In a potentially controversial area, reporting accuracy is of central importance. It is therefore important to alert you to what may be a misinterpretation of our recent research in your editorial on revalidation.
Biased surveys will wreck revalidation
This was a study of 1,065 doctors, not just GPs. Patients did not ‘rate their GP more harshly if they were of Asian origin'. On the contrary, doctors of Asian origin performed just as well as white doctors in patient feedback after adjusting for other factors we considered. And patients did not adversely rate locums, although it was certainly true that doctors who had a higher proportion of patients seeing their ‘usual doctor' had higher scores than otherwise – perhaps highlighting the importance of continuity of care.
It is important to appreciate the true balance of these findings. Overall, we were reassured that doctors' age, gender and ethnic background were of less importance in influencing patient and colleague feedback than some other variables, such as the region of the world a doctor had obtained their medical degree, or their specialty. We reported that psychiatrists might face particular challenges on patient feedback.
It is true we have advised caution in how multi-source feedback data is collected and managed, but our results support rather than undermine judicious use of such approaches. They highlight the importance of taking account of the context care is delivered in. We believe our results support the GMC's view of patient and colleague feedback as having potential in identifying areas at which doctors may wish to target professional development.
From Professor John Campbell
Professor of general practice and primary care, Peninsula Medical School
Editor's note
I'm grateful to Professor Campbell for the clarification. In ascribing findings to GPs, rather than doctors generally, and to locums, rather than ‘not the usual doctor', I was using shorthands that did not properly describe the results.
The study concludes ‘doctors who had trained in South Asia were likely to score lower on patient feedback' and that ‘colleagues gave doctors of Asian origin lower scores'. The editorial should have referred to Asian training, rather than Asian origin, when it discussed patient feedback, although it did include the caveat: ‘The researchers point out that, overall, race was not an independent predictor of nature of feedback.'
Professor Campbell rightly highlights the importance of describing data precisely. Pulse, however, remains of the view that surveys showing evidence of ‘systematic bias' should not currently be used in revalidation.
