Diabetes Mellitus: A Fundamental and Clinical Text
3rd Edition
© 2004 Lippincott Williams & Wilkins
←↑→
73
Preventing Type 2 Diabetes: The Diabetes Prevention Program
David M. Nathan
Richard F. Hamman
Sanford Garfield
Anne Brenneman
Sarah F. Fowler
William C. Knowler
The Diabetes Prevention Program Research Group
Type 2 diabetes mellitus has grown from an underappreciated, poorly understood chronic disease that was thought to have relatively few consequences to a highly prevalent, degenerative disorder that is the cause of much of the blindness, renal failure, amputations, and cardiovascular disease in the United States and Europe. Moreover, the diabetes epidemic is increasingly affecting the developing world as a “westernized” lifestyle, with its attendant obesity and decreased level of physical activity, is adopted. The combination of genetic and environmental risk factors promises to make type 2 diabetes one of the most widespread epidemics of the twenty-first century.
Although the major risk factors for the development of type 2 diabetes have been recognized for more than 50 years, the concept of preventing the disease is relatively new. In the past 5 years, however, several high-quality clinical trials have demonstrated that the development of diabetes is not inevitable in those at particularly high risk. Type 2 diabetes can be prevented or, at a minimum, delayed. The interventions that have been used successfully reflect our understanding of the pathophysiology of the disease and should provide effective strategies to help manage the diabetes epidemic.
Background
Type 2 diabetes, previously labeled as non–insulin-dependent or adult-onset diabetes, has only been clearly recognized as a distinct clinical entity since the 1930s (1). Insidious in onset and without the drama offered by the occurrence of ketoacidosis, this form of diabetes historically has escaped the attention given to type 1 diabetes. Although long considered a milder form of diabetes, type 2 diabetes has proved to be similar to type 1 diabetes as a cause of retinopathy, nephropathy, and neuropathy (2). In addition, because of the generally older age of most patients with type 2 diabetes and the common cooccurrence of cardiovascular disease (CVD) risk factors, type 2 diabetes is accompanied by a high prevalence of CVD, the major cause of death of patients with diabetes (3). The complications of diabetes cause the greatest human suffering and economic cost associated with the disease (4).
Type 2 diabetes affects 6% to 8% of the adult population in much of North America and Europe (5,6). Its prevalence is age related, with almost 20% of the population over 60 years of age affected. In addition, specific racial and ethnic groups are affected disproportionately. For example, African Americans, Hispanic Americans, Asian Americans and Pacific Islanders, and Native American Indians have a 1.5- to 8-fold greater risk of having type 2 diabetes than whites in the United States (7,8) (Fig. 73.1). The predilection for specific ethnic and racial groups and for first-degree relatives to be at greater risk for developing type 2 diabetes suggests a genetic basis for the disease. Studies of monozygotic twins have reinforced its genetic roots (9). Despite an intensive search for the genetic causes of this highly complex disease, the genetic basis for only a small fraction of type 2 diabetes has been discovered (10). The autosomally inherited maturity-onset diabetes of youth (MODY), which represents no more than 5% of all type 2 diabetes has, not surprisingly, been the first form of type 2 diabetes to yield its genetic secret (11). Several genetic polymorphisms that may contribute modestly to a larger fraction of type 2 diabetes have been described (12).
In order for the genetic predilection to develop type 2 diabetes to become manifest, a number of superimposed environmental or acquired factors appear to be necessary in most patients. The major acquired risk factors include overweight and obesity, as well as decreased physical activity, which contributes to obesity and plays an independent role as well (13,14,15). The underlying “westernized” lifestyle that contributes to these risk factors and its components, such as a diet high in fat and refined carbohydrates, and increasingly sedentary jobs, with decreased physical labor and increased television watching, have also been identified as contributors to the occurrence of type 2 diabetes (16,17).
Figure 73.1. Prevalence of diabetes in racial-ethnic groups in the United States. (Data from
Diabetes in special populations. In: Harris M, Cowie C, Stern M, et al. Diabetes in America, 2nd ed. (NIH publication no. 95–1468). Bethesda, MD: National Institutes of Health, 1995:613–702
; and
Diabetes Care 1993;16:157
.)
