Resistance training effect on functional decline in older adults with dementia

Abstract

BACKGROUND

Dementia is commonly accompanied by impaired mobility and increased falls, contributing substantially to disability and dependence. Pharmacological treatments have provided limited benefit for physical function, increasing interest in non-pharmacological interventions aimed at preserving mobility and independence. Resistance training may represent a promising intervention for preserving mobility in dementia, although evidence regarding its effectiveness remains incompletely synthesised.

AIM

To determine whether resistance training decreases functional decline in older adults with dementia, particularly gait speed, falls, mobility, and independence.

METHODS

A meta-analysis was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and was prospectively registered with PROSPERO (CRD420251138318). Randomised controlled trials on resistance training in adults aged ≥ 60 years with dementia were identified through PubMed, CINAHL, PsycINFO, and the Cochrane Central Register of Controlled Trials. Primary outcomes were gait speed and falls; secondary outcomes included physical function, daily activities, muscle strength, quality of life, cognition, and adverse events. Risk of bias and meta-analysis were completed using standard tools.

RESULTS

Eight randomised controlled trials involving 1059 participants were included. Meta-analysis of five trials (n = 329) demonstrated a statistically significant improvement in gait speed favouring resistance training (mean difference 0.10 m/second, 95% confidence interval: 0.04-0.15; P = 0.0004) with negligible heterogeneity (I² = 0%). Four trials reporting fall outcomes showed a consistent trend toward reduced fall incidence (incidence rate ratios = 0.86, 95% confidence interval: 0.71-1.05), although statistical significance was not reached. Secondary outcomes, including physical performance, activities of daily living, and muscle strength, demonstrated variable findings across studies.

CONCLUSION

Resistance training improves gait speed and may reduce falls in older adults with dementia, supporting its safe integration into the dementia care pathway, to enhance mobility and independence.

Key Words: Dementia; Resistance training; Gait speed; Accidental falls; Physical functional performance

Core Tip: Resistance training results in consistent and clinically meaningful improvements in gait speed in older adults with dementia, with negligible heterogeneity across trials. Although a non-significant reduction in fall incidence is observed, the direction and magnitude of effect suggest a probable benefit limited by current sample sizes. These findings support resistance training as a key intervention to preserve mobility and potentially reduce fall risk in this population.

INTRODUCTION

Dementia represents one of the most significant global public health challenges of the twenty-first century, affecting more than 55 million people across the globe, with prevalence expected to rise substantially as populations age[1]. The condition comprises a group of progressive neurodegenerative disorders characterised by cognitive impairment, behavioural symptoms, and declining functional capacity leading to nursing home admission[2,3]. Although cognitive deterioration is dominant to dementia, loss of physical function and mobility often has the greatest impact on independence, caregiver burden, institutionalisation, and quality of life[3,4].

Functional impairment in dementia is multifactorial and commonly involves deficits in mobility due to altered gait, balance, coordination, and muscle strength in addition to cognitive dysfunction, of which studies show can be prevented[5,6]. Among these impairments, gait dysfunction is one of the most consistently observed mobility abnormalities which are highly linked to falls. Individuals with dementia frequently demonstrate slower gait speed, shorter stride length, and greater gait variability compared with cognitively healthy older adults[7-9]. Reduced gait speed is clinically important, as it is strongly associated with falls, disability, hospitalisation, and mortality in older populations and early institutionalisation[3,10].

Falls represent a major source of morbidity in individuals living with dementia, occurring at rates substantially higher compared to cognitively intact older adults[11-13]. Fall-related injuries commonly lead to fractures, hospitalisation, reduced confidence, and accelerated loss of independence, which exercise has shown to reduce[12,14]. Increased risk of fall in dementia is influenced by multiple factors, including impaired balance, gait abnormalities, reduced muscle strength, visuospatial deficits, and medication use[15,16]. Consequently, interventions capable of improving mobility and reducing fall risk are of considerable clinical importance in the dementia context and can help address dementia attitudes[16].

Current pharmacological therapies for dementia primarily target cognitive symptoms and provide limited benefit for physical function and mobility impairment, which is a concern as costs for dementia care continue to increase[17,18]. As a result, increasing attention has focused on non-pharmacological interventions that address modifiable contributors to functional decline in this domain[19]. Exercise-based interventions have emerged as a promising strategy for preserving physical function in older adults, particularly those living with dementia[20-22].

