Head To Head

Should we screen for type 2 diabetes: Yes

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e4514 (Published 09 July 2012) Cite this as: BMJ 2012;345:e4514
  1. Kamlesh Khunti, professor 1,
  2. Melanie Davies, professor 2
  1. 1Department of Health Sciences, University of Leicester, Leicester Diabetes Centre, Leicester LE5 4PW, UK
  2. 2Department of Cardiovascular Sciences, University of Leicester
  1. Correspondence to K Khunti kk22{at}le.ac.uk

Kamlesh Khunti and Melanie Davies believe that screening for diabetes is feasible and cost effective, but Elizabeth Goyer and colleagues (doi:10.1136/bmj.e4516) think that a population approach to modifying risk factors would be better

Type 2 diabetes is common, increasing in prevalence, and associated with high morbidity, mortality, and healthcare costs. Current estimates suggest that there are about 3.1 million adults in England with diabetes, of whom roughly a quarter are undiagnosed.1 For several years many countries have been debating whether to screen for diabetes.2 The disease certainly meets many of the criteria for screening: its natural course is well characterised, it can be asymptomatic, many people remain undiagnosed, high numbers of people have complications at diagnosis, and there is now good quality evidence regarding effective management for those with established diabetes. There is also evidence that screening for diabetes does not result in any harm to patients.3 4 Although there has been some uncertainty about the uptake of a screening programme in the real world setting,5 we believe it should be implemented.

Evidence for screening

No trials have compared screening for diabetes with no screening, which would be the ideal platform to establish whether screening is cost effective. The recent ADDITION Europe randomised study of 3055 people investigated the effect of intensive treatment of multiple risk factors in people with screen detected type 2 diabetes. The patients in the control arm were also screen detected but received routine care from primary care practitioners.4 At five year follow-up, the intensive group had significant improvements in cardiovascular risk factors compared with the control group, although the 17% relative risk reduction in the primary outcome (first cardiovascular event) was non-significant (13.5/1000 person years in the intensive treatment group compared with 15.9/1000 in the control group).4 However, the UK Prospective Diabetes Study showed that the effects of tight glycaemic control on development of diabetes related complications are longer term6 and therefore would be unlikely to be seen within the five year follow-up of the ADDITION study. The ACCORD and Veterans Administration diabetes randomised trials also showed benefits of tight control for people without previous cardiovascular events who were earlier in their disease trajectory (duration of diabetes less than 15 years).7 8

In the absence of a control arm without screening, it is difficult to ascertain the true benefits of screening. Current recommendations have therefore been based on modelled cost effectiveness analyses. Any screening programme for diabetes will also identify people at high risk of type 2 diabetes, and the evidence for prevention of diabetes in this group is unequivocal.9 One recent economic decision analysis model estimated that compared with no screening, the cost of screening for diabetes and impaired glucose tolerance followed by lifestyle interventions was £6242 (€7750; $9750) for each quality adjusted life year (QALY) gained.10 A simulated study using the Archimedes model suggested that compared with no screening, screening for type 2 diabetes is cost effective when started between the ages of 30 years and 45 years, with screening repeated every three to five years.11 Compared with no screening, all screening strategies reduced the incidence of myocardial infarction (3-9 events prevented per 1000 people screened) and microvascular complications (3-9 events prevented per 1000 people screened) and deaths (2 to 5 events per 1000 people).11

Practicalities

Another review suggested that two stage screening for diabetes is cost effective for patients aged 40-70 years compared with no screening (£10 216/QALY).12 However, most of the economic modelling studies have used the oral glucose tolerance test for diagnosis, which has been shown to be impractical and reduces uptake.13 Recently the World Health Organization has recommended using glycated haemoglobin (A1c) concentrations to diagnose diabetes,14 which should increase the uptake of screening.

The recent evidence makes the case for screening for diabetes (and those at high risk) much stronger, and two major initiatives in the UK are likely to help implement screening. The National Health Service Health Checks programme for adults aged 40-74 years was introduced in England in 2009 with the objective of assessing the risk of developing vascular or metabolic disease, including diabetes. The second initiative is the imminent publication of the National Institute for Health and Clinical Excellence’s guidance on identification and prevention of type 2 diabetes in people at high risk.15The guidance makes practical recommendations on risk identification using simple self assessment risk scores or computer based risk scores in people aged 40 to 75 years and younger people from black and minority ethnic groups followed by a fasting glucose or a random haemoglobin A1c assessment. The guidance also makes pragmatic recommendations to help people make long term lifestyle changes to reduce risk and delay onset of type 2 diabetes.15 The comprehensive modelling in this programme has also suggested that this strategy is likely to be cost effective.

Direct evidence of the benefits of early detection of type 2 diabetes by screening are lacking and are now unlikely to be obtained. However, the ADDITION study, the recent robust modelling studies, and a move to using non-invasive tests followed by haemoglobin A1c as a diagnostic test for diabetes suggest that screening and intensive risk factor control of people with diabetes is feasible in primary care and is likely to be cost effective.

Notes

Cite this as: BMJ 2012;345:e4514

Footnotes

  • Competing interests: Both authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that that had no support from any organisation for the submitted work; MJD has received funds for research, honorariums for speaking at meetings, and has served on advisory boards for Lilly, Sanofi Aventis, MSD, Novo Nordisk, BMS, BI, and Roche. KK has received funds for research, honorariums for speaking at meetings, or served on advisory boards for Astra Zeneca, GSK, Lilly, Novartis, Pfizer, Servier, Sanofi Aventis, MSD, and Novo Nordisk. KK (chair) and MJD are members of the National Institute for Health and Clinical Excellence public health guidance on preventing type 2 diabetes and both are advisers to the UK Department of Health for the NHS health checks programme.

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

References