Intended for healthcare professionals


Tight control of blood glucose in long standing type 2 diabetes

BMJ 2009; 338 doi: (Published 06 March 2009) Cite this as: BMJ 2009;338:b800
  1. Richard Lehman, general practitioner1,
  2. Harlan M Krumholz, Harold H Hines Junior professor of medicine and epidemiology and public health2
  1. 1Hightown Surgery, Banbury OX16 9DB
  2. 2Section of Cardiovascular Medicine and Robert Wood Johnson Clinical Scholars Program, Department of Medicine, Section of Health Policy and Administration, School of Public Health, Yale University School of Medicine, PO Box 208088, New Haven, CT 06520-8088, USA
  1. richard.lehman{at}

    Reducing glycated haemoglobin below 7% is not supported by evidence and may even be harmful

    During the past year, three important studies have provided evidence that tighter glycaemic control (to <7% glycated haemoglobin) in older adults with type 2 diabetes does not provide substantial benefit and may increase the risk of adverse outcomes. These findings, which some experts and policy makers found surprising, should lead to the re-evaluation of recommendations about what constitutes high quality care for these patients.

    The management of type 2 diabetes in the United Kingdom takes place largely in primary care and is strongly influenced by the requirements of the quality and outcomes framework (QOF)—an annual reward and incentive programme, which although voluntary provides a substantial proportion of general practitioners’ income. From April 2009, general practitioners in the UK will need to reduce glycated haemoglobin in half of their patients with type 2 diabetes to below 7% to earn the same amount that they are currently paid for achieving a target of 7.5%. The average practice that achieves this level of performance will be paid around £3000 (€3375; $4250). Tens of thousands of patients will need to be given additional oral treatment or will be treated with insulin. Treatment with insulin brings with it an increased risk of hypoglycaemia and the additional costs of daily blood glucose monitoring and the insulin itself. It may also result in people who drive for a living losing their jobs if the new target leads them to be treated with insulin.

    This new glycaemic target was agreed on by NHS employers and the general practitioners’ committee of the BMA in October 2008, ironically just when evidence was gathering that tight glycaemic control in established type 2 diabetes has little benefit and can even be harmful. In June last year, two large randomised controlled trials of intensive glucose lowering in type 2 diabetes were published. ACCORD (Action to Control Cardiovascular Risk in Diabetes; 10 251 patients, mean age 62, median disease duration 10 years, mean follow-up 3.5 years) found no evidence of a lower risk of non-fatal myocardial infarction, non-fatal stroke, or death from cardiovascular causes in the group in which glycated haemoglobin was lowered to a median of 6.4% compared with a standard treatment group maintained at a median of 7.5%.1 In fact, the trial was terminated early because of higher mortality in the intensively treated group. ADVANCE (Action in Diabetes and Vascular Disease; 11 140 patients, mean age 66, mean disease duration eight years, median follow-up five years) found a small reduction in total events in the tight control group (mean glycated haemoglobin 6.4% compared with 7.3%), which was mainly accounted for by a 21% reduction in nephropathy (mostly new onset microalbuminuria), but no effect on major cardiovascular events.2 In January 2009, another randomised trial from the United States randomised military veterans with type 2 diabetes of 11.5 years’ mean duration to different levels of glycaemic control (VADT; 1791 mainly male patients, mean age 60.5, median follow-up 5.6 years).3 It found no significant differences in macrovascular or microvascular outcomes over 5.6 years between patients in the “standard treatment” group (median glycated haemoglobin of 8.4%) and those who were intensively treated (mean value 6.9%). Each of the trials showed an excess risk of hypoglycaemia in the group that was treated more intensively. Taken together, the three trials show that no reduction of clinically meaningful adverse outcomes occurred in patients with long standing type 2 diabetes treated to a glycated haemoglobin below 7.0% in the time periods studied. Moreover, intensive treatment is accompanied by substantial costs and an increased risk of hypoglycaemia and perhaps mortality.

    At first glance, these findings seem to disagree with the recently published long term results of the UK Prospective Diabetes Study (UKPDS).4 This study reported that some patients who achieved good early control show benefit at a median follow-up of 17 years, even though tight glycaemic control is usually lost after the first year. However, this study was designed to compare dietary control with intensive drug treatment in younger patients (mean age 54 years at enrolment) who were newly diagnosed with type 2 diabetes, and it provides no guidance on appropriate strategies for older patients with long standing disease.

    The idea that tight glycaemic control for everyone would improve outcomes was a hypothesis that needed to be tested 30 years ago, when the UKPDS was set up in response to the finding of increased mortality from cardiovascular disease in the earlier University Group Diabetes Programme.5 The first reports from UKPDS indicated that the strategy was successful,6 although tight control, defined as a fasting plasma glucose concentration of 6 mmol/l, using sulfonylureas and insulin had no effect on cardiovascular disease and improved microvascular outcomes only minimally—mainly the need for retinal photocoagulation.7 But in the wake of the three recent studies of tight glycaemic control in older patients with long standing diabetes, it is certainly “time to challenge conventional wisdom,” as the subtitle of a recent commentary on the ACCORD trial states.8 The optimal approach for patients remains to be elucidated.

    Another question is whether all strategies to reduce glycated haemoglobin have the same effect; a recent systematic review of cardiovascular outcomes in oral drugs for diabetes shows that the level of our current understanding in this area is unsatisfactory.9 We need better evidence to evaluate the balance of risk and benefit for individual patients, and we need to move away from the simplistic idea that the value of a particular drug or strategy can be predicted by its glycaemic lowering effects. Even when a drug lowers blood glucose, it may make a difference how this occurs. In UKPDS the strategy seemed to matter, with better long term outcomes seen in the overweight group treated with metformin.

    The QOF in the UK has been a successful driver of evidence based improvement in the care of diabetes, particularly tight control of blood pressure and the prescription of statins. But by encouraging tighter glycaemic control in all patients with type 2 diabetes, regardless of disease duration and the drugs used to achieve control, the new QOF target encourages an outdated strategy and one that may not provide a net benefit to patients. Moreover, given the evidence and the complexity of the decision, patient preference should play a strong role in the strategy that is pursued. The change of target from 7.5% to 7% should be withdrawn before it wastes resources and possibly harms patients.


    Cite this as: BMJ 2009;338:b800


    • Competing interests: HMK has research contracts for developing outcome measures for public health reporting with the Centers for Medicare and Medicaid Services, and he chairs a scientific advisory board for UnitedHealthcare. Neither of these roles is related specifically to the management of diabetes.

    • Provenance and peer review: Commissioned; not externally peer reviewed.