1. Overall Design

The study is a randomized clinical trial to test the safety and efficacy of each of two interventions designed to prevent NIDDM. Participants with IGT and FPG values of 95-125 mg/dL [5.3-6.9 mmol/L] will be included; in addition we will emphasize recruitment of individuals with particularly high risk of development of NIDDM including those with obesity, the elderly, women with a history of gestational diabetes, and members of minority groups such as African Americans, Hispanic Americans, Asian and Pacific Islander Americans, and Native Americans.

Eligible volunteers will be stratified according to center and assigned to one of the two intervention or control groups during a 2-2/3-year recruitment period. The pharmacological intervention will be double blind and placebo controlled. Masking intensive lifestyle intervention assignment to the participants is not possible and masking the investigators is not practical.

After randomization, participants will have quarterly clinical evaluations with a fasting plasma glucose at semi-annual visits and a 75 gram oral glucose tolerance test at annual visits. All participants will be followed for a minimum of 3 1/3 years after the close of recruitment resulting in 3 1/3 to 6 years of participant follow-up.

A total of 3,000 participants (1,000 per group) will provide 90% power to detect a 33% reduction in the progression to diabetes, assuming an annual rate of progression to diabetes in the control group of at least 6.5% and a level of significance of 5% (two-sided) with adjustment for pairwise comparison of the three treatment groups.

2. Participation Criteria

2.1 Principles Guiding Selection of Criteria

• Ensure selection of individuals with a high risk of progression to NIDDM to provide adequate statistical power within the planned duration of the DPP.
• Exclude individuals with conditions or treatments that would interfere with participation in or completion of the protocol, or that have a confounding effect on the measurement of the primary outcomes of the study.
• Exclude individuals who would be at high risk for adverse effects from the proposed interventions.
• Design criteria that can be measured accurately and efficiently in the setting of a multicenter trial that includes heterogeneous populations.

Detailed criteria and methods are described in the Manual of Operations.

2.2 Inclusion Criteria

1. Ethnicity: All ethnic groups



2. Gender: Men and Women



3. Age
   
   
   1. Lower age limit: 25 years

      Recommend focus on ages 35 years and older, except for Native Americans, women with a history of gestational diabetes, or women with polycystic ovary syndrome.

   2. Upper age limit: NONE

      Will only exclude for cause (e.g. diseases, functional limitations detailed below).

4. Impaired glucose tolerance with elevated fasting plasma glucose based on a single 75 gm OGTT
   
   
   1. Fasting plasma glucose 95-125 mg/dL (5.3-6.9 mmol/L)
      

      AND

   2. 2-hr plasma glucose 140-199 mg/dL (7.8-11.0 mmol/L)

      Individuals who meet only one of these glucose criteria may be rescreened after 6 months. Because of the relative higher rate of progression from impaired glucose tolerance to diabetes in Native Americans and the small size of the population, the glucose requirement for eligibility in the Southwest American Indian Center will be fasting glucose <126 mg/dL and 2-hr. plasma glucose 140-199 mg/dL.

5. Body-mass Index
   
   
   1. Lower BMI limit: 24 kg/m²
      

      Because Asian-Americans develop diabetes and IGT at a BMI that is lower than the BMI of the general U.S. population with IGT and NIDDM, the BMI criteria for eligibility for Asian Americans will be 22 kg/m² or greater.

   2. Upper BMI limit: NONE

2.3 Exclusion Criteria

1. Exclusion for underlying disease likely to limit life span and/or increase risk of interventions
   
   
   1. Cancer requiring treatment in the past 5 years, with the exception of cancers which have been cured or, in the opinion of the investigator, carry a good prognosis.

      For example, non-melanoma skin cancer, papillary thyroid carcinoma, and cervical carcinoma in situ.

   2. Infectious diseases
      1. Self-reported HIV positivity

         No serologic testing.

      2. Active tuberculosis

   3. Cardiovascular disease
      1. Hospitalization for treatment of heart disease in the past 6 months

         Includes therapeutic procedures (e.g. CABG, PTCA), not diagnostic procedures (e.g. coronary angiogram).

