What is the Diabetes Prevention Program Outcomes Study (DPPOS)?

The DPPOS aims to assess the long-term effects of interventions used in the Diabetes Prevention Program (DPP) on the development of type 2 diabetes and its complications. Eighty-eight percent of all surviving DPP participants who were eligible joined the DPPOS.

DPPOS Phase 1

The DPP Outcomes Study (DPPOS), an eleven-year follow-up study, was funded in 2002 for a five-year period with the understanding that completion of the study would require re-funding via competitive renewal.  (In 2004, NIDDK administratively approved an additional year of funding recognizing that collecting and analyzing five-year data in a study of this size would require six years.)  The overarching goal of the DPPOS was to study whether the relatively short-term benefits of delaying diabetes demonstrated in the DPP would translate into a more long-lasting impact that would reduce the public health burden of the diabetes epidemic.  Specifically, DPPOS had the following major goals: 1) to examine the long-term effects and durability of prior DPP interventions on further diabetes development; 2) to determine the effects of DPP interventions on aggregate diabetes-associated microvascular outcomes; 3) to determine the effects of DPP interventions on aggregate clinical cardiovascular disease (CVD), atherosclerosis, and CVD risk factors; and 4) to describe the incidence of CVD, CVD risk factors and atherosclerosis in IGT and new onset type 2 diabetes.  In addition to numerous other secondary outcomes, including the impact of these primary outcomes on health economics and quality of life, the diversity of the DPPOS cohort provided the opportunity to address the goals in diverse ethnic-racial groups and by age and sex. 

During a bridge period between the end of DPP and the beginning of DPPOS, all participants were instructed in the lifestyle intervention.  The DPPOS design included an effort to maintain the original DPPOS metformin group on metformin treatment and to continue a lifestyle "boost" program for the original lifestyle intervention group.  During DPPOS, participants are seen semi-annually in clinic, with visits similar to those during DPP.

Results from DPPOS Phase 1 have already shown that the effects of the original interventions persisted for at least 6 years after the end of the original trial, albeit with somewhat reduced efficacy (Lancet, 2009).  Although determining the effects of diabetes prevention on CVD outcomes was considered a high priority, the cohort was recognized to be at relatively low risk for CVD events, and thus DPPOS was underpowered for CVD outcomes.  However, mid-way through the DPPOS, CVD risk factors were shown to be reduced in the original lifestyle intervention group in the setting of fewer hypolipidemic and anti-hypertensive agents being used (Lancet 2009).

What are the main findings of the DPPOS Phase 1?

After an average of 10 years' follow up, intensive lifestyle changes aimed at modest weight loss

  • reduced the rate of developing type 2 diabetes by 34 percent compared with placebo.
  • reduced the rate of developing type 2 diabetes by 49 percent in those age 60 and older compared with placebo.
  • delayed type 2 diabetes by about 4 years compared with placebo.
  • reduced cardiovascular risk factors.
  • reduced hemoglobin A1c (A1C) and fasting glucose compared with placebo. The A1C test gives information about average blood glucose levels for the past 2 to 3 months.

After an average of 10 years' follow up, treatment with metformin

  • reduced the rate of developing diabetes by 18 percent compared with placebo.
  • delayed diabetes by 2 years compared with placebo.
  • reduced A1C and fasting glucose compared with placebo.

DPPOS Phase 2

DPPOS Phase 2 examined the longer-term effects of the original DPP interventions on microvascular and neuropathic outcomes (jointly, as primary outcome), diabetes prevention, conventional CVD risk factors, biochemical markers of atherosclerosis and inflammation, and atherosclerosis measured by coronary artery and visceral calcification.  The question of whether there is a beneficial impact on the aggregate early-stage microvascular (retinopathy, nephropathy, and neuropathy) outcomes will be answered and the analyses of cost-effectiveness will be considerably enhanced through this understanding.  By the end of this phase, 87 percent of the surviving DPP cohort joined DPPOS with minimal loss to follow-up among those enrolled in DPPOS.  There has been 86-99% attendance at annual follow-up visits, and >99% of study data have been collected at those visits.  

What are the main findings of the DPPOS Phase 2?

DPPOS Phase 2 found no significant differences in the prevalence of the aggregate microvascular outcome (comprised of fundus photography measured retinopathy, nephropathy based on eGFR or microalbuminuria, or neuropathy detected by abnormal monofilament sensation) among the treatment groups (Lancet D&E, 2015). However, in a prespecified analysis by sex, women had a 21% and 22% reduced risk of this composite microvascular outcome in the ILS group compared with the placebo and metformin groups, respectively (p<0.05). Moreover, among the participants who had not developed diabetes during DPP/DPPOS, the prevalence of the aggregate microvascular outcome was 28% lower compared with those who had developed diabetes (p<0.001). In a post-hoc analysis among those whose latest HbA1c was ≥6.5%, representing ~25% of the cohort, lifestyle showed significant (p<0.05) reductions in the aggregate microvascular outcome (41% reduction), retinopathy (49%), and neuropathy (61%), compared with placebo. These observations highlight that the DPP/DPPOS population is being observed early in the long-term course of diabetes, with accelerating glycemia-related risks over time.

DPPOS Phase 3

DPPOS Phase 3 was designed to evaluate the long-term effects of the original DPP interventions, in particular the effect of metformin compared with placebo on total cancer at the end of the first 5 years of Phase 3 and MACE (Major Adverse Cardiovascular Events: MI, stroke or CVD death) at the end of the second 4 years. Secondary aims include the evaluation of the long-term intervention effects on continued diabetes prevention, conventional CVD risk factors, biochemical markers of atherosclerosis and inflammation, and individual components of the microvascular outcome (retinopathy, nephropathy, and neuropathy). Additional major secondary outcomes include aging-related outcomes, such as quality of life, activities of daily living, physical function, and cognitive function. At the request of the National Institute on Aging (NIA), DPPOS added spirometry as a measure of pulmonary function, a 6-minute walk test as a measure of cardiorespiratory fitness, and conducted a pilot study to obtain medical records to adjudicate hospitalized infections and fractures in 2016-17. NIA also funded another DXA measurement in DPPOS Year 17. These new outcomes enable DPPOS Phase 3 to compute the Healthy Aging Index, as well as a multi-morbidity index, to determine the long-term effects of the original DPP interventions and to evaluate predictors of long-term healthy aging. Beginning in DPPOS Year 17, new measures of neuropathy were also added to identify symptomatic distal symmetric polyneuropathy (DSPN).  


DPPOS Protocol

The protocol for the Diabetes Prevention Program Outcomes Study (DPPOS), describes the background, design and organization of the DPPOS. The protocol is maintained by the Coordinating Center (CoC) at the George Washington University Biostatistics Center through new releases of the entire protocol or issuance of supplemental protocol memoranda. The preface contains a summary of the protocol modifications made during the DPPOS. Comments or questions regarding aspects of the DPPOS protocol, including distribution, should be directed to the staff of the CoC .

Below is the most up-to-date version:





Research Group
Information for each clinical site and research core is listed under Research Group.

Program Staff Only: To access the research group website, please click here.


DPPOS Coordinating Center
George Washington University Biostatistics Center
6110 Executive Blvd. Suite 750
Rockville, MD 20852


Technical Support
Please contact the DPPOS webmaster with any problems with login or accessing documents from the website.