The British Medical Association today complained that family doctors stand to lose thousands of pounds of revenue as a result of a flawed survey.

The GP Patient survey results carried out for the Department of Health show that patients are, in general, happy with their GPs and with the arrangements for getting appointments. But, the BMA charged, some practices will lose money as a result of the responses of a handful of patients.
 
The report and technical annexes suggest the BMA has a point. The overall response rate to the postal survey was 38 per cent (5,660,217 sent out, 2,163,456 returned) but in many areas it was much lower – 30 per cent in London, for example.
 
There were also huge variations between practices. In 32 practices, less than ten per cent of patients who were sent a form actually completed and returned it. The response rate was below 30 per cent in 1,345 practices – 16 per cent of the total number of practices in England.
 
Interestingly - given that no explanation is provided for it - response rates are falling year by year. In 2006-07, only half as many practices (688) recorded response rates below 30 per cent as in 2008-09. And in 2006-07, 703 practices achieved response rates of 60-69 per cent; this year, only 60 did.
 
The survey, carried out by Ipsos-Mori, cost £8 million in 2008-09, according to an answer given in Parliament by Ben Bradshaw in response to a question by Andrew Selous MP. This was a saving of 20 per cent on the previous year, the minister boasted. Maybe that’s why the response rate fell.
 
The LibDems says the survey is costly and unrepresentative and should be scrapped. The BMA points to one practice in rural Lancashire with 14,000 patients that stands to lose £10,000 in income on the basis of questionnaires returned by less than 1 per cent of its patients. Hardly sounds like evidence-based medicine, does it?