Estimating the number of drug injectors from needle exchange data

Background. Information on the size of a drug injecting population is required for resource allocation and health service evaluation. The truncated Poisson (tP) estimator is a method of prevalence estimation which, in contrast to the capture-recapture method, can be used when only a single data source is available. Objective. We present a study on the size of a drug injecting population to illustrate how the tP estimator can be calculated and how some of its underlying assumptions can be checked. Results. Although 647 individuals attended the needle exchange in 1997, the total number of drug injectors was estimated to be 1041 within a 95% confidence interval of 960 similar to 1137. Discussion. The truncated Poisson estimator can be an easy and quick method of providing a prevalence estimate of drug injecting and may produce valid estimates of relative changes in prevalence rates over time.