Resistance training, also referred to as strength or progressive resistance exercise, involves performing exercises against external resistance such as weights, resistance bands, machines, or body weight to preserve body functioning[7,23]. In healthy older adults, resistance training has consistently been shown to improve muscle strength, gait speed, balance, and overall physical performance[24-26]. These effects are particularly relevant in later life, where age-related declines in muscle mass and strength contribute significantly to impaired mobility and increased fall risk[27,28].

Several physiological mechanisms may explain the potential benefits of resistance training in dementia. Improvements in lower-limb muscle strength may enhance gait propulsion and walking stability, while adaptations in neuromuscular coordination may improve balance and postural control[14,29-32]. In addition, emerging evidence suggests that exercise may influence cerebral blood flow and neuroplasticity, potentially contributing to broader neurological benefits, though not without adequate nutrition[32-34].

Despite these potential advantages, the effectiveness of resistance training in individuals with dementia remains uncertain and unexploited[7,33], although cognitive impairment, behavioural symptoms, and medical comorbidities may affect exercise participation and responsiveness to training interventions[35,36]. Furthermore, previous systematic reviews investigating exercise interventions in dementia have frequently combined aerobic, balance, and strength-based programmes, making it difficult to isolate the specific effects of resistance training on clinically meaningful outcomes[37-39].

Several recent randomised controlled trials have examined resistance-based exercise interventions in individuals with dementia, providing additional evidence regarding their effects on mobility and fall risk[40,41]. A focused synthesis of this evidence may therefore provide clearer insight into the role of resistance training as a therapeutic strategy in dementia care and guide the existing pathways. The aim of this meta-analysis was to evaluate the effects of resistance training on functional decline in older adults with dementia, with particular emphasis on gait speed and falls as clinically meaningful indicators of mobility and independence.

MATERIALS AND METHODS

This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[42,43] and prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO; No. CRD420251138318).

Eligibility criteria

Studies were eligible if they met a predetermined criterion as follows.

Population: Adults aged 60 years or older with a clinical diagnosis of dementia based on recognized diagnostic criteria such as Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, or International Statistical Classification of Diseases and Related Health Problems, 10^th Revision[44,45].

Intervention: Structured resistance training programme involving exercises performed against external resistance such as weights, resistance bands, machines, or body weight[40].

Comparison: Usual care, social activities, or alternative exercise interventions not including resistance training.

Outcomes: Primary outcomes were gait speed and falls. Secondary outcomes included physical performance, activities of daily living, muscle strength, quality of life, cognition, and adverse events[46-48].

Study design: Randomised controlled trials.

Search strategy

A comprehensive literature search was conducted in PubMed, CINAHL, PsycINFO, and the Cochrane Central Register of Controlled Trials in September 2025 with no date or language restrictions. The search strategy combined concepts related to dementia, resistance training, functional decline, and randomised controlled trials. Boolean operators, MeSH terms, and controlled vocabulary tailored to each database were applied during the literature search[49]. Reference lists of relevant articles and systematic reviews were also screened to identify additional eligible studies (Figure 1).

Figure 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses: Flow diagram of study selection and inclusion process. RCT: Randomized controlled trial; SR: Systematic review; MA: Meta-analysis.

Study selection

Articles from the search results were exported to a Microsoft Excel spreadsheet, and duplicates were removed. Titles and abstracts were screened against the eligibility criteria, and a consensus on the final articles was reached by the two authors. Full-text articles were then retrieved and reviewed to determine final inclusion. Any disagreements were resolved through discussion between the student and supervisor.

Data extraction

Data extraction was performed using a standardised form. Extracted data included study design and characteristics, participant demographics, intervention and comparator details, outcome measures, settings, duration and reported results. The extracted data were reviewed by the supervisor to ensure accuracy.

Risk of bias assessment

Risk of bias was assessed using the Cochrane Risk of Bias 2 tool. This tool evaluates bias across five domains: Randomisation process, deviations from intended interventions, missing outcome data, measurement of outcomes, and selection of reported results[50].

Statistical analysis

A meta-analysis was conducted using random-effects models to address variability between studies. Gait speed outcomes were analysed using the mean difference in metres per second (m/second). Falls were analysed as recurrent events using incidence rate ratios (IRRs) based on total falls and approximated person-time (sample size × follow-up duration), allowing pooling across studies with different follow-up periods. Study-specific log (IRR) estimates and variances were calculated assuming a Poisson distribution. A continuity correction of 0.5 was applied where zero events occurred in one study arm. Pooled effects were estimated using restricted maximum likelihood and presented as IRRs. Statistical heterogeneity was assessed using the I² statistic[51-54]. Sensitivity analyses were performed to assess robustness, and overall and subgroup findings were additionally synthesised narratively where appropriate.

