Stroke Recovery’s Hidden Divide: Frailty Paths Predicted by Mind and Mood

When an older adult survives a stroke, the visible battle often ends at discharge. But a new longitudinal study from China suggests that the quieter war against frailty is only beginning, and that the trajectory it follows over the following year may be charted within the first days of hospitalization. Researchers tracking more than three hundred stroke survivors aged 60 and older found that frailty does not follow a single path after stroke. Instead, patients split into three distinct courses over twelve months, and the forces steering them toward the worst outcomes were strikingly non-physical: impaired cognition, damaged mood, heavy comorbidity burdens, and thin social support.

The study, conducted at the Department of Neurology of the First Affiliated Hospital of Nanjing Medical University, enrolled 314 consecutive older adults hospitalized for acute stroke between April and December 2024. Baseline assessments were completed within 48 hours of admission, capturing demographic and clinical characteristics, stroke severity measured on the National Institutes of Health Stroke Scale, comorbidity burden, perceived social support, and five domains of what geriatricians call intrinsic capacity. Frailty itself was scored using the Fried Frailty Phenotype, a five-criterion instrument assessing unintentional weight loss, exhaustion, low physical activity, slowness, and grip weakness. Follow-up assessments repeated the frailty measurement at three, six, and twelve months after discharge, and 293 participants—93.3 percent of the cohort—completed the full year of observation.

To make sense of the repeated measurements, the team turned to latent class growth modeling, a statistical technique that searches heterogeneous longitudinal data for hidden subgroups of people whose scores change in similar ways over time. Rather than averaging everyone’s frailty scores into a single trend line, the method asks whether the population actually contains several trend lines. Model selection relied on standard fit criteria, including the sample-size adjusted Bayesian Information Criterion and the adjusted Lo-Mendell-Rubin likelihood ratio test, which together determine how many trajectory classes the data genuinely support. The answer was three, and the differences between them were clinically dramatic.

The largest group, comprising 67.24 percent of participants, followed a low-decreasing trajectory: they began with low frailty scores and improved further over the year, suggesting that for roughly two-thirds of older stroke survivors, recovery of physical resilience is the default course. A second group, 20.48 percent, was moderate-stable, entering the study with intermediate frailty and remaining there through all four assessments. The most concerning group, 12.29 percent of the sample, followed a high-increasing trajectory, starting with high frailty and growing progressively frailer across the twelve months after discharge. One in eight older stroke patients, in other words, does not merely fail to recover—they deteriorate steadily, and the study set out to identify who they are before the deterioration begins.

The answer came from multinomial logistic regression, which compared each trajectory group against the low-decreasing reference while adjusting for demographic and clinical covariates. Two findings stood out for their consistency across both unfavorable trajectories. Cognitive impairment, assessed with the Montreal Cognitive Assessment, roughly doubled the odds of belonging to the moderate-stable group (odds ratio 2.632, 95 percent confidence interval 1.263 to 5.488) and more than quadrupled the odds of landing in the high-increasing group (odds ratio 4.506, 95 percent confidence interval 1.846 to 11.002). Psychological impairment, screened with the 15-item Geriatric Depression Scale, showed a parallel pattern, tripling the odds of moderate-stable membership (odds ratio 3.671) and roughly tripling the odds of high-increasing membership (odds ratio 3.161). The mind, it appears, is not a bystander in physical recovery—it is a load-bearing pillar.

The high-increasing trajectory had its own additional risk architecture. Patients who had experienced moderate or severe strokes, rather than normal or mild ones, were far more likely to deteriorate; conversely, a normal or mild stroke reduced the odds of high-increasing membership by nearly three-quarters (odds ratio 0.271, 95 percent confidence interval 0.095 to 0.773). A high comorbidity burden, quantified with the age-adjusted Charlson Comorbidity Index, nearly quadrupled the odds of the worsening trajectory (odds ratio 3.801, 95 percent confidence interval 1.610 to 8.978). And social connection mattered: low or moderate perceived social support, measured with the Perceived Social Support Scale, more than tripled the odds of belonging to the high-increasing group (odds ratio 3.171, 95 percent confidence interval 1.310 to 7.672). Frailty after stroke, the data imply, is not simply a matter of damaged brain tissue—it is the product of neurological injury layered on chronic disease, eroded cognition, depressed mood, and isolation.

