Final Recommendation Statement

Abnormal Blood Glucose and Type 2 Diabetes Mellitus: Screening, 1996

January 01, 1996

Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

This topic is being updated. Please use the link(s) below to see the latest documents available.

Recommendation Summary

Population Recommendation Grade
Nonpregnant adults There is insufficient evidence to recommend for or against routine screening for non-insulin-dependent diabetes (NIDDM) in nonpregnant adults. C
Pregnant women There is also insufficient evidence to recommend for or against routine screening for gestational diabetes mellitus (GDM). C
Asymptomatic adults Screening with immune markers to identify asymptomatic individuals at risk for developing insulin-dependent diabetes mellitus (IDDM) is not recommended in the general population. D

Full Recommendation:

Recommendations made by the USPSTF are independent of the U.S. government. They should not be construed as an official position of the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services.

Expand All

Approximately 14 million persons in the U.S. have diabetes mellitus.1 Non-insulindependent diabetes mellitus (NIDDM) or Type II diabetes accounts for 90-95% of all cases of diabetes in the U.S., while insulin-dependent diabetes mellitus (IDDM) or Type I diabetes accounts for the remaining 5-10%.1-4 An estimated half of all persons with diabetes (primarily patients with NIDDM) are currently unaware of their diagnosis.2 Diabetes may cause life-threatening metabolic complications, and is the seventh leading cause of death in the U.S., contributing to roughly 160,000 deaths each year.1,3 It is also an important risk factor for other leading causes of death such as coronary heart disease and cerebrovascular disease.4 Diabetes is the most common cause of polyneuropathy, with approximately 50% of diabetics affected within 25 years of diagnosis,5 and is responsible for over 50% of the 120,000 annual nontraumatic amputations in the U.S.6 Diabetic nephropathy is now the leading cause of end-stage renal disease in the U.S.7 and, if current trends continue, will soon account for 50% of all patients with renal failure.8 Diabetes is the leading cause of blindness in adults ages 20-74 and accounts for over 8,000 new cases of blindness each year.9 Infants born of diabetic women are at increased risk of fetal malformation, prematurity, spontaneous abortion, macrosomia, and metabolic derangements.10,11 Compared to persons without diabetes, diabetic patients have a higher hospitalization rate, longer hospital stays, and increased ambulatory care visits.3,12 The total annual economic burden of diabetes is believed to approach $100 billion in the U.S.13

The onset of NIDDM is usually after age 30, and the prevalence steadily increases with advancing age. It is estimated that nearly 20% of the U.S. population aged 65-74 has diabetes.2 The prevalence of NIDDM is markedly increased in Native Americans and is also higher among black and Hispanic populations.3 The prevalence of NIDDM is greater than 70% in Pima Indians 55 years of age and older.14 Other risk factors for diabetes include family history, obesity, and a previous history of gestational diabetes or impaired glucose tolerance. IDDM has an earlier onset (usually before age 30), a much shorter asymptomatic period, and a more severe clini cal course than NIDDM.

Gestational diabetes mellitus (GDM), the development of glucose intolerance during pregnancy, occurs in 3-5% of all pregnancies and is the most common medical problem of pregnancy.3,15 Risk factors for GDM include obesity, increased maternal age, hypertension, glucosuria, a family history of diabetes, and a history of a macrosomic, stillborn, or congenitally malformed infant. GDM is a risk factor for fetal macrosomia and is associated with other neonatal complications, such as hyperbilirubinemia and hypoglycemia. Macrosomia – most commonly defined as birth weight above 4,000 or 4,500 g – is not itself a morbid condition but is associated with increased risk of operative delivery (cesarean section or vacuum or forceps delivery) and birth trauma (e.g., clavicular fracture, shoulder dystocia, and peripheral nerve injury).16-19 In some series, the incidence of shoulder dystocia in infants over 4,000 g is close to 2%.20 Women with a history of GDM are also at increased risk for developing NIDDM later in life.21

Return to Table of Contents

The diagnosis of diabetes in many nonpregnant patients is based on typical symptoms (polyuria, polydypsia) in association with clear elevation of glucose (fasting plasma glucose > 140 mg/dL [7.8 mM]). Many asymptomatic persons, however, may have abnormal glucose metabolism and be at increased risk for complications of diabetes.

Diagnosis of Diabetes in Asymptomatic Persons

The National Diabetes Data Group (NDDG) 22 and World Health Organization (WHO)23 have issued similar criteria for diagnosing diabetes in asymptomatic persons, based on elevated fasting plasma glucose (>140 mg/dL) or an abnormal plasma or serum glucose following an 75 g oral glucose tolerance test (OGTT). NDDG criteria for a positive OGTT (> 200 mg/dL at 2 hours and before 2 hours) differ slightly from WHO criteria (glucose > 200 mg/dL at 2 hours alone). Abnormal glucose measurements on more than one occasion are required for a diagnosis of diabetes.22,23 The complex diagnostic criteria reflect both the difficulty in distinguishing diabetic from nondiabetic patients on the basis of a single measurement, and the substantial test-retest variability of the OGTT. The coefficient of variation for OGTT ranges from 20% to 35%.24,25 To improve reliability of the OGTT in nonpregnant adults, the American Diabetes Association (ADA) recommends that patients eat an unrestricted diet for 3 days preceding the test and fast overnight before the test.26

Both the NDDG and WHO recognize an intermediate form of disordered glucose metabolism, impaired glucose tolerance (IGT), based on intermediate results of the OGTT (140-200 mg/dL).22,23 Patients with IGT are at increased risk of developing frank diabetes, but rates of progression are highly variable. IGT is also a risk factor for cardiovascular disease.25 A significant number of individuals diagnosed with IGT revert to normal on repeat testing,25 and the treatment implications of IGT alone are uncertain.

