Aims

To support the free and open dissemination of research findings and information on alcoholism and alcohol-related problems. To encourage open access to peer-reviewed articles free for all to view.

For full versions of posted research articles readers are encouraged to email requests for "electronic reprints" (text file, PDF files, FAX copies) to the corresponding or lead author, who is highlighted in the posting.

___________________________________________

Saturday, February 25, 2012

Scaling Properties of the Combined ICD-10 Dependence and Harms Criteria and Comparisons With DSM-5 Alcohol Use Disorder Criteria Among Patients in the



The Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV), and the International Statistical Classification of Diseases and Related Health Problems, 10th Edition (ICD-10), alcohol use disorders (AUDs) classifications offer competing and somewhat overlapping diagnostic tools for assessing alcohol dependence and harms/abuse. Both systems are in active stages of development in anticipation of their next respective iterations. Although much psychometric work has been done studying DSM-IV criteria, efforts toward the ICD-11 have been less prevalent.

Data from 3,191 drinkers in seven emergency department sites in four countries (United States, Mexico, Argentina, and Poland) were used to study the psychometric properties of the combined ICD-10 dependence and harms criteria. Comparisons with the proposed set of DSM-5 criteria and diagnostic thresholds are also included.

Item response theory analyses of the combined ICD-10 dependence and harms criteria suggested a single underlying factor, both overall and for each site separately, with only moderate differential item functioning across sites. Overall agreement between the summative combined ICD-10 dependence and harms criteria and the proposed 11-criteria DSM-5 scale was very high (r = .97), as was agreement between proposed diagnostic threshold levels of 0–1 (negative AUD), 2–3 (moderate AUD), and 4 or more (severe AUD) criteria endorsed (κ = .84). Although disagreement between the two three-level diagnoses was only 5.5% of the sample, a majority of these were because of differences between endorsement rates of abuse versus harms criteria.

Although there is support for efforts to align the two schemes, results are consistent with other studies finding the largest differences between the two systems emanating from differences between the abuse and harms domains. More research is needed before differences between the two systems can be reconciled.



Read Full Abstract

Request Reprint E-Mail: jbond@arg.org