The real-world evidence of heart failure: findings from 41 413 patients of the ARNO database

Abstract

Aims: Patients with heart failure (HF) randomized in controlled trials are generally selected and do not fully represent the 'real world'. The purpose of this study is to better describe the characteristics of HF by analysing administrative data of a population of nearly 2 500 000 subjects.

Methods and results: Data came from the ARNO Observatory including inhabitants of five Local Health Units of the Italian National Health Service (INHS). Patients were selected when discharged for HF (1 January 2008-31 December 2012) and prescribed at least one HF treatment. Clinical characteristics, pharmacological treatments, rehospitalization, and direct costs for the INHS were described during 1-year follow-up (FU). Of the 2 456 739 subjects included in the database, 54 059 (2.2%) were hospitalized for HF: 41 413 were discharged alive and prescribed HF treatments. Mean age was 78 ± 11 years and 51.4% were females. Just 26.6% were managed in a cardiology setting. The most frequent co-morbidities were diabetes (30.7%), COPD (30.5%), and depression (21%). ACE inhibitors/ARBs, beta-blockers, and mineralocorticoid antagonists were prescribed in 65.8, 49.7, and 42.1% of patients, respectively. During 1-year FU, at least one rehospitalization occurred in 56.6% of patients, 49% of them due to non-cardiovascular causes. The direct cost per patient per year to the INHS was €11 867, of which 76% was related to hospitalizations.

Conclusions: Real-world evidence provides a description of patient characteristics and treatment patterns that are different from those reported by randomized clinical trials. Costs for the INHS are mainly driven by hospitalizations, which are often due to non-cardiovascular reasons.

Keywords: Heart failure; Pharmacological management; Real-world registries.