Death and quality in cardiac surgery
Quality monitoring in medicine was a neglected field until the last two decades. Doctors traditionally did their best, but how good that was could not be evaluated. This situation remains in some areas of medicine, but specialties with clearly-defined interventions and outcomes have progressed in quality of care evaluation, and cardiac surgery leads the way. Measuring the risk of an intervention allows prediction of the outcome and is essential for quality monitoring: without knowing the predicted outcome, the actual outcome cannot be evaluated. Cardiac surgery risk models like EuroSCORE and others have been adopted worldwide, so that measuring risk-adjusted performance is now an integral part of the delivery of good cardiac surgical care. When mortality for a procedure is higher for one surgeon (or hospital) than another, this can be due to one of three reasons, or a combination of the three: the difference is due to chance, or variation in risk profile, or better and safer service. We now have the tools to distinguish between the above factors. We can also observe performance over time: cusum curves plot the performance of surgeons and hospitals by showing hypothetical ‘lives saved’. This provides early warning of deterioration in performance before a problem reaches statistical significance. With the appropriate tools, it is possible not only to identify a problem, but also to anticipate and thus prevent a problem from happening. Monitoring clinical performance is an exciting and rewarding field for future development, and one that will yield real improvements in patient outcomes.