Estimating Size of Injecting Drug User Population
Estimating the size of most at risk populations is important for assessing program prevention needs. PLACE data can be used to estimate the number of people visiting PLACE venues including the numbers of injecting drug users, commercial sex workers, clients of commercial sex workers, and men who have sex with men. For example, Tate and Hudgens (American Journal of Epidemiology, 2007) used information from people surveyed at PLACE venues in Osh, Kyrgyzstan in a mathematical model to estimate that the number of all persons socializing at PLACE venues over a four-week period was 56,171 (95% confidence interval 43,634, 68,708). Of these patrons, 27%, or approximately 15,000 reported that they had ever injected drugs and 17% (approximately 950) reported that they had injected in the past week.
Simpler methods to estimate the crude size of risk populations can be used if mathematical modeling is not an option. Crude estimates can be obtained from information obtained in Step 3 of the PLACE method about the number of people visiting the venue at peak times, and from interviews with site patrons in Step 4 about frequency of attendance and whether they inject drugs, engage in commercial sex or engage in anal sex with a man.
For example, site representatives at each of the 237 venues identified in Osh estimated the number of men and women who visit the venue during a busy time. These estimates ranged from less than ten people to over 500. The mean number reported was 53. Consequently, we estimate that 53x237 people (or approximately 12,500) visit the sites during a busy time. However, this underestimates the number visiting the site over a four-week period because not all people who attend the site in a four-week period were at the site at the time of the interview.
To adjust the estimate to take into account the fact that some people visit the site every day during a four-week period and some people visit rarely, data from a question asked of patrons about how frequently they visit the site to is used to calculate a “multiplier” figure. If everyone reported visiting the site every day, the multiplier figure would be one. If everyone reported visiting the site once in a four-week period the multiplier number would be 16, reflecting the fact that sites reported on average 16 busy times in a four week period. In Osh, the multiplier figure is approximately five. Thus the crude estimate of the number of patrons is 62,500 (5x12,500), which is slightly higher than the estimate obtained using a more sophisticated approach. The estimated number of injecting drug users is 27% of 62,500 or 16,875.
Although estimates of the size of populations can be readily derived from PLACE data, the validity of these estimated depends on the accuracy of data from site representatives and patrons, and the proportion of people in the risk group of interest (e.g. injecting drug users) who visit venues.
