1. Background
One of the items that CQC will review with practices on inspections is the QOF exception rates. The data for these is available from NHS Digital where you can search by practice and look at historical QOF data, including exception rates.
Practices have recently contacted our support desk following a CQC inspection asking why their exception rates are so high, particularly for cancer clinical indicators. In the example above the practice had not recorded any exceptions for cancer patients in that year so how does that equate to an exception rate of 26.5%?
Practices can also run QOF achievement reports in CQRS and extract exception / exclusion data.
Again, in the example above, from the same practice, you can see that it appears that there were no exceptions recorded that year yet the exception rate is showing as 26.52%
2. Exclusion that are Exceptions?
Having communicated with NHS Digital it transpires that there is a misconception of what are exceptions and what are exclusions. Exclusions are normally for reasons that are beyond the control of the practice whereas an exception is a result of something that has been recorded by the practice. Exception rates are used as a quality measure of the practice.
Many indicators have business rules to identify either newly registered or newly diagnosed patients.
Example - CAN003: Patients diagnosed in the 6 months before the 'Payment Period End Date' (PPED)
In the CQRS data screen shot above you will see there were 35 patients diagnosed with cancer in the 6 months leading up to the end of the QOF year. These patients would have been filtered out by the QOF business rules and would not have had an alert in the QOF alert box. At the end of the QOF year, if they have not had a review, these patients are automatically excepted from the Cancer QOF targets and are used in the practice exception rate calculation and this is why you may end up with a high exception rate without realising.
In addition, the 2019/20 QOF business rules include new 'Personalised Care Adjustments'. These have added automatic exceptions where a patient has had two invites, more than 7 days apart, coded with the correct QOF invite codes and have not had a review coded. According to NHS digital these patients will also count as ‘excepted’ so potentially practices could end up with very high exception rates this year.
NHS Digital’s logic in both cases above, either when a patient is diagnosed towards the end of the QOF year or has two invites, is that these are patients that could be reviewed and the necessary work completed. Therefore, if you don't, then they are classed as exceptions.
3. How to avoid this situation
3.a Excluded patients
EMIS mix together ‘Exclusions’ and ‘Exceptions’ which makes it very difficult to identify patients who are going to count towards your exception rates. We are developing searches which will help to make this clear for you. Please note - Any QOF work completed over-rides an exception (whether that be exception due to addition of a code, due to 2 invites or due to recent diagnosis or registration) and does count towards end of year achievements.
3.b Two QOF invites and no review
Patients who have had 2 QOF invites coded more than 7 days apart and still have outstanding QOF work will be exception from any domain areas which are outstanding.
3.c Monitoring
A process should be in place to regularly monitor the 'Excluded Patients'. Any QOF work completed overrides an exclusion or exception and does count towards the practice end of year achievements.
4. Exceptions and Prevalence
Your chronic disease prevalence for each QOF Domain has a direct bearing on the QOF point value for that domain. A larger register size directly results in more £ per QOF point achieved.
Prevalence, however, is a tricky area to understand and one of the most frequently asked questions is whether prevalence is taken before or after exception reporting. Prevalence is, in fact, based on the entire register size regardless of the number of Exceptions. Exceptions (or exclusions) only apply to the ‘denominators’ for each indicator.
This data may be useful if you like figures: https://digital.nhs.uk/data-and-information/publications/statistical/quality-and-outcomes-framework-achievement-prevalence-and-exceptions-data/2018-19-pas. It provides spreadsheets with raw data prevalence and exception rates. The same data is made into a more digestible format by Dr Gavin Jamie who run the GPContract website (www.gpcontract.co.uk). This shows that e.g. for AF:
- A practice had a list size 2017-18 of 18561, the AF register was 298, the AF prevalence was 1.61.
- The same practice had a list size 2018-19 of 18959, AF register was 316 and AF prevalence was 1.67
It then goes on to provide details of the individual domains within AF and does not mention prevalence again, just exceptions and denominators. Thus we can be certain that prevalence refers to the number on the register. We know that there are no exceptions which remove patients from the entire register as it just looks for patients with AF:
Prevalence is not, therefore, affected by any exceptions or exclusions whether that be manual addition of a 'personalised care adjustment code' or whether it be due to a patient having had 2 recall letters.
If you require any further assistance on the process above, please contact Ardens support on: [email protected]