Global assessment of existing HIV and key population stigma indicators: A data mapping exercise to inform country-level stigma measurement

Background
Stigma is an established barrier to the provision and uptake of HIV prevention, diagnostic, and treatment services. Despite consensus on the importance of addressing stigma, there are currently no country-level summary measures to characterize stigma and track progress in reducing stigma around the globe. This data mapping exercise aimed to assess the potential for existing data to be used to summarize and track stigma, including discrimination, related to HIV status, or key population membership at the country level.

Methods and findings
This study assessed existing indicators of stigma related to living with HIV or belonging to 1 of 4 key populations including gay men and other men who have sex with men, sex workers, people who use drugs, and transgender persons. UNAIDS Strategic Information Department led an initial drafting of possible domains, subdomains, and indicators, and a 3-week e-consultation was held to provide feedback. From the e-consultation, 44 indicators were proposed for HIV stigma; 14 for sexual minority stigma (including sexual behavior or orientation) related to men who have sex with men; 12 for sex work stigma; 10 for drug use stigma; and 17 for gender identity stigma related to transgender persons. We conducted a global data mapping exercise to identify and describe the availability and quality of stigma data across countries with the following sources: UNAIDS National Commitments and Policies Instrument (NCPI) database; Multiple Indicator Cluster Surveys (MICS); Demographic and Health Surveys (DHS); People Living with HIV Stigma Index surveys; HIV Key Populations Data Repository; Integrated Biological and Behavioral Surveys (IBBS); and network databases. Data extraction was conducted between August and November 2020. Indicators were evaluated based on the following: if an existing data source could be identified; the number of countries for which data were available for the indicator at present and in the future; variation in the indicator across countries; and considerations of data quality or accuracy. This mapping exercise resulted in the identification of 24 HIV stigma indicators and 10 key population indicators as having potential to be used at present in the creation of valid summary measures of stigma at the country level. These indicators may allow assessment of legal, societal, and behavioral manifestations of stigma across population groups and settings. Study limitations include potential selection bias due to available data sources to the research team and other biases due to the exploratory nature of this data mapping process.

Conclusions
Based on the current state of data available, several indicators have the potential to characterize the level and nature of stigma affecting people living with HIV and key populations across countries and across time. This exercise revealed challenges for an empirical process reliant on existing data to determine how to weight and best combine indicators into indices. However, results for this study can be combined with participatory processes to inform summary measure development and set data collection priorities going forward.