assessment score for patients to evaluate their eczema, based on subjective and objective criteria from the SCORAD (see also SCORAD below) (Stalder 2011) |
| Preservative | A natural or synthetic ingredient added to products such as foods, pharmaceuticals, paints, biological samples, wood, etc. which help to prevent decomposition caused by microbial growth or by undesirable chemical changes |
| Propylene glycol | Propylene glycols attract water and by enhancing skin penetration they behave as moisturisers to improve the appearance of the skin |
| Protease | An enzyme that breaks down proteins (via proteolysis) |
| Pruritus | Itch |
| Quality of life | The general well‐being of individuals and societies. Health‐Related Quality of Life (HRQoL) looks at quality of life in relation to health |
| QoLIAD | Quality of Life Index for Atopic Dermatitis (QoLIAD). An assessment tool to evaluate the impact of eczema and its treatment on quality of life (Whalley 2004) |
| Remission | A temporary or permanent decrease or absence of the symptoms and signs of disease activity |
| Sensitisation | Exposure to an allergen that results in the development of hypersensitivity, i.e. an increased or disproportionate response to the allergen |
| SCORAD‐index | An assessment tool used by clinicians to evaluate the extent and severity of eczema (SCORing Atopic Dermatitis) (European Task Force on Atopic Dermatitis 1993) |
| Staphyloccocus aureus | A type of bacterium that is often found on the skin |
| Subjective | Something experienced by the participant not perceived by the investigator or physician |
| TEWL | Trans‐epidermal water loss (TWL or TEWL) is the quantity of water that diffuses through and evaporates from the epidermis |
| Topical corticosteroid | Corticosteroids applied to the skin; these are effective in controlling inflammation and used to treat eczema and many other skin conditions |
| Urea | Urea absorbs water, helps to reduce the amount of water lost though the skin and increases skin penetration of other substances. It softens the horny layer and also has anti‐itch (anti‐pruritic) properties. |
| Vehicle | In this review 'vehicle' means a moisturiser that has the same composition as the studied moisturiser, but lacks the ingredient that is considered to be the most beneficial |
| Volar | The inside surface of the forearm, i.e. the same side as the palm of the hand |
| Xerosis | Dry skin |
| Study ID | Response | Additional | Comment |
|---|---|---|---|
| Emails sent: 26 June 2014, 7 February 2016, 12 February 2016, 26 February 2016, 12 March 2016, 19 March 2016, to [email protected] Regarding allocation concealment and method of blinding Replies received: 24 March 2016 with responses; 29 March 2016 with additional information | Yes | ‐ | |
| Emails sent: 13 February 2016, 26 February 2016, to irena.angelova‐fischer@uk‐sh.de Regarding sequence generation and allocation concealment Reply received: 1 March 2016 with response to sequence generation Several emails sent regarding allocation concealment, but this remains unclear | Yes | ‐ | |
| [email protected] no need to contact, but this is recent email for future update of the review | Not applicable | ‐ | |
| Email sent: 13 February 2016, to [email protected] Regarding sequence generation and allocation concealment Email received: 15 February 2016 with responses | Yes | ‐ | |
| Emails sent: 13 February 2016, 26 February 2016, to [email protected] Regarding method of blinding Email received: 27 February 2016 with responses | Yes | ‐ | |
| Emails sent: 12 May 2014, 14 June 2014, 21 January 2016, to [email protected] and [email protected] Regarding sequence generation, allocation concealment and method of blinding Replies received: 2 February 2015 and 15 February 2016; received all responses and additional material | Yes | ‐ | |
| Emails sent: 14 February 2016, 26 February 2016, 12 March 2016, 19 March 2016, to [email protected]. Also did not reply to questions about the Abramovits 2008 study Regarding allocation concealment and method of blinding, EASI scores at day 43 and subjects/care givers assessment of global response at day 43 in vehicle group No reply received | Not applicable | ‐ | |
| Emails sent: 15 February 2016, 26 February 2016, 12 March 2016, 26 March 2016, to franck.boralevi@chu‐bordeaux.fr Regarding P‐VAS scores, SCORAD, Objective SCORAD, and HI at day 28 as well as SDs Replies received: 31 March 2016 and 5 April 2016 responses and additional information | Yes | ‐ | |
| Emails sent: May 2014, 15 February 2016, to elsner@derma‐jena.de and [email protected] Regarding sequence generation, allocation concealment and method of blinding, mean SCORAD, TEWL, capacitance and SDs Replies received: 16 May 2014 and 15 February 2016, received responses and additional information | Yes | ‐ | |
| Emails sent: 15 February 2016, 26 February 2016, 15 March 2016, 19 March 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, TEWL values and SDs after 2 weeks, number of dropouts, numbers of male/female and age of participants Reply received: 22 March 2016 with responses to our questions | Yes | ‐ | |
| Email sent: 16 February 2016 to [email protected] and clarence.de‐[email protected] Regarding allocation concealment and method of blinding Replies received: 24 February 2016 and 12 March 2016 from [email protected] and [email protected], with responses to our questions | Yes | ‐ | |
| Emails sent: 16 February 2016, 12 March 2016, 19 March 2016, 26 March 2016, to [email protected] Regarding allocation concealment and method of blinding No reply received | Not applicable | ‐ | |
| Emails sent: 19 February 2016, 12 March 2016, 19 March 2016, 26 March 2016, to [email protected] Regarding allocation concealment and method of blinding, more precise baseline data and data at end of study means and SDs No reply received | Not applicable | ‐ | |
| Emails sent: 19 February 2016, 12 March 2016, 19 March 2016, 26 March 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, precise baseline data and data at end of study, means and SDs, Participant assessments of target site skin appearance for redness, peeling, dryness, stinging/burning, and overall skin irritation, sponsoring and declaration of interest No reply received | Not applicable | ‐ | |
| Emails sent: 5 March 2016, 12 March 2016, 19 March 2016, 26 March 2016 to Jason Emer (email not current anymore), with last 2 emails to A Frankel ([email protected]) Regarding sequence generation, allocation concealment, method of blinding, colour version of pdf and SDs at week 4 No reply received | Not applicable | ‐ | |
| Emails sent: 7 March 2016, 12 March 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, more precise data Reply received: 12 March 2015 "it is so long time ago that I performed this study so I do not remember. I retired last November 2015 and I do not have assess to any data now." | No | ‐ | |
| Email sent 9 March 2016 to [email protected] Regarding sequence generation, allocation concealment, method of blinding and incomplete data Reply received: 14 March 2014 with responses to our questions | Yes | ‐ | |
| Email sent: 13 May 2014 to giordano.labadie.f@chu‐toulouse.fr, frederic.cambazard@chu‐st‐etienne.fr, gerard.guillet@chu‐poitiers.fr, [email protected], and valerie.mengeaud@pierre‐fabre.com Regarding sequence generation and allocation concealment Reply received: 26 May 2014 with responses to our questions | Yes | ‐ | |
| Emails sent:14 March 2016, 19 March 2016, 26 March 2016, 3 April 2016, to [email protected] Regarding allocation concealment, method of blinding and inconsistencies in text page 64 and table 1 No reply received | Not applicable | ‐ | |
| Email sent: 13 April 2014, to [email protected], frederic.cambazard@chu‐st‐etienne.fr, valerie.mengeaud@pierre‐fabre.com Regarding sequence generation and allocation concealment Reply received: 24 June 2015, after several emails we received responses to everything we asked for | Yes | ‐ | |
| Email sent: 17 April 2016, to [email protected] (no longer correct, no more recent email, so sent again to [email protected], but this is also no longer correct) Regarding sequence generation, allocation concealment and baseline data for TEWL and corneometry No recent email addresses could be found | Not applicable | ‐ | |
| We could not find a current email address of any of the authors listed | Not applicable | ‐ | |
| Emails sent: 18 March 2016, 26 March 2016, 3‐4 February 2016,10 April 2016, to [email protected] Regarding sequence generation and allocation concealment, frequency of use, precise data after 3 months (means and SDs), more data on adverse events No reply received | Not applicable | ‐ | |
| Emails sent: 19 March 2016, 26 March 2016, 3 April 2016, to [email protected] Regarding allocation concealment, method of blinding, exact data for POEM and quality of life at baseline, and SCORAD, POEM and quality of life at day 28 | Not applicable | ‐ | |
| Emails sent: 8 January 2016, 24 January 2016, 12 February 2016, 8 April 2016, 18 April 2016, to [email protected] Regarding sequence generation and allocation concealment No reply received | Not applicable | ‐ | |
| Email sent: 19 March 2015, to [email protected] Regarding full text publication Reply received: 21 March 2015, Principal investigator could not remember anymore | No | ‐ | |
| Email sent: 19 March 2015, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, exact TEWL values and corneometry with SDs at week 4 Reply received: 21 March 2016 with responses to some of our questions, the rest was no longer accessible | Yes and No | We did not receive exact data | |