The concept binding these threads together is intrinsic capacity, a framework promoted by the World Health Organization through its Integrated Care for Older People initiative. Intrinsic capacity refers to the composite of all an individual’s physical and mental capacities—locomotion, cognition, psychological state, vitality or nutrition, and sensory function—and the study assessed all five domains at baseline using validated instruments including the Short Physical Performance Battery, the Mini Nutritional Assessment Short-Form, and the Montreal Cognitive Assessment. The WHO’s model treats aging not as a disease to be cured but as a dynamic balance between capacity and environment, and this study provides some of the most granular evidence yet that the framework has predictive teeth in a specific clinical population. Where capacity is compromised—particularly in the cognitive and psychological domains—the trajectory of frailty bends downward.

For geriatric nursing, the implications are concrete. The study’s authors argue that early multidomain assessment, performed within the first 48 hours of admission, can flag patients at risk of unfavorable frailty trajectories before they leave the hospital. A nurse armed with a cognitive screen, a depression scale, a comorbidity index, and a social support questionnaire can identify the roughly one in eight patients destined for progressive frailty and tailor post-discharge monitoring accordingly: closer follow-up intervals, cognitive rehabilitation referrals, psychological support, caregiver engagement, and structured social interventions. The alternative—waiting for frailty to declare itself in the community—may mean intervening a year too late, after slowness, exhaustion, and weight loss have already compounded.

The study’s design carries both strengths and limits worth noting. Its prospective longitudinal structure, near-complete follow-up, and use of a validated phenotype instrument lend credibility to the trajectory findings, and the statistical approach is well suited to the question of heterogeneous recovery. Yet the cohort was drawn from a single tertiary hospital in Nanjing, participants ranged only from 60 to 86 years with a mean age just under 69, and observational associations cannot prove that treating depression or bolstering social support would reroute a patient’s frailty course. Those questions demand intervention trials. Still, the core message is hard to dismiss: recovery from stroke in later life is a year-long process with predictable failure modes, and the earliest warning signs are written not in the legs but in the mind, the mood, and the web of people around the patient.

As populations age and stroke survival improves, the number of older adults living in the aftermath of stroke will only grow, and frailty will remain one of the strongest predictors of disability, institutionalization, and death in this group. What this study adds is a map—a way to see, within days of admission, which road a patient is likely to travel. Turning that map into better outcomes will require health systems to treat cognition, mood, and social connection not as soft extras but as core vital signs of post-stroke recovery, monitored and managed with the same urgency as blood pressure and mobility. The patients most at risk of a fading year are identifiable on day two. The task now is to act on what the data already show.

Subject of Research: Frailty trajectories and intrinsic capacity in older stroke survivors during the first year after hospital discharge

Article Title: Frailty trajectories from hospitalization to 12 months after discharge in older stroke survivors and their associations with intrinsic capacity: implications for geriatric nursing

Article References: Tang, X.-X., Tang, Q.-Y., Liu, L., Li, B.-B., Mei, J., & Yu, L. (2026). Frailty trajectories from hospitalization to 12 months after discharge in older stroke survivors and their associations with intrinsic capacity: implications for geriatric nursing. BMC Geriatrics. https://doi.org/10.1186/s12877-026-08415-5

Image Credits: AI Generated

DOI: 10.1186/s12877-026-08415-5

Keywords: stroke, frailty, older adults, intrinsic capacity, geriatric nursing, longitudinal study, cognitive impairment, depression, social support, comorbidity, Fried Frailty Phenotype, post-discharge care