A detailed systematic review protocol was developed a priori to define the research question, inclusion and exclusion criteria, search strategy, and data synthesis methods. However, the protocol was not prospectively registered in an international prospective register of systematic reviews such as PROSPERO. We acknowledge this as a limitation to the transparency and reproducibility of our review. To mitigate this concern and ensure the completeness of reporting, this review strictly adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The completed PRISMA checklist is provided as Supplementary file 1.

Studies were selected based on the following PICOS framework:

For clarity: RCTs were defined as studies where participants were allocated to intervention or control groups using a random method. CCTs were defined as studies with a control group (receiving usual care, no intervention, or an alternative intervention) but without random allocation of participants (e.g., allocation by alternation, birth date, or medical record number). This distinction is maintained throughout the review to ensure transparency.

Exclusion criteria included: studies published before 2015; non-English articles; studies without full-text available; and interventions not primarily based on physical exercise (e.g., those focusing solely on nutrition or physical therapy without a structured exercise component).

The literature search was conducted between May 21 and June 4, 2025, using the MEDLINE、PMC, Pubmed Central Canada electronic database. The search strategy employed combinations of English-language keywords related to the target population, exercise modalities, and outcomes of interest, namely:

The study selection process was conducted in three sequential phases, following the PRISMA 2020 guidelines, as illustrated in the flow diagram (Figure 1).

First, all duplicate records were automatically and manually removed from the initial search results. Second, the titles and abstracts of the remaining unique records were screened by two independent reviewers against the eligibility criteria. Studies that clearly did not meet the PICOS criteria were excluded at this stage. Third, the full texts of the remaining potentially eligible reports were retrieved and thoroughly assessed for eligibility by the review team.

Any disagreements between reviewers at the second or third stage were resolved through discussion or by consulting a third reviewer. The reasons for excluding full-text articles were systematically recorded and are detailed in Figure 1. This rigorous process culminated in the final inclusion of 95 studies for data synthesis.

The data extraction and coding procedures for the studies included in this systematic review were conducted in a structured and standardized manner to ensure consistency and methodological rigor. A primary reviewer independently performed a full-text reading of all selected articles and extracted relevant data according to a predefined coding framework. Any uncertainties or discrepancies arising during this process were discussed and resolved with a second reviewer, thereby increasing the reliability and internal validity of the review.

Extracted data included:

Whenever reported, additional contextual details, such as environmental setting, level of supervision, and progression criteria, were also extracted to facilitate interpretation and comparison across studies.

The included studies employed a variety of validated instruments to assess functional fitness outcomes. The most frequently used instruments for each domain were:

Due to heterogeneity in interventions and outcomes, a narrative synthesis was employed. Findings are presented in structured tables and summarized by intervention type.

With respect to the instruments used to assess functional fitness, the most frequently employed were the TUG, present in 18 studies; the SPPB, used in 12 studies; the 6MWT, in 9 studies; the handgrip strength test, in 15 studies; the Chair Stand Test / 30-Second Sit-to-Stand, in 16 studies; and balance scales such as the Berg Balance Scale, used in at least 10 studies. Some authors chose to employ combined functional batteries or tests adapted to the context and condition of the participants, such as the Senior Fitness Test.

The effects of the included interventions on functional fitness outcomes are comprehensively summarized in Table 4. The synthesis of evidence reveals several key patterns:

In summary, the evidence robustly indicates that tailored exercise programs are a potent intervention for improving physical function in older adults. The observed heterogeneity in outcomes underscores the importance of personalized exercise prescription based on an individual’s health status and specific goals.

To appraise the robustness of the evidence underpinning our primary conclusions, we conducted a focused assessment of key methodological quality indicators for a selection of studies that were pivotal to our synthesis, based on their sample size, relevance to the review’s aim, and influence on the overall findings. The results of this assessment are summarized in Table 5.

The evaluation revealed a mixed picture of methodological rigor. Among the key studies assessed, the majority provided adequate descriptions of randomization and reported high rates of participant adherence, which are strengths. However, blinding of outcome assessors was not consistently implemented across these studies, and the use of intention-to-treat analysis was variable. These aspects represent potential sources of bias and should be considered when interpreting the corresponding results. The overall methodological remarks for each study are detailed in Table 5.

Approximately 29 studies consistently supported the effectiveness of structured physical exercise programs in enhancing functional fitness. These programs were shown to be safe, well-tolerated, and accessible across various settings, including home-based environments, care institutions, and clinical contexts. Additionally, higher adherence rates were observed in studies that included personalized and supervised programs, particularly those with longer intervention durations, underscoring the importance of individualized, structured approaches for optimal results.

The synthesized evidence robustly confirms the foundational efficacy of structured exercise for improving functional fitness in older adults. However, a critical appraisal reveals that this benefit is not uniform and is significantly moderated by intervention modality, population characteristics, and outcome specificity.

Dose–Response and Modality Specificity: While multicomponent and combined aerobic-RT [e.g., (5, 16)] appear to provide the broadest benefits across functional domains (mobility, strength, aerobic capacity), a clear dose–response relationship remains obscured by heterogeneous reporting of intensity and volume. The consistent, robust improvements in muscle strength from RT highlight its non-negotiable role in combating sarcopenia and functional decline.

The Critical Role of Baseline Status: The impact of exercise is profoundly influenced by baseline health. We observed a stark contrast between substantial functional gains in clinically compromised populations (e.g., those with rheumatoid arthritis, institutionalized elderly) and the absence of significant benefit in already high-functioning older adults, as illustrated by the null findings of the large DO-HEALTH trial. This suggests that the principle of diminishing returns may apply, with the greatest absolute gains occurring in those with the lowest initial functional capacity.

Heterogeneity in Cognitive Outcomes: The discordant findings on cognitive function—ranging from null effects (13) to positive gains (21)—underscore that “exercise” is not a monolithic intervention for the brain. The positive outcomes associated with complex, coordinative activities like Taekwondo lend support to the hypothesis that cognitive benefits are more likely to emerge from exercises that provide a high level of motor-cognitive integration, potentially through mechanisms such as the upregulation of neurotrophic factors (e.g., BDNF). This remains a fertile ground for hypothesis-testing in future research.

Individualization in Clinical Populations: The modest results in challenging clinical contexts like HFpEF (19) reinforce that physiological ceilings and pathologies can limit adaptations. In such cases, and in frail populations, the primary goal should shift from maximizing intensity to optimizing movement safety and feasibility, where even low-intensity exercise can yield meaningful improvements in quality of life and mobility.

In conclusion, the prevailing “one-size-fits-all” prescription is inadequate. Future practice and research must pivot toward personalized exercise medicine, explicitly accounting for an individual’s clinical profile, functional baseline, and personal goals to optimize outcomes.

The efficacy of exercise in frail and institutionalized older adults is well-established; however, our synthesis critically underscores that MCT represents the most potent and adaptable model for this population. Its superiority lies in its synergistic targeting of the multifactorial nature of functional decline.

Synergistic Effects for Frail Populations: The significant improvements in mobility (TUG) and endurance (6MWT) observed in institutionalized settings (5) are not merely additive but likely synergistic. For individuals with generalized frailty, an isolated improvement in strength may not translate to better mobility if balance remains impaired. MCT concurrently addresses these interconnected domains (strength, balance, endurance), thereby creating a more robust and functional adaptation, as theorized by Cadore et al. (6).

Beyond Physical Exercise: The Role of Adjuvant Strategies: The discussion on integrating nutritional strategies like HMB supplementation is pertinent but requires critical nuance. The evidence suggests that such adjuvants may potentiate gains primarily in populations at high risk for malnutrition or sarcopenia. Their role in otherwise well-nourished, institutionalized individuals is less clear and should not be assumed to provide universal additive benefit beyond the exercise stimulus itself.

Feasibility as a Key Outcome in Complex Patients: The demonstration of high adherence and safety in extremely vulnerable populations, such as advanced cancer patients (22), is a finding of paramount importance. It challenges therapeutic nihilism and shifts the success metric from sheer efficacy to safety and feasibility. This evidence is crucial for justifying the allocation of clinical resources to implement exercise oncology programs and for setting realistic patient expectations.

