International bodies including the United Nations, the World Health Organisation and the European Monitoring Centre for Drugs and Drug Addiction have highlighted the increasing problem of opiate use and have recommended the truncated Poisson methods and capture-recapture methods, with more than two data sources, for prevalence estimation.

This recommendation is difficult to implement however in regions where data sources are limited. The truncated Poisson method, which requires a single data source, is less publicised but is also endorsed by the EMCDDA. Within this paper, we implement both of these methods in Western Australia for the first time.

We provide the first multi-source enumeration of opiate prevalence for this region, and in addition develop the two methods for use with in-patient hospital data and police records. We found from the multi-source enumeration that a total of 2826 unique individuals were identified as opiate users in the 2-year study period. In addition, using unstratified capture-recapture models we estimated that the population of hidden opiate users was 17,233 (95% CI of 8960-35,055) during the same period. This gives us a total 2-year prevalence estimate of 20,059 opiate users in Western Australia in the period from the 1st July 1996 to 30th June 1998. This corresponds to an estimated prevalence rate of approximately 18 per 1000 of the population aged between 14 and 55.

The need for these methods in the coming decade will increase as we see expansion of the European Union and the increased reporting of drug use in developing and emerging nations.