Sex and gender differences in technology-based rehabilitation for people with stroke: A scoping review

Abstract

Objectives

This scoping review maps evidence on sex and gender differences in technology-assisted stroke rehabilitation and identifies gaps to inform equitable approaches.

Data sources

Searches were conducted in two rounds (up to March 2025 and April 2025–March 2026) on MEDLINE, CINAHL, Scopus, Web of Science, Cochrane Library, and Google Scholar.

Review methods

Following PRISMA-ScR guidelines, we included primary studies reporting sex and gender outcomes in adults with stroke receiving technological interventions, guided by the PCC framework.

Results

From 7071 records, 11 studies (1626 stroke survivors, 39.9% female) met inclusion criteria. Interventions included robotic exoskeletons, soft robotic gloves, virtual reality, exergaming, functional electrical stimulation, and biofeedback. Five studies found no significant sex differences. However, others reported female advantages in robotic gait training (p = 0.007), functional electrical stimulation (OR = 3.92), soft robotic glove hand function (3.43-fold greater improvement), and Virtual Reality-based balance recovery (p = 0.03). Younger women (<62 years) outperformed men in fine motor tasks, a pattern reversing with age.

Conclusion

Gender and sex disparities in rehabilitation outcomes vary by technology type, underscoring the need for inclusive device designs and improved female representation in trials. Future research must integrate biological and sociocultural perspectives and standardize outcome measures to optimize tailored rehabilitation for all stroke survivors.

Keywords

Stroke gender differences sex differences technology-based interventions personalized rehabilitation

Introduction

Stroke is a leading cause of death and disability worldwide, with over 7 million deaths and 160 million disability-adjusted life years lost annually.¹ Post-stroke disability, particularly in motor and balance domains, is expected to rise with population ageing.^2,3 Technology-based rehabilitation, including robotics, exoskeletons, virtual reality, augmented reality, telerehabilitation, and wearable sensors, offers intensive and adaptive training that supports motor recovery, gait, balance, and upper limb function.^4–7 Recent evidence highlights the significant potential for improvement in functional mobility and gait adaptability through these advanced systems. For example, gait adaptation training on an augmented reality treadmill⁸ and digital exercise combined with augmented reality training systems⁹ have demonstrated substantial benefits in improving mobility and walking performance among patients with stroke. Efficacy, however, depends on matching the right device to the right patient.¹⁰

Sex refers to biological attributes that influence cardiovascular and neurological function, whereas gender encompasses socially constructed roles and behaviors that shape health experiences.¹¹ Both dimensions affect outcomes: sex determines vulnerability and recovery potential, while gender influences risk exposure, access to care, adherence, and social support.¹¹

While men have higher incidence rates in midlife for stroke, women, especially postmenopausal, experience poorer outcomes, influenced by biological, psychosocial, and systemic sampling bias factors.^12–14 Yet, sex and gender remain under-investigated in rehabilitation research.¹⁵ Male-centered device design may compromise fit or efficacy for women, while virtual therapy and gaming interfaces may engage genders differently, influencing adherence.¹⁶ Similarly, acceptance of home-based wearable technologies may vary by gender.¹⁷

This scoping review maps current evidence on sex and gender differences in technologically assisted stroke rehabilitation, focusing on functional outcomes such as motor recovery, gait, balance, endurance, and upper limb function.

Methods

This scoping review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Scoping Reviews. The protocol was prospectively registered on the Open Science Framework (https://doi.org/10.17605/OSF.IO/2EGXB) and structured according to the Population–Concept–Context framework.

Studies were eligible if they included adults (≥18 years) with ischemic or hemorrhagic stroke and reported sex- or gender-based outcomes. The review focused on functional recovery measures: motor recovery, gait, balance, endurance, and upper limb function after technology-based rehabilitation. Eligible interventions included robotics, exoskeletons, virtual reality, telerehabilitation, brain–computer interfaces, neuromodulation, wearable sensors, and haptic feedback. Both clinical and home-based settings were included, whereas secondary literature and studies using only conventional rehabilitation were excluded.

