Drug arrest statistics: a proposal for their use.

The extent to which drug use has increased, decreased, or leveled in the recent past is still subject to debate. This inconclusiveness is at least partly the result of the researchers' inability to monitor drug use in a constant and ongoing fashion. Instead, cross-sectional studies of different populations taken at different points in time are most often used to examine "trends." This paper suggests that arrest data, easily obtainable and immediately analyzable, may be used with caution to observe change in drug use.