The Diabetes Prevention Program: Findings and Public Health Implications

1
The Diabetes Prevention Program Findings and
Public Health Implications

• Gloria L.A. Beckles, MD, MSc
• Division of Diabetes Translation
• NCCDPHP, CDC

12th Annual Maternal Child Health Epidemiology
Conference December 6-8, 2006 Atlanta, GA
2
Disclaimer

• The findings and conclusions in this
  presentation have not been formally
• disseminated by the Centers for Disease Control
  and Prevention and should not be construed to
  represent any agency determination or policy.

3

• Background

4
There are entirely too many diabetic patients in
the country. Statistics for the last thirty years
show so great an increase in the number that,
unless this were in part explained by a better
recognition of the disease, the outlook for the
future would be startling.
Joslin EP. The Prevention of Diabetes Mellitus
JAMA 19217679-84.
5
Trends in Prevalence of Total, Diagnosed, and
Undiagnosed Diabetes U.S Adults aged 20-74 years
Age-, Sex-, Race/Ethnicity -standardized
Beckles, GLA, et al. Diabetes 200655.
6
Burden of Diabetes U.S. mid-1990s

• In the mid-1990s,
• An estimated 16 million adults had DM
• Huge individual and societal burden recognized
• Associated with 2-3-fold excess risk of heart
  attack and stroke
• the leading cause of new blindness, end-stage
  renal disease, and LE amputations
• Accounted for 17 of all deaths after age 25
• Cost approx. 100 billion per year

7
Rationale for a Prevention Trial

• Evidence is that the increasing trend in
  prevalence is being driven by increasing
  incidence of type 2 DM
• Once type 2 DM develops, it is difficult to
  treat.
• Prevention of type 2 DM would result in a
  significant reduction in social and economic
  costs.

8
Feasibility of Prevention

• Prevention of type 2 DM should be feasible since
  
• Long asymptomatic period that precedes
  development of the disease
• Screening tests can identify persons at high
  risk
• Safe, potentially effective interventions for
  the modifiable risk factors
• Observational studies and 2 clinical trials of
  lifestyle change

9
Primary Research Questions

• Does a lifestyle intervention or treatment with a
  hypoglycemic agent prevent or delay the onset of
  diabetes in persons at high risk?
• If yes, do the two interventions have different
  effectiveness?
• Does effectiveness of either intervention differ
  by age, sex, or race/ethnicity?

10

• Population Methods

11
The Diabetes Prevention Program A Randomized
Clinical Trial to Prevent Type 2 Diabetes in
Persons at High Risk 27 sites
Sponsors NIH, NIDDK, NIA, NICHD, IHS, CDC, ADA
and other agencies and corporations
12
Eligibility Criteria

• Impaired Glucose Tolerance (IGT)
• Fasting glucose 95-125 mg/dL (5.3 - 6.9 mmol/L)
• AND
• 2-hour glucose 140-199 mg/dL (7.8 - 11.0 mmol/L)
• Body mass index 24 kg/m2
• Age 25 years
• Men and Women
• All ethnic groups
• goal of 50 from high risk populations

13
Characteristics of Ramdomized Cohort
The DPP Research Group. Diabetes Care
2000231619-29
14
GDM Study Cohorts in DPP
History of GDM
No history of GDM
Hispanic American 17
Hispanic American 16
American Indian 10
Asian 3
Asian 2
American Indian 7
African American 18
African American 23
Caucasian 50
Caucasian 54
15
Study Interventions

• Intensive Lifestyle program with specific aims
• Body Weight 7 loss
• Dietary fat lt 25 calories per day from fat
• Caloric intake 1200 1800 kcal per day
• Physical activity 150 minutes per week
• Metformin ( standard lifestyle recommendations)
  twice daily
• Placebo ( standard lifestyle recommendations)
  twice daily

16
Study Interventions
Eligible participants Randomized Standard
lifestyle recommendations
Metformin (n 1073)
Placebo (n 1082)
Intensive Lifestyle (n 1079)
17

• Results

18
Primary Outcome Incidence of Diabetes
Placebo
Metformin vs. Placebo (lt0.001)
Lifestyle vs. Metformin (plt0.001) vs. Placebo
(plt0.001)
The DPP Research Group, NEJM 346393-403, 2002
19
Effect of Treatment on Incidence of Diabetes

• Placebo Metformin Lifestyle
• Incidence of diabetes 11.0
  7.8 4.8
• (percent per year)
• Reduction in incidence ---- 31
  58
• compared with placebo
• Number needed to treat ---- 13.9
  6.9
• to prevent 1 case in 3 years

The DPP Research Group, NEJM 346393-403, 2002
20
Diabetes Incidence Rates by Sex
The DPP Research Group, NEJM 346393-403, 2002
21
Diabetes Incidence Rates by Age
The DPP Research Group, NEJM 346393-403, 2002
22
Diabetes Incidence Rates by Ethnicity
The DPP Research Group, NEJM 346393-403, 2002
23
Diabetes Incidence Rates by BMI

The DPP Research Group, NEJM 346393-403, 2002
24
Consistency of Treatment Effects

• Lifestyle intervention was beneficial regardless
  of sex, age, ethnicity, or BMI
• The efficacy of lifestyle relative to metformin
  was greater in older persons and in those with
  lower BMI (lt 30 kg/m2)
• The efficacy of metformin relative to placebo was
  greater in those with BMI 30 kg/m2

25
Implications for Public Health - 1

• Identification of those at risk
• Screening for Impaired Glucose Tolerance?
• Those with poor access to health care?
• Ethics of just screening without follow-up?
• Translation of intervention(s)
• What dose (frequency, intensity, duration) can
  be achieved in real world settings?
• What is minimally necessary compliance?
• Role of extra-clinical interventions?
• Environmental change to support /sustain
  lifestyle?
• Community-level activities vs. go-it-alone
  lifestyle

26
Implications for Public Health - 2

• Evaluation of overall efforts
• Time-Place-Person Trends in Type-2 Diabetes
• e.g. population Surveillance
• Adequacy of the Intervention (s) Process
• e.g., population surveys of exposure