Usefulness of administrative databases in identifying the predictive factors of repeated hospitalizations for heart failure

OBJECTIVES: To investigate heart failure (HF) hospitalizations, the following one-year follow-up, and any possible connection between rehospitalizations due to HF and patients’ characteristics derived from administrative databases. DESIGN: Retrospective longitudinal design. SETTING AND PARTICIPANTS: The study was conducted analyzing public hospital records of a district in Abruzzo Region (Central Italy), which counts more than 300,000 inhabitants. Patients hospitalized for HF from 01.01.2016 to 31.12.2017 (index event) were included in the study and followed- up for one year. MAIN OUTCOME MEASURES: Frequency of repeated hospital admissions, time intervals from the index HF hospitalizations, and causes of readmissions were investigated. RESULTS: A total of 1,587 patients discharged alive after an index hospitalization for HF were included in the study. The mean age of the patients was 79.6 years and the majority of them were females (53.7%). The mean length of stay (LOS) for the index hospitalizations was 8.8 ±6.8 days. During the follow- up period, 336 (21.2%) patients underwent one to four repeated hospitalizations for HF. The first readmission due to HF occurred after a median time of 106.5 days from the index event discharge, and for 20.0% of all cases it occurred within 31 days; 453 patients (28.6%) were readmitted exclusively for other causes, and 67 (4.2%) died out of hospital without any previous HF re-hospitalization. When the outcome was considered as a composite endpoint (out-of-hospital death/HF re-hospitalization), age 75 (HR 1.737; 95%CI 1.330-2.267), LOS at the index hospitalization 8 days (HR 1.302; 95%CI 1.066-1.591), and repeated hospitalizations for other causes (HR 1.789; 95%IC 1.465-2.185) were associated with the risk of repeated hospitalizations for HF. CONCLUSIONS: This study shows that about one HF patient out of five experienced at least one re-hospitalization for HF within one year from index hospitalization. In addition to having a longer index hospitalization, these patients were older and frequently suffered from comorbidities which also led to hospitalizations. The results underline the need for a close and careful follow-up after the discharge of old HF patients with multiple pathologies in order to avoid further HF admissions in a short time.