Screening for diabetes

BMJ 1994; 308 doi: http://dx.doi.org/10.1136/bmj.308.6937.1160b (Published 30 April 1994) Cite this as: BMJ 1994;308:1160
  1. Melanie Davies,
  2. John Day
  1. Senior registrar, Leicester Royal Infirmary, Leicester LE1 5WW Consultant physician, Ipswich Hospital, Ipswich IP4 5PD.

    EDITOR,--Baldev Singh and colleagues reported the results of the British Diabetic Association's study on the effects of advertising on the general public's awareness of diabetes, and they suggested that this approach should be further evaluated as a means of achieving earlier diagnosis of non-insulin dependent diabetes mellitus.1 We have doubts about this conclusion.

    One of the reasons for these conclusions was that the approach could be cost effective (pounds sterling 1000 per new case of diabetes). This method has a high false positive rate and a low specificity, as of those presenting to the general practitioner, only 17 (17%) were found to have diabetes. The cost does not compare well with other methods and presumably does not include the cost of performing diagnostic tests on the 82 patients without diabetes.2

    In terms of achieving an earlier diagnosis of diabetes, the 10 week campaign revealed an additional 17 cases of diabetes. The total population aged 15-75 years subjected to the campaign was approximately 300500 according to the 1991 population census.3 It has often been assumed that there are as many undiagnosed cases of diabetes as there are diagnosed, and if we assume an overall prevalence of diabetes of 1% (probably an under-estimate in this age group) then we would expect that over 3000 cases of diabetes remain undetected. The ability of this campaign to identify only a predicted 22 (0.7%) of these subjects must be disappointing.

    In contrast, we have shown in a large study based in general practice that a postal request system with self testing for postprandial glycosuria is an effective method of screening for diabetes.2 In a target population of 13795 subjects aged 45-70 years, 99 new cases of diabetes were identified by this method at a cost of pounds sterling 81 per new case (1990 prices). We have proposed ways in which this cost could be reduced.2 We have since shown that this method can be effectively repeated after 30 months, suggesting that public cooperation in repeated testing for diabetes can be maintained.4

    The current study, as a means of raising public awareness regarding diabetes, seemed to achieve its aim; it may also have been beneficial in terms of reducing anxiety about symptoms of diabetes. However, it does not seem effective in making an early diagnosis of--that is, screening for--non-insulin dependent diabetes. When other more effective and inexpensive methods based in general practice have already been shown to have promising results, stretched resources should firstly be channelled into these approaches.5


    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
    Sign up for a free trial