Six databases were searched: MEDLINE, CINAHL, Scopus, Web of Science, Cochrane Library, and Google Scholar for grey literature. No limits on publication year or language were applied. The strategy involved three stages: preliminary searches to identify keywords, refined search strings with Boolean operators applied across all databases, and manual screening of reference lists (full syntax is available in Supplemental material).

All records were imported into Rayyan for deduplication and screening.¹⁸ Two reviewers independently screened titles, abstracts, and full texts; disagreements were resolved with a third reviewer. The study selection process is presented in the PRISMA flow diagram.

Results

The systematic search was conducted in two consecutive rounds: the first up to March 2025 and the second from April 2025 to March 2026. Overall, 7071 records were identified. After duplicate removal, 5735 records were screened by title and abstract, of which 5715 were excluded. Twenty full-text reports were assessed for eligibility. Following exclusion of reports without sex- or gender-stratified analyses and one review article, 11 studies met the inclusion criteria: 10 from the first search and 1 from the second (Figure 1).

Figure 1.

PRISMA-ScR flow diagram.

The included studies, published between 2014 and 2026, were conducted across Italy, Poland, Canada, Russia, Japan, Hungary, and Iraq. Overall, 1626 stroke survivors were enrolled across the 11 studies. Female representation ranged from 32.4% to 47%, with an overall mean of approximately 39.9%. Study designs included randomized controlled trials,^19–22 one pseudorandomized trial,²³ retrospective cohorts,^24–25 and prospective cohorts.^26–29

Participants were predominantly in the subacute phase (eight studies), while chronic stroke populations were investigated in three studies.^27–29 Mean age ranged from 51.9 (12.5) years²⁶ to 66.5 (5.9) years.²³

A broad range of rehabilitation technologies was employed. Robotic systems included InMotion 2.0,²⁴ KINARM exoskeleton,²⁹ Lokomat,²¹ Gait Trainer GT,^19–20 and the ARAMIS exoskeleton.²⁵ Treadmill-based biofeedback was delivered through Gait Trainer 2,²⁸ while immersive virtual reality was implemented using the GRAIL system.²⁶ Functional electrical stimulation with WalkAide was used to improve gait speed.²⁷ Additional interventions included Xbox 360-based exergaming²³ and a soft robotic glove for upper-limb recovery.²² All included studies were conducted in clinical rehabilitation settings.

Intervention intensity and duration varied substantially across studies. The number of sessions ranged from 4²⁹ to 50,²³ with most protocols delivered five times per week. Examples included 20 sessions over 4 weeks (45 min/session) in robotic upper-limb rehabilitation,²⁴ 10 sessions over 2 weeks (90 min/session) for treadmill biofeedback,²⁸ twice-daily exergaming sessions over 5 weeks,²³ and 8 weeks of soft robotic glove training performed 3–4 times weekly.²²

Primary outcomes mainly focused on motor recovery, including Fugl-Meyer Assessment scores^{22,24–25,27} and Functional Ambulation Category.^19–21 Additional measures assessed activities of daily living, such as the Modified Barthel Index^23–24 and Functional Independence Measure,^{21,26,28–29} as well as gait and balance outcomes, including the 10-meter walk test,^27,29 Berg Balance Scale,^23,26 and 6-minute walk test.^23,26 Psychosocial outcomes, particularly anxiety and depression, were also evaluated in selected studies.^23,26

Sex and gender related analyses were reported in all 11 studies, although no study included data on non-binary individuals. Across the evidence base, findings were mixed but suggest a recurring trend toward greater responsiveness among women in selected technology-assisted rehabilitation outcomes. Five studies reported no significant sex differences in their primary outcomes.^{19,23,25,28–29} However, the remaining studies identified clinically relevant female advantages in specific domains.

