Administración intermitente versus diaria de corticosteroides inhalados para el asma persistente en niños y adultos
Resumen
Antecedentes
La administración diaria de corticosteroides inhalados (CSI) es la base del tratamiento recomendado en niños y adultos con asma persistente. Sin embargo, a menudo los pacientes utilizan CSI de forma intermitente o los médicos recomiendan su administración solamente al comienzo de las exacerbaciones.
Objetivos
El objetivo de esta revisión fue comparar la eficacia y la seguridad de la administración intermitente versus diaria de CSI en el tratamiento de los niños y los adultos con asma persistente y de los niños en edad preescolar con sospecha de asma persistente.
Métodos de búsqueda
Se hicieron búsquedas en el registro especializado de ensayos del Grupo Cochrane de Vías Respiratorias (Cochrane Airways Group, CAGR) y en el sitio web de ClinicalTrials.gov hasta octubre de 2012.
Criterios de selección
Se incluyeron los ensayos controlados aleatorios (ECA) que compararon la administración intermitente versus diaria de CSI en niños y adultos con asma persistente. No se permitieron cointervenciones con otros calmantes y corticosteroides orales de rescate utilizados durante las exacerbaciones.
Obtención y análisis de los datos
Dos revisores evaluaron forma independiente los ensayos para determinar su inclusión y su calidad y extrajeron los datos. El resultado primario de eficacia fue el número de pacientes con una o más exacerbaciones que requirieron corticosteroides orales y el resultado primario de seguridad fue el número de pacientes con eventos adversos graves de salud. Los resultados secundarios incluyeron exacerbaciones, pruebas de la función pulmonar, control del asma, efectos adversos, tasas de retiros y marcadores inflamatorios. Se asumió que existía equivalencia cuando la estimación del cociente de riesgos (CR) y su intervalo de confianza (IC) del 95% estaban entre 0,9 y 1,1. La calidad de las pruebas se evaluó mediante GRADE.
Resultados principales
Seis ensayos (incluido un ensayo que probó dos protocolos relevantes) cumplieron con los criterios de inclusión para un total de siete comparaciones de grupos. Los cuatro ensayos pediátricos (dos con niños en edad preescolar y dos con niños en edad escolar) y dos ensayos de grupos paralelos en adultos, con una duración de 12 a 52 semanas, fueron de calidad metodológica alta. Un total de 1211 pacientes con asma persistente confirmada o presunta contribuyeron a los metanálisis. No hubo diferencias estadísticamente significativas entre los grupos en cuanto al riesgo de que los pacientes experimentaran una o más exacerbaciones que requirieran la administración de corticosteroides orales (1204 pacientes; CR 1,07; IC del 95%: 0,87 a 1,32; el intervalo de confianza amplio se traduce en un riesgo de exacerbaciones en el grupo de CSI intermitente que varía entre 17% y 25%, lo que supone un riesgo del 19% con la administración diaria de CSI). La edad, la gravedad de la obstrucción de las vías respiratorias, el protocolo de incremento gradual utilizado durante las exacerbaciones y la duración de los ensayos no influyeron significativamente en el resultado primario de eficacia. No se observaron diferencias entre los grupos en cuanto al riesgo de que los pacientes experimentaran eventos adversos graves de salud (1055 pacientes; CR 0,82; IC del 95%: 0,33 a 2,03). En comparación con el grupo de administración diaria de CSI, el grupo de administración intermitente de CSI mostró una mejoría más pequeña en el cambio desde el inicio en la tasa de flujo espiratorio máximo (TFEM) del 2,56% (IC del 95%: ‐4,49% a ‐0,63%), menos días sin síntomas (diferencia de medias estandarizada [DME] ‐0,15; IC del 95%: ‐0,28 a ‐0,03), menos días de control del asma (‐9%; IC del 95%: ‐14% a ‐4%), más uso de β2‐agonistas de rescate (0,12 inhalaciones/día; IC del 95%: 0 a 0,23) y un aumento mayor desde el inicio en el óxido nítrico exhalado de 16,80 partes por mil millones (IC del 95%: 11,95 a 21,64). No hubo diferencias significativas entre los grupos en el volumen espiratorio forzado en un segundo (VEF1), la calidad de vida, la hiperreactividad de las vías respiratorias, los efectos adversos, las hospitalizaciones, las visitas al departamento de urgencias ni los retiros. En los ensayos pediátricos, la administración intermitente de CSI (budesonida y beclometasona) se asoció con un mayor crecimiento, cambio de 0,41 cm desde el inicio (532 niños; IC del 95%: 0,13 a 0,69) en comparación con el tratamiento diario.
Conclusiones de los autores
En los niños y adultos con asma persistente y en los niños en edad preescolar con sospecha de asma persistente hubo pruebas de baja calidad de que las estrategias de administración intermitente y diaria de CSI fueron igual de efectivas en cuanto al uso de corticosteroides orales de rescate y la tasa de eventos adversos graves de salud. La solidez de las pruebas significa que actualmente no es posible asumir una equivalencia entre las dos opciones. La administración diaria de CSI fue superior a la administración intermitente de CSI en varios indicadores de la función pulmonar, la inflamación de las vías respiratorias, el control del asma y el uso de calmantes. Ambos tratamientos parecieron seguros, aunque se observó una supresión moderada del crecimiento asociada con la administración diaria de budesonida y beclometasona inhaladas, en comparación con su administración de forma intermitente. Los médicos deben sopesar cuidadosamente los efectos beneficiosos y perjudiciales potenciales de cada opción de tratamiento y considerar el impacto desconocido a largo plazo(> un año) del tratamiento intermitente en el crecimiento pulmonar y el deterioro de la función pulmonar.
PICOs
Resumen en términos sencillos
¿La administración de corticosteroides inhalados solamente cuando los síntomas de asma empeoran es igual de segura y efectiva que la administración diaria de corticosteroides inhalados?
El asma crónica es una enfermedad causada por la inflamación subyacente de las vías respiratorias (tubos pequeños en los pulmones) y las crisis asmáticas ocurren cuando las vías respiratorias se contraen, lo cual dificulta la respiración en el paciente. En los pacientes con asma leve, a menudo se recomienda la administración diaria de corticosteroides inhalados para controlar la inflamación subyacente. Sin embargo, muchos pacientes con asma utilizan corticosteroides inhalados solo cuando aparecen síntomas. Se deseó analizar las pruebas disponibles de los ensayos controlados aleatorios que compararon la administración diaria de corticosteroides inhalados con el uso de estos fármacos solamente en el momento del empeoramiento de los síntomas en niños y adultos con asma persistente (seis ensayos con 1211 pacientes).