| Email sent: 20 March 2016, to [email protected]‐muenchen.de Regarding allocation concealment and blinding Email address is no longer correct, and further searches showed that he died in 2012. None of the other authors could be contacted | Not applicable | ‐ | |
| Emails sent: 20 March 2016, 26 March 2016, to a‐[email protected], [email protected] [email protected] Regarding sequence generation, allocation concealment, method of blinding and precise study data Reply received: 26 March 2016 with responses to our questions | Yes | ‐ | |
| Emails sent: 15 April 2016, 19 April 2016, 23 April 2016, 30 April 2016, to [email protected] Regarding allocation concealment and blinding Reply received: 2 May 2016 with responses to our questions | Yes | ‐ | |
| Email sent: 14 May 2014, to [email protected]. Latest email address is [email protected] (since October 2015) Regarding sequence generation and allocation concealment Reply received: 14 May 2014 with responses to our questions | Yes | ‐ | |
| Email sent: 15 May 2014, to [email protected], clarence.de‐[email protected], [email protected] Regarding sequence generation, stratifying, allocation concealment, means and SDs for some outcomes and clarification about data losses for the following outcomes in particular IDQOL, DFI, IGE Reply received: 30 June 2014 with additional information | Yes | ‐ | |
| Email sent: 30 June 2014, to [email protected] and [email protected] Regarding clarification of correct citation i.e. author string/journal page numbers to the 2 citations listed in Included studies, plus sequence generation, allocation concealment and method of blinding Reply received: 30 June 2014 and 5 July 2014, with responses to our questions | Yes | ‐ | |
| Emails sent: 2 January 2016, 5 January 2016, 15 January 2016, 20 January 2016, to [email protected], [email protected], [email protected] Regarding missing data of control, and clarification regarding study design, and data regarding EASI, IGA Repy received: 20 January 2016 from Dr Nebus with responses to our questions and additional documents | Yes | ‐ | |
| Email sent: 7 January 2016, to [email protected] Regarding sequence generation and allocation concealment Reply received: 18 January 2016, from [email protected], with responses to our questions | Yes | ‐ | |
| Email sent: 26 March 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, dropouts, EASI, VAS scores at baseline and day 21 and additional study details Replies received: 26 March 2016, and 30 March 2016 with responses to our questions | Yes | ‐ | |
| Emails sent: 26 March 2016, 3 April 2016, 10 April 2016, 16 April 2016, to Dr Seo: [email protected] Regarding sequence generation, allocation concealment, method of blinding and precise data of IGA, VAS, TEWL and corneometry with SD at 4 weeks No reply received | Not applicable | ‐ | |
| Emails sent: 27 March 2016, 3 April 2016, 10 April 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding and precise data of EASI at day 43, and data of appraisal patients Reply received: 12 April 2016, with responses regarding sequence generation, allocation concealment and blinding, but not the rest. 14 April 2016 we sent mails to the sponsor at [email protected] (no longer working) and [email protected] No further details retrieved | In part | ‐ | |
| Email sent 27 March 2016, 3 April 2016 jennifer.theunis@pierre‐fabre.com. Regarding sequence generation, allocation concealment, and precise data of TEWL at day 43, and to which group drop‐out was randomised Response 7 April 2016 we received responses to our questions | Yes | ‐ | |
| Email sent 28 March 2016, 3 April 2016, 10 April 2016, 16 April 2016 [email protected] Regarding allocation concealment, precise data at baseline,day 10 and 28 for TEWL, PGA, EASI, adverse events in Cis UCA group and in vehicle No reply received | Not applicable | ‐ | |
| Email sent: 29 March 2016, to [email protected]‐bonn.de Regarding sequence generation, allocation concealment, method of blinding as one group had two different tubes, precise data at baseline and week 16 for VAS, EASI, DLQI and CDLQI Reply received: 30 March 2016, saying that cannot help us, the study was too long ago, and Schering Dermatology does not exist anymore. Intendis (the pharmaceutical company) did not reply | No | ‐ | |
| Email sent: 2 April 2016, to [email protected], [email protected] Regarding sequence generation, allocation concealment, differences between publication and protocol and exact baseline values for data after 15, 30 and 60 minutes for TEWL, and corneometry Reply received: 12 April 2016, with responses to our questions | Yes | ‐ | |
| Email sent: 18 May 2014, to [email protected], SOTIRIOS‐[email protected] Regarding sequence generation and allocation concealment, mean reduction in EASI and corresponding standard deviations at day 28, and dropouts Reply received: 18 July 2014, with responses to our questions | Yes | ‐ | |
| Emails sent: 2 April 2016, 10 April 2016, 16 April 2016, 23 April 2016, to [email protected], [email protected], [email protected] Regarding exact baseline values (and SD) at day 28 for dryness scale, TEWL, and corneometry Reply received: 2 May 2016, with responses to our questions and additional information | Yes | ‐ | |
| Email sent: 3 April 2016, to [email protected] Regarding sequence generation, allocation concealment, method of blinding, precise data at baseline and day 28 for IGA and patient/family self‐assessments Reply received: 4 April 2016, with responses to our questions and additional information | Yes | ‐ | |
| Email sent: 14 May 2014, to [email protected], Adam Reich Regarding sequence generation and allocation concealment | Yes | Quasi randomised, exclude | |
| Emails sent: 3 April 2016, 10 April 2016, 16 April 2016, 23 April 2016, 30 April 2016, to [email protected]‐u.ac.jp Regarding sequence generation and allocation concealment, and SCORAD (mean and SD) at start maintenance phase and at day 28 No reply received | Not applicable | ‐ | |
| Emails sent: 3 April 2016, 10 April 2016, to [email protected] Regarding allocation concealment Reply received: 11 April 2016, with response to our question | Yes | ‐ | |
| Email sent: 8 April 2016, to [email protected] Regarding allocation concealment Reply received: 8 April 2016, with response to our question | Yes | ‐ | |
| Emails sent: 2 July 2014, 7 July 2014, 24 January 2016, 12 February 2016, 8 April 2016, 18 April 2016, to [email protected] ([email protected] is no longer in use) Regarding sequence generation, allocation concealment and standard deviations (or SEM) of the SCORAD at baseline, week 2, week 4 and week 6? Reply received: 18 April 2016, with some data but no information regarding sequence generation, or allocation concealment and there were no data for week 6 | In part | ‐ | |
| Email sent: 10 April 2016, to tweberQbdfusa.com Regarding sequence generation and allocation concealment | Yes | ‐ | |
| Email sent: 15 April 2016, to [email protected], karin.wiren@omega‐pharma.se Regarding sequence generation and allocation concealment | Yes | ‐ | |
| Emails sent: 5 January 2016, 15 January 2016, 31 January 2016, 12 February 2016, 10 April 2016, to [email protected] Regarding sequence generation and allocation concealment No reply received | Not applicable | ‐ | |
| Emails sent: 10 April 2016, 16 April 2016, to [email protected] Regarding allocation concealment, method of blinding, SCORAD values at start maintenance phase and follow‐up, item EQ‐5D Reply received: 18 April 2016, with responses to our questions | Yes | ‐ | |
| Abbreviations CDLQI: Children’s Dermatology Life Quality Index | |||
| MOISTURISER VERSUS VEHICLE, PLACEBO OR NO TREATMENT | ||||
|---|---|---|---|---|
| Moisturisers versus no treatment (i.e. no moisturiser) | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Exomega lotion (oat) twice daily 6 weeks | No treatment for 6 weeks | No | Moderate‐ and high‐potency corticosteroids allowed | |
| Exomega moisturising milk (oat) twice daily for 2 months | No treatment for 2 months | No | Moderate‐ and high‐potency corticosteroids allowed | |
| Eucerin Eczema Relief body cream (oat and licochalcone) once a day for 6 months + cleanser | Only cleanser for 6 months | No | Eucerin Eczema Relief Instant Therapy allowed for active lesions, | |
| Canoderm (urea 5%) twice daily 6 months | No treatment 6 months | No | Permitted only at areas other than target lesion | |
| Cethaphil Restoraderm Body moisturiser (ceramide precursors etc) twice daily 27 days | No treatment 27 days | Yes | No | |
| Emollient balm twice daily for 28 days | Hygiene product for 28 days | No | Not mentioned | |
| Atopiclairversus vehicle | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Atopiclair three times daily for 50 days | Vehicle three times daily for 50 days | No | No | |
| Atopiclair three times daily for 21 days | Vehicle three times daily for 21 days | No | Oral medication continued | |
| Atopiclair three times daily for 43 days | Vehicle three times daily for 43 days | No | If really needed, low‐potency topical corticosteroids allowed | |
| Patrizi 20083‐arm (1st comparison) As Atopiclair 'light' is not marketed and clearly less effective, it is not included in the comparison Atopiclair versus vehicle. The comparison Atopiclair 'light' versus vehicle (2nd comparison) will therefore not be further discussed | Atopiclair three times daily for 43 days | Vehicle three times daily for 43 days | No | If really needed, low‐potency topical corticosteroid allowed |
| Other moisturisers versus vehicle or placebo | ||||
| Urea‐containing moisturisers | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Laceran (10% urea) twice daily for 4 weeks | Vehicle twice daily for 4 weeks | Yes | No | |
