Will Proposed DSM-5 Changes to Assessment of Alcohol Problems Do Any Better?

Proposed changes to the upcoming fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) will affect the criteria used to assess alcohol problems. One change would collapse the two diagnoses of alcohol abuse (AA) and alcohol dependence (AD) into a single diagnosis called alcohol use disorder (AUD). A second change would remove "legal problems," and a third would add a criterion of "craving." A study of the potential consequences of these changes has found they are unlikely to significantly change the prevalence of diagnoses.

Results will be published in the March 2013 issue of Alcoholism: Clinical & Experimental Research and are currently available at Early View.

"Updating the DSM could be advantageous if changes are made based on improvements in our understanding of a disorder's etiology, and/or if changes improve the accuracy of the diagnosis," said Alexis C. Edwards, assistant professor in the Department of Psychiatry at Virginia Commonwealth University School of Medicine as well as corresponding author for the study. "It would probably be a little disappointing if no changes were ever made, because that might suggest that we haven't made much headway in understanding and accurately diagnosing psychiatric disorders, despite all our efforts."

"The DSM represents the 'rulebook' for assigning a psychiatric diagnosis, including one of AA or AD," noted Arpana Agrawal, assistant professor of psychiatry at Washington University School of Medicine. "DSM definitions of psychiatric disorders determine how common or 'prevalent' psychiatric disorders are in the population and who might be eligible for treatment-related benefits. For example, AD is a psychiatric disorder currently diagnosed by the presence of three or more criteria occurring within a single 12-month period versus problematic aspects of behavior such as binge drinking."

Edwards added that there is empirical evidence that AA and AD exist on a continuum. "That is, 'abuse' is different from 'dependence' in degree but not in kind," she said. "If this is the case, then it might be helpful for researchers and clinicians to talk about a general AUD phenomenon rather than operating under a false dichotomy."

Edwards and her colleagues used a population-based sample of twins assessed for lifetime AA and AD diagnoses to explore phenotypic differences across DSM-Fourth Edition and DSM-5. Craving was not included in the analysis due to its exclusion in the fourth edition. "One potential downside to changing the criteria is that information on 'craving' is unavailable in many existing datasets, so many researchers will be limited to examining DSM-IV criteria," observed Edwards.

"One of the key findings is that the prevalence of the new diagnosis, AUD, is unlikely to be much higher than the prevalences we've seen for abuse and dependence combined," said Edwards. "Another key finding is that the genetic etiology of the new diagnosis is almost entirely consistent with that of the current diagnoses. Familial aggregation is an important validator for psychiatric problems, and our findings indicate that the genetic risk captured by AA and AD diagnoses is also captured by the modified AUD diagnosis we used. Finally, our results suggest that the removal of the legal problems criterion in unlikely to adversely affect diagnostic validity as that criterion does not appear to be contributing anything critical to the current diagnosis."

"There are two novel aspects to this particular study," added Agrawal. "First, the authors posit that DSM-5 does not represent a marked improvement in diagnostic validity as those individuals who are newly diagnosed with AUDs do not represent a more severely affected group than those who will no longer be diagnosed - this assertion deserves more research attention and additional study. Second, and particularly unique to this study, they also find that heritable influences on DSM-IV and DSM-5 AUDs, while largely overlapping, do include some genetic factors that are specific to DSM-5 alcohol use disorders - this may turn out to be a particularly exciting research finding."

"It isn't clear that the people who would now be diagnosed with AUD are really any more 'deserving' of that diagnosis than the people who will now be left out diagnostically; rather, it is possible that these changes will result in an exchange of one mildly affected group of people for another," said Edwards. "In addition, craving, which we were unable to include here since it wasn't available in the dataset, could very well add a lot to our understanding of alcohol problems. Furthermore, our findings confirm previous reports that asking about family history can be critical in assessing an individual's risk for alcohol problems. Finally, it's clear that alcohol use problems rarely occur in isolation, so people should be aware that if they've experienced drinking problems, they are also likely to have problems with depression, anxiety, and other substances."

"Clinicians should find it reassuring that an overwhelming majority of those who are currently diagnosed with DSM-IV alcohol abuse/dependence, particularly dependence, will continue to receive a diagnosis with DSM-5 criteria," said Agrawal. "It is important to recognize that the changes coming with the 2013 publication of DSM-5 are based on the accumulated wisdom gleaned from the scientific literature. That said, one group of individuals previously diagnosed with DSM-IV alcohol abuse will no longer receive a diagnosis: these individuals tend to be male, belong to a higher socioeconomic stratum, and frequently endorse drunk driving as the sole symptom of alcohol problems. As these individuals may no longer be eligible for treatment-related benefits, their wellbeing and, in particular, their possible progression to AUDs, should be carefully monitored. Yet overall, the changes are unlikely to have significant impact on the treatment of AUDs."