Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: assignment of gas composition was randomized in blocks of 9

Follow‐up period: 24 h "all investigations were carried out within a day after arrival at 4559 m"

Diagnosis of AMS

1. Acute mountain sickness (AMS) score. “A score of more than 3 was required for entry to the trial”

Scale used for assessing AMS

1. Subscore (AMS‐C) including 11 statements (Appendix 8)

2. Environmental symptom questionnaire (Appendix 8)

Participants

Number of participants randomized: 20

Sex: men = 19 (95%)

Age: median 32 years (range 22 to 51)

Baseline data

1. Not reported

Inclusion criteria

1. Patients with an AMS score more than 3

Exclusion criteria

1. Not clearly reported

Interventions

Intervention group 1 (n = 6)

1. Oxygen in nitrogen. Dose: 33% (1.6%); route: breathing through a tightly fitted face mask; duration: 30 min; frequency: not clearly reported, apparently a single dose

Intervention group 2 (n = 7)

1. Carbon dioxide in air. Dose 3% (0.15%); breathing through a tightly fitted face mask; duration: 30 min; frequency: not clearly reported, apparently a single dose

Control group (n = 7)

1. Compressed "normal" air. Duration: 30 min; frequency: not clearly reported, apparently a single dose

Co‐intervention

1. Room air was provided for 30 min right before the experimental gas. The gas was humidified and the flow adjusted manually to the ventilation of the subject by the maintenance of a 50 litre reservoir‐balloon at a constant size

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Severity of AMS (AMS‐C score, Appendix 8). Before and immediately after the treatment

2. Symptoms of environmental stress (Appendix 8). After each clinical examination

3. Physiological variables: ventilation, PaCO₂, PaO₂, Oxygen saturation

4. Mean blood flow velocity in the median cerebral artery (MCA)

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Swiss‒Italian border, Capanna "Regina Margherita" in the Alps Valais

Altitude setting: 4559 m (barometric pressure 430 mmHg to 440 mmHg)

Study dates: not reported

Identifier number: not reported

A priori sample estimation: no

Conflicts of interest: not reported

Funding/Support

1. "This study was supported by grant 3200‐0092.85 from the Swiss National Science Foundation"

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "assignment of gas composition was randomised in blocks of nine" (page 773)

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Quote: "the person responsible for the gas supply, the gas bottles, and the reservoir‐balloon were hidden behind a curtain from the subjects and the examiners" (page 773)

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

Quote: "four investigators carried out one each of four different measurements throughout the study‐clinical examinations, ventilation, blood gas analysis, and transcranial doppler ultrasound examination. They were not aware of each other’s results during treatment of any particular patient" (page 773)

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

High risk

Results are reported in a figure. Exact numbers could not be retrieved. In the text, authors reported interpretation of data and P values

Other bias

Unclear risk

Bias in the presentation data: baseline characteristics by groups was not shown

Methods

Three‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: 12 h

Diagnosis of AMS

1. "Headache and one additional sign or symptom were required for entering the trial"

Scales used for assessing acute mountain sickness

1. Clinical score (Appendix 8)

2. AMS‐C score of the questionnaire (Appendix 8)

Participants

Number of participants randomized: 64

Sex: men = 49 (77%)

Age : mean 31 years (range 18 to 52)

Baseline data

1. Clinical score mean: Intervention group — 4.1, 95% CI 3.7 to 4.5; Control group 1 — 4.3, 95% CI 3.7 to 4.8; Control group 2 — 4.5, 95% CI 4.0 to 5.0

2. AMS‐C score mean: Intervention group — 1.8, 95% CI 1.5 to 2.2; Control group 1 — 1.6, 95% CI 1.2 to 1.9; Control group 2 — 1.4, 95% CI 0.7 to 2.6

Inclusion criteria

1. Mountaineers planning to stay overnight "who had ascended by foot and who stayed at 4559 m for at least 12 hours after treatment were eligible to enter the trial if they suffered from headache and one or more additional symptoms of acute mountain sickness". "Headache and one additional sign or symptom were required for entering the trial"

Exclusion criteria

1. "Subjects with clinical signs of high altitude pulmonary oedema (dyspnoea at rest, respiration rate > 25/min, and rales) and those who had taken acetazolamide or nifedipine during ascent"

Interventions

Intervention group (n = 31)

1. Hyperbaric chamber at a pressure of 193 mbar; dose: 193 mbar; frequency: once during the trial; duration of the intervention: 1 h

Control group 1 (n = 23)

1. Hyperbaric chamber at a pressure of 20 mbar; dose: 20 mbar; frequency: once during the trial; duration of the intervention: 1 h

Control group 2 (n = 10)

1. Name: bed rest; dose: NA; frequency: once during the trial; duration of the intervention:

Cointervention

1. Analgesic: paracetamol (intervention group = 18; control group 1 = 15; control group 2 = 8),

2. Antiemetic: thiethylperazine (intervention group = 2; control group 1 = 6; control group 2 = 2)

**Characteristics of the chamber: fabric hyperbaric chamber made by Certec (F‐692 10 Sourcieux‐les‐Mines, France)

Outcomes

Do the authors define outcomes as 'primary' or 'secondary'?: yes

Primary

1. Symptoms of acute mountain sickness before, immediately after, and 12 h after treatment

2. Permitted intake of analgesic

3. Antiemetic drugs in the follow‐up period

Secondary

1. Arterial oxygen saturation

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Swiss‒Italian border, Capanna "Regina Margherita" in the Alps Valais

