Dr Jonathan Graffy explains the rationale for the revised quality and outcomes framework regarding three targets for glycaemic control in patients with diabetes

Diabetes mellitus affects over 2.3 million people in the UK1 and its prevalence is rising, partly as a result of increasing levels of obesity and decreased physical activity, but also because people live longer after diagnosis. This is a tribute to better cardiovascular risk protection. It is well known that successful management of diabetes requires an organised, proactive approach, and also a partnership with patients who have a thorough understanding of their condition.

Since the quality and outcomes framework (QOF) was introduced in 2004, there have been significant improvements in the quality of diabetes care in the UK.2 More people with the condition are now diagnosed, and there is evidence that risk factors, such as raised blood pressure, hyperglycaemia, and raised cholesterol levels are being managed more effectively.3 Most of these improvements were, however, achieved during the first 2 years of the QOF, and since 2005, the rate of improvement has slowed.

Revised targets

In May 2008, NICE published a revised guideline on the management of type 2 diabetes that emphasised the need to set individual targets.4 Although the NICE Guideline Development Group (GDG) viewed 6.5% as ideal for glycated haemoglobin (HbA1c), it cautioned against intensive efforts to get below this value. The NICE guideline recognised that for many patients, higher targets are more appropriate. However, as a result of this new guidance, the previous QOF targets for blood-glucose levels of 7.5% and 10% looked increasingly out of step.5

Up until April 2009, the mechanism for reviewing QOF performance measures involved an expert panel reviewing relevant evidence and advising on changes that might be included. These amendments were then considered at annual negotiations between NHS Employers and the profession’s representatives (BMA General Practitioners Committee). As a result of this review, and in line with the NICE guideline, the targets for glycaemic control were revised for the 2009/10 update of the QOF. Three new indicators were introduced: DM 23, DM 24, and DM 25 (see Table 1), with new targets of 7%, 8%, and 9%.6

The new targets were introduced for three reasons:

  • There is evidence that poor glycaemic control is associated with more complications and the previous levels were significantly above those advised in the NICE guideline
  • The old targets only measured HbA1c at two points on the distribution curve (7.5% and 10%) so there was no incentive to improve levels from 9% to 8%
  • Most practices in the UK were already doing better than the previous targets.
Table 1: New indicators for glycaemic control in 2009/10 QOF update6
DM 23 The percentage of patients with diabetes in whom the last HbA1c is 7% or less (or equivalent test/reference range depending on local laboratory) in the previous 15 months (replaces DM 20) 17 40–50%
DM 24 The percentage of patients with diabetes in whom the last HbA1c is 8% or less in the previous 15 months (new) 8 40–70%
DM 25 The percentage of patients with diabetes in whom the last HbA1c is 9% or less in the previous 15 months (replaces DM 7) 10 40–90%
HbA1c=glycated haemoglobin

The case for tightening glycaemic control

The importance of glycaemic control in reducing both microvascular and macrovascular complications in type 1 diabetes was demonstrated in the DCCT/EDIC (Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications) study.7 In patients with type 2 diabetes, the UKPDS (United Kingdom Prospective Diabetes Study) reported that at 10-years’ follow up, mortality was significantly lower in those initially managed intensively (13% less in the sulfonylurea and insulin group [p=0.007] and 27% less in the metformin group [p=0.002]).8 Reductions in cardiovascular risk appear only after several years of intensive treatment.

At the population level, observational studies also demonstrate a continuous relationship between glycaemia and risk of mortality and cardiovascular events. For example, an increase of 1% in HbA1c was independently associated with a 28% higher risk of death in the European Prospective Investigation of Cancer and Nutrition–Norfolk cohort.9 This association extends below the diagnostic cut off for diabetes, suggesting that as with blood pressure and cholesterol, the lower the HbA1c the greater the benefit in the long term, providing that treatment does not cause problems in the short term.

