An intensive lifestyle intervention was associated with a lower long-term burden of multimorbidity among adults with prediabetes, while metformin was not associated with a statistically significant reduction vs. placebo, according to a 21-year follow-up analysis of the Diabetes Prevention Program (DPP) published June 15 in JAMA.

The observational, multicenter cohort analysis included 2021 morbidity data for 1,173 participants from the DPP and DPP Outcomes Study (DPPOS) (68% women). Participants were Medicare enrollees who were at high risk for type 2 diabetes at enrollment between 1996 and 1999 and had originally been randomized to intensive lifestyle intervention (n=385), metformin 850 mg twice daily (n=385) or placebo (n=403).

The intensive lifestyle intervention comprised 16 individual core sessions, followed by two years of monthly maintenance sessions. The intervention targeted at least 7% weight loss during the first six months, with maintenance supported by dietary modification and at least 150 minutes per week of moderate-intensity physical activity.

The randomized DPP intervention phase had a mean duration of 3.2 years, and DPPOS follow-up has continued since 2002. Median participant age was 51 years at randomization and 74 years at 2021 outcome ascertainment.

During DPPOS, the placebo intervention was discontinued, metformin was offered open label to eligible participants, and those originally assigned to lifestyle intervention attended twice-yearly booster sessions.

At 21 years, multimorbidity – defined as at least two of 15 prespecified prevalent conditions, including hypertension, heart failure, coronary artery disease/ischemic heart disease, cardiac arrhythmias and hyperlipidemia – was present in 85% of the cohort: 82% of participants assigned to lifestyle intervention, 85% assigned to metformin and 87% assigned to placebo. Three or more conditions were present in 72% of participants assigned to lifestyle intervention vs. 81% in both the metformin and placebo groups. The median number of conditions was four in the lifestyle intervention group and five in the metformin and placebo groups.

The most prevalent chronic conditions were hyperlipidemia (76%), hypertension (75%) and diabetes (67%).
In covariate-adjusted analyses, assignment to lifestyle intervention was associated with a lower risk of multimorbidity vs. placebo (hazard ratio [HR], 0.79), whereas assignment to metformin was not associated with a significant reduction vs. placebo (HR, 0.91). Results were similar when diabetes was excluded from the multimorbidity definition. The association was stronger in an analysis limited to the highest-cost multimorbidity dyads, including combinations of stroke, chronic obstructive pulmonary disease, chronic kidney disease, heart failure and asthma (HR for lifestyle vs. placebo, 0.57).

"Because lifestyle modification can be safe and cost-effective, sustaining those behaviors among Medicare beneficiaries at risk of diabetes may help to reduce health burden and health care spending," write Marcel E. Salive, MD, and colleagues for the DPP Research Group.

"The lifestyle intervention strategies implemented in the [U.S.] DPP offer a practical and effective approach that may reduce the health care burden associated with aging," write Hermes Florez, MD, PhD, MPH, MA; Caroline Foster, PhD; and Gilberto Vizcaino, MD, PhD, in an accompanying editorial comment. "Modest weight loss through healthy nutrition and increased physical activity remains a cornerstone to preserve quality of life and to potentially promote healthy longevity in this population."