P.1078

Other less prevalent risk factors or diseases that increase the risk for type 2 diabetes, such as acromegaly, Cushing’s disease and syndrome, and a large number of medications, have been identified (18). A history of gestational diabetes is also a major risk factor for the future development of type 2 diabetes (19). Although some of these conditions are rare, they have provided insights into the pathogenesis of type 2 diabetes.
Thus, an increasingly obese, sedentary, and aging population has spawned the current diabetes epidemic in the United States and Europe, with similar patterns emerging in the developing world. Improved understanding of the risk factors, pathogenesis, and clinical course of the prediabetic state paved the way to identify modifiable risk factors, the populations that might respond to interventions, and the interventions that might effectively prevent the development of type 2 diabetes (20,21).
Pathogenesis and Clinical Course of Prediabetes and Type 2 Diabetes: Relevance to Prevention
Two separate but interconnected pathophysiologic characteristics underlie the majority of cases of type 2 diabetes: insulin resistance and relatively reduced insulin secretion (22,23). Although both of these factors probably have a genetic basis, only a small fraction of patients with type 2 diabetes with specific, causal abnormalities in the insulin receptor or in insulin secretion (e.g., MODY) have been identified (11). Despite the absence of a clear understanding of the specific causes for the majority of patients with type 2 diabetes, the usual course of metabolic abnormalities during the development of type 2 diabetes is well understood. Most persons who develop type 2 diabetes pass through a prediabetic state characterized by insulin resistance and consequent fasting, postchallenge, and postprandial hyperglycemia. During this period, plasma glucose levels increase progressively, secondary to a combination of decreased peripheral disposal into muscle and increased flux of glucose out of the liver (24). Using currently accepted prediabetic glycemic categories (18), the prediabetic state is most sensitively detected by measuring plasma glucose levels after a glucose challenge, such as a 75-g oral glucose tolerance test. Persons with impaired glucose tolerance (IGT), defined as a plasma glucose level between 140 and 200 mg/dL (7.8–11 mM) 2 hours after ingesting the 75 g of glucose, make up a large fraction of the prediabetic population. For most persons in the prediabetic state, fasting plasma glucose levels remain relatively normal, according to current criteria.
Although only a few long-term longitudinal studies have been performed, it appears that approximately one half of insulin-resistant persons remain in this stable metabolic state owing to a compensatory increase in insulin secretion in response to the insulin resistance and increased glucose levels. Thus, insulin resistance need not progress to diabetes if insulin secretion continues unabated. Metabolic stability is maintained with hyperinsulinemia.
At Least 50% of Persons with Impaired Glucose Tolerance Develop Diabetes
Those who develop diabetes do so because insulin secretion fails to match the heightened insulin requirements associated with insulin resistance (25). The high insulin levels that characterized the prediabetic state begin to decrease, postprandial glycemia worsens, and hepatic glucose output increases, resulting in fasting hyperglycemia (24). The waning insulin secretion probably has a genetic basis—abnormal insulin secretory patterns can be identified in nondiabetic family members of patients with type 2 diabetes (26,27)—and may also be acquired (23). As glucose and other intermediate metabolite levels increase, an acquired, and partially reversible, abnormality in insulin secretion develops. The so-called glucotoxicity or lipotoxicity that affects the β-cell may be specific to type 2 diabetes; at least some studies have demonstrated no obvious effect of experimental hyperglycemia for 72 hours on β-cell function in nondiabetics (28).
P.1079

The relatively well-delineated course of prediabetes as it progresses to diabetes and the pathophysiologic changes that underlie metabolic worsening provide target populations and potential interventions for the prevention of type 2 diabetes.
The Diabetes Prevention Program
The proposal by the National Institute of Diabetes, Digestive and Kidney Diseases (NIDDK) to explore whether type 2 diabetes could be prevented was motivated by a number of factors. The recognition of the need to stem the type 2 diabetes epidemic, and in particular seminal studies in the Pima Indians, provided a strong impetus to decrease the occurrence of diabetes. The successful conduct and completion of the Diabetes Control and Complications Trial in 1993 (29) also provided a template for successful clinical trials in diabetes. A Research Funding Announcement was issued in 1994, and 22 centers successfully competed for funding. An Native American Indian center, based at the Phoenix branch of NIDDK and encompassing four separate geographic sites, and a site in Hawaii were included to enrich the study cohort with Native American Indians and Pacific Islanders, respectively, bringing the total number of clinical centers to 27. The consensus study protocol was developed between 1994 and 1996 by the steering committee, which included the principal investigators from the clinical centers and coordinating center and representation from NIDDK, and by the Diabetes Prevention Program (DPP) subcommittees (30).