The recently published ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial suggests, however, that there are risks in setting glycaemic levels too low in patients with type 2 diabetes.10 In this study, an aggressive treatment strategy resulted in a median HbA1c level of 6.4% in the intervention group, but this was associated with increased mortality. Although this has led some to question whether the revised QOF glycaemic targets are safe,11 the concern appears unfounded because the trial protocol was very different from everyday practice. Participants had poor baseline glycaemic control and HbA1c was reduced rapidly (1.4% over 4 months) towards a target of <6%, often using rosiglitazone.10 The study was not directly relevant to the QOF target of 7% for 50% of the whole population with diabetes.

In contrast, there was no between-group difference in all-cause mortality in ADVANCE12 or the Veteran Affairs Diabetes Trial,13 which investigated the effect of intensive glucose control on vascular outcomes. The UKPDS, mortality was significantly lower in the intensive-treatment arm.8 Nevertheless, the ACCORD results certainly cast doubt on the benefits of rapidly aiming for normoglycaemia (HbA1c below 6.5%) in patients who have had poorly controlled diabetes and perhaps cardiovascular disease for many years.

The case for split-level targets

The purpose of shifting from two targets in HbA1c (7.5% and 10%) to three (7%, 8%, and 9%) in the 2009/10 QOF clinical indicators was to promote a downward shift in blood glucose across the whole distribution range. The importance of setting multiple targets is illustrated by the distribution of HbA1c readings from the National Diabetes Audit in 2005–2006 (Figure 1).14 These data also show that many patients are already meeting these new targets.

Separate QOF targets for HbA1c at 7%, 8%, and 9% allow clinicians to select the most appropriate level in discussion with each individual patient. These individual judgements remain important and clinicians should not try to lower every patient’s HbA1c to ?7% as this may be inappropriate or unsafe. Patients do need to know, however, that any reduction in HbA1c towards their agreed target should benefit their future health, unless side-effects (including hypoglycaemia), or their efforts to achieve it, impair their quality of life.

HbA1c ?7%
According to the 2005–2006 National Diabetes Audit data, 45% of patients had HbA1c values of ?7%, a percentage that is improving as people are diagnosed earlier. Earlier diagnosis and better management have together served to reduce the mean HbA1c for the population as a whole, making the targets easier to attain. The year-on-year improvements in glycaemic control noted in recent years are likely to have continued since the audit was conducted.

HbA1c ?8%
The new HbA1c mid-level target of ?8% is important because merely reducing the lower reading without monitoring performance at an interim level would reduce the incentives to improve care for those unable to reach 7%. The thresholds were set based on the 2005–2006 National Diabetes Audit data, which showed that 71.5% of people already had an HbA1c level below 8%.14

HbA1c ?9%
The reduction in the upper target level from 10% should encourage healthcare professionals to do more for people with poor control of HbA1c. If their blood-glucose level can be lowered to 9%, patients should face fewer complications in the long term. This group of patients often has complex needs and may require more intensive personalised approaches to care.

Figure 1: Distribution of HbA1c levels in England and the new diabetes targets and payment thresholds14
Graffy graph
HbA1c=glycated haemoglobin
Data taken from the National Diabetes Audit 2005–2006

Achieving the new targets in practice

For most practices, the new indicators will not be difficult to achieve, but those working in areas with many patients of South Asian descent may find it harder. Type 2 diabetes is between four and six times more prevalent in this group and there are various cultural and other barriers to achieving better glycaemic control.15 For these practices, ensuring that half their patients have an HbA1c below 7% may prove challenging. Indeed, if there is evidence that more patients are exception reported, or that QOF scores fall in deprived areas, there might be a case for reviewing the upper threshold for this indicator.

Understanding the needs of patients with diabetes is always important, but particularly so for the small minority of patients who have poor control of blood glucose. Patients with type 1 diabetes may be more concerned about avoiding hypoglycaemic episodes than their HbA1c level. Individuals with type 2 diabetes may be reluctant to switch to insulin, although tablets may not provide sufficient control as their beta-cell function declines over time. Despite good intentions, many patients do not follow healthy diets or lead active lives, and some baulk at taking multiple medications. Alcohol, bingeing, depression, and family stress can also contribute to poor control. However, it is part of the GP’s job to help people tackle their complex problems. This means finding out why patients find it hard to control their diabetes and developing a shared plan to help them manage this. Healthcare professionals can promote better self management in a number of ways; for example, ensuring that the patients:16,17

  • attend structured education
  • agree on a personalised care plan.

