Application of the network scale‐up method to estimate the sizes of key populations for HIV in Singapore using online surveys

Abstract

Abstract Introduction Singapore lacks robust data on the sizes of the key populations that are most at risk for HIV. Using the network scale‐up method for hidden or hard‐to‐reach populations, we estimate the sizes of five key populations—male clients of female sex workers (MCFSW), men who have sex with men (MSM), female sex workers (FSW), people who inject drugs (PWID) and transgender people—and profile the ages and ethnicities of respondents with the high‐risk contacts they report knowing. Methods We conducted a cross‐sectional online survey between March and May 2019 (n = 2802) using a network scale‐up instrument previously developed for Singapore. Participants were recruited using an existing panel and online advertising, and the sample reweighted by age, sex, ethnicity and education attained to represent the general adult population. We built a Bayesian hierarchical model to estimate the sizes of the five key populations for HIV in Singapore. Results After adjustment, the sizes of the at‐risk populations are estimated to be: 76,800 (95% credible interval [CI]: 64,200–91,800) MCFSW; 139,000 (95% CI: 120,000–160,000) MSM; 8030 (95% CI: 3980–16,200) FSW; 3470 (95% CI: 1540–7830) PWID and 18,000 (95% CI: 14,000–23,200) transgender people. Generally, men reported knowing more people in all the high‐risk groups; older people reported knowing more MCFSW, FSW and transgender people; and younger people reported knowing more MSM. There was a bimodal effect of age on those who reported knowing more PWIDs: people in their 20s and 60s reported more contacts. Conclusions This study demonstrates that a size estimation study of hidden populations is quickly and efficiently scalable through using online surveys in a socially conservative society, like Singapore, where key populations are stigmatized or criminalized. The approach may be suitable in other countries where stigma is prevalent and where barriers to surveillance and data collection are numerous.

Keywords

• modelling
• key and vulnerable populations
• men who have sex with men
• sex workers
• transgender people
• stigma