Screening for Diabetes

See also: Insulin Resistance, Impaired Glucose Tolerance and Gestational Diabetes.

Our affluent and sedentary society has led to an increasing prevalence of obesity. This has caused an increase in the relative frequency of diabetes in the general population, particularly Type 2 diabetes in older adults. There may also be an increasing incidence of Type 1 diabetes, but this is uncertain and the cause(s) for such an increase are unclear. However, Type 1 diabetes tends to present symptomatically in a relatively short space of time as insulin reserves are exhausted and significant hyperglycaemia develops, thus it is of less concern in terms of a target for screening.

Type 2 diabetes, in contrast, can remain hidden and asymptomatic for long periods (many years), either as the pre-diabetic 'metabolic syndrome'/impaired glucose tolerance (IGT), or mild-to-moderate frank Type 2 diabetes. Recent evidence shows that many people with Type 2 diabetes have the condition for 9–12 years before diagnosis.¹ Given that chronic damage to many of the body's organ systems can occur during this time, and that many patients presenting with Type 2 diabetes already have complications, or may present with life-threatening metabolic derangements such as hyperglycaemic, hyperosmolar, non-ketotic coma (HONK), screening for Type 2 diabetes should be an effective way to reduce the burden of this disease in the appropriately selected population.

Screening of an older, predominantly white, socially representative cohort of patients enrolled in population-based heart disease studies has revealed a prevalence of frank undiagnosed Type 2 diabetes in this group of around 7%. IGT had a prevalence of about 20%, ie 1 in 5 of this population.² Thus it appears there is a relatively urgent need to detect this at-risk group and diagnose and ameliorate their condition before irreversible complications set in. There is, as yet, no definitive evidence that such a policy will lead to better outcomes in terms of mortality and morbidity, but intuitively it seems a good bet and worthy of formal investigation.

There is good evidence that an appropriately designed and targeted screening strategy is effective at detecting undiagnosed Type 2 diabetics in a UK-based primary care setting, with the number of patients needing to be screened to detect one case of Type 2 diabetes, or impaired fasting glucose, being relatively low at 7–13.^{3 4} Undiagnosed diabetes rates were about 20% of those already diagnosed.³ The Department of Health has announced a series of pilot programs based in inner-city GP practices to assess the real-world effectiveness of such screening strategies.⁵ It is likely to take many years or decades of prospective research and follow-up to confirm improved population outcomes as a result of such screening. This does not preclude individual practitioners from instituting their own strategies to detect the disease in those at high risk in the meantime. Indeed, it is likely to be a highly satisfying form of screening, in that it is virtually certain that undiagnosed individuals will be detected and start therapy or lifestyle alterations that one would consider in general terms to be likely to enhance their long-term prospects of continuing health.

There are no agreed hard-and-fast criteria for selection of the screening population. Screening on the basis of age alone has been shown to have a low yield.⁶ Most studies have used some or all of the following criteria:

Again there are no definitely agreed protocols as yet. Fasting venous plasma glucose measurements are usually used as the primary measure, although it may be sufficient, and certainly cheaper/quicker, to perform fasting capillary sampling, but there is no good evidence available on this yet. The large UK-based study⁴ used the following results interpretations:

Type 1 or Type 2?

It should not be assumed that all those detected by screening have Type 2 diabetes. There are many cases of people who develop Type 1 diabetes in their 30s, 40s, or even older (the author of this article included, which caused my GP practice-nurse no end of confusion, based upon the concept of mature-onset diabetes mellitus, an outmoded and useless term). In population screening, such patients will be detected occasionally and should not be lumped in with Type 2 diabetics, just because they were detected by screening and not illness. Clinical factors such as the degree of hyperglycaemia, symptoms associated with both Type 1 and Type 2 diabetes, BMI, family history and other biochemical markers should be used to decide on the Type 1/Type 2 divide. Admittedly, the vast majority of patients detected by screening will have Type 2 diabetes, but this is not a sine qua non. If unsure, refer the patient for specialist diabetic advice.

IGT detected by screening

Patients with IGT should have the nature of their condition explained to them and be made aware that if they do not alter their lifestyle they have a reasonable chance of progressing to Type 2 diabetes. Diet modification, increased exercise and modification of risk factors for cardiovascular disease are the mainstays of management of this condition. See separate article on managing impaired glucose tolerance in primary care.

Type 2 diabetes detected by screening

Patients with Type 2 diabetes should be managed as for all patients with this condition. Community- and hospital outpatients-based follow up, appropriate dietary and pharmacological treatment, along with cardiovascular risk factor assessment and modification, are the most important therapeutic avenues. See separate article on managing diabetes in general practice, and other Type-2 diabetes-related articles.