Improving the effectiveness of HIV prevention : a randomised controlled trial of a small group behavioural intervention for 'high-risk' gay men attending genitourinary medicine clinics
Gay men are one group that is seriously affected by HIV in the UK. Prevention interventions have been widely implemented, but few have been rigorously evaluated for their effectiveness. This thesis describes the first randomised controlled trial (RCT) of a small group behavioural intervention for 'high-risk' gay men attending genitourinary medicine (GUM) clinics to use incident sexually transmitted infections (STI) as the primary endpoint and examines its impact on local prevention policy and practice. At total of 343 gay men GUM clinic attenders were randomised to either standard management (brief session with a counsellor) or standard management plus a brief small group behavioural intervention delivered by trained co-facilitators. Follow up was by questionnaire at 6 and 12 months for self-reported behavioural outcomes and record review of the clinic databases for new STI diagnoses and treatment. The participation rate was 72.3%. Of those randomised to the intervention, 70.9% attended. Questionnaire follow up was 80.5% at 6 months and 71.1% at 12 months. In total, 89.8% of men returned one or both follow up questionnaires or had an STI screen (with results available) during the follow up. At baseline, 36.6% of intervention participants vs. 30.1% of controls reported unprotected anal intercourse (UAI) in the last month. At 12 months, these proportions were 27.2% and 31.5% respectively (adjusted /7-value = 0.32). However, 30.8% of intervention participants vs. 20.8% of controls had at least one STI diagnosed at the clinic during follow up (Odds Ratio = 1.692; 95% Confidence Interval: 1.033-2.772). Although intervention participants were slightly more likely to have an STI screen during the follow up period (52.9% vs. 48.2%), they were also more likely to screen positive for a new STI (Adjusted odds ratio = 1.841; 95% Confidence Interval 0.996-3.40). Process evaluation and secondary analysis identified 3 possible explanations for these unanticipated results. These were: 1) inappropriate choice and use of behavioural change theories in determining the intervention content; 2) a mis-match between intervention facilitators? and participants? expectations and perceptions of the intervention; and 3) an unintended intervention effect of the trial?s retention and follow up procedures. The study concluded that the intervention did not reduce risk of acquiring an STI and demonstrated that even carefully conceived interventions should be rigorously evaluated. The trial results did not result in immediate changes in clinic practice or local HIV prevention policy. The thesis concludes that provision of rigorous research evidence alone cannot bring about changes in either HIV prevention policy or prevention practice.