Patient surveys 'are unreliable'

GP efforts to improve services in line with patient survey findings may backfire because the accuracy of the data is 'questionable', according to UK statisticians.

Last month, health minister Lord Warner said patient surveys would be used to measure improved care under practice-based commissioning.

Survey data are to be used this year to measure access and patient choice.

GP practices are already expected to develop two-year action plans based on survey outcomes under new GMS, which awards practices up to £20,000 a year based on patient opinion (GP, 24 February).

Speaking at the annual scientific meeting of the Society for Academic Primary Care, held in Staffordshire, last week, Matthew Hankins, a statistician at Brighton and Sussex Medical School, questioned the validity of both the IPQ and the GPAQ surveys used in GMS.

'We can't find any evidence that these questionnaires have been validated against any external criterion. They showed consistency but that doesn't bear on the issue of validity,' he said.

Although estimates for their reliability exist, Mr Hankins said 'whether they're adequate or not is open for discussion'.

Moreover, existing peer-reviewed evidence for the surveys' validity and reliability relates to earlier versions, not those included in the contract, he said.

However, GPC deputy chairman Dr Laurence Buckman branded claims the surveys were not validated 'outrageous'. He said: 'That's a very unfair slur and it's very untrue. They were validated on enormous numbers of UK patients.'

Dr David Jenner, Devon GP and clinical director of Client-Focused Evaluation Programme UK Surveys, agreed that the questionnaires probably did not meet the stringent validity standards some academics would like.

Nevertheless he was 'confident that the IPQ is reliable, consistent and valid'.

Professor Martin Roland, director of the National Primary Care Research and Development Centre at the University of Manchester, who was involved with the development of GPAQ, said: 'It has as much validity as any other available instrument in the UK and has a significant number of papers published in support of its validity.'

Mr Hankins also questioned whether the new patient survey that the DoH plans to introduce at the beginning of next year to evaluate access and choice would provide data of sufficient quality for reliable conclusions to be drawn.

Five draft questions have already been formed for this survey with a yes/no format in mind.

Mr Hankins said: 'It wouldn't be unreasonable to interview patients and use a multimodal approach.'


  • Statisticians have questioned the validity of IPQ and GPAQ.
  • They suggest that an assessment system that includes interviews might be better.
  • Patient surveys count for 75 quality framework points.
  • A patient survey to evaluate access and choice is being developed by the GPC and NHS Employers.

Have you registered with us yet?

Register now to enjoy more articles and free email bulletins


Already registered?

Sign in

Follow Us: