Revisiting Screening for Type 2 Diabetes Mellitus: To Screen or Not to Screen, That Is the Question

To the Editor: We read with interest the commentaries on screening for diabetes by Sheehy et al¹ and Lawler² in the January 2009 issue of Mayo Clinic Proceedings. Sheehy et al debated the recent US Preventive Services Task Force (USPSTF) guidelines³ that diabetes screening should be done only in adults with blood pressure (BP) higher than 135/80 mm Hg compared with the broader recommendations of the American Diabetes Association. Lawler provided reasons to exercise caution in screening for diabetes.

We think it is ironic that screening for diabetes is still being debated when the disease is reaching epidemic proportions with its attendant serious economic and health consequences. Although it is possible to delay or prevent diabetic complications with timely treatment, diabetes must first be diagnosed because 1 or more of these complications may be present at the time of diagnosis. Thus, the call should be for more liberal, not restrictive, screening guidelines.

Sheehy et al provided “10 good reasons” to advocate for broader diabetes screening, arguing that the criteria used by the USPSTF were more restrictive than those used for obesity. Lawler voiced reservations about broader screening, raising concerns about evidence, guideline recommendations, competing priorities, and testing errors.

Lawler emphasizes that neither argument (the one by the USPSTF or by Sheehy et al) is predicated on evidence-based medicine, ie, randomized controlled trials, simply because designing such studies would be unethical. Thus, he advocated switching efforts from screening for diabetes to focusing on lifestyle modifications but admits that recidivism is common and long-term success is unusual. If physicians are already struggling to help patients with (established) diabetes achieve this goal, it is unlikely that this recommendation would work.

A reasonable argument against the USPSTF's new recommendations is the established association among the 5 components (not only hypertension) of the metabolic syndrome (MetS) and macrovascular complications. A recent study (N=14,993) found that the risk of incident stroke is incrementally increased with each additional component of MetS.⁴ The greatest risk of developing stroke was found in individuals with MetS with either elevated BP or fasting plasma glucose (ie, either component and not necessarily both). Therefore, limiting screening to adults with elevated BP would miss those at high risk (ie, those with MetS) who do not yet have this component.

Finally, Lawler's other concern was disease mislabeling (ie, diagnosing diabetes when a patient does not have the disease), which may result in psychological or financial (insurance) consequences. This was argued by Vinicor⁵ a decade ago: “It is important not to tell a person that diabetes is present when it is not; likewise, to not identify diabetes when it exists is to ultimately deny benefits of improved metabolic control.”

In conclusion, we agree with Sheehy et al in advocating the broader American Diabetes Association recommendations for diabetes screening. We think that the USPSTF's recommendations are too restrictive and that wider screening should be under-taken. Perhaps other interested organizations, such as the American Association of Clinical Endocrinologists and American College of Physicians, will address this issue.