Abstract 3168: Repeated Heart Failure Hospitalizations Predict Mortality in Regional Population
Context: Prediction of outcomes for patients with heart failure (HF) helps physicians to guide therapy and patients to plan. The prognostic importance of repeated HF hospitalizations has not been established for a large regional population.
Method: We identified a cohort of patients with a first HF hospitalization among all residents of British Columbia (2000–2004) using health care utilization data and followed for subsequent hospitalizations and death. Survival times were measured after the 1st and each subsequent HF hospitalization, and Kaplan-Meier curves were plotted. The effect of the number of HF hospitalizations was estimated using a multivariate Cox model.
Results: The mean age at first HF hospitalization was 77, 45% were women, 47% had coronary artery disease, 11% had cerebrovascular diseases, and 21% had chronic kidney diseases (CKD). Deaths occurred in 7,401 of 14,374 patients, which were attributed to cardiovascular causes in 61% cases, over half of which occurred in hospital. The risk of death increased after each HF hospitalization [Figure⇓]. After adjusting for age, gender and major comorbidities, the number of HF hospitalizations was a strong predictor of all-cause death and cardiovascular death. Advanced age (>=75) and CKD attenuated the effect of the number of HF hospitalizations. Most patients were alive 2 years after the 1st HF hospitalization but half were dead by 1 year after 3 hospitalizations.
Conclusions: The number of HF hospitalizations predicts outcome and should be tracked as part of HF management. Three or more HF hospitalizations should trigger re-evaluation of therapies and discussion with patient/family regarding prognosis.