Abstract
Background/Aims The Essentia Health Heart Failure Program (HFP) began in 1998, with the primary goal of improving care, with the expectation that both mortality and rehospitalization would be better controlled. This effort reviewed the electronic health records (EHR) of Essentia Health East heart failure patients ages 65 and over (65+), following hospital discharge, to assess the relationships between HFP participation and post-discharge survival (all-cause mortality) 30 days, six months and one year, and rehospitalization at 30 and 60 days.
Methods A case series design used EHR to identify heart failure patients, DRG 291-293, ages 65+, admitted to St. Mary’s Medical Center, Duluth, MN, discharged alive between January 1, 2006 and December 31, 2010. Discharges to hospice were excluded. HFP electronic data classified participation within each post-discharge study period. Due to age and sex differences in HFP participation, logistic regression was used to evaluate sex and age (<85 vs. 85+ years) adjusted odds ratios and associated 95% confidence intervals (CI) for survival at 30 days, six months and one year, and rehospitalization at 30 and 60 days.
Results EHR review identified 770 patients meeting inclusion/exclusion criteria, 397 (51.6%) were male and 257 (33.4%) were 85 years of age and over. HFP participation was 231, 258, 281, and 288 during the first 30 days, 60 days, 6 months, and 1 year post-discharge, respectively. Adjusted odds ratios (CI) for survival were 2.39 (1.19, 4.82), 1.73 (1.21, 2.48), and 1.41 (1.03, 1.93), at 30 days, 6 months and 1 year, respectively. Adjusted odds ratios (CI) for any rehospitalization were 0.58 (0.32, 1.05) and 0.95 (0.59, 1.55), at 30 and 60 days, respectively.
Discussion HFP participation was consistently associated with increased rates of survival, with the association decreasing over time. The association between HFP participation and 30 day rehospitalization was marginal, with the odds ratio well below 1.00, but the CI barely including 1.00. No apparent association existed between HFP participation and 60 day rehospitalization. The 60 day readmission rate is potentially affected by elective readmission for device therapy. Further evaluation including disease characteristics, characteristics of rehospitalizations and underlying cause of death is warranted.
