Do ACE Inhibitors/Angiotensin II type 1 receptor antagonists reduce hospitalisations in older patients with heart failure? A propensity analysis

Abstract

Background: Randomised controlled trials have shown a reduced risk of heart failure (HF) hospitalisation among users of ACE inhibitors (ACEIs) or angiotensin II type 1 receptor antagonists (angiotensin receptor blockers [ARBs]), but these results have limited generalisability. Some observational studies have also demonstrated reductions in hospitalisation but are potentially affected by non-random treatment selection.

Objective: To assess the effect of ACEI/ARB therapy on all-cause and HF-related hospitalisations among older adults using a propensity model to adjust for treatment-selection bias and focusing on consistent medication use as the exposure of interest.

Methods: A retrospective cohort study of continuously enrolled, older (age > or =60 years) Kansas Medicaid beneficiaries with HF, using data from May 1999 to April 2000. A propensity analysis was used to identify a comparison group of untreated persons that were otherwise clinically similar to treated persons. The effect of regular ACEI/ARB use on hospitalisations was estimated using multivariable logistic regression models. The HF sample included 887 subjects, of whom 235 (27%) received regular ACEI/ARB therapy. To be considered a regular user of ACEI/ARB therapy ('treated'), we required evidence that a subject obtained at least 80% of their intended daily supply. The main outcome measure was the effect of regular ACEI/ARB use on all-cause and HF-related hospitalisations.

Results: Treated subjects were matched against an equal number of untreated persons, for a final sample of 470 persons. The mean age of both treated and untreated subjects was 81 years. Regular ACEI/ARB use did not alter the adjusted odds ratio (AOR) of all-cause hospitalisation (AOR = 1.04, 95% CI 0.71, 1.52), which occurred in 40% of the sample, or the odds of an HF-related hospitalisation (AOR = 1.01, 95% CI 0.65, 1.57), which occurred in 22.6% of both groups.

Conclusion: Although randomised controlled trials have shown that ACEI/ARB treatment is associated with reduced hospitalisations in patients with HF, this benefit was not observed in our study. Further study of ACEI/ARB outcomes is needed in a larger sample of older subjects with HF.