Outcomes of cardiac surgery: associations with physician characteristics, institutional characteristics, and transfers of care.

BACKGROUND

Although there are several studies of the human and system factors that influence the outcomes of cardiac surgery, it remains difficult to draw conclusions because many do not simultaneously adjust for the characteristics of patients, physicians, and institutions. The current study explores the associations between these factors and inhospital mortality, with a particular focus on whether patients had the same operating and attending physician.

METHOD AND RESULTS

This is a retrospective observational study of 114,751 hospitalizations from 2003 to 2009 in Pennsylvania that included a coronary artery bypass graft, valve surgery, or both. The study included 70 teaching and nonteaching hospitals, 289 operating physicians who were also the attending physicians for 75% of the hospitalizations, and 2654 attending physicians for the remaining hospitalizations. After adjustment, there was a 38.4% decrease (95% CI, 20.3%-56.5%) in mortality when the operating and attending physician were the same. For the operator, each procedure performed was associated with a 0.05% (95% CI, 0.04%-0.06%) decrease in mortality and each year since medical school was associated with a 0.9% (95% CI, 0.02%-1.8%) increase in mortality. For the attending, each year since medical school was associated with a 0.67% (95% CI, 0.01%-1.4%) decrease in patient mortality.

CONCLUSIONS

The findings indicated that an increase in the log odds of mortality was associated with the transfer of care between an attending and operating physician. Better patient outcomes were associated with an operator with higher volume who was closer to medical school graduation and an attending physician with more clinical experience.