| Canoderm (urea 5%) twice daily for 6 months | No treatment for 6 months | No | Only at other areas than target lesion | |
| Laceran (10% urea) twice daily for 4 weeks | Vehicle twice daily for 4 weeks | Yes | No | |
| Lodén 20023 arm (1st comparison) | Urea saline 4% cream once daily for 30 days | Placebo (cream base) cream once daily for 30 days | No | Topical steroids allowed |
| Glycerol‐containing moisturisers | ||||
| Lodén 20023 arm (2nd comparison) | Glycerol 20% once daily for 30 days | Placebo (cream base) once daily for 30 days | No | Topical steroids allowed |
| Dexeryl (glycerol 15%) twice daily for 4 weeks | Vehicle without glycerol twice daily for 4 weeks | No | If really needed, moderate‐potency topical corticosteroid allowed | |
| Glycerol 20% twice daily for 4 weeks | Vehicle without glycerol twice daily for 4 weeks | Yes | No | |
| Oat‐containing moisturisers | ||||
| Exomega lotion (oat) twice daily for 6 weeks | No treatment for 6 weeks | No | Moderate‐ and high‐potency corticosteroids allowed | |
| Exomega moisturising milk (oat) twice daily for 2 months | No treatment for 2 months | No | Moderate‐ and high‐potency corticosteroids allowed | |
| Oatmeal based occlusive cream twice daily for 8 weeks | Occlusive vehicle for 8 weeks | No | Topical medications allowed | |
| Eucerin Eczema Relief body cream (oat and licochalcone) once a day for 6 months + cleanser | Cleanser only for 6 months | No | Eucerin Eczema Relief Instant Therapy was allowed for active lesions | |
| Remaining moisturisers versus vehicle or placebo | ||||
| Ammonium lactate 6% in water‐in‐oil emulsion twice daily for 4 weeks | Vehicle twice daily for 4 weeks | Yes | No | |
| Pale sulfonated 4% shale oil cream three times daily for 4 weeks | Vehicle three times daily for 4 weeks | No | No | |
| Atoderm Intensive cream twice daily for 6 months | Moisturiser base twice daily for 6 months | No | Topical corticosteroid and immunomodulators could be continued | |
| Triclosan 1% moisturiser twice daily for 41 days | Vehicle cream twice daily for 41 days | No | Low‐potency corticosteroid allowed | |
| Thumm 20003 arm (1st comparison) | Hippophae rhamnoides 10% cream for 4 weeks | Placebo cream for 4 weeks | No | No |
| Thumm 20003 arm (2nd comparison) | Hippophae rhamnoides 20% cream for 4 weeks | Placebo cream for 4 weeks | No | No |
| Oils versus placebo | ||||
| Gehring 1999study 1 | Primrose oil amphilic o/w emulsion twice daily for 4 weeks | Placebo oil twice daily for 4 weeks | Yes | No |
| Gehring 1999study 2 | Primrose oil amphilic w/o emulsion twice daily for 4 weeks | Placebo oil twice daily for 4 weeks | Yes | No |
| Camellia oil spray for 2 weeks | Purified water spray for 2 weeks | No | All allowed without changing | |
| ONE MOISTURISER VERSUS ANOTHER MOISTURISER | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Patrizi 20083 arm (3rd comparison) | Atopiclair three times daily for 43 days | Atopiclair 'light' three times daily for 43 days | No | If really needed, low‐potency topical corticosteroid allowed |
| Miller 20113 arm (1st comparison) | Atopiclair three times daily for 3 weeks | EpiCeram three times daily for 3 weeks | No | No |
| Miller 20113 arm (2nd comparison) | Atopiclair three times daily for 3 weeks | Aquaphor (petrolatum 41%, glycerol, lanolin etc.), three times daily for 3 weeks | No | No |
| Miller 20113 arm (3rd comparison) | EpiCeram (high ceramides) three times daily for 3 weeks | Aquaphor three times daily for 3 weeks | No | No |
| EpiCeram twice daily for 4 weeks | Hyalotopic (hyaluronic acid, glycerol, propylene glycol etc) twice daily for 4 weeks | Yes | No | |
| EpiCeram twice daily for 3 weeks | Oatmeal‐containing cream twice daily for 3 weeks | No | No | |
| EpiCeram for 4 weeks | Eucerin for 4 weeks | Yes | No | |
| MimyX + Eucerin twice daily for 12 weeks | Eucerin cream twice daily for 12 weeks | Yes | If really needed, low‐potency topical corticosteroid allowed | |
| Albolene twice daily for 4 weeks | MimyX twice daily for 4 weeks | Yes | Low‐potency topical corticosteroid allowed | |
| Fredriksson 1975(2 studies) | Aquacare twice daily for 4 weeks | Calmurid twice daily for 4 weeks | Yes | No |
| Locobase repair twice daily for a year | Atoderma twice daily for a year | No | Moderate‐potency topical corticosteroid allowed | |
| Canoderm (urea 5%) twice daily for 6 months | Miniderm (no urea) twice daily for 6 months | No | No | |
| Urea 5% moisturiser twice daily for 6 weeks | Urea 10% lotion for 6 weeks | No | Topical steroids allowed | |
| Urea 4% + NaCl in o/w twice daily for 2 weeks | Urea 4% in o/w twice daily for 2 weeks | Yes | No | |
| Lodén 20023 arm (3rd comparison) | Glycerol 20% once daily for 30 days | Urea saline 4% cream once daily for 30 days | No | Topical corticosteroids allowed |
| Propyless (20% propylene glycol) twice daily for 2 weeks | Fenuril (urea 4% and NaCl 4%) twice daily for 2 weeks | Yes | No | |
| Ceramide‐containing moisturiser twice daily for 6 weeks | Control moisturiser (?) twice daily for 6 weeks | No | Topical corticosteroids allowed | |
| Furfuryl palmitate‐enriched moisturiser twice daily for 2 weeks | Moisturiser twice daily for 2 weeks | No | No | |
| Pro‐AMP cream (rhamnosoft ceramides) twice daily for 4 weeks | Hydrating cream (glycerol, vaseline, paraffin twice daily for 4 weeks | No | No | |
| Thumm 20003 arm (3rd comparison) | Hippophae rhamnoides 10% cream for 4 weeks | Hipophae rhamnoides 20% cream for 4 weeks | No | No |
| Lactobacillus sakei‐containing moisturiser twice daily for 4 weeks | Control moisturiser for 4 weeks | Yes | Topical corticosteroids allowed | |
| Virgin coconut oil twice daily for 8 weeks | Mineral oil twice daily for 8 weeks | No | No | |
| Virgin coconut oil twice daily for 4 weeks | Virgin olive oil twice daily for 4 weeks | No | No | |
| Bleach bath with moisturiser on one occasion | Water bath with moisturiser on one occasion | No | No | |
| MOISTURISERS VERSUS ACTIVE TREATMENT | ||||
| Moisturisers versus topical corticosteroids | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Within‐participant studies comparing licochalcone containing moisturiser versus hydrocortisone | ||||
| O/W formulation containing licochalcone A (Glycyrrhiza Inflata root extract) twice daily for 1 week | Hydrocortisone cream twice daily for 1 week | Yes | No | |
| Licochalcone twice daily for 6 weeks | Hydrocortisone cream twice daily for 6 weeks | Yes | No | |
| Licochalcone twice daily for 4 weeks | Hydrocortisone cream twice daily for 4 weeks | Yes | No | |
| Parallel studies comparing moisturisers versus topical corticosteroids | ||||
| Stelatopia (2% sunflower oil distillate, fatty acids, ceramides) twice daily for 3 weeks | Hydrocortisone butyric propionate twice daily for 3 weeks | No | No | |
| EpiCeram twice daily for 4 weeks | Fluticasone 0.5% cream twice a day for 4 weeks | No | No (Cetaphil lotion applied to uninvolved lesions) | |
| 20% petrolatum in cetomacrogol + wet wrap for 4 weeks | Mometasone furoate 0.1% + wet wrap for 4 weeks | No | No | |
| w/o emulsion Excipial twice a day for 1 week | Hydrocortisone 1% in w/o emulsion (Excipial) twice daily for 1 week | No | No | |
| Moisturiser containing spent grain wax, spinose kernel oil, etc. twice a day for 4 weeks | Hydrocortisone 1% cream twice a day for 4 weeks | Yes | No | |
| Moisturiser (Advabase) twice a day for 16 weeks | Methylprednisolone aceponate cream 2 days a week, on other days used moisturiser twice a day for 16 weeks | No | No | |
| Moisturiser versus topical immunomodulators | ||||
| Eletone (high lipid) three times daily for 4 weeks | Pimecrolimus three times daily for 4 weeks | Yes | No | |
| Moisturiser therapy (?) for 4 weeks | Tacrolimus for 4 weeks | No | No | |
| Hyalotopic (ceramide) three times daily for 4 weeks | Pimecrolimus twice a day for 4 weeks | Yes | No | |
| VEHICLE + MOISTURISER VERSUS TOPICAL CORTICOSTEROID + MOISTURISER | ||||
| Berth‐Jones 20034 arm (1st comparison) | Vehicle cream twice weekly + moisturiser for 16 weeks | Fluticasone propionate 0.05% cream twice weekly + moisturiser for 16 weeks | No | No |
| Berth‐Jones 20034 arm (2nd comparison) We did not consider other possible comparisons of the 4 arms important for this review | Vehicle ointment twice weekly + moisturiser for 16 weeks | Fluticasone propionate 0.005% ointment twice weekly + moisturiser for 16 weeks | No | No |
| Vehicle twice a week + moisturiser for 20 weeks | Fluticasone propionate 0.05% cream twice weekly + moisturiser for 20 weeks | No | No | |
| Placebo ointment twice weekly + moisturiser for 16 weeks | Fluticasone propionate 0.005% ointment twice weekly + moisturiser for 16 weeks | No | No | |
| TOPICAL ACTIVE TREATMENT + MOISTURISER VERSUS TOPICAL ACTIVE TREATMENT ALONE | ||||
| Study | Intervention | Comparator | Within‐participant? | Active treatment allowed? |
| Draelos 20083 arm (1st comparison) | Fluocinonide 0.05% twice a day + ceramide cleanser+ moisturising cream for 4 weeks | Fluocinonide 0.05% twice a day plus cleansing bar for 4 weeks | No | No |
| Draelos 20083 arm (2nd comparison) 3rd possible comparison did not include moisturiser i.e. fluocinonide + cleansing bar vs fluocinonide + ceramide cleanser | Fluocinonide 0.05% twice a day + ceramide cleanser+ moisturising cream for 4 weeks | Fluocinonide 0.05% twice a day plus ceramide cleanser for 4 weeks | No | No |
| Moisturising and softening cream + flumethasone ointment twice a day for 3 weeks | Flumethasone ointment twice a day for 3 weeks | No | No | |
| Simpson 2011 study D | Restoraderm moisturiser twice a day + topical corticosteroids for 4 weeks | Routine use of topical corticosteroids for 4 weeks | Yes | No |
