This site is intended for health professionals only

Avoiding pitfalls in QOF exception coding


Reviewing your QOF exception reporting is a complex but necessary chore at this time in the practice year. But with the rules constantly changing, ensuring that you accurately code – and are not caught out by PCT inspectors – can be difficult.

This article provides an up-to-date guide on how to maximise your QOF points and avoid the most common pitfalls in exception reporting.

It is correct up until version 18 of the QOF rules (published in December 2010). Five byte codes have been used for illustration, but the rules are the same whichever version of Read coding is used.

Missing expired codes

There is nothing more frustrating than doing QOF work and then not having it recognised, and the risk of this happening increases if expired exception codes are not being actively followed up.

Expiring codes generally apply to things such as informed dissent or where a patient is declared as unsuitable by the practice and – depending on the time they are entered – often apply only for the current QOF year.

Non-expiring codes apply to areas such as the recording of allergies that do not vary from year to year. Dr Simon Clay has an excellent list of the different types of codes that fits on an A4 sheet and every practice should have one available for easy reference [1].

As an example of how you can be caught out, the mental health register includes all patients who have ever had a psychotic illness, but often patients have been recovered for years – we have had a couple of patients who appear due to puerperal psychosis in the 1970s. In these cases, annual reviews and plans are inappropriate and the unsuitable patients can be coded with 9h91. But this is an expiring code (there is no non-expiring code) and needs to be entered every year. Having done this once, it is fairly easy to search for patients with the code to apply it again the following year.

Doing coding twice

Expiring exception codes last for 15 months, and so those entered in January, February or March will count for two consecutive QOF years, whereas codes entered in December will expire just before the final reckoning in April. In most practices, exception reporting tends to be something that happens at the end of the year, but this means your codes will be valid for only one year and you risk needlessly duplicating the work the following year. If your clinical system will allow you to ‘look forward' and run the QOF as if it were April, use this to avoid duplicating any work. Otherwise, a manual search for codes expiring in December will do the same job.

Using old codes

The rules that govern the use of Read codes change a couple of times a year, making it almost impossible for practices to keep up. The latest version (18) came out in the first week of December 2010, and these rules may not reach your system until February or March this year – leaving little time for practices to make the required changes.

For instance, although previously practices could exception report patients from the diabetes indicators where there was no local service with the code 9h43 (service unavailable), from version 18 this will no longer apply. Most practices will not be using this code, but those that do could be in for a shock when the rules are implemented.

One solution to this is to ensure that your practice manager keeps up to date with all the changes as soon as they come out. I cover most of the new developments in version 18 here, and you can check for regular updates on my website [2].

Local data-quality staff at the PCT can also be helpful at keeping you up to date, and of course there is often relevant information in Pulse [3].

Using too many codes

There are some cases where recording two separate exception codes is unnecessary. For example, in the latest QOF rules the codes 8IAn and 8Iao (codes for diabetic patients with declining neuropathy and foot pulse assessments respectively) are now both valid exception codes. Although they differ in meaning, they both exclude from both examinations.

Also new in version 18 of the rules is an extra exception code for cervical cytology – 9NiT (DNA cervical cytology). Like all exception codes for smears, it lasts for a full five years. It is unlikely a single missed appointment will exclude that patient from the national recall programme, but it will be worth coding these DNAs for QOF purposes.

Recording drug intolerances inaccurately

Both QOF indicators CHD9 and Stroke12 provide points for anticoagulation treatment, yet obviously there are times where anticoagulation is inadvisable due to high risk of bleeding.

There is no single code in this situation. Three codes must be entered to indicate that neither aspirin nor warfarin nor clopidogrel are indicated. If all three are not present, then the patient will not be excluded from the indicator.

In a similar way, CHD11, DM15 and CKD5 concern treatment with ACE inhibitors or angiotensin receptor blockers. Patients may be intolerant of both of these drug classes or renal disease may preclude their use. In this case, again both codes must be present to exclude the patient. This applies both to expiring and non-expiring codes but you can mix them – for example, a patient can be allergic to one and refuse the other.

Inadequate record keeping

Exception reporting is a controversial area, with pressure on primary care organisations to monitor and investigate practices they suspect of being over-eager to exclude patients from treatment. Yet there are increasing reasons for exception reporting.

Nationally, NICE guidance is at best ambivalent about treating cholesterol to target and the desirability of pushing HbA1c below 7% is increasingly questioned. Local guidance may also lead to differences – my PCT has recently declared that not all patients newly diagnosed with stable angina need exercise testing, effectively nullifying CHD 2.

This is why it is vital that individual exceptions can be justified – putting a reason in free text alongside the exception code can make things much easier later, when the exception examiners question a code entered 14 months ago.

Creating a template for exception coding – with a prompt to enter the reasoning behind the decision – can encourage the use of correct codes throughout the practice, but equally important is staff training to ensure that code are used correctly.

Not utilising the system's flexibility

The rules around who should have a depression assessment are complex, but you can make them work to your advantage.

Many patients don't come back for the second questionnaire, and in many cases this may be because they no longer have significant problems. The rules take account of this and if you can code the depression as resolved (212S) then no further assessment is necessary. This does not have to be coded in the 12 weeks following diagnosis – it can be at any point in the QOF year, and will save you a lot of fruitless time and effort.

Likewise, screening for depression is not always necessary in patients with chronic diseases, and the QOF rules recognise this.

QOF mandates an annual two-question depression screen in patients with diabetes or heart disease, but those patients with a new diagnosis in that QOF year are automatically excluded.

Also, those you are already treating for depression can be specifically excluded as unsuitable for screening with the code 9hC0.

For more tips on QOF coding read Dr Simon Clays series of articles here

Dr Gavin Jamie is a GP in Swindon and runs the QOF Database website


QOF exception reporting