The model presented here extends previous work which attempt to derive from the number of observed institutional contacts with drug abusers the total level of unobserved drug use activity. The mathematical model relating drug use and institutional contact presented here could be used to make local estimates of the number of hardcore drug users. These local estimates would be used as the anchor points in developing synthetic estimates for higher levels of geographic aggregation, eventually arriving at a national figure. The first section of the article deals with the problem of modeling drug use and institutional contact; the second section outlines a plan for sampling and data collection. 1 table, 1 figure, 5 notes, and 63 references