Diagnosis of Gestational Diabetes

The diagnosis of GDM is traditionally based on two or more abnormal values during a 3-hour glucose tolerance test using 100 g glucose.22,27 NDDG diagnostic criteria are based on extrapolations from standards for whole blood glucose originally derived by O'Sullivan28 to identify mothers at risk of developing diabetes in long-term follow-up. The conversion factor used to develop criteria for plasma glucose measurements may have been incorrect, however,29 and others have proposed modified criteria with lower thresholds as more sensitive predictors of adverse pregnancy outcomes.30 Outside of North America, the diagnosis of GDM is usually based on WHO criteria using a 2-hour 75 g glucose tolerance test.23 The prevalence of GDM varies considerably depending on whether WHO, NDDG, or modified criteria of Carpenter and Coustan30 are used.15,31 In addition to poorly standardized criteria for a positive OGTT in pregnancy, the lack of studies on the reproducibility of the 100 g glucose tolerance test contributes to ongoing controversy over the diagnosis of GDM.32-34

Because diagnostic glucose tolerance testing is too time-consuming and expensive for routine screening, various blood or urine tests have been examined for their ability to identify three distinct at-risk populations among asymptomatic persons: persons with undiagnosed NIDDM, pregnant women with GDM, and individuals at high risk of developing IDDM.

Screening for Non-Insulin-Dependent Diabetes

The most commonly used screening tests for NIDDM include measurement of serum or plasma glucose in fasting or postprandial specimens, measurement of glycosylated proteins in blood, and detection of glucose in urine. The sensitivity and specificity of the fasting plasma glucose (compared to diagnostic oral glucose tolerance testing) depends on the threshold set to define an abnormal screening result. A single fasting glucose above 140 mg/dL is specific for diabetes (>99%) but sensitivity varies widely among different populations (21-75%).35-40 Using a lower threshold (>123 mg/dL) improves sensitivity (40-88%), while maintaining reasonably high specificity (97-99%).35-40 A random (i.e., nonfasting) plasma glucose greater than 140 mg/dL has a sensitivity of 45% and a specificity of 86%.41 The ADA recommends that a fasting plasma glucose greater than 115 mg/dL, or a random glucose greater than 160 mg/dL, be considered a positive screen to be confirmed with OGTT.26

The nonenzymatic attachment of glucose to circulating proteins, primarily hemoglobin and albumin, reflects overall metabolic control in diabetic populations. A number of studies have evaluated HbA1c and serum fructosamine as screening tests for diabetes.40,42-45 Test characteristics are more variable than fasting plasma glucose, with sensitivity ranging from 15% to 93% and specificity from 84% to 99%.

Presence of glucose in the urine is fairly specific but less sensitive than most blood tests for NIDDM. In population-based screening using semiquantitative urine dipstick, a "trace positive" dipstick result or greater has a reported sensitivity of 23-64% and specificity of 98-99%.40,46 In a high-risk population, quantitative assays of urine glucose achieved high sensitivity (81%) with high specificity (98%), comparable to both fasting plasma glucose and glycosylated protein assays.40

Sensitivity of all screening tests increases with the severity of hyperglycemia among the diabetic population.40 Both the sensitivity and positive predictive value of screening tests will be highest in high-risk populations such as Native Americans and African Americans, where undiagnosed diabetes and severe hyperglycemia are more prevalent.40 In the asymptomatic general population, where the prevalence of undiagnosed diabetes is only 1-3%, a greater proportion of diabetic patients may be missed by screening, and many persons with a positive screening test will not have diabetes. Screening asymptomatic persons may have some harmful effects, including an increase in false-positive diagnoses in a review of 112 patients being treated for diabetes in a general practice, nine (8%) patients, all without classic symptoms, were found not to have diabetes on further evaluation.47 Even a true-positive diagnosis could have adverse consequences for an asymptomatic person if it causes "labeling" effects48 or difficulty obtaining insurance.

Screening for GDM

The Third International Workshop Conference on Gestational Diabetes has recommended screening pregnant women at 24-28 weeks of gestation with a 50 g 1-hour oral glucose challenge test, performed in fasting or nonfasting state.27 Patients with plasma glucose of 140 mg/dL (7.8 mM) or greater at 1 hour should undergo a diagnostic 3-hour OGTT. There is no single threshold that accurately separates normal from abnormal results on the glucose challenge test, however.[[27] Estimates of sensitivity of screening under this protocol range from 71% to 83% with a specificity of 78%-87%.30,49,50 Sensitivity is increased by using a lower threshold for a positive screen30,51 and by testing in the fasting state.49 A large prospective study of nearly 4,300 pregnant women reported that using higher cutpoints (142-149 mg/dL) and adjusting for time since last meal could reduce the misclassification of patients based on initial screening tests.52 Reproducibility of the 1 hour glucose challenge test is only fair,53 but it improves with advancing gestational age.54 In an unselected pregnant population (prevalence of GDM approximately 3%), fewer than one in five women with a positive glucose challenge test will meet criteria for gestational diabetes on a full OGTT.52

The elevations in plasma glucose in GDM are less pronounced than in IDDM or NIDDM. As a result, neither serum glycosylated proteins51,55-58 nor urine glucose34 are sufficiently sensitive for detecting GDM. In addition, glucosuria is common among nondiabetic pregnant women. Random blood glucose has been advocated as a simpler and less costly screening test for GDM59,60 but its test performance has not been fully evaluated. A large prospective study is comparing fasting and random plasma glucose to oral glucose challenge for detecting GDM and for predicting adverse perinatal outcomes.52

Screening for Patients at Risk for IDDM

A growing body of evidence indicates that IDDM is a genetically linked autoimmune disorder, in which progressive destruction of insulin-producing pancreatic islet cells eventually leads to complete dependence on exogenous insulin.61 Islet cell autoantibodies and insulin autoantibodies are pres- ent in the majority of patients with newly diagnosed IDDM,62 and may precede the onset of clinical symptoms by months to years. Immunoassays for islet cell antibodies remain difficult to standardize63 however, and appear to be of limited value for screening in the general population. In individuals without a family history of IDDM, the prevalence of islet cell autoantibodies ranges from 0.3% to 4.0% and the chance of developing IDDM in antibody-positive individuals is estimated to be less than 10%.64 The potential value of immune markers is greater in high-risk individuals (i.e., first-degree relatives of affected patients). Several studies report that a combination of immune markers and measures of insulin responsiveness can identify a population at very high risk (up to 70%) of developing IDDM.62,63,65 This high risk may make such persons appropriate candidates for experimental interventions to reduce the risk of progression to IDDM. Only 10% of all cases of IDDM, however, occur in persons with a positive family history.

