Electronic health record identification of prediabetes and an assessment of unmet counselling needs
Version of Record online: 20 JUN 2011
© 2011 Blackwell Publishing Ltd
Journal of Evaluation in Clinical Practice
Volume 18, Issue 4, pages 861–865, August 2012
How to Cite
Zimmermann, L. J., Thompson, J. A. and Persell, S. D. (2012), Electronic health record identification of prediabetes and an assessment of unmet counselling needs. Journal of Evaluation in Clinical Practice, 18: 861–865. doi: 10.1111/j.1365-2753.2011.01703.x
- Issue online: 2 JUL 2012
- Version of Record online: 20 JUN 2011
- Accepted for publication: 25 March 2011
- diabetes prevention;
- electronic health record;
- impaired fasting glucose;
- lifestyle modification;
Rationale, aims and objectives Large clinical trials demonstrate that lifestyle modification can prevent or delay the onset of diabetes in those with prediabetes. However, recent National Health and Nutrition Survey data suggest that prediabetes often goes unrecognized, and the majority of prediabetic individuals do not report having received lifestyle advice from physicians. We explored whether electronic health record (EHR) query of glucose measurements can identify prediabetic patients, and we estimated rates of prediabetic lifestyle counselling in a large, urban, primary care practice.
Methods Electronic search identified patients with plasma glucose levels of 100 to 199 mg dL−1 between 1 June 2007 and 1 June 2009, excluding those with diabetes or diabetic medications/supplies. From these 5366 patients, 100 randomly selected patients underwent classification into provisional categories based on available EHR data: likely prediabetes, likely diabetes, glucose abnormality in the setting of acute illness, or normal glucose metabolism. In those likely to have prediabetes, we assessed lifestyle modification counselling.
Results Fifty-eight per cent (95% CI 48% to 68%) of patients sampled were likely to have prediabetes. Fourteen per cent of those sampled were likely to have diabetes. Thirty-one per cent of prediabetics (95% CI 22% to 42%) had documented lifestyle counselling. Counselled patients had a significantly higher baseline mean body mass index compared to those not counselled (34.1 versus 29.9, P = 0.037).
Conclusions EHR query using glucose measurements can identify prediabetic patients and those requiring further glucose metabolism evaluation, including those with undiagnosed diabetes. Future research should investigate EHR-based, population-level interventions to facilitate prediabetes recognition and counselling.