. 2017 Mar 17;15:64. doi: 10.1186/s12916-017-0823-7

Table 4.

Summary table of findings of update randomised controlled trials and their relevance to the meta-review questions

Reference and weighting; Participants, n; Risk of bias	Comparison	Relevance to meta-review questions:	Study type and interventions included	Target group(s)	Main results [1^o] is the defined primary outcome
Reference and weighting; Participants, n; Risk of bias	Comparison	What is the impact? Target groups? Which components? Context?	Study type and interventions included	Target group(s)	Main results [1^o] is the defined primary outcome
Al-Sheyab 2012 [48] n = 261 HIGH risk of bias	Adolescent Asthma Action programme vs. standard care. FU: 3 mo	Target: Adolescents Components: Peer education	Cluster RCT. Triple A. Peer leaders from year 11 were trained to deliver programme to years 8, 9 and 10.	Adolescents in Jordanian high school. I group had fewer females, fewer symptoms and higher English proficiency.	Compared to control improvements QoL score improved [I: 5.42 (SD 0.14) vs C: 4.07 (SD 0.14) MD 1.35 (95%CI 1.04–1.76)].
Baptist 2013 [49] n = 70 HIGH risk of bias	Personalised asthma self-regulation intervention vs. education session. FU 12 mo	Target: Older adults Components: Health educator	RCT. 6-session programme (group telephone). Patients selected an asthma-specific goal, and addressed potential barriers. Control is single session basic education + 2 telephone calls.	Aged ≥65 y. Physician diagnosis of asthma, no restriction in severity. Majority Caucasian.	No between-group differences in A&E visits or hospitalisations. Healthcare utilisation was lower at 6 mo but not 12 mo. ACQ was similar at 1 mo and 6 mo. At 12 mo, I participants were 4.2 times more likely to have an ACQ score <0.75. [1^o] QoL (mAQLQ) was significantly higher in the I than in C at all time points (1, 6 and 12 mo).
Ducharme 2011 [50] n = 219 LOW risk of bias	‘Take-home plan’ post A&E visit with PAAP + prescription information vs. prescription but no PAAP/information. FU: 28 days	Target: Children, A&E attendees Components: PAAP with prescription	RCT. Intervention is written PAAP with a ‘formatted’ prescription for ICS (i.e. including information about use) issued by A&E doctor on discharge following asthma exacerbation.	Canadian children 1–17 y recruited during A&E attendance for acute asthma (78% were under the age of 6 y).	No between-group differences in unscheduled care at 28 days. Compared to control, at 28 days children given the PAAP had better asthma control (proportion with Asthma Quiz Score <2 I: 58% vs. C: 41%; RR 1.36, 95% CI 1.04–1.86). No between-group differences in child/caregiver QoL at 28 days. [1^o] Adherence to ICS declined from 90% (day 1) to 50% at day 14, with no significant group difference.
Goeman 2013 [51] n = 114 Low risk of bias	Person-centred education vs. written information. FU: 12 mo	Target: Older adults Components: Personalised education	RCT. Personally tailored education session with asthma educator based on responses to a questionnaire; inhaler technique.	≥55 y, community-based asthmatics with no restriction in asthma severity.	[1^o] At 12 mo I participants had better asthma control than C (ACQ MD 0.3, 95% CI 0.06–0.5, p = 0.01) and better asthma-related QoL (p = 0.01). No significant difference in number of steroid courses (p = 0.17). At 12 mo, more I participants (n = 36, 61%) owned a PAAP compared to C (n = 21, 38%; p = 0.015). [1^o] Similar adherence to ICS at 12 mo (p = 0.015).
