Can Activity Counseling Cut Mortality in Type 2 Diabetes?

Physical activity is consistently linked with longer life in people with type 2 diabetes. Yet large lifestyle trials have struggled to prove that prescribing diet and exercise reduces major cardiovascular events or mortality. A new post-hoc analysis from Italy offers an intriguing explanation: the durable intervention may be behavior change itself, not a supervised workout plan.

In the Italian Diabetes and Exercise Study 2 (IDES_2), behavioral counseling was associated with an approximately 50% to 59% lower rate of all-cause mortality over 11 years than standard care. The finding deserves attention. It does not settle the question. The mortality analysis was not planned in advance, involved only 300 participants and roughly 53 deaths, and did not measure whether participants actually stayed more active after the trial ended.

A striking result from a small, post-hoc trial

Balducci et al.'s 2026 IDES_2 analysis followed 300 Italian adults with established type 2 diabetes who had originally been randomly assigned to standard care or a structured behavioral counseling program. The program aimed for 150 minutes a week of moderate physical activity and used education, motivational interviewing, and practical work on environmental barriers to activity.

Over 11 years, the counseling group had a reported all-cause mortality hazard ratio of roughly 0.41 to 0.50 versus standard care. In relative terms, that corresponds to about a 50% to 59% reduction. The publication describes this as a post-hoc analysis rather than a prespecified mortality endpoint. That distinction changes how much confidence the number earns.

Randomization helps reduce baseline differences between groups, but it cannot rescue every later analytical choice. With about 53 deaths total, a Cox regression model has limited room for adjustment before estimates become unstable. Small post-hoc analyses also tend to produce larger-looking effects than later replication studies. The article does not establish how many deaths were prevented in absolute terms, so readers should not translate the relative reduction into a personal survival forecast.

The original five-year trial did improve cardiorespiratory fitness and metabolic markers. Those are useful outcomes, but they are not mortality. Nor did the long-term analysis measure physical activity after the formal intervention, leaving a missing link in the proposed chain: counseling, sustained activity, then lower mortality.

Why this does not erase Look AHEAD

The obvious challenge comes from Look AHEAD, the far larger type 2 diabetes lifestyle trial. About 5,000 adults received either intensive lifestyle intervention focused on weight loss, caloric restriction, and supervised exercise or a comparison program. After a median 9.6 years, the intervention did not significantly reduce its primary composite cardiovascular endpoint.

A 2022 analysis of Look AHEAD data also found that greater cardiorespiratory fitness predicted lower cardiovascular risk, while the parent randomized trial remained null for cardiovascular events. That is an important split: fitter participants may have had lower risk, but an association within a trial does not prove that the intervention caused the lower risk.

Look AHEAD weighs more heavily than IDES_2 for cardiovascular claims because it was much larger and tested a prespecified clinical endpoint. Still, the trials did not ask the same question. Look AHEAD emphasized weight loss and evaluated composite cardiovascular events; IDES_2 emphasized adoption and maintenance of activity and later examined all-cause mortality. Their follow-up periods also differed slightly, though 11 years versus 9.6 years is not enough to prove that duration explains the discrepancy.

The defensible conclusion is narrower: IDES_2 does not overturn Look AHEAD. It raises the possibility that counseling designed around durable activity habits could produce benefits that a weight-centered program and a cardiovascular composite endpoint miss.

Why a decade may matter

Long delays are biologically plausible. In 2026 post-hoc analyses of the Diabetes Prevention Program Outcome Study and the Da Qing Diabetes Prevention Outcome Study, lifestyle-related cardiovascular morbidity and mortality benefits emerged during extended follow-up and persisted for more than 20 years after the formal interventions ended.

Those results support the idea that lifestyle effects can take a long time to appear. They do not directly validate IDES_2, because the source populations had prediabetes or elevated metabolic risk rather than established type 2 diabetes. Prediabetes may allow more metabolic reversal than diabetes that has already developed. The DPPOS and Da Qing analyses were also post-hoc, so they strengthen a hypothesis about timing rather than provide definitive confirmation.

