Screening for Type 2 Diabetes > Effectiveness/Harms of Screening Tests Author: Lorraine Lipscombe, MD, FRCPC; Denice S. Feig, MD, MSc, FRCPC
Editorial changes - 2009-05-01
Author information and module status
Key Points
Population at Risk
Effectiveness/Harms of Screening Tests
Effectiveness/Harms of Early Treatment
Direct Evidence that Screening Reduces Adverse Outcomes
Timeline
Cost-Effectiveness
Patient Education
Referral/Consultation
Guidelines
Tables
References
Glossary
What's New
Patient Information
Additional Resources
Tools
Recommendation
Recognize that the HbA1c test has good specificity but only moderate sensitivity for detecting undiagnosed diabetes, and its performance varies depending on the population and cutpoint used.B

Evidence:

  • Results from NHANES III, a survey that included 7832 U.S. adults who underwent testing for diabetes, found that an HbA1c value of >=5.6% (1 SD above the mean) was 83.4% sensitive and 84.4% specific for detecting undiagnosed diabetes. This specificity increased to 97.4% and the sensitivity decreased to 63.2% for an HbA1c value of >=6.1% (2 SD above the mean) (63).
  • In a study of 936 Canadians of European, South Asian, and Chinese origin, an HbA1c value of >=5.9% had a sensitivity of 75.0% and a specificity of 79.1% for detection of diabetes. The specificity increased to 95% but the sensitivity decreased to 71.7% when this was combined with an FBG of >=103 mg/dL (5.7 mmol/L) (62).
  • A meta-analysis of 18 studies comprising 8984 subjects found that a HbA1c >=4 SD above normal was associated with a sensitivity of 36% and a specificity of 100% for a diagnosis of diabetes (64).

Comments:

  • International variability in HbA1c assays had previously precluded its validation for diagnosis of diabetes, but the National Glycohemoglobin Standardization Program has now made the measurement of HbA1c more precise for use in studies and clinical practice (65).

FAQs
Denice S. Feig, MD, MSc, FRCPC has no financial relationships with pharmaceutical companies, biomedical device manufacturers, or health-care related organizations. Lorraine Lipscombe, MD, FRCPC has no financial relationships with pharmaceutical companies, biomedical device manufacturers, or health-care related organizations. Sonal Singh, MD, editorial consultant, has no financial relationships with pharmaceutical companies, biomedical device manufacturers, or health-care related organizations.
Steven E. Weinberger, MD, FACP, Acting Editor, PIER, has stock holdings in Glaxosmithkline and Abbott.

The information included herein should never be used as a substitute for clinical judgment and does not represent an official position of ACP. Because all PIER modules are updated regularly, printed web pages or PDFs may rapidly become obsolete. Therefore, PIER users should compare the date of the last update on the website with any printout to ensure that the information being referred to is the most current available.
PIER is copyrighted (c) 2009 by the American College of Physicians,
190 N. Independence Mall West, Philadelphia, PA 19106-1572, USA.