This site is intended for health professionals only


QOF: Exception reporting

30 May 2013

Share this article

 

There is no doubt that the quality and outcomes framework (QOF) is going to be a greater challenge to practices this year than it ever has been. Increases to thresholds and the introduction of a significant number of new indicators will increase workload if practices want to continue to achieve points. One hundred points have been removed bringing the risk of reduced income.

With more work for the same or even reduced returns, it has become more important that practices tackle the myriad of tests, reviews and prescribing required in the most efficient way possible. Timing can be crucial in ensuring that the work done and codes entered are recognised at the end of the QOF year.

It is not only practices that will have more to do. As the number of indicators have risen there is also more that is required from patients. Nobody enjoys having a blood test, and getting to an appointment can require significant organisation for patients. More frequent reviews, tests and questionnaires will increase the demands on patient time and attendance. Many patients will be listed on several different registers, each with their own sets of requirements. If patients need to make fewer visits to the surgery they are more likely to turn up to them all, leading to better health for them and higher levels of achievement for the practice. Careful scheduling can help here.

Twenty areas have had raised thresholds this year. In England these are mostly well over 95%, and a couple are 100%. The rest of the indicators will follow next year. To get all of the points will require either achieving the indicator or exception reporting every qualifying patient. 

In many cases exception reporting can require more effort than achieving the indicator would – for instance, to send three letters. In any case it can take time to show that all of the appropriate steps have been taken. This should be integrated into how patients are called for tests or reviews.

Changes to the timings of indicators this year will change the most effective way to call patients for review. Previously practices could start to work on a year’s QOF achievement 15 months in advance, which is from the first of January. This has been reduced to 12 months in most cases – from the previous April. In effect there will be a sharp division between QOF years on the stroke of midnight of March 31 each year. From the point of view of collecting points, it has become troublesome to recall patients for review around this date.

If a patient is due for review at the start of April and books their appointment slightly early, then a further review would be needed within the QOF year. The same is true for reviews due at the end of March. In previous years it would not matter too much if this was a little late as they would likely have been covered by the previous year’s review.

So long as you don’t start too early, the first step should be to define what work will need to be done by the practice during the year. In some cases indicators are dependent on the results of other indicators. Knowing which patients are eligible for the dependant indicator will leave more time to achieve the target.

One example of this is the atrial fibrillation area. The first indicator is for calculating a ‘congestive heart failure, hypertension, age and diabetes’ (CHADS2) score for each patient with atrial fibrillation. The other two indicators are for the prescription of either aspirin or anticoagulation, dependant on the score.

Calculating the score earlier in the year will allow practices to look at these groups and start prescribing or commencing anticoagulation. Although prescriptions are only counted for QOF after October 1, commencing blood thinning agents as early as possible will make that easier to achieve.

There is a three-month period of automatic exception reporting following a new diagnosis of atrial fibrillation or a new registration with the practice, but there is no similar grace period after a new score. If a patient has a new higher score calculated at the end of the year, then the practice may only have a couple of weeks to initiate anticoagulation. As age is a component of the score, then it will rise in some patients during the year.

Other indicators which are dependent on others include the prescribing of angiotensin-converting-enzyme (ACE) inhibitors in diabetic patients with microalbuminuria, or the giving of smoking cessation advice. Similarly, new diagnoses will need specific reviews.

It is a fundamental principle of the QOF rules that, when it comes to measured values, only the most recent measurement counts towards the final calculation. This makes having plenty of time useful. The optimal mechanism to gain points (although clinical considerations should obviously take precedence) is to continually optimise treatment and test until the target is reached and then to stop testing.

Measuring HbA1c for blood glucose early in the year could be one example of this. If the first reading is less than 54mmol/L then there is no reason under QOF to test for the rest of the year, and in a stable patient this could be clinically quite appropriate also. Indeed there a disincentive to measure again as it is only possible to lose points and not to gain any further. Where a patient has a high blood sugar, the treatment should be optimised and further testing performed after this has time to work. There is no limit to the number of times that a test can be performed, only the last result is counted. However there are probably some limits to what the lab, and the patient, will allow.

