Comparing databases: determinants of sexually transmitted infections, HIV diagnoses, and lack of HIV testing among men who have sex with men.

backgroundEarly detection and treatment of STI/HIV are public health priorities. Our objective was to compare characteristics of men who have sex with men (MSM) in Dutch data available in 2010 from EMIS, an international internet survey, Schorer Monitor, a Dutch internet survey, and data from STI- clinic visits, since these might be subject to different and unknown biases.

methodsData from Dutch MSM Internet Surveys (EMISNL N = 3,787; Schorer Monitor, SMON N = 3,602), and 3,800 STI clinic visits (SOAP) were combined into one dataset. We included factors that were measured in all three databases. The socio-demographics included were age (at the time of the survey), zip code, and ethnicity. Behavioural variables included were the number of sexual partners, condom use with last sexual partner, drug use, being diagnosed with STI, being diagnosed with HIV, and HIV testing. Outcomes we investigated were being diagnosed with STI, HIV, and never been tested for HIV.

resultsLogistic regressions showed that determinants for being diagnosed with STI were having more sexual partners, drug use, and having had an HIV test (aORs 1.3 to 17.1) in EMIS and SMON. Determinants for being diagnosed with HIV in all three databases were older age, living in Amsterdam, and having more partners (aORs 1.8 to 4.4). In EMIS and SMON, drug use, non-condom use, and having STI were additional determinants (aORs 1.6 to 8.9). Finally, determinants associated with never been tested for HIV were being younger (only SOAP), living outside of Amsterdam, having fewer partners, no drug use, and no STI (aORs 0.2 to 0.8).

conclusionsRisk factors from internet surveys were largely similar, but differed from STI clinics, possibly because it involves self-reports rather than diagnoses or because of differences in timing. The difference between the internet surveys and STI clinic data is much less pronounced for having never been tested, suggesting both are appropriate for this outcome. These findings shed light on conclusions drawn from different data sources, as well as the comparability of recruitment strategies, the robustness of risk factors, consequences of phrasing questions differently, and on (policy) implications based on different data sources.