Esta revisión de los ensayos controlados aleatorios no encontró diferencias significativas en el número de crisis asmáticas de gravedad moderada entre los pacientes que recibieron corticosteroides inhalados todos los días y los que los recibieron "cuando fue necesario". Sin embargo, no hubo información suficiente para concluir que los dos enfoques fueron equivalentes. Se encontró que los pacientes que recibieron corticosteroides inhalados diariamente tuvieron un control del asma ligeramente mejor y una mejor función pulmonar, menos uso de calmantes y más días sin síntomas que los que recibieron corticosteroides inhalados de forma intermitente. También se observó que en comparación con la administración intermitente de corticosteroides inhalados, los niños crecieron ligeramente menos con la administración diaria de budesonida y beclometasona inhaladas (se sabe que los corticosteroides inhalados afectan el crecimiento), lo que destaca la importancia del uso de una dosis más segura y más baja de corticosteroides inhalados. No se observaron diferencias significativas entre los grupos en la tasa de retiros ni los efectos adversos. Estos resultados no proporcionan conclusiones firmes, aunque la mejoría en el control del asma, la función pulmonar y la inflamación de las vías respiratorias apoyarían ligeramente más el uso diario de corticosteroides inhalados comparado con su administración solo cuando los síntomas empeoran. Se recomienda que los médicos y los pacientes analicen cuidadosamente los riesgos y los efectos beneficiosos de cada opción de tratamiento y vigilen la respuesta de los pacientes individuales para ajustar el tratamiento según sea necesario.
Authors' conclusions
Summary of findings
| Intermittent 'as needed' ICS versus daily ICS for persistent asthma in children and adults | ||||||
| Patient or population: Children and adults with persistent asthma. | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Daily ICS | Intermittent ICS | |||||
| Patients with 1 or more exacerbations requiring oral corticosteroids (duration 12 to 52 weeks) | 19 per 100 | 20 (17 to 25) per 100 | RR 1.07 | 1204 | ⊕⊕⊝⊝ | |
| Patients with serious adverse health events (duration 12 to 52 weeks) | 2 per 100 | 2 (1 to 4) per 100 | RR 0.82 (0.33 to 2.03) | 1055 (6 studies) | ⊕⊕⊝⊝ | |
| Patients with at least 1 exacerbation requiring emergency department/acute care visit (duration 12 to 52 weeks) | 15 per 100 | 18 (14 to 20) per 100 | RR 1.08 | 1055 | ⊕⊕⊝⊝ | |
| Change from baseline AM PEFR (%) (duration 44 to 52 weeks) | The mean change from baseline am PEFR (%) in the intermittent groups was 2.56% lower than daily ICS group | MD ‐2.56 (‐4.49 to ‐0.63) | 350 | ⊕⊕⊕⊝ | ||
| Proportion of asthma control days over the period (duration 44 to 52 weeks) | The mean proportion of asthma control days over the period in the intermittent groups was 9% lower than daily ICS group (4% to 14%) | MD ‐0.09 (‐0.14 to ‐0.04) | 330 | ⊕⊕⊕⊝ | ||
| Change from baseline mean daily use of β2‐agonists (puffs/day) (duration 24 to 44 weeks) | The mean change from baseline mean daily use of β2‐agonists (puffs/day) in the intermittent groups was 0.12 puffs/day higher than daily ICS group | MD 0.12 (0.00 to 0.23) | 442 | ⊕⊕⊕⊝ moderate2 | ||
| Change in height (cm) (duration 44 to 52 weeks) | The mean change in height (cm) in the intermittent groups was 0.41 cm higher than daily ICS group (0.13 to 0.69 higher) | MD 0.41 (0.13 to 0.69) | 532 | ⊕⊕⊕⊝ | ||
| Overall withdrawals (duration 12 to 52 weeks) | 14 per 100 | 14 (11 to 19) per 100 | OR 1.05 | 1210 | ⊕⊕⊝⊝ | |
| *The basis for the assumed risk was the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1. Confidence intervals were too wide to exclude important differences between treatments for this outcome. 2. Diversity of ways in which intermittent ICS regimens were implemented and the variety of participants in the trials limits the confidence in our estimate of the treatment effects | ||||||
Background
National and international asthma guidelines recommend daily inhaled corticosteroids (ICS) as the mainstay of treatment in children and adults with mild persistent asthma as well as in preschool children in whom persistent asthma is suspected or predicted (Australia 2006; BTS 2011; GINA 2011; Lougheed 2012; NAEPP 2011). However, people often discontinue their recommended daily ICS treatment when asymptomatic and re‐start treatment when deemed required, that is, at the onset of exacerbations. Although frequent, sub‐optimal adherence to asthma medication is often underestimated by physicians as a cause of poor asthma control in individual patients (Lasmar 2009). A number of risk factors for non adherence have emerged including low socioeconomic status (Blais 2006), fear of side effects associated with corticosteroids (Rabe 2004), the erroneous conception that no symptoms equate no disease (Halm 2006), forgetfulness, etc. Faced with the difficulty of convincing patients to be adherent to daily controller therapy or when unsure of the inefficacy of intermittent therapy (Lasmar 2009), many physicians have endorsed intermittent therapy, prescribing short‐term ICS to be initiated at the onset of an exacerbation as part of a self‐management plan rather than daily controller therapy.
With the publication of several trials comparing intermittent to daily ICS in children (Martinez 2011; Papi 2009; Turpeinen 2008; Zeiger 2011) and adults (Boushey 2005; Papi 2007) with persistent asthma, we believed that a systemic review evaluating the efficacy and safety of this approach was timely, considering the popularity of intermittent therapy with patients and physicians.
Objectives
To compare the efficacy and safety of daily ICS versus intermittent ICS in the management of children and adults with persistent asthma. In addition, we wished to identify the characteristics of patients or treatment that may influence the magnitude of response attributable to either strategy.
Methods
Criteria for considering studies for this review
Types of studies
We included randomised controlled trials (RCTs) comparing daily versus intermittent ICS (that is, initiated for a short duration only at the onset of exacerbations) in children and adults with asthma; only trials with a parallel‐group design were eligible.
Types of participants
We included children aged one to 18 years and adults with persistent asthma, including preschool children with suspected, or at risk of developing, persistent asthma.
Types of interventions
The treatment group received patient‐initiated intermittent ICS at the onset of exacerbations and continued for a short duration. The control group received daily ICS between and during exacerbations. We allowed trials in which the dose of ICS was stepped‐up or was unequal between groups during exacerbations, or both. No co‐interventions were permitted other than rescue β2‐agonists and oral corticosteroids. We specifically excluded use of any long‐acting β2‐agonists (e.g. formoterol), as part of daily or step‐up management.
Types of outcome measures
The following outcomes were considered.
Primary outcomes
-
Patients experiencing one or more exacerbations requiring rescue oral corticosteroids.
-
Patients with serious adverse health events.
Secondary outcomes
-
Exacerbations requiring hospital admission (and number of people experiencing one or more events).
-
Exacerbations requiring an emergency department visit or unscheduled acute care visit (and number of people experiencing one or more events).
-
Time to exacerbation requiring rescue oral corticosteroids.
-
Clinical or physiological outcomes reflecting chronic asthma control including period or change in: pulmonary function tests, symptoms, night‐time awakening, quality of life, β2‐agonist use, days with/without symptoms and days with/without β2‐agonist use.
-
Biological markers of inflammation as change in: eosinophil count in blood and sputum, leukotrienes in biological fluids, expired nitric oxide, eosinophilic cationic protein, etc.
-
Safety profile comprised of overall and specific adverse effects, including growth and adrenal suppression.
-
Withdrawal rates: overall withdrawals, withdrawals due to poor asthma control and withdrawals due to adverse effects;
Search methods for identification of studies
Electronic searches
We identified trials from the Cochrane Airways Group Specialised Register of trials (CAGR), which is derived from systematic searches of bibliographic databases including the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, CINAHL, AMED, and PsycINFO, and handsearching of respiratory journals and meeting abstracts (see Appendix 1 for further details). All records in the CAGR coded as 'asthma' were searched using the following terms: (intermittent* or as‐needed* or "as needed" or prn or irregular* or occasional* or sporadic* or short‐course*) and (daily* or regular* or routine*).
