Effectiveness and cost-effectiveness of an interactive web-based sexual health intervention in Chinese young women

Abstract

Sexual health is a major concern in young adults worldwide, and its promotion could help prevent sexually transmitted infections (STIs). Young women, in particular, are susceptible to contracting STIs and engaging in risky sexual behaviors. However, interventions seldom addressed the effects of sexual coercion on sexual health in the Chinese population, which are highly related to safer sex practice. In this thesis, I aim to report the effectiveness and cost-effectiveness of an interactive web-based sexual health intervention (Smart Girlfriend) combined the component to address sexual coercion to promote safer sex practice in Chinese young women, and to examine potential mediators and moderators to understand how sexual coercion affects condom use. A multicenter randomized controlled trial was conducted at 5 universities in Hong Kong. Female university students aged 18 years or older, who had intimate partners, and had not received any sexual health information in the past 12 months were included. Eligible participants were randomly assigned to the Smart Girlfriend or the control (one-page information about condom use). The consistency of condom use (Primary Outcome) was defined as the percentage of male condom protected sex with every partner during the past 3 months. Knowledge, attitude, norms, and self-efficacy concerning condom use (Secondary Outcome) which were assessed by the Multidimensional Condom Attitudes Scales (MCAS). Intention to treat was applied in the analyses. The Zero/One Inflated Beta (ZOIB) regression model was used to analyze the primary outcome and a t-test was used to analyze the secondary outcome. Further analysis included moderated mediation analyses to test if sexual orientation moderates the indirect effect of condom negotiation on the relationship between sexual coercion and condom use, and the cost-effectiveness analysis to estimate the cost-effectiveness of the Smart Girlfriend with the control by using the decision analytic model over a 4-year time horizon. The main model outcome was the incremental cost per quality-adjusted life-year (QALY). Of the 1503 screened students, 781 were randomized (Intervention: 384, Control:397). The average age of participants was 21.5 (SD 2.6) years. ZOIB results regarding consistency of condom use indicated that there was no significant intervention effect. MCAS scores at 3-month follow-up were significantly higher in the Smart Girlfriend group. The moderated mediation analysis indicated that sexual orientation moderated the mediation, as the indirect effect of condom negotiation was significant in heterosexual but not in sexual minority women. The cost-effectiveness analysis showed that the introduction of the Smart Girlfriend would result in cost savings over a 4-year time horizon, compared with the control. A 99% probability of being cost-effective was detected for a willingness to pay of 133,915 HKD per QALY. This study found the Smart Girlfriend did not significantly improve condom use compared with one-page information about condom use during a 6-month follow-up. The potential reasons are the low engagement, the ceiling effect, the difference in different sexual orientations and the less personalized and passively delivered intervention content. Given the low attrition rate and estimated cost-effectiveness, the Smart Girlfriend is considered as a feasible option to prevent STIs in young women.