Clinical utility as a criterion for revising psychiatric diagnoses

Changes in DSM-IV were guided by empirical data that mostly focused on improving diagnostic validity and reliability. Although many changes were made explicitly to improve clinical utility, no formal effort was made to empirically determine actual improvements in clinical utility. The authors propose that future revisions of DSM empirically demonstrate improvement in clinical utility to clarify whether the advantages of changing the diagnostic criteria outweigh potential negative consequences.

The authors provide a formal definition of clinical utility and then suggest that the merits of a proposed change to DSM be evaluated by considering 1) its impact on the use of the diagnostic system, 2) whether it enhances clinical decision making, and 3) whether it improves clinical outcome.

Evaluating a change based on its impact on use considers both user acceptability and accuracy in application of the diagnostic criteria. User acceptability can be measured by surveying users' reactions, assessing user acceptability in a field trial setting, and measuring the effects on ease of use. Assessment of the correct application of diagnostic criteria entails comparing the clinician's diagnostic assessment to expert diagnostic assessment. Assessments of the impact on clinical decision making use methods developed for evaluating adherence to practice guidelines. Improvement in outcome entails measuring reduction in symptom severity or improvement in functioning or in documenting the prevention of a future negative outcome.

Empirical methods should be applied to the assessment of changes that purport to improve clinical utility in future revisions of DSM.