We also conducted an advanced search of ClinicalTrials.gov using 'intermittent' as keyword, 'asthma' as condition and 'interventional studies' as study type. All databases were searched from their inception to October 2012 and there was no restriction on language of publication.
Searching other resources
We checked reference lists of all primary studies and reviewed articles for additional references. We contacted authors of identified included trials for additional data and to confirm quality of methodology, and asked them to identify other similar published and unpublished studies.
Data collection and analysis
Selection of studies
From the titles, abstracts or descriptors, one review author (BFC) reviewed the literature searches. Those that were clearly not RCTs or did not fit the inclusion criteria were excluded. Two review authors (BFC and FMD) obtained full‐text articles for all potentially eligible trials and independently assessed them for inclusion. We resolved any disagreement through discussion.
Data extraction and management
Two review authors independently extracted data (BFC and FMD or CC) and disagreements were dealt with by consensus between the two review authors and with the input of a third review author, when needed.
Assessment of risk of bias in included studies
Two review authors (BFC and FMD or CC) independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Any disagreements were resolved by discussion or by involving the third review author. We assessed the risk of bias according to the following domains:
-
random sequence generation;
-
allocation concealment;
-
blinding of participants and personnel;
-
blinding of outcome assessment;
-
incomplete outcome data;
-
selective outcome reporting; and
-
other bias.
We graded each potential source of bias as high, low or unclear risk of bias.
Measures of treatment effect
We calculated treatment effects for dichotomous variables as risk ratio (RR) or risk difference (RD), or both with 95% confidence interval (CI). We arbitrarily assumed equivalence if the RR estimate and its 95% CI were between 0.9 and 1.1. For continuous outcomes, such as pulmonary function tests, we calculated pooled statistics as mean differences (MD) or standardised mean differences (SMD) with 95% CI. We summarised differences between groups in event rates using rate ratios.
Unit of analysis issues
The unit of analysis was the participant.
Dealing with missing data
Where possible, we contacted investigators or study sponsors to obtain missing numerical outcome data. We did not use imputation to replace missing data.
Assessment of heterogeneity
We tested homogeneity of outcome effect measures between studies being meta‐analysed using both the Chi2 test for heterogeneity (Cochrane Q test) and the I2 statistic; P < 0.10 in the former or an I2 > 40% was deemed indicative of significant heterogeneity. If significant heterogeneity was present, a random‐effects model was used to aggregate the outcome effect measures, otherwise, a fixed‐effect model was reported.
Assessment of reporting biases
We used funnel plots to examine the possibility of publication bias and reported the Egger test (Egger 1997). We also planned to perform sensitivity analysis by excluding trials with poor methodology and those unpublished to examine the possibility of bias.
Data synthesis
The analysis compared intermittent versus daily ICS. In case of a trial with more than one intervention or control group, we considered additional comparisons, if appropriate. In such cases, we halved the number of participants in the group that served twice as comparator to avoid over‐representation. For dichotomous outcomes, we halved both the numerator and denominator of the group that served twice as comparator.
We performed the meta‐analysis using Review Manager (RevMan 2011).
Subgroup analysis and investigation of heterogeneity
We planned to perform subgroup analyses to explore possible reasons for heterogeneity of study results and, in the absence of heterogeneity, to identify potential effect modifiers, that is, variables that change the magnitude of benefit using the following a priori defined subgroups:
-
age (preschool children versus school‐aged children versus adults) (preschool children being a subgroup added post hoc);
-
severity of asthma obstruction at baseline as determined by the mean group percent of predicted forced expiratory volume in one second (FEV1) classified as mild (FEV1 ≥ 80%), moderate (FEV1 61% to 79%), or severe (FEV1 ≥ 60%);
-
protocols used during exacerbations (pre‐emptive four‐fold step‐up ICS dose in both groups, pre‐emptive four‐fold step‐up ICS dose in the intermittent ICS group only, 'as needed' ICS plus β2‐agonist use in both groups and 'as needed' ICS plus β2‐agonist use in the intermittent ICS group only) (added post hoc);
-
trial duration (12 to 24 weeks versus 44 to 52 weeks).
Sensitivity analysis
For the primary outcome, we planned to perform two sensitivity analyses: (1) one to determine the effect of study methodological quality by removing trials with inadequate or unclear information on random sequence generation, blinding procedure, or a large or differential loss to follow‐up in both groups and (2) the other to explore the impact of publication status, by removing the unpublished trials. Potential publication bias was explored by using a funnel plot and the Egger test (Egger 1997).
Results
Description of studies
Results of the search
The search for literature conducted until to December 2011 identified a total of 206 citations and abstracts through database searching and 26 citations from clinicaltrials.gov. Of them, 16 full‐text potential trials were reviewed and finally six trials (seven comparisons) were included for the meta‐analysis (Figure 1). We updated the literature search in October 2012. There were 6 additional references, but no new included studies.
Study flow diagram.
Included studies
Six trials (Boushey 2005; Martinez 2011a; Martinez 2011b; Papi 2007; Papi 2009; Turpeinen 2008; Zeiger 2011) reporting seven comparisons and enrolling 1211 patients with confirmed or suspected persistent asthma, met the inclusion criteria for this review. All trials were published in full text. We described below the characteristics of all included trials.
Design: all trials had a parallel‐group design.
Participants: two comparisons involved preschool children (aged one to five years) (Papi 2009; Zeiger 2011), three comparisons involved school‐aged children (aged five to 18 years) (Martinez 2011a; Martinez 2011b; Turpeinen 2008) while two trials involved only adult patients (aged 18 to 65 years) (Boushey 2005; Papi 2007). With the exception of the two preschool‐aged trials, all studies enrolled individuals with symptomatic mild persistent asthma (although one paediatric trial admitted that, in retrospect, their participants probably had mild or moderate airway obstruction at baseline) (Turpeinen 2008). The remaining two trials pertained to preschool children with recurrent wheezing in whom persistent asthma was not confirmed; one trial enrolled only children considered at high risk of subsequent asthma (based on the Asthma Predictive Index (Castro‐Rodriguez 2000),a scoring system to predict the probability to develop asthma at six years of age) (Zeiger 2011). Most trials described a gender ratio varying between 38% to 69% males. Three comparisons (Martinez 2011a; Martinez 2011b; Papi 2009) reported atopy in 36% to 61% of participants. Participants were stepped down to placebo and as needed β2‐agonist in three comparisons (Martinez 2011a; Martinez 2011b; Papi 2007) while the rest of the trials used anti‐asthma treatments to meet inclusion criteria for the run‐in period of two to four weeks.
Intervention duration: the duration of intervention varied from 12 to 52 weeks. The intermittent treatment was used for one to two weeks during exacerbations.
Intervention drugs: the ICS used was either budesonide or beclomethasone. During exacerbations, the following step‐up protocols were used: (1) pre‐emptive four‐fold ICS dose step‐up to 800 to 1600 μg/day in both groups (Boushey 2005; Turpeinen 2008); (2) pre‐emptive four‐fold ICS dose step‐up to 2000 μg/day in intermittent group only (Zeiger 2011), (3) 'as needed' ICS plus β2‐agonist in both groups (Martinez 2011a) and (4) 'as needed' ICS plus β2‐agonist in intermittent group only (Martinez 2011b; Papi 2007; Papi 2009). The information on dose, drug, device and duration of treatment is provided in the Characteristics of included studies table.
