Multimorbidity Patterns Determine Who Benefits Most from Frailty Interventions
A SPRINTT trial analysis shows that disease-pattern profiles in frail older adults predict responsiveness to lifestyle interventions for mobility preservation.
Summary
A post hoc analysis of the SPRINTT randomized trial examined whether the mix of chronic diseases in frail older adults affected how well a combined physical activity and nutrition intervention prevented mobility disability. Researchers used latent class analysis to sort 1,199 community-dwelling adults aged 70 or older into four multimorbidity patterns: unspecific, psychiatric, cardiometabolic, and respiratory. Overall, the intervention cut the risk of mobility disability or death by 22% and extended disability-free life by about two months. However, those with a psychiatric multimorbidity pattern gained the most — nearly an extra three months of mobility — while those with cardiometabolic or respiratory patterns showed no statistically significant benefit. These findings suggest that knowing a patient's disease-combination profile could help clinicians better target preventive lifestyle programs for frail older adults.
Detailed Summary
Frailty and sarcopenia in older adults represent a major public health challenge, as loss of mobility independence sharply reduces quality of life and increases mortality. Lifestyle interventions combining physical activity and nutritional support have shown promise, but it has remained unclear whether the specific constellation of chronic diseases a person carries — their multimorbidity pattern — shapes how much they benefit.
This post hoc analysis of the SPRINTT randomized controlled trial (NCT02582138) enrolled 1,199 community-dwelling adults aged 70 or older who had frailty, sarcopenia, and at least two chronic conditions. The multicomponent intervention included structured physical activity with technological support and individualized nutritional counseling. Latent class analysis was used to identify distinct multimorbidity clusters among participants.
Four patterns emerged: unspecific (55% of participants), psychiatric (24%), cardiometabolic (13%), and respiratory (7%). Across the full cohort, the intervention reduced the hazard of mobility disability or death by 22% (HR 0.78, 95% CI 0.66–0.92) and extended time free from disability by approximately 2.1 months. Subgroup analyses revealed the greatest benefit in participants with the psychiatric multimorbidity pattern, who gained roughly 3.1 extra disability-free months (HR 0.65, 95% CI 0.47–0.92). Bayesian partial pooling supported benefit in the unspecific pattern (HR 0.80). Estimates for the cardiometabolic and respiratory patterns were imprecise and non-significant.
These results carry important implications for clinical practice. Rather than applying the same lifestyle prescription to all frail older patients, clinicians may achieve better outcomes by considering a patient's underlying disease architecture. Patients with psychiatric comorbidities appear particularly responsive, possibly because physical activity and structured support address behavioral and neurological contributors to decline.
Caveats include the post hoc and observational nature of the subgroup analysis, the possibility that smaller subgroups (cardiometabolic, respiratory) were underpowered, and that the summary is based on the abstract only, limiting full methodological appraisal.
Key Findings
- Overall, the intervention reduced mobility disability or death risk by 22% and added ~2.1 disability-free months in frail older adults.
- Frail adults with psychiatric multimorbidity gained ~3.1 extra disability-free months — the largest benefit of any subgroup.
- Four distinct multimorbidity patterns were identified: unspecific (55%), psychiatric (24%), cardiometabolic (13%), respiratory (7%).
- Cardiometabolic and respiratory pattern groups showed no statistically significant benefit from the intervention.
- Disease-pattern profiling could allow clinicians to better target lifestyle interventions for frail older populations.
Methodology
Post hoc analysis of the SPRINTT randomized controlled trial (NCT02582138) involving 1,199 community-dwelling adults aged ≥70 with frailty, sarcopenia, and ≥2 chronic conditions. Latent class analysis identified multimorbidity patterns; intervention effects within each pattern were assessed using Cox proportional hazards models and Bayesian partial pooling analyses.
Study Limitations
This is a post hoc, hypothesis-generating subgroup analysis, so findings should be interpreted cautiously and not as confirmatory evidence. Smaller subgroups (cardiometabolic, respiratory) may have been underpowered to detect meaningful effects. The summary is based on the abstract only, precluding full assessment of methodology, covariate adjustment, or sensitivity analyses.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: