Am Fam Physician. 2010;82(6):684-686
Background: According to 2005–2006 National Health and Nutrition Examination Survey data, the national prevalence of diabetes mellitus in ambulatory patients 20 years and older was an estimated 12.9 percent. Approximately 40 percent of these persons are unaware they have the disease, meaning 5.1 percent of U.S. adults 20 years and older have diabetes but do not know it. In recent years, type 2 diabetes has been confirmed to have the “legacy effect” associated with type 1 diabetes, which is defined as worsened cardiovascular morbidity and mortality after a period of untreated hyperglycemia, even if blood glucose levels are controlled at a later time. Sheehy and colleagues assessed the American Diabetes Association (ADA) and the U.S. Preventive Services Task Force (USPSTF) diabetes screening guidelines to determine their case-finding ability.
The Study: The authors analyzed electronic health records from a midwestern group practice from January 1, 2005, through December 31, 2007. The practice treated about 2 million patients in 48 million encounters since implementing electronic health records in 2003. Patients were included if they were 20 years and older on January 1, 2005; had seen their primary care physician for any reason at least twice in the previous 36 months, with one visit being within the previous 24 months; did not have a diagnosis of diabetes, prediabetes, or preexisting comorbidities in 2003 or 2004; did not have a visit for pregnancy between 2003 and 2007; and did not die during the study years. Among patients in the study who had one or more screening tests, 74.3 percent had random glucose testing; 9.1 percent, A1C measurement; 0.8 percent, fasting plasma glucose assessment; and 0.8 percent, glucose tolerance testing. ADA and USPSTF criteria for screening are shown in the accompanying table. The authors used pre-2008 USPSTF guidelines during the study; additionally, they estimated how the newer 2008 USPSTF guidelines would compare in the same population by excluding hyperlipidemia as an inclusion criterion for patients who met the pre-2008 USPSTF criteria. The authors also studied how the number and type of high-risk factors affected the results, as well as if insurance status affected the rates of screening.
| American Diabetes Association | |
| Testing should be considered in all adults who are overweight (body mass index ≥ 25 kg per m2) and have additional risk factors: | |
| Physical inactivity | |
| First-degree relative with diabetes | |
| Members of high-risk ethnic populations | |
| Women who delivered a newborn weighing > 9 lb (4.1 kg) or were diagnosed with gestational diabetes | |
| Hypertension | |
| High-density lipoprotein cholesterol < 35 mg per dL (0.91 mmol per L) or triglyceride level > 250 mg per dL (2.82 mmol per L) | |
| Women with polycystic ovary syndrome | |
| Impaired glucose tolerance or impaired fasting glucose on previous tests | |
| Other clinical conditions associated with insulin resistance | |
| History of cardiovascular disease | |
| In the absence of the above criteria, testing for diabetes and prediabetes should begin at 45 years of age | |
| If the results are normal, testing should be repeated at least at three-year intervals, with consideration of more frequent testing dependent on initial results and risk status | |
| Pre-2008 U.S. Preventive Services Task Force | |
| Screening for type 2 diabetes is recommended in adults with hypertension or hyperlipidemia | |
| Evidence is insufficient to recommend for or against routinely screening asymptomatic adults for type 2 diabetes, impaired glucose tolerance, or impaired fasting glucose | |
| 2008 U.S. Preventive Services Task Force | |
| Screening is recommended for asymptomatic adults with sustained blood pressure > 135/80 mm Hg | |
| No recommendation for asymptomatic adults with blood pressure ≤ 135/80 mm Hg | |
Results: The study included 46,991 patients, of whom 59.4 percent were women and 99.5 percent were insured. A total of 33,823 patients met at least one of the three screening criteria (ADA criteria, 65.5 percent; pre-2008 USPSTF criteria, 58.0 percent; and 2008 USPSTF criteria, 25.6 percent). Of patients meeting the ADA criteria, 26,597 (86.4 percent) were screened, and 1,329 (5.0 percent) were diagnosed with diabetes. Among those meeting the pre-2008 USPSTF criteria, 24,221 (88.9 percent) were screened, and 1,293 (5.3 percent) were diagnosed with diabetes. Of those meeting the 2008 USPSTF criteria, 11,333 (94.0 percent) were screened, and 869 (7.7 percent) were diagnosed with diabetes. Of all patients who were eligible and tested by any criteria, 1,390 patients (4.8 percent) had newly diagnosed diabetes. Of those patients, the ADA screening criteria missed 4.4 percent, the pre-2008 USPSTF screening criteria missed 7.0 percent, and the 2008 USPSTF screening criteria missed 37.5 percent of patients who could have been diagnosed.
When individual high-risk factors were evaluated, prediabetes (15.8 percent), polycystic ovary syndrome (12.6 percent), and vascular disease (10.0 percent) indicated the highest rates of newly diagnosed diabetes. The authors' database did not include physical inactivity, family history of diabetes, or other conditions associated with insulin resistance, so they were unable to include patients with these possible risk factors in their analysis. The number of high-risk factors strongly correlated with a diagnosis of diabetes. Patients without insurance were less likely to be screened (54.9 versus 85.4 percent with insurance), but were more likely to be diagnosed with diabetes (14.0 versus 4.8 percent of the insured population).
Conclusion: The authors conclude that the ADA screening guidelines identify a significantly higher number of patients eligible for screening, and have an increased case-finding ability compared with the 2008 USPSTF guidelines. The number of risk factors and presence of certain high-risk factors can help determine prediabetes status. Patients without health insurance are screened significantly less often than those who are insured.
editors note: It is useful to note the rationale behind the 2008 USPSTF guidelines for screening adults for type 2 diabetes. The USPSTF concluded that there is no direct evidence suggesting that health outcomes for patients are altered when screening asymptomatic persons.1 Additionally, the indirect evidence they reviewed did not show a benefit for screening general populations. The justification for eliminating hyperlipidemia from the 2008 screening guidelines includes a randomized controlled trial and a meta-analysis, both of which concluded that patients with hyperlipidemia benefit equally from lipidlowering agents regardless of whether they have diabetes. 2,3 This is in contrast to patients with hypertension, because it has been demonstrated that they experience a greater benefit from more aggressive blood pressure control when they have a clinical diagnosis of diabetes.3,4 Whereas intensive lifestyle modifications and some pharmacologic interventions have been shown to decrease the rate of progression from prediabetes to diabetes, the evidence of their impact on long-term health outcomes is lacking.1 Despite their differing recommendations, the USPSTF and ADA guidelines should serve as tools to assist physicians in making a collaborative decision about diabetes screening with their patients.—C.H. and SUMI SEXTON, MD, Associate Medical Editor, American Family Physician
