Risica 2013.

Methods	Study design: RCT Year of study: not reported Intervention period: participants had access to 12 weekly cable TV shows and 4 months follow‐up intervention Follow‐up period (postintervention): assessment at 3 months (end of TV shows), 8 months (end of follow‐up intervention) and 12 months (4 months postintervention)
Participants	Country: USA (Massachusetts) Language: English Target population: African American women Inclusion criteria: self‐identified as African American or black women, aged 18–70 years, resided in catchment area of the cable TV company, planned to stay in area for ≥ 1 year, access to telephone, television and videocassette recorder (VCR), available to watch the TV programme at its airtime, BMI ≥ 22 kg/m2 Exclusion criteria: pregnant or < 4 months postpartum, physical problems that would prevent mild physical activity, previous history of treatment for eating disorders, unable to speak and read English, participation in other weight‐control research project Participants: control, N = 82; intervention 1, N = 71, Intervention 2, N = 210 (3 intervention arms combined) Sex: women Socioeconomic characteristics: all self‐described African American, black or West Indian/Caribbean ethnicity. 12% foreign‐born. 12% had not completed high school, 40% with college education. 44% in lowest income category. The study paid for cable TV service for participants who could not afford it (19% of participants) Other: 71% obese. Self‐reported hypertension > 25%, diabetes 12%, and on medication for hypertension or heart condition > 25%.
Interventions	Study objective/aim: to evaluate the effectiveness of a culturally tailored weight control cable TV programme for black women Control: wait‐list/attention placebo comparison. Biweekly mailings for 12 weeks with other health‐related information. Received all the TV shows as videos and other intervention material after the 12‐month follow‐up (i.e. after final outcome assessment) Intervention 1: participants were given access to 12 one‐hour live programmes on cable TV and printed material corresponding to the shows biweekly by post. After the shows, participants received 4 monthly mailings with written material and booster videotapes including an exercise video. Intervention 2 (combination of 3 study arms): in addition to all intervention content described under intervention 1, participants received either access to a toll‐free number to call during a live sharing part of show; 12 weekly and 4 monthly telephone support calls from a community outreach educator; or both access to call the 'live' sharing and the 16 telephone support calls Content of the mass media message: educational content regarding nutrition and physical activity to improve health and weight control as defined by each woman. Practical cooking demonstration or physical activity breaks. 20 min of each show devoted to a live 'sharing' between social worker, featured guest, and live callers. Media channel(s) of dissemination: cable TV, print material, video Theoretical basis: intervention based on Social Action Theory. Behavioural change, stress reduction and self‐management principles integrated into nutrition and physical activity sections Targeting approach: designed to be culturally appropriate based on formative research with 500 people from the target population. All African American female cast in the TV shows, including all experts; recruited only target population
Outcomes	Indicators of behavioural change: BMI (kg/m2), weight (kg) Self‐reported behavioural change: food habits questionnaire (fat behaviours), leisure activity score Adverse outcomes: none reported
Notes	Intervention fidelity: random participants were called during shows to monitor who was watching. Calls to live sharing part and support calls logged. At 3 months follow‐up, 69% of participants had watched 7‐12 shows and 57% had read most or all written material. Participants in intervention 2 were more reluctant to call live than expected. < 50% of the planned support calls were completed. Sources of funding: the National Cancer Institute (USA)
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation sequence adequately generated	Low risk	After baseline assessment, participants drew an envelope from a container with concealment of allocation.
Allocation adequately concealed	Low risk	After baseline assessment, participants drew an envelope from a container with concealment of allocation.
Baseline outcome measurements similar	Low risk	Baseline outcome measurements presented and analysed as balanced
Baseline characteristics similar	Low risk	Baseline characteristics presented and analysed as balanced
Incomplete outcome data adequately addressed	Low risk	ITT analyses conducted
Knowledge of allocated interventions prevented	High risk	Unclear risk for the BMI measurements. High risk for self‐reported behaviours in non‐blinded study
Adequately protected against contamination	Low risk	Participants reported to be geographically spread
Free of selective outcome reporting	Low risk	Results compared with published protocol
Other bias	Low risk	None considered