Altitude setting 4559 m (barometric pressure 430 mmHg to 440 mmHg)

Identifier number: not reported

Study dates: 1990 to 1991

A priori sample estimation: no

Conflicts of Interest: not reported

Funding/Support

1. "This study was supported by a grant from the research institute of the Swiss School of Sports, Magglingen, and by grant 3200‐0092.85 from the Swiss National Science Foundation"

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "randomisation was performed in blocks of six (in 1990) and nine (in 1991)." Page 1098

Allocation concealment (selection bias)

High risk

Quote: "the investigator assigned the treatment by drawing a lot from an envelope containing the assignments of one block. When the remaining lots could be predicted they were added to the envelope containing the next randomisation block." Page 1099

Blinding of participants and personnel (performance bias)
All outcomes

High risk

No blinding method reported. Hyperbaric chamber compared to bed rest has not been masked. Outcomes are dependent on subjective assessment

Blinding of outcome assessment (detection bias)
All outcomes

High risk

No blinding method reported; however, hyperbaric chamber compared to bed rest has not been masked and the outcome is dependent on subjective assessment

Incomplete outcome data (attrition bias)
All outcomes

High risk

Quote: "in 1990 the first seven subjects assigned to low pressure were unintentionally treated with 39 mbar (equivalent to a descent of 500 m) until the inaccuracy of the built in manometer in the low pressure range was discovered. Their results were excluded from analysis, although they were not significantly different from those obtained in subjects treated with 16 or 23 mbar." Page 1099. Outcome data was not available for seven participants in the intervention group (unbalanced attrition)

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Protocol not available

Other bias

Unclear risk

The use of analgesics and antiemetics was permitted during the study period as an option in the three groups. Authors found no significant statistical difference among groups in the use of these drugs

Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period

1. Not clearly reported

2. Adverse events were measured up to 90 min after intervention

Diagnosis of AMS

1. Lake Louise Score > 6 with a headache score > 2 and/or a gastrointestinal score > 2 and/or an ataxia score > 2

Scale used for assessing acute mountain sickness

1. Lake Louise Score (minimum score is 0 and the maximal score is 25)

Participants

Number of participants randomized: 25

Sex: not reported

Age: adults, details about age were not reported

Baseline data

1. AMS Lake Louise score: Intervention group: mean 11.6 (SD = 1.7)

2. Control group: mean 10.9 (SD = 3)

Inclusion criteria

1. "Subjects from the prevention trial who had consented to take part in the treatment trial providing prevention failed". Prevention failed: "Lake Louise Score > 6 with a headache score > 2 and/or a gastrointestinal score > 2 and/or an ataxia score > 2"

2. "Volunteers at the Capanna Regina Margherita who had not taken part in the prevention trial, but who had a Lake Louise Score > 6 with a headache score > 2 and/or a gastrointestinal score > 2 and/or an ataxia score > 2"

Exclusion criteria

1. Residency above 600 m

2. A stay above 2000 m

3. Medication, including vitamins or magnesium, during the last 3 months

4. Cardiac, pulmonary, neurological, renal, hepatic or psychiatric disease

Interventions

Intervention group (n = 12)
Name: magnesium sulphate (16 mmol)
Route: intravenous infusion
Dose: 4 grams. Ampoules drawn into bags of 100 ml of physiological saline and infused over 30 min
Frequency: single dose
Duration of the intervention: infusion for 30 min

Control group (n = 13)
Name: matching placebo

Cointerventions

1. 20 patients (80%) received Magnesium as prophylaxis, in a prevention trial, hours or few days before the treatment trial

2. If the treatment failed, or if the volunteers wished so, they received rescue medication (oxygen by face mask, 4 mg of dexamethasone intravenously and 500 mg of acetazolamide orally)

Outcomes

Primary

1. Treatment success. Defined as number of subjects who had a drop in the Lake Louise Score > 50% 60 min after the start of the treatment (i.e. 30 min after the end of the infusion)

Secondary

1. Number of subjects who had a drop in the Lake Louise Score > 25% and whether or not there was a significant decrease in the score after treatment compared with before the start of the treatment

Adverse effects

1. Outcomes of interest in the review: reduction in illness severity scores of AMS and adverse events

Notes

Country: Swiss‐Italian border, Capanna "Regina Margherita" in the Alps Valais

Altitude setting: 4559 m (barometric pressure 430 mmHg to 440 mmHg)

Identifier number: not reported

Study dates: not reported

A priori sample estimation: yes

Conflicts of interest: not reported

Funding/Support

Study was supported by

1. research funds from the Department of Anaesthesiology, Pharmacology and Surgical Intensive Care, Geneva University Hospitals, Geneva, Switzerland; and

2. the Carlos and Elsie De Reuter Fund, Switzerland

M.R.T.

