Clinical characteristics and follow-up of subjects with non-insulin-dependent diabetes mellitus diagnosed by screening

Diabetes risk score: towards earlier detection of type 2 diabetes in general practice

BACKGROUND

Type 2 diabetes is common, costly and often goes unrecognised for many years. When patients are diagnosed, the majority exhibit associated tissue damage or established cardiovascular risk. Evidence is accumulating that earlier detection and management of diabetes and related metabolic abnormalities may be beneficial. We aimed to develop and evaluate a score based on routinely collected information to identify people at risk of having undetected diabetes.

METHODS

A population-based sample of 1077 people, aged 40 to 64 years, without known diabetes, from a single Cambridgeshire general practice, underwent clinical assessment including an oral glucose tolerance test. In a separate 12-month study, 41 practices in southern England reported clinical details of patients aged 40 to 64 years with newly diagnosed Type 2 diabetes. A notional population was created by random selection and pooling of half of each dataset. Data were entered into a regression model to produce a formula predicting the risk of diabetes. The performance of this risk score in detecting diabetes was tested in an independent, randomly selected, population-based sample.

RESULTS

Age, gender, body mass index, steroid and antihypertensive medication, family and smoking history contributed to the score. In the test population at 72% specificity, the sensitivity of the score was 77% and likelihood ratio 2.76. The area under the receiver-operating characteristic curve was 80%.

CONCLUSIONS

A simple score, using only data that are routinely collected in general practice, can help identify those at risk of diabetes. This score could contribute to efficient earlier detection through case-finding or targeted screening.

Screening for Type 2 diabetes mellitus in the UK Indo-Asian population

AIMS

Type 2 diabetes mellitus (DM) has a high prevalence in Asian subjects. A simple method of screening using self-testing for postprandial glycosuria achieved a good response rate and a sensitivity which compared favourably to more expensive and invasive methods in a semirural Caucasian population. We examined its effectiveness in Asian subjects.

METHODS

Caucasian and Indo-Asian subjects aged 35-70 years in two general practices in Leicester (n=9896 (6198=Asian subjects, 3698=Caucasian)) were screened. Those known to have diabetes were excluded. Subjects were asked to self-test for glycosuria 1 h after their main meal. Instruction and response cards were translated in Punjabi and Gujarati and sent to the Asian subjects, depending on age and surname.

RESULTS

Response rate was 34.4% in Asian subjects compared to 54.0% in Caucasian subjects. Prevalence of glycosuria was 8.2% in Asian subjects and 3.2% in Caucasian subjects. Two hundred and thirty-nine subjects recorded glycosuria and 202 (84.5% of the total, 86.9% of Asian subjects, 78.1% of Caucasian) attended for oral glucose tolerance test (OGTT). Sixty-three (31.2%) were found to have diabetes (46, 73% Asian), 29 (14.4%) impaired glucose tolerance (24, 82.8% Asian) and 110 (54.4%) normal glucose tolerance (82, 74.6% Asian). Thus 30% of Asian subjects and 34% of Caucasian subjects had diabetes on OGTT. The prevalence of diabetes in 35-70 years in the total population after screening was 5.6% (6.8% in Asian subjects, 3.6% in Caucasian) and in the screened population was 12.7% (17.9% in Asian subjects, 6.3% in Caucasian).

CONCLUSIONS

Screening for diabetes using this method, in terms of response rate, is not as effective in a large city setting, particularly in the Asian population. However, the yield of diabetes in the age group 35-64 years compares well to much more expensive and labour intensive approaches and its use in this population in a primary care setting is justified.