Return to Table of Contents

Asymptomatic NIDDM

Up to 20% of patients with newly diagnosed NIDDM already have early retinopathy, suggesting that the onset of diabetes may be many years (estimated 9-12 years) before clinical diagnosis, and that the microvascular changes may precede overt symptoms in many patients.66 Earlier detection through screening might provide an opportunity to reduce the progression of microvascular or macrovascular disease due to asymptomatic hyperglycemia. Animal models of diabetes suggest that hyperglycemia is the underlying cause of microvascular complications,67 and numerous epidemiologic studies confirm that the degree of hyperglycemia and duration of disease are associated with microvascular complications such as nephropathy, retinopathy, and neuropathy.5,68-72 Direct evidence that improving glucose control reduces the incidence of these complications has only recently become available, and only for patients with IDDM. In the Diabetes Control and Complications Trial (DCCT), over 1,400 subjects with IDDM were randomized to intensive insulin therapy versus conventional treatment. Intensive insulin therapy improved average blood glucose, significantly reduced progression of existing retinopathy, and significantly lowered the incidence of retinopathy, neuropathy, and nephropathy in all patients.73,74

The DCCT study is generally regarded as providing strong evidence of the role of hyperglycemia in diabetic microvascular disease, but questions remain about extrapolating its results to the management of patients with NIDDM.75 The incidence of microvascular complications is lower in NIDDM than IDDM, and the largest controlled trial to date of treatment of NIDDM (the University Group Diabetes Program study) found no effect of improved glucose control with insulin or drug thera py on retinopathy.76 More definitive results may come from the U.K. Prospective Diabetes Study (UKPDS), which randomized 2,520 patients with newly diagnosed NIDDM controlled with diet to diet alone, or additional therapy with chlorpropamide, glibenclamide, metformin, or insulin.77 Three-year results indicated that patients receiving drug or insulin therapy had significantly better glucose control but greater weight gain and more frequent episodes of hypoglycemia.78 Data on other clinical outcomes are not yet available.

Patients with diabetes are at significantly increased risk for coronary heart disease, stroke, and peripheral vascular disease cardiovascular diseases combined account for the majority of deaths in diabetic patients. The risk of cardiovascular disease, however, is not clearly associated with either disease duration or degree of glycemic control. The rate of increase in coronary heart disease risk over time is similar in patients with NIDDM and in nondiabetic patients.79,80 In 8-year follow-up of almost 500 diabetic men and women, disease duration was associated with risk of ischemic heart disease in patients with IDDM but not in those with NIDDM,81 and there was no correlation between cerebrovascular and peripheral vascular events and diabetes duration. Detecting such an association may be complicated by difficulty in accurately ascertaining the onset of diabetes in patients with NIDDM. Insulin resistance and hyperinsulinemia may be more important determinants of macrovascular complications than degree of glucose control.79,82 In the UGDP study, neither cardiovascular disease nor mortality was reduced by improved glucose control in the intervention groups,76 but the interpretation of these findings has been criticized.83 Drug therapy for NIDDM carries the risk of hypoglycemia. In the UKPDS study, the annual incidence of hypoglycemia was 28% for patients on glibenclamide, and 33% for those on insulin episodes requiring medical therapy occurred in 1.4% of subjects each year.78

The majority of individuals in the U.S. who have disordered glucose metabolism have IGT.84 Untreated, most persons with IGT do not develop diabetes, but the reported cumulative incidence of diabetes at 10 years has varied from 15% to 61%.25 Progression to diabetes is highest in some Native American populations.85 There is little direct evidence of a benefit of detecting and treating IGT.86,87 Prospective studies of interventions to prevent progression to frank diabetes in patients with IGT have produced conflicting results. One trial of dietary and pharmacologic treatment88 and a nonrandomized trial of diet and physical activity training89 each reported a reduced incidence of diabetes, whereas other prospective studies have reported no effect on the rate of progression to diabetes.90-92

Gestational Diabetes

GDM is associated with increased risk of fetal macrosomia, birth trauma, neonatal hypoglycemia, and perinatal mortality.93-96 No properly controlled trial has examined the benefit of universal or selective screening compared to routine care without screening. In two retrospective analyses, no significant difference in macrosomia or in birth trauma was found in women screened for GDM compared to unscreened control populations.97,98 Because women screened for GDM are more likely to be at high risk, such studies cannot reliably exclude a benefit of screening.98

The clearest benefit of screening is the potential for treatment to reduce the incidence of fetal macrosomia in women with GDM. Although modified diet can reduce hyperglycemia in GDM, only one controlled trial has examined the effect of dietary therapy on clinical outcomes in GDM.99 A total of 158 women with mild GDM (positive by NDDG criteria but not WHO criteria) were randomized to diet treatment or no therapy there were no significant differences in perinatal outcomes, although slightly fewer infants over 4,000 g were born to diet-treated mothers (3 vs. 5).100 Several randomized controlled trials have demonstrated that diet and insulin (compared to diet alone) results in improved glucose control and reduced incidence of macrosomia in women with GDM.94,101,102 Macrosomia was not significantly reduced in a fourth trial, but 15% of the women assigned to diet therapy received insulin because glucose control was inadequate.103 An overview of four randomized trials estimated that treatment of GDM with diet and insulin, compared to diet alone, reduced the incidence of macrosomia by two thirds (6% vs. 17%).104 Despite a reduction in macrosomia, there were no significant differences in rates of cesarean section, forceps delivery, or birth trauma between treated and control groups in any of the prospective trials, however. There was only one reported instance of shoulder dystocia among 140 births in the two trials reporting this outcome.101,104 In a retrospective analysis of 445 gestational diabetics, women who received both insulin and dietary treatment had significantly lower rates of birth trauma and operative delivery than women who received dietary treatment alone or no intervention.96 Since treatment was not randomly assigned, factors other than treatment may have contributed to the differences in outcomes.