Co‐intervention: no trials reported the use of additional anti‐asthma drugs other than 'as‐needed' short‐acting inhaled β2‐agonists and rescue oral corticosteroids for treatment of symptoms or exacerbations, or both. In three trials, inhaled β2‐agonists were a prompt to use ICS: in two comparisons, inhaled salbutamol was used simultaneously with, but using a separate device than, beclomethasone (Martinez 2011a; Martinez 2011b); in two trials, salbutamol was used in the same device as the ICS, namely an inhaler of beclomethasone 250/salbutamol 100 μg (Papi 2007) and a nebule containing beclomethasone 800/salbutamol 1600 μg (Papi 2009). In other words, in these three trials (four comparisons), the dose of ICS during exacerbations was not fixed, but rather was driven by patients' need for rescue β2‐agonists. In one trial, oral corticosteroids could be initiated by patients as part of a written self‐management plan (Boushey 2005), and the trigger for self‐initiation of oral corticosteroids is described in the Characteristics of included studies table. The remaining trials appeared to require a physician to initiate a course of corticosteroids (Martinez 2011a; Martinez 2011b; Papi 2007; Papi 2009; Turpeinen 2008; Zeiger 2011).
Outcomes: the primary outcomes, namely the number of patients experiencing one or more exacerbations requiring rescue oral corticosteroids and patients with serious adverse health events were reported in the original publication (Boushey 2005) or directly obtained from authors (Martinez 2011a; Martinez 2011b; Papi 2007; Papi 2009; Turpeinen 2008; Zeiger 2011) and therefore available for all trials. Other measures of asthma exacerbation (e.g. hospital admissions, emergency department visits, time to exacerbations), change in asthma control (e.g. lung function tests, symptoms, use of rescue β2‐agonist, asthma control days, quality of life), airway hyper‐reactivity (PC20 (provocative concentration of a substance causing a 20% fall in FEV1)), airway inflammation (exhaled nitric oxide) as well as withdrawals and adverse health events were reported or obtained from authors in several trials; albeit not always in a format or with sufficient details to allow aggregation.
Excluded studies
The search strategy conducted until December 2011 yielded a total of 226 citations. Of these, 220 citations were excluded for the following mutually exclusive reasons: duplicate references (N = four); not RCTs (N = 20); subjects did not have asthma (N = 87); patients had an acute asthma exacerbation or were treated in emergency department, or both (N = four); daily ICS was not one of the tested strategies (N = 59); intermittent ICS was not one of tested strategies (N = 41); patients received non‐permitted co‐interventions (N = four); treatment was administered for less than four weeks (N = one). We added 6 new excluded studies from the update search run in Oct 2012. The references and reasons for exclusion are provided in the Characteristics of excluded studies table.
Risk of bias in included studies
Full details of the risk of bias for each trial can be found in the Characteristics of included studies tables. A graphical summary of our 'Risk of bias' judgements can be found in Figure 2. All trials described their method of randomisation.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Allocation
All but one trial provided sufficient information to confirm allocation concealment either in the published manuscript or in the protocol provided upon request.
Blinding
All trials reported double‐blinding with convincing details either in the published manuscript or in the protocol provided upon request. This implied that all trials used double‐dummies and that patients had different sets of (active or placebo) inhalers/nebules to be used between exacerbations than those used during exacerbations.
Incomplete outcome data
All trials reported the number and reasons of withdrawals in both groups. With one exception (Zeiger 2011), the proportion of withdrawals was small (< 10%) and well balanced between groups. In the trial reported by Zeiger et al, an overall withdrawal rate of 23% was observed with an imbalance in withdrawals between the two groups (19% versus 28%).
Selective reporting
All trials reported all outcomes mentioned in the method section, without any apparent bias.
Other potential sources of bias
We did not encounter any other significant source of bias in the included trials.
Effects of interventions
Primary outcomes
Efficacy: patients with exacerbations requiring oral corticosteroids
There was no statistically significant group difference in the number of patients experiencing one or more exacerbations requiring oral corticosteroids (1204 patients; RR 1.07; 95% CI 0.87 to 1.32; fixed‐effect model). In other words, with a 19% risk of exacerbation in patients treated with daily ICS, intermittent ICS treatment would lead to between 17 and 25% of participants experiencing an exacerbation. The magnitude of effect was not significantly affected by the different protocols used at the time of exacerbation (test for subgroup differences: Chi2 = 0.41; degrees of freedom (df) = 3; P = 0.94; I2 = 0%) (Analysis 1.1; Figure 3), age (preschool children versus school‐aged children versus adults; test for subgroup differences: Chi2 = 2.64; df = 2; P = 0.27; I2 = 24.2%) (Analysis 1.2), duration of trials (12 to 24 weeks versus 44 to 52 weeks; test for subgroup differences: Chi2 = 0.01; df = 1; P = 0.93; I2 = 0%) (Analysis 1.3), and severity of airway obstruction (mild versus moderate airway obstruction; test for subgroup differences: Chi2 = 0.81; df = 1; P = 0.37; I2 = 0%) (Analysis 1.4). There was no evidence of systematic bias identified by the test for funnel plot asymmetry (Figure 4) (intercept: ‐1.07; 95% CI ‐4.23 to 2.10). Because all trials were published and were of high methodological quality, we could not perform sensitivity analyses to assess bias due to poor methodology or publication status.
Forest plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.1 Patients with 1 or more exacerbations requiring oral corticosteroids.
Funnel plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.1 Patients with 1 or more exacerbations requiring oral corticosteroids.
Safety: patients with serious adverse health events
There was no statistically significant group difference in serious adverse health events (1055 patients; RR 0.82; 95% CI 0.33 to 2.03; fixed‐effect model) (Analysis 1.5; Figure 5) with no apparent effect of the pre‐emptive ICS step‐up protocols (Chi2 = 1.08; df = 3; P = 0.78). There was no heterogeneity across trials.
Forest plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.2 Patients with serious adverse health events.
Secondary outcomes
Exacerbations
There was no statistically significant group difference in: the number of patients with exacerbation requiring emergency department visits (1055 patients; RR 1.08; 95% CI 0.90 to 1.30) (Analysis 1.6), the number of patients experiencing at least one exacerbation requiring hospital admission (1204 patients; RR 0.85; 95% CI 0.29 to 2.49) (Analysis 1.7); the number of exacerbations (event rate) requiring emergency department visits (264 patients; RR 0.69; 95% CI 0.14 to 3.44) (Analysis 1.8), and the time to first exacerbation requiring oral corticosteroids (492 patients; HR 0.88; 95% CI 0.55 to 1.40; random‐effects model) (Analysis 1.9).
Lung function
There was a statistically significant group difference in the percent change from baseline in the peak expiratory flow rate (PEFR) (350 patients; MD ‐2.56%; 95% CI ‐4.49% to ‐0.63%) (Analysis 1.10). No group difference was observed in the percent change from baseline in FEV1 (365 patients; MD ‐0.49%; 95% CI ‐5.82% to 4.84%; random‐effects model) (Analysis 1.12) and mean diurnal variation in PEFR (442 patients; MD 0.02; 95% CI ‐0.71 to 0.76) (Analysis 1.14). The paucity of studies reporting these outcomes prevented the aggregation of data for absolute change from baseline in PEFR (L/min) (Analysis 1.11), FEV1 (L) (Analysis 1.13) and PC20 (Analysis 1.15).
