Two brief fitness tests tied to mortality in older adults

Clinical takeaway: Watching an older patient rise from a chair and walk a short course may flag vulnerability that a diagnosis list does not. Fitness is modifiable, so the result is a prompt to intervene, not just to stratify.

Blood pressure, weight, and A1c often get assessed at a visit with an older patient, with each number predicting health risks. Physical function predicts outcomes as well or better in older adults but often isn't taken into account. Measuring it takes a stopwatch and under a minute, which makes it a feasible addition.

Risk stratification after 65 focuses on disease burden, which counts diagnoses rather than capacity. That leaves the patients who are independent and undiagnosed, but losing ground, without many clear indicators of relative capacity. A new large cohort study is one of the few to score several fitness domains in the same older population and link them to death records in an effort to provide further insight.

The composite index pooled percentile ranks across all seven assessments; it showed the steepest gradient. Mortality reached 25.8% in the least-fit fifth of the cohort and 5.1% in the fittest. After adjustment for age, sex, BMI, self-reported activity, 10 chronic conditions, and overall comorbidity score, the fittest fifth had 61% lower mortality risk than the least fit. Balance and agility led the individual tests.

The eight-foot up-and-go, which times a patient rising from a chair, walking around a cone, and sitting back down, split survival most sharply of the seven assessments: 4.9% of the fastest fifth died, against 25.2% of the slowest. One-leg stance came next, then the 30-second chair stand, a measure of lower-body strength. Gradients were shallower for aerobic endurance on a two-minute step test and shallowest for the two flexibility tests. Nearly all of the difference sat between the bottom fifth and everyone above it.

Participants were 13,423 community-dwelling adults aged 65 years or older who completed standardized fitness testing at national testing stations across Taiwan in 2015 and 2016, with records linked to national health insurance and death registry data through the end of 2022. Certified examiners administered all seven assessments under fixed protocols that included practice trials and pretest safety screening. Adults older than 90 years and those with dementia, stroke, prior amputation, cancer, or other conditions likely to limit performance were excluded. Over a median seven years, 1,631 participants (12.2%) died.

Two gaps stand between this finding and a change in practice. Fitness scores were never compared against the comorbidity tools already in use, so whether they sharpen a risk estimate or merely echo it is untested. And fitness was measured once, leaving open whether patients who improve their scores lower their risk. Trials pairing performance-based screening with exercise referral would answer both.

Source: Wu M, et al. (2026 Aug 10) JAMA Netw Open. Physical Fitness and All-Cause Mortality in Older Adults