      2. New York Heart Association Functional Class > 2
      3. Left bundle branch block on ECG
      4. Third degree atrioventricular block on ECG
      5. Uncontrolled hypertension: SBP >180 mmHg or DBP >105 mmHg on treatment

         Milder degrees of hypertension would prompt referral to primary care provider for treatment (with DPP compatible agent) but would not exclude.

      6. Stroke or transient ischemic attack in the past 6 months
      7. New York Heart Association Functional Class 2 in persons who are currently treated with a loop diuretic or digitalis preparation
   
   
   4. Gastrointestinal disease
      1. Self-reported chronic hepatitis or cirrhosis, or serum AST or ALT elevated by the following criteria:
         
         • serum AST 66 U/L
           
         • serum ALT 58 U/L if over 47 years
           
         • serum ALT 118 U/L if male 47 years
           
         • serum ALT 46 U/L if female 47 years
      2. Episode of alcoholic hepatitis or alcoholic pancreatitis ever
      3. Inflammatory bowel disease requiring treatment in the past year
      4. Recent or significant abdominal surgery (e.g., gastrectomy)

         Would not exclude surgery for conditions with no long-term adverse effects (e.g., appendectomy, cholecystectomy, polypectomy).

   5. Renal disease
      1. Serum creatinine 1.4 mg/dL (124 mmol/L) for men; 1.3 mg/dL (115 mmol/L) for women
      2. Urine protein 2+ on one occasion (dipstick), in the absence of infection or vaginal contamination
      3. In individuals who are or will become 80 years of age during the study, a direct measure of creatinine clearance, based on a 24-hour urine collection, will be required. Creatinine clearance levels 75 mL/min will be required in order for these individuals to be eligible.
   
   
   6. Lung disease
      1. Chronic obstructive airways disease or asthma requiring daily therapy
      2. New York Heart Association Functional Class > 2
      3. Use of oxygen at home
   
   
   7. Electrolyte abnormality

      Serum potassium <3.2 or >5.5 mmol/L

      Serum sodium and bicarbonate levels are reported to the clinical center during screening to alert investigators to participants with electrolyte or acid-base levels outside expected ranges.

   8. Anemia

   Hematocrit < 36.0% in men or <33.0% in women

   9. Other chronic disease or condition likely to limit life span to < 6 years.
   
   
   
   10. Conditions not specifically mentioned above may serve as criteria for exclusion at the discretion of the clinic site.


2. Exclusion for conditions or behaviors likely to effect the conduct of the DPP
   1. Unable or unwilling to give informed consent
   2. Unable to communicate with the pertinent clinic staff
   3. Another household member is a participant or staff member in DPP
   4. Unwilling to accept treatment assignment by randomization
   5. Current or anticipated participation in another intervention research project that would interfere with any of the interventions offered in DPP
   6. Weight loss of >10% in past 6 months for any reason except post-partum weight loss
   7. Likely to move away from participating clinics in next 5 years
   8. Unable to walk 0.25 mile in 10 minutes
   9. Unable to complete remainder of DPP run-in tasks
   10. Pregnancy and childbearing

   None of these criteria are necessarily permanent. Women may be considered potentially eligible after sufficient time has elapsed or after other relevant conditions have changed.

   1. Currently pregnant or less than 3 months post-partum
   2. Currently nursing or within 6 weeks of having completed nursing
   3. Pregnancy anticipated during study

      Recognizing the difficulty of foreseeing events this far in the future, women who answer "probably not" would be included. Women with definite plans to become pregnant would be excluded.

   4. Unwilling to undergo pregnancy testing or to report possible or confirmed pregnancies promptly during the course of the DPP
   5. Unwilling to take adequate contraceptive measures, if potentially fertile

      Clinic sites would determine how far to probe regarding type of method, conduct of method, and likelihood of fertility (medical history, level and type of sexual activity).

   11. Major psychiatric disorder which, in opinion of clinic staff, would impede conduct of the DPP

       Concern regarding such a disorder would be prompted initially either by clinic staff perceptions or by self-report of psychiatric hospitalization. Confirmatory evidence could include scores on standardized instruments (e.g., Brief Symptom Inventory or Mini-Mental Status Examination) or clinical evaluation.

