RESUMEN
HIV prevalence estimates from bathhouse testing programmes differ widely, ranging from 3% to 11%. The observed difference may be a consequence of whether the programme was part of a research project or a community-based programme. A research-funded testing programme was offered at a bathhouse between May 2001 and December 2002. A local community-based organization began a testing programme in July 2006 at the same bathhouse; the data for this analysis cover the period July 2006 through December 2007. County-wide HIV prevalence estimates were available across the two time periods (i.e. 2001-2002 and 2006-2007). The research-funded testing programme recruited fewer men but identified more positive individuals (10.7% of those testing in the research programme) than were identified among men who tested in the area clinics (3.8% of those men who have sex with men [MSM] testing throughout the county in the same time period). However, the community-based testing programme identified about the same proportion of positive MSM (2.6%) as county clinics (2.7%) in the same time period. In conclusion, results confirmed that even in the same venue, a community-based HIV testing programme identified a similar proportion of positive MSM as the area clinics; however, the research-funded programme identified appreciably more. Incentives may contribute to the difference.