Comparative Potential of the 2-Item Versus the 9-Item Patient Health Questionnaire to Predict Death or Rehospitalization in Heart Failure

Abstract

BACKGROUND: Depression is common in heart failure and associated with adverse clinical outcomes. We investigated the potential of the 2-item patient health questionnaire (PHQ-2) versus that of the 9-item version (PHQ-9) to predict death or rehospitalization. METHODS AND RESULTS: Participants of the Interdisciplinary Network for Heart Failure program were eligible, if they completed the PHQ-9 during baseline assessment. All participants were hospitalized for cardiac decompensation and had a left ventricular ejection fraction ≤40% before discharge. PHQ-2 scores were extracted from the answers to the first 2 PHQ-9 questions. To analyze associations of PHQ-2 and PHQ-9 with both, death and rehospitalization, univariable Cox regression models were used. To compare screening efficacy of both tools, c-statistics were computed. The sample consisted of 852 patients, (67.6±12.1 years; 27.7% women; 42.3% New York Heart Association class III/IV). Follow-up was 18 months (100% complete). During follow-up, 152 patients died and 482 were rehospitalized. Both, PHQ-2 and PHQ-9, predicted death in univariable analysis (hazard ratio, 1.18; 95% confidence interval, 1.09-1.29; P<0.001 and hazard ratio, 1.07; 95% confidence interval, 1.04-1.09; P<0.001, respectively), as well as rehospitalization (hazard ratio, 1.07; confidence interval, 1.01-1.21; P=0.02 and hazard ratio, 1.03; confidence interval, 1.01-1.04; P=0.001, respectively). These results were confirmed by c-statistics. CONCLUSIONS: In univariable models and confirmed by c-statistics the potential of both PHQ-2 and PHQ-9 to predict death and hospitalization was similar. In clinical practice, PHQ-2 screening seems thus sufficiently reliable and more feasible than the time-consuming PHQ-9 to identify patients at an increased risk of adverse outcomes. CLINICAL TRIAL REGISTRATION: URL: http://www.controlled-trials.com. Unique identifier: ISRCTN 23325295.