| Desonide 0.05% twice a day + three times daily moisturiser for 3 weeks | Desonide 0.05% cream twice a day for 3 weeks | Yes | No | |
| Msika 20085 arm (1st comparison) | Desonide 0.05% twice a day plus moisturiser + sunflower oil 2% twice a day for 21 days | Desonide 0.05% twice a day for 21 days | No | No |
| Msika 20085 arm (2nd comparison) We did not consider other possible comparisons of the 5 arms to be important for this review | Desonide 0.05% once daily plus moisturiser + sunflower oil 2% twice a day for 21 days | Desonide 0.05% once daily for 21 days | No | No |
| BoPao cream + 10% urea ointment once a day or twice a day for 2 weeks | BoPao cream only (antifungal/anti‐inflammatory cream) | No | No | |
| o/w: oil in water w/o: water in oil | ||||
| Study ID | Interventions & comparisons | N | Comments |
|---|---|---|---|
| 5% urea as active substance versus 4% urea and 4% NaCl | 50 | The data were reported in box‐and‐whisker plots, and no precise data were provided, too much estimation | |
| Emollient + fresh expressed milk versus moisturiser only | 9 | None of our outcomes were assessed | |
| Aqueous cream BP versus Oilatum Junior Bath additive | 38 | Poster with limited information. The principal investigator said, "The study itself was a purely mechanistic study, and not meant to provide clinical evidence" | |
| Nioleol (10% primrose oil, 8%‐9% γ‐linolenic acid) versus Uriage (borage oil and 24% γ‐linolenic acid) versus Atopic (35%‐40% γ‐linolenic acid) versus Atoderm control moisturiser once daily for 12 weeks | 23 | Unclear how many participants were randomised to each arm | |
| Proderm versus no treatment | 24 | No baseline data nor end value data were reported. The data were reported in box‐and‐whisker plots, and were not interpretable | |
| Oilatum Plus versus Oilatum Emollient | 30 | Unclear how many participants were randomised to each arm, inconsistencies in reporting of data | |
| Study 1 Emulsifying ointment with aqueous cream versus emulsifying ointment with baby oil Study 2 Cetomacrogol versus emulsifying ointment versus glycerol/petrolatum versus petroleum jelly | 120 | Frequency of use during day or week were not reported. There were quite some inconsistencies in text and figures. Study 2 reported no end data, just that all scores tended to decline. We mailed investigators numerous times to clarify study details, but received no response. | |
| Midpotency corticosteroid cream versus midpotency corticosteroid cream combined with a hydrolipid cream | 6 | Poster with limited information, principal investigator was not able to provide missing study details | |
| Glycerol 20% cream versus 4% urea and 4% NaCl | 110 | Unclear how many participants were randomised to each arm. The data all need to be estimated from box‐and‐whisker plots, too much estimation | |
| PPARα activator and ceramide versus moisturiser without these ingredients | 31 | Only 5 participants with eczema included, no individual patient data, not our prespecified outcomes | |
| Cis‐urocanic acid 5% emulsion cream versus control vehicle | 14 | Data provided need to be estimated from figures (for transepidermal water loss (TEWL)), or no precise data were provided other than that there were no significant differences. We mailed investigators numerous times to clarify study details, but received no response | |
| Cream containing 10% urea versus control cream | 70 | Unclear how many were randomised to each treatment arm, no separate data for healthy subjects and atopic subjects | |
| Two different formulations of polidocanol‐ and urea‐containing creams against each other | 54 | Unclear how many were randomised to each treatment arm, no separate data for healthy subjects and atopic subjects | |
| Urea 10% ointment versus base OR versus urea 20% ointment | 552 | The data were confusingly reported in this study and did not lend themselves to further analysis. As the study was 39 years old we have not contacted the investigators for data |
| Analysis | Comparison | MD/RR/HR/SMD | 95% confidence interval | P value |
|---|---|---|---|---|
| Change from baseline in SCORAD | Moisturisers versus no treatment Pooled data | MD ‐2.51 | ‐3.66 to ‐1.37 | P < 0.0001 |
| Number of participants experiencing a flare | Moisturisers versus no treatment Pooled data | RR 0.39 | 0.22 to 0.68 | P = 0.0008 |
| Amount of topical steroids used | Moisturisers versus no treatment Pooled data for the first 3 to 4 weeks | MD ‐5.34 | ‐7.79 to ‐2.89 | P < 0.0001 |
| Moisturisers versus no treatment Single study data last 3 to 4 weeks | MD 0.50 | ‐4.70 to 5.70 | P = 0.85 | |
| Moisturisers versus no treatment Pooled data for 6 to 8 weeks | MD ‐8.11 | ‐11.00 to ‐5.22 | P < 0.00001 | |
| Change from baseline in quality of life | Moisturisers versus no treatment Pooled data | SMD ‐0.14 | ‐0.44 to 0.16 | P = 0.35 |
| Number of participants who experienced good improvement to total resolution | Atopiclair versus vehicle Pooled data | RR 4.16 | 2.96 to 5.86 | P < 0.00001 |
| Change from baseline itch measured on a VAS | Atopiclair versus vehicle Pooled data | MD ‐2.08 | ‐2.35 to ‐1.81 | P < 0.00001 |
| Number of participants reporting an adverse event | Atopiclair versus vehicle Pooled data | RR 1.04 | 0.80 to 1.35 | P = 0.78 |
| Change from baseline in EASI | Atopiclair versus vehicle Pooled data | MD ‐4.05 | ‐5.00 to ‐3.10 | P < 0.00001 |
| Number of participants experiencing a flare | Atopiclair versus vehicle Pooled data | RR 0.18 | 0.11 to 0.31 | P < 0.00001 |
| Change from baseline in skin capacitance | Urea‐containing versus vehicle | MD 3.13 | 2.13 to 4.13 | P < 0.00001 |
| Numbers of participants reporting an adverse event | Glycerol versus placebo cream Pooled data | RR 0.89 | 0.67 to 1.18 | P = 0.46 |
| Change in disease severity as assessed by the investigators | Oat‐containing cream versus vehicle or no treatment Pooled data | SMD ‐0.19 | ‐0.43 to 0.05 | P = 0.12 |
| Change from baseline in quality of life | Oat‐containing cream versus vehicle or no treatment Pooled data | SMD ‐0.09 | ‐0.35 to 0.17 | P = 0.51 |
| Number of participants that experienced improvement | All moisturisers versus vehicle, placebo or no treatment Pooled data | RR 2.20 | 1.84 to 2.62 | P < 0.00001 |
| Change from baseline in itch | All moisturisers versus vehicle, placebo or no treatment Pooled data | SMD ‐0.88 | ‐1.04 to ‐0.72 | P < 0.00001 |
| Number of participants that expressed treatment satisfaction | All moisturisers versus vehicle, placebo or no treatment Pooled data | RR 1.69 | 1.35 to 2.11 | P < 0.00001 |
| Number of participants reporting an adverse event | All moisturisers versus vehicle, placebo or no treatment Pooled data | RR 1.06 | 0.88 to 1.27 | P = 0.56 |
| Change in disease severity as assessed by the investigators | All moisturisers versus vehicle, placebo or no treatment Pooled data | SMD ‐0.62 | ‐0.73 to ‐0.51 | P < 0.00001 |
| Number of participants experiencing a flare | All moisturisers versus vehicle, placebo or no treatment Pooled data | RR 0.35 | 0.26 to 0.47 | P < 0.00001 |
| Change from baseline in quality of life | All moisturisers versus vehicle, placebo or no treatment Pooled data | SMD ‐0.40 | ‐0.64 to ‐0.17 | P = 0.0006 |
| Change from baseline in TEWL | Primrose oil versus placebo oil Pooled data | MD ‐0.34 | ‐1.44 to 0.76 | P = 0.55 |
| Change from baseline in skin hydration | Primrose oil versus placebo oil Pooled data | MD 0.34 | ‐2.54 to 3.21 | P = 0.82 |
| Change from baseline in itch (VAS) | Licochalcone versus hydrocortisone Pooled data | MD ‐0.37 | ‐0.75 to ‐0.00 | P = 0.05 |
| Change from baseline in SCORAD | Licochalcone versus hydrocortisone Pooled data | MD ‐0.12 | ‐0.77 to 0.54 | P = 0.73 |
| Change from baseline in TEWL | Licochalcone versus hydrocortisone Pooled data | MD ‐2.04 | ‐3.60 to ‐0.49 | P = 0.010 |
| Number of participants reporting an adverse event | Vehicle plus moisturiser versus fluticasone propionate plus moisturiser Pooled data | RR 0.60 | 0.42 to 0.85 | P = 0.004 |
| Number of participants experiencing a flare | Vehicle plus moisturiser versus fluticasone propionate plus moisturiser Pooled data | RR 2.27 | 1.91 to 2.71 | P < 0.00001 |
| Hazard ratio for rate of flare | Vehicle plus moisturiser versus fluticasone propionate plus moisturiser Pooled data | HR 3.67 | 2.78 to 4.84 | P < 0.00001 |
| Change in disease severity as assessed by the investigators | Active treatment in combination with a moisturiser versus active treatment only Pooled data | SMD ‐0.87 | ‐1.17 to ‐0.57 | P = 0.00001 |
| Change in quality of life IDQOL | Active treatment in combination with a moisturiser versus active treatment only Pooled data | MD ‐1.31 | ‐2.70 to 0.09 | P = 0.07 |
| Change of quality of life DFI | Active treatment in combination with a moisturiser versus active treatment only Pooled data | MD ‐1.03 | ‐2.47 to 0.42 | P = 0.17 |
| Abbreviations DFI: dermatitis family impact | ||||
| MOISURISER VERSUS NO MOISTURISER | ||||||
|---|---|---|---|---|---|---|
| Change from baseline in SCORAD | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | MD (95% CI) | Heterogeneity I² | P value |
| All trials (Giordano‐Labadie 2006; Grimalt 2007; Patrizi 2014) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Allocation concealment | ||||||
| Low risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Blinding of participants and personnel | ||||||
| High risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Blinding of outcome assessment | ||||||
| High risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Incomplete outcome data | ||||||
| Low risk (Giordano‐Labadie 2006; Patrizi 2014) | 2 | 63 | 65 | ‐3.39 (‐4.73 to ‐2.05) | 0% | P < 0.00001 |