From Evidence to Implementation: The high adherence rates reported across these studies are not incidental but are a direct result of deliberate adaptations in intensity, logistics, and supervision. This translates to a clear practical implication: the successful implementation of MCT in geriatric care requires a supportive infrastructure that prioritizes professional supervision and individualization. Therefore, the next major barrier is not proving efficacy, but overcoming policy and resource constraints to integrate these programs into standard care within long-term care facilities and specialized clinics.

In summary, MCT should be considered the foundational exercise prescription for complex geriatric patients. Future work should focus on dismantling implementation barriers and refining the targeted application of adjuvant strategies like nutritional supplementation to specific, high-risk subgroups.

The broader analysis of included clinical trials provides robust scientific support for the central role of RT in promoting functionality, metabolic health, and well-being among older adults. Evidence indicates that even in contexts of frailty, chronic disease, or institutionalization, regular structured RT is safe, feasible, and effective, yielding benefits across a wide spectrum of health domains.

Foundational Benefits for Muscle and Function: RT consistently produces the most robust improvements in muscular strength and functional fitness. Studies demonstrate its efficacy even in challenging populations, such as improving isokinetic strength in obese older women (23) and enhancing walking speed and functional strength in women with sarcopenia through accessible elastic-band training (3). These findings underscore its essential role in combating sarcopenia and promoting autonomy.

Cognitive and Multimodal Integration: Concerning cognitive function, findings are heterogeneous and highlight a critical nuance. While conventional RT alone shows limited cognitive effects (13), integrating cognitive tasks during training (dual-task training) proves more effective for cognitive improvement (18). This suggests that the cognitive benefits of exercise are not automatic but are contingent upon the nature of the stimulus, emphasizing the value of motor-cognitive integration.

Metabolic and Body Composition Benefits: From a metabolic perspective, RT—even at moderate intensities—effectively improves blood pressure, functional capacity, and metabolic biomarkers (11, 15). Furthermore, it is a potent stimulus for favorable body composition changes, including reductions in fat mass and increases in lean mass (9).

Feasibility and Safety Considerations: The demonstrated efficacy of simple, accessible approaches like elastic-band training (11, 15) confirms that RT can be successfully implemented with high adherence even in resource-limited settings. Nevertheless, certain precautions are warranted. Emerging evidence raises a hypothesis that high-intensity RT may induce transient cellular alterations (24), underscoring the importance of personalized, and possibly monitored, prescriptions in vulnerable populations.

In conclusion, RT serves as a cornerstone intervention for aging adults, with proven benefits for physical function, metabolism, and body composition. Its application is highly adaptable, but optimal outcomes require personalized prescription that considers individual goals, clinical status, and the nuanced evidence regarding its effects on cognitive and cellular health.

This review consolidates RT as a non-negotiable, central dogma in geriatric health. However, moving beyond this affirmation, our synthesis crystallizes key insights about its scope, limitations, and the sophisticated framework required for its optimal application.

Reaffirming Universality and Accessibility: The consistent efficacy of RT across a spectrum of modalities—from machines to elastic bands—is a finding of profound practical importance. It demonstrates that the principle of progressive overload is universally applicable, democratizing access to its benefits regardless of setting or resources. This evidence should empower clinicians to prescribe RT with confidence in diverse environments, from high-tech gyms to resource-limited nursing homes.

Defining the Boundary of Cognitive Transfer: The repeated divergence between physical and cognitive outcomes [e.g., (13, 18)] allows us to define a critical boundary: RT does not automatically confer cognitive benefits. Cognitive improvement is not a passive byproduct but an active acquisition that likely requires explicit cognitive engagement during training. This reframes the role of RT in brain health from a direct mediator to a potent platform for delivering concurrent cognitive stimulation.

The Primacy of the Exercise Stimulus Over Passive Supplements: The consistent failure of isolated micronutrient supplementation (e.g., vitamin D, omega-3) to augment the effects of RT (20) delivers a clear message. In the absence of deficiency, the active contractile stimulus is the primary and most potent driver of adaptation. This firmly places the focus—and resources—on the quality and supervision of the exercise program itself, rather than on ancillary, often ineffective, supplements.