1. received a Programme for Social Medicine, Preventive and Edpidemiological Research (PROSPER) grant from the Swiss National Science Foundation (No. 3233‐051939.97/2)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "randomized, placebo‐controlled, double‐blind trial". "The Geneva University Hospital Pharmacy was responsible for randomization (table of random numbers) and preparation of the study drugs" Page 270

Allocation concealment (selection bias)

Low risk

Quote: "the Geneva University Hospital Pharmacy was responsible for randomization (table of random numbers) and preparation of the study drugs."

Comment: Central allocation

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Quote: "randomized, placebo‐controlled, double‐blind trial". Page 270.

Quote: "study drugs were provided in identical, numbered 20 ml ampoules" Page 271

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

Quote: "randomized, placebo‐controlled, double‐blind trial". Page 270.

Quote: "study drugs were provided in identical, numbered 20 ml ampoules" Page 271

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Protocol not available.

Other bias

High risk

20 patients (80%) received magnesium as prophylaxis, in a prevention trial, hours or few days before the treatment trial. This may be a confusion variable.

It is not clearly stated the timing of rescue medication or the reason (either treatment failure or volunteers' wish)

Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: patients

Follow‐up period: unclear, apparently 12 to 16 h

Diagnosis of AMS

1. 3 or more points in a symptoms severity score

Scale used for assessing Acute Mountain Sickness

1. The name of the scale is not reported. Authors described "The presence of the symptoms listed was scored as follows: one point for mild headache, nausea, dizziness, shortness of breath and insomnia and two points for severe headache (not relieved by paracetamol 500 mg) and for vomiting. Responses were checked with one of the investigators. Subjects then underwent a clinical examination for tachypnoea (two points), facial or peripheral oedema (one location one point, two or more locations two points), ataxia (heel to toe walking test and Romberg test two points), and pulmonary rales (discreet one point, pronounced two points). Patients with three or more points were selected for the drug trial"

Participants

Number of participants randomized: 35

Sex: men = 28 (80%)

Age: "the two groups were comparable in age"

Baseline data (mean symptom score per group)

1. Dexamethasone group = mean 5.4 (SD = 1.7)

2. Placebo group = mean 4.8 (SD = 1.0)

Inclusion criteria

1. Climbers with symptoms of acute mountain sickness (AMS: 3 or more points in the in the symptoms severity score)

Exclusion criteria

1. Frank high altitude pulmonary or cerebral oedema, or both

Interventions

Intervention group 1 (n = 17)

1. Dexamethasone, route: by mouth; dose: "8 mg initially and another 4 mg after six and 12 hours"; frequency: initially and then every 6 h; duration of the intervention: 12 to 16 h

Control group (n = 18)

1. Identical placebo, route: not clearly reported, "identical placebo"; dose: not clearly reported, "identical placebo"; frequency: not clearly reported, "identical placebo"; duration of the intervention: 12 to 16 h

Cointervention

1. None reported

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Acute mountain sickness score: baseline and after 12 to 16 h of intervention

2. Number of patients becoming totally asymptomatic

3. Arterial oxygen saturation

4. Spirometric measurements: resting minute ventilation, forced vital capacity, and forced expiratory volume in 1 second

5. Physiological measures: weight, pulse rate, blood pressure, arterial oxygen saturation

6. Retinal photography

Outcomes of interest in the review

1. Complete relief of AMS symptoms

2. Reduction in illness severity scores of AMS

Notes

Country: Swiss‒Italian border, Capanna "Regina Margherita" in the Alps Valais

Altitude setting: 4559 m (barometric pressure 430 mmHg to 440 mmHg)

Identifier number: not reported

Study dates: not reported

Priori sample estimation: no

Conflicts of Interest: not reported.

Funding/Support

1. "This study was supported by a grant from the EMDO Stiftung" (University of Zurich)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "patients were randomly assigned". Page 1381

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "patients were randomly assigned". Page 1381

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "A double blind, randomised, placebo controlled trial". Page 1380

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "A double blind, randomised, placebo controlled trial". Page 1380

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

High risk

Protocol not available. Data presented graphically for individuals. No data available for each group for symptomatic scores

Other bias

Unclear risk

Baseline characteristics poorly presented

Methods

Two‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: not clearly specified. Probably 24 h after intervention

Diagnosis of AMS

1. "The AMS Symptom Questionnaire was used to diagnose acute mountain sickness and to evaluate severity”

Scale used for assessing Acute Mountain Sickness

1. AMS Symptom Questionnaire (not Lake Louise), a weighted severity scale of 1 to 3 (least severe to most severe) with 9 items and a maximum score of 18 (Appendix 8)

Participants

Number of participants randomized: 12

Sex: men = 10 (91%)

Age: median 32 years (range 25 to 46)

Baseline data

Mean symptom scores

1. Acetazolamide group: 3.8 +/‐ 0.4

2. Placebo group: 3.8 +/‐ 1.7

Inclusion criteria

1. Patients with acute mountain sickness (score of 2 or greater on a weighted severity scale of 1 to 3), onset of symptoms at 4200 m within 24 h of inclusion

Exclusion criteria

1. Use of acetazolamide within the previous week, defined high altitude pulmonary oedema or high altitude cerebral oedema, serious medical illness

Interventions

Intervention group (n = 6)