The benefit of improved glucose control on other outcomes in GDM, including perinatal mortality, remains uncertain. Although several case series have reported marked improvements in perinatal death rates with treatment of GDM,95,97,105-107 none of these studies employed an appropriate control group. The use of historical controls (i.e., outcomes of prior pregnancies) or general population controls is likely to exaggerate the apparent benefits of treatment. In an overview of five randomized trials, there was no significant difference in perinatal mortality among women treated with diet and insulin (2.7%) and those treated with diet alone (3.2%).104 Moreover, in trials conducted after 1975, there were no perinatal deaths in treated or control groups.100,104 In one trial, insulin treatment was associated with lower rates of neonatal jaundice and nonsignificant reductions in admissions to the neonatal ICU.108 At the same time, treatment of GDM may have adverse effects for some women. In one retrospective analysis, women with GDM who maintained tight glucose control (mean glucose < 87 mg/dL) had a higher incidence of small-for-gestational age infants than nondiabetic controls.109

Degrees of hyperglycemia more subtle than in GDM may result in increased maternal and neonatal complication rates.110-112 The incidence of macrosomia and preeclampsia/eclampsia is higher in women who demonstrate at least one abnormal result among the four points on a glucose tolerance test. The prevalence of mildly hyperglycemic pregnant women who do not meet the criteria for GDM but are at increased risk during pregnancy is unknown.

Although treatment of GDM can reduce macrosomia, the impact of widespread screening and treatment on the overall incidence of macrosomia and dystocia may be quite small. The reported incidence of macrosomia in the general population varies from 1% to 8%,93,113 and most macrosomic infants are born to women without GDM.114 Gestational diabetes was responsible for only 5% of infants over 4,500 g in one study,115 and it is estimated to account for only 5% of shoulder dystocia cases in this country.116 Other factors such as maternal obesity, gestational weight gain, and maternal age may be more important determinants of macrosomia and adverse outcomes.117 In a prospective study of GDM controlled with diet, the only significant predictor of birth weight was maternal weight at delivery plasma glucose levels were poor predictors of birth weight.118

Persons at Risk for IDDM

Earlier diagnosis of IDDM could be of considerable benefit if treatment could arrest the disease process before severe insulinopenia and hyperglycemia had developed. A number of recent trials have examined whether immunosuppressive agents can delay disease progression in patients with new-onset IDDM.61 Although some patients have experienced prolonged remissions, the benefit has not been sustained in most patients, and the serious adverse effects of immunosuppressive agents are likely to preclude their use in completely asymptomatic persons. There have been several promising small trials of other interventions to prevent IDDM in high-risk asymptomatic persons, enrolling individuals identified by autoantibodies levels and other physiologic measures.63,119,120 Multicenter randomized clinical trials are currently underway to determine whether prophylactic regimens of insulin or nicotinamide can prevent progression to IDDM in such high-risk subjects.61

Return to Table of Contents

The Canadian Task Force on the Periodic Health Examination (CTF),121 the American College of Physicians (ACP),122 and the American Academy of Family Physicians123 recommend against routine screening for diabetes among asymptomatic nonpregnant adults each of these organizations concluded that selective screening may be reasonable among individuals at high risk of developing diabetes (e.g., older obese persons, those with a strong family history). AAFP policy is currently under review. The ADA recommends screening all individuals with a careful history and measuring fasting glucose on those with identified risk factors for developing diabetes, including obesity, family history, history of GDM, selected medical conditions, or selected ethnic background.124 A 1994 report of the WHO concluded that population screening for NIDDM was not justified, but that opportunistic screening of high-risk persons may be useful to permit earlier intervention.125

The ACP,122 the ADA,124 and the Third International Workshop Conference on Gestational Diabetes 27 recommend universal screening for GDM in pregnant women between weeks 24 and 28 using a 1-hour glucose tolerance test. The American College of Obstetricians and Gynecologists and the American Academy of Pediatrics do not recommend universal screening in pregnancy but strongly recommend screening pregnant women in certain high-prevalence populations (e.g., Native Americans) and those with specific risk factors (age over 30, family history of diabetes, previous macrosomia, malformed or stillborn infants, hypertension, or glucosuria).126,127 The CTF concluded that there was insufficient evidence to recommend for or against universal screening for GDM, but suggested close monitoring of women with risk factors for GDM.121

Return to Table of Contents

Screening for diabetes in asymptomatic adults suffers from two important limitations: the lack of a practical screening test that is both sensitive and specific, and insufficient evidence that detection of diabetes in the asymptomatic period significantly improves long-term outcomes. Even if improving glucose control can reduce long-term complications of NIDDM, many other factors must be considered in determining the likely benefits and risks of screening in asymptomatic persons: efficacy of diet or medications in reducing glucose levels compliance of asymptomatic persons with lifestyle advice possible risks of drug or insulin therapy inconvenience and costs of screening, follow-up, and treatment and the potential adverse effects of screening (false-positive diagnoses, "labeling" of asymptomatic persons). Targeting screening to high-risk groups (certain ethnic populations, older overweight subjects) and emphasizing interventions that are inexpensive and safe (exercise, prudent diet, and weight loss) are likely to minimize the potential adverse effects of screening. Since most of these interventions are recommended for all adults, the additional benefit of screening to promote lifestyle interventions remains uncertain. If the ongoing UKPDS trial demonstrates important clinical benefits from more intensive interventions (i.e., drug or insulin therapy) in patients with minimally symptomatic NIDDM, this would provide stronger support for screening for diabetes among asymptomatic adults.

The value of widespread screening for GDM is also unproven. Important questions remain about the diagnostic gold standard, the optimal screening test, and the appropriate management of GDM. Although there is good evidence that insulin treatment can reduce the incidence of macrosomia in GDM, evidence of a benefit on clinically important perinatal outcomes (birth trauma, operative delivery, neonatal metabolic derangements, or perinatal mortality) is much weaker. The high risk associated with GDM in earlier cohorts primarily reflects adverse outcomes in women who were older, overweight, or otherwise at increased risk. Universal screening is likely to have only a small impact on the overall incidence of macrosomia and birth trauma and may subject many low-risk women to the inconvenience, costs and possible risks of follow-up testing, dietary restriction, or insulin management. A 1988 study estimated that universal screening would cost $8,000 per case of macrosomia prevented.122 By one estimate, however, up to 10,000 women would need to be screened to prevent 50 cases of macrosomia, 6 cases of shoulder dystocia, and 1 case of shoulder girdle injury (few of which cause lasting problems).128 Targeting screening to women with risk factors for GDM (including older age), with emphasis on dietary management of GDM, is likely to minimize the adverse effects and costs of screening. Direct evidence of a benefit of screening on important clinical outcomes is not available for any group, however.

Immune markers are not sufficiently specific to recommend their use in the general population at this time. Screening persons with a family history of IDDM using immune markers and physiologic measurements can identify a small number of persons at very high risk of developing IDDM. Patients with a family history account for only 10% of all cases of IDDM, however, and trials of interventions to prevent IDDM in high-risk patients have not yet been completed.