   12. Excessive alcohol intake, either acute or chronic

       Defined as any one of the following: (1) average consumption of 3 or more alcohol containing beverages daily; (2) consumption of 7 or more alcoholic beverages within a 24 hr period in the past 12 months; or (3) clinical assessment of alcohol dependence based on two or more positive responses to the CAGE questionnaire (if confirmed by further probing) or on other evidence available to clinic staff. If any of these exclusion criteria are met, the subject may still be considered eligible if, after an explanation of the importance of limiting alcohol intake during the DPP, the clinic staff believes that the volunteer can and will limit future alcohol intake to acceptable levels.

   13. Other condition which, in opinion of clinic staff, would effect the conduct of the DPP


3. Exclusions related to metabolism
   
   1. Diabetes at baseline evaluation evidenced by any of the following:
      1. Fasting plasma glucose 126 mg/dL [7.0 mmol/L]
      2. 2-hour plasma glucose 200 mg/dL [11.1 mmol/L] based on 75 gm OGTT
      3. Diabetes diagnosed by a physician and confirmed by other clinical data (e.g. documentation of a fasting plasma glucose 126 mg/dL [7.0 mmol/L] or a positive OGTT)
         
         Persons who report having been diagnosed with diabetes but who have neither taken hypoglycemic medication nor have confirmatory clinical evidence of diabetes may undergo OGTT at the discretion of the clinic site.
      
      
      
      4. Ever used hypoglycemic medication (oral agents or insulin; except during GDM)
   
   
   2. Disease associated with disordered glucose metabolism
      1. Cushing's syndrome
      2. Acromegaly
      3. Pheochromocytoma currently under treatment (i.e., not surgically cured)
      4. Chronic pancreatitis
   
   
   3. Thyroid disease, suboptimally treated

      Suboptimal treatment is defined as a sensitive TSH assay outside the clinically acceptable range (except for low TSH levels in patients being treated with suppressive doses because of a history of thyroid nodules or thyroid cancer). TSH would be assessed in these settings: (1) reported use of anti-thyroid medication, (2) reported use of thyroid hormone.

   4. Fasting plasma triglyceride level > 600 mg/dL (6.77 mmol/L) on one occasion despite treatment
   5. Conditions not specifically mentioned above may serve as criteria for exclusion at the discretion of the clinic site


4. Exclusions related to medications
   
   1. Antihypertensives
      1. Thiazide diuretics
      2. Beta-blockers
   
   
   2. Lipid-lowering agents--Niacin only

      Statins, bile acid sequestrants, and gemfibrozil would not be excluded. 

   3. Glucocorticoids other than topical, ophthalmic, and inhaled preparations
   
   
   
   4. Antibiotics
      1. HIV-related agents, e.g., AZT, DDI, pentamadine
      2. Antituberculous agents, e.g., INH, ethambutol (except INH alone as prophylaxis)
   
   
   5. Antineoplastic agents
   
   
   
   6. Psychoactive agents
      1. Antipsychotic agents

         These include, but are not limited to Haldol, Loxitane, Mellaril, Navane, Prolixin, Stelazine, Thorazine, and Trilafon.

      2. Fluoxetine (Prozac) at doses exceeding 20 mg daily, or other selective serotonin reuptake inhibitors (SSRI) at comparable doses.

         Low dose fluoxetine and other antidepressant medications will not be excluded.

   7. Bronchodilators
      1. Aminophylline, if used daily
      2. Inhaled beta-agonists, if used daily
   
   
   8. Other medications
      1. Phenytoin
      2. Amphetamines
      3. Prescription weight-loss drugs
   
   
   9. Medications not specifically mentioned above may serve as criteria for exclusion at the discretion of the clinic site

3. Principles Guiding Selection of Interventions

The interventions considered for the DPP included all available therapies that had demonstrated efficacy in improving the abnormalities in glucose metabolism that characterize IGT and NIDDM by decreasing fasting and/or post-challenge glucose levels. The available modalities were scrutinized with regard to the following additional criteria: strength and consistency of data supporting efficacy, safety, and tolerability including side-effect profile at the doses required for efficacy; potential acceptance by participants; potential for long-term adherence; ability to translate to a non-research setting; other desirable effects such as ability to improve cardiovascular risk factors; and ability to mask the intervention (availability of placebo). No single intervention fulfilled all of the criteria above. However, three interventions were considered to have the requisite qualities necessary for the DPP.

