Professor Helen Lester's recent research on US clinical incentives suggest removing them QOF indicators could have mean a decline in clinical care. But Dr Terry McCormack argues that care can only start to improve when some indicators are culled.
The difficulty with addressing this question is that this is an ‘evidence light' area, which is surprising when you consider how many countries now use pay for performance within their health care systems, and how much money the UK in particular spends on QOF. QOF currently includes 134 indicators, worth 1000 points, 697 of which are focused on clinical areas. Although indicators could remain indefinitely within QOF, this would restrict the potential benefits of financial incentives to a limited number of clinical conditions. This would be a shame because we do know, from data collected in English practices and published in the NEJM that QOF effectively ‘kick started' care above secular trend for people with asthma and diabetes, but only for a year or so. We also know that achievement levels in some indicators have reached a ceiling with no real possibility of further improvement. So we are left with a dilemma- should we remove indicators once certain statistical criteria are met (high achievement, low exception reporting, little variation between practices and consistently high performance over time) or do we leave them in, knowing that the incentive is essentially paying for more of the same? At heart, this dilemma has been created by a lack of agreement over the purpose of QOF- is it a quality improvement mechanism or a payment scheme for primary care?
Evidence from Kaiser Permanente in California suggests that removing financially incentivised indicators may lead to a decline in performance levels. Part of the reason for working with Kaiser was because there was no equivalent data set for QOF. We found four ‘shared' indicators (yearly assessment of the level of glycaemic control, screening for diabetic retinopathy, control of hypertension and cervical screening) and looked at performance as financial incentives were attached and removed over time. Across the 35 Kaiser facilities, the removal of incentives was associated with a decrease in performance of about 3% per year on average for screening for diabetic retinopathy and about 1.6% per year for cervical screening. However there are problems with trying to make direct comparisons with QOF. For a start, the payments were far smaller and crucially they were to the organization not directly to the doctors.
So what does this all mean? Personally I think that indicators should be removed once we can see that they have effectively ‘run their course'- that is, that they cannot possibly improve patient care any further. Indeed the GPC have already agreed to remove up to eight clinical indicators worth 28 QOF points from April 2011. These indicators are all process measures and reward actions such as taking blood pressure or measuring blood cholesterol or glucose. I appreciate that a strong counter argument is that practices will still be expected to provide that care- and who pays? But I hope that after 7 years, these actions are now so embedded in the fabric of primary care, and are probably largely performed by non medical staff, that the public money spent on rewarding such actions can be more usefully focused on new areas of health care.
However removing indicators requires carefully thought through safety nets. Policy makers may need to consider a stepwise reduction of payments for indicators with a significant number of points. Patients also know about and care which clinical areas are in QOF. Between 2005-7, over 500 ideas for new QOF areas were submitted to the Department of Health, 25% of which came from patient groups. If a clinical domain has indicators removed, then patients need to understand why. Above all, a system of monitoring achievement in areas where indicators have been removed needs to be in place and negotiators need to decide, a priori, the level of decline that triggers a review and possible reintroduction.
Professor Helen Lester is a GP in Birmingham and professor of primary care at the National Primary Care Research and Development Centre.
Quite the opposite, culling some carefully selected QOF indicators will improve patient care by opening the door for new indicators. We all have our pet hates about QOF. Pointless box ticking being one of them and some indicators seem to me to be just that. We all want to do useful and meaningful work. A good example is the atrial fibrillation indicator which encourages us to prescribe an antiplatelet or an anticoagulant. The relative reduction in stroke risk when using aspirin is 22% whilst the relative reduction in stroke risk when using warfarin is a massive 64%.1 Which would you want to take? The current indicator encourages us take the simple route of using aspirin and should be changed to prompt us into using the CHADS2 scoring system and then to initiate warfarin when indicated. That would improve patient care.
As a general principle QOF should be about incentivising us to initiate good care rather than just mindlessly checking on compliance or endlessly nagging patients. So I would favour reducing some of the smoking and epilepsy indicators in favour of new approaches to conditions such as peripheral vascular disease and dementia management. Patients with peripheral vascular disease require all the steps used in CHD and stroke such as good blood pressure control and statin therapy. We also need to encourage the use of ankle brachial pressure measurement using dopplers in primary care. Newly diagnosed dementia patients require numerous investigations such as thyroid function, B12 and folate testing as well as an assessment of depression.
We could further refine some of the present cardiovascular indicators. We should be encouraging all primary care clinical staff to take the patients pulse in order to find cases of atrial fibrillation and in coronary heart disease to ensure a reasonable control of heart rate. Modern automatic blood pressure machines have removed pulse awareness. We have become very good at checking if an angina patient has had yet another cholesterol test despite them having been on a statin for many years but do we think to ask them if they have had any chest pain recently, information the patient will not necessarily volunteer. I fear that in our current national financial difficulties the NHS Health Checks programme will be an easy target for cut backs. Further primary prevention QOF indicators would be needed to bolster PP1 and PP2 if that did happen. In fact I have always felt that QOF would have worked better for primary prevention than the local enhanced service route.
Some indicators need to be culled because of changing or lacking evidence.
Having said all that I do feel careful selection is required. There are dynamic situations where regular checks are needed and patient management often requires a change, hypertension being a good example. Blood pressure management in the United Kingdom has improved considerably under QOF and I would not tamper with the indicators too much.
One last point is that we also need to return to an element of being trusted to do as a good a job as we can for our patients. I doubt any of us would start dismantling the care we have provided just because QOF is not looking over our shoulders in that particular aspect anymore.
Dr Terry McCormack is a GP in Whitby and former chair of the Primary Care Cardiovascular Society
Blood pressure being taken Debate debate no
