Reliable and valid psychiatric diagnoses are central to clinical practice and research, and these are defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM), a common language for clinicians and researchers. In DSM-IV, substance dependence was a reliable and valid diagnosis. However, problems were identified with DSM-IV abuse, and with the DSM-IV distinction between abuse and dependence. In DSM-5, just published in May, 2013, the abuse-dependence distinction was replaced by a single Substance Use Disorder (SUD). The new DSM-5 SUD is defined by 11 criteria: all DSM-IV dependence criteria, three abuse criteria, and craving, with a threshold of e2 criteria, and severity indicated by a criteria count. While many problems were solved by the DSM-5 SUD changes, they proved surprisingly controversial, with data lacking from current populations, especially treated ones, to answer some of the questions. During the initial funding for R01DA018652, we analyzed existing data (largely epidemiologic) on the relationship of abuse and dependence. This study was productive and successful, contributing some of evidence regarding the DSM-5 SUD changes. However, to address controversies and questions about DSM-5 SUD, more knowledge is needed about its reliability and validity, particularly among patients currently in treatment. We propose a renewal to investigate the strengths and potential weaknesses of the new DSM-5 SUD in clinical samples, and to identify ways to improve it in DSM-5.1, 5.2 etc. (anticipated at briefer intervals than the time between DSM- IV and DSM-5). We will address these issues by collecting new data from 600 substance-using patients in four New York City community treatment settings, all with successful prior research collaborations. The sample will be diverse in patient characteristics and in substances used. We will focus on alcohol, cannabis, cocaine, nicotine and opioids. Trained clinician interviewers will give diagnostic interviews covering DSM-IV and DSM-5 SUD criteria at baseline, retest (1-21 days later), 3 months and 6 months. Validation will utilize antecedent variables (e.g., family history); concurrent variables (e.g., substance use level, functioning, comorbidity); and prediction/prognosis (e.g., remission and relapse in symptoms and use, functioning). Technology-assisted data collection will include Computer-Assisted Personal Interviews (CAPI); Audio Computer-Assisted Self- administered measures for private assessment of validation variables (A-CASI); and, for prospective validation, brief daily self-reports of substance use, craving, and functioning via interactive voice response (IVR). We will also investigate thresholds; whether patient characteristics affect reliability or validity; and importantly, whethe a valid subset of the 11 DSM-5 SUD criteria can be identified to provide a shorter, easier-to-use list. The team is highly experienced with the issues and methods. Our pilot work demonstrates project feasibility. This revised proposal for a renewal of R01DA018652 responds to PA-11-230 (drug phenotypes for prevention, services and etiologic research), of which conditions defined by DSM-5 form a large, important subset.
In the United States, psychiatric diagnoses are operationalized in the Diagnostic and Statistical Manual of Mental Disorders, which has many purposes, providing a common language and educational tool for clinicians, and an important basis of measurement for researchers, e.g., eligibility criteria for clinical trials. DSM-5 implemented many changes in the definitions of Substance Use Disorders (SUD), but the impact of these changes on reliability and validity is currently unknown. We propose an investigation of these issues to inform the substance abuse and mental health fields, and indicate future improvements in SUD diagnoses in DSM-5.1, DSM-5.2 and beyond.
Showing the most recent 10 out of 53 publications
