OBJECTIVES: To review the scientific data on the role of sexually transmitted diseases (STDs) in sexual transmission of HIV infection and discuss the implications of these findings for HIV and STD prevention policy and practice. METHODS: Articles were selected from a review of Medline, accessed with the OVID search engine. The search covered articles from January 1987 to September 1998 and yielded 2101 articles. Methods used to uncover articles which might have been missed included searching for related articles by author, and combing literature reviews. In addition, all abstracts under the category "sexually transmitted diseases" from the XI and XII International Conferences on AIDS (Vancouver 1996 and Geneva 1998) and other relevant scientific meetings were reviewed. Efforts were made to locate journal articles which resulted from the research reported in the identified abstracts. All original journal articles and abstracts which met one of the following criteria were included: (1) studies of the biological plausibility or mechanism of facilitation of HIV infectiousness or susceptibility by STDs, (2) prospective cohort studies (longitudinal or nested case-control) which estimate the risk of HIV infection associated with specific STDs or STD syndromes, or (3) intervention studies which quantitate the effect which STD treatment can have on HIV incidence. RESULTS: Strong evidence indicates that both ulcerative and non-ulcerative STDs promote HIV transmission by augmenting HIV infectiousness and HIV susceptibility via a variety of biological mechanisms. These effects are reflected in the risk estimates found in numerous prospective studies from four continents which range from 2.0 to 23.5, with most clustering between 2 and 5. The relative importance of ulcerative and non-ulcerative STDs appears to be complex. Owing to the greater frequency of non-ulcerative STDs in many populations, these infections may be responsible for more HIV transmission than genital ulcers. However, the limited reciprocal impact of HIV infection on non-ulcerative STDs and the evidence that non-ulcerative STDs may increase risk primarily for the receptive partner (rather than bidirectionally) may modulate the impact of these diseases. The results of two community level randomised, controlled intervention trials conducted in Africa suggest that timely provision of STD services can substantially reduce HIV incidence, but raise additional questions about the optimal way to target and implement these services to achieve the greatest effect on HIV transmission. CONCLUSIONS: Available data leave little doubt that other STDs facilitate HIV transmission through direct, biological mechanisms and that early STD treatment should be part of a high quality, comprehensive HIV prevention strategy. Policy makers, HIV prevention programme managers, and providers should focus initial implementation efforts on three key areas: (i) improving access to and quality of STD clinical services; (ii) promoting early and effective STD related healthcare behaviours; and (iii) establishing surveillance systems to monitor STD and HIV trends and their interrelations.