Alcohol use in hazardous situations: implications for DSM-IV and DSM-5 alcohol use disorders.

BACKGROUND: The use in hazardous situations criterion (hazard) is 1 of 4 criteria related to alcohol abuse in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), and 1 of the 11 criteria related to the new alcohol use disorder (AUD) proposed for DSM-5. The current study aims to evaluate the hazard criterion in the context of both DSM-IV alcohol abuse and DSM-5 AUD.

METHODS: Data came from the 2007 Australian National Survey of Mental Health and Well-Being (n = 8,841) as a stratified, multistage area probability sample of persons aged 16 to 85 years. Logistic regressions were used to: (i) compare the clinical characteristics of those with alcohol abuse including hazard and those with alcohol abuse because of other criteria; (ii) investigate the relationship between the hazard criterion and indices of socioeconomic status (SES); and (iii) investigate the effect of eliminating the hazard criterion on the epidemiology and correlates of the proposed DSM-5 AUD diagnosis.

RESULTS: When compared with the other abuse subgroup, those with abuse including hazard (irrespective of other abuse criteria endorsed) were more likely to report another drug use disorder. The 2 abuse subgroups could not be differentiated by any other clinical characteristics. There were no systematic relationships between the hazard criterion and indices of SES. The elimination of the hazard criterion would lead to a considerable decrease in the prevalence of AUD, with those no longer receiving a diagnosis more likely to be young males with drug use disorders and suicidal behaviors.

CONCLUSIONS: The current study failed to replicate previous analyses that indicated problems with the hazard criterion when assessed in the U.S. population. Many of the problems identified in the hazard criterion appear to be due to operationalizations of this criterion that includes items specifically related to drink-driving. The current results indicate that the elimination of the hazard criterion would lead to a considerable decline in the prevalence of DSM-5 AUD and risk excluding a potentially clinically significant subtype of AUD from future diagnosis.