An intensive lifestyle intervention, focusing on dietary changes to promote weight loss and on increased physical activity, was selected because of the recognized short-term salutary effects of such a regimen on glucose metabolism, dyslipidemia, and other cardiovascular risk factors. The scientific literature (reviewed previously) provides support for the hypothesis that decreases in body weight and increases in physical activity ameliorate hyperglycemia and may prevent or delay the development of NIDDM. In addition, decreasing waist-to-hip ratio may decrease the risk of NIDDM; however, modification of waist-to-hip ratio is best accomplished by losing weight and increasing physical activity. In addition, the safety profile and low frequency of adverse side effects complemented the appeal of a nonpharmacological intervention. Although other features of intensive lifestyle intervention, such as the inability to mask participants and investigators from knowing treatment assignment, did not fulfill the criteria, the positive attributes were considered to outweigh any shortcomings.

Two pharmacological interventions were selected. The biguanide metformin was selected on the basis of its demonstrated efficacy to improve glucose intolerance (UKPDS, 1995; DeFronzo, et al., 1995) and the expectation that it would prevent or delay progression of IGT to NIDDM. In addition, use of metformin was supported by its safety profile, the low frequency of adverse effects at the doses necessary to obtain the desired metabolic effect, the strength and consistency of data supporting efficacy, potential for long-term adherence, ability to translate in the non-research setting, other desirable effects such as the ability to improve cardiovascular risk factors, and the availability of placebo. Troglitazone, a member of a new class of thiazolidinedione drugs termed insulin sensitizers, was selected on the basis of its potent effect to lower glycemia in NIDDM patients without an increase in insulin levels, its ability to improve lipid levels, the absence of significant side-effects or adverse events in short-term (1-2 year) human studies, the potential for translation in the non-research setting, and the availability of placebo (Nolan, et al., 1994; Valiquett, et al., 1995). Sulfonylureas, even in the relatively low doses that have been used in past studies of prevention, are associated with hypoglycemia. Because of this potential side-effect, sulfonylureas were not considered further for use in the DPP.

Randomization to the troglitazone pharmacological intervention was suspended on May 27, 1998, and discontinued by the NIDDK on June 3, 1998. The NIDDK, with input from the DPP Data Monitoring Board, decided to discontinue use of troglitazone in the DPP based on the following: "Associated with troglitazone use in the DPP, there is an increased risk of liver toxicity resulting in serum ALT levels greater than or equal to 8 times the upper limit of normal. In the DPP, there has been one case of hepatic failure requiring liver transplantation. Within the context of this research trial, safety monitoring, even if intensified, is not likely to eliminate the risk. It is also too early in the trial to estimate reliably and compare the absolute risk or benefit of continuing troglitazone in this study population. We are not willing to continue to study a drug for the purpose and benefit of preventing progression from impaired glucose tolerance to diabetes when the drug has demonstrated liver toxicity with hepatic failure."

Participants randomized to troglitazone prior to May 27, 1998, were unmasked to their intervention assignment, monitored to ensure that no liver toxicity developed after discontinuing the troglitazone medication, and continue follow-up by the DPP Research Group (see the DPP substudy protocol "Follow-up of DPP Participants Randomized to Troglitazone"). Pharmacological participants randomized to metformin or double-placebo, discontinued their troglitazone-placebo, continue their coded metformin medication, and remained masked to their pharmacological assignment (metformin or metformin-placebo).

[Top]

4. Assignment to Treatment Groups

4.1 Stratification

Randomization will be stratified by clinical center. This will ensure balance between the three treatment groups with respect to anticipated differences in the participant populations and possible differences in participant management.

