The classification of DSM-5 alcohol use disorders: Construct and predictive validity, sub-types and comorbidity

• Main Author: Casey, Martina
• Published: Ulster University 2014
• Subjects: 616.861
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.604553

By first identifying and then addressing problems within a classificatory system for Alcohol Use Disorders (AUDs) knowledge of the disorder can be improved. This enhanced information can be used to ultimately improve patient diagnoses and prognoses. Proposed amendments for the DSM-5 , published in 201 0, presented the opportunity to assess the resulting implications to the AUD classification and to examine various aspects of the proposed collective representation of AUDs. Utilising the proposed changes, this thesis explored a number of issues using sophisticated statistical techniques to analyse data from the NESARC survey. Changes to the representation of AUDs and their diagnostic criteria in the DSM-5 resulted in an improved model of AUDs over the DSM-IV. Utilising 15 putative grouping indicators, three AUD latent Classes were identified, differentiated on a severity gradient. This severity gradient differentiation provided further support for the new AUD unidimensional latent construct envisaged in the DSM-5. The Classes were found to differ in their presentation of covariates, with high levels of comorbidity for other clinical disorders evident, particularly additional substance use disorders (SUDs). In assessing the predictive validity of a DSM-5 AUD diagnosis, diagnostic latent status stability was found to be extremely high. Predictors of the baseline statuses showed that younger, male smokers with additional multiple comorbidity were at increased risk for an AUD diagnosis. However males and those with an additional SUD were also more likely to transition, with the dominant transition pattern being one of regression. Some evidence of natural recovery was also found, as having previously sought help for drinking was not predictive of a transition. Whilst findings from these studies suggest that the DSM-5 classification of AUDs is an improved representation of the disorder, a number of issues remain. Highlighted areas for further research include the severity range captured by the criteria plus possible existent criteria redundancy, Longitudinal research is also needed to establish temporal patterns of AUD diagnostic transitions, their stability and underlying motivational factors. This will add to the cannon of knowledge on natural disorder remission and maintenance.