Current depressive symptoms but not history of depression predict hospital readmission or death after discharge from medical wards: a multisite prospective cohort study

Abstract

Objective: Although death or readmission shortly after hospital discharge is frequent, identifying inpatients at higher risk is difficult. We evaluated whether in-hospital depressive symptoms (hereafter "depression") are associated with short-term readmission or mortality after discharge from medical wards.

Methods: Depression was assessed at discharge in a prospective inpatient cohort from 2 Canadian hospitals (7 medical wards) and defined as scores ≥ 11 on the 27-point Patient Health Questionnaire (PHQ-9). Primary outcome was all-cause readmission or mortality 90 days postdischarge.

Results: Of 495 medical patients [median age 64 years, 51% women, top 3 admitting diagnoses heart failure (10%), pneumonia (10%) and chronic obstructive pulmonary disease (8%)], 127 (26%) screened positive for depression at discharge. Compared with nondepressed patients, those with depression were more frequently readmitted or died: 27/127 (21%) vs. 58/368 (16%) within 30 days and 46 (36%) vs. 91 (25%) within 90 days [adjusted odds ratio (aOR) 2.00, 95% confidence interval 1.25-3.17, P=.004, adjusted for age, sex and readmission/death prediction scores]. History of depression did not predict 90-day events (aOR 1.05, 95% CI 0.64-1.72, P=.84). Depression persisted in 40% of patients at 30 days and 17% at 90 days.

Conclusions: Depression was common, underrecognized and often persisted postdischarge. Current symptoms of depression, but not history, identified greater risk of short-term events independent of current risk prediction rules.

Keywords: Depression; Inpatients; Interviews; Mortality; Patient readmission.