Design of the Diabetes Prevention Program and Major Goals
The DPP was designed as a four-arm, placebo-controlled randomized clinical trial to determine whether the development of diabetes could be prevented by any one of three interventions—intensive lifestyle modification, metformin, or troglitazone—compared with a placebo control group (30). The three medication arms were double blind and placebo controlled, but the intensive lifestyle intervention was not masked. The study had sufficient power to compare each of the active interventions and determine whether any of them was more effective than the others. An important secondary goal was to determine whether the interventions had an effect on the development of cardiovascular disease, atherosclerosis, or their risk factors. Other a priori goals included subanalyses to determine the relative efficacy of the interventions in men and women, older versus younger age participants, and in the minority racial-ethnic groups compared with whites.
The troglitazone intervention arm of the study was terminated in June 1998 owing to concern regarding potential liver toxicity associated with the drug. The 585 participants who had been randomly assigned to troglitazone therapy were unmasked to their treatment assignment at that time and were offered a lifestyle intervention program. Although they continued to be followed in the DPP, the premature termination of their intervention, after a mean drug exposure of approximately 9 months, and the additional lifestyle intervention made interpretation of their results problematic and not easily comparable with the remaining three arms of the study. For these reasons, the DPP continued as a three-arm trial, and the major results presented at study end (31) and reviewed here are for the three-arm trial.
The study was designed to have at least 90% power to detect a difference of at least 33% in the development of diabetes among treatment arms. Recruitment was to take place over a 3-year period, with follow-up ranging from 3 to 6 years. The primary outcome, time to development of diabetes as defined by the American Diabetes Association criteria (18), was to be analyzed by life-table analysis with an intent-to-treat approach.
Recruitment and Eligibility
The major recruitment criteria were designed to provide a study cohort that would fulfill the following requirements: a sufficiently high conversion rate to diabetes to make the study practical within the time and budgetary constraints provided by the funding; representativeness of the population with IGT in the United States, focusing on those minority populations that are disproportionately affected by diabetes; a high likelihood of adherence to the study requirements and completion of the study; and absence of clinical or other factors that might interfere with the conduct and completion of the study. The major recruitment criteria, compared with the other major prevention trials, are summarized in Table 73.1.
Potential volunteers were recruited through advertising, mass mailings, and community-based programs. A staged screening process determined whether volunteers fulfilled the recruitment criteria and were likely to comply with the study protocol (Table 73.1). Recruitment was initiated in May 1996 and was completed 3 years later, with a total of 3,815 volunteers,3,234 in the three-arm study and 585 in the troglitazone arm.
The study cohort reflected the eligibility criteria. The baseline characteristics have been reported in detail (32) and are summarized in Table 73.2, with a comparison to the populations recruited for the other major prevention trials (33,34,35).
Interventions
The interventions used by the DPP were selected after an exhaustive search for potentially effective interventions that had to fulfill the following criteria: preliminary evidence of efficacy in improving the metabolic state that underlies IGT and diabetes; safety, tolerability, and absence of significant side effects; ease of application in and potential acceptance by the diverse populations to be recruited; potential for translation in the general IGT population; and availability of a placebo to mask the medications. Although a series of hypoglycemic and antidiabetic medications were considered, the medications metformin and the thiazolidinedione troglitazone appeared to satisfy these criteria best. Because metformin was not approved for prevention and troglitazone had not been approved for any purpose at the inception of the study, the DPP operated under an investigational new drug (IND) application issued by the U.S. Food and Drug Administration. Medication therapy with metformin and metformin placebo was initiated with one 850-mg tablet per day, or one half of a tablet to reduce gastrointestinal side
P.1080

effects, and titrated to 850 mg twice per day. Programs to enhance adherence to medications were developed and implemented during the study. The medications used in the DPP are summarized in Table 73.3, along with the medications used in the other prevention trials.
