Darrow 2011.

Methods	Study design: controlled before‐after cross‐sectional study   Sampling frame: Florida adults Collection method: State of Florida HIV incidence data Description of the community coalition: Florida International University and 3 community‐based organizations that served African American, Hispanic, Haitian, and Afro‐Caribbean communities formed a coalition in response to REACH 2010 funding with the aim of reducing disparities in HIV among minority young adults through community awareness and education. The University served as the lead agency and contracted with the CBOs annually to provide educational outreach
Participants	Communities: Broward County, Florida Country: USA Ages included in assessment: adults   Reasons provided for selection of intervention community: minority population with high HIV/AIDS rates  Intervention community (population size): South Florida counties of Broward, Palm Beach, and Miami‐Dade (population size not provided) Comparison community (population size): North Florida counties of Duval, Hillsborough, and Orange (population size not provided)
Interventions	Name of intervention: Broward County REACH HIV Prevention Theory: none reported Aim: to improve levels of awareness of the HIV/AIDS problems among minority young adults, to increase perceptions of the seriousness of the local AIDS problem, and to stimulate action to reduce HIV/AIDS in minority communities  Description of costs and resources: 6 community health outreach workers, graduate students, and other university staff delivered interventions. US$103,500 provided annually to 3 CBOs; funding amount to University not reported   Components of the intervention: educational materials in the form of posters, pamphlets, bus advertisements, newspaper articles, and radio and television public service announcements, as well as community health fairs and poetry competitions Start date: 2000  Duration: 4 years
Outcomes	Outcomes and measures: annual HIV incidence rates per 100,000  Time points: 1998 to 2006
Notes	Outcomes measured at population level using state HIV incidence data Funding source: government
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Non‐randomized
Allocation concealment (selection bias)	High risk	Non‐randomized
Baseline outcome measurement similar	Low risk	Statewide HIV surveillance data
Baseline characteristics similar	High risk	Baseline differences in HIV incidence between intervention and control
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Statewide HIV surveillance data
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Statewide surveillance system data
Blinding of participants and personnel (performance bias) All outcomes	High risk	Non‐randomized
Protection against contamination	Low risk	Counties not adjacent
Selective reporting (reporting bias)	Low risk	Statewide HIV surveillance data
Other bias	High risk	Included counties in the intervention group that received community outreach