Population Size Estimations Among Hidden Populations Using Respondent-Driven Sampling Surveys: Case Studies From...

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Estimating the size of hidden populations (FSW, MSM, PWID) is vital for understanding health vulnerabilities and needs.
• Accurate population size estimates inform targeted interventions and resource allocation for public health initiatives.

Purpose of the Study:

• To advance the successive sampling-population size estimation (SS-PSE) method by incorporating visibility imputation.
• To evaluate the performance of the modified SS-PSE method using 15 datasets from respondent-driven sampling (RDS) surveys in Armenia.
• To compare SS-PSE estimates with other methods and provide guidelines for improved implementation.

Main Methods:

• The SS-PSE method approximates RDS sampling using a successive sampling model.
• Visibility imputation was incorporated to account for varying propensities to participate in studies.
• The study analyzed RDS data from female sex workers (FSW), men who have sex with men (MSM), and people who inject drugs (PWID) across three Armenian cities.

Main Results:

• SS-PSE showed good fits for FSW and PWID populations in multiple cities compared to prior estimates.
• Inconsistencies were observed for MSM populations, potentially due to the inclusion of transgender individuals in datasets.
• The study highlights the need to carefully define population eligibility criteria and consider subgroup variations.

Conclusions:

• Expert-based prior estimations may not always align with empirical data.
• RDS surveys require rigorous assessment for assumption adherence and convergence.
• SS-PSE should be used alongside other estimation methods and historical data for robust population size estimation.