Epidemiological Impact and Cost-effectiveness of Long-acting Antiretrovirals for Pre-exposure Prophylaxis Among Key Populations in Thailand

Discrete choice experiments (DCE) will be part of a survey, assessing sociodemographic characteristics, risk behavior, PrEP related information. Two separate surveys and DCEs will be designed, one for MSM, and one for TGW. Information collected through DCEs will allow us to estimate user values in the absence of observations or readily available data, and use these preferences to inform mathematical modeling on the impact and cost-effectiveness of long-acting PrEP. DCEs will be designed as per best-practice guidance by the International Society for Pharmacoeconomics and Outcomes Research (ISPOR). To maximize statistical and response efficiency, the number of attributes selected for evaluation must be restricted, i.e. the most important attributes must be known prior to construction. Therefore, the DCE design and implementation of the survey will be conducted in different phases:

PHASE I - Identification of attributes In-depth interviews (IDIs) with up to 40 potential, current, previous PrEP users (20 MSM,20 TGW) will assess which attributes are most relevant among each population. Posters will be used to recruit participants for IDIs. Self-reported HIV-negative MSM, TGW who either report interest in PrEP use, current PrEP use, or past PrEP use, will be eligible to participate in the IDIs. IDIs will be conducted by the study staff by in-person or on-line interviews. Focus group discussion (FGD) will prepare to discuss the interested information from IDIs. Following IDIs, participants will be informed and invited to the focus group discussion. A FGD will recruit 6-10 IDI participants from each MSM, TGW group. A list of 10 random numbers will be generated using statistical software if more than 10 individuals show interest in participating in the FGD. If there are less than 6 accepted participants for the FGD, the backup list or a new participant will be invited. The study staff will perform FGD by in-person or on-line conference. Data obtained during this process will be used to create a final list of attributes for each population and their levels to include in the choice tasks, ensuring that the choices are plausible and exchangeable. The number of DCE choice sets will be about 6 per person.

PHASE II - Pilot testing of DCE (The result from Phase I will be used for finalize an initial draft in this part) Initial drafts of the choice tasks (one draft for MSM and one draft for TGW) will be administered to convenience samples of 30 MSM, 30 TGW (10% of phase III sample size), respectively, to obtain feedback about difficulty of tasks, plausibility of levels, length, wording and clarity using a survey form. Final DCEs for each population will be agreed upon by consensus, and incorporated in the surveys.

PHASE III - Survey implementation (The result from Phase II will be used for finalize a survey in this part).

Two final surveys will be launched:1 survey for MSM, 1 survey for TGW. Both surveys will collect data on: