SYSTEM DYNAMICS MODEL FOR COCAINE PREVALENCE ESTIMATION AND TREND PROJECTION

• NCJ Number: 142328
• Journal: Journal of Drug Issues Volume: 23 Issue: 2 Dated: (Spring 1993) Pages: 251- 280
• Author(s): J B Homer
• Date Published: 1993
• Length: 30 pages
• Grant Number(s): 87-IJ-CX-0042; 90-IJ-CX-0014
• Sponsoring Agency: National Institute of Justice (NIJ); US Dept of Justice NIJ Pub
• Corporate Author: University of California, Los Angeles
• Publication Format: Article
• Publication Type: Report (Technical)
• Language: English
• Country: United States of America

Annotation

This article presents a system dynamics simulation model that reproduces a variety of national indicator data that reflect cocaine use and supply over a 15-year period and provides detailed estimates of underlying prevalence.

Abstract

The system dynamics model of cocaine prevalence presented is a new application of a general methodology first developed in the mid-1950's to analyze dynamic phenomena through the use of simulation models (Forrester, 1961). A system dynamics model consists of an interconnected set of algebraic and difference equations that represent the continuous ebb and flow of people, materials, and information. After being assigned initial conditions consistent with historical data, this set of equations is used to generate output over time. If the model is valid, this output will mimic the true course of events, and the model may be used for the analysis of trends and alternative scenarios. System dynamics is of special interest as a modeling method because of its emphasis on endogenous feedback relationships between constituent variables, such as the relationships among population variables, supply and demand variables, and other indicator variables. In the reported application of the model to cocaine use, sensitivity tests were conducted to clarify observed trends such as growth in the compulsive use of crack cocaine and decline in the casual use of cocaine powder. Alternative scenarios with possible policy implications were simulated and projected through the year 2002; the results are assessed. The author concludes that the model presented apparently has value as a tool for improving understanding of national cocaine prevalence trends. It may also be useful as a starting point for additional studies of illicit drug use and drug market dynamics. 13 notes and 18 references