Anticipating Community Drug Problems

The research developed a three-stage public health model of drug diffusion and the community drug indicators as aggregate measures of individual drug abuse use careers. The analysis used monthly data on drug indicators from the two jurisdictions (April 1984 through September 1990 for Washington and January 1988 through September 1990 for Portland) to estimate the correlations among drug problem indicators over time and to examine these correlations at different stages in the spread of a new form of drug abuse. The analysis also estimated lagged models in which the results of arrestee urinalysis were used to predict subsequent community drug problems. Results revealed that indicators of cocaine and PCP use peaked in the same years and began to decline in the same years. Arrestee urinalysis was the first indicator to signal a significant period of increasing problems, with both PCP and cocaine. Beyond the initial phase, consistent short-term relationships were not found. Based on conservative time-series models, drugs that exhibited little long-term trends across the study period appeared uncorrelated with other community drug problems. Figures, tables, appended codebook information, and 39 references (Author abstract modified)