Halterman 2014 [52] n = 638 LOW risk of bias	Personalised prompts for clinicians and parents, practice training and feedback vs. written guidelines. FU: 6 mo	Target: Children, deprived communities Components: Feedback Context: Community-based, clinical training	Cluster RCT. Intervention practices received personalised clinician and parent prompts + blank PAAP; practice training; feedback. Control practices sent guidelines.	Urban, primary care practices in deprived communities. Parents/children 2–12 y with persistent, poorly controlled asthma. Recruited from waiting room over 4 ystudy.	11% in both groups had an A&E visit or hospitalisation. [1^o] Compared to control practices, at 2 mo children in the PAIR-UP practices had more symptom-free days [I: 10.2 days/2 weeks (SD 4.8) vs. C: 9.5 days/2 weeks (SD 5.1); MD 0.78, 95% CI 0.29–1.27] but the difference was not significant at 6 mo. Nights with symptoms remained significant at 6 mo [I: 1.4 (SD 3.0) vs. C: 1.8 (SD 3.2); MD −0.43; 95% CI −0.77 to −0.09].
Horner 2014 [53] n = 183 UNCLEAR risk of bias	Asthma plan for kids vs. teaching on general health and well-being. FU: 12 mo	Target: Children, rural communities	Cluster RCT. Programme delivered in 16 × 15 min sessions, 3 days/week for 5.5 weeks, by school nurses during lunch break + home visit.	Grades 2–5 (ages 7–11 y) with physician diagnosis of asthma.	No between-group difference for admissions or A&E visits. No between-group difference in QoL scores. Inhaler skill improved in the intervention group compared to control after 4 mo, with reported higher self-efficacy.
Joseph 2013 [54] n = 422 UNCLEAR risk of bias	Web-based asthma management intervention vs. control. FU: 12 mo	Target: Adolescents, urban deprived, ethnic groups Components: Web-based, behavioural change	RCT. Internet-based programme targeted at African-Americans/urban adolescents with traits (low motivation; low perceived emotional support; resistance to change; rebelliousness).	Grades 9–12 (ages 14–18 y) with physician diagnosis of asthma and report >4 days of restricted activity in the past 30 days at baseline.	No difference in reported A&E visits/hospitalisations at 12 mo. [1^o] Compared to C, at 12 mo the I participants had fewer symptom-days (RR 0.8, 95% CI 0.6–1.0). No difference in nights with symptoms, schooldays missed, days of restricted activity or days had to change plans. Students characterised with rebelliousness or low perceived emotional support reported fewer symptom-days.
Khan 2014 [55] n = 91 HIGH risk of bias	Asthma education + individualised written PAAP vs. asthma education (excluding PAAP).	Target: Ethnic groups Components: Written PAAP	RCT. Both groups received individual asthma education during an OPD visit from a paediatrician + monthly FU. Intervention group trained in using a PAAP.	1–14 y. Recruited via A&E OPD with partly controlled asthma (daytime or nocturnal symptoms, activity limitation, lung function < 0% best or exacerbation in previous year).	[1^o] Trend for improved outcomes at 6 mo but no significant between-group difference in proportion of children attending A&E (I: 36% vs. C: 52%; p = 0.141). There was no between-group difference in unscheduled doctor visits, asthma attacks, missed school days or night-time awakenings.
Rhee 2011 [56] n = 112 UNCLEAR risk of bias	Peer-led asthma education provided by peers at a day camp vs. adult-led camp.	Target: Adolescents. Components: Peer leaders	RCT. Asthma self-management skills + psychosocial skills taught at a day camp by peer leaders + monthly peer telephone contact. Control: Similar education delivered by adults. No telephone.	13–17 y (including low-income families). Mild/moderate/severe asthma. Asthma diagnosis for 1 y. Able to understand spoken and written English.	[1^o] Both groups reported significantly increased QoL over time (F = 4.31, p = 0.002), with I group having significantly higher QoL at 6 mo (MD 11.38, 95% CI 0.96–21.79, p = 0.03) and 9 mo (MD 12.97, 95% CI 3.46–22.48, p = 0.008). Both groups reported improved attitude to asthma (F = 11.94, p = 0.001), with greater improvement in I at 6 mo (MD 4.11, 95% CI 0.65–7.56, p = 0.02).
