UPDATE on discrepancies noticed in CVDPREVENT data
A number of CVDPREVENT users have flagged that they have noticed that CVDPREVENT data does not completely align with other sources. In particular, discrepancies have been noticed when comparing CVDPREVENT with Quality and Outcomes Framework (QOF) data.
Some of the discrepancy can be attributed to expected differences. CVDPREVENT is designed to audit CVD prevention activity across England, whilst QOF is an incentivisation scheme. You can read about some of the expected differences in the section below.
In some cases where a patient has moved practice, activity (including diagnoses or blood tests) that took place at their previous GP practice has not been picked up in the audit data collection. Excluding this information was not an intended part of the design of the audit. For a complete dataset, information from patients’ former practices do need to be included.
The CVDPREVENT team are currently working with partners on a resolution to ensure the audit collects all the data that is required for the reporting of CVDPREVENT indicators. The intention is that a resolution should be in place for the reporting of data up to June 30th 2024 (scheduled for release in October 2024).
We will notify our users with any updates as soon as we have them and apologise for any inconvenience caused.
If you have any questions or concerns, please contact the CVDPREVENT Support Team at nhsbn.cvdprevent@nhs.net
Current underreporting in CVDPREVENT
The set of circumstances described above is contributing to underreporting in CVDPREVENT, particularly for indicators of diagnosed prevalence, where an external denominator is used. There may also be an impact on treatment measures which use the prevalent population as a denominator, although at national level differences in proportions for these indicators with QOF are minor. It is also likely that case finder indicators contain some individuals whose diagnosis for the relevant condition has not been extracted.
Expected differences between CVDPREVENT and QOF
Many indicators in CVDPREVENT are intended to align with indicators in QOF, to provide in-year data updates and additional information on health inequalities. However, although similar, small differences between QOF and CVDPREVENT are anticipated.
CVDPREVENT extracts patient records from GP systems which is then analysed centrally, whereas QOF indicators are reported in aggregate form to NHS England having been analysed already. As CVDPREVENT is extracted differently, pragmatic decisions have been taken to align as closely as possible with QOF within the limitations of data collection. In some cases, the benefit of exact alignment with QOF may be outweighed by the amount of data that needs to be collected to make this happen.
Furthermore, CVDPREVENT has 98% coverage of practices, but patients can choose to opt out of their data being submitted to the audit, while it continues to be included in aggregate counts for QOF.
More information about the differences between CVDPREVENT and QOF can be found in the CVDPREVENT methodology annex, pages 5 – 6.