Early detection of undiagnosed diabetes mellitus: a US perspective.

Undiagnosed Type 2 diabetes has become a common condition in the US, comprising one-third of all cases of the disease. We believe that screening for and detection of undiagnosed Type 2 diabetes is an important endeavor. In this review we provide evidence that diabetes is a condition that is appropriate for population screening and detection. This includes evidence that: 1. Type 2 diabetes is a significant health problem. It affects more than 16 million adults in the US and places these individuals at high risk for serious complications of the eyes, nerves, kidneys, and cardiovascular system. 2. There is a latent phase before diagnosis of Type 2 diabetes. During this period of undiagnosed disease, risk factors for diabetic micro- and macrovascular complications are markedly elevated and diabetic complications are developing. 3. Diagnostic criteria for diabetes have been established and are based on plasma glucose values. These criteria define a group of individuals with significant hyperglycemia who also have a high frequency of risk factors for micro- and macrovascular disease. 4. The natural history of Type 2 diabetes is understood. In most patients, diabetes proceeds inexorably from genetic predisposition, through the stage of insulin resistance and hyperinsulinemia, to beta cell failure and overt clinical disease. 5. There are effective and acceptable therapies available for Type 2 diabetes and its complications. Treating hyperglycemia to prevent complications is more effective than treating these complications after they have developed. Furthermore, guidelines for treatment to prevent cardiovascular disease in people known to have diabetes are more stringent than in those individuals who are not known to have diabetes. 6. There is a suitable test for screening for undiagnosed Type 2 diabetes that has high sensitivity and specificity - measurement of fasting plasma glucose. Guidelines for identifying persons at high risk for diabetes have been established.

The GlucoWatch biographer: a frequent automatic and noninvasive glucose monitor.

The GlucoWatch (Cygnus, Inc, Redwood City, CA, USA) biographer provides automatic, frequent and noninvasive blood glucose measurements for up to 12 h. The device extracts glucose through intact skin where it is measured by an amperometric biosensor. Clinical trials in a variety of environments have shown that the biographer provides accurate and precise glucose measurements when compared with serial fingerstick blood glucose measurements. Mean difference between these measurements was 0.26 mmol/L in the home environment (r = 0.80). Over 94% of biographer readings were in the clinically acceptable A+B region of the Clarke Error Grid. A slight positive bias is observed for the biographer readings at low glucose levels. Biographer precision, as measured by coefficient of variation (CV)%, is approximately 10%. The low glucose alert function of the biographer was able to detect up to 75% of hypoglycaemic episodes with a low false alert level. Skin irritation, characterized by erythema and oedema was either nonexistent or mild in over 87% of subjects and resolved in virtually all subjects without treatment in several days. The GlucoWatch biographer has been shown to be a safe and effective method to track glucose level trends and patterns, which should enable improved glycaemic control for many patients.

Prevention and treatment of microvascular and neuropathic complications of diabetes.

The most important factors that influence development of the eye, kidney, and nerve complications of diabetes are the duration and degree of exposure to hyperglycemia. Hypertension is also very important. The risk of complications appears to be similar for all patients with diabetes; differences in complication rates between patients with type 1 and type 2 diabetes are largely due to differences in duration of diabetes and glycemic control. The benefits of lowering blood sugar are also similar for patients with diabetes, regardless of the type. Significant reductions in complications can be seen with relatively short-term treatment of hyperglycemia.

Racial and ethnic differences in glycemic control of adults with type 2 diabetes.