4.2 Randomization Method

There are several alternative methods to assign the participants randomly within clinical center (e.g., simple randomization or permuted block designs). The urn method of randomization provides a high probability of balance in treatment assignments, is unpredictable in unmasked studies, and allows an explicit randomization analysis to be conducted with relative ease (Wei and Lachin, 1988). For these reasons, the urn method will be used to randomly assign participants to the three treatment groups.

A sequence of randomization numbers within a clinical center will be constructed of the form XXYZZZ, where XX is the clinical center number, Y is a number that indicates assignment to either the intensive lifestyle intervention or pharmacological treatment, and ZZZ is a three digit sequence number within each XXY combination. The DPP Coordinating Center will prepare the master randomization list with assignments to the three treatment groups within a clinical center using the standard urn design.

The sequence of pharmacological randomization numbers within a clinical center with the specific pharmacological treatment assignment (i.e., metformin or placebo) will be forwarded, in confidence, to the drug distribution center for drug labeling and distribution. Pharmacological treatment assignment to the sequence of pharmacological randomization numbers will be known only by the staff of the DPP Coordinating Center and the drug distribution center.

[Top]

5. Timing and Conditions of Outcome Assessment

5.1 Primary Outcome

The primary outcome, progression from impaired glucose tolerance (IGT) to diabetes, is assessed by OGTT testing annually, by fasting plasma glucose (FPG) every 6 months, and at the time that symptoms consistent with hyperglycemia occur. Conditions for the OGTT are specified in the Manual of Operations. The annual OGTT and 6-month FPG will be postponed for up to 6 weeks if a temporary concomitant condition exists that would affect glucose tolerance. An OGTT that is positive for diabetes, or a 6-monthly FPG that is 126 mg/dL [7.0 mmol/L], will be repeated for confirmation before the participant is considered to have reached the primary outcome. When a participant has been in a "time-out" (other than pregnancy), such as for a concomitant disease known to affect glucose tolerance, the primary outcome will be assessed at the time of the next regularly scheduled 6-monthly FPG or annual OGTT after that time-out ends.

5.2 Secondary Outcome

Secondary outcomes will be assessed according to the Section 12 schedule.

6. Level of Masking

6.1 Treatment Groups

Pharmacological treatment assignment (metformin or placebo) will be double masked. Masking participants to the intensive lifestyle intervention versus pharmacological treatment is not possible and masking the investigators is not practical. At no time will the code of the specific pharmacological treatment assignment be broken without the express knowledge and consent of the clinical center's principal investigator. The exception to this policy is that if a woman becomes pregnant while on study medication, the medication will be discontinued and she, her PI, and her Primary Care Provider will be unmasked. Information on the potential teratogenicity of the medication will be provided to both the participant and her care providers.

6.2 Central Laboratory Outcomes

Primary outcome data (OGTT and FPG results) measured centrally will remain masked to the investigators and to the participants until confirmed progression from IGT to diabetes. Plasma lipid levels and HbA_1c measured centrally will remain masked to the investigators and to the participants during the study. Two exceptions to this policy are (1) if an individual result falls outside a pre-determined, clinically acceptable range representing a significant risk to the participant, it will be confirmed. If the result remains abnormal, the investigator and the participant will be notified and appropriate clinical steps will be taken. (2) If an individual participant insists on knowing specific secondary outcome results, such as plasma lipid values, and the person's continued participation in DPP is considered by the investigator to be in jeopardy, the investigator may request an exception from the Coordinating Center.

6.3 Clinical Outcomes

Some of the secondary outcome data, such as blood pressure, weight, and waist circumference, will be measured at the clinical center. It is not feasible to keep these data masked from the participants, if they want to know their result. Therefore, participants will be unmasked to measurements of body size and blood pressure.

6.3.1 Data Collectors

In order to promote objectivity of data collection and to minimize the opportunity for bias, the intent is to separate outcome measurement from the intensive lifestyle intervention case managers. This is particularly important for dietary intake data, blood pressure, interview questionnaires, and anthropomorphic measures, where the potential exists for subjectivity. Intensive lifestyle intervention case managers must not perform these outcome measures for participants with whom they are intervening.