A new version of the GRASP-AF audit tool has been released today, which incorporates important updates relating to the CHADS2 and CHA2DS2-VASc algorithms.
We are pleased to announce that an instructional guide (workbook) to accompany the audit tool has also been released, supporting you in the analysis and interpretation of practice data.
The inclusion criteria for the algorithms have been altered. A review of the original research papers has revealed that whilst the 'S' represents stroke, there is a degree of ambiguity as to whether this includes haemorrhagic stroke and subsequent validation studies specifically exclude it. NHSIQ and PRIMIS would like to point out that patients with haemorrhagic strokes should now be excluded from the Stroke element of the CHADS2 and CHA2DS2-VASc algorithms.
The implication is that for practices using previous versions of the tool, a small number of patients with haemorrhagic stroke may have had an inflated CHADS2 and CHA2DS2-VASc score and possibly flagged for a review for consideration for anticoagulation therapy. However, please note that the audit tool has always identified patients with a Read code of haemorrhagic stroke to ensure practices are signposted to individuals for who anticoagulation might not be appropriate (codes listed in a separate column in the datasheet and highlighted in magenta).
Furthermore, our research has uncovered a significant data quality problem relating to the coding of stroke. Within the CHART Online database, over a third of patients with atrial fibrillation, who also have a recording of stroke, have been coded using an unspecified stroke code such as 'stroke NOS'. Practices using the tool need to be aware that patients who have had a haemorrhagic stroke may continue to be included in the algorithms, as without a specific Read coded entry, they cannot be excluded. Practices are advised to enter a Read code onto the patient's record that represents the level of detail known to them. The PRIMIS Data Quality Indicators audit tool will help practices to identify patients who have a recording of 'stroke NOS' but no other Read coded entries to indicate the type of stroke.
To clarify, only patients with a coded entry for ischaemic stroke or stroke unspecified are included in the CHADS2 algorithm. For CHA2DS2-VASc, only patients with a coded entry for ischaemic stroke, stroke unspecified or thromboembolism (as outlined in reference 1) are included in the algorithm.
1Am J Med. 2010 Jun;123(6):484-8. doi: 10.1016/j.amjmed.2009.12.013. Improving stroke risk stratification in atrial fibrillation. Lip GY, Halperin JL.
The new version of the GRASP-AF audit tool can be obtained by clicking on CHART Updates within the CHART tool itself.
The new workbook can be downloaded via a link on the CHART home screen or from the GRASP-AF page on this website.
Please note: from 12th March 2014 data uploads to CHART Online will only be accepted from thus new version of the audit tool.
GRASP-AF is one of three audit tools that comprise the GRASP suite of tools, delivered in partnership with NHS Improving Quality. Information on GRASP-AF, including documentation, can be found on the GRASP-AF page.
If you, your practice or your CCG would like further information on the GRASP suite of tools, please contact enquiries@nhsiq.nhs.uk
Applied Health Research Building University Park University of Nottingham Nottingham, NG7 2RD