OBJECTIVE: To evaluate glycemic control in a representative sample of U.S. adults with type 2 diabetes. RESEARCH DESIGN AND METHODS: The Third National Health and Nutrition Examination Survey included national samples of non-Hispanic whites, non-Hispanic blacks, and Mexican Americans aged > or = 20 years. Information on medical history and treatment of diabetes was obtained to determine those who had been diagnosed with type 2 diabetes by a physician before the survey (n = 1,480). Fasting plasma glucose and HbA1c were measured, and the frequencies of sociodemographic and clinical variables related to glycemic control were determined. RESULTS: A higher proportion of non-Hispanic blacks were treated with insulin and a higher proportion of Mexican Americans were treated with oral agents compared with non-Hispanic whites, but the majority of adults in each racial or ethnic group (71-83%) used pharmacologic treatment for diabetes. Use of multiple daily insulin injections was more common in whites. Blood glucose self-monitoring was less common in Mexican Americans, but most patients had never self-monitored. HbA1c values in the nondiabetic range were found in 26% of non-Hispanic whites, 17% of non-Hispanic blacks, and 20% of Mexican Americans. Poor glycemic control (HbA1c > 8%) was more common in non-Hispanic black women (50%) and Mexican-American men (45%) compared with the other groups (35-38%), but HbA1c for both sexes and for all racial and ethnic groups was substantially higher than normal levels. Those with HbA1c > 8% included 52% of insulin-treated patients and 42% of those taking oral agents. There was no relationship of glycemic control to socioeconomic status or access to medical care in any racial or ethnic group. CONCLUSIONS: These data indicate that many patients with type 2 diabetes in the U.S. have poor glycemic control, placing them at high risk of diabetic complications. Non-Hispanic black women, Mexican-American men, and patients treated with insulin and oral agents were disproportionately represented among those in poor glycemic control. Clinical, public health, and research efforts should focus on more effective methods to control blood glucose in patients with diabetes.

The effects of treatment on the direct costs of diabetes.

Treatment of diabetic complications consumes health care resources. Intensive therapy was shown by the Diabetes Control and Complications Trial (DCCT) to avert complications. Economic analyses and models have been used to evaluate the cost-effectiveness of intensive therapy for people with type 1 and type 2 diabetes. An economic analysis of the DCCT estimated the cost of intensive therapy to be two to three times greater than that of conventional therapy. In contrast, an economic model predicts that intensive therapy, as compared with conventional therapy, could reduce blindness from 34 to 20% or by 41%, end-stage renal disease from 24 to 7% or by 71%, and lower-extremity amputations from 7 to 4% or by 43%. Although intensive therapy is more expensive, when the costs of complications are factored in, it becomes cost-effective for treatment of type 1 diabetes. Similarly, a model to evaluate the cost-effectiveness of intensive therapy for people with type 2 diabetes found that the lifetime costs of general and diabetes-related medical care would be approximately two times greater. However, the reduction in lifetime costs of complications, which would produce substantial reductions in costs of treatment, largely offsets the difference. Intensive therapy for type 1 and type 2 diabetes may be more expensive than conventional therapy, but from an economic perspective, it is comparable in cost to pharmacological therapies for people with hypertension and hypercholesterolemia. From a health system viewpoint, intensive therapy represents a fruitful long-term financial investment.

Comparison of diabetes diagnostic categories in the U.S. population according to the 1997 American Diabetes Association and 1980-1985 World Health Organization diagnostic criteria.