Primary prevention may be a more effective means to reduce diabetes-associated morbidity than widespread screening. Diet, exercise, and weight reduction can safely improve glucose tolerance and are likely to have independent benefits on other important chronic diseases (see Chapters 55 and 56). Whether diabetes screening improves compliance with generally recommended lifestyle interventions has not been determined.

Clinical Intervention

There is insufficient evidence to recommend for or against routine screening for NIDDM in nonpregnant adults ("C" recommendation). Although evidence of a benefit of early detection is not available for any group, clinicians may decide to screen selected persons at high risk of NIDDM on other grounds, including the increased predictive value of a positive test in individuals with risk factors and the potential (although unproven) benefits of reducing asymptomatic hyperglycemia through diet and exercise. Individuals at higher risk of diabetes include obese men and women over 40, patients with a strong family history of diabetes, and members of certain ethnic groups (Native Americans, Hispanics, African Americans). In persons without risk factors, screening for asymptomatic disease is much less likely to be of benefit, due to the low burden of disease and the poor predictive value of screening tests in low-risk persons. Measurement of fasting plasma glucose is recommended by experts as the screening test of choice the frequency of screening is left to clinical discretion.

There is also insufficient evidence to recommend for or against routine screening for GDM ("C" recommendation). Although a clear benefit of screening on perinatal morbidity has not been demonstrated for any group, clinicians may decide to screen high-risk pregnant women on other grounds, including the higher burden of disease, and the potential clinical benefits from reducing macrosomia due to GDM. Risk factors for GDM include obesity, older maternal age, a family history of diabetes, and a history of macrosomia, fetal malformation, or fetal death. The 1-hour 50 g glucose challenge test, with confirmation of abnormal results with a 3-hour 100 g oral glucose tolerance test, is the screening test recommended by expert panels in the U.S.

Screening with immune markers to identify asymptomatic individuals at risk for developing IDDM is not recommended in the general population ("D" recommendation).

Return to Table of Contents

The draft update of this chapter was prepared for the U.S. Preventive Services Task Force by M. Carrington Reid, MD, PhD, Harold C. Sox, Jr., MD, Richard Comi, MD, and David Atkins, MD, MPH. 

