Background: Heart failure readmission has continued to be a growing chronic condition in the United States and has reached epidemic proportions. The disease is considered a major clinical and public health problem and the future cost to treat and manage people with heart failure is of great concern to many health experts. The problem addressed in this research study was the increasing rate of readmissions for people with heart failure, and further understanding the factors associated with those readmissions.

Objective: The purpose of this quantitative study was to investigate potential relationships between 30-day heart failure readmissions and non-clinical factors, to include follow-up appointment status, clinic sites, geographic regions, hospitals, insurance type, mental health status and annual period (2018 and 2019). The theoretical conceptual framework utilized in this research study was the Expanded Chronic Care Model (ECCM). Given the focus of this research study around non-clinical factors to predict heart failure readmissions, this theoretical orientation correlates to the most significant determinants of heart failure health which are non-clinical attributes, designs of delivery systems, and decision support factors.

Design: A retrospective study was utilized where data was extracted from nine hospitals within a healthcare system located in the Mid-Atlantic region. Data from people between the ages of 55 and 84 with a diagnosis of heart failure based on the ICD-10-CM codes were included in the population within a two-year period: 2018 and 2019. Bivariate and multivariate statistical analyses were conducted using binomial logistic regression test to estimate the relative odds of re-admission associated with each patient characteristic and controlling for covariates: Age, Sex, Ethnicity, and Comorbidity. To assess the strength of association, the odds ratios (OR) effect size and 95% confidence intervals (CI) were calculated.

Results: This quantitative retrospective study had a study population of 5,799 people who were admitted to one of nine hospitals in a healthcare system with a diagnosis of heart failure. Binomial logistic regression was conducted to assess the relative odds of 30-day HF readmission (DV) associated with each patient characteristic. The nine hospitals reflected the average person with HF readmitted within 30 days was 67%, with a range of 48% (lowest) to 86% (highest). Bivariate Binomial models that were statistically significant include ClinicSite χ²(1) = 417.22, p < .001; Hospitals χ²(8) = 342.99, p < .001; Insurance Type χ²(2) = 36.29, p < .001; and Period χ²(1) = 11.41, p < .001. All multivariate models controlling for the covariates were statistically significant with the overall model χ²(22) = 640.01, p < .001.

Conclusion: There is a statistically significant relationship of non-clinical variables to determine the odds of observing a 30-day hospital readmission for people with heart failure when controlling for age, sex, ethnicity and CCI. Keywords: readmission, ejection fraction, NYHA (HF) Class, risk prediction models, Charlson Comorbidity Index (CCI).