Assessing recovery in depression: Validity of symptom inventories

Depression inventories are commonly used to assess outcome in studies of the treatment and course of affective disorders. Unfortunately, different investigators use different cutoff scores to define recovery, limiting the comparability of studies. Moreover, not even the most commonly employed cutoffs have been shown to correspond to qualitative assessments of recovery. The present study addressed this issue by exploring the relationship between scores on two quantitative depression inventories, the Beck Depression Inventory and the Carroll Rating Scale for Depression, and qualitative interview-based assessments of recovery in a 6-month follow-up of 79 depressed outpatients. Kappa and the predictive power of various cutoffs for each inventory were determined. Using traditional cutoffs, both inventories were moderately good predictors of recovery, suggesting that their use in outcome studies is warranted. However, alternative cutoffs may provide somewhat better indices of recovery from depression.