1. Acetazolamide. Dose, frequency and duration: 250 mg, at time 0 and 8 h after inclusion in the study. Route: oral

Control group (n = 6)

1. Placebo. Dose, frequency and duration: at time 0 and 8 h after inclusion in the study. Route: oral

Co‐intervention

1. None reported

Outcomes

Main outcome measures

1. Acute mountain sickness score at baseline and at 24 h

2. Pulmonary gas exchange at baseline and at 24 h

Secondary outcomes

1. Other physiologic measurements

2. Side effects

Outcomes of interest in the review

1. Complete relief of AMS symptoms

2. Reduction in illness severity scores of AMS

3. Adverse events

Notes

Country: Alaska. Denali Medical Research Project high altitude research station, McKinley

Altitude setting: 4200 m
Identifier number: not reported

Study dates: June 1989

A priori sample estimation: no

Conflicts of Interest: not reported

Funding/Support

1. “In part by the Carles S. Houston Award from the Wilderness Medical Society; the American Heart Association Alaska Affiliate; and the United States Army Research and Development Command”

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "participants were randomly assigned to receive either acetazolamide or placebo in a double‐blind fashion". Quote: "randomization was done in blocks of four to ensure equivalent numbers in each group". Page 462

Allocation concealment (selection bias)

High risk

Quote: "one participant reported a history of sulfa‐drug allergy and was assigned (non‐randomly) to the placebo group" Page 462

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Quote: "participants were randomly assigned to receive either acetazolamide or placebo in a double‐blind fashion" Page 462. However, authors reported "several participants reported increased urination and suspected that they were receiving acetazolamide" Page 463, this situation may have influenced results like acute mountain sickness score

Blinding of outcome assessment (detection bias)
All outcomes

High risk

Not reported. However, authors reported "several participants reported increased urination and suspected that they were receiving acetazolamide" Page 463, this situation may have influenced results like the acute mountain sickness score

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

Unclear risk

Design bias: not sample size calculation

Methods

Two‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: climbers

Follow‐up period: 2 h

Scale used for assessing acute mountain sickness score

1. Lake Louise Acute Mountain Sickness (AMS) criteria

Scale used for assessing high altitude headache

1. Visual analogue scale (VAS)

Participants

Number of participants randomized: 74

Sex: men = 30 (40%)

Age: mean 33 years (range 13 to 61)

Baseline data

1. AMS score: Ibuprofen = 5.9 (SD not reported); paracetamol = 5.9 (SD not reported)

2. VAS score: Ibuprofen = 4.9, CI 95% 4.1 to 5.7); paracetamol = 4.7, CI 95% 4.0 to 5.5

Inclusion criteria

1. "Trekkers experiencing headache"

Exclusion criteria

1. History of chronic headache disorder, migraine headache, NSAID/paracetamol allergy, previous use of same within the prior 8 h

No cases of HAPE or HACE were noted during the study period

Interventions

Intervention group 1 (n = 39)

1. Ibuprofen, route: oral; dose: 400 mg; frequency: not clearly reported, apparently a single dose

Intervention group 2 (n = 35)

1. Paracetamol, route: oral; dose: 1000 mg; frequency: not clearly reported, apparently a single dose

Cointerventions

1. Not reported

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Relief high altitude headache

2. VAS at time 0, 30, 60 and 120 min

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Nepal

Altitude setting: 4243 m

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of Interest: not reported

Funding/Support

1. "This work was supported through an unrestricted grant provided by McNeil CPC. Absolute control of study design, data acquisition, analysis, and interpretation, as well as manuscript preparation, resided exclusively with the named authors at all times." (Page 383)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "randomly assigned rapid‐release capsules" Page 384

Allocation concealment (selection bias)

Unclear risk

Quote: "each was given an envelope containing a detailed history questionnaire, followed by four separate, identical pages containing 10 cm visual analogue scales (VAS). The envelope also contained identical, randomly assigned rapid‐release capsules..." Page 384.

Comments: it is not stated whether the envelope was opaque or not

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Insufficient information about the blinding of participants and personnel to permit judgment of 'Low risk' or High risk'

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Insufficient information about the blinding of participants and personnel to permit judgment of 'Low risk' or High risk'

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Lost after randomization: 1 (1%) ("after choosing to leave the study area (decided to hike further during the day"), study group not reported. Outcome data (all outcomes) were available for the rest of the participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

Unclear risk

Design bias: not sample size calculation

Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: low risk of bias

Follow‐up period: 18 h

Scale used for assessing acute mountain sickness score

1. Lake Louise Acute Mountain Sickness criteria, severity of High Altitude Headache (HAH) based on visual analogue scale pain score (VAS)

Participants

Number of participants randomized: 24

Sex: men = 14 (58%)

Age: mean 29.1 years (SD = 1.7, range 18 to 50 years)

Inclusion criteria

1. ≥ 18 years, suffering high altitude headache before 24 h of ascent

Exclusion criteria

1. Severe cardiac, pulmonary or liver disease

2. Severely impaired kidney function

3. History of migraine

4. Current history of alcohol or drug abuse

5. Allergy to gabapentin

6. Treatment with anticonvulsants or tricyclic antidepressants

Interventions

Intervention group (n = 12)

