Abstract 18182: Patient Health Questionnaire-9 Cut-off Scores for Identifying Patients at Highest Risk for Adverse Health Status Outcomes 1 Year After Acute Myocardial Infarction
Introduction: Depression is strongly associated with adverse cardiovascular outcomes. The Patient Health Questionnaire-9 (PHQ-9) quantifies depressive symptoms and a score of ≥ 10 is considered the optimal threshold for diagnosing major depression. Whether this PHQ-9 threshold optimally identifies patients at risk for poor health status outcomes after acute myocardial infarction (AMI) is unknown.
Methods: In the 24-center TRIUMPH registry of AMI patients, 4062 patients responded to the PHQ-9 during their index hospitalization. Using receiver operating characteristic (ROC) curve analyses of PHQ-9 scores, optimal cut-off scores and areas under the curve (AUC) were calculated for 1-year health status outcomes, as assessed by the Short Form-12 (SF-12) physical and mental component scales (PCS; MCS) and the Seattle Angina Questionnaire angina frequency and quality of life (SAQ-AF; SAQ-QoL) scales. Adverse health status outcomes were defined as the lowest tertiles for PCS (≤ 39.1) and MCS (≤ 50) scores and the clinical thresholds of daily/weekly angina (SAQ-AF≤60) and fair/poor QoL (SAQ-QoL<50). The optimal PHQ-9 threshold for each criterion was identified using the Youden Index and bootstrapping was used to calculate 95% confidence intervals (CI).
Results: ROC analysis revealed optimal cut-off scores of ≥ 4 for SF-PCS (CI = 3, 6), 5 for SF-MCS (CI = 4, 7), and 6 for both SAQ AF (CI = 3, 8) and QoL (CI = 3, 9). In each case, the 95% CI did not include a score of 10. AUC for optimal cut-off scores were similar for all health status outcomes (range 0.63 to 0.68) and higher than AUC for the commonly used cut-off of ≥ 10.
Conclusions: When prognosticating adverse health status outcomes after AMI, PHQ-9 threshold scores of 4-6 are the best for identifying high-risk patients. Although only modest in discriminative ability, scores of 4-6 discriminate patients at high risk for adverse health status outcomes better than the commonly used cut-off of ≥ 10. Further validation studies are needed.
- © 2010 by American Heart Association, Inc.