Asthma control
There was a statistically significant group difference in disfavour of intermittent ICS compared to daily ICS in the change from baseline in asthma control days (214 patients; MD ‐7% days; 95% CI ‐14% to ‐1%) (Analysis 1.16), the proportion of asthma control days (330 patients; MD ‐9%; 95% CI ‐14% to ‐4%) (Analysis 1.17), the change from baseline in the mean daily use of β2‐agonists (442 patients; MD 0.12 puffs/day; 95% CI 0.00 to 0.23) (Analysis 1.18), cumulative dose of rescue albuterol over the period (214 patients; MD 51.47 μg; 95% CI 11.36 to 91.57) (Analysis 1.19) and the change in the proportion of symptom‐free days (984 patients; SMD ‐0.15; 95% CI ‐0.28 to ‐0.03) (Analysis 1.22). No statistically significant group difference was observed in: the change from baseline in daytime symptoms scores (591 patients; SMD 0.13; 95% CI ‐0.04 to 0.29) (Analysis 1.20), the change from baseline in night‐time awakenings (448 patients; MD ‐3%; 95% CI ‐8% to 2%) (Analysis 1.23) and quality of life (389 patients; MD ‐0.16; 95% CI ‐0.36 to 0.04) (Analysis 1.24). The meta‐analysis of the change from baseline in the asthma symptoms utility index (Analysis 1.21) and in percent days with β2‐agonist use (Analysis 1.25).could not be performed because of the paucity of data.:
Airway inflammation
A statistically significant group difference was observed in the change in exhaled nitric oxide favouring daily over intermittent ICS (214 patients; MD 16.80 parts per billion (ppb); 95% CI 11.95 to 21.64) (Analysis 1.26).
Safety profile
Withdrawals
There was no statistically significant group difference in overall withdrawals (1210 patients; OR 1.05; 95% CI 0.75 to 1.46) (Analysis 1.32); withdrawals due to poor asthma control (1063 patients; OR 1.66; 95% CI 0.54 to 5.13) (Analysis 1.37) or withdrawals owing to adverse effects (1063 patients; OR 0.78; 95% CI 0.21 to 2.93) (Analysis 1.34).
Adverse effects
As expected by the protocol, the cumulative dose of ICS over the period (μg HFA‐beclomethasone), (214 patients; MD ‐16,334.36 μg; 95% CI ‐19,189.04 to ‐13,479.69) (Analysis 1.35) was statistically higher in the daily ICS group compared to intermittent group.
There was no statistically significant group difference in overall adverse effects (726 patients; OR 1.01; 95% CI 0.70 to 1.44) (Analysis 1.36) or any of the specific effects: oral candidiasis (343 patients; OR 0.30; 95% CI 0.03 to 2.95) (Analysis 1.33), nausea (393 patients; OR 1.16; 95% CI 0.54 to 2.51) (Analysis 1.39), upper respiratory tract infections (393 patients; OR 1.47; 95% CI 0.89 to 2.43) (Analysis 1.41). However, a statistically significant group difference was observed favouring intermittent over daily inhaled budesonide or beclomethasone in the change in height from baseline over 44 to 52 weeks (532 patients; MD 0.41 cm; 95% CI 0.13 to 0.69) (Analysis 1.27). A meta‐analysis could not be performed due to paucity of data or events for: change in weight (kg) (Analysis 1.30), pneumonia (Analysis 1.38), headache (Analysis 1.40), change in salivary cortisol (Analysis 1.31) and death (Analysis 1.42).
Discussion
This meta‐analysis, based on four paediatric trials (with five comparisons) and two adult trials did not identify a statistically significant difference in the risk of patients experiencing exacerbations requiring oral corticosteroids or serious adverse health events between patients allocated to daily ICS versus intermittent ICS. The results of our analyses did not meet our a priori definition of equivalence. We observed statistically significant group differences in favour of daily ICS on several markers of asthma control (asthma control days, use of β2‐agonists, cumulative dose of rescue albuterol and symptom‐free days), certain markers of lung function (e.g. PEFR), and airway inflammation (exhaled nitric oxide). However, these benefits were achieved at the cost of a small, statistically significant, suppression of growth.
No firm conclusions can be made regarding the superiority or equivalence of these dosing strategies in preventing exacerbations requiring patient‐ or physician‐initiated systemic corticosteroids. While the absence of a statistically significant effect in exacerbations requiring rescue oral corticosteroids might suggest that intermittent and daily ICS are equally effective , they cannot be viewed as equivalent: the wide confidence interval around the pooled risk ratio of 1.07 means that the rate of exacerbations may be reduced by as much as 17% or increased by as much as 32% with intermittent compared therapy. Assuming a control group risk of 19% for exacerbations in people treated with daily ICS, intermittent ICS treatment would lead to between 17 and 25% participants experiencing an exacerbation. .
Despite the difficulty in diagnosing persistent asthma in preschool‐aged children (and consequently, an expected greater risk of misclassification), age groups did not significantly impact on the magnitude of effect. Equally, duration of intervention, baseline severity of airway obstruction and the step‐up protocols during exacerbation did not appear to influence the magnitude of response. The latter finding is in line with the conclusion of a Cochrane Review , in which a self‐initiated increase in ICS to 1000 to 2000 μg/day at the onset of an exacerbation was not associated with a statistically significant reduction in the risk of an exacerbation requiring rescue oral corticosteroids compared to usual dose of ICS (Quon 2010). This observation may explain why step‐up protocols to four‐fold ICS doses in both groups (Boushey 2005; Turpeinen 2008) or in only in the intermittent group (Zeiger 2011) or using ICS whenever relief with β2‐agonist was needed, did not reduce the risk of exacerbations. The notable absence of significant heterogeneity within and between step‐up protocols in our review supports the argument against differences in step‐up protocols as important determinants of the observed response.
With regards to asthma control between exacerbations, seven out of 17 secondary efficacy outcomes supported daily over intermittent ICS therapy, namely the change in PEFR, change in asthma control days, the proportion of asthma control days, use of β2‐agonists, cumulative use of rescue albuterol (μg), symptom‐free days and the level of exhaled nitric oxide. The remaining 10 secondary efficacy outcomes showed no significant group difference, possibly due to insufficient power. Admittedly, the magnitude of benefit of daily over intermittent ICS was clinically more important on asthma symptoms (7% more asthma control days and 9% more symptom‐free days) than on lung function (2.56% higher PEFR) or rescue β2‐agonists (0.12 puffs/day), which were of dubious clinical significance. The increase in exhaled nitric oxide would suggest more eosinophilic airway inflammation with intermittent compared to daily therapy. We suspect that the two preschool‐aged studies, which primarily selected children on the basis of repeated wheezing would or could have included a significant proportion of children with intermittent asthma in whom the best treatment remains to be determined. This is particularly true for Zeiger 2011 that specifically enrolled children with no or minimal interim symptoms and to a lesser extent Papi 2009 that recruited patients with symptoms over the 2‐week run‐in but who improved significantly within 2 weeks even on placebo, raising the possibility of recruitment during the symptomatic period following a previous wheezing episode and/or a regression towards the mean. Indeed, in children with intermittent viral‐induced asthma, intermittent high‐dose ICS have been demonstrated to be superior to placebo (Ducharme 2009) and daily ICS not superior to placebo (McKean 2000) or intermittent montelukast (Bacharier 2008). The mixed population of children with intermittent and persistent asthma in Zeiger 2011 may have diluted the effect. Indeed, with the exception of Zeiger 2011, all other included trials concluded that daily ICS were superior to intermittent ICS. Although it might be anticipated that age, severity, criteria used to initiate rescue treatment, and step‐up protocols during exacerbations would have led to variation between the study results, most secondary outcomes in our review showed little statistical heterogeneity. In view of these potential sources of variation, the occasionally modest, yet consistent, findings that support daily over intermittent ICS for maintaining control between exacerbations, constitute supportive evidence to the superiority of daily over intermittent ICS for maintaining asthma control.