Return to Table of Contents

1. National Diabetes Information Clearinghouse Diabetes statistics. Bethesda: National Institute of Diabetes and Digestive and Kidney Diseases. 1994; (NIH Publication no. 94-3822.)
2. Harris MI. Undiagnosed NIDDM: clinical and public health issues. Diabetes Care. 1993;16:642–652.
3. American Diabetes Association. Alexandria, VA: Alexandria, VA; 1995. Diabetes—1996 vital statistics.
4. Alberti KGMM, DeFronzo RA, Zimmet P, eds. International textbook of diabetes mellitus. New York: John Wiley and Sons. 1995
5. Harati Y. Diabetic peripheral neuropathies. Ann Intern Med. 1987;107:546–559.
6. Centers for Disease Control. Lower extremity amputations among persons with diabetes mellitus–Washington, 1988. MMWR. 1991;40:737–741.
7. Viberti GC, Yip-Messent J, Morocutti A. Diabetic nephropathy: future avenue. Diabetes Care. 1992;15:1216–1225.
8. Breyer JA. Diabetic nephropathy in insulin-dependent patients. Am J Kidney Dis. 1992;20:533–547.
9. Centers for Disease Control and Prevention. Public health focus: prevention of blindness associated with diabetic retinopathy. MMWR. 1993;42:191–195.
10. Garner P.Type I diabetes mellitus and pregnancy. Lancet. 1995;346:157–161.
11. Miodovnik M, Mimouni F, Dignan PSJ, et al. Major malformations in infants of IDDM women: vasculopathy and early-trimester poor glycemic control. Diabetes Care. 1988;11:713–718.
12. Centers for Disease Control and Prevention. Surveillance for diabetes mellitus–United States, 1980-1989. MMWR. 1993;42(SS-2):1–20.
13. American Diabetes Association. Alexandria, VA: Alexandria, VA; 1993. Direct and indirect costs of diabetes in the United States in 1992.
14. Knowler WC, Saad MS, Pettitt DJ, et al. Determinants of diabetes mellitus in the Pima Indians. Diabetes Care. 1993;16(Suppl 1):216–226.
15. Magee MS, Walden CE, Benedetti TJ, et al. Influence of diagnostic criteria on the incidence of gestational diabetes and perinatal morbidity. JAMA. 1993;269:609–615.
16. Gross TL, Sokol RJ, Williams T, et al. Shoulder dystocia: a fetal-physician risk. Am J Obstet Gynecol. 1987;156:1408–1418.
17. McFarland LV, Raskin M, Daling JR, et al. Erb/Duchenne's palsy: a consequence of fetal macrosomia and method of delivery. Obstet Gynecol. 1986;68:784–788.
18. Modanlou HD, Dorchester WL, Thorosian A, et al. Macrosomia–maternal, fetal, and neonatal implications. Obstet Gynecol. 1980;55:420–424.
19. Sandmire HF, O'Halloin TJ. Shoulder dystocia: its incidence and associated risk factors. Int J Gynaecol Obstet. 1988;26:65–73.
20. Cunningham FG, ed. Norwalk, CT: Norwalk, CT; 1993. Williams obstetrics. 19th ed.
21. Harris MI. Gestational diabetes may represent discovery of preexisting glucose intolerance. Diabetes Care. 1988;11:402–411.
22. National Diabetes Data Group Classification and diagnosis of diabetes mellitus and other categories of glucose tolerance. Diabetes. 1979;28:1039–1057.
23. Expert Committee on Diabetes Mellitus—World Health Organization WHO tech report series 646. Geneva: World Health Organization. 1980.
24. Home P.The OGTT: gold that does not shine. Diabetes Med. 1988;5:313–314.
25. Yudkin JS, Alberti KG, McLarty DG, et al. Impaired glucose tolerance. Is it a risk factor for diabetes or a diagnostic ragbag? BMJ. 1990;301:397–402.
26. American Diabetes Association. Office guide to diagnosis and classification of diabetes mellitus and other categories of glucose intolerance. Diabetes Care. 1993;16(Suppl 2):4–6.
27. Metzger BE. Summary and recommendations of the Third International Workshop Conference on Gestational Diabetes Mellitus. Diabetes. 1991;40(Suppl 2):197–201.
28. O'Sullivan JB, Gellis SS, Dandrow RV, et al. The potential diabetic and her treatment in pregnancy. Obstet Gynecol. 1966;27:683–689.
29. Sacks DA, Salim AF Aba-Fadil S, Fotheringham N. Do the current standards for glucose tolerance testing in pregnancy represent a valid conversion of O'Sullivan's original criteria? Am J Obstet Gynecol. 1989;161:638–641.
30. Carpenter MW, Coustan DR.Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol. 1982;144:768–773.
31. Li DFH, Wong VCW, O'Hoy KM, et al. Evaluation of the WHO criteria for 75 g oral glucose tolerance test in pregnancy. Br J Obstet Gynaecol. 1987;94:847–850.
32. Canadian Task Force on the Periodic Health Examination Periodic health examination, 1992 update: 1. Screening for gestational diabetes mellitus. Can Med Assoc J. 1992;147:435–443.
33. Naylor CD.Diagnosing gestational diabetes: is the gold standard valid? Diabetes Care. 1989;12:565–572.
34. Singer DE, Coley CM, Samet JH, et al. Tests of glycemia in diabetes mellitus. Ann Intern Med. 1989;110:125–137.
35. Cockram CS, Lau JT, Chan AY, et al. Assessment of glucose tolerance test criteria for diagnosis of diabetes in Chinese subjects. Diabetes Care. 1992;15:988–990.
36. Blunt BA, Barrett-Connor E, Wingard DL. Evaluation of fasting plasma glucose as screening test for NIDDM in older adults. Diabetes Care. 1991;14:989–993.
37. Sekikawa A, Tominaga M, Takahashi K, et al. Is examination of fructosamine levels valuable as a diagnostic test for diabetes mellitus. Diabet Res Clin Pract. 1990;8:187–192.
38. Swai ABM, Harrison K, Chuwa LM, et al. Screening for diabetes: does measurement of serum fructosamine help? Diabet Med. 1988;5:648–652.
39. Hanson RI, Nelson RG, McCance DR, et al. Comparison of screening tests for non-insulin dependent diabetes mellitus. Arch Intern Med. 1993;153:2133–2140.
40. Modan M, Harris MI. Fasting plasma glucose in screening for NIDDM in the U.S. and Israel. Diabetes Care. 1994;17:436–439.
41. Bourn D, Mann J. Screening for noninsulin dependent diabetes mellitus and impaired glucose tolerance in a Dunedin general practice—is it worth it? NZ Med J. 1992;105:208–210.
42. Croxson SC, Absalom S, Burden AC. Fructosamine in diabetes screening of the elderly. Ann Clin Biochem. 1991;28:279–282.
43. Guillausseau PJ, Charles MA, Paolaggi F, et al. Comparison of HbA1 and fructosamine in diagnosis of glucose-tolerance abnormalities. Diabetes Care. 1990;13:898–900.
44. Forrest RD, Jackson CA, Gould BJ, et al. Four assay methods for glycated hemoglobin compared as screening tests for diabetes mellitus: the Islington diabetes survey. Clin Chem. 1988;34:145–148.
45. Little RR, England JD, Wiedmeyer H, et al. Relationship of glycosylated hemoglobin to oral glucose tolerance. Diabetes. 1988;37:60–64.
46. Anderson Andersson DKG DK, Lundblad E, Svardsudd K. A model for early diagnosis of type II diabetes mellitus in primary health care. Diabetes Med. 1993;10:167–173.
47. Patchett P, Roberts D. Diabetic patients who do not have diabetes: investigation of register of diabetic patients in general practice. BMJ. 1994;308:1225–1226.
