Objective: To evaluate outcomes of a sexual health promotion intervention to prevent transmission of sexually transmitted diseases (STD) including HIV among genitourinary medicine clinic attenders. Design: A randomized controlled trial using pre-test and post-test measurements after 4 months, with clinical follow-up at 18 months. Setting: A genitourinary medicine clinic in Nottingham, UK. Patients: A total of 492 patients were randomly assigned either to an individually focused counselling and skills training intervention, including written materials (n = 148), to receive written materials only (n = 162), or to usual clinic procedure (n = 182). Intervention: Social learning theory provided the theoretical framework for the intervention, which was informed by previous research in this setting, and aimed to alter perception of risk for HIV infection, increase knowledge and attitudes to condoms, and increase condom use. Main outcome measures: The main outcomes of interest were self-reported condom use and behaviour change. Other outcomes of interest were knowledge and attitudes to condoms, and re-attendance at the clinic with a diagnosis of STD. Results: Intervention subjects were significantly more likely than controls to report carrying condoms when anticipating sexual intercourse with a new sexual partner (P less than or equal to 0.05), and were more likely to perceive themselves at risk of HIV infection (P less than or equal to 0.001). There were no significant effects of the intervention on levels of knowledge about correct condom use, attitudes to condoms, self-reported condom use or incidence of STD. Conclusions: The limited effectiveness of the intervention suggests that condom promotion should continue, but that additional investment in clinic-based health promotion is unlikely to result in consequent health gain. (C) 1998 Lippincott-Raven Publishers.