Ethical considerations

Ethical approval for this systematic review and meta-analysis was obtained in accordance with the university guidelines. The ethics application form was fully completed, submitted, and approved by the relevant ethics committee prior to commencement of the study[55].

RESULTS

Study selection

Database searches identified 1015 records. After removal of duplicates and screening procedures, eight randomised controlled trials met the inclusion criteria[56-59] and were included in the systematic review[60-63].

Study characteristics

The eight included trials comprised a total of 1059 participants of all genders. The mean participant age was 82.4 years, and approximately 68% were female. Studies were conducted across several countries of different continents, including the United Kingdom, Sweden, Germany, France, Spain, the Netherlands and Australia.

Participants, across genders, included both community-dwelling individuals and residents of long-term care facilities. Dementia severity ranged from mild to severe, although most participants had mild-to-moderate cognitive impairment[44,45]. Resistance training interventions varied in frequency, intensity, and duration. Most programmes involved supervised exercise sessions targeting major muscle groups, particularly lower limbs. Intervention frequency ranged from two to five sessions per week, and programme duration ranged from eight weeks to twelve months.

Risk of bias

Three studies were assessed as having low risk of bias, three studies had some concerns, and two studies were considered to have high risk of bias due to issues related to missing outcome data or deviations from intended interventions.

Primary outcomes

Gait speed: Five trials involving 329 participants reported gait speed outcomes and were suitable for meta-analysis[57-61]. A random-effects meta-analysis of post-intervention gait speed was performed using mean difference (m/second) as the effect measure. Four studies reported mean ± SD directly, while one study[61] reported medians with interquartile ranges, which were conservatively converted to approximate mean ± SD; the two intervention arms were combined to avoid double-counting the control group. The pooled analysis demonstrated a statistically significant improvement in gait speed favouring resistance training [mean difference 0.10 m/second, 95% confidence interval (CI): 0.04-0.15; P = 0.0004]. Heterogeneity was negligible (I² = 0%), indicating highly consistent findings across studies despite differences in participant characteristics, interventions, and follow-up duration. Sensitivity analysis excluding the converted study[61] yielded virtually identical results (mean difference 0.10 m/second, 95%CI: 0.04-0.15), confirming the robustness of the finding. The magnitude of improvement exceeded the commonly reported minimal clinically important difference for gait speed in older adults[64], demonstrating a clinically meaningful benefit (Figure 2).

Figure 2 Forest plots. A: Effect of resistance training on gait speed in older adults with dementia, forest plot for the random-effects meta-analysis; B: Effect of resistance training on fall incidence in older adults with dementia, forest plot for the random-effects meta-analysis. CI: Confidence interval; IRR: Incidence rate ratio.

Falls: Four trials involving 803 participants reported falls outcomes[59,61-63]. The results were pooled using a random-effects IRR model, appropriate for recurrent events and differing follow-up durations. The pooled analysis demonstrated a trend toward fewer falls favouring resistance training compared with control (IRR = 0.86, 95%CI: 0.71-1.05), corresponding to an approximate 14% relative reduction in fall incidence, although this did not reach conventional statistical significance. Heterogeneity was low to moderate (I² approximately 25%), indicating broadly consistent findings across studies despite differences in populations, interventions, and follow-up periods. Larger studies contributed most of the statistical weight and generally showed small-to-moderate reductions in falls, while a small zero-event study contributed minimal influence due to imprecision. Sensitivity analyses excluding this study and varying continuity corrections produced similar findings, confirming the robustness. Overall, resistance training may reduce fall incidence, although the current evidence base may be underpowered to demonstrate a definitive effect (Figure 2).

Secondary outcomes

Secondary outcomes, which included the short physical performance battery, timed up and go test, activities of daily living measures, muscle strength, quality of life, and cognition, showed greater variability across studies. Several trials reported improvements in physical performance measures and muscle strength following resistance training interventions. However, results for activities of daily living were inconsistent, with some studies reporting modest improvements and others finding no significant differences between groups. Quality of life and cognitive outcomes generally showed no significant improvements associated with resistance training[56,62,65].

Adverse events

Resistance training interventions were generally well tolerated. Minor musculoskeletal complaints, such as transient muscle soreness, were occasionally reported. No serious adverse events were directly attributed to the exercise interventions.