No preregistered trial in established type 2 diabetes has used all-cause mortality as its primary endpoint with more than 10 years of follow-up. That is the evidence gap behind the confident headlines this topic tends to attract.

The cancer signal is plausible, but exploratory

In IDES_2, cancer deaths appeared to account for more of the mortality difference than cardiovascular deaths. That pattern fits broader population evidence: a 2025 wearable-accelerometry study linked greater physical activity across intensities with lower mortality, cardiometabolic disease, and cancer risk.

Type 2 diabetes often involves hyperinsulinemia, higher IGF-1 signaling, and chronic low-grade inflammation, all of which could contribute to an environment that favors cancer growth and survival. Physical activity could affect several of these pathways. But plausible biology is not proof that cancer prevention caused the IDES_2 mortality result.

The cancer subgroup likely contained only about 10 to 20 deaths. That is too few events for a reliable causal conclusion, and the reported uncertainty included the possibility of no effect. Treat the cancer finding as a lead for larger trials, not proof that walking is an oncology intervention.

What the wider activity evidence can—and cannot—tell you

Observational evidence supports regular movement even when mortality trial evidence remains incomplete. A 2025 dose-response meta-analysis of daily steps found that higher step counts were associated with lower all-cause mortality, cardiovascular disease, and cancer across adult populations, without an apparent plateau at moderate step counts. Association still leaves room for healthier people being more able to move, but the consistency of the pattern makes the IDES_2 result biologically credible.

Activity need not arrive in one heroic gym session. A meta-analysis of objectively measured light-intensity activity found a lower risk of cardiovascular mortality with more light activity; each additional 30 minutes per day was associated with a hazard ratio of 0.80 (95% confidence interval, 0.67 to 0.96). This was population-level observational evidence, not a type 2 diabetes counseling trial.

For glucose management, timing may help. A 2023 systematic review and meta-analysis found that exercise close to a meal can reduce acute post-meal glucose excursions in people with and without impaired glucose tolerance. A 10- to 15-minute walk after meals is a practical way to build activity into a day, though the study evidence addresses short-term glucose responses rather than longevity.

Exercise also improves outcomes that matter before any mortality curve separates. A systematic review of exercise training in type 2 diabetes found improvements in heart-rate-variability measures, which reflect cardiac autonomic function. A 2024 meta-analysis found that combined aerobic and resistance exercise improved metabolic health, cognition, physical function, and quality of life in middle-aged and older adults with type 2 diabetes. These are meaningful benefits, but none proves that behavioral counseling cuts mortality.

Build a routine that can outlast motivation

The useful lesson from IDES_2 is not that one counseling protocol has been proven to halve mortality. It is that the design of an activity plan may matter. A calendar full of supervised sessions can work; so can walking routes, post-meal routines, resistance training at home, social accountability, and a step tracker. The better system is the one a person can keep doing when enthusiasm has gone on vacation.

• Aim for the 150 minutes per week of moderate activity used as the IDES_2 counseling target, adjusting for medical limitations and clinician advice.
• Use steps or active minutes as feedback, not as a moral scorecard.
• Pair short walks with meals if post-meal glucose is a concern.
• Include aerobic and resistance work for fitness, function, and metabolic health.
• Seek individualized guidance before changing an exercise program if you use glucose-lowering drugs that can cause hypoglycemia, have neuropathy, or have cardiovascular symptoms.

The verdict

IDES_2 provides the most direct randomized signal that behavioral physical-activity counseling might lower long-term mortality in established type 2 diabetes. Its reported hazard ratio is large, but the analysis is post-hoc, small, and unable to confirm sustained activity during follow-up. Look AHEAD's null cardiovascular result remains a major constraint on stronger claims.

Exercise and daily movement still have ample reason to belong in diabetes care: better fitness, metabolic health, function, acute glucose control, and consistent observational links with lower mortality. Calling behavioral counseling a proven mortality treatment would outrun the evidence. Calling durable activity habits worth building would not.