This site is intended for health professionals only

At the heart of general practice since 1960

Will relaxing QOF HbA1c target harm care of patients with diabetes?

Moving the target from 7% to 7.5% may see some patients miss out on best care, says Dr Martin Hadley-Brown. But Dr Richard Lehman argues that the 7% target should never have been introduced and it is right to scrap it

Moving the target from 7% to 7.5% may see some patients miss out on best care, says Dr Martin Hadley-Brown. But Dr Richard Lehman argues that the 7% target should never have been introduced and it is right to scrap it



The QOF indicator advisory committee recommended the HbA1c target for diabetes should be raised from 7% to 7.5% because of concerns that some patients were being treated too aggressively.

But if that is the worry, there are other ways you could address it that would ensure all patients get the most appropriate treatment for them. One would have been to keep the lowest target at 7% but relax the threshold to earn the top payment from 50% to, say, 40% or 45%. I feel it would have been better to relax the thresholds. It would have encouraged GPs still to aim for less than 7% in patients where it was safely achievable, but with the extra leeway, meaning that they would not feel that they had to exception report to achieve the target. Exception reporting went up significantly when the target was introduced and I think that's why.

My worry is that getting below 7.5% is often reasonable and appropriate. Many of our patients – especially when they are first diagnosed – may well be able to get down below 7% (to 6.5% in some cases) just on metformin, rather than the combinations of different drugs that put them at risk of hypoglycaemia.

I'm convinced they benefit from that level of control, and that tight control early in their journey with diabetes improves long-term outcomes. The extension to the UK Prospective Diabetes Study published in 2008 confirmed the persistence of benefit in people who had had reasonable control right at the beginning of their time with the disease.

If I have someone who comes in now, newly diagnosed with type 2 diabetes, I am likely to set a target as close to 6.5% – essentially normal – as possible, and to try to achieve it with lifestyle changes, exercise and diet, and starting metformin early. And rather than waiting for them to get to 7.5% before increasing their metformin dose or adding a second agent, that 7% target acts as a reminder to do it early on.

I think putting the target back up to 7.5% may be interpreted as saying ‘actually, that's not worth doing or that's not important' and suggest that slacker control in the early stage of diabetes is justified. And I'm simply not convinced by that.

I also think that lowering the threshold from 50% to 40% or 45% would have had another advantage in terms of exception reporting. In our practice we were hitting 45% below an HbA1c of 7% without having to exception report that much.

But that last 5% made us look through our list to find people who were genuinely appropriate for exception reporting to make sure we weren't being penalised. We ended up with exception reporting rates round about 8%, which was more than double what it had been the year before.

There are dangers in exception reporting people. They may be excluded from services they would otherwise get such as retinal screening. Being exception reported is not ideal because it gives the message that those patients are not part of best care.

It is for those reasons I'm convinced keeping 7% but dropping the threshold would be the better option. We don't want to lose the message that – for some patients at least – aiming for 7% is entirely appropriate and good care.

Dr Martin Hadley-Brown is a GP in Thetford, Norfolk, and chair of the Primary Care Diabetes Society



The QOF incentive to lower the level of HbA1c below 7% in 50% of patients with diabetes should never have come into force. Nine months before it was introduced, two papers in the New England Journal of Medicine showed that such a strategy in people with long-standing type 2 diabetes did not improve outcomes and could actually increase mortality. This was followed by a deafening silence from the diabetes specialist community, which for decades had argued the opposite. Even the NEJM itself failed to provide an editorial to accompany these groundbreaking studies, ACCORD and ADVANCE.

In the UK, NICE produced an update on type 2 diabetes in May 2008, just two months too soon to include these studies. Instead, from some observational evidence and over-extrapolation from the UKPDS trial, it recommended that the HbA1c of all diabetic patients should be reduced to a target level of 6.5%. Later in 2008, well after the evidence was in that such a policy might be harmful, this now outdated chapter of the NICE guideline was used as justification for a new indicator in 2009.

I grew increasingly worried, especially when a third study – the Veterans Affairs Diabetes Trial – confirmed lack of benefit and an increase in severe hypoglycaemia.

I co-wrote a BMJ editorial that pointed out the potential harm of the new QOF target. This appeared immediately after the target was introduced and led to much comment but no immediate action.

Since then the diabetology community has been slowly changing its tune in light of the evidence. Many primary care organisations were challenged to waive this QOF target on the grounds of patient safety.

Finally in June this year, two years after the publication of ACCORD and ADVANCE, the QOF committee was reconvened to reconsider the 7% HbA1c target.

The decision by the committee to relax the target was, to say the least, unsurprising. In fact, the incoming evidence supports a policy of maintaining HbA1c in people with long-standing type 2 diabetes at a higher range than 7.5%. A Lancet study published this year based on the UK general practice research database – so in fact the very people that we in UK primary care are treating with additional drugs to lower HbA1c – finds higher mortality both above and below the range of 7.5-9.4%. This may still come as a shock to many people accustomed to the ‘lower is better' way of thinking, which has been drummed into us for the last 20 years.

We have swallowed this idea because it seems logical and is strongly supported by observational evidence that shows a straight-line relation between HbA1c and diabetic complications right down to a level of about 5%.

Unfortunately, that does not equate to being able to reverse such complications by reducing HbA1c by any means at our disposal. Some treatments accelerate ß-cell depletion; others cause an increase in cardiovascular events by other mechanisms; most lead to weight gain and hypoglycaemia, especially insulin.

The QOF target has almost certainly caused harm to patients. This is not only suggested by the mortality figures from the Lancet study, but is also reflected in national audit figures that show a misdirection of effort in trying to reduce HbA1c in well-controlled patients with long-standing disease, while ignoring the high-risk poorly compliant patients who will never fall neatly into the QOF system of targets. These are the people we should concentrate on, whether or not we get paid for doing so.

Dr Richard Lehman is a GP in Banbury, Oxfordshire, and member of the Primary Care Cardiovascular Society

Will relaxing QOF HbA1c target harm care of patients with diabetes? Will relaxing QOF HbA1c target harm care of patients with diabetes? Yes No

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say