Confronting and Contextualizing Molecular Trade-offs: The observations of potential genomic instability (24) should not be alarmist but should instead catalyze a more nuanced understanding of exercise physiology in aging. These findings hypothesize a potential molecular trade-off between the potent functional anabolism of high-intensity RT and cellular stress in very frail individuals. This underscores that “one-size-fits-all” high-intensity prescriptions are obsolete, mandating a future where exercise prescription is not only personalized to function but also informed by biomarker feedback.

In conclusion, the era of simply recommending “strength training” is over. The frontier now lies in implementing precision exercise prescription: a sophisticated integration of modality, intensity, cognitive demand, and nutritional support, all carefully calibrated to an individual’s clinical, functional, and potentially even biological profile, to maximize benefit and mitigate any nascent risks.

While RT is crucial for strength, this review underscores that aerobic training constitutes the foundational pillar for enhancing cardiorespiratory endurance and functional capacity in older adults. A critical synthesis, however, reveals that its role is both complementary and subject to important specificities.

The Primacy of Aerobic Exercise for Functional Mobility: The most consistent benefit of aerobic exercise lies in its unparalleled ability to improve walking capacity (e.g., 6MWT) and functional endurance. This is not merely a performance metric; it is the physiological basis for performing extended activities of daily living independently. The efficacy of even low-intensity, seated aerobic protocols highlights its unique applicability to the most frail individuals for whom high-impact or high-load activities are not feasible.

The Synergy with RT: A Non-Competitive Partnership: It is critical to frame aerobic and RT not as competitors but as synergistic partners. While aerobic exercise optimally improves central cardiorespiratory function and endurance, RT addresses peripheral musculoskeletal strength. The multicomponent interventions showing the greatest overall functional gains [e.g., (5)] validate that the combination of these two modalities produces a more comprehensive physiological adaptation than either alone.

Aerobic Exercise and Cognition: Clarifying the Pathway: The inconsistent cognitive outcomes associated with aerobic training suggest its effects are not automatic. The improvements in specific domains like executive function are hypothesized to be mediated by enhanced cerebrovascular flow and BDNF upregulation. However, this pathway appears most robust when the aerobic stimulus is coupled with an inherent cognitive challenge, such as navigating an environment or following complex routines, moving beyond simple, repetitive motions.

The Critical Role of Feasibility and Adherence: The success of home-based and low-cost aerobic interventions (e.g., using DVDs) is a testament to the critical importance of feasibility. For a behavior as crucial as sustained aerobic activity, an easily accessible, moderate-intensity program that fosters long-term adherence is often superior to an intensive, supervised program that is ultimately abandoned.

In conclusion, aerobic training is not a secondary option but a core component of geriatric exercise prescription. Its primary mandate is to sustain the cardiorespiratory engine that powers functional independence. The future of its application lies in intelligently integrating it with RT and cognitive activities to create synergistic, feasible, and personalized multimodal interventions.

In summary, despite the heterogeneity of the included literature, several key findings consistently emerge: (1) Structured exercise, particularly resistance and MCT, is a safe and effective means to improve functional fitness in older adults. (2) The most significant functional gains are often observed in frail or clinically compromised populations. (3) High adherence, facilitated by supervision and program individualization, is a critical mediator of success. (4) While physical benefits are robust, cognitive benefits are less consistent and may require specifically designed interventions that integrate cognitive challenges.

First, the protocol for this systematic review was not prospectively registered, which may introduce concerns regarding potential reporting bias and transparency. Second, our search was limited to three major databases (MEDLINE, PMC, and PubMed Central Canada). While this focused approach captured a substantial body of literature, it may have missed some relevant studies indexed in other databases such as Embase or CINAHL. Despite the methodological rigor of this systematic review—particularly the exclusive inclusion of RCTs and the use of validated instruments to assess functionality—several important limitations should be acknowledged. First, most included studies feature relatively small and heterogeneous samples, which may compromise the generalizability of results. Moreover, there is substantial variability in exercise protocols (duration, intensity, type, and frequency), hindering direct comparisons and the establishment of an optimal dose–response relationship. Finally, we were unable to adequately explore the potential moderating effect of individual characteristics such as sex or baseline health status on the outcomes, as the majority of included studies did not report results stratified by these factors.