Hospital Readmissions and Mortality Patterns in a Cohort with Heart Failure: 1 Year Follow-Up from the Brazilian Public Healthcare System Database

Abstract

Heart failure (HF) affects 1.6 million people in Brazil. About 30%-40% progress to death in a year. This study aims to report 1-year retrospective analyses of hospital readmissions and mortality patterns associated to HF or all-other causes (OC) in a Brazilian cohort of patients with HF. Brazilian Hospital Information System (SIH/SUS) database was used to collect hospital readmissions and mortality data associated to HF or OC, according to ICD-10 codes (I50.0 and I50.9, and “others”, respectively). Patients were followed during 1 year (January-December 2014, regardless the entry data). Eligible criteria included patients with one previous HF hospital admission. A cohort of 21,015 patients was considered eligible. No HF or OC hospital readmissions were observed for 13,050 (62%) and 5,241 (25%) patients, respectively. For HF, 1, 2, 3, 4, and 5 re-hospitalizations were observed for 6,053, 1,255, 395, 152, and 59 patients. One patient required ten re-hospitalizations and the same occurred for 12 and 15 hospital readmissions. Therefore, about 38%, 9%, 3% and 1% of all patients had 1, 2, 3 or 4 hospital readmissions at least. OC readmissions showed higher number of patients for each amount of hospitalizations (11,046, 2,965, 989, 415, and 184 for 1, 2, 3, 4, and 5, respectively). Seven patients required ten re-hospitalizations and 1 required 19. Compared to HF, OC showed approximately twice of all patients with 1 hospital readmission at least (75%). Mortality rate was 19.23%, 702 (3.34%) due to HF and 2,632 (12.52%) due to OC. The number of patients requiring hospitals readmissions for HF remain significant, despite available treatment options in Brazilian Public Healthcare System. OC results showed higher numbers, however, for this analysis all-other hospitalizations causes were included (except HF) indicating a high presence of comorbidities among this population and/or a misregistration at hospitalization time.