Table 73.1. Major eligibility criteria of prevention trials
  DPP FDPS DaQing STOP-NIDDM
Age (yr) ≥25 40–65 >25 40–70
BMI (kg/m2) ≥24a ≥25   25–40
Glucose levels (mM)
   Fasting 5.8–6.9 <7.8 <7.8 5.6–7.7
   2-hour OGTT 7.8–11.0 7.8–11.0 7.8–11.0 7.8–11.0
a≥22 in Asian Americans.
DPP, Diabetes Prevention Program (U.S.) (31); FDPS, Finnish Diabetes Prevention Study (33);
DaQing study (35); STOP-NIDDM study (34). OGTT, oral glucose tolerance test.
The choice of goals for the intensive lifestyle intervention was predicated on previous studies that had identified the level of weight loss and physical activity necessary to ameliorate the abnormal metabolism of IGT and the levels that could be practically achieved. The means of achieving these goals and the specific elements of the intensive lifestyle program are described in Table 73.3, again contrasting them with the lifestyle interventions in the the Da Qing study (35) and Diabetes Prevention Study in Finland (33).
In the DPP, all volunteers were provided with standard healthy lifestyle recommendations, including advice regarding weight loss and exercise, and smoking cessation, at the beginning of the study and annually. Participants were encouraged to follow the food pyramid guidelines; to consume the equivalent of a National Cholesterol Education Program step 1 diet (36); to lose 5% to 10% of their initial weight through diet and exercise; to increase their activity gradually with a goal of at least 30 minutes of moderate intensity activity, such as walking, 5 days each week; to avoid excessive alcohol intake; and to stop smoking.
Table 73.2. Baseline characteristics of prevention trials
  DPP FDPS DaQing STOP-NIDDM
Number 3,234 522 530 1,429
Age (mean years) 51 55 45 51
Sex (% women) 68 67 47 51
Race (%)
   White 55 100   98
   African-American 20      
   Hispanic-American 16      
   American Indian 5      
   Asian-American or 4      
      Pacific Islander/Asian     100  
BMI (kg/m2) 34 31 26 31
Glycemia (mM)
   Fasting 5.9 6.1 5.6 6.2
   2-hour OGTT 9.1 8.8 9.0 9.3
DPP, Diabetes Prevention Program (U.S.) (31); FDPS, Finnish Diabetes Prevention Study (33); DaQing study (35); STOP-NIDDM study (34). BMI, body mass index; OGTT, oral glucose tolerance test.
Participants who were randomly assigned to the intensive lifestyle program had essentially the same goals as the standard lifestyle recommendations above, but the program was more intensive. The specific goals were to achieve and maintain a weight reduction of at least 7% of initial body weight through healthy eating and exercise and to achieve and maintain a level of activity of at least 150 minutes per week of moderate-intensity activity, such as walking or bicycling. Case managers provided a 16-session core curriculum program, administered at individual meetings over approximately the first 24 weeks. They contacted the participants at least monthly, with in-person contacts at least every 2 months, for the remainder of the study. The core curriculum incorporated a behavioral modification approach that taught participants to shop for, cook, and eat less fat and to consume fewer calories. The initial goal was to reduce dietary fat to less than 25% of total calories. If weight loss goals were not achieved, a calorie goal was added. Incorporating increased physical activity into daily activities was another major focus. The DPP intensive lifestyle program is available in its entirety (http://www.bsc.qwu.edu/dpp/manuals.htmlvdoc).
Outcomes
The masked intervention phase of the DPP was terminated in August 2001, 1 year ahead of schedule, at the recommendation of its Data Safety Monitoring Board and based on the results obtained. Retention of subjects during the average 2.8 years of the study (range 1.8–4.6 at study termination) was high, with 99.6% of subjects surviving and more than 93% attending annual visits and completing a follow-up visit within 6 months of study end.
The lifestyle intervention accomplished the goals established. Specifically, mean weight loss at the end of 6 and 12 months was at least 7%, with maintenance of more than 5% weight loss at the time of study end (Fig. 73.2). Similarly, the mean activity level exceeded the 150 minutes per week goal, with approximately 221 minutes per week of self-reported activity at 6 months and 189 minutes per week by study end. Adherence to the medication arms, based on pill count, was also high, with
P.1081

approximately 80% of the treatment groups assigned to placebo or metformin taking at least 80% of assigned medications.