Women showed superior gains in robotic gait rehabilitation,¹⁹ greater session-normalized improvements in functional independence after Lokomat training,²¹ and a markedly higher likelihood of responding to functional electrical stimulation for gait recovery, with almost fourfold greater odds compared with men.²⁷ Similarly, immersive virtual reality rehabilitation was associated with better balance recovery in women.²⁶ In upper-limb rehabilitation, women receiving soft robotic glove therapy demonstrated substantially greater improvements in hand opening rate than men, while no sex-related differences emerged in the control condition.²²

Some findings suggest that the relationship between sex and recovery may be further modified by age. One study reported a significant age by sex interaction in manual dexterity, whereby women younger than 62 years outperformed men, whereas the opposite pattern was observed in older participants.²⁹ A summary of the included studies is presented in Table 1.

Table 1.

Summary of included studies.

Author, year	N (F%)	Age (mean (SD) or range)	Technology	Intervention (type, duration, frequency)	Primary outcomes	Key sex/Gender-related findings
Franceschini et al., 2018	60 (41.7%)	65.8 (15.4)	InMotion 2.0(end-effect or robot)	4 weeks, 5 sessions/week, 45 min/session and conventional physiotherapy	Modified Barthel Index	Sex was not a predictor of favorable outcome (p > 0.05); no significant sex differences.
Guzik et al., 2019	50 (36%)	≈60 (11.7)	Treadmill with biofeedback (Gait Trainer 2)	2 weeks, 5 sessions/week, 90 min/session	10-metre walk test, TUG, 2-minute walk test, FIM, Barthel Index	Sex did not influence any outcome (p > 0.05 for all comparisons).
Hawe et al., 2020	160 (32.5%)	Males: 59.6 (12.8); Females: 58.0 (14.6)	KINARM exoskeleton robot (robotic assessment)	Longitudinal, observational assessments at 1, 6, 12, 26 weeks post-stroke	Robotic task scores (z-scores), FIM, CMSA, Purdue Pegboard, SAFE score	No sex differences in robotic tasks; exception: sex × age interaction for Purdue Pegboard (younger females outperformed younger males; older males outperformed older females).
Lamberti et al., 2021	236 (39%)	Females: 64 (59–67); Males: 64 (61–66) (median, IQR)	Lokomat (robotic gait trainer)	Robot-assisted gait training (30 min walking/session) + multidisciplinary rehabilitation; mean sessions: 14 (F) vs 15 (M); 1–5 sessions/week	Functional Ambulation Category, FIM (total, motor, cognitive)	No overall sex difference; but ΔFAC normalized per session significantly greater in females (p = 0.025). Females with ischemic stroke improved more than males with ischemic stroke.
Mayouf & Al-Jubouri, 2025	68 (not reported, but ∼50% female in glove group)	>20 years (mean not reported)	Soft robotic glove (SR Glove) vs conventional exercises	8 weeks, 3–4 sessions/week	Hand opening and closing scores (derived from FMA-UE, ARAT, AMAT)	In the glove group, females improved hand opening 3.43-fold vs 1.91-fold in males (p < 0.05); no sex differences in the control group
Morone et al., 2014	74 (32.4%)	RAGT: 61.5 (12.4); FAGT: 62.3 (13.4)	Gait Trainer (GT)	4 weeks, 5 sessions/week, 20 active min/session	FAC effectiveness (% of maximal possible improvement)	In the robotic group: females improved more than males (p = 0.0048); patients ≥65 years improved more than younger ones (p = 0.040).
Morone et al., 2018	100 (34%)	62.7 (12.4)	Gait Trainer (GT)	4 weeks, 5 sessions/week, 20 active min/session and conventional therapy	FAC, Trunk Control Test, Barthel Index	Sex was not significant in logistic regression (p = 0.722).
Pignolo et al., 2021	102 (38% F)	63.6 (13.1)	Exoskeleton robot (ARAMIS) + conventional therapy	7 weeks, 60 min/day, 22.7 ± 10.6 sessions total	FMA-UE, Motricity Index (MI), TCT, Barthel Index, FIM	Sex was associated with Barthel Index (females showed greater gains).
Shurupova et al., 2021	34 (47% F)	51.9 (12.5)	Immersive VR (GRAIL, Motekforce)	10 sessions, 40 min/session, over 2 weeks	Berg Balance Scale (BBS), TUG, FIM, RADLS, 6MWT, postural stability tests, STAI, BDI	Sex was a significant predictor of BBS improvement: females improved more than males (p = 0.03).
Sota et al., 2018	101 (32.7% F)	59.8 (12.9)	Functional Electrical Stimulation (WalkAide)	Minimum 20 min/session; 26.6 ± 19.6 sessions; 19.4 ± 18.2 total hours	10-metre walk test (gait speed), SIAS-FP, FMA-LE, MAS, ROM, TUG, 6MWT	Female sex was a predictor of being a responder (OR = 3.92, 95% CI 1.43–12.25, p = 0.007).
Tollár et al., 2021	641 (45.5% F)	66.5 (5.9)	Exergaming (Xbox 360: Reflex Ridge, Space Pop, Just Dance)	5 weeks, 5 sessions/week. EX2: twice daily (50 sessions); EX1: once daily (25 sessions); CON: 25 standard sessions	modified Rankin Scale (mRS), BI, EQ5-VAS, BBS, 6MWT, center of pressure (COP), blood pressure	No sex × group interaction (p > 0.05). Males and females improved equally on all outcomes.

