P1636Causes and risk factors for non-cardiovascular hospital readmissions among heart failure patients

Abstract Background/Introduction Despite better patient-management and technological improvements, patients with heart failure (HF) have poor prognosis and impaired health-related quality of life (HRQoL). HF patients also experience significant non-cardiovascular morbidity leading to higher disease burden and recurrent hospitalizations, which also put a heavy burden on healthcare systems. Purpose To describe causes and risk factors for non-cardiovascular hospital readmissions among HF patients. Methods We analyzed 3,172 hospital discharge summaries for 927 patients enrolled in a clinical trial testing the efficacy of a disease management program compared to usual care between August 2007 and June 2011. The study intervention had no effect on hospital admission and mortality rates. Baseline information on patient characteristics included age, gender, LVEF, NYHA classification, 6-minute walk test, BNP level, BMI, chronic comorbidity and HRQoL score. Analysis of hospital discharge summaries included identification and coding of the primary hospital admission cause. These causes were further grouped to HF-related, other cardiovascular disease-related and non-cardiovascular causes. Information on follow-up prescription purchases of HF drugs was also collected. Factors associated with non-cardiovascular hospital admission rate were tested in multivariable models, adjusted for study group, study period and year of recruitment. Results Non-cardiovascular hospital admissions (of which 77% were non-elective) accounted for 45% of all hospital readmissions. Of these hospitalizations, 23% were due to infectious diseases, 13% due to gastrointestinal causes, 12% due to respiratory causes and 8% due to falls and associated trauma. Patients with two or more non-cardiovascular hospital admissions during follow-up were significantly older, more commonly female, had higher proportion of preserved LVEF (≥50%), worse baseline functional capacity and HRQoL score, and higher co-morbidity score, BNP level and mortality rates compared to other patients. Factors significantly and independently associated with higher acute non-cardiovascular hospitalization rate included [rate ratios (RRs), 95% confidence intervals (CIs)]: non-preserved vs. reduced LVEF [1.48 (1.14, 1.92)]; higher baseline comorbidity score [1.14 (1.08, 1.21) per 1-point increment]; and higher loop diuretics, beta-adrenergic receptor blockers and ACE-I/ARBs treatment level during follow-up; 4th vs. 1st quartiles: 4.13 (2.93, 6.32), 1.51 (1.07, 2.13) and 1.32 (0.96, 1.82), respectively. Baseline overweight/obesity and greater 6-min. walking distance [0.93 (0.89, 0.93) per 50m increment] were found protective. Conclusion This study indicates a higher disease burden and mortality rate among HF patients with recurrent non-cardiovascular hospital admissions. It provides pertinent information needed for the design and implementation of effective interventions to reduce disease burden in HF patients.