It is known that there is a world epidemic of diabetes and projections have been made that there will be 2.9 million people with diabetes by the year 2010 (Amos et al 1997). As will be clear from Chapter 1, about 80% of these people will have type 2 diabetes, which presents a major challenge to any healthcare system.

Most healthcare professionals would have no difficulties in diagnosing diabetes in a person presenting with the classic symptoms of type 1 diabetes of polyuria, polydipsia, weight loss and tiredness. In such people, the diagnosis is easily confirmed by a clearly elevated blood glucose level confirmed by laboratory testing (Box 2.1) and appropriate treatment commenced. Similarly, it is easy to identify people who clearly do not have diabetes by finding a random plasma glucose below 5.5 mmol/L. Problems most frequently arise when a person is found to have glucose levels between 7.0 mmol/L and 11.1 mmol/L and does not complain of any of the usual signs and symptoms of diabetes.

Diagnosing type 2 diabetes is often more problematic because the onset is very gradual and a person can have diabetes for many years before presenting with complications of the disease (Harris et al 1992, United Kingdom Prospective Study Group (UKPDS) 1998). Hence the challenge to the healthcare team is to identify people with type 2 diabetes before they present with complications of the disease.

Undiagnosed diabetes is not a benign condition. It carries with it costs to the individual – through the onset of complications – and costs to the healthcare system in the treatment of these. Therefore, to diagnose type 2 diabetes prior to complications arising seems a sensible way forward. Coupled with this, the costs of managing people with diabetes globally can be measured in clinical costs, indirect costs and the costs to carers on looking after people with diabetes (WHO 2004). Costs to the UK NHS vary between 8.7% of acute sector costs (Currie et al 1997) to 4.1% for type 2 diabetes alone (Williams et al 2001).

Although the prevalence of type 2 diabetes increases with age, it is now recognised in teenagers (Peters & Davidson 1992, Sinha et al 2002). The UKPDS (1998) found that up to 50% of people presented at diagnosis of diabetes with established complications, particularly retinopathy, indicating that they had diabetes for as long as 12 years. Diabetes remains the leading cause of blindness in working age adults, of end-stage renal disease and of non-traumatic lower extremity amputations; in addition it produces a 2- to 4-fold increase in cardiovascular risk. It is also recognised that mortality rates are higher in the population of people with diabetes whether diagnosed with diabetes or not than in the non-diabetic population (Jarrett & Shipley 1988). Hence the morbidity associated with type 2 diabetes is costly for the individuals concerned.

A random plasma glucose equal to, or exceeding, 11.1 mmol/L is usually indicative of diabetes but a diagnosis should not be made on the basis of only one abnormal finding in the absence of clinical symptoms (WHO 1999). Fiona will be informed that she has diabetes if she has one additional glucose test on another day with a value in the diabetic range (Box 2.2). It is strongly recommended that only plasma venous samples are used to diagnose diabetes (Table 2.1).

Fiona’s case study is not uncharacteristic, as people usually present in primary care with other associated medical conditions. The primary healthcare team (PHCT) should be alert to the possibility of undiagnosed diabetes hindering a person’s anticipated recovery, e.g. the slow-to-heal leg ulcer. Likewise, members of the PHCT attending people with a sore toe, vaginal or penile thrush, blurred vision, boils and carbuncles or complaints of weight loss must be proactive in their role to assist in the early detection of diabetes. The PHCT should be aware of the risk factors for developing type 2 diabetes that are frequently associated with undiagnosed diabetes (Box 2.3). These risk factors are only pertinent to the person with type 2 diabetes.

The oral glucose tolerance test (OGTT) remains the definitive standard for diagnosing diabetes despite the concerns expressed regarding its poor reproducibility and the fact that it is subject to a wide variety of influences (Yudkin et al 1990). The OGTT is still recommended if there is any doubt about the diagnosis (Vaccaro et al 1999, WHO 1999). As a test, however, it is time consuming for the person, can be rather unpleasant and can be quite inconvenient for both people and staff.

It should be remembered that the OGTT is conducted only where there is an equivocal blood glucose result in the absence of any symptoms. It is usually performed within general practice; those hospitals involved in administering the test do so on an outpatient basis. Educating the person before the test is essential as several factors can influence its reproducibility (Box 2.4). This is usually done by nursing staff and can be reinforced by literature. It is important that the person knows who to contact with any queries regarding the test as it can be seen that many factors can influence the result.