48. Haynes RB, Sackett DL, Taylor DW, et al. Increased absenteeism after detection and labelling of hypertensive patients. N Engl J Med. 1978;299:741–744.
49. Coustan DR, Widness JA, Carpenter MW, et al. Should the fifty-gram, one-hour plasma glucose screening test for gestational diabetes be administered in the fasting or fed state? Am J Obstet Gynecol. 1986;154:1031–1035.
50. O'Sullivan JB, Mahan CM. Criteria for the oral glucose tolerance test in pregnancy. Diabetes. 1964;13:278–285.
51. Roberts AB, Baker JR, Metcalf P, et al. Fructosamine compared with a glucose load as a screening test for gestational diabetes. Obstet Gynecol. 1990;76:773–775.
52. Sermer M, Naylor CD, Gare DJ, et al. Impact of time since last meal on the gestational glucose challenge test. The Toronto Tri-Hospital Gestational Diabetes Project. Am J Obstet Gynecol. 1994;171:607–616.
53. Sacks DA, Abu-Fadil S, Greenspoon JS, et al. How reliable is the fifty-gram one-hour glucose screening test? Am J Obstet Gynecol. 1989;161:642–645.
54. Espinosa de los Monteros A, Carino N, Ramirez A. The reproducibility of the 50-gram, 1-hour glucose screen for diabetes in pregnancy. Obstet Gynecol. 1993;82:515–518.
55. Menon U, Ranjan M, Jasper P, et al. Evaluation of plasma fructosamine as a screening test for gestational diabetes. Aust NZ J Obstet Gynaecol. 1991;31:25–26.
56. Corcoy R, Cerqueira MJ, Pedreno J, et al. Serum fructosamine is not a useful screening test for gestational diabetes. Eur J Obstet Gynecol Reprod Biol. 1990; 1991;38:217–220.
57. Comtois R, Desjarlais F, Nguyen M, et al. Clinical usefulness of estimation of serum fructosamine concentration as screening test for gestational diabetes. Am J Obstet Gynecol. 1989;160:651–654.
58. Cefalu WT, Prather KL, Chester DL, et al. Total serum glycosylated proteins in detection and monitoring of gestational diabetes. Diabetes Care. 1990;13:872–875.
59. Lind T. Antenatal screening using random blood glucose values. Diabetes. 1985;34(Suppl 2):17–20.
60. Hatem M, Dennis KJ. A random plasma glucose method for screening abnormal glucose tolerance in pregnancy. Br J Obstet Gynaecol. 1987;94:213–216.
61. Atkinson MA, Maclaren NK. The pathogenesis of insulin-dependent diabetes mellitus. N Engl J Med. 1994;331:1428–1444.
62. Riley WJ, Maclaren NK, Krischer J, et al. A prospective study of the development of diabetes in relatives of patients with insulin-dependent diabetes. N Engl J Med. 1990;323:1167–1172.
63. Eisenbarth GS, Verge CF, Allen H, et al. The design of trials for prevention of IDDM. Diabetes. 1993;42:941–947.
64. Bingley PJ, Bonifacio E, Shattock M, et al. Can islet cell antibodies predict IDDM in the general population? Diabetes Care. 1993;16:45–50.
65. Bonifacio E, Bingley PJ, Shattock M, et al. Quantification of islet-cell antibodies and prediction of insulin-dependent diabetes. Lancet. 1990;335:147–149.
66. Harris MI, Klein R, Welbourn TA, et al. Onset of NIDDM occurs at least 4-7 years before clinical diagnosis. Diabetes Care. 1992;15:815–819.
67. Brownlee M.Glycation and diabetic complications. Diabetes. 1994;43:836–841.
68. Takazakura E, Nakamoto Y, Hayakawa H, et al. Onset and progression of diabetic glomerulosclerosis: a prospective study based on serial renal biopsies. Diabetes. 1975;24:1–9.
69. Miki E, Fukuda M, Kuzuya T, et al. Relation of the course of retinopathy to control of diabetes, age, and therapeutic agents in diabetic Japanese patients. Diabetes. 1969;18:773–780.
70. Chase HP, Jackson WE, Hoops SL, et al. Glucose control and the renal and retinal complications of insulin-dependent diabetes mellitus. JAMA. 1989;261:1155–1160.
71. Reichard P, Berglund B, Britz A, et al. Intensified conventional insulin treatment retards the microvascular complications of insulin-dependent diabetes mellitus (IDDM): the Stockholm Diabetes Intervention Study (SDIS) after 5 years. J Intern Med. 1991;230:101–108.
72. Dahl-Jorgensen K, Bjoro T, Kierulf P, et al. Long term glycemic control and kidney function in insulin-dependent diabetes mellitus. Kidney Int. 1992;41:920–923.
73. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977–986.
74. The Diabetes Control and Complications Trial The effect of intensive diabetes therapy on the development and progression of neuropathy. Ann Intern Med. 1995;122:561–568.
75. Lasker RD.The Diabetes Control and Complications Trial. Implications for policy and practice. N Engl J Med. 1993;329:1035–1036.
76. University Group Diabetes Program. Effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. VIII. Evaluation of insulin therapy: final report. Diabetes. 1982;31(Suppl 5):1–81.
77. U.K. Prospective Diabetes Study Group. UK Prospective Diabetes Study (UKPDS) VIII. Study design, progress and performance. Diabetologia. 1991;34:877–890.
78. United Kingdom Prospective Diabetes Study Group. United Kingdom prospective diabetes study (UKPDS) 13: relative efficacy of randomly allocated diet, sulphonylurea, insulin or metformin in patients with newly diagnosed non-insulin dependent diabetes followed for three years. BMJ. 1995;310:83–88.
79. Donahue RP, Orchard TJ. Diabetes mellitus and macrovascular complications. Diabetes Care. 1992;15:1141–1155.
80. Jarrett RJ, Shipley MJ. Non-insulin dependent diabetes mellitus and cardiovascular disease–putative association versus common antecedents, further evidence from the Whitehall Study. Diabetologia. 1988;31:737–740.
81. Morrish NJ, Stevens LK, Jarrett RJ, et al. Risk factors for macrovascular disease in diabetes mellitus: the London follow-up to the WHO multinational study of vascular disease in diabetics. Diabetologia. 1991;34:590–594.
82. Welborn TA, Wearne K Coronary heart disease incidence and cardiovascular mortality in Busselton with reference to glucose and insulin concentrations. Diabetes Care. 1979;2:154–160.
83. Kilo C, Miller JP, Williamson JR. The Achilles heel of the University Group Diabetes Program. JAMA. 1980;243:450–457.
84. Harris MI. Impaired glucose tolerance in the U.S. population. Diabetes Care. 1989;12:464–474.
85. Saad MF, Knowler WC, Pettitt DJ, et al. Transient impaired glucose tolerance in Pima indians: is it important? BMJ. 1988;297:1438–1441.
86. Genuth SM, Houser HB, Carter JR, et al. Observations on the value of mass indiscriminate screening for diabetes mellitus based on a five-year follow-up. Diabetes. 1978;27:377–383.
87. Bennett PH, Knowler WC. Early detection and intervention in diabetes mellitus: is it effective? J Chronic Dis. 1984;37:653–656.
88. Sartor G, Schersten B, Carlstrom S, et al. Ten-year follow-up of subjects with impaired glucose tolerance: prevention of diabetes by tolbutamide and diet regulation. Diabetes. 1980;29:41–49.
