Rhodes 2011.

Methods	Study design: cluster RCT Sampling frame: adult Latino males in rural central North Carolina Sampling method: random; participants recruited from community locations frequented by Latinos Description of the community coalition: community‐based participatory research partnership of more than 50 members located mainly in northwest and central North Carolina.  Members represent the lay community, including African American/black, Latino, and white gay men; organizational representatives; and academic researchers. Partnership has numerous studies ongoing
Participants	Communities: rural central North Carolina Country: USA Ages included in assessment: 18+ (n = 139) Reasons provided for selection of intervention community: Latinos in USA disproportionately affected by HIV/AIDS, North Carolina has one of the fastest growing Latino populations Intervention community (population size): rural central North Carolina (unknown) Comparison community (population size): rural central North Carolina (unknown)
Interventions	Name of intervention: HoMBReS‐2 (Hombres Manteniendo Bienestar y Relaciones Saludables‐2) Theory: Social Cognitive and Education Empowerment Aim: to test the efficacy of a small‐group HIV prevention intervention regarding condom use and HIV testing behavior Description of costs and resources: not provided Components of the intervention: 2 small group sessions involving trust‐building activities, didactic teaching, DVD modules, role plays, and group discussion, as well as skills building, practice, and feedback Start date: not provided Duration: 2 sessions
Outcomes	Outcomes and measures (follow‐up months) Consistent condom use (3‐month follow‐up) HIV testing (3‐month follow‐up)  Years of pre and post measurements: not provided
Notes	Participants entered the study and were randomly assigned in waves to avoid delays between recruitment and intervention delivery. Unclear why missing outcome data appear extensive
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Randomization occurred in waves; each participant selected an envelope from a discrete pool of assignments during each wave
Allocation concealment (selection bias)	High risk	No allocation concealment
Baseline outcome measurement similar	High risk	Consistent condom use much greater in intervention group (34.7%) than in control group (21.4%) at baseline
Baseline characteristics similar	Unclear risk	Intervention group significantly younger; all other presented characteristics appeared similar
Blinding of outcome assessment (detection bias) All outcomes	High risk	No blinding
Incomplete outcome data (attrition bias) All outcomes	High risk	Very low rate of loss to follow‐up reported, but final outcome measures show extensive missing observations
Blinding of participants and personnel (performance bias) All outcomes	High risk	No blinding
Protection against contamination	High risk	No protection against contamination was detailed; 6 participants attended the incorrect intervention arm
Selective reporting (reporting bias)	Low risk	Main outcomes reported