Rikkers-Mutsaerts 2012 [57] n = 90 UNCLEAR risk of bias	Internet-based self –management vs. usual care. FU: 12 mo	Target: Adolescents. Components: Internet-based	RCT. Internet-based self-monitoring with algorithm-based advice. Programme included education (web-based + group), self-monitoring (FEV₁ + ACQ), PAAP and 3–6 mo review.	12–18 y with mild to severe persistent asthma on regular ICS medication and poorly controlled at recruitment.	No between-group differences in exacerbations, physicians’ visits or telephone contacts. [1^o] QoL was better in I group at 3 mo (PAQLQ I: 6.00 vs. C: 5.68; MD 0.40, 95% CI 0.17–0.62) but not at 12 mo (I: 5.93 vs. C: 6.05; MD 0.05, 95% CI 0.50–0.41). Asthma control was improved in I group at 3 mo (ACQ I: 0.96 vs. C: 1.19; MD −0.32, 95% CI −0.56 to −0.08) but not at 12 mo (I: 0.83 vs. C: 0.79; MD −0.05, 95% CI −0.35 to 0.25).
Shah 2011 [58] 150 GPs and 201 children LOW risk of bias	GP training (PACE study) vs. no training. FU: 12 mo	Targets: Children Components: GP training	Cluster RCT. GPs participated in 2 × 3-h workshops on communication and education strategies to facilitate quality asthma care.	150 GPs and 221 children with asthma in their care.	No between-group difference in hospitalisation/A&E visits (I: 18% vs. C: 12%; difference 6%, 95% CI −4 to 15). No between-group differences in school absence or parent absenteeism for child’s asthma. [1^o] More patients in I group GPs had a PAAP (I: 61% vs. C: 46%; difference 15%, 95% CI 2–28).
van Gaalen 2013 [59] n = 107 HIGH risk of bias	Internet-based self –management vs. control (FU of SMASHING trial). FU: 30 mo	Target: Adults Components: Internet-based	RCT (FU study). Education + PAAP, self-monitoring and regular review. The 200 patients in original 12-mo trial were invited for FU after 18 mo.	Adults with asthma aged 18–50 y, using ICS. 107/200 (54%) participated: I group: 47/101 (47%); C group: 60/99 (61%). Participants ACQ was similar, but AQLQ was greater than in non-participants.	At 30 mo after baseline, there was a slightly attenuated improvement for both QoL (AQLQ adjusted between-group MD 0.29, 95% CI 0.01–0.57) and ACQ (adjusted MD of −0.33, 95% CI −0.61 to −0.05) scores in favour of the intervention. No between-group differences in FEV₁.
Wong 2012 [60] n = 80 HIGH risk of bias	Symptom-based written PAAP vs. verbal counselling. FU: 6 mo	Target: Children, ethnic groups Components: Written PAAP	Single blinded RCT. Intervention was symptom-based PAAP given out at initial contact. Outcomes measured at baseline, 3, 6 and 9 mo.	Malaysian children (mix of Malay, Chinese and Indian) with all severities of asthma. Aged 6–17 y. Recruitment process not described.	At 6 mo there was no difference in A&E visits/unscheduled care [intervention 4 (SD 10.8) vs. control 6 (SD 21.1); p = 0.35]. At 6 mo there was no difference in proportion controlled (ACT ≥ 20 I: 81% vs. C: 87%; p = 0.50), with no exacerbations (ACT ≥ 20 I: 89% vs. C: 82%; p = 0.62) or in QoL [mean PAQLQ I: 6.11 (SD 0.88) vs. 6.11 (SD 1.09); p = 0.99].

Abbreviations: A&E accident and emergency, ACQ Asthma Control Questionnaire, ACT Asthma Control Test, AQLQ Asthma-related Quality of Life Questionnaire, C control, CI confidence interval, FEV ₁ forced expiratory volume in one second, FU follow-up, GP general practitioner, I intervention, ICS inhaled corticosteroid, mAQLQ mini Asthma-related Quality of Life Questionnaire, MD mean difference, mo months, PAAP personalised asthma action plan, PAQLQ paediatric asthma-related quality of life, QoL quality of life, RCT randomised controlled trial, RR risk ratio, SD standard deviation, y years