Background: In-hospital mortality is frequently used as an outcomes measure for surgical procedures. Recently, hospitals have developed subacute care facilities to allow earlier discharge. Outcomes of patients discharged (transferred) to these units or to other similar facilities may not be captured in reports of in-hospital mortality.
Study design: The California Office of Statewide Health Planning and Development (OSHPD) patient discharge abstract database was examined to determine the rates of discharge to other facilities (transfer) and the number of in-hospital deaths occurring during the index hospitalization and after transfer in patients undergoing cardiac surgery procedures. Data were collected for 1997, 1998, and 1999 for coronary artery bypass grafting (CABG-only, n = 82,897), CABG plus additional procedures (CABG-plus, n = 11,869), and valve repair or replacement (Valve-only, n = 14,872). In-hospital mortality and transfer rates (same-day discharge and readmission to another facility) were determined for all hospitals through the index hospitalization and subsequent transfers.
Results: Aggregated 3-year in-hospital mortality rates for the index hospitalization were 2.98% for CABG-only, 9.25% for CABG-plus, and 4.85% in Valve-only groups. Transfer rates were 12.41%, 23.16%, and 13.43%, respectively. The percentages of all in-hospital deaths occurring after transfer from the index hospital were 13.5% (385 of 2,857) in CABG-only, 13.3% (168 of 1,266) in CABG-plus, and 11.0% (89 of 811) in Valve-only patients. When corrected for these additional deaths, the actual in-hospital mortality rate was 3.45% for CABG-only, 10.67% for CABG-plus, and 5.45% for Valve-only procedures.
Conclusions: Transfer to another healthcare facility rather than discharge home is a common practice after cardiac surgery. A substantial percentage of in-hospital deaths occurs after discharge from the primary institution.