Table 73.3. Summary of interventions in primary prevention trials
No. of interventions DPP 3 FDPS 2 DaQing 4a STOP-NIDDM 2
Lifestyle
   Exercise goal Moderate intensity ≥150 min/wk Moderate exercise 30 min/day Mild to very strenuous activity per day for 5–30 min, depending on level of activity  
Dietary
   Weight loss goal ≥7% ≥5% BMI of 23 for those with BMI ≥25 at baseline  
   Visit frequency 16 in first 24 wk, then monthly 7 in first year, then every 3 mo 7 in first 4 mo, and then every 3 mo  
   Diet prescription 25% fat, 1,200–2,000 kcal/day <30% fat, ≥15 g fiber/1,000 kcal 25%–30% fat, 55%–65% carbohydrate  
Medications Metformin 850 mg b.i.d.     Acarbose 100 mg t.i.d.
aControl, diet, exercise, and diet plus exercise.
DPP, Diabetes Prevention Program (U.S.) (31); FDPS, Finnish Diabetes Prevention Study (33);
DaQing study (35); STOP-NIDDM study (34). BMI, body mass index; b.i.d., twice daily; t.i.d., three times daily.
The intensive lifestyle and metformin regimens were both significantly more effective than placebo in the prevention of diabetes (Fig. 73.3). Lifestyle was also more effective than metformin. Compared with placebo treatment, lifestyle intervention and metformin reduced the development of diabetes by 58% and 31%, respectively (4.8 and 7.8 vs. 11.1 cases per 100 patient-years). The beneficial effects of these interventions were the same in men and women and the various ethnic-racial groups included. Of note, the lifestyle intervention was at least as effective in the oldest participants (≥60 years of age at baseline), with a 71% reduction compared with placebo, as in the younger participants. Metformin was less effective in the less overweight participants [body mass index (BMI) 24–30 kg/m2] and older participants and most effective in the heaviest participants (BMI ≤ 35 kg/m2).
Other Intervention Studies
The paucity of well-designed clinical trials investigating the feasibility of preventing type 2 diabetes in the past has now been addressed by the DPP and three other recent clinical trials seeking to determine the feasibility of diabetes prevention (33,34,35). [A fourth, relatively small, study that focused on Hispanic women with a history of gestational diabetes has also been completed (37).] In concert, they support the ability to prevent, or delay, the development of type 2 diabetes in diverse populations. A brief review of those studies and a comparison of differences and similarities with the DPP are summarized in Table 73.2 and Fig. 73.4. Lifestyle interventions have been the most common intervention applied and appear to be effective in diverse populations with different lifestyles. The benefit of lifestyle intervention has been shown to delay the development of diabetes in studies of 3 to 6 years’ duration. The two medications, metformin and acarbose, have been effective, albeit not as potent as the lifestyle interventions.
Conclusions
Type 2 diabetes, a chronic degenerative disease that is a consequence of longer lifespan and altered lifestyle superimposed on genetic risk, is likely to increase unless effective interventions are used to delay or stop its development. The recently concluded clinical trials have provided convincing proof that type 2 diabetes is not inevitable, even in those at high risk, and that interventions designed to reverse or ameliorate the pathophysiology of glucose intolerance can delay its onset. The uniform effect of the interventions in men and women and in all racial-ethnic groups studied in the DPP, and in the other trials, support the universal effectiveness of the interventions. On the other hand, the differential effects of lifestyle intervention and metformin by age and body mass suggest that selection of the interventions may be tailored to provide maximal efficacy and cost effectiveness.
Major questions remain in the wake of these clinical trials. The long-term effects of these interventions regarding the duration of diabetes prevention and their efficacy in preventing organ-specific complications, compared with waiting to treat
P.1082

P.1083

diabetes when it develops, need to be established. Similarly, the potential benefits of the interventions on the development of cardiovascular disease must be determined. With the longer-term outcomes better understood, cost-benefit analyses must be performed in order to determine the most efficient application of health-care dollars. Finally, the translation of these clinical trials into clinical care, including the most effective and efficient means of identifying persons at risk and of providing the interventions, must be carefully considered. The identification of effective strategies in the DPP and other clinical trials to prevent the development of type 2 diabetes is the first step in curtailing the diabetes epidemic.
