Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.
Skip to main content

ABSTRACT Background: In the US, it is estimated that up to 25 percent of the diabetes population may be undiagnosed, adding an additional $18 billion to the already high $174 billion in annual costs for diabetes care. Objectives: To assess in primary care practices in the US: 1) the extent of undiagnosed diabetes; and 2) the clinical management of patients with known diabetes, based on the frequency of meeting selected indicators of diabetes care quality and comparing these outcomes to results in England. Design: Retrospective descriptive analysis of clinical and laboratory data derived from a nationally representative database of electronic medical records (GE Centricity) of patients attending primary care practices in the US in 2008/09. Three modifications of a simple algorithm were tested for the first objective, and 16 quality indicators were selected to test for the second. Key Results: Records from 10,430,056 non-diabetic patients were assessed for fasting and random glucose values diagnostic of diabetes. Regarding Objective 1, at least two abnormal values were present in 0.39 percent (n=40,359), and of the remaining patients, 0.22 percent (n=23,261) had at least one documented HbA1c >6.5. Regarding Objective 2, of 622,260 patients with known diabetes for at least 15 months, a sampling of findings show: 59 percent had HbA1c testing in that interval; 34 percent were at goal (<7 percent); BMI was recorded in 57 percent; blood pressure was recorded in 88 percent, with 68 percent <145/85 mmHg; 33 percent had urine checked for micro-albuminuria, 69 percent had a serum creatinine recorded. Regional differences in prevalence of undiagnosed diabetes and satisfaction of indicators of care quality were noted. Conclusions: It is feasible to apply a simple algorithm to a nationwide database of electronic medical records to identify patients with potentially undiagnosed diabetes. An analysis of quality of care indicators in patients with known diabetes suggests a wide gap exists between recommended and actual practice.

Original publication




Journal article



Publication Date



60(Suppl 1)


A633 - A633