No difference was observed in withdrawal rates and in overall, or specific adverse effects between both approaches. It must be noted that some adverse effects typically associated with daily ICS, such as osteopenia and adrenal suppression, were not measured in most of trials. However, a statistically significant, yet small, growth suppression was identified in three paediatric trials in groups using a daily nebulised/inhaled budesonide or beclomethasone at doses of 100 to 500 μg/day over the period of 44 to 52 weeks. The observed suppression was concordant with, yet smaller than, the observed growth suppression of 1.54 cm/year and 1.1 cm/year with 400 μg of daily inhaled beclomethasone and 200 μg of daily inhaled budesonide, respectively, in placebo‐controlled paediatric trials (Sharek 2000; The Childhood Asthma Program Research Group 2000). The lower than expected group difference on growth may suggest: (1) that both daily and intermittent ICS impact on growth, (2) a lower growth suppression is attributed to a lower dose of daily ICS dose, or (3) that enrolled children in these studies were not comparable to those enrolled in placebo‐controlled trials. Unfortunately, none of the included paediatric trials used newer molecules such as fluticasone, ciclesonide or mometasone. Until new trials comparing daily to intermittent ICS using newer molecules are available, care must be taken not to attribute the growth suppression to the daily strategy without recognising that existing trials used molecules with a known growth suppression effect. Importantly, as included trials were of a maximum of one year, the long‐term impact of intermittent ICS on lung growth (children) and lung function decline over the next five to 10 years and the potential impact on airway remodelling remain important concerns in children and adults.
Summary of main results
The six included trials (four paediatrics; two adults), lasting 12 to 52 weeks (1211 patients) tested four exacerbation protocols, namely pre‐emptive four‐fold ICS dose step‐up or 'as needed' ICS plus β2‐agonist, either in both groups or only in the intermittent group. There was no significant group difference in patients experiencing at least one exacerbation requiring oral corticosteroids. The magnitude of effect did not appear to be significantly influenced by the step‐up protocol used during exacerbations, age, duration of intervention or the severity of baseline airway obstruction. Compared to daily ICS, the intermittent ICS group displayed fewer symptom‐free days, less improvement in asthma control days and greater increase in exhaled nitric oxide as well as statistically significant, but modest, lower improvement in PEFR and higher rescue β2‐agonists use. There was no significant group difference in severe adverse events, FEV1, quality of life, airway hyper‐reactivity, adverse effects, hospitalisations, emergency department visits and withdrawals. However, the beneficial effects of daily budesonide or beclomethasone was achieved at the cost of lower linear growth in children (278 preschool children and 330 school‐aged children) over 44 to 52 weeks (MD 0.41 cm; 95% CI 0.13 to 0.69).
Overall completeness and applicability of evidence
This review summarises the best evidence available up to October 2012 derived from six trials (1211 patients with suspected or confirmed persistent asthma) of high methodological quality. The results pertain to children and adults with persistent asthma, and preschoolers with repeated wheezing suspected of persistent asthma. The systematic search to identify eligible trials and unpublished reports minimise the risk of inclusion bias. The outstanding collaboration of the authors/funders of six of the seven comparisons (Martinez 2011a; Martinez 2011b; Papi 2007; Papi 2009; Turpeinen 2008; Zeiger 2011) allowed us to obtain additional unpublished data and confirmation of methodological quality which strengthened the meta‐analysis. Due to the paucity of trials or the absence of events, 11 of 37 secondary outcomes could not be aggregated. While study authors reported enrolling patients with confirmed or suspected persistent asthma, the criteria used in paediatric trials (frequency of exacerbations with or without atopy, family history of asthma and eosinophilia) may have included an unknown proportion of preschool children with intermittent viral‐induced asthma that may have diluted the effect. The review is heavily weighted towards preschool‐ and school‐aged children, with only two trials pertaining to adults. The long‐term impact of intermittent versus daily ICS on lung growth, airway remodelling, bone mineralisation and adrenal function in children and lung function decline in adults beyond one year of follow‐up remain to be addressed.
Quality of the evidence
The included trials were of high methodology and were generally at low risk of bias. The confirmation of methodology by almost all authors or funders (with supportive evidence such as study protocols) and the provision of additional unpublished data allowed more precise estimates. The quality of evidence for our key outcomes reflects a lack of power from the studies that we included in the analysis (statistical imprecision) and variation in the different approaches used (indirectness).
Potential biases in the review process
None were identified.
Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
Forest plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.1 Patients with 1 or more exacerbations requiring oral corticosteroids.
Funnel plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.1 Patients with 1 or more exacerbations requiring oral corticosteroids.
Forest plot of comparison: 1 Intermittent 'as needed' ICS versus daily ICS, outcome: 1.2 Patients with serious adverse health events.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 1 Patients with 1 or more exacerbations requiring oral corticosteroids.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 2 Exacerbations requiring oral corticosteroids: subgroup analysis for age.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 3 Exacerbations requiring oral corticosteroids: subgroup analysis for duration of treatment.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 4 Exacerbations requiring oral corticosteroids: subgroup analysis for severity of airway obstruction.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 5 Patients with serious adverse health events.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 6 Patients with at least 1 exacerbation requiring emergency department/acute care visit.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 7 Patients with at least 1 exacerbation requiring hospital admission.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 8 Number of exacerbations requiring emergency department visits.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 9 Time to exacerbation requiring oral corticosteroids.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 10 Change from baseline AM PEFR (%).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 11 Change from baseline AM PEFR (L/min).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 12 Change from baseline FEV1 (%).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 13 Change from baseline FEV1 (L).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 14 Change from baseline mean diurnal variation in PEFR.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 15 Change in PC20.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 16 Change from baseline in asthma control days.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 17 Proportion of asthma control days over the period.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 18 Change from baseline mean daily use of β2‐agonists (puffs/day).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 19 Cumulative dose of rescue albuterol (μg) over the period.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 20 Change from baseline in daytime symptom scores.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 21 Change from baseline in asthma symptoms utility index.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 22 Change in proportion of 'symptom‐free days'.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 23 Change from baseline in night‐time awakenings.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 24 Change from baseline in quality of life (QoL).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 25 Days with use of β2‐agonists (%).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 26 Change in exhaled nitric oxide (ppb).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 27 Change in height (cm).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 28 Change in height (Z‐score).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 29 Change in weight (Z‐score).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 30 Change in weight (kg).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 31 Salivary cortisol level.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 32 Overall withdrawals.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 33 Oral candidiasis.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 34 Withdrawal owing to adverse events.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 35 Cumulative dose of ICS over the period (μg beclomethasone equivalent).