Figure 73.2. Changes in body weight (A), leisure physical activity (B) and adherence to medication (C) according to DPP intervention. Each point represents the mean value for all participants examined at that time. The number of participants decreased over time because of the variable length of time that participants were in the study. For example, data on weight were available on 3,085 people at 6 months,3,064 at 1 year,2,887 people at 2 years, and 1,510 at 3 years. Changes in weight and leisure physical activity over time differed significantly among the treatment groups (p < 0.001 for each comparison). (Reprinted from
DPP Research Group. The Diabetes Prevention Program: reduction in the incidence of type 2 diabetes with lifestyle modification or metformin. N Engl J Med 2002;346:393–403
, with permission.)
Figure 73.3. Cumulative incidence of diabetes in the Diabetes Prevention Program according to intervention. The diagnosis of diabetes was based on the criteria of the American Diabetes Association (18). The incidence of diabetes differed significantly among the three groups (p < 0.001 for each comparison). (Reprinted from
DPP Research Group. The Diabetes Prevention Program: reduction in the incidence of type 2 diabetes with lifestyle modification or metformin. N Engl J Med 2002;346:393–403
, with permission.)
Figure 73.4. Reduction in incidence of diabetes in four major clinical trials compared with control intervention. The DaQing study included four treatment arms (control, diet, exercise, and diet plus exercise) (35). FDPS, Finnish Diabetes Prevention Study (33); STOP-NIDDM study (34); and DPP, Diabetes Prevention Program (U.S.) (31).
Acknowledgment
The Diabetes Prevention Program was supported by the National Institutes of Health through the National Institute of Diabetes and Digestive and Kidney Diseases, National Center on Minority Health and Health Disparities, the National Institute of Child Health and Human Development, and the National Institute on Aging; the Indian Health Service; the Centers for Disease Control and Prevention; the General Clinical Research Center Program, National Center for Research Resources; the American Diabetes Association; Bristol-Myers Squibb; and Parke-Davis. The DPP investigators are indebted to the participants for their dedication to the goal of preventing diabetes.
References
1. Himsworth HP. Diabetes mellitus: its differentiation into insulin-sensitive and insulin insensitive types. Lancet 1936;1:117.
2. Nathan DM. Long-term complications of diabetes mellitus. N Engl J Med 1993;328:676–685.
3. Panzram G. Mortality and survival in type 2 (non–insulin-dependent) diabetes mellitus. Diabetologia 1987;30:123–131.
4. Eastman RC, Javitt JC, Herman WH, et al. Model of complications of NIDDM. II. Analysis of the health benefits and cost-effectiveness of treating NIDDM with the goal of normoglycemia. Diabetes Care 1997;20:735–744.
5. Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in U.S. adults: the Third National Health and Nutrition Examination Survey, 1988–1994. Diabetes Care 1998;28:518–524.
6. King H, Rewers M, WHO Ad Hoc Diabetes Reporting Group. Global estimates for prevalence of diabetes mellitus and impaired glucose tolerance in adults. Diabetes Care 1993;16:157–177.
7. Edelstein SL, Knowler WC, Bain RP, et al. Predictors of progression from impaired glucose tolerance to NIDDM: an analysis of six prospective studies. Diabetes 1997;46:791–810.
8. Diabetes in special populations. In: Harris M, Cowie C, Stern M, eds. Diabetes in America, 2nd ed. National Institutes of Health Publication No 95–1468. Bethesda, MD: National Institutes of Health, 1995: 613–702.
9. Barnett AH, Eff C, Leslie RDG, et al. Diabetes in identical twins. Diabetologia 1981;20:87–93.
10. Lindgren CM, Hirschhorn JN. The genetics of type 2 diabetes. Endocrinologist 2001;11:178–187.
11. Stoffel M, Froguel P, Takeda J, et al. Human glucokinase gene: isolation, characterization, and identification of two missense mutations linked to early-onset non–insulin-dependent (type 2) diabetes mellitus. Proc Natl Acad Sci USA 1992;89:7698–7702.
12. Altshuler D, Hirschhorn J, Klannemark M, et al. The common PPARgamma Pro12Ala polymorphism is associated with decreased risk of type 2. Nat Genet 2000;26:76–80.