OBJECTIVE: To compare the 1997 American Diabetes Association (ADA) and the 1980-1985 World Health Organization (WHO) diagnostic criteria in categorization of the diabetes diagnostic status of adults in the U.S. RESEARCH DESIGN AND METHODS: Analyses are based on a probability sample of the U.S. population age 40-74 years in the 1988-1994 Third National Health and Nutrition Examination Survey (NHANES III). People with diabetes diagnosed before the survey were identified by questionnaire. For 2,844 people without diagnosed diabetes, fasting plasma glucose was obtained after an overnight 9 to < 24-h fast, HbA1c was measured, and a 2-h oral glucose tolerance test was administered. RESULTS: Prevalence of diagnosed diabetes in this age-group is 7.9%. Prevalence of undiagnosed diabetes is 4.4% by ADA criteria and 6.4% by WHO criteria. The net change of -2.0% occurs because 1.0% are classified as having undiagnosed diabetes by ADA criteria but have impaired or normal glucose tolerance by WHO criteria, and 3.0% are classified as having impaired fasting glucose or normal fasting glucose by ADA criteria but have undiagnosed diabetes by WHO criteria. Prevalence of impaired fasting glucose is 10.1% (ADA), compared with 15.6% for impaired glucose tolerance (WHO). For those with undiagnosed diabetes by ADA criteria, 62.1% are above the normal range for HbA1c compared with 47.1% by WHO criteria. Mean HbA1c is 7.07% for undiagnosed diabetes by ADA criteria and 6.58% by WHO criteria. CONCLUSIONS: The number of people with undiagnosed diabetes by ADA criteria is lower than that by WHO criteria. However, those individuals classified by ADA criteria are more hyperglycemic, with higher HbA1c values and a greater proportion of values above the normal range. This fact, together with the simplicity of obtaining a fasting plasma glucose value, may result in the detection of a greater proportion of people with undiagnosed diabetes in clinical practice using the new ADA diagnostic criteria.

The cost-effectiveness of intensive therapy for diabetes mellitus.

Although persons with diabetes constitute only 3.1% of the US population, costs for their care account for 11.9% of total US health care expenditures. Approximately half of the expenditures for medical care for diabetes are for treatment of the metabolic condition and half for the treatment of chronic complications. Intensive therapy for persons with diabetes uses more resources and is more expensive than conventional therapy. On the other hand, intensive therapy is associated with a lower incidence of costly chronic complications. Formal economic analyses have demonstrated that intensive therapy is cost-effective for the treatment of diabetes. In IDDM, intensive therapy costs approximately $20,000 per QALY gained; in NIDDM, it costs approximately $16,000 per QALY gained. From an economic perspective, intensive therapy for persons with diabetes compares favorably with pharmacologic therapy for high-risk individuals with hypertension and hypercholesterolemia. Health policy should foster the use of such therapy for persons with diabetes mellitus.

Model of complications of NIDDM. I. Model construction and assumptions.

OBJECTIVE: To develop a model of NIDDM for analyzing prevention strategies for NIDDM. RESEARCH DESIGN AND METHODS: A Markov type model with Monte Carlo techniques was used. Age, sex, and ethnicity of cohort was based on U.S. data. Incidence rates of complications were also based on community and population studies. RESULTS: Nonproliferative retinopathy, proliferative retinopathy, and macular edema are predicted in 79, 19, and 52%, respectively, of people with NIDDM; 19% are predicted to develop legal blindness. Microalbuminuria, gross proteinuria, and end-stage renal disease related to diabetes are predicted in 53, 40, and 17%, respectively. Symptomatic sensorimotor neuropathy and lower-extremity amputation are predicted in 31 and 17%, respectively. Cardiovascular disease is predicted in 39%. Higher rates of complications (1.1-3.0x) are predicted in minority populations. Predicted average life expectancy is 17 years after diagnosis. CONCLUSIONS: A probabilistic model of NIDDM predicts the vascular complications of NIDDM in a cohort representative of the incident cases of diabetes in the U.S. before age 75 years. Predictions of complications and mortality are consistent with the known epidemiology of NIDDM. The model is suitable for evaluating the effect of preventive interventions on the natural history of NIDDM.

Model of complications of NIDDM. II. Analysis of the health benefits and cost-effectiveness of treating NIDDM with the goal of normoglycemia.

