[Skip to Content]
[Skip to Content Landing]
Letters to the Editor
March 2007

Errors in Assessing DSM-IV Substance Use Disorders

Arch Gen Psychiatry. 2007;64(3):379-380. doi:10.1001/archpsyc.64.3.379

In the June 2005 issue of the ARCHIVES, Kessler and colleagues1 reported results from the National Comorbidity Survey Replication (NCS-R), a 2001-2002 US psychiatric epidemiology survey of DSM-IV mental and substance use disorders.2-4 Included are prevalences and associated features, such as treatment of current (last 12 months) disorders. Given the fundamental nature of this study and the potential policy relevance, it is essential that rates estimated by the study be accurate. We are concerned about the accuracy of the rates reported for alcohol and drug abuse and dependence disorders. The specific problem is that the study purports to apply DSM-IV criteria, but the questionnaire used by Kessler and colleagues, the World Mental Health–Composite International Diagnostic Interview (WMH-CIDI),5 skips all respondents’ past questions on DSM-IV dependence if they do not respond positively to questions on DSM-IV abuse, effectively using abuse as a screen for dependence. This same procedure has been used in the recent National Institute of Mental Health–sponsored NCS-A adolescent study; the NCS-2, a 10-year follow-up of the original NCS conducted in 1990-19926; and the National Study of African American Life, and the National Study of Latino and Asian Americans.7 Beyond the United States, the NCS-R methods have been used in epidemiological surveys under way in 27 other countries under the aegis of the World Health Organization (WHO) World Mental Health Survey Initiative.8