Even where several tests through the year are expected, only one will count. It is advised that patients taking lithium have the levels measured every three months. To allow for this, only results recorded after December 1 are valid for QOF. Often there are quite small numbers of patients and if one patient on the register were to miss the target then the financial loss could be relatively high.

These blood levels do vary – there would be little point in testing otherwise – and patients will occasionally miss the target range. It is most advantageous to practices to take tests during December to allow some time to modify drug doses and measure once again. In the patients that did get into the correct range it would be best not to test again until April.

The other indicator with a shorter timescale than most is for blood pressure measurement in hypertension. In this case, only readings taken after the July 1 will count towards the end-of-year calculation. The blood pressure area has a higher prevalence than any other area and so this will affect a lot of patients.

Several other areas, such as diabetes, stroke or heart disease require a measurement of blood pressure after the start of April. It would be most efficient to schedule reviews for those patients not on the hypertension register in the first three months of the year and then call the others later.

Exception reporting will be more important than ever as the upper thresholds are rising. It will no longer be enough to fail to record when a patient does not make an appointment after an invitation. In previous years, practices have tended to only exception report to the minimum number required rather than coding each and every patient. Exception reporting has also tended to happen at the end of year when it was clear how many patients would be needed.

Having an efficient system to send invitations and record responses will be essential to getting all of the points, and practices will need to be much more proactive from now on.

It makes most sense to record patients who are unsuitable for intervention at the start of the QOF year. This does not remove the incentive to achieve targets in these patients but will allow the practice to see which patients are suitable to be called in.

The regulations specify that there should either be an explicit rejection of review or, as is more commonly the case, there should be no response to at least three invitations. The guidelines do not specify the form of the invitation. Most often this will be by letter but could equally be by electronic means. Verbal invitations are also valid, although they will need to be documented. Letters are still the most common method but all these principles are the same whatever the method of invitation.

Sending the first letter is relatively easy, although there is a balance to be had on the timing of invitations. Many patients will come in during the normal course of reviews – the expense of sending letters for review may not be justified. However there needs to be enough time to potentially send out three letters. It may be optimal to send out invitations to patients at risk around the time that the influenza vaccinations arrive.

There are some software packages and practice systems which can assist in targeting these recalls at patients. The combination of indicators which apply to each patient will depend on their diagnoses and, in some cases, their age. Making sure that patients are called in for each of their checks could potentially be complicated.

Fortunately many of the indicators are similar across several areas. Blood pressure checks along with blood tests for cholesterol and renal function are fairly standard, but other tests such high-density lipoprotein (HDL) measurement are more restricted to particular areas. Urine samples are needed from patients with diabetes to measure the albumin/creatinine ratio. 

It may be worthwhile grouping patients before sending out letters. Patients with diabetes are likely to require a full range of checks. Once this group has been identified, a second group of patients with coronary heart disease, stroke or peripheral arterial disease could be called for blood pressure and cholesterol measurement. A final group may include patients who do not need any laboratory investigations at all, such as those with only chronic obstructive pulmonary disease (COPD), asthma or younger patients with mental health problems.

Much of this can be dealt with at the time of the review with alert and adaptive people doing the review. There has to be a system to look up what would be required for each patient.

Even more complicated is keeping track of which invitations have been sent, or given verbally to each patient. There are no official read codes specifically for these invitations being sent. The sending of each invitation should be recorded as well as the check for which the patient has been invited. Local codes can be used if your computer system supports them.

Where there are three recorded invitations in any area then the patient can be recorded as informed dissent. There will need to be a different code for each disease area.

While there are a slew of new clinical indicators this year, it will be practice organisation as much as delivery of clinical services that will determine success in the coming year. Early and systematic scheduling of chronic disease monitoring will benefit practice and patients alike.