Revising Substance-Related Disorders in the DSM-5: A History.

OBJECTIVE

This article narrates a history of several important changes to the substance-related disorders chapter in the (DSM-5), based on interviews with people involved in the pre-planning and the development of the revisions. These changes include collapsing substance abuse and substance dependence into a single substance use disorder, adding craving as a diagnostic criterion, and incorporating a behavioral addiction--gambling disorder--into the substance-related disorders chapter. Studies using Item Response Theory (IRT) supported the new substance use disorder diagnosis. The IRT analyses demonstrated that the abuse and dependence items can be ordered on a single latent dimension and that some of the presumably milder abuse items indexed a greater level of severity than the presumably more pathological dependence items. Those who opposed collapsing abuse and dependence emphasized the validity and clinical utility of the dependence syndrome on which much important treatment research was based. Both those who favored and those who opposed adding craving agreed that it was redundant with the other diagnostic criteria and did not improve the performance of the criterion set. Nevertheless, some clinicians supported adding craving because of its importance in the conceptualization of substance use disorders, and some researchers supported it because of its potential to be validated as a diagnostically useful biomarker. Those who opposed adding craving argued that considering the validity of an individual criterion alone rather than its contribution to the incremental validity of the criterion set represented a major shift in diagnostic philosophy that had potentially far-reaching implications for future revisions of the DSM.

CONCLUSIONS

We conclude by observing that, unlike what occurred in the broader DSM-5 process, despite differences of opinion the work group reached consensus. In part, this may be explained by some shared standards within the work group versus the disagreement about standards across the broader DSM-5 process.