Evaluation of the predictors of readmission in Korean patients with heart failure

Abstract

What is known and objectiveVarious factors contribute to the high rate of readmission among patients hospitalized with heart failure (HF). Determination of these factors is fundamental to identify potential targets for intervention in hospitalized patients. MethodsThe retrospective cohort study used a large national insurance database to identify episodes of HF. Clinical information up to 12 months from the index hospitalization was obtained. Depending on their outcome, eligible patients were classified into a 30-day readmission group after discharge or a non-readmission group. Potential predictors of 30-day readmission were categorized by patient, drug therapy and health system utilization factors. Results and discussionHeart failure was identified in 19 128 inpatients. Of these, 27<bold></bold>6% were readmitted within 30 days after discharge. The mean Charlson comorbidity index (CCI) score was 5<bold></bold>2 2<bold></bold>9 for the readmission group and 4<bold></bold>3 +/- 2<bold></bold>5 for the non-readmission group. The strongest predictors included paralysis [adjusted odds ratio (AOR) 2<bold></bold>27, 95% confidence interval (CI) 1<bold></bold>97-2<bold></bold>62], followed by metastatic cancer (AOR 2<bold></bold>22, 95% CI 1<bold></bold>81-2<bold></bold>72) and loop diuretic therapy (AOR 1<bold></bold>52, 95% CI 1<bold></bold>29-1<bold></bold>79). A prescription of ACE inhibitor or angiotensin receptor blocker at discharge was associated with a 17% decreased risk (AOR 0<bold></bold>83, 95% CI 0<bold></bold>77-0<bold></bold>89). What is new and conclusionsHospitalized patients with HF have a 30-day all-cause readmission rate exceeding a quarter. Post-discharge care should focus on patients with advanced age, acuity of admission, enrolled medical aid, hospitalization exceeding 14 days, higher CCI score, more than 10 prescription drugs at discharge, presence of several comorbidities and loop diuretic therapy, which are independent predictors for 30-day readmission.