Efficient assessment of depressive symptoms and their prognostic value in myocardial infarction patients

Background: Identification of post-myocardial infarction (MI) depressive symptoms is a major concern and clinicians are in need of a short, easy-to-use assessment tool for this population. We evaluated the utility and prognostic power of a brief 10-item version of the Beck Depression Inventory (BDI10) in post-MI patients. Methods: Patients (n = 416) were assessed on demographic/clinical variables and completed the BDI10, BDI-21 and STAI (anxiety symptoms) two months post-MI. The end point was a composite of cardiac death and recurrent MI. The average follow-up period was 2.7 years and follow-up data was complete for all patients. Results: Factor analysis of the BDI10 yielded a one-factor model. The BDI10 had good reliability (alpha = .82) and correlated highly (r = .89) with the standard BDI measure of depressive symptoms; a BDI-derived proxy measure of the BDI10 correlated .94 with the actual BDI10 score. There were 41 events attributable to cardiac death (n = 24) or recurrent MI (n = 20). The mean level of depressive symptoms as measured by the BDI10 was significantly higher in patients who experienced an event (4.3 +/- 4.4) compared with event-free patients (2.6 +/- 2.8); p = 0.015. The BDI10 (HR:1.18; 95%CI:1.08-1.29, p<0.0001) independently predicted death/recurrent MI adjusting for disease severity. Age, BMI and cardiac history were also independent predictors of death/recurrent MI in these analyses. Limitations: The low number of women (22%), relatively healthy sample (mean LVEF = 52%) and lack of a Structured interview as gold standard for assessment of depression. Conclusions: The BDI10 is a brief, valid, and easy-to-use self-report measure of depressive symptoms that predicts post-MI outcomes and hence can be used clinically for risk stratification purposes, while reducing assessment burden. (C) 2009 Elsevier B.V. All rights reserved.