Martin Roland et al1 raise a number of concerns about our review,2 perhaps the most serious of which is that we ‘missed’ four papers.3–5 We did consider these papers but found no data relevant to the reliability and validity of the GPAQ. Including them would not have changed our conclusions, as a citation demonstrating that a questionnaire has simply been used does not endorse its validity.
While we did not discuss face and content validity (space did not permit), these are the absolute minimum one might expect of a questionnaire. Our concern that the GPAQ has not been validated against an external criterion is not unique. It was a concern originally voiced by the GPAQ development team but unfortunately was never subsequently addressed.
Finally, Roland et al state that a group of ‘independent academic advisors’ recommended that the GPAQ be used in the GP contract. This is an appeal to authority rather than a challenge to our conclusions. It was the same group who also recommended the IPQ, and we note that our conclusions about the reliability and validity of this questionnaire have not been challenged.
We welcome the commitment of Roland et al to further research and development of the GPAQ. But our fundamental point remains that this should have taken place before national adoption of the GPAQ and the IPQ by the Quality and Outcomes Framework.
- © British Journal of General Practice, 2007.
In this issue
