What factors influence HIV testing? Modeling preference heterogeneity using latent classes and class-independent...

Area of Science:

• Public Health
• Epidemiology
• Behavioral Science

Background:

• Increased HIV testing rates are essential for HIV epidemic control, particularly among high-risk populations.
• Tailoring HIV testing interventions to population preferences can improve uptake and effectiveness.

Purpose of the Study:

• To elicit and analyze the preferences of high-risk populations for different HIV testing options.
• To inform the design of targeted HIV testing strategies by understanding key attribute preferences.

Main Methods:

• A discrete choice experiment (DCE) was conducted with 740 female barworkers and male Kilimanjaro mountain porters in northern Tanzania.
• Participants completed 12 choice tasks evaluating hypothetical HIV testing options based on attributes like privacy, invasiveness, and accessibility.
• Data were analyzed using a random effects latent class logit model to identify distinct preference profiles.

Main Results:

• Eight distinct preference classes were identified, indicating significant heterogeneity in testing preferences among participants.
• Substantial variation in preferences was observed both between and within these classes.
• Attribute importance varied across classes, with privacy, accessibility, and perceived accuracy significantly influencing choices.

Conclusions:

• HIV testing preferences are diverse within high-risk populations, necessitating tailored intervention strategies.
• Discrete choice experiments provide a robust method for systematically designing heterogeneity-focused HIV testing interventions.
• Matching testing options to identified preference profiles can enhance engagement and improve HIV testing rates.