Using respondent driven sampling in high risk areas to recruit heterosexuals at high risk of HIV: Demographic results from the Houston site of the National HIV Behavioral Surveillance Project

Objectives: To develop a sampling frame to recruit heterosexuals at high risk of HIV, census tracts with high rates of poverty and HIV infection were identified. We report demographic and economic factors among individuals recruited in high risk areas of Houston.

Methods: This cross-sectional surveillance project used respondent driven sampling to recruit heterosexuals living in 133 high-risk census tracts. Each participant could recruit up to three peers; there are dual incentives, $20 for the interview and $10 for each recruited peer. The survey included questions about sexual behaviors, drug use, and previous HIV testing. Voluntary HIV testing was also offered.

Results: We interviewed 560 women and 333 men; 98% were Black, 63% were female, and 67% were less than 35 years of age. 57% had never married and 36% had less than a 12th grade education. In the last year, men and women had a mean of 6.1 and 5.2 sexual partners. 50% used condoms at their last sex with a casual partner. Half reported that they knew the HIV status of their last sexual partner; and 16% had never tested for HIV. We found 1.8% to be infected with HIV; 80% of those infected had been unaware of their infection.

Conclusions: We established that our heterosexual population was at high risk of HIV and had a seroprevalence of HIV of 1.8%. Therefore, our sampling approach using RDS in high risk areas was successful in identifying high risk individuals.