| High risk (Grimalt 2007) | 1 | 78 | 70 | ‐0.16 (‐2.36 to 2.04) | NA | P = 0.89 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 141 | 135 | ‐2.42 (‐4.55 to ‐0.28) | 68% | P = 0.03 |
| ATOPICLAIR VERSUS VEHICLE | ||||||
| Number of participants who considered their skin to have improved | ||||||
| Variable | Number of studies | Number of participants in Atopiclair group | Number of participants in vehicle group | RR (95% CI) | Heterogeneity I² | P value |
| All trials (Abramovits 2008; Belloni 2005; Boguniewicz 2008) | 3 | 232 | 158 | 4.51 (2.19 to 9.29) | 64% | P < 0.0001 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 232 | 158 | 4.51 (2.19 to 9.29) | 64% | P < 0.0001 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Belloni 2005) | 2 | 160 | 88 | 3.09 (2.08 to 4.59) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 8.06 (3.95 to 16.42) | NA | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005) | 2 | 160 | 88 | 3.09 (2.08 to 4.59) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 8.06 (3.95 to 16.42) | NA | P < 0.00001 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005) | 2 | 160 | 88 | 3.09 (2.08 to 4.59) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 8.06 (3.95 to 16.42) | NA | P < 0.00001 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Boguniewicz 2008) | 2 | 87 | 85 | 6.95 (3.69 to 13.07) | 0% | P < 0.00001 |
| High risk (Abramovits 2008) | 1 | 145 | 73 | 3.02 (2.00 to 4.56) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 232 | 158 | 4.51 (2.19 to 9.29) | 64% | P < 0.0001 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 232 | 158 | 4.51 (2.19 to 9.29) | 64% | P < 0.0001 |
| Change from baseline in itch measured on a VAS | ||||||
| Variable | Number of studies | Number of participants in Atopiclair group | Number of participants in vehicle group | MD (95% CI) | Heterogeneity I² | P value |
| All trials (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Patrizi 2008) | 4 | 235 | 161 | ‐2.65 (‐4.21 to ‐1.09) | 97% | P = 0.0008 |
| Sequence generation | ||||||
| Low risk (all trials) | 4 | 235 | 161 | ‐2.65 (‐4.21 to ‐1.09) | 97% | P = 0.0008 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 163 | 91 | ‐2.25 (‐3.83 to ‐0.68) | 95% | P = 0.005 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐3.80 (‐4.36 to ‐3.24) | NA | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 163 | 91 | ‐2.25 (‐3.83 to ‐0.68) | 95% | P = 0.005 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐3.80 (‐4.36 to ‐3.24) | NA | P < 0.00001 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 163 | 91 | ‐2.25 (‐3.83 to ‐0.68) | 95% | P = 0.005 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐3.80 (‐4.36 to ‐3.24) | NA | P < 0.00001 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Boguniewicz 2008; Patrizi 2008) | 3 | 106 | 104 | ‐2.33 (‐4.13 to ‐0.52) | 97% | P = 0.01 |
| High risk ((Abramovits 2008) | 1 | 129 | 57 | ‐3.70 (‐4.66 to ‐2.74) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 4 | 235 | 161 | ‐2.65 (‐4.21 to ‐1.09) | 97% | P = 0.0008 |
| Other bias | ||||||
| Low risk (all trials) | 4 | 235 | 161 | ‐2.65 (‐4.21 to ‐1.09) | 97% | P = 0.0008 |
| Change from baseline in EASI | ||||||
| Variable | Number of studies | Number of participants in Atopiclair group | Number of participants in vehicle group | MD (95% CI) | Heterogeneity I² | P value |
| All trials (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Patrizi 2008) | 4 | 251 | 175 | ‐4.00 (‐5.42 to ‐2.57) | 51% | P < 0.00001 |
| Sequence generation | ||||||
| Low risk (all trials) | 4 | 251 | 175 | ‐4.00 (‐5.42 to ‐2.57) | 51% | P < 0.00001 |
| Allocation concealment | ||||||
| Low risk Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 179 | 105 | ‐3.36 (‐4.47 to ‐2.25) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐5.99 (‐7.85 to ‐4.13) | NA | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 179 | 105 | ‐3.36 (‐4.47 to ‐2.25) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐5.99 (‐7.85 to ‐4.13) | NA | P < 0.00001 |
| Blinding of outcome assessment | ||||||
| Low risk Abramovits 2008; Belloni 2005; Patrizi 2008) | 3 | 179 | 105 | ‐3.36 (‐4.47 to ‐2.25) | 0% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐5.99 (‐7.85 to ‐4.13) | NA | P < 0.00001 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005) | 1 | 15 | 15 | ‐3.30 (‐5.67 to ‐0.93) | NA | P = 0.006 |
| Unclear risk (Boguniewicz 2008; Patrizi 2008) | 2 | 91 | 89 | ‐4.42 (‐7.73 to ‐1.10) | 77% | P = 0.009 |
| High risk (Abramovits 2008) | 1 | 145 | 71 | ‐3.62 (‐5.06 to ‐2.18) | NA | P < 0.0001 |
| Selective reporting | ||||||
| Low risk (all trials) | 4 | 251 | 175 | ‐4.00 (‐5.42 to ‐2.57) | 51% | P < 0.00001 |
| Other bias | ||||||
| Low risk (all trials) | 4 | 251 | 175 | ‐4.00 (‐5.42 to ‐2.57) | 51% | P < 0.00001 |
| OAT‐CONTAINING MOISTURISERS VERSUS VEHICLE OR NO TREATMENT (NO MOISTURISER) | ||||||
| Change from baseline in disease severity as assessed by the investigators (EASI and SCORAD) | ||||||
| Variable | Number of studies | Number of participants in oat‐containing moisturiser group | Number of participants in control group | SMD (95% CI) | Heterogeneity I² | P value |
| All trials (Giordano‐Labadie 2006; Grimalt 2007; Nebus 2009) | 3 | 138 | 134 | ‐0.23 (‐0.66 to 0.21) | 65% | P = 0.30 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 138 | 134 | ‐0.23 (‐0.66 to 0.21) | 65% | P = 0.30 |
| Allocation concealment | ||||||
| Low risk (all trials) | 3 | 138 | 134 | ‐0.23 (‐0.66 to 0.21) | 65% | P = 0.30 |
| Blinding of participants and personnel | ||||||
| Low risk (Nebus 2009) | 1 | 25 | 25 | 0.01 (‐0.55 to 0.56) | NA | P = 0.98 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 113 | 109 | ‐0.33 (‐0.98 to 0.32) | 81% | P = 0.32 |
| Blinding of outcome assessment | ||||||
| Low risk (Nebus 2009) | 1 | 25 | 25 | 0.01 (‐0.55 to 0.56) | NA | P = 0.98 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 113 | 109 | ‐0.33 (‐0.98 to 0.32) | 81% | P = 0.32 |
| Incomplete outcome data | ||||||
| Low risk (Giordano‐Labadie 2006; Nebus 2009) | 2 | 60 | 64 | ‐0.36 (‐1.03 to 0.32) | 71% | P = 0.30 |
| High risk (Grimalt 2007) | 1 | 78 | 70 | ‐0.02 (‐0.35 to 0.30) | NA | P = 0.98 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 138 | 134 | ‐0.23 (‐0.66 to 0.21) | 65% | P = 0.30 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 138 | 134 | ‐0.23 (‐0.66 to 0.21) | 65% | P = 0.30 |
| Effect | ||||||
| No difference (Grimalt 2007; Nebus 2009) | 2 | 103 | 95 | ‐0.02 (‐0.29 to 0.26) | 0% | P = 0.91 |
| Difference in favour of oat‐containing moisturiser (Giordano‐Labadie 2006) | 1 | 35 | 39 | ‐0.69 (‐1.16 to ‐0.22) | NA | P = 0.004 |
| Change from baseline in quality of life | ||||||
| Variable | Number of studies | Number of participants in oat‐containing moisturiser group | Number of participants in control group | SMD (95% CI) | Heterogeneity I² | P value |
| All trials (Giordano‐Labadie 2006; Grimalt 2007; Nebus 2009) | 3 | 110 | 116 | ‐0.09 (‐0.37 to 0.19) | 12% | P = 0.53 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 110 | 116 | ‐0.09 (‐0.37 to 0.19) | 12% | P = 0.53 |
| Allocation concealment | ||||||
| Low risk (all trials) | 3 | 110 | 116 | ‐0.09 (‐0.37 to 0.19) | 12% | P = 0.53 |
| Blinding of participants and personnel | ||||||
| Low risk (Nebus 2009) | 1 | 25 | 25 | 0.10 (‐0.46 to 0.65) | NA | P = 0.74 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 85 | 91 | ‐0.16 (‐0.55 to 0.24) | 42% | P = 0.44 |
| Blinding of outcome assessment | ||||||
| Low risk (Nebus 2009) | 1 | 25 | 25 | 0.10 (‐0.46 to 0.65) | NA | P = 0.74 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 85 | 91 | ‐0.16 (‐0.55 to 0.24) | 42% | P = 0.44 |
| Incomplete outcome data | ||||||
| Low risk (Giordano‐Labadie 2006; Nebus 2009) | 2 | 60 | 64 | ‐0.17 (‐0.63 to 0.29) | 39% | P = 0.48 |
| High risk (Grimalt 2007) | 1 | 50 | 52 | 0.03 (‐0.36 to 0.41) | NA | P = 0.89 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 110 | 116 | ‐0.09 (‐0.37 to 0.19) | 12% | P = 0.53 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 110 | 116 | ‐0.09 (‐0.37 to 0.19) | 12% | P = 0.53 |
| ALL MOISTURISERS VERSUS VEHICLE TO PLACEBO OR NO MOISTURISER | ||||||
| Number of participants who considered their skin to have improved | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | RR (95% CI) | Heterogeneity I² | P value |
| All studies (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Lodén 2002; Nebus 2009) | 5 | 323 | 249 | 2.46 (1.16 to 5.23) | 95% | P = 0.02 |
| Sequence generation | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Nebus 2009 | 4 | 257 | 183 | 3.10 (0.98 to 9.82) | 95% | P = 0.05 |
| Unclear risk (Lodén 2002) | 1 | 66 | 66 | 1.24 (1.03 to 1.49) | NA | P = 0.02 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Nebus 2009) | 3 | 185 | 113 | 2.19 (0.75 to 6.39) | 95% | P = 0.15 |
| Unclear risk (Boguniewicz 2008; Lodén 2002) | 2 | 138 | 136 | 3.11 (0.25 to 38.71) | 98% | P = 0.98 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Nebus 2009) | 3 | 185 | 113 | 2.19 (0.75 to 6.39) | 95% | P = 0.15 |
| Unclear risk (Boguniewicz 2008; Lodén 2002) | 2 | 138 | 136 | 3.11 (0.25 to 38.71) | 98% | P = 0.98 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Nebus 2009) | 3 | 185 | 113 | 2.19 (0.75 to 6.39) | 95% | P = 0.15 |
| Unclear risk (Boguniewicz 2008; Lodén 2002) | 2 | 138 | 136 | 3.11 (0.25 to 38.71) | 98% | P = 0.98 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Lodén 2002; Nebus 2009) | 3 | 106 | 106 | 1.23 (0.94 to 1.62) | 48% | P = 0.13 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 8.06 (3.95 to 16.42) | NA | P < 0.00001 |
