A recent decline in cocaine use among youthful arrestees in Manhattan, 1987 through 1993.

Abstract

Cocaine use among youths as measured by several annual surveys was most popular from about 1979 to 1986, after which it declined. This study carefully examines the nature of the decline by focusing on microdata for youthful arrestees in Manhattan. Multiple statistical analyses examine whether the decline in cocaine use detected by urinalysis is attributable to fewer arrested youths born more recently having become regular users (cohort effect), to regular users decreasing their consumption (period effect), or to changes in arrest or sampling priorities (artifact). All analyses suggest that the dramatic decline in detected cocaine use among arrestees--from 69% in 1987 to 17% in 1993--was a cohort effect. Detected cocaine use, which was highest (78%) among arrestees reaching 18 in 1986 at the height of the crack epidemic in New York City, subsequently declined to a low of 10% among arrestees reaching 18 in 1993. These findings suggests that the epidemic in use of cocaine and crack entered a decline in the late 1980s. However, widespread use of these drugs will probably continue to prevail as an aging population with established habits persists in its use.