For the GP, treatment of diabetes is becoming ever more complex. Advice on diet, exercise, and avoiding obesity is important, but there are several other factors to consider:

  • Smokers may need help to quit
  • Blood pressure and lipids may require drug treatment
  • Complications such as retinopathy, nephropathy, neuropathy, and other foot problems demand attention.

There have been real advances in new medicines, but it would be costly if these were widely adopted, and side-effects (e.g. concerns over heart failure and fractures with rosiglitazone treatment) can take time to emerge. Although it is beyond the scope of this article, the 2009 NICE guideline on the control of blood glucose levels helps to clarify the place of the newer therapeutic agents.18 The guideline is particularly useful for the way in which it sets out options for combining drugs and monitoring whether new agents are effective; it covers:

  • thiazolidinediones (pioglitazone, rosiglitazone)
  • dipeptidyl peptidase-4 (DPP-4) inhibitors (sitagliptin, vildagliptin)
  • the glucagon-like peptide-1 (GLP-1) mimetic licensed at publication (exenatide)
  • options for insulin.

Should other indicators change?

The other indicators involved in the management and monitoring of patients with diabetes are shown in Table 2.6 These cover most of the measurable risk factors and will prompt early detection of complications. Most practices have established systems to check these regularly and use templates to organise diabetes consultations. This may be a reason not to make further changes, but there is a case for reducing duplication between measures (especially when people have more than one condition). Looking ahead, if the QOF stagnates, there is a risk that politicians and the public will question whether they are getting value for money. As health spending comes under greater scrutiny, funds may leach away if they are not delivering further improvements in care. So the debate about HbA1c targets should prompt a wider discussion about how the QOF might adapt and change for other components of diabetes care.

Since April 2009, a new process of reviewing the indicators, coordinated by NICE, has been adopted. Final decisions will continue to remain with NHS Employers and the profession’s negotiators. Previously, indicators could be adopted if both sides agreed; now they will be tested in pilot practices before they are implemented more widely. Changes under consideration include:

  • Split-level targets for blood pressure, as suggested in previous expert panel reports.14 These would operate at 140/80 mmHg and 150/90 mmHg, rewarding tighter control as the ideal, but accepting that not everyone can reach 140/80 mmHg without adverse effects
  • Removing payments for measuring levels of HbA1c, lipids, and blood pressure (as opposed to achieving certain levels of control)19
  • New measures for foot care: switching from testing pulses and neuropathy to a fuller assessment of the risk of foot problems. This was suggested in previous expert panel reports, but will now be piloted by the National Primary Care Research and Development Centre.20

Over the next 2 years, new measures of HbA1c (expressed as mmol per mol of haemoglobin) will be implemented, with laboratories reporting only the new units after June 2011.21 This changeover to meet the International Federation of Clinical Chemistry guidelines means that changes to the QOF targets will also be needed. The equivalent levels of 53 mmol/mol (7%), 64 mmol/mol (8%), and 75 mmol/mol (9%) may prove harder for clinicians and patients to remember, but the further revision seems unavoidable.

