Head To Head

Should we screen for type 2 diabetes? No

BMJ 2012; 345 doi: http://dx.doi.org/10.1136/bmj.e4516 (Published 09 July 2012) Cite this as: BMJ 2012;345:e4516
  1. Elizabeth Goyder, professor1,
  2. Les Irwig, professor 2,
  3. Nick Payne, honorary senior lecturer1
  1. 1ScHARR, University of Sheffield, Sheffield S1 4DA, UK
  2. 2University of Sydney, Screening and Test Evaluation Program, Sydney School of Public Health, Sydney, New South Wales, Australia
  1. Correspondence to: E Goyder e.goyder{at}sheffield.ac.uk

Kamlesh Khunti and Melanie Davies (doi:10.1136/bmj.e4514) believe that screening for diabetes is feasible and cost effective, but Elizabeth Goyer and colleagues think that a population approach to modifying risk factors would be better

The main evidence for the benefits of identifying people at increased risk of type 2 diabetes, as well as those who already have diabetes, comes from trials of early intervention and modelling. These suggest that screening followed by lifestyle interventions promoting changes in diet and physical activity is both effective and cost effective in people at increased risk.1 2 However, none of the models directly examines the relative benefits of these individualised approaches compared with population approaches. We suggest that screening has several disadvantages and that a mean population approach to risk reduction may be more appropriate.

Disadvantages of screening

Cost effectiveness modelling clearly shows the advantages of making management decisions on the basis of overall cardiovascular risk rather than on the presence or absence of individual cardiovascular risk factors such as hypertension.3 Recent modelling suggests that it may be as effective to treat on the basis of age alone as on the basis of a more complex risk assessment including blood pressure and cholesterol levels.4 Similarly, diabetes specific risk assessment and measurement of blood glucose may contribute little to management decisions. Changes to diet or physical activity levels will always be advisable for people who are overweight or sedentary, whatever their overall diabetes or cardiovascular risk score and whatever their glucose result.

Moreover, focusing on diabetes risk alone will underestimate the health risks for someone who is overweight but has a normal glucose level (notably those of cardiovascular disease and cancer)5 and ignores the additional benefits of weight loss and exercise (including the effect on psychological and social wellbeing).6 Given that the benefits of lifestyle intervention also depend on the extent to which someone can change their diet and physical activity patterns, a focus on diabetes risk will not always identify those who could benefit most.

Giving individuals an accurate assessment of their personal risk of diabetes may still be valuable because it could motivate behaviour change, but the evidence for this is equivocal. Evaluations of screening programmes that have measured psychological outcomes and behavioural intentions suggest that a negative screening result, carefully explained, will not cause “false reassurance”7 but that, equally, a personalised risk score doesn’t increase the chance of successful behaviour change.8

With an increasingly overweight and sedentary population and lower thresholds for intervention, the “population at risk” becomes the majority of the population. This makes the cost of individualised assessment and intensive interventions for everyone who could benefit an increasingly expensive option. It also makes a strategy that focuses on individual behaviour change even more likely to exacerbate health inequalities, as those with more resources and support will find it easier to change their lifestyle.9

Advantages of a population approach

The main interventions proposed for reducing the risk of diabetes among those at increased risk are increasing physical activity (and reducing sedentary activities) and dietary change (less sugar and fat; more fruit, vegetables, and fibre).1 Since these interventions would benefit the whole population and impact on a wide range of conditions beyond diabetes, it seems rational to consider interventions to facilitate population-wide behaviour change rather than focus on one risk group just because we can identify them.10

Moreover, given recent secular changes in our environment and habits that have led to even the average individual being overweight, a policy that focuses on interventions to reverse changes responsible for the increasing risk makes sense.11 An additional benefit of such population level interventions is that not only do they facilitate individual efforts to change behaviour but they may have other benefits in terms of environmental sustainability—for example, through promotion of active travel or more local food production.12 Regulation of the food and drink industry to create a level playing field for healthier products and more generally promoting the availability of healthier foods at reasonable prices might also be a promising place to start.13

Focusing on population approaches rather than individual risk assessment would free up health service resources for prevention of complications in people with clinical diabetes.14 For example, modelling of the current Health Check programme in England (screening every five years between the ages of 40 and 74 years, which includes diabetes screening as part of cardiovascular risk assessment) suggests 1600 vascular events and 650 deaths will be avoided annually.15 This is at a cost of around £30m (€37m; $47m) in direct payments to primary care providers (assuming 1.1 million people attend for a check each year at £25-£28 per check).16 17 Spending limited health service resources on a diabetes risk assessment and prevention programme reduces resources available for specialist management of these serious preventable diabetes complications. While, in the short term, the theoretical cost savings from reduced complication rates generated by screening programmes can rarely be realised in practice, the real savings generated by not screening could immediately be spent on diabetes patients’ care.

The striking variations in obesity rates between communities18 offer hope that increasing obesity and diabetes prevalence are not inevitable. Screening is a relatively expensive investment, targeting only a proportion of those who could benefit, whereas population level interventions hold promise in terms of potentially both reducing risk and reducing health inequalities.


Cite this as: BMJ 2012;345:e4516


  • Competing interests: All authors have completed the ICMJE unified disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: LI is part funded by the Australian National Health and Medical Research Council (grant 633003); EG and NP are currently both part funded by a National Institute for Health and Clinical Excellence (NICE) contract for evidence reviews and economic modelling to inform forthcoming NICE public health guidance on preventing type 2 diabetes in people at high risk.

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