Diagnostic accuracy of self-rating scales for screening of depression in coronary artery disease patients

Abstract

Objective: We evaluated the internal consistency and psychometric properties of the Hospital Anxiety and Depression Scale (HADS) and the Beck Depression Inventory-II (BDI-II) for screening of major depressive episodes (MDE) in coronary artery disease (CAD) patients undergoing rehabilitation.

Methods: Five-hundred and twenty-two consecutive CAD patients (72% men; mean age 58±9 years) attending a rehabilitation program 2 weeks after inpatient treatment for acute ischemic cardiac events completed the HADS depression subscale (HADS-D), HADS anxiety subscale (HADS-A) and the BDI-II. Interview outcome using the Mini International Neuropsychiatric Interview (MINI) for current MDE according to the DSM-IV-TR criteria was considered as the gold standard.

Results: Fifty-six (11%) patients had a current MDE. The HADS-D, HADS-A, HADS-total and BDI-II had high internal consistency. Area under the ROC curve was the highest for the BDI-II followed by the HADS. Optimal cut-off values for screening of MDE were ≥5 for the HADS-D, ≥8 for the HADS-A and ≥14 for the HADS-total and for the BDI-II. At optimal cut-off values the BDI-II had slightly superior psychometric properties when compared to the HADS. However, positive predictive values were low for the HADS and for the BDI-II.

Conclusions: In CAD patients undergoing rehabilitation, the HADS and BDI-II had high internal consistency. Screening for MDE at optimal cut-off values the BDI-II was slightly superior when compared to the HADS. Positive predictive values for the BDI-II and for the HADS were low indicating that a large proportion of patients with positive screening results did not meet criteria for MDE.