Effect of the readmission primary diagnosis and time interval in heart failure patients: analysis of English administrative data

Aims To compare the predictors of unplanned readmission by primary diagnosis and time since discharge in heart failure ( HF ) patients. Methods and results We used national hospital administrative data for England to analyse unplanned readmission by primary diagnosis ( HF and non‐ HF ) at 7, 30, 90, 182, and 365 days after the index discharge. A total of 84 212 adult patients had their first HF admission between April 2008 and March 2010; 14 104 (16.8%) died during the index admission and were excluded. Of the remaining 70 108, half were readmitted and 28.7% died during 1 year from discharge (overall mortality rate of 40.6%). Patients had an average of three co‐morbidities. Hierarchical logistic regression showed that arrhythmias [odds ratio ( OR ) = 1.13] and valvular disease ( OR  = 1.12) had significantly higher odds only for HF readmission; dementia ( OR  = 1.29), stroke ( OR  = 1.29), and mental health conditions ( OR  = 1.25) had higher odds only for non‐ HF . Ischaemic heart disease, renal disease, and chronic lung disease predicted both. Same‐day discharge occurred for 6% of patients and was strongly associated with higher readmission for HF at 7 days, less so thereafter, and not for non‐ HF after 7 days. Other relationships changed little between 7 and 365 days. Prior outpatient non‐attendance was associated with 5–10% higher odds of any readmission per appointment missed. Conclusion In HF patients, some predictors of readmission for HF , especially some common co‐morbidities, differ from those for non‐ HF . In contrast, the time since discharge made little difference to the results.