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 36 Overall adverse effects.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 37 Withdrawal owing to poor asthma control/exacerbations.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 38 Pneumonia.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 39 Nausea.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 40 Headache.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 41 Upper respiratory tract infections.
Comparison 1 Intermittent 'as needed' ICS versus daily ICS, Outcome 42 Death.
| Intermittent 'as needed' ICS versus daily ICS for persistent asthma in children and adults | ||||||
| Patient or population: Children and adults with persistent asthma. | ||||||
| Outcomes | Illustrative comparative risks* (95% CI) | Relative effect | No of participants | Quality of the evidence | Comments | |
| Assumed risk | Corresponding risk | |||||
| Daily ICS | Intermittent ICS | |||||
| Patients with 1 or more exacerbations requiring oral corticosteroids (duration 12 to 52 weeks) | 19 per 100 | 20 (17 to 25) per 100 | RR 1.07 | 1204 | ⊕⊕⊝⊝ | |
| Patients with serious adverse health events (duration 12 to 52 weeks) | 2 per 100 | 2 (1 to 4) per 100 | RR 0.82 (0.33 to 2.03) | 1055 (6 studies) | ⊕⊕⊝⊝ | |
| Patients with at least 1 exacerbation requiring emergency department/acute care visit (duration 12 to 52 weeks) | 15 per 100 | 18 (14 to 20) per 100 | RR 1.08 | 1055 | ⊕⊕⊝⊝ | |
| Change from baseline AM PEFR (%) (duration 44 to 52 weeks) | The mean change from baseline am PEFR (%) in the intermittent groups was 2.56% lower than daily ICS group | MD ‐2.56 (‐4.49 to ‐0.63) | 350 | ⊕⊕⊕⊝ | ||
| Proportion of asthma control days over the period (duration 44 to 52 weeks) | The mean proportion of asthma control days over the period in the intermittent groups was 9% lower than daily ICS group (4% to 14%) | MD ‐0.09 (‐0.14 to ‐0.04) | 330 | ⊕⊕⊕⊝ | ||
| Change from baseline mean daily use of β2‐agonists (puffs/day) (duration 24 to 44 weeks) | The mean change from baseline mean daily use of β2‐agonists (puffs/day) in the intermittent groups was 0.12 puffs/day higher than daily ICS group | MD 0.12 (0.00 to 0.23) | 442 | ⊕⊕⊕⊝ moderate2 | ||
| Change in height (cm) (duration 44 to 52 weeks) | The mean change in height (cm) in the intermittent groups was 0.41 cm higher than daily ICS group (0.13 to 0.69 higher) | MD 0.41 (0.13 to 0.69) | 532 | ⊕⊕⊕⊝ | ||
| Overall withdrawals (duration 12 to 52 weeks) | 14 per 100 | 14 (11 to 19) per 100 | OR 1.05 | 1210 | ⊕⊕⊝⊝ | |
| *The basis for the assumed risk was the mean control group risk across studies. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect (and its 95% CI). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1. Confidence intervals were too wide to exclude important differences between treatments for this outcome. 2. Diversity of ways in which intermittent ICS regimens were implemented and the variety of participants in the trials limits the confidence in our estimate of the treatment effects | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Patients with 1 or more exacerbations requiring oral corticosteroids Show forest plot | 7 | 1204 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.87, 1.32] |
| 1.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 2 | 264 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.96 [0.44, 2.13] |
| 1.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.03 [0.80, 1.34] |
| 1.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.17 [0.67, 2.03] |
| 1.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 555 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.20 [0.71, 2.04] |
| 2 Exacerbations requiring oral corticosteroids: subgroup analysis for age Show forest plot | 7 | 1204 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.87, 1.32] |
| 2.1 Preschool children | 2 | 498 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.09 [0.85, 1.41] |
| 2.2 School‐aged children | 3 | 329 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.26 [0.84, 1.88] |
| 2.3 Adults | 2 | 377 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.60 [0.26, 1.34] |
| 3 Exacerbations requiring oral corticosteroids: subgroup analysis for duration of treatment Show forest plot | 7 | 1204 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.87, 1.32] |
| 3.1 12 to 24 weeks | 2 | 448 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.13 [0.37, 3.39] |
| 3.2 44 to 52 weeks | 5 | 756 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.87, 1.32] |
| 4 Exacerbations requiring oral corticosteroids: subgroup analysis for severity of airway obstruction Show forest plot | 7 | 1204 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.87, 1.32] |
| 4.1 Mild airway obstruction | 6 | 1089 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.06 [0.86, 1.30] |
| 4.2 Moderate airway obstruction | 1 | 115 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.95 [0.32, 27.51] |
| 5 Patients with serious adverse health events Show forest plot | 6 | 1055 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.82 [0.33, 2.03] |
| 5.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.04, 3.06] |
| 5.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.25 [0.34, 4.56] |
| 5.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.65 [0.03, 15.53] |
| 5.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 555 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.10, 6.22] |
| 6 Patients with at least 1 exacerbation requiring emergency department/acute care visit Show forest plot | 6 | 1055 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.08 [0.90, 1.30] |
| 6.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.06, 15.34] |
| 6.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.08 [0.90, 1.30] |
| 6.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 6.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 555 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 7 Patients with at least 1 exacerbation requiring hospital admission Show forest plot | 7 | 1204 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.85 [0.29, 2.49] |
| 7.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 2 | 264 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.04, 3.06] |
| 7.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.25 [0.34, 4.56] |
| 7.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 7.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 555 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 8 Number of exacerbations requiring emergency department visits Show forest plot | 2 | 264 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.14, 3.44] |
| 8.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 2 | 264 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.69 [0.14, 3.44] |
| 9 Time to exacerbation requiring oral corticosteroids Show forest plot | 3 | Hazard Ratio (Random, 95% CI) | 0.88 [0.55, 1.40] | |
| 9.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Hazard Ratio (Random, 95% CI) | 0.97 [0.68, 1.38] | |
| 9.2 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | Hazard Ratio (Random, 95% CI) | 0.60 [0.36, 1.01] | |
| 9.3 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | Hazard Ratio (Random, 95% CI) | 1.93 [0.55, 6.76] | |
| 10 Change from baseline AM PEFR (%) Show forest plot | 3 | 350 | Mean Difference (IV, Fixed, 95% CI) | ‐2.56 [‐4.49, ‐0.63] |
| 10.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 136 | Mean Difference (IV, Fixed, 95% CI) | ‐1.20 [‐6.61, 4.21] |
| 10.2 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐2.42 [‐5.34, 0.50] |
| 10.3 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐3.10 [‐6.02, ‐0.18] |
| 11 Change from baseline AM PEFR (L/min) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 11.1 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 12 Change from baseline FEV1 (%) Show forest plot | 2 | 365 | Mean Difference (IV, Random, 95% CI) | ‐0.49 [‐5.82, 4.84] |
| 12.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 137 | Mean Difference (IV, Random, 95% CI) | ‐3.3 [‐6.49, ‐0.11] |
| 12.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 228 | Mean Difference (IV, Random, 95% CI) | 2.14 [‐0.40, 4.68] |
| 13 Change from baseline FEV1 (L) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 13.1 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 14 Change from baseline mean diurnal variation in PEFR Show forest plot | 3 | 442 | Mean Difference (IV, Fixed, 95% CI) | 0.02 [‐0.71, 0.76] |
| 14.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐0.14 [‐1.42, 1.14] |
| 14.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 335 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.79, 1.00] |