13. Rewers M, Hamman R. Risk factors for non–insulin-dependent diabetes. In: Harris M, ed. Diabetes in America, 2nd ed. National Institutes of Health Publication No. 95–1468. Bethesda, MD: National Institutes of Health, 1995:79–230.
14. Colditz G, Willett W, Rotnitzky A, et al. Weight gain as a risk factor for clinical diabetes mellitus in women. Ann Intern Med 1995;122;481–486.
P.1084

15. Manson JE, Nathan DM, Krolewski AS, et al. A prospective study of exercise and incidence of diabetes among US male physicians. JAMA 1992;268:63–67.
16. vanDam R, Rimm EB, Willett WC, et al. Dietary pattern and risk for type 2 diabetes mellitus in U.S. men. Ann Intern Med 2002;136: 201–209.
17. Hu FB, Leitzman MF, Stampfer M, et al. Physical activity and television watching in relation to risk of type 2 diabetes mellitus in men. Arch Intern Med 2001;161:1542–1548.
18. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183–1197.
19. Coustan D, Carpenter M, O’Sullivan P, et al. Gestational diabetes: predictors of subsequent disordered glucose metabolism. Am J Obstet Gynecol 1993;168:1139–1145.
20. Knowler WC, Narayan KVM, Hanson RL, et al. Preventing non–insulin-dependent diabetes mellitus. Diabetes 1995;44:483–488.
21. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 2001;345:790–797.
22. DeFronzo RA. The triumvirate: β-cell, muscle, liver: a collusion responsible for NIDDM. Diabetes 1988;37:667.
23. Leahy JL. Natural history of β-cell dysfunction in NIDDM. Diabetes Care 1990;13:992.
24. DeFronzo RA. The triumvirate: β-cell, muscle, liver: a collusion responsible for NIDDM. Diabetes 1988;37:667–687.
25. Brunzell JD, Robertson RP, Lerner RL, et al. Relationships between fasting plasma glucose levels and insulin secretion during intravenous glucose tolerance tests. J Clin Endocrinol Metab 1976;42:222.
26. Pimenta W, Korytkowski M, Mitrakou A, et al. Pancreatic beta-cell dysfunction as the primary genetic lesion in NIDDM: evidence from studies in normal glucose tolerant individuals with a first-degree NIDDM relative. JAMA 1995;273:1855.
27. O’Rahilly S, Turner RC, Matthews DR. Impaired pulsatile secretion of insulin in relatives of patients with non–insulin dependent diabetes. N Engl J Med 1988;318:1225.
28. Del Prato S, Leonetti F, Simonson D, et al. Effect of sustained physiologic hyperinsulinemia and hyperglycemia on insulin secretion and insulin sensitivity in man. Diabetologia 1994;37:1025–1035.
29. DCCT Research Group. The effect of intensive diabetes treatment on the development and progression of long-term complications in insulin-dependent diabetes mellitus: the Diabetes Control and Complications Trial. N Engl J Med 1993;329:978–986.
30. Diabetes Prevention Program Research Group. The Diabetes Prevention Program: design and methods for a clinical trial in the prevention of type 2 diabetes. Diabetes Care 1999;22:623–634.
31. DPP Research Group. The Diabetes Prevention Program: reduction in the incidence of type 2 diabetes with lifestyle modification or metformin. N Engl J Med 2002;346:393–403.
32. Diabetes Prevention Program Research Group. The Diabetes Prevention Program: baseline characteristics of the randomized cohort. Diabetes Care 2000;23:1619–1629.
33. Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001;344:1343–1350.
34. Chiasson J-L, Josse RG, Gorris R, et al. for the STOP-NIDDM Trial Research Group. Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomized trial. Lancet 2002;359:2072–2077.
35. Pan X-R, Li G-W, Hu Y-H, et al. Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. Diabetes Care 1997;20:537–544.
36. Step by step: eating to lower your high blood cholesterol. Bethesda, MD: National Heart, Lung, and Blood Institute Information Center, 1987.
37. Buchanan TA, Xiang HA, Peters RK, et al. Preservation of pancreatic β-cell function and prevention of type 2 diabetes by pharmacological treatment of insulin resistance in high risk Hispanic women. Diabetes 2002;51:2796–2803.
←↑→