DISCUSSION

This meta-analysis shows that resistance training produces clinically meaningful improvements in gait speed in older adults suffering from dementia. The pooled improvement of approximately 0.10 m/second exceeded established thresholds for minimal clinically important change and represents a meaningful enhancement in mobility. Gait speed is widely recognised as an important indicator of functional health in older adults. Slower walking speed has been associated with increased risk of hospitalisation, disability, institutionalisation, and mortality[57-61,66,67]. Consequently, interventions capable of improving gait speed may contribute to improved independence and quality of life.

The findings of this review indicate that individuals with dementia still have the capacity to achieve significant physical adaptations in response to resistance training. Importantly, the magnitude of improvement observed in this review is comparable to that reported in studies involving cognitively healthy older adults[68-70]. This suggests that cognitive impairment does not substantially limit the physiological benefits of resistance training. The analysis also indicated a potential reduction in fall incidence associated with resistance training. Although the pooled estimate did not reach statistical significance, the direction of effect was consistent across studies and suggests a clinically relevant reduction in fall risk[59,61-63].

Several physiological mechanisms may explain the improvements observed in gait speed. Resistance training increases muscle strength, particularly in the lower limb musculature responsible for gait propulsion and stability[66]. Improved muscle strength may enhance walking efficiency and enable individuals to maintain faster walking speeds. Resistance training may also improve neuromuscular coordination and balance, both of which are critical for maintaining safe mobility in older adults[40]. Enhanced balance control may reduce compensatory gait adaptations that typically result in slower walking speeds among individuals with dementia[71-73].

Despite these encouraging findings, several limitations should be considered, especially the heterogeneity of resistance training protocols across included studies that included variations in frequency, intensity, duration, supervision, and exercise type. Follow-up periods were also relatively short, limiting conclusions regarding long-term effects, and differences in dementia severity and study settings from community to institutional care may contribute to residual variability in outcomes. Such variability suggests that resistance training interventions may require adaptation to individual patient needs, care settings, and resource availability to allow for real-world implementation. Also, subgroup analyses were limited by the small number of included studies and inconsistent reporting, highlighting the need for future research to explore these factors.

Nevertheless, resistance training represents a feasible and relatively low-cost intervention that can be implemented in both community and residential care settings as researchers continue to address the highlighted limitations. Given the limited pharmacological options available to address functional decline in dementia, exercise interventions may represent an important component of comprehensive dementia care. Future research should focus on larger randomised controlled trials with longer follow-up periods and standardised outcome measures, and tailored dementia care settings to better evaluate the long-term benefits of resistance training in individuals with dementia and what suits personalised care.

CONCLUSION

This meta-analysis demonstrates that resistance training produces consistent and clinically meaningful improvements in gait speed in older adults with dementia and may contribute to reductions in fall risk. The pooled improvement in gait speed exceeded established thresholds for minimal clinically important change, indicating that resistance training can generate meaningful benefits in mobility and functional performance in this population. Given that gait speed is strongly associated with adverse outcomes such as disability, hospitalisation, institutionalisation, and mortality, these improvements may translate into important gains in independence and quality of life for individuals living with dementia. Importantly, resistance training interventions were generally safe and well-tolerated across the included trials, supporting their feasibility for implementation in both community and residential care settings.

Despite these encouraging findings, the current evidence base remains limited by relatively small sample sizes, heterogeneity in intervention protocols, and relatively short follow-up periods. Although a trend toward reduced fall incidence was observed, the available trials were not sufficiently powered to establish a definitive effect on fall outcomes. Larger and adequately powered randomised controlled trials are therefore required to more clearly determine the effects of resistance training on falls and other clinically important endpoints.

Future research should also focus on identifying optimal training parameters, including exercise intensity, frequency, and programme duration, as well as evaluating the long-term sustainability of functional improvements. In addition, studies exploring strategies to enhance adherence and implementation of exercise programmes in real-world care settings will be essential. Overall, resistance training represents a promising non-pharmacological intervention for preserving mobility, supporting functional independence, and improving quality of life in older adults with dementia.

ACKNOWLEDGEMENTS

The author sincerely thanks Almighty God, DSH sisters for the opportunity to study, and my biological family back in Kenya, for their moral support. Appreciation is also extended to my manager and the OPAU team at Morriston Hospital for their support, and to Ranges Academy Munyuini, Morriston Hospital Library, and Morriston Council Library for their valuable resources and support throughout this work. We extend our appreciation to the Faculty of Life Sciences and Education at the University of South Wales, in association with Learna Ltd., for the Acute Medicine MSc program and their invaluable support in our work. We sincerely acknowledge the efforts of the University of South Wales and commend them for their commitment to providing lifelong learning opportunities and advanced life skills to healthcare professionals.