Risk stratification and outcome assessment of the adult cardiac surgical patient.

Abstract

Comparisons between institutions are already occurring, and comparisons between individual providers may also become a reality. In spite of the negative views of such comparisons, they are likely to be mandated because of pressure from health care regulators, insurers, and patients. Despite awareness of the importance of demographic variables and concurrent medical problems in influencing outcome, valid comparisons are presently difficult to conduct in the open heart surgical population. Current methods of risk stratification each have limitations. A method for risk assessment based on multivariate analysis from a large group of patients that can be prospectively validated at multiple institutions would be valuable, not only for mortality rate comparisons, but also for patient counseling, research, and hospital management uses. Caution must be applied when using risk assessment in individual patients. Physicians need to be involved in the development of such severity stratifying systems, since inclusion of inappropriate or medically irrelevant data can influence the outcome of multivariate analyses. Ongoing research and evolution of scoring systems also need to occur since therapy changes over time. It is likely that models will need to be developed for application preoperatively, at ICU admission, and for the complex, long-term patient at 7 days or beyond, in order to fully inform medical decision-making.