Short term risk adjusted outcomes for heart failure

Background Cardiovascular diseases are still the leading cause of death worldwide. Heart failure (HF) is now recognized as a major problem in industrialized countries. Short-term adjusted outcomes are indicators of the quality of the diagnostic and therapeutic process during/after hospitalization. The aim of the study is to evaluate the in-hospital mortality and hospital 30-day readmissions for heart failure using two different Risk Adjustment (RA) tools. Methods We used data from the hospital discharge abstract (HD) of a retrospective cohort of heart failure patients (2002-2007) admitted in Tuscan hospitals, in Italy. The outcomes considered were: in-hospital mortality and hospital read- mission at 30 days after discharge. We compare two RA tools: All-Patients Refined Diagnosis Related Groups (APR-DRG) system, based only on data of index hospitalization, and Elixhauser Index (EI), that also include informations on previous hospitalization. Logistic regression models were performed and models were compared using the C statistic (C). Results The study population include 58.202 hospitalizations. The crude in-hospital mortality was 9.7%, lower in females and increasing with age. Thirty-day readmissions was 5.1%, lower in females and higher in patients 85+. The APR-DRG class of risk of death was a predictive factor for in-hospital mortality, while the APR-DRG class of severity was not significantly associated with 30-day readmissions(p > 0,05). EI comorbid- ities associated with outcomes were : non metastatic cancer (OR 2,25, p < 0,05) for in-hospital mortality and diabetes (ranging 1,20-1,24, p < 0,05) for 30 day hospital readmissions. The discriminative abilities for in-hospital mortality were sufficient for both models considered, while were low for 30- day readmissions rate. Conclusions Our findings showed that:1) age, gender, APR-DRG risk of death and some Elixhauser comorbidities are predictive factors of outcomes; 2) of the two RA models in heart failure patients, the APR-DRG showed an acceptable ability to predict hospital mortality while none of them was satisfactory in predicting the readmissions within 30 days. The recognition of patients at risk for these outcomes is a definite advantage to the improvement of health services. Key messages Both the APR-DRG model and the EI model are good predictors for in-hospital mortality. Readmission at 30 days for heart failure were not successfully predictable with APR-DRG and Elixahuser Index