SPRINTT analysis finds mobility benefit, but disease-pattern differences remain uncertain

A new analysis of the SPRINTT randomized trial found that a multicomponent lifestyle program reduced mobility disability among older adults living with frailty, sarcopenia and multiple chronic conditions. The findings were published August 31 in Nature Aging, more than four years after the trial’s main results appeared in The BMJ.

The analysis adds a clinically relevant healthspan question: does the particular combination of chronic diseases influence who benefits from an exercise- and nutrition-based intervention? Although some disease-pattern subgroups appeared to benefit more than others, the statistical test designed to determine whether treatment effects truly differed across patterns was not significant. The subgroup findings therefore remain exploratory.

A post hoc analysis of a randomized trial

SPRINTT enrolled 1,519 community-dwelling adults aged 70 or older across 16 sites in 11 European countries. Participants had physical frailty and sarcopenia but could independently complete a 400-meter walk when they entered the trial.

Participants were randomly assigned to either a multicomponent program or healthy-aging education. The program combined individualized aerobic, resistance, flexibility and balance activities with nutritional counseling and activity monitoring. Exercise was delivered through supervised sessions and home activities, while counseling focused on adequate protein and energy intake. Treatment and follow-up continued for up to 36 months.

The new post hoc analysis included 1,199 participants from the trial’s primary functional group who also had at least two chronic conditions. Their median age was 79, 71% were women, and the median participant had six chronic conditions.

Researchers used baseline diagnoses to identify four statistical clusters: an unspecific pattern without strongly overrepresented diseases, representing 55% of participants; a psychiatric pattern, 24%; a cardiometabolic pattern, 13%; and a respiratory pattern, 7%.

The overall result favored the intervention

The analysis treated incident mobility disability or death as an event. Mobility disability was primarily defined as inability to complete 400 meters within 15 minutes under specified conditions.

Across the full analytical population, the multicomponent intervention was associated with a 22% lower event hazard than healthy-aging education, with a hazard ratio of 0.78 and a 95% confidence interval from 0.66 to 0.92. At 36 months, the estimated probability of remaining free from mobility disability or death was 8.2 percentage points higher in the intervention group. Participants assigned to the program averaged 2.09 additional months without an event over that period.

These overall findings are consistent with the trial’s prespecified primary analysis, published in 2022. In that report, mobility disability occurred in 46.8% of intervention participants and 52.7% of controls within the primary functional group. Persistent mobility disability, however, did not reach conventional statistical significance, and serious adverse events were reported in 39.2% of intervention participants and 36.0% of controls.

Subgroup signals require restraint

In conventional subgroup analyses, the psychiatric cluster showed a statistically significant estimated benefit. A Bayesian partial-pooling analysis supported benefit in both the psychiatric and unspecific clusters, while estimates for the smaller respiratory and cardiometabolic groups were imprecise.

Those contrasts do not establish that multimorbidity type determines treatment response. The formal treatment-by-pattern interaction had a P value of 0.870, providing no statistical evidence that the intervention’s effect differed among the four clusters. The authors reported that the analysis was likely underpowered for interaction testing; simulations suggested that substantially more participants would have been needed to detect differences of the observed magnitude reliably.

Other limitations include the post hoc design, limited diagnostic coding detail and the absence of biomarkers that might explain the proposed disease patterns. The participants were already frail, sarcopenic and at least 70 years old, limiting generalization to younger or healthier populations.

The practical conclusion is narrower than precision-treatment language might imply. SPRINTT supports a structured, multicomponent program as a way to reduce mobility disability in a selected high-risk older population. Whether baseline disease clusters can reliably identify people who benefit more—or need a different intervention—requires prospective confirmation.