Random capillary whole blood glucose test as a screening test for diabetes mellitus in a middleaged population

Abstract

This study was aimed at assessing the adequacy of a random capillary whole blood glucose (RCBG) test as a screening test for diabetes mellitus in a middle-aged (55-year-old) Finnish population. Both the screening test (RCBG) and a standard 2-h oral glucose tolerance test (OGTT) were performed on all the participants according to the WHO criteria. The prevalence based on the 2-h OGTTs was 8.5% (11.7% for men and 6.0% for women). The sensitivity, specificity and predictive values of the screening test for diabetes and the receiver operator characteristic (ROC) curves were studied. The likelihood ratios for diabetes within the test categories were also calculated. When a cut-off point of 6.2 mmol l-1 was used, which proved to be the most appropriate in the present study, the specificity was 92% and the sensitivity 63% (79% for men and 40% for women). The men had much higher sensitivities than the women at all cut-off points, but the specificities were comparable. The present study indicated that the RCBG test could be used as a screening test in a subpopulation with a high prevalence of diabetes and severe hyperglycaemia, but it would be quite insensitive in a subpopulation with a considerable overlap of RCBG values between the diabetic group and the non-diabetic one. It would be inappropriate to use the RCBG test as a routine screening test for diabetes in a general population, particularly among subjects with a known low prevalence of diabetes.

Screening for Type 2 diabetes—should it be now?

“…The optimal cut‐off ranged from 4.4 to 6.7 mmol/l (80–120 mg/dl), but still the sensitivity ranged from 67% to 83% and specificity from 65% to 80%. Similar results have been reported from other groups [17]. Due to these problems with standardization and the relatively low sensitivity and specificity, random blood glucose alone is of limited value as a screening test for diabetes.…”

“… Prediction may be improved by the use of a risk score [17–21], but the risk score should be tested and validated within the specific population. …”

Hypertension, hyperinsulinaemia and obesity in middle-aged Finns with impaired glucose tolerance

“…The details of the initial survey and the follow-up study have been reported previously. [14][15][16] A postal questionnaire concerning the previous history of diabetes, hypertension, anti-hypertensive medication and other questions was sent to each of the participants, and the answers to the questions and the anti-hypertensive agents used previously were checked during the interviews.…”

Testing for type 2 diabetes in Indigenous Australians: guideline recommendations and current practice

“…Two studies have examined this and performed an OGTT in the whole population irrespective of the RCG result. Qiao et al (1995) studied 1,008 people and using a cutoff level for RCG of 5.8 mmol/L achieved a sensitivity of 79% in men but only 40% in women, while specificity was 86% and 84%, respectively, for men and women. The authors Case Detection and Diagnosis, concluded that RCG is too insensitive to use for routine screening for diabetes in a general population, particularly in populations with a known low prevalence of diabetes.…”