Risk of Hospitalization among Elderly Heart Failure Patients initiating Beta-Blockers or ACE-Inhibitors
MetadataShow full item record
Objective: For elderly(age≥65 years) Heart Failure(HF) patients an early initiation of Beta-Blockers(BBs) vs. Angiotensin-Converting Enzyme Inhibitor(ACEIs) could have an impact on the risk of HF hospitalization. Thus the aim of this study is to estimate the determinants of initiating a therapy and their risk of first HF hospitalization, of recurrent HF hospitalization and composite outcome of HF hospitalization with death as the terminal event. Methods: Using 2008-2011 Medicare Advantage Prescription Drug Plan, elderly HF patients initiating ACEI or BBs, defined as category 1(BBc1 e.g. Metoprolol, Bisoprolol etc.) and category 3(BBc3 e.g. Carvedilol and Labetalol), were identified. The baseline determinants of initiating either of these drugs were identified by Multinomial logistic regression analysis. The survival probabilities of first HF hospitalization, recurrent HF hospitalization and of composite outcome were obtained by inverse probability weighted count process model, by stratified total time and gap model and by the marginal model respectively for a follow-up of a year using SAS 9.3 at the p-value of 0.05. Results: Of all 6430 eligible patients, there were 248(3.86%) events of first HF hospitalization with 55(0.86%) events of ≥2 recurrent hospitalization. Of these, ACEI, BBc1, BBc3 and a combination of ACEI with BBc1(ABC1) and with BBc3(ABC3) were initiated by 1194(18.57%), 1519(23.62%), 490(7.62%), 257(4.00%) and by 126(1.96%) patients respectively. This likelihood for initiating BBc1 compared to ACEI increased statistically significantly by atrial fibrillation/flutter(OR:1.348,95%CI:1.09-1.667), Age(OR:1.017,95%CI:1.005-1.028), with vasodilators(OR:1.373,95%CI:1.051-1.795) and with a missing BNP value(OR:1.97,95%CI:1.085-3.578). For initiating BBc3 compared to ACEI the likelihood increased statistically significantly for males(OR:1.294,95%CI:1.039-1.612), for patients with chronic atherosclerosis(OR:1.502,95%CI:1.12-2.015), for those taking vasodilators(OR:1.600,1.144-2.238) or Diuretics (OR:1.306,95%CI:1.05-1.625) or having an elevated BNP value(OR:2.465,95%CI:1.008-6.029). It decreased statistically significantly for patients taking Calcium Channel Blockers(OR:0.718,95%CI:0.547-0.942) in the washout period. The hazard ratio for 3 respective outcomes were 0.562(95%CI:0.28- 1.13), 0.969(95%CI: 0.47- 1.996) and 1.32(95%CI: 0.702- 2.483) for BBc1 as the initiation therapy. For BBc3, these were 1.242(95%CI:0.64- 2.43), 0.871(95%CI: 0.51- 1.48) and 1.41(95%CI: 0.84- 2.36) respectively. Conclusions: There is a statistically non-significant difference in the effect of an early initiated BB in comparison to ACEI for the risk of HF hospitalization.