|Home | About | Journals | Submit | Contact Us | Français|
A new study shows that the introduction of publicly available performance tables in the United Kingdom showing mortality after major cardiac surgery by individual surgeons did not result in fewer procedures being performed on high risk patients, as critics had predicted. The study also shows an association between the introduction of the tables and a decrease in mortality.
The study, published online in Heart (http://heart.bmj.com, doi: 10.1136/hrt.2006.106393), analysed data that were collected prospectively from all NHS centres in northwest England that undertake cardiac surgery. The data covered 25730 patients undergoing coronary artery bypass grafting for the first time between April 1997 and March 2005. Figures were for 30 different surgeons in four major NHS sites.
The researchers compared surgery carried out before and after individual cardiac surgeons' outcomes became public in 2001, to determine whether some surgeons had become more averse to risk, operating only on patients with a lower risk of complications or death. They also assessed the effect of the introduction of the tables on patients' mortality.
They used the EuroSCORE risk scoring system for patients undergoing cardiac surgery to divide them into low risk (EuroSCORE 0-5), high risk (6-10), and very high risk (>11) patients. Analysis of data before and after public disclosure of surgeons' performance showed that the number of high risk patients undergoing cardiac surgery rose rather than fell—from 449 (14% of all patients who underwent surgery) before public disclosure to 547 (17%) afterwards (P<0.001). The number of patients at very high risk who underwent surgery also rose slightly, from 41 (1.3%) to 47 (1.4%). The proportion of patients aged over 80 and of those with kidney disease, a recent heart attack, or peripheral vascular disease all increased significantly. In contrast, the number of patients at low risk who underwent surgery fell slightly, from 2694 (85%) before disclosure to 2654 (82%) afterwards.
Observed mortality fell from 2.4% in 1997-8, before disclosure, to 1.8% in 2004-5 (P=0.014)—even though the expected mortality (based on EuroSCORE) rose from 3.0 to 3.5 (P<0.001), indicating that more complicated cases or more elderly people were being taken on. Overall, the ratio of observed to expected mortality decreased from 0.8 to 0.5 (P<0.05).
Ben Bridgewater, a consultant cardiac surgeon at the University Hospital of South Manchester NHS Foundation Trust and lead author of the study, said: “Making cardiac surgeons' results publicly available has been controversial, and most clinicians have not embraced the concept readily. But if you look at the data there is quite strong evidence that collecting, analysing, and feeding back outcome data to clinicians improves quality. We found a quite marked improvement in mortality, despite an increase in predicted risk.”
There was no evidence that high risk patients had been denied surgery, Mr Bridgewater added. “In fact, the number of high risk patients having surgery increased.”
Because of the positive effects of public disclosure of results shown in the study, he suggested that such performance tables should be applied in other areas of medicine. “Given that the downside of disclosure is small and the upside is big, the results of the study should encourage other clinical groups to take this forward, rather than being driven by politicians or the media.”
Cardiac surgery is currently the only area of medicine in the UK for which data on individual surgeons are disclosed. This followed a public inquiry into the deaths of children undergoing cardiac surgery at Bristol Royal Infirmary in 2001, which recommended making individual heart surgeons' performance public (BMJ 2001;323:181, doi: 10.1136/bmj.323.7306.181).
Julian Le Grand, professor of social policy at the London School of Economics and former adviser on health policy to the prime minister, agreed that public disclosure of outcomes should be extended to other areas of medicine.
But he argued that the aim must be to give useful information back to doctors and hospitals rather than to the government as part of setting targets or managing performance. “The dangers of publicising this sort of information —that doctors may fudge their figures or take less risky cases—have not been seen in this study. The evidence shows that, on the whole, this type of scheme does more good than harm.”
Michael Summers, vice chairman of the Patients' Association, said that his organisation would also welcome more open reporting of outcomes, particularly in general surgery. “The figures are useful for GPs and patients, as well as the doctors receiving feedback on their performance—particularly as patients can now choose from up to four hospitals for procedures.”