1. Gabapentin; route: oral; dose: 300 mg; frequency: not clearly reported, apparently a single dose; duration of the intervention: not clearly reported, apparently a single dose

Control group (n = 12)

1. Placebo (monohydrate lactose identical capsule); route: oral; frequency: not clearly reported, apparently a single dose; duration of the intervention: not clearly reported, apparently a single dose

Co‐intervention

1. 400 mg ibuprofen after 1 h of gabapentin/placebo intake (page 1275)

Outcomes

Primary endpoints

1. Need of supplementary analgesics after 1 h of gabapentin/placebo

2. Severity of HAH based on VAS score

3. Duration of HAH‐free phase

Secondary endpoints

1. AMS incidence (Lake Louise score ≥ 3 with headache and any other symptom)

2. Incidence of severe AMS (Lake Louise score ≥ 5)

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS and adverse events

Notes

Country: Iran (Tochal Hotel)

Altitude setting: 3500 m
Identifier number: not reported

Study dates: 1 to 7 January and 10 to 20 February 2007

A priori sample estimation: no

Financial disclosures: not reported

Funding/Support

1. Darou Darman Pars Pharmaceuticals (providing gabapentin and placebo)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "the computer‐generated randomisation codes" (page 1275)

Allocation concealment (selection bias)

Low risk

Quote: "only the pharmacist who provided the drugs knew the details of the computer‐generated randomisation codes" (page 1275)

Quote: "medications were in identical opaque boxes labelled with randomisation codes that were not disclosed to investigators or assessor." (page 1275)

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Quote: "medications were in identical opaque boxes labelled with randomisation codes that were not disclosed to investigators or assessor." (page 1275)

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

Quote: "medications were in identical opaque boxes labelled with randomisation codes that were not disclosed to investigators or assessor." (page 1275)

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available.

Other bias

High risk

"We acknowledge Dr Alireza Madjd, managing director of Darou Darman Pars Pharmaceuticals, for providing gabapentin and placebo". There was no statement addressing the independence of authors with regard to those providing funding (source of industry bias)

Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: "patients were monitored for only one hour after treatment"

Diagnosis of AMS

1. "Patients with nausea or headache who had arrived to altitude within 72 hours". Also "AMS patients with mild HAPE as diagnosed by chest radiography and clinical examination"

Scale used for assessing acute mountain sickness

1. "A mild headache was assigned one point, and two points were given for a severe headache. Nausea was given one point"

Participants

Number of participants randomized: 29; "because of mechanical and technical errors, complete data were available in only 24 of the subjects"

Sex: men = 17 (71%)

Age: mean 37 years

Baseline data

1. 6 participants had AMS and HAPE (hyperbaric chamber = 3; supplementary oxygen = 3).

Inclusion criteria

1. Patients with AMS: "patients with nausea or headache who had arrived to altitude within 72 hours"

2. "AMS patients with mild HAPE as diagnosed by chest radiography and clinical examination"

Exclusion criteria

1. Severe altitude illness (requiring prompt evacuation to a lower‐altitude treatment facility)

2. Previous treatment with oxygen, acetazolamide, or dexamethasone

3. Acute or chronic heart or lung disease (not including HAPE)

4. Less than 18 years of age

5. Pregnancy or nursing mother

6. Evidence of acute upper respiratory infection

Interventions

Intervention group (n = 13)

1. Hyperbaric chamber (simulated descent of 1432 m) dose: 120 mmHg of pressurisation above ambient pressure; route: breathing air inside the chamber; frequency and duration: 2 h

Control group (n = 11)

1. Supplementary oxygen; dose: 4 L (30% to 35%); route: by facemask; frequency and duration: 2 h

Co‐intervention

1. None reported

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Symptom response: "symptoms of headache and nausea were monitored at 15‐minute intervals for the first hour and at 30‐minute intervals for an additional hour of treatment and for one hour after treatment". Symptoms were scored "using the same point system that was used for entry of patients into the study"

2. Speed of symptom resolution

3. Haemodynamic variables: blood pressure, heart rate, arterial oxygen saturation (SaO₂)

4. Complications

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

2. Adverse events

Notes

Country: USA (Snake River Health Clinic, Keystone, Colorado)

Altitude setting: 2850 m

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of interest: not reported

Funding/Support

1. This study was funded in part by a grant from DuPont de Nemours and Company, Inc, and the University of Colorado Department of Chemical Engineering

2. The chamber was donated to the institute by Hyperbaric Mountain Technologies Inc.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "patients agreeing to participate signed informed consent and then were randomly assigned to oxygen or hyperbaric treatment protocols." Page 1110

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "patients agreeing to participate signed informed consent and then were randomly assigned to oxygen or hyperbaric treatment protocols." Page 1110

Blinding of participants and personnel (performance bias)
All outcomes

High risk

The study was not blinded. Quote: "we did not attempt to blind either the oxygen or the hyperbaric therapy." Page 1111

Blinding of outcome assessment (detection bias)
All outcomes

High risk

The study was not blinded. Quote: "we did not attempt to blind either the oxygen or the hyperbaric therapy." Page 1111

Incomplete outcome data (attrition bias)
All outcomes

High risk

Outcome data was missing from: hyperbaric chamber group = 2 out of 13 (15%) participants; oxygen group = 3 out of 11 (27%) participants.

