Little is known about the effectiveness of HIV testing and counseling programs to inspire behavior change among patients seen at US clinics for Sexually Transmitted Diseases (STDs). We studied changes in gonorrhea incidence from 1989 through 1993 in a historical cohort of 4,031 patients seen at a public STD clinic in New Orleans, Louisiana. To determine if HIV testing and counseling promote risky behaviors, we designed a pretest-posttest observational study which included a treatment group and an untreated comparison group. The frequency, timing, and results of the HIV tests performed during follow-up were taken into account in the analysis. We used Cox's proportional hazard regression techniques for multivariate failure time data to model the occurrence of gonorrhea infections over time. The models adjusted the comparisons for history of gonorrhea, and HIV testing and counseling, as well as important confounding factors. We found that being tested and learning about a negative test result may have prevented up to 1,318 gonorrhea infections (34.7 percent of the expected total number of infections) among individuals who never tested HIV positive. The risk of gonorrhea remained constant at a low level among 911 individuals who were not tested during follow-up (22.6 percent of all patients). The risk of gonorrhea decreased markedly after HIV testing among 2,041 individuals at low baseline risk (50.6 percent of all patients), decreased moderately after each of the first two HIV tests among 672 subjects at intermediate baseline risk (16.7 percent of all patients), but increased after the first test among 358 subjects at high baseline risk (8.9 percent of all patients). Posttest counseling was associated with a marginally significant decline in risk of gonorrhea among subjects at low baseline risk, but had no apparent effect on the other patients. No disinhibition effect was observed after posttest counseling. Our study suggests that HIV testing and counseling promote safer behaviors in most patients diagnosed with gonorrhea in a public clinic for sexually transmitted diseases. Nevertheless, a possible rebound effect was observed after HIV testing among patients who had the highest baseline risk of gonorrhea
