We must say how disappointed we were on reading the paper ‘Can we trust studies using audit software?’ by Rice et al.1 We certainly hope it does not have an adverse effect on the use of the GRASP tools in the future, as our experience with them has been very positive.
First, one must question how representative this experience of one practice is.
Second, much of the difference between the GRASP-AF results and the ‘manual’ search used in the paper (which actually was an electronic audit too as they used the practice computer system to do this) is clearly due to a difference in the definition of AF used in the two methods and in the high use of free text in this practice to record exclusions.
The manual search described looked for patients with ‘unresolved’ AF, which allows GPs to exclude some high-risk patients from the audit, whereas GRASP-AF searches for a ‘history of AF’. Therefore, the GRASP search includes patients that the QOF would exclude under the category of ‘resolved’. As we know, no audit software can search free-text entries, hence the need to ensure accurate coding is used in the first place. The use of free text rather than predefined Read codes to record patient refusal or contraindication explains why the software could not identify patients. This highlights issues about appropriate use of coding in this practice rather than any inherent problem with the audit software.
One important feature of the GRASP-AF audit software was omitted from this paper: that GRASP offers the opportunity to easily share anonymised data for benchmarking using CHART Online. We feel the article missed an important opportunity to show that the issue is not about comparing automated versus manual processes, but rather how best to make use of our most precious resource — clinical time. The paper’s proposal to return to manual audit will exasperate already busy GPs. Clearly the only sensible option for the future is to use audit software and ensure that coding is improved so that the results are as accurate as possible.
- © British Journal of General Practice 2016
REFERENCE
- 1.↵