Reason: Quote: "because of mechanical and technical errors, complete data were available in only 24 of the subjects, and the remainder was excluded from data analysis." "These errors occurred in the monitoring equipment, not with the hyperbaric chamber" Page 1111

Selective reporting (reporting bias)

High risk

Protocol not available. Data presented graphically for individuals. No data available for each group for symptomatic scores

Other bias

High risk

Design bias: not sample size calculation

There was no statement considering the independence of authors with respect to those providing funding (source of industry bias)

Methods

Two‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: patients

Follow‐up period: at least 11 h

Scales for assessing acute mountain sickness score

1. Lake Louise score, clinical score, and AMS‐C score of the environmental symptom questionnaire of Sampson 1983

Participants

Number of participants randomized: 31 climbers with symptoms of acute mountain sickness

Sex: men = 22 (71%)

Age: mean 31.5 years

Inclusion criteria

1. Mountaineers planning to stay overnight; symptoms or signs of acute mountain sickness; clinical score of 3 or more for clinical acute mountain sickness

Exclusion criteria

1. Clinical signs of high altitude pulmonary oedema

Quote: "most subjects had ascended to high altitude without prior acclimatisation... by using a cable car to an altitude of 3200." (page 1232)

Interventions

Intervention group (n = 15)

1. Name: hyperbaric chamber (Certec, F‐69210 Sourcieux‐les‐Mines, France); dose: 193 mbar (equivalent to a descent of 2250 m); frequency: once during the trial; duration of the intervention: 1 h

Control group (n = 16)

1. Name: dexamethasone, route: oral administration; dose: 8 mg initially; frequency: 4 mg every 6 h; duration of the intervention: not clearly reported. Due to severe vomiting in 4 subjects the initial dose was administered intravenously

Co‐intervention

1. "After they entered the trial subjects were allowed to take mild analgesics (paracetamol) for headache, but this had to be reported to the investigator"

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Acute mountain sickness relief: Lake Louise score, clinical score and AMS‐C score. Measured before the intervention and after 1 h and 11 h

2. Permitted intake of mild analgesics before treatment and in the follow‐up period

3. Physiological variables: pulse rate, blood pressure and arterial oxygen saturation

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Swiss‒Italian border. Capanna "Regina Margherita" located at an altitude of in the Alps Valais

Altitude setting: 4559 m (barometric pressure 430 mmHg to 440 mmHg)
Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Financial disclosures: not reported

Funding/Support

1. Swiss National Science Foundation (grant 32‐33729.92)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "randomisation was performed in blocks of eight by drawing lots from an envelope containing the assignments of one block." Page 310

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "randomisation was performed in blocks of eight by drawing lots from an envelope containing the assignments of one block." Page 310

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

No blinding. Quote: "the volunteers completed a questionnaire on environmental symptoms' and the Lake Louise self assessment questionnaire directed towards the symptoms of acute mountain sickness. The responses were checked with the investigator, and subsequently a clinical examination for peripheral oedema, pulmonary rales, and ataxia (Romberg test and heel to toe walking test) was performed." "Interviews and clinical examinations were always performed by the same investigator" Page 1233

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

Unclear risk

Design bias: not sample size calculation

Methods

Two‐group parallel RCT, 1 centre

ITT: no
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: not clearly reported

Diagnosis of AMS

1. “Patients who met the mountain sickness diagnosis criteria according to the Chinese Medical Association (1996)”

Scale used for assessing acute mountain sickness

1. Lake Louise questionnaire

Participants

Number of participants randomized: 47

Sex: men = 47 (100%)

Age: mean 18 (range 16 to 21)

Baseline data

1. Lake Louise questionnaire: standard intervention group = 4.39 (SD = 2.31), nitric oxide Intervention Group = 4.43 (SD = 2.04)

Inclusion criteria

1. “The patients who met the mountain sickness diagnosis criteria according to the China Plateau Medical Association (1996)”

Exclusion criteria

1. Not reported

Interventions

Intervention group (n = 24)

1. Nitric oxide plus standard treatment; route: inhalation; dose: 0.001% nitric oxide, 3 L/min; nitric oxide inhalation balanced with air at 3658 m; frequency: twice/day; duration of the intervention: 1 h/time

Control group (n = 23)

1. Standard intervention group (oxygen inhalation; aminophylline; dexamethasone; furosemide); route, dose, frequency and duration of the intervention: details not reported

Co‐intervention

1. Not reported

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Change in Lake Louise Score

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Tibet, China

Altitude setting: 3658 m

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of Interest: not reported

Funding/Support

1. Not reported

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: “47 male participants were randomised into 2 groups”. Page 1631. Authors did not specify if a random sequence generation was used

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

Unclear risk

Design bias: not sample size calculation

Methods

Two‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: up to 12 h after medication

Diagnosis of AMS

1. "Medical history, history of headache at low altitude, headache score and Lake Louise AMS score "

Scale used for assessing Acute Mountain Sickness

1. Lake Louise AMS score (LL score)

2. Headache score

Participants

Number of participants randomized: 29

Sex: men = 23 (79%)