OBJECTIVE: To analyze the health benefits and economics of treating NIDDM with the goal of normoglycemia. RESEARCH DESIGN AND METHODS: Incidence-based simulation model of NIDDM was used. Hazard rates for complications were adjusted for glycemia using risk gradients from the Diabetes Control and Complications Trial. Treatment costs were estimated from national survey data and clinical trials. Incremental costs and benefits were expressed in present value dollars (3% discount rate). Life-years were adjusted for quality of life, yielding quality-adjusted life-years (QALYs). RESULTS: Comprehensive treatment of NIDDM that maintains an HbA1c value of 7.2% is predicted to reduce the cumulative incidence of blindness, end-stage renal disease, and lower-extremity amputation by 72, 87, and 67%, respectively. Cardiovascular disease risk increased by 3% (no effect of treating glycemia is assumed). Life expectancy increased 1.39 years. The cost of treating hyperglycemia increased by almost twofold, which is partially offset by reductions in the cost of complications. The estimated incremental cost/QALY gained is $16,002. Treatment is more cost-effective for those with longer glycemic exposure (earlier onset of diabetes), minorities, and those with higher HbA1c under standard care. CONCLUSIONS: The incremental effectiveness of treating NIDDM with the goal of normoglycemia is estimated to be approximately $16,000/QALY gained, which is in the range of interventions that are generally considered cost-effective.

Non-insulin-mediated glucose disappearance in subjects with IDDM. Discordance between experimental results and minimal model analysis.

Both insulin and glucose contribute to the regulation of glucose metabolism in vivo. We directly measured the ability of glucose per se to promote glucose disposal in subjects with insulin-dependent diabetes mellitus (IDDM). We compared our results with predictions of the minimal model of glucose metabolism. To identify minimal model parameters, a frequently sampled intravenous glucose tolerance test (FSIVGTT) was administered to each subject while they were connected to a Biostator (a device that monitors blood glucose and gives insulin to mimic normal insulin secretion). Data from this test reflected normal glucose tolerance and were in excellent agreement with minimal model predictions. The FSIVGTT was then repeated without the Biostator in the same diabetic subjects in order to directly measure the effect of glucose per se to promote glucose disposal in the absence of an incremental insulin effect (a basal insulin drip was maintained). To compare these results with minimal model predictions, the equations describing glucose disappearance in the absence of an incremental insulin effect were solved using parameters identified from the Biostator experiment. The glucose disappearance measured in the absence of an incremental insulin response was much slower than the minimal model predictions. Thus, the minimal model appears to overestimate the effect of glucose per se on glucose uptake and underestimate the contribution of incremental insulin.

Direct comparison of standard and insulin modified protocols for minimal model estimation of insulin sensitivity in normal subjects.

Estimates of in vivo insulin sensitivity (S1) can be derived from minimal model analysis of a frequently sampled intravenous glucose tolerance test (FSIVGTT). Modification of the FSIVGTT by administration of tolbutamide improves the correlation between S1 and more traditional glucose clamp measurements. Recently, a modified FSIVGTT using insulin infusion (instead of tolbutamide) has been described. This modification may be useful for studying subjects with poor insulin secretion. We directly compare the standard FSIVGTT with the insulin modified FSIVGTT in ten normal subjects (age 29 +/- 2 yr, BMI 22.5 +/- 1.2 kg/m2). Each subject received both tests. The order of the studies was randomized and the interval between paired studies was approximately one week. After an overnight fast, glucose (0.3 g/kg) was infused from time 0 to 2 min. For the modified FSIVGTT, insulin (4 mUkg-1.min-1) was infused from time 20 to 25 min. None of the subjects became hypoglycaemic after insulin infusion. Paired t-test analysis did not reveal significant differences between the two protocols for any of the minimal model parameters. The standard FSIVGTT gave estimates for S1 of 5.63 +/- 1.32 x 10(-4) min-1 per microU/ml while the insulin modified FSIVGTT gave estimates for S1 of 5.11 +/- 0.83 x 10(-4) min-1 per microU/ml. However, the fractional standard deviation for S1 was significantly smaller with the insulin modified protocol (3.6 +/- 1.0 vs. 22.2 +/- 9.0, p < 0.037). Thus, under our experimental conditions, insulin infusion during FSIVGTT appears to improve the precision of minimal model estimation of S1 in normal subjects.