Table 2: Indicators for diabetes mellitus excluding new measures of HbA1c control6
No. Indicator Points Payment stages
DM 19 The practice can produce a register of all patients aged 17 years and over with diabetes mellitus, which specifies whether the patient has type 1 or type 2 diabetes 6  
DM 2 The percentage of patients with diabetes whose notes record BMI in the previous 15 months 3 40–90%
DM 5 The percentage of patients with diabetes who have a record of HbA1c or equivalent in the previous 15 months 3 40–90%
DM 21 The percentage of patients with diabetes who have a record of retinal screening in the previous 15 months 5 40–90%
DM 9 The percentage of patients with diabetes with a record of the presence or absence of peripheral pulses in the previous 15 months 3 40–90%
DM 10 The percentage of patients with diabetes with a record of neuropathy testing in the previous 15 months 3 40–90%
DM 11 The percentage of patients with diabetes who have a record of the blood pressure in the previous 15 months 3 40–90%
DM 12 The percentage of patients with diabetes in whom the last blood pressure is ?145/85 mmHg 18 40–60%
DM 13 The percentage of patients with diabetes who have a record of micro-albuminuria testing in the previous 15 months (exception reporting for patients with proteinuria) 3 40–90%
DM 22 The percentage of patients with diabetes who have a record of estimated glomerular filtration rate or serum creatinine testing in the previous 15 months 3 40–90%
DM 15 The percentage of patients with diabetes with a diagnosis of proteinuria or micro-albuminuria who are treated with ACE inhibitors (or angiotensin II receptor blockers) 3 40–80%
DM 16 The percentage of patients with diabetes who have a record of total cholesterol in the previous 15 months 3 40–90%
DM 17 The percentage of patients with diabetes whose last measured total cholesterol within the previous 15 months is ?5 mmol/l 6 40–70%
DM 18 The percentage of patients with diabetes who have had influenza immunisation in the preceding 1 September to 31 March 3 40–85%
BMI=body mass index; HbA1C=glycated haemoglobin; ACE=angiotensin-converting enzyme

Strategies for lipid management (DM 16 and DM 17)

Currently, the QOF indicators are based on a ‘test and treat’ approach, rewarding reductions in the level of total cholesterol to 5.0 mmol/l, but this seems increasingly out of step with current practice. Low-density lipoprotein (LDL) cholesterol is a better marker of vascular risk, but even this is less useful than risk scores, which incorporate multiple risk factors. Also, because the benefits come from treatment, not measurement, there is a case for incentive payments rewarding this. One option would be to measure the percentage of patients with diabetes whose total cardiovascular risk is over 20% who are prescribed a statin. It is time for a wider debate in this area, and any change should be carefully piloted if it is to be put into practice.

Measuring education and self-management support

Most of the QOF targets cover processes that can be easily measured, but some of the most important aspects of diabetes care are the activities that healthcare professionals do to engage patients: how they explain the condition and medication, and how they help people to change their diet and levels of physical activity. Providing that it does not encourage a ‘tick-box’ mentality, the QOF could do more to encourage this type of support. Potential new indicators worth exploring include systematic recording of referral to structured education, particularly for people recently diagnosed, and ensuring that the patient has a copy of an agreed personalised care plan. These ideas should be high on the agenda for the next round of indicators to pilot.


The QOF has supported significant improvements for diabetes care in the UK. This year’s changes to the HbA1c indicators provide an additional incentive to build on this, but controlling glycaemia is only part of the story. Managing blood pressure, weight, and lipid levels are also important, as is the early detection of complications. It is now time for a wider discussion about how the QOF might better reward best practice for these other aspects of care.