1. Sumatriptan group: 14 men

2. Placebo: 9 men + 6 women. Quote " ...all 6 women participating in the study were assigned to the placebo group" (page 389)

Age: mean 34.5 years (18 to 56 years)

Baseline data

1. Lake Louise Score: mean 6.1 +/‐ 1.4

2. Headache score: mean 2.4 +/‐ 0.5

Inclusion criteria

1. Moderate headache on a 4‐point scale (0 = none, 1 = light, 2 = moderate, 3 = severe)

2. Stay overnight at the hut

3. Written informed consent

Exclusion criteria

1. Age under 18 or over 55 years

2. Symptomatic ischaemic heart disease or heart disease of other aetiology

3. Multiple cardiovascular risk factors, Prinzmetal angina

4. Raynaud’s phenomenon

5. Diseases that influence metabolism or excretion of sumatriptan

6. Severe, acute mountain sickness with ataxia, focal neurological symptoms, or changes in mental status

7. Pregnancy or nursing women

8. Known intolerance to sumatriptan

9. Abuse of opioids or ergotamine analgesics

10. Abuse of alcohol or intake of > 15 grams of ethanol in the past 24 h

11. Co‐medication with serotonin‐interferent drugs

12. Intake of analgesics 4 h or ergotamine 24 h preceding the study

13. Acetazolamide, corticosteroid, or nifedipine medication

Interventions

Intervention group (n = 14)

1. Name: sumatriptan; route: oral; dose: 100 mg; frequency: every 3 h; duration of the intervention: not clearly reported

2. Control group (n = 15): placebo 100 mg; route: oral; dose: 100 mg; frequency and duration not clearly reported

Co‐intervention

1. Paracetamol (500 mg) if headache pain was not relieved 3 h after administration of the study drug. Quote: "When headache had improved within 1 hours after administration of the study drug but recurred after 3 hours, an oral dose of 100 mg of sumatriptan was given" (page 389)

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Improvement of high altitude headache ("defined as reduction of the headache score by 1 point"). Assessed at 1 h, 3 h, and 12 h after medication

2. Lake Louise AMS score (LL score)

3. Blood pressure

4. Heart rate

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Italy, Capanna "Regina Margherita" in the Alps Valais

Altitude setting: 4559 m (barometric pressure 430 mmHg to 440 mmHg)

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of interest: not reported

Funding/Support: the Sezione Varallo del Club Alpino Italiano and of the Glaxo‐Wellcome Company. Study drug was supplied by Glaxo‐Wellcome company (Bad Oldesole, Germany)

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "randomization, without stratification, was performed in blocks of 4 subjects." Page 389

Allocation concealment (selection bias)

Unclear risk

Insufficient information about the allocation concealment process to permit judgment of 'Low risk' or High risk'

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Quote: "...sumatriptan and placebo had identical appearance..." Page 389

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Insufficient information about blinding of outcome assessment to permit judgment of 'Low risk' or High risk'

Incomplete outcome data (attrition bias)
All outcomes

High risk

Outcome data was not available in:

1. Sumatriptan group: 4 out of 14 (29%) participants

2. Placebo group: 5 out of 15 (33%) participants

Reason:

1. severe acute mountain sickness (intense headache, vomiting, ataxia, or clouded consciousness)

2. acute illness of the examiner

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

High risk

Baseline differences:

Quote: # 1: "despite strict randomisation in blocks of 4 subjects, all 6 women participating in the study were assigned to the placebo group, resulting in a significant difference of gender distribution between treatment groups" Page 389
Quote: # 2: "subjects of the sumatriptan group were somewhat older (mean age 38 versus 31 yr),..." Page 389

There was no statement considering the independence of authors with respect to those providing funding (source of industry bias)

Methods

Three‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: not clearly reported, apparently until recovery

Diagnosis of AMS

1. Not reported

Scale used for assessing Acute Mountain Sickness

1. None

Participants

Number of participants randomized: 65 soldiers and railway workers

Sex: men = 65 (100%)

Age: mean 25 years

Baseline data

Symptom duration before recruitment: nifedipine: 9 days ± 3; nitric oxide: 8 days ± 3; conventional therapy: 8 days ± 3

Inclusion criteria

1. Patients diagnosed with high altitude pulmonary oedema. No additional inclusion criteria were applied

Exclusion criteria

1. Not reported

Interventions

Intervention group 1 (n = 24)

Name: nifedipine in addition to conventional therapy; nifedipine route: oral; dose: set at 20 mg at the first time, then 10 mg; frequency: every 8 h; duration of the intervention: until fully recovered

Intervention group 2 (n = 22)

Nitric oxide In addition to oral nifedipine. Nitric oxide (BG‐951, co‐developed by Guangzhou General Hospital and Beijing Factory of Analytical Machinery): dose: 10 ppm; route: inhalation, balanced with oxygen at 80% concentration level, inhalation rate was set at 8 L/min to 10 L/min; frequency and duration of the intervention: during 30 min

Control group (n = 19)

Conventional therapy: oxygen, intravenous furosemide, aminophylline and dexamethasone. Dose: not reported; route: inhalation in the case of oxygen; intravenous injection for furosemide, aminophylline and dexamethasone; frequency and duration: not reported

