The research, led by Xiangyu Zhai and Doris S. F. Yu at The University of Hong Kong, examined 4,477 community-dwelling older adults classified as prefrail or frail, with a mean age of 72.2 years and 72.5 percent of participants being women. Participants were drawn from two large healthy-aging programs spanning all 18 administrative districts of Hong Kong, and frailty was assessed using a modified version of the well-known Fried frailty phenotype, which counts criteria including unintentional weight loss or muscle shrinking, weakness, slowness, low physical activity, and exhaustion. Those meeting one or two criteria were classed as prefrail, while three or more indicated frailty. In total, 3,385 participants were prefrail and 1,092 were frail. The scale of the sample is notable, because most previous exercise studies in frail populations have been small trials with inconsistent findings, leaving clinicians without clear guidance on how to balance efficacy against tolerability.

The study’s central innovation lies in how it disentangled two distinct fitness components that reflect two distinct exercise modalities. Cardiorespiratory fitness, or CRF, is the physiological outcome of aerobic exercise and was measured here with a two-minute walk test, in which participants walked at their comfortable usual pace and the maximum distance covered was recorded. This field test is well suited to frail populations who cannot manage treadmill or cycle ergometry, and it correlates moderately with laboratory-measured peak oxygen uptake. Muscle strength, the outcome of resistance training, was assessed with a digital handgrip dynamometer, the most widely used clinical proxy for overall strength. Cognitive function was screened with the five-minute Montreal Cognitive Assessment protocol, using age- and education-adjusted cutoffs for the Hong Kong population to classify cognitive impairment. In total, 315 participants, or 7.0 percent, met criteria for cognitive impairment.

The results reveal a striking dose-response relationship for cardiorespiratory fitness. Compared with the least fit quartile, participants in the second, third, and fourth quartiles of walking performance had progressively lower odds of cognitive impairment, with fully adjusted odds ratios of 0.57, 0.32, and 0.15 respectively. In other words, the fittest quartile showed roughly 85 percent lower odds of impairment than the least fit, even after accounting for age, sex, education, marital status, living arrangement, body mass index, physical activity, frailty status, and muscle strength. Treated as a continuous variable, every additional 5 meters covered in the two-minute walk was associated with 7 percent lower odds of cognitive impairment. The association held consistently across prefrail and frail subgroups, and across nearly every demographic and clinical stratum the researchers examined, with the sole exception of a small underweight subgroup.

Muscle strength told a more complicated story. On its own, higher handgrip strength was associated with lower odds of cognitive impairment, with a significant linear trend across quartiles and roughly 3 percent lower odds per additional kilogram of grip strength. But when the researchers additionally adjusted for cardiorespiratory fitness, the strength association was substantially attenuated, and only the third and fourth quartiles retained statistical significance. The pattern suggests that much of the apparent protective effect of grip strength may be shared with, or mediated through, overall aerobic capacity. This matters because grip strength predominantly reflects upper-limb muscle function, whereas the two-minute walk test requires adequate lower-limb strength to complete, meaning the fitness measure already incorporates a component of leg strength. The two physiological markers are therefore not fully independent, and the overlap appears to favor CRF as the more comprehensive indicator of the physical fitness relevant to cognitive health.

The joint analysis drove the point home. Participants were divided into four groups: unfit and weak, unfit but strong, fit but weak, and fit and strong, with unfit and weak defined as the bottom quartile of each distribution. Compared with the unfit-and-weak reference group, the fit-but-weak group had an odds ratio of 0.47 for cognitive impairment, while the unfit-but-strong group had an odds ratio of 0.71. Being fit and strong conferred the lowest odds of all, at 0.38. Crucially, when the researchers used the unfit-but-strong group as the reference instead, the fit-but-weak group still fared better, with an odds ratio of 0.66. High cardiorespiratory fitness alone thus conferred about 56 percent lower odds of cognitive impairment than high muscle strength alone. For a population in which every bout of exertion carries a cost, that hierarchy is potentially practice-changing.

