Sensitivity to detect change and the correlation of clinical factors with the Hamilton Depression Rating Scale and the Beck Depression Inventory in depressed inpatients

Abstract

Discrepancies between scores on the Hamilton Depression Rating Scale (HAMD) and the Beck Depression Inventory (BDI), as well as differences regarding their sensitivity to detect change, have been reported. This study investigates discrepancies and their potential prediction on the basis of demographic, personality, and clinical factors in depressed inpatients and analyzes the sensitivity to change. The HAMD and the BDI were administered to 105 inpatients with major depressive disorder randomized to 5 weeks of either interpersonal psychotherapy or clinical management. Personality was assessed with the NEO Five-Factor Inventory. Low extraversion and high neuroticism were associated with relatively higher endorsement of depressive symptoms on the BDI compared with the HAMD. The HAMD presented a greater reduction of symptom scores than the BDI. Patients with high BDI scores, high HAMD scores or both revealed the greatest change, possibly due to a statistical effect of regression to the mean. Restricted by sample size, analyses were not differentiated by treatment condition. Regression to the mean cannot be tested directly, but it might be considered as a possible explanation. The HAMD and the BDI should be regarded as two complementary rather than redundant or competing instruments as the discrepancy is associated with personality characteristics. Attributing large effect sizes solely to effective treatment and a sensitive measure may be misleading.