| High risk (Abramovits 2008) | 1 | 145 | 73 | 3.02 (2.00 to 4.56) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 5 | 323 | 249 | 2.46 (1.16 to 5.23) | 95% | P = 0.02 |
| Other bias | ||||||
| Low risk (all trials) | 5 | 323 | 249 | 2.46 (1.16 to 5.23) | 95% | P = 0.02 |
| Change from baseline in itch | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | SMD (95% CI) | Heterogeneity I² | P value |
| All studies (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Boralevi 2014; Nebus 2009; Patrizi 2008; Patrizi 2014) | 7 | 412 | 337 | ‐1.10 (‐1.83 to ‐0.38) | 94% | P = 0.003 |
| Sequence generation | ||||||
| Low risk (all trials) | 7 | 412 | 337 | ‐1.10 (‐1.83 to ‐0.38) | 94% | P = 0.003 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Nebus 2009; Patrizi 2008; Patrizi 2014) | 6 | 340 | 267 | ‐0.89 (‐1.56 to ‐0.23) | 91% | P = 0.009 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐2.22 (‐2.64 to ‐1.80) | NA | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Nebus 2009; Patrizi 2008) | 5 | 312 | 241 | ‐0.98 (‐1.79 to ‐0.18) | 93% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐2.22 (‐2.64 to ‐1.80) | NA | P < 0.00001 |
| High risk (Patrizi 2014) | 1 | 28 | 26 | ‐0.52 (‐1.06 to 0.03) | NA | P = 0.06 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Nebus 2009; Patrizi 2008) | 5 | 312 | 241 | ‐0.98 (‐1.79 to ‐0.18) | 93% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐2.22 (‐2.64 to ‐1.80) | NA | P < 0.00001 |
| High risk (Patrizi 2014) | 1 | 28 | 26 | ‐0.52 (‐1.06 to 0.03) | NA | P = 0.06 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Boralevi 2014; Nebus 2009; Patrizi 2014) | 4 | 192 | 191 | ‐0.38 (‐0.94 to 0.17) | 80% | P = 0.18 |
| Unclear risk (Boguniewicz 2008; Patrizi 2008) | 2 | 91 | 89 | ‐2.29 (‐2.67 to ‐1.91) | 0% | P < 0.00001 |
| High risk (Abramovits 2008) | 1 | 129 | 57 | ‐1.33 (‐1.67 to ‐0.99) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 7 | 412 | 337 | ‐1.10 (‐1.83 to ‐0.38) | 94% | P = 0.003 |
| Other bias | ||||||
| Low risk (all trials) | 7 | 412 | 337 | ‐1.10 (‐1.83 to ‐0.38) | 94% | P = 0.003 |
| Number of participants who expressed treatment satisfaction | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | RR (95% CI) | Heterogeneity I² | P value |
| All trials (Abramovits 2008; Belloni 2005; Nebus 2009) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Allocation concealment | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Blinding of participants and personnel | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Blinding of outcome assessment | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Nebus 2009) | 2 | 40 | 40 | 1.04 (0.77 to 1.42) | 0% | P = 0.79 |
| High risk (Abramovits 2008) | 1 | 145 | 73 | 2.14 (1.58 to 2.89) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 185 | 113 | 1.35 (0.77 to 2.36) | 83% | P = 0.29 |
| Number of participants who reported an adverse event | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | RR (95% CI) | Heterogeneity I² | P value |
| All trials (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Boralevi 2014; Gayraud 2015; Grimalt 2007; Korting 2010; Lodén 2002; Patrizi 2008; Tan 2010) | 10 | 680 | 595 | 1.03 (0.82 to 1.30) | 21% | P = 0.80 |
| Sequence generation | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Boralevi 2014; Gayraud 2015; Grimalt 2007; Korting 2010; Patrizi 2008; Tan 2010) | 9 | 614 | 529 | 0.96 (0.74 to 1.24) | 16% | P = 0.76 |
| Unclear risk (Lodén 2002) | 1 | 66 | 66 | 1.31 (0.89 to 1.91) | NA | P = 0.17 |
| Allocation concealment | ||||||
| Low risk ((Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Grimalt 2007; Korting 2010; Patrizi 2008; Tan 2010) | 7 | 491 | 411 | 1.00 (0.65 to 1.55) | 35% | P = 0.99 |
| Unclear risk (Boguniewicz 2008; Korting 2010; Lodén 2002) | 3 | 189 | 184 | 1.08 (0.82 to 1.43) | 15% | P = 0.59 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Patrizi 2008; Tan 2010) | 6 | 400 | 329 | 0.94 (0.72 to 1.24) | 0% | P = 0.67 |
| Unclear risk (Boguniewicz 2008; Lodén 2002) | 2 | 138 | 136 | 1.11 (0.83 to 1.48) | 26% | P = 0.49 |
| High risk (Grimalt 2007; Korting 2010) | 2 | 142 | 130 | 2.27 (0.06 to 90.70) | 80% | P = 0.66 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Patrizi 2008; Tan 2010) | 6 | 400 | 329 | 0.94 (0.72 to 1.24) | 0% | P = 0.67 |
| Unclear risk (Boguniewicz 2008; Lodén 2002) | 2 | 138 | 136 | 1.11 (0.83 to 1.48) | 26% | P = 0.49 |
| High risk (Grimalt 2007; Korting 2010) | 2 | 142 | 130 | 2.27 (0.06 to 90.70) | 80% | P = 0.66 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Boralevi 2014; Gayraud 2015; Korting 2010; Lodén 2002; Tan 2010) | 6 | 352 | 350 | 0.99 (0.70 to 1.40) | 23% | P = 0.96 |
| Unclear risk (Boguniewicz 2008; Patrizi 2008) | 2 | 92 | 90 | 0.95 (0.69 to 1.30) | 0% | P = 0.73 |
| High risk (Abramovits 2008; Grimalt 2007) | 2 | 236 | 155 | 3.04 (0.24 to 38.72) | 71% | P = 0.39 |
| Selective reporting | ||||||
| Low risk (all trials) | 10 | 680 | 595 | 1.03 (0.82 to 1.30) | 21% | P = 0.80 |
| Other bias | ||||||
| Low risk (all trials) | 10 | 680 | 595 | 1.03 (0.82 to 1.30) | 21% | P = 0.80 |
| Change in disease severity as assessed by the investigators | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | SMD (95% CI) | Heterogeneity I² | P value |
| All studies (Abramovits 2008; Belloni 2005; Boguniewicz 2008; Boralevi 2014; Gayraud 2015; Giordano‐Labadie 2006; Grimalt 2007; Korting 2010; Nebus 2009; Patrizi 2008; Patrizi 2014; Tan 2010) | 12 | 683 | 598 | ‐0.65 (‐0.89 to ‐0.41) | 75% | P < 0.00001 |
| Sequence generation | ||||||
| Low risk (all trials) | 12 | 683 | 598 | ‐0.65 (‐0.89 to ‐0.41) | 75% | P < 0.00001 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Giordano‐Labadie 2006; Grimalt 2007; Nebus 2009; Patrizi 2008; Patrizi 2014; Tan 2010) | 10 | 561 | 481 | ‐0.53 (‐0.76 to ‐0.30) | 66% | P = 0.009 |
| Unclear risk (Boguniewicz 2008; Korting 2010) | 2 | 122 | 117 | ‐1.15 (‐1.43 to ‐0.88) | 0% | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Nebus 2009; Patrizi 2008; Tan 2010) | 7 | 420 | 346 | ‐0.53 (‐0.77 to ‐0.30) | 52% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐1.04 (‐1.39 to ‐0.69) | NA | P < 0.00001 |
| High risk ( Giordano‐Labadie 2006; Grimalt 2007; Korting 2010; Patrizi 2014) | 4 | 191 | 182 | ‐0.77 (‐1.41 to ‐0.12) | 88% | P = 0.02 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Belloni 2005; Boralevi 2014; Gayraud 2015; Nebus 2009; Patrizi 2008; Tan 2010) | 7 | 420 | 346 | ‐0.53 (‐0.77 to ‐0.30) | 52% | P < 0.00001 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | ‐1.04 (‐1.39 to ‐0.69) | NA | P < 0.00001 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007; Korting 2010; Patrizi 2014) | 4 | 191 | 182 | ‐0.77 (‐1.41 to ‐0.12) | 88% | P = 0.02 |
| Incomplete outcome data | ||||||
| Low risk (Belloni 2005; Boralevi 2014; Gayraud 2015; Giordano‐Labadie 2006; Korting 2010; Nebus 2009; Patrizi 2014; Tan 2010 | 8 | 369 | 368 | ‐0.66 (‐0.96 to ‐0.36) | 71% | P < 0.0001 |
| Unclear risk (Boguniewicz 2008; Patrizi 2008) | 2 | 91 | 89 | ‐0.93 (‐1.29 to ‐0.57) | 17% | P < 0.00001 |
| High risk (Abramovits 2008; Grimalt 2007) | 2 | 223 | 141 | ‐0.41 (‐1.17 to 0.35) | 92% | P = 0.29 |
| Selective reporting | ||||||
| Low risk (all trials) | 12 | 683 | 598 | ‐0.65 (‐0.89 to ‐0.41) | 75% | P < 0.00001 |
| Other bias | ||||||
| Low risk (all trials) | 12 | 683 | 598 | ‐0.65 (‐0.89 to ‐0.41) | 75% | P < 0.00001 |
| Number of participants who experienced a flare | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | RR (95% CI) | Heterogeneity I² | P value |
| All studies (Abramovits 2008; Boguniewicz 2008; Gayraud 2015; Patrizi 2008; Weber 2015; Wirén 2009) | 6 | 341 | 266 | 0.33 (0.17 to 0.62) | 73% | P = 0.0006 |
| Sequence generation | ||||||
| Low risk (all trials) | 6 | 341 | 266 | 0.33 (0.17 to 0.62) | 73% | P = 0.0006 |
| Allocation concealment | ||||||
| Low risk (Abramovits 2008; Gayraud 2015; Patrizi 2008; Weber 2015; Wirén 2009) | 5 | 269 | 196 | 0.33 (0.15 to 0.71) | 78% | P = 0.005 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 0.29 (0.12 to 0.68) | NA | P = 0.005 |
| Blinding of participants and personnel | ||||||
| Low risk (Abramovits 2008; Gayraud 2015; Patrizi 2008) | 3 | 227 | 151 | 0.27 (0.06 to 1.20) | 89% | P = 0.09 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 0.29 (0.12 to 0.68) | NA | P = 0.005 |
| High risk (Weber 2015; Wirén 2009) | 2 | 42 | 45 | 0.40 (0.23 to 0.70) | 0% | P = 0.001 |
| Blinding of outcome assessment | ||||||
| Low risk (Abramovits 2008; Gayraud 2015; Patrizi 2008) | 3 | 227 | 151 | 0.27 (0.06 to 1.20) | 89% | P = 0.09 |
| Unclear risk (Boguniewicz 2008) | 1 | 72 | 70 | 0.29 (0.12 to 0.68) | NA | P = 0.005 |
| High risk (Weber 2015; Wirén 2009) | 2 | 42 | 45 | 0.40 (0.23 to 0.70) | 0% | P = 0.001 |
| Incomplete outcome data | ||||||
| Low risk (Gayraud 2015; Weber 2015; Wirén 2009) | 3 | 104 | 106 | 0.54 (0.31 to 0.92) | 47% | P = 0.02 |
| Unclear risk (Boguniewicz 2008; Patrizi 2008) | 2 | 92 | 89 | 0.26 (0.12 to 0.57) | 0% | P = 0.0007 |
| High risk (Abramovits 2008) | 1 | 145 | 71 | 0.14 (0.07 to 0.28) | NA | P < 0.00001 |
| Selective reporting | ||||||
| Low risk (all trials) | 6 | 341 | 266 | 0.33 (0.17 to 0.62) | 73% | P = 0.0006 |
| Other bias | ||||||
| Low risk (all trials) | 6 | 341 | 266 | 0.33 (0.17 to 0.62) | 73% | P = 0.0006 |
| Change from baseline in quality of life | ||||||
| Variable | Number of studies | Number of participants in moisturiser group | Number of participants in control group | SMD (95% CI) | Heterogeneity I² | P value |
| All trials (Gayraud 2015; Giordano‐Labadie 2006; Grimalt 2007) | 3 | 146 | 154 | ‐0.39 (‐0.90 to 0.12) | 79% | P = 0.13 |
| Sequence generation | ||||||
| Low risk (all trials) | 3 | 146 | 154 | ‐0.39 (‐0.90 to 0.12) | 79% | P = 0.13 |
