Variation in readmission and mortality following hospitalisation with a diagnosis of heart failure: prospective cohort study using linked data

BACKGROUND: Hospitalisation for heart failure is common and post-discharge outcomes, including readmission and mortality, are often poor and are poorly understood. The purpose of this study was to examine patient- and hospital-level variation in the risk of 30-day unplanned readmission and mortality following discharge from hospital with a diagnosis of heart failure. METHODS: Prospective cohort study using data from the Sax Institute's 45 and Up Study, linking baseline survey (Jan 2006-April 2009) to hospital and mortality data (to Dec 2011). Primary outcomes in those admitted to hospital with heart failure included unplanned readmission, mortality and combined unplanned readmission/mortality, within 30 days of discharge. Multilevel models quantified the variation in outcomes between hospitals and examined associations with patient- and hospital-level characteristics. RESULTS: There were 5074 participants with a heart failure admission discharged from 251 hospitals; 1052 (21%) had unplanned readmissions, 186 (3.7%) died, and 1146 (23%) had either/both outcomes within 30 days of discharge. Crude outcomes varied across hospitals, but between-hospital variation explained little of the total variation in outcomes (intraclass correlation coefficients (ICC) after inclusion of patient factors: 30-day unplanned readmission ICC = 0.0125 (p = 0.24); death ICC = 0.0000 (p > 0.99); unplanned readmission/death ICC = 0.0266 (p = 0.07)). Patient characteristics associated with a higher risk of unplanned readmission included: being male (male vs female, adjusted odds ratio (aOR) = 1.18, 95% CI: 1.00-1.37); prior hospitalisation for cardiovascular disease (aOR = 1.44, 1.08-1.91) and for anemia (aOR = 1.36, 1.14-1.63); comorbidities at admission (severe vs none: aOR = 1.26, 1.03-1.54); lower body-mass-index (obese vs normal weight: aOR = 0.77, 0.63-0.94); and lower social interaction scores. Similarly, risk of 30-day mortality was associated with patient- rather than hospital-level factors, in particular age (≥85y vs 45-< 75y: aOR = 3.23, 1.93-5.41) and comorbidity (severe vs none: aOR = 2.68, 1.82-3.94). CONCLUSIONS: The issue of high readmission and mortality rates in people with heart failure appear to be system-wide, with the variation in these outcomes essentially attributable to variation between patients rather than hospitals. The findings suggest that there are limitations in using these outcomes as hospital performance measures in this patient population and support the need for patient-centred strategies to optimise heart failure management and outcomes.This specific project was supported by NSW Agency for Clinical Innovation. Emily Banks is supported by the National Health and Medical Research Council of Australia