89. Eriksson KF, Lindgarde F. Prevention of type 2 (non-insulin-dependent) diabetes mellitus by diet and physical exercise. The 6-year Malmo feasibility study. Diabetologia. 1991;34:891–898.
90. Keen H, Jarrett RJ, McCartney P. The ten-year follow-up of the Bedford survey (1962-1972) glucose tolerance and diabetes. Diabetologia. 1982;22:73–78.
91. Jarrett RJ, Keen H, Fuller JH, et al. Treatment of borderline diabetes: controlled trial using carbohydrate restriction and phenformin. BMJ. 1977;2:861–865.
92. Jarrett RJ, Keen H, McCartney P. The Whitehall Study: ten year follow-up report on men with impaired glucose tolerance with reference to worsening of diabetes and predictors of death. Diabetic Med. 1984;1:279–283.
93. Langer O, Berkus MD, Huff RW, et al. Shoulder dystocia: Should the fetus weighing > 4000 grams be delivered by cesarean section? Am J Obstet Gynecol. 1991;165:831–837.
94. O'Sullivan JB, Charles D, Mahan CM, et al. Gestational diabetes and perinatal mortality rate. Am J Obstet Gynecol. 1973;116:901–904.
95. Hod M, Merlob P, Friedman S, et al. Gestational diabetes mellitus—a survey of perinatal complications in the 1980's. Diabetes. 1991;40(Suppl 2):74–78.
96. Coustan DR, Imarah J. Prophylactic insulin treatment of gestational diabetes reduces the incidence of macrosomia, operative delivery, and birth trauma. Am J Obstet Gynecol. 1984;150:836–842.
97. Gabbe SG, Mestman JH, Freeman RK, et al. Management and outcome of Class A diabetes mellitus. Am J Obstet Gynecol. 1977;127:465–469.
98. Santini DL, Ales KL. The impact of universal screening for gestational glucose intolerance on outcome of pregnancy. Surg Gynecol Obstet. 1990;170:427–436.
99. Hunter DJS, Keirse MJNC. Gestational diabetes. In: Chalmers I, Enkin MW, Keirse MJNC, eds.. Effective care in pregnancy and childbirth. , Oxford: Oxford University Press, 1989:403–410.
100. Li DFH, Wong VCW, O'Hoy KMKY, et al. Is treatment needed for mild impairment of glucose in pregnancy? A randomized controlled trial. Br J Obstet Gynaecol. 1987;94:851–854.
101. Coustan DR, Lewis SB. Insulin therapy for gestational diabetes. Obstet Gynecol. 1978;51:306–310.
102. Thompson DJ, Porter KB, Gunnells DJ, et al. Prophylactic insulin in the management of gestational diabetes. Obstet Gynecol. 1990;75:960–964.
103. Persson B, Strangenberg M, Hansson U, et al. Gestational diabetes mellitus (GDM) comparative evaluation of two treatment regimens, diet versus insulin and diet. Diabetes. 1985:11(Suppl 2)101–105.
104. Walkinshaw SA. Diet + insulin vs diet alone for 'gestational diabetes. In: Enkin MW, Keirse MJNC, Renfrew MJ, Neilson JP, eds. Pregnancy and childbirth module. Cochrane Database of Systematic Reviews: Review No. 06650, 20 April 1993, "Cochrane Updates on Disk,", Oxford: Update Software, 1994 Disk Issue 1.
105. Adashi EY, Pinto H, Tyson JE. Impact of maternal euglycemia on fetal outcome in diabetic pregnancy. Am J Obstet Gynecol. 1979;133:268–274.
106. Roversi GD, Gargiulo M, Nicolini U, et al. A new approach to the treatment of diabetic women: report of 479 cases seen from 1963 to 1975. Am J Obstet Gynecol. 1979;135:567–576.
107. Gyves MT, Rodman HM, Little AB, et al. A modern approach to management of pregnant diabetics: a two-year analysis of perinatal outcomes. Am J Obstet Gynecol. 1977;128:606–616.
108. Langer O, Anyaegbunam A, Brustman L, et al. Management of women with one abnormal oral glucose tolerance test value reduces adverse outcome in pregnancy. Am J Obstet Gynecol. 1989;161:593–599.
109. Langer O, Levy J, Brustman L, et al. Glycemic control in gestational diabetes mellitus—how tight is tight enough: small for gestational age vs. large for gestational age? Am J Obstet Gynecol. 1989;161:646–653.
110. Berkus MD, Langer O.Glucose tolerance test: degree of glucose abnormality correlates with neonatal outcome. Obstet Gynecol. 1993;81:345–348.
111. Langer O, Brustman L, Anyaegbunam A, et al. The significance of one abnormal glucose tolerance test value on adverse outcome in pregnancy. Am J Obstet Gynecol. 1987;157:758–763.
112. Lindsay MK, Graves W, Klein L. The relationship of one abnormal glucose tolerance test value and pregnancy outcome. Obstet Gynecol. 1989;73:103–106.
113. Shelley-Jones DC, Beischer NA, Sheedy MT, et al. Excessive birth weight and maternal glucose tolerance—a 19-year review. Aust NZ J Obstet Gynaecol. 1992;32:318–324.
114. Braveman P, Showstack J, Browner W, et al. Evaluating outcomes of pregnancy in diabetic women: epidemiologic considerations and recommended indicators. Diabetes Care. 1988;11:281–287.
115. Spellacy WN, Miller S, Winegar A, et al. Macrosomia: maternal characteristics and infant complications. Obstet Gynecol. 1985;66:158–161.
116. Hopwood HG. Shoulder dystocia: fifteen years' experience in a community hospital. Am J Obstet Gynecol. 1982;144:162–166.
117. Johnson JW, Longmate JA, Frentzen B. Excessive maternal weight and pregnancy outcome. Am J Obstet Gynecol. 1992;167:353–370.
118. Jacobson JD, Cousins L. A population-based study of maternal and perinatal outcome to patients with gestational diabetes. Am J Obstet Gynecol. 1989;161:981–986.
119. Keller RJ, Eisenbarth GS, Jackson RA. Insulin prophylaxis in individuals at high risk of type I diabetes. Lancet. 1993;341:927–928.
120. Elliott RB, Chase HP.Prevention or delay of type I (insulin-dependent) diabetes: progression to overt IDDM despite oral nicotinamide. mellitus in children using nicotinamide. Diabetologia. 1991;34:362–365.
121. Canadian Task Force on the Periodic Health. Examination Canadian guide to clinical preventive health care. Ottawa: Canada Communication Group. 1994:16–23, 601-609.
122. Singer DE, Samet JH, Coley CM, Nathan DM.Screening for diabetes mellitus. In: Eddy DM, ed. Common screening tests. , Philadelphia: American College of Physicians, 1991;154–178, 404-405.
123. American Academy of Family Physicians. Kansas City, MO; 1994. Age charts for periodic health examination. (Reprint no. 510.)
124. American Diabetes Association. Position Statement, screening for diabetes. Diabetes Care. 1993;16(Suppl 2):7–9.
125. World Health Organization Prevention of diabetes mellitus: report of a WHO study group. Technical Report Series no. 844. Geneva: World Health Organization. 1994
126. American College of Obstetricians and Gynecologists Diabetes and pregnancy. Washington DC: Washington DC; Technical Bulletin no. 200. 1994
127. American Academy of Pediatrics and American College of Obstetricians and Gynecologists. Elk Grove Village, IL; 1992. Guidelines for perinatal care, 3rd ed.
128. Blank A, Metzger BE, Grave GD. Effects of gestational diabetes on perinatal morbidity reassessed. Diabetes Care. 1995;18:127–129.

Return to Table of Contents