Psychometric properties of DSM assessments of illicit drug abuse and dependence: results from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC)

Abstract

DSM-IV criteria for illicit drug abuse and dependence are largely based on criteria developed for alcohol use disorders and there is a lack of research evidence on the psychometric properties of these symptoms when applied to illicit drugs. This study utilizes data on abuse/dependence criteria for cannabis, cocaine, stimulants, sedatives, tranquilizers, opiates, hallucinogens and inhalants from the National Epidemiological Survey on Alcohol and Related Conditions (NESARC, n=43 093). Analyses included factor analysis to explore the dimensionality of illicit drug abuse and dependence criteria, calculation of item difficulty and discrimination within an item response framework and a descriptive analysis of 'diagnostic orphans': individuals meeting criteria for 1-2 dependence symptoms but not abuse. Rates of psychiatric disorders were compared across groups. Results favor a uni-dimensional construct for abuse/dependence on each of the eight drug classes. Factor loadings, item difficulty and discrimination were remarkably consistent across drug categories. For each drug category, between 29% and 51% of all individuals meeting criteria for at least one symptom did not receive a formal diagnosis of either abuse or dependence and were therefore classified as 'orphans'. Mean rates of disorder in these individuals suggested that illicit drug use disorders may be more adequately described along a spectrum of severity. While there were remarkable similarities across categories of illicit drugs, consideration of item difficulty suggested that some alterations to DSM regarding the relevant severity of specific abuse and dependence criteria may be warranted.