  1. NHS Choices. Diabetes, type 1. www.nhs.uk/conditions/diabetes/Pages/Introduction.aspx (accessed 28 September 2009).
  2. Campbell S, Reeves D, Kontopantelis E et al. Quality of primary care in England with the Introduction of pay for performance. N Engl J Med 2007; 357 (2): 181–190.
  3. Campbell S, Reeves D, Kontopantelis E et al. Effects of pay for performance on the quality of primary care in England. N Engl J Med 2009; 361 (4): 368–378.
  4. National Institute for Health and Care Excellence. Type 2 diabetes: the management of type 2 diabetes. Clinical Guideline 66. London: NICE, 2008. Available at: www.nice.org.uk/guidance/CG66/Guidance/pdf/English
  5. British Medical Association, NHS Employers. Revisions to the GMS contract, 2006/07. Delivering investment in general practice. London: BMA, 2006.
  6. British Medical Association, NHS Employers. Quality and outcomes framework guidance for GMS contract 2009/10. London: BMA, NHS Employers, 2009.
  7. The Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study Research Group. Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes. N Engl J Med 2005; 353 (25): 2643–2653.
  8. Holman R, Paul S, Bethel M et al. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med 2008; 359 (15): 1577–1589.
  9. Khaw K, Wareham N, Luben R et al. Glycated haemoglobin, diabetes, and mortality in men in Norfolk cohort of European Prospective Investigation of Cancer and Nutrition (EPIC-Norfolk). BMJ 2001; 322 (7277): 15–18.
  10. The ACCORD study group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008; 358 (24): 2545–2559.
  11. Lehman R, Krumholz H. Tight control of blood glucose in long standing type 2 diabetes. BMJ 2009; 338: b800.
  12. The ADVANCE Collaborative Group. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008; 358 (24): 2560–2572.
  13. Duckworth W, Abraira C, Moritz T et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009; 360 (2): 129–139.
  14. Graffy J, Griffin S. Review of the quality and outcomes framework for diabetes: current indicators 2007–2008. Manchester: University of Manchester, 2008. Available at: www.npcrdc.ac.uk/Publications/Diabetes_200708.pdf
  15. Khunti K, Kumar S, Brodie J. Diabetes UK and South Asian Health Foundation recommendations on diabetes research priorities for British South Asians. London: Diabetes UK, 2009. Available at: www.sahf.org.uk/uploads/docs/files/26.pdf
  16. Davies M, Heller S, Skinner T et al. Effectiveness of the diabetes education and self management for ongoing and newly diagnosed (DESMOND) programme for people with newly diagnosed type 2 diabetes: cluster randomised controlled trial. BMJ 2008; 336 (7642): 491–495.
  17. Graffy J, Eaton S, Sturt J, Chadwick P. Personalized care planning for diabetes: policy lessons from systematic reviews of consultation and self-management interventions. Primary Health Care Research & Development 2009; 10 (3): 210–222.
  18. National Institute for Health and Care Excellence. Type 2 diabetes: newer agents for blood glucose control in type 2 diabetes. Clinical Guideline 87. London: NICE, 2009. Available at: www.nice.org.uk/guidance/index.jsp?action=download&o=44318
  19. Anekwe, L. Hunter of new QOF targets emerges from the shadows. Pulse 19 Aug 09.
    www.pulsetoday.co.uk/story.asp?sectioncode=20&storycode=4123467 (accessed 23 October 2009).
  20. National Institute for Health and Care Excellence. Quality and outcomes framework indicators in development. London: NICE, 2009. Available at: www.nice.org.uk/aboutnice/qof/IndicatorsInDevelopment.jsp
  21. News Distribution Service. Change in measurement of HbA1c for people with diabetes. nds.coi.gov.uk/content/detail.aspx?NewsAreaId=2&ReleaseID=402333&SubjectId=2 (accessed 9 October 2009). G
written by Dr David Jenner, NHS Alliance GMS contract/PBC Lead
  • The new QOF targets for glycaemic control in dabetes are likely to drive expenditure on prescribing budgets
  • Newer agents such as DPP-4 inhibitors, thiazolidinediones, and GLP mimetics are much more expensive than metformin and sulfonylureas
  • The QOF target DM 23 is out of line with current NICE guidance, which sets a threshold for HbA1c of 7.5% for a range of more aggressive hypoglycaemic agents
  • PCTs should consider local exception reporting for DM 23 to avoid inappropriate prescribing and costs in cases where control cannot be reached using the agents as defined in NICE guidance
  • Practices should be encouraged to screen for diabetes in at-risk groups and through the NHS Health Check—earlier diagnosis helps increase the proportion of patients with diabetes with HbA1c under 7%, and improves outcomes
  • There is scope for professional education and perhaps prescribing incentives for professionals to use the new medications appropriately—use of newer agents in individuals should be reviewed in line with the updated NICE guideline on type 2 diabetes before their long-term use is considered
  • Local commissioners may want to review their provision for structured education and to ensure that this is offered to newly diagnosed patients
  • The pressure to improve glycaemic control may lead to more patients with type 2 diabetes being switched to insulin. Commissioners should ensure that the local arrangements to support this process work well and are well-integrated into primary care—the setting where most patients with type 2 diabetes are seen