Design and Analysis in Epidemiology and Prevention Research for Substance Abuse and Addiction

• Activity Number: 2
• Type: Invited
• Date/Time: Sunday, August 11, 2002 : 2:00 PM to 3:50 PM
• Sponsor: National Institue on Drug Abuse-NIH
• Author(s): Beth Reboussin*+ and James Anthony
• Affiliation(s): Wake Forest University School of Medicine and Johns Hopkins Bloomberg School of Public Health
• Keywords: diagnostic criteria ; drug dependence ; latent class regression

Abstract

Understanding the mechanisms underlying the transition from drug use to drug dependence is a topic of growing interest among epidemiology and prevention scientists. The development of effective prevention and intervention programs relies on a better characterization of the conditions and processes that underlie the stages leading toward drug dependence. Current diagnostic criteria require the co-occurrence of several drug use behaviors and problems. This presentation will focus on the use of latent class models to study associations among a set of drug use behaviors and problems so that underlying subpopulations (classes) with particular response profiles might be defined. Our approach will allow for the exploration of latency periods of risk rather than restricting analyses to threshold-based diagnostic criteria. Methods for determining whether classes correspond to different severity levels or represent qualitatively different stages of dependence and drug involvement will be examined. Factors associated with membership in different stages of drug involvement and dependence will be explored through the use of latent class regression models.