Why might aerobic capacity be so tightly linked to brain health in this vulnerable group? The authors point to several plausible biological pathways. Frailty is accompanied by accelerated physiological decline, elevated oxidative stress, and heightened neuroinflammation, processes that can precede and potentiate cognitive deterioration. Higher CRF may buffer these insults by enhancing cerebral blood flow, reducing oxidative damage, promoting synaptogenesis, and modulating neurotrophic factors and neurotransmitter systems. Aerobic exercise is also known to optimize metabolic regulation and support neurogenesis. Resistance training, by contrast, is thought to act mainly by reducing systemic inflammation and stimulating the release of neuroprotective myokines from contracting muscle. Both pathways are real, but the integrated performance of the cardiovascular, respiratory, and musculoskeletal systems captured by CRF may more fully reflect the bodily reserve on which the aging brain depends.

The clinical implications are considerable. Because frail older adults face fatigue, limited activity tolerance, and high dropout rates in demanding multicomponent programs, the findings suggest that aerobic exercise aimed at improving CRF may warrant evaluation as a first-line strategy in future intervention trials, with individualized resistance training added when feasible. The authors emphasize that low-impact or chair-based aerobic programs involving continuous movement and mild exertion could offer suitable entry points for those unable to manage conventional training. Just as important, the two-minute walk test itself could serve as a cheap, scalable screening tool: a simple measure of walking distance, requiring no specialized equipment, might identify prefrail and frail adults at elevated risk of cognitive decline and flag them for timely intervention. Whether raising CRF yields cognitive benefits before strength improves remains an open question that only randomized trials can answer.

The researchers are careful about the limits of their evidence. This was a cross-sectional study, so it cannot establish causation; early subclinical cognitive decline could plausibly reduce physical activity and thereby lower fitness, rather than low fitness causing impairment. The walk test was performed at a self-selected pace, making it a submaximal proxy rather than a direct measure of maximal aerobic capacity, and grip strength captures only upper-limb function. The cognitive screen assessed global performance without distinguishing specific domains, and all participants were recruited in Hong Kong, which may limit generalizability. Sensitivity analyses, including reclassifying participants excluded from the walk test as having the lowest fitness, nonetheless supported the robustness of the main findings. The study was funded by the Hong Kong Jockey Club Charities Trust and approved by the Hospital Authority Clinical Research Ethics Committee.

Even with those caveats, the study is the first to compare the independent and combined associations of CRF and muscle strength with cognitive impairment specifically in prefrail and frail older adults, and its message is unusually actionable. Notably, participants with cognitive impairment in the sample were older, less educated, more likely to live alone, less likely to meet physical activity guidelines, and had higher frailty prevalence and lower fitness and strength than their cognitively intact peers, underscoring how these risks cluster. For clinicians, caregivers, and older adults navigating the exhausting reality of frailty, the findings suggest a pragmatic hierarchy: if resources and energy are limited, prioritize the heart and lungs, and let the walk test tell you where you stand. The next step, the authors argue, is to test directly in prospective cohorts and randomized controlled trials whether aerobic exercise, potentially combined with tailored resistance training, can actually slow cognitive decline in this population rather than merely marking those at risk.

Subject of Research: Associations of cardiorespiratory fitness and muscle strength with cognitive impairment in prefrail and frail older adults

Article Title: Optimizing frailty care: an analysis of dual exercise targets and cognitive function among frail older adults

Article References: Zhai, X., Xi, J., Miao, M., & Yu, D. S. F. (2026). Optimizing frailty care: an analysis of dual exercise targets and cognitive function among frail older adults. GeroScience. https://doi.org/10.1007/s11357-026-02520-4

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02520-4

Keywords: frailty, cardiorespiratory fitness, muscle strength, cognitive impairment, older adults, aerobic exercise, resistance training, handgrip strength, two-minute walk test, dementia prevention, geroscience, exercise prescription