The Men's Safer Sex project: intervention development and feasibility randomised controlled trial of an interactive digital intervention to increase condom use in men

Background: This report details the development of the Men's Safer Sex website and the results of a feasibility randomised controlled trial (RCT), health economic assessment and qualitative evaluation. Objectives: (1) Develop the Men's Safer Sex website to address barriers to condom use; (2) determine the best design for an online RCT; (3) inform the methods for collecting and analysing health economic data; (4) assess the Sexual Quality of Life (SQoL) questionnaire and European Quality of Life-5 Dimensions, threelevel version (EQ-5D-3L) to calculate quality-adjusted life-years (QALYs); and (5) explore clinic staff and men's views of online research methodology. Methods: (1) Website development: we combined evidence from research literature and the views of experts (n = 18) and male clinic users (n = 43); (2) feasibility RCT: 159 heterosexually active men were recruited from three sexual health clinics and were randomised by computer to the Men's Safer Sex website plus usual care (n = 84) or usual clinic care only (n = 75). Men were invited to complete online questionnaires at 3, 6, 9 and 12 months, and sexually transmitted infection (STI) diagnoses were recorded from clinic notes at 12 months; (3) health economic evaluation: we investigated the impact of using different questionnaires to calculate utilities and QALYs (the EQ-5D-3L and SQoL questionnaire), and compared different methods to collect resource use; and (4) qualitative evaluation: thematic analysis of interviews with 11 male trial participants and nine clinic staff, as well as free-text comments from online outcome questionnaires. Results: (1) Software errors and clinic Wi-Fi access presented significant challenges. Response rates for online questionnaires were poor but improved with larger vouchers (from 36% with (sic) 10 to 50% with (sic)30). Clinical records were located for 94% of participants for STI diagnoses. There were no group differences in condomless sex with female partners [incidence rate ratio (IRR) 1.01, 95% confidence interval (CI) 0.52 to 1.96]. New STI diagnoses were recorded for 8.8% (7/80) of the intervention group and 13.0% (9/69) of the control group (IRR 0.75, 95% CI 0.29 to 1.89). (2) Health-care resource data were more complete using patient files than questionnaires. The probability that the intervention is cost-effective is sensitive to the source of data used and whether or not data on intended pregnancies are included. (3) The pilot RCT fitted well around clinical activities but 37% of the intervention group did not see the Men's Safer Sex website and technical problems were frustrating. Men's views of the Men's Safer Sex website and research procedures were largely positive. Conclusions: It would be feasible to conduct a large-scale RCT using clinic STI diagnoses as a primary outcome; however, technical errors and a poor response rate limited the collection of online self-reported outcomes. The next steps are (1) to optimise software for online trials, (2) to find the best ways to integrate digital health promotion with clinical services, (3) to develop more precise methods for collecting resource use data and (4) to work out how to overcome barriers to digital intervention testing and implementation in the NHS.