Co‐intervention

1. Penicillin, streptomycin were also used to prevent bacteria infection (mode of delivery: intramuscular injection)

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Disease course (duration of symptoms)

2. Time until pulmonary rales disappear

3. Time until shadows on chest radiograph disappear

Outcomes of interest in the review: none

Notes

Country: China (military hospital at Kunlun Mountain at Sinkiang province)

Altitude setting: 3700 m

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of interest: not reported

Funding/Support: Military Medical and Health Research Fund

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: “65 participants were randomised into 3 groups” Page 212, without specifying how the random sequence was generated

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of participants and personnel (performance bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

In the Results section, authors summarized, “All of the 65 participants were fully recovered”. Comment: outcome data were available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Protocol not available

Other bias

Low risk

No other sources of bias identified

Methods

*Results presented here correspond to the second of three experiments carried out by the researchers

Experiment 1

1. Corresponds to an expedition 1 "a preliminary experiment comparing the effect of acetazolamide with methazolamide on blood gases was done in ten subjects" (acute mountain sickness was not an inclusion criteria)

Experiment 2

1. Corresponds to an expedition 2 "A placebo‐controlled trial of acetazolamide for AMS was done in 13 subjects"

Experiment 3

1. Corresponds to an expedition 3 "randomised double blind comparison of methazolamide and dexamethasone. Data from the dexamethasone arm of the trial were insufficient and have been omitted"

Two‐group parallel RCT, 1 centre

ITT: yes
Overall study quality: high risk of bias

Unit of randomization: participants

Follow‐up period: five days

Diagnosis of AMS

1. "Headache, anorexia/nausea and insomnia were scored; 1 = mild, 2 = moderate and 3 = severe. A score of 3 was given for any degree of ataxia, confusion or disorientation. For entry into the acute therapy trial on these expeditions, a persistent AMS score of 3 determined at clinical interview was required"

Scale used for assessing Acute Mountain Sickness

1. Self administered questionnaire of 18 questions (0: not present, 5: extreme) and a maximum score of 180

Participants

Number of participants randomized: 13

Sex: not reported for experiment 2

Age: not reported for experiment 2. They reported subjects aged 22 to 58 for the three experiments (see methods above)

Baseline data

1. Self‐administered AMS questionnaires

2. Placebo: 21.6 (+/‐ 11)

3. Acetazolamide: 33.3 (+/‐ 13.7)

Inclusion criteria

1. Healthy; non obese; unacclimatized subjects; persistent AMS score of 3

Exclusion criteria

1. Not reported

Interventions

Intervention group (n = 6)

1. Acetazolamide; route: oral; dose: 20 mg Kg^‐1 (1 to 1.5 grams) initially and then 500 mg daily

Control group (n = 7)

1. Placebo: "all drugs and placebo were prepared in identical gelatin capsules"

Co‐intervention

1. None reported

Outcomes

Not pre‐fixed as 'primary' or 'secondary'

1. Headache worsening (proportion of participants per group)

2. Response to acute therapy: using self‐administered AMS questionnaires of 18 questions scored 0 (not present) to 5 (extreme) (mean, SD and proportion with improvement)

3. Blood gases

4. Cerebral blood flow after allocation, before the administration of study drug and 20 h to 24 h later

Outcomes of interest in the review

1. Reduction in illness severity scores of AMS

Notes

Country: Karakoram mountains, located in the borders between Pakistan, India and China

Altitude setting: 3200 to 5486 m

Identifier number: not reported

Study dates: not reported

A priori sample estimation: no

Conflicts of Interest: not reported

Funding/Support

1. "The work was supported by grants from the Arthur Thomson and the Wellcome Trusts." "Lederle Laboratories UK kindly supplied the acetazolamide"

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "subjects were randomly allocated on a double‐blind basis" Page 51

Allocation concealment (selection bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'

Quote: "subjects were randomly allocated on a double‐blind basis" Page 51

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Quote: "all drugs and placebo were prepared in identical gelatin capsules" Page 51

However, blinding was discontinued "Six of the placebo group were given 1.5 grams oral acetazolamide 24 hours after entry into the trial because AMS symptoms had persisted. At this point all subjects were aware of their treatment status" Page 51

Blinding of outcome assessment (detection bias)
All outcomes

High risk

Quote: "all drugs and placebo were prepared in identical gelatin capsules" Page 51

However, blinding was discontinued "Six of the placebo group were given 1.5 grams oral acetazolamide 24 hours after entry into the trial because AMS symptoms had persisted. At this point all subjects were aware of their treatment status" Page 51

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All outcomes: no withdrawals. Outcome data was available for all participants

Selective reporting (reporting bias)

Unclear risk

Insufficient information to permit judgment of 'Low risk' or 'High risk'. Protocol not available

Other bias

High risk

Some participants in experiment 2 could have participated in the other two experiments (see methods above). Quote: "23 were studied during one of the three expeditions, six in two expeditions and three subjects in all three expeditions." Page 50. There is not enough information regarding how many participants from the second expedition were involved in the other two, and how much time passed between one expedition and another to identify a carry‐over effect.

There was no statement considering the independence of authors with respect to those providing funding (source of industry bias)