In theory, progress in reducing the number of people newly infected with HIV is best assessed by monitoring the changes in incidence over time. In practice, however, prevalence data rather than incidence data are available. In analysing the prevalence data from key populations for assessing the impact of prevention programmes, it is desirable not to restrict analysis to young people but to report on the people newly initiating behaviour that puts them at higher risk of infection, such as by restricting the analysis to people participating in sex work for less than one year, to men who first had sex with another man within the past year or to people initiating injecting drug use within the past year. This type of analysis also has the advantage of not being affected by antiretroviral therapy increasing survival and thereby increasing prevalence.
If prevalence estimates are available, disaggregated by greater than and less than one year in sex work, one year of sexual activity with other men or one year of injecting drugs, countries are strongly encouraged to report this disaggregation in their country progress report and to use the comments field in the reporting tool for this indicator to present disaggregated estimates.
Because of the difficulties in accessing key populations, biases in serosurveillance data are likely to be more significant than in data collected from a less stigmatized population, such as women attending antenatal clinics. If there are concerns about the data, the interpretation should reflect these concerns.
Understanding how the sampled populations relate to any larger populations sharing similar high-risk behaviour is critical to interpreting this indicator. The period during which people belong to a key population is more closely associated with the risk of acquiring HIV than their age. It is therefore desirable not to restrict analysis to young people but to report on other age groups as well.
Trends in HIV prevalence among key populations in the capital city provide a useful indication of the performance of HIV prevention programmes in that city. However, they are not representative of the situation in the country as a whole.
The addition of new sentinel sites will increase the sample’s representativeness and therefore provide a more robust point estimate of HIV prevalence. However, adding new sentinel sites reduces the comparability of values over time. As such, using consistent sites when undertaking trend analysis is important.
In previous reporting rounds, several countries have reported the HIV prevalence among subpopulations of transgender women through the additional comments field in the Global AIDS Response Progress Reporting online reporting tool. This demonstrates that the data are feasible to obtain in different settings.
Surveys exclusively covering transgender people are rare. Most data for transgender communities are drawn from surveys of men who have sex with men or sex workers. The risk environment reported in most transgender communities is great, placing transgender women at especially high risk of becoming HIV- positive and transmitting the infection. Examples from several Latin American countries demonstrate that successful surveys can be conducted in transgender communities. If transgender women are respondents in surveys of sex workers, include the data with sex workers as a disaggregation. If transgender people are respondents in surveys of men who have sex with men, include the data under the transgender tab.
Prisoners are easily reached with services, while released individuals can be efficiently linked to appropriate care and prevention services. The HIV prevalence can be readily estimated and quickly provide information that can be acted on.
In settings where high-risk behaviours for HIV transmission are criminalized, there is potential for high HIV prevalence and over-interpreting the results. Full understanding of the prison population is helpful during the analysis, especially the reasons for detention.