| Allocation concealment | ||||||
| Low risk (all trials) | 3 | 146 | 154 | ‐0.39 (‐0.90 to 0.12) | 79% | P = 0.13 |
| Blinding of participants and personnel | ||||||
| Low risk (Gayraud 2015) | 1 | 62 | 61 | ‐0.81 (‐1.18 to ‐0.44) | NA | P < 0.0001 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 84 | 93 | ‐0.15 (‐0.55 to 0.24) | 42% | P = 0.44 |
| Blinding of outcome assessment | ||||||
| Low risk (Gayraud 2015) | 1 | 62 | 61 | ‐0.81 (‐1.18 to ‐0.44) | NA | P < 0.0001 |
| High risk (Giordano‐Labadie 2006; Grimalt 2007) | 2 | 84 | 93 | ‐0.15 (‐0.55 to 0.24) | 42% | P = 0.44 |
| Incomplete outcome data | ||||||
| Low risk (Gayraud 2015; Giordano‐Labadie 2006) | 2 | 97 | 100 | ‐0.62 (‐1.04 to ‐0.19) | 52% | P = 0.004 |
| High risk (Grimalt 2007) | 1 | 49 | 54 | 0.03 (‐0.36 to 0.41) | NA | P = 0.89 |
| Selective reporting | ||||||
| Low risk (all trials) | 3 | 146 | 154 | ‐0.39 (‐0.90 to 0.12) | 79% | P = 0.13 |
| Other bias | ||||||
| Low risk (all trials) | 3 | 146 | 154 | ‐0.39 (‐0.90 to 0.12) | 79% | P = 0.13 |
| LICOCHALCONE‐CONTAINING MOISTURISERS VERSUS HYDROCORTISONE ACETATE 1% CREAM | ||||||
| Change from baseline in disease severity as assessed by the investigators (SCORAD) | ||||||
| Variable | Number of studies | Number of participants in licochalcone group | Number of participants in hydrocortisone group | MD (95% CI) | Heterogeneity I² | P value |
| All trials (Angelova‐Fischer 2014; Udompataikul 2011; Wanakul 2013) | 3 | 96 (within‐participant) | 96 (within‐participant) | 0.08 (‐1.96 to 2.13) | 85% | P = 0.94 |
| Sequence generation | ||||||
| Low risk (Angelova‐Fischer 2014; Wanakul 2013) | 2 | 70 (within‐participant) | 70 (within‐participant) | ‐0.90 (‐2.85 to 1.05) | 82% | P = 0.32 |
| Unclear risk (Udompataikul 2011) | 1 | 26 (within‐participant) | 26 (within‐participant) | 2.57 (0.59 to 4.55) | NA | P = 0.01 |
| Blinding of participants and personnel | ||||||
| Low risk (Wanakul 2013) | 1 | 52 (within‐participant) | 52 (within‐participant) | ‐2.00 (‐3.47 to ‐0.53) | NA | P = 0.008 |
| Unclear risk (Angelova‐Fischer 2014; Udompataikul 2011) | 2 | 44 (within‐participant) | 44 (within‐participant) | 1.12 (‐1.38 to 3.61) | 82% | P = 0.38 |
| Blinding of outcome assessment | ||||||
| Low risk (Wanakul 2013) | 1 | 52 (within‐participant) | 52 (within‐participant) | ‐2.00 (‐3.47 to ‐0.53) | NA | P = 0.008 |
| High risk (Angelova‐Fischer 2014; Udompataikul 2011) | 2 | 44 (within‐participant) | 44 (within‐participant) | 1.12 (‐1.38 to 3.61) | 82% | P = 0.38 |
| Incomplete outcome data | ||||||
| Low risk (Angelova‐Fischer 2014; Wanakul 2013) | 2 | 70 (within‐participant) | 70 (within‐participant) | ‐0.90 (‐2.85 to 1.05) | 82% | P = 0.32 |
| Unclear risk (Udompataikul 2011) | 1 | 26 (within‐participant) | 26 (within‐participant) | 2.57 (0.59 to 4.55) | NA | P = 0.01 |
| Selective reporting | ||||||
| All studies | 3 | 96 (within‐participant) | 96 (within‐participant) | 0.08 (‐1.96 to 2.13) | 85% | P = 0.94 |
| Other bias | ||||||
| All studies | 3 | 96 (within‐participant) | 96 (within‐participant) | 0.08 (‐1.96 to 2.13) | 85% | P = 0.94 |
| VEHICLE TREATMENT + MOISTURISER VERSUS FLUTICASONE TREATMENT TWICE WEEKLY + MOISTURISER | ||||||
| Number of participants reporting an adverse event | ||||||
| Variable | Number of studies | Number of participants in vehicle + moisturiser group | Number of participants in fluticasone propionate + moisturiser group | RR (95% CI) | Heterogeneity I² | P value |
| All studies (Berth‐Jones 2003 (2 studies); Glazenburg 2009; Hanifin 2002) | 4 | 312 | 406 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Sequence generation | ||||||
| Low risk (Berth‐Jones 2003 (2 studies); Glazenburg 2009) | 3 | 193 | 177 | 0.30 (0.12 to 0.73) | NA | P = 0.008 |
| Unclear risk (Hanifin 2002) | 1 | 119 | 229 | 0.70 (0.48 to 1.04) | NA | P = 0.08 |
| Allocation concealment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | Not estimable | NA | NA |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Blinding of participants and personnel | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | Not estimable | NA | NA |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Blinding of outcome assessment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | Not estimable | NA | NA |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Incomplete outcome data | ||||||
| Unclear risk (all studies) | 4 | 312 | 406 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Selective reporting | ||||||
| Low risk (all studies) | 4 | 312 | 406 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Other bias | ||||||
| Low risk (all studies) | 4 | 312 | 406 | 0.51 (0.22 to 1.14) | 67% | P = 0.10 |
| Number of participants experiencing a flare | ||||||
| Variable | Number of studies | Number of participants in vehicle + moisturiser group | Number of participants in fluticasone propionate + moisturiser group | RR (95% CI) | Heterogeneity I² | P value |
| All studies (Berth‐Jones 2003 (2 studies); Glazenburg 2009; Hanifin 2002) | 4 | 312 | 406 | 2.17 (1.51 to 3.11) | 74% | P < 0.0001 |
| Sequence generation | ||||||
| Low risk (Berth‐Jones 2003 (2 studies); Glazenburg 2009) | 3 | 193 | 177 | 2.02 (1.24 to 3.30) | 76% | P = 0.005 |
| Unclear risk (Hanifin 2002) | 1 | 119 | 229 | 2.62 (2.03 to 3.39) | NA | P < 0.00001 |
| Allocation concealment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | 2.17 (0.88 to 5.37) | 88% | P = 0.09 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 2.27 (1.62 to 3.19) | 54% | P < 0.00001 |
| Blinding of participants and personnel | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | 2.17 (0.88 to 5.37) | 88% | P = 0.09 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 2.27 (1.62 to 3.19) | 54% | P < 0.00001 |
| Blinding of outcome assessment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 157 | 138 | 2.17 (0.88 to 5.37) | 88% | P = 0.09 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 155 | 268 | 2.27 (1.62 to 3.19) | 54% | P < 0.00001 |
| Incomplete outcome data | ||||||
| Unclear risk (all studies) | 4 | 312 | 406 | 2.17 (1.51 to 3.11) | 74% | P < 0.0001 |
| Selective reporting | ||||||
| Low risk (all studies) | 4 | 312 | 406 | 2.17 (1.51 to 3.11) | 74% | P < 0.0001 |
| Other bias | ||||||
| Low risk (all studies) | 4 | 312 | 406 | 2.17 (1.51 to 3.11) | 74% | P < 0.0001 |
| Hazard ratio for rate of flare | ||||||
| Variable | Number of studies | Number of participants in fluticasone propionate + moisturiser group | Number of participants in vehicle + moisturiser group | HR (95% CI) | Heterogeneity I² | P value |
| All studies (Berth‐Jones 2003 (2 studies); Glazenburg 2009; Hanifin 2002) | 4 | 406 | 312 | 3.69 (1.80 to 7.55) | 85% | P = 0.0004 |
| Sequence generation | ||||||
| Low risk (Berth‐Jones 2003 (2 studies); Glazenburg 2009) | 3 | 177 | 193 | 2.84 (1.44 to 5.61) | 76% | P = 0.003 |
| Unclear risk (Hanifin 2002) | 1 | 229 | 119 | 7.70 (4.62 to 12.84) | NA | P < 0.00001 |
| Allocation concealment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 138 | 157 | 3.26 (1.09 to 9.74) | 87% | P = 0.03 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 268 | 155 | 4.16 (1.21 to 14.31) | 89% | P = 0.02 |
| Blinding of participants and personnel | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 138 | 157 | 3.26 (1.09 to 9.74) | 87% | P = 0.03 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 268 | 155 | 4.16 (1.21 to 14.31) | 89% | P = 0.02 |
| Blinding of outcome assessment | ||||||
| Low risk (Berth‐Jones 2003 (2 studies)) | 2 | 138 | 157 | 3.26 (1.09 to 9.74) | 87% | P = 0.03 |
| Unclear risk (Glazenburg 2009; Hanifin 2002) | 2 | 268 | 155 | 4.16 (1.21 to 14.31) | 89% | P = 0.02 |
| Incomplete outcome data | ||||||
| Unclear risk (all studies ) | 4 | 406 | 312 | 3.69 (1.80 to 7.55) | 85% | P = 0.0004 |
| Selective reporting | ||||||
| Low risk (all studies) | 4 | 406 | 312 | 3.69 (1.80 to 7.55) | 85% | P = 0.0004 |
| Other bias | ||||||
| Low risk (all studies) | 4 | 406 | 312 | 3.69 (1.80 to 7.55) | 85% | P = 0.0004 |
| NA not applicable; MD mean difference; SMD standardised mean difference; RR risk ratio; HR hazard ratio | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Change from baseline in SCORAD Show forest plot | 3 | 276 | Mean Difference (IV, Random, 95% CI) | ‐2.42 [‐4.55, ‐0.28] |
| 1.2 Number of participants experiencing a flare Show forest plot | 2 | 87 | Risk Ratio (M‐H, Random, 95% CI) | 0.40 [0.23, 0.70] |
| 1.3 Rate of flare Show forest plot | 2 | 87 | Hazard Ratio (IV, Random, 95% CI) | 3.74 [1.86, 7.50] |
| 1.4 Amount of topical steroids used Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | Subtotals only | |
| 1.4.1 Amount of topical corticosteroids first 3‐4 weeks | 2 | 222 | Mean Difference (IV, Random, 95% CI) | ‐8.25 [‐17.22, 0.72] |
| 1.4.2 Amount of topical corticosteroids used last 3‐4 weeks | 1 | 74 | Mean Difference (IV, Random, 95% CI) | 0.50 [‐4.70, 5.70] |
| 1.4.3 Total amount of topical corticosteroids used in 6 to 8 weeks | 2 | 222 | Mean Difference (IV, Random, 95% CI) | ‐9.30 [‐15.33, ‐3.27] |
| 1.5 Change from baseline in quality of life Show forest plot | 2 | 177 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.15 [‐0.55, 0.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Number of participants who experienced good improvement to total resolution Show forest plot | 3 | 390 | Risk Ratio (M‐H, Random, 95% CI) | 4.51 [2.19, 9.29] |
| 2.1.1 Number of participants who experienced good improvement to total resolution (low risk of bias) | 1 | 30 | Risk Ratio (M‐H, Random, 95% CI) | 4.00 [1.01, 15.81] |
| 2.1.2 Number of participants who experienced good improvement to total resolution (unclear risk of bias) | 1 | 142 | Risk Ratio (M‐H, Random, 95% CI) | 8.06 [3.95, 16.42] |