| 15 Change in PC20 Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 15.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 16 Change from baseline in asthma control days Show forest plot | 2 | 214 | Mean Difference (IV, Fixed, 95% CI) | ‐0.07 [‐0.14, ‐0.01] |
| 16.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐0.06 [‐0.15, 0.04] |
| 16.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐0.09 [‐0.18, 0.01] |
| 17 Proportion of asthma control days over the period Show forest plot | 3 | 330 | Mean Difference (IV, Fixed, 95% CI) | ‐0.09 [‐0.14, ‐0.04] |
| 17.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 116 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 17.2 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐0.09 [‐0.16, ‐0.02] |
| 17.3 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | ‐0.09 [‐0.16, ‐0.02] |
| 18 Change from baseline mean daily use of β2‐agonists (puffs/day) Show forest plot | 3 | 442 | Mean Difference (IV, Fixed, 95% CI) | 0.12 [0.00, 0.23] |
| 18.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 0.15 [‐0.05, 0.35] |
| 18.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 335 | Mean Difference (IV, Fixed, 95% CI) | 0.10 [‐0.04, 0.24] |
| 19 Cumulative dose of rescue albuterol (μg) over the period Show forest plot | 2 | 214 | Mean Difference (IV, Fixed, 95% CI) | 51.47 [11.36, 91.57] |
| 19.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 59.30 [4.83, 113.77] |
| 19.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 42.2 [‐17.06, 101.46] |
| 20 Change from baseline in daytime symptom scores Show forest plot | 3 | 591 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.13 [‐0.04, 0.29] |
| 20.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 143 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.15 [‐0.18, 0.48] |
| 20.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 448 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.12 [‐0.07, 0.30] |
| 21 Change from baseline in asthma symptoms utility index Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 21.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 22 Change in proportion of 'symptom‐free days' Show forest plot | 5 | 984 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.15 [‐0.28, ‐0.03] |
| 22.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 2 | 258 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.32 [‐0.57, ‐0.07] |
| 22.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Std. Mean Difference (IV, Fixed, 95% CI) | 0.0 [‐0.24, 0.24] |
| 22.3 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 448 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.15 [‐0.34, 0.03] |
| 23 Change from baseline in night‐time awakenings Show forest plot | 2 | 448 | Mean Difference (IV, Fixed, 95% CI) | ‐0.03 [‐0.08, 0.02] |
| 23.1 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 448 | Mean Difference (IV, Fixed, 95% CI) | ‐0.03 [‐0.08, 0.02] |
| 24 Change from baseline in quality of life (QoL) Show forest plot | 2 | 389 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.16 [‐0.36, 0.04] |
| 24.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 137 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.24 [‐0.58, 0.10] |
| 24.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 252 | Std. Mean Difference (IV, Fixed, 95% CI) | ‐0.11 [‐0.36, 0.14] |
| 25 Days with use of β2‐agonists (%) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 25.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 26 Change in exhaled nitric oxide (ppb) Show forest plot | 2 | 214 | Mean Difference (IV, Fixed, 95% CI) | 16.80 [11.95, 21.64] |
| 26.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 18.16 [11.09, 25.23] |
| 26.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 15.59 [8.94, 22.24] |
| 27 Change in height (cm) Show forest plot | 4 | 532 | Mean Difference (IV, Fixed, 95% CI) | 0.41 [0.13, 0.69] |
| 27.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 95 | Mean Difference (IV, Fixed, 95% CI) | 0.61 [0.12, 1.10] |
| 27.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 223 | Mean Difference (IV, Fixed, 95% CI) | 0.25 [‐0.17, 0.67] |
| 27.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 0.46 [‐0.37, 1.29] |
| 27.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 107 | Mean Difference (IV, Fixed, 95% CI) | 0.40 [‐0.43, 1.23] |
| 28 Change in height (Z‐score) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 28.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 29 Change in weight (Z‐score) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 29.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 30 Change in weight (kg) Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 30.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 31 Salivary cortisol level Show forest plot | 1 | Mean Difference (IV, Fixed, 95% CI) | Totals not selected | |
| 31.1 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 32 Overall withdrawals Show forest plot | 7 | 1210 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.05 [0.75, 1.46] |
| 32.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 2 | 264 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.54 [0.71, 3.34] |
| 32.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.59 [0.34, 1.04] |
| 32.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.90 [0.63, 5.74] |
| 32.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 561 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.32 [0.75, 2.30] |
| 33 Oral candidiasis Show forest plot | 2 | 343 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.30 [0.03, 2.95] |
| 33.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.32 [0.01, 8.07] |
| 33.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 228 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.29 [0.01, 7.12] |
| 34 Withdrawal owing to adverse events Show forest plot | 6 | 1063 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.78 [0.21, 2.93] |
| 34.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 117 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.01, 8.35] |
| 34.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.20 [0.01, 4.14] |
| 34.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 34.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 561 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.85 [0.29, 27.61] |
| 35 Cumulative dose of ICS over the period (μg beclomethasone equivalent) Show forest plot | 2 | 213 | Mean Difference (IV, Random, 95% CI) | ‐16334.36 [‐19189.04, ‐13479.69] |
| 35.1 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Mean Difference (IV, Random, 95% CI) | ‐17787.0 [‐20164.04, ‐15409.96] |
| 35.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 1 | 106 | Mean Difference (IV, Random, 95% CI) | ‐14874.0 [‐17269.16, ‐12478.84] |
| 36 Overall adverse effects Show forest plot | 3 | 726 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.70, 1.44] |
| 36.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.94 [0.46, 1.90] |
| 36.2 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 2 | 448 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.03 [0.68, 1.57] |
| 37 Withdrawal owing to poor asthma control/exacerbations Show forest plot | 6 | 1063 | Odds Ratio (M‐H, Random, 95% CI) | 1.66 [0.54, 5.13] |
| 37.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 117 | Odds Ratio (M‐H, Random, 95% CI) | 3.43 [0.88, 13.38] |
| 37.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Random, 95% CI) | 0.28 [0.08, 1.06] |
| 37.3 'As needed' ICS+β2‐agonist use during exacerbations in both groups | 1 | 107 | Odds Ratio (M‐H, Random, 95% CI) | 2.70 [0.68, 10.76] |
| 37.4 'As needed' ICS+β2‐agonist use during exacerbations in intermittent group only | 3 | 561 | Odds Ratio (M‐H, Random, 95% CI) | 2.44 [0.39, 15.41] |
| 38 Pneumonia Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 38.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 39 Nausea Show forest plot | 2 | 393 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.16 [0.54, 2.51] |
| 39.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.98 [0.23, 4.13] |
| 39.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.24 [0.50, 3.10] |
| 40 Headache Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Totals not selected | |
| 40.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] | |
| 41 Upper respiratory tract infections Show forest plot | 2 | 393 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.47 [0.89, 2.43] |
| 41.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 41.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.47 [0.89, 2.43] |
| 42 Death Show forest plot | 2 | 393 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 42.1 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in both groups | 1 | 115 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 42.2 Pre‐emptive 4‐fold step‐up ICS dose during exacerbations in intermittent group only | 1 | 278 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
