Joseph 2013.

Methods	Included as process evaluation Intervention study design: parallel‐group randomised controlled trial Unit of allocation: child Process evaluation methods: survey based, including multi‐variate analyses of outcomes
Participants	Setting: 6 high schools in Detroit, Michigan, USA Age of children: mean age, 15.6 years Child characteristics (BME/SES): 98% African American; 74% of children were in receipt of free or reduced price school meals Asthma status: asthmatic only; severity unclear Intervention recipients: children only
Interventions	School type: high school Intervention description: this is an adapted version of a tailored computer programme (Puff City) that was tested in Joseph 2010. Puff City focusses on 3 behaviours: controller medication adherence, keeping an inhaler nearby, and smoking reduction or cessation. This new intervention included new submodules designed to target teens with characteristics shown to be associated with lack of behaviour change in the previous trial, and who exhibited no change after 1 or more sessions. Students were provided 4 sessions in total Control description: controls received 4 sessions of generic asthma education to match the experience of students in the treatment group Theoretical framework: Puff City uses tailoring to apply behavioural theory. Also includes Health Belief Model, Attribution Theory, and motivational interviewing
Outcomes	Core processes evaluated (child level): attrition, dosage, adherence
Notes	Process evaluation category: stand‐alone and integrated (2 papers) Breadth and depth: breadth and depth Voice of children given prominence: featured but not sufficiently Note: study is not included as an outcome evaluation because the comparison group received asthma education (this study evaluated the added impact of providing tailored messaging) Funding source: National Institutes of Health; National Heart, Lung, and Blood Institute
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	N/A
Allocation concealment (selection bias)	Unclear risk	N/A
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	N/A
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	N/A
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	N/A
Selective reporting (reporting bias)	Unclear risk	N/A
Other bias	Unclear risk	N/A
Transparent and clearly stated aims	Low risk	Study aims were clearly stated
Explicit theories underpinning and/or literature review	Low risk	Behavioural theory informed the intervention
Transparent and clearly stated methods and tools	High risk	Information collected from caregivers is unclear
Selective reporting	High risk	Data were collected from caregivers, but study authors did not state what these data included
Harmful effects	Low risk	Subgroup analyses of the impact of potential risk group were undertaken
Population and sample described well	Unclear risk	More could have been done to describe caregivers
Continuous evaluation	Low risk	Pre‐post assignment data were collected
Evaluation participation equity and sampling	Unclear risk	Data collected are unclear
Design and methods overall approach	Unclear risk	Data were collected but were not presented; a clear outline of the research design was not presented
Tools and methods of data collection reliable/credible	High risk	Information collected from caregivers is unclear
Tools and methods of data analysis reliable/credible	Low risk	Data presented were analysed in a straightforward way
Performance bias/neutrality/credibility/conformability	Unclear risk	Steps taken to address this are unclear
Reliability of findings and recommendations	High risk	Very low proportion of eligible students took part, so reliability of study findings was compromised
Transferability of findings	Unclear risk	Subgroup analyses were conducted; however study authors did not address transferability of findings, and the high level of non‐response does impede data transferability
Overall risk of bias of process evaluation	High risk	Data collected were not presented clearly; high level of non‐participation impinges on ability to generalise, even to the population in question