| 2.1.3 Number of participants who experienced good improvement to total resolution (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 3.02 [2.00, 4.56] |
| 2.2 Change from baseline in itch measured on a VAS Show forest plot | 4 | 396 | Mean Difference (IV, Random, 95% CI) | ‐2.65 [‐4.21, ‐1.09] |
| 2.2.1 Change from baseline in itch measured on a VAS (low risk of bias) | 1 | 30 | Mean Difference (IV, Random, 95% CI) | ‐0.80 [‐1.20, ‐0.40] |
| 2.2.2 Change from baseline in itch measured on a VAS (unclear risk of bias) | 2 | 180 | Mean Difference (IV, Random, 95% CI) | ‐3.10 [‐4.47, ‐1.73] |
| 2.2.3 Change from baseline in itch measured on a VAS (high risk of bias) | 1 | 186 | Mean Difference (IV, Random, 95% CI) | ‐3.70 [‐4.66, ‐2.74] |
| 2.3 Number of participants reporting an adverse event Show forest plot | 4 | 430 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.79, 1.33] |
| 2.4 Change from baseline in EASI Show forest plot | 4 | 426 | Mean Difference (IV, Random, 95% CI) | ‐4.00 [‐5.42, ‐2.57] |
| 2.4.1 Change from baseline in EASI (low risk of bias) | 1 | 30 | Mean Difference (IV, Random, 95% CI) | ‐3.30 [‐5.67, ‐0.93] |
| 2.4.2 Change from baseline in EASI (unclear risk of bias) | 2 | 180 | Mean Difference (IV, Random, 95% CI) | ‐4.42 [‐7.73, ‐1.10] |
| 2.4.3 Change from baseline in EASI (high risk of bias) | 1 | 216 | Mean Difference (IV, Random, 95% CI) | ‐3.62 [‐5.06, ‐2.18] |
| 2.5 Number of participants experiencing a flare Show forest plot | 3 | 397 | Risk Ratio (M‐H, Random, 95% CI) | 0.18 [0.11, 0.31] |
| 2.5.1 Number of participants experiencing a flare (unclear risk of bias) | 2 | 181 | Risk Ratio (M‐H, Random, 95% CI) | 0.26 [0.12, 0.57] |
| 2.5.2 Number of participants experiencing a flare (high risk of bias) | 1 | 216 | Risk Ratio (M‐H, Random, 95% CI) | 0.14 [0.07, 0.28] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 3.1 Change from baseline in skin capacitance Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | 1.23 [‐7.39, 9.86] | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 4.1 Number of participants reporting an adverse event Show forest plot | 2 | 385 | Risk Ratio (M‐H, Random, 95% CI) | 0.90 [0.68, 1.19] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 5.1 Change in disease severity as assessed by the investigators (SCORAD and EASI) Show forest plot | 3 | 272 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.23 [‐0.66, 0.21] |
| 5.1.1 Change in disease severity as assessed by the investigator (EASI) (low risk of bias) | 1 | 50 | Std. Mean Difference (IV, Random, 95% CI) | 0.01 [‐0.55, 0.56] |
| 5.1.2 Change in disease severity as assessed by the investigators (SCORAD) (high risk of bias) | 2 | 222 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.33 [‐0.98, 0.32] |
| 5.2 Change from baseline in quality of life Show forest plot | 3 | 226 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.09 [‐0.37, 0.19] |
| 5.2.1 Change from baseline in quality of life (low risk of bias) | 1 | 50 | Std. Mean Difference (IV, Random, 95% CI) | 0.10 [‐0.46, 0.65] |
| 5.2.2 Change from baseline in quality of life (high risk of bias) | 2 | 176 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.16 [‐0.55, 0.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 6.1 Number of participants who experienced improvement Show forest plot | 5 | 572 | Risk Ratio (M‐H, Random, 95% CI) | 2.46 [1.16, 5.23] |
| 6.1.1 Number of participants who experienced improvement (low risk of bias) | 2 | 80 | Risk Ratio (M‐H, Random, 95% CI) | 1.86 [0.41, 8.31] |
| 6.1.2 Number of participants who experienced improvement (unclear risk of bias) | 2 | 274 | Risk Ratio (M‐H, Random, 95% CI) | 3.11 [0.25, 38.71] |
| 6.1.3 Number of participants who experienced improvement (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 3.02 [2.00, 4.56] |
| 6.2 Change from baseline in itch Show forest plot | 7 | 749 | Std. Mean Difference (IV, Random, 95% CI) | ‐1.10 [‐1.83, ‐0.38] |
| 6.2.1 Change from baseline in itch (low risk of bias) | 3 | 329 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.36 [‐1.16, 0.43] |
| 6.2.2 Change from baseline in itch (unclear risk of bias) | 2 | 180 | Std. Mean Difference (IV, Random, 95% CI) | ‐2.29 [‐2.67, ‐1.91] |
| 6.2.3 Change from baseline in itch (high risk of bias) | 2 | 240 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.95 [‐1.75, ‐0.16] |
| 6.3 Number of participants who expressed treatment satisfaction Show forest plot | 3 | 298 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [0.77, 2.36] |
| 6.3.1 Number of participants who expressed treatment satisfaction (low risk of bias) | 2 | 80 | Risk Ratio (M‐H, Random, 95% CI) | 1.04 [0.77, 1.42] |
| 6.3.2 Number of participants who expressed treatment satisfaction (high risk of bias) | 1 | 218 | Risk Ratio (M‐H, Random, 95% CI) | 2.14 [1.58, 2.89] |
| 6.4 Number of participants reporting an adverse event Show forest plot | 10 | 1275 | Risk Ratio (M‐H, Random, 95% CI) | 1.03 [0.82, 1.30] |
| 6.4.1 Number of participants reporting an adverse event (low risk of bias) | 4 | 471 | Risk Ratio (M‐H, Random, 95% CI) | 0.83 [0.58, 1.18] |
| 6.4.2 Number of participants reporting an adverse events (unclear risk of bias) | 3 | 314 | Risk Ratio (M‐H, Random, 95% CI) | 1.08 [0.82, 1.42] |
| 6.4.3 Number of participants reporting an adverse events (high risk of bias) | 3 | 490 | Risk Ratio (M‐H, Random, 95% CI) | 1.35 [0.37, 4.96] |
| 6.5 Change in disease severity as assessed by the investigators Show forest plot | 12 | 1281 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.65 [‐0.89, ‐0.41] |
| 6.5.1 Change in disease severity as assessed by the investigators (low risk of bias) | 5 | 512 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.45 [‐0.74, ‐0.15] |
| 6.5.2 Change in disease severity as assessed by the investigators (unclear risk of bias) | 2 | 180 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.93 [‐1.29, ‐0.57] |
| 6.5.3 Change in disease severity as assessed by the investigators (high risk of bias) | 5 | 589 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.77 [‐1.23, ‐0.30] |
| 6.6 Number of participants experiencing a flare Show forest plot | 6 | 607 | Risk Ratio (M‐H, Random, 95% CI) | 0.33 [0.17, 0.62] |
| 6.6.1 Number of participants experiencing a flare (low risk of bias) | 1 | 123 | Risk Ratio (M‐H, Random, 95% CI) | 0.80 [0.48, 1.34] |
| 6.6.2 Number of participants experiencing a flare (unclear risk of bias) | 2 | 181 | Risk Ratio (M‐H, Random, 95% CI) | 0.26 [0.12, 0.57] |
| 6.6.3 Number of participants experiencing a flare (high risk of bias) | 3 | 303 | Risk Ratio (M‐H, Random, 95% CI) | 0.27 [0.12, 0.59] |
| 6.7 Change from baseline in quality of life Show forest plot | 3 | 300 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.39 [‐0.90, 0.12] |
| 6.7.1 Change from baseline in quality of life (low risk of bias) | 1 | 123 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.81 [‐1.18, ‐0.44] |
| 6.7.2 Change from baseline in quality of life (high risk of bias) | 2 | 177 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.15 [‐0.55, 0.24] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 7.1 Change from baseline in TEWL Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | ‐0.34 [‐1.44, 0.76] | |
| 7.2 Change from baseline in skin hydration Show forest plot | 1 | Mean Difference (IV, Random, 95% CI) | 0.34 [‐2.54, 3.21] | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 8.1 Change from baseline in itch (VAS) Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | ‐0.48 [‐1.46, 0.50] | |
| 8.1.1 Change from baseline in itch (VAS) (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | 0.00 [‐0.47, 0.47] | |
| 8.1.2 Change from baseline in itch (VAS) (high risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐1.00 [‐1.61, ‐0.39] | |
| 8.2 Change from baseline in SCORAD Show forest plot | 3 | Mean Difference (IV, Random, 95% CI) | 0.08 [‐1.96, 2.13] | |
| 8.2.1 Change from baseline in SCORAD (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐2.00 [‐3.47, ‐0.53] | |
| 8.2.2 Change from baseline in SCORAD (high risk of bias) | 2 | Mean Difference (IV, Random, 95% CI) | 1.12 [‐1.38, 3.61] | |
| 8.3 Change from baseline in TEWL Show forest plot | 2 | Mean Difference (IV, Random, 95% CI) | ‐0.50 [‐5.88, 4.87] | |
| 8.3.1 Change from baseline in TEWL (unclear risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | ‐3.00 [‐4.71, ‐1.29] | |
| 8.3.2 Change from baseline in TEWL (high risk of bias) | 1 | Mean Difference (IV, Random, 95% CI) | 2.51 [‐1.21, 6.23] | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 9.1 Rate of flare Show forest plot | 1 | Hazard Ratio (IV, Random, 95% CI) | Totals not selected | |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 10.1 Number of participants reporting an adverse event Show forest plot | 3 | 718 | Risk Ratio (M‐H, Random, 95% CI) | 0.51 [0.22, 1.14] |
| 10.2 Number of participants experiencing a flare Show forest plot | 3 | 718 | Risk Ratio (M‐H, Random, 95% CI) | 2.17 [1.51, 3.11] |
| 10.3 Rate of flare Show forest plot | 3 | 723 | Hazard Ratio (IV, Random, 95% CI) | 3.69 [1.80, 7.55] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 11.1 Change in disease severity as assessed by the investigators Show forest plot | 2 | 192 | Std. Mean Difference (IV, Random, 95% CI) | ‐0.87 [‐1.17, ‐0.57] |
| 11.2 Change in quality of life IDQOL Show forest plot | 1 | 67 | Mean Difference (IV, Random, 95% CI) | ‐1.31 [‐2.70, 0.09] |
| 11.3 Change in quality of life DFI Show forest plot | 1 | 67 | Mean Difference (IV, Random, 95% CI) | ‐1.03 [‐2.47, 0.42] |