RAGT: robot-assisted gait training; FAGT: floor-assisted gait training; FMA-UE: Fugl-Meyer Assessment Upper Extremity; FIM: Functional Independence Measure; CMSA: Chedoke McMaster Stroke Assessment; ARAT: Action Research Arm Test; AMAT: Arm Motor Ability Test; STAI: State-Trait Anxiety Inventory; BDI: Beck Depression Inventory; SIAS-FP: Stroke Impairment Assessment Set Foot-Pat Test; MAS: Modified Ashworth Scale; TUG: Timed Up & Go; ROM: range of motion.

Discussion

Our findings indicate that female stroke survivors achieved superior gains in several specific interventions, including robotic gait training,^19,21 functional electrical stimulation,²⁷ soft robotic glove hand therapy,²² and virtual reality-based balance training.²⁶ In contrast, no sex differences emerged in upper-limb robotics,²⁴ exergaming,²³ treadmill biofeedback,²⁸ or exoskeleton path analysis.²⁵

Several hypotheses from the included studies may help explain why women appeared to benefit more from certain technologies. One such hypothesis relates to hormonal neuroprotection. Hawe et al.²⁹ observed that women under 62 years outperformed men in fine motor tasks, a pattern that reversed in older age. This finding suggests that estrogen-related neuroplasticity and post-injury protection are more pronounced in premenopausal women. Another hypothesis concerns ischemic stroke pathophysiology. Lamberti²¹ and Shurupova²⁶ reported that female survivors of ischemic stroke responded better to intensive, repetitive training, possibly because of differences in cerebral collateral circulation or inflammatory responses. A further contributing factor may be orientation toward ADLs. Pignolo et al.²⁵ noted that women showed greater gains in the Barthel Index, which could reflect a higher motivation toward practical autonomy and self-care, thereby translating motor recovery into functional improvement more rapidly.

Moreover, operator bias is reduced when technology is used. Franceschini²⁴ and Tollár²³ suggested that robotics eliminate unconscious therapist bias, ensuring that women receive the same training intensity as men. Finally, a crucial distinction emerges between impairment and function. Hawe²⁹ found no sex differences in robotic measures of pure motor impairment, whereas Lamberti²¹ and Sota²⁷ found female advantages in functional outcomes. This divergence suggests that biological recovery alone may not fully explain the observed effects; rather, social roles, behavioral adaptation, caregiving expectations, and the relevance of outcome measures may shape how similar motor gains are translated into everyday function. Prior rehabilitation evidence indicates that women may achieve similar global disability scores yet outperform men in selected self-care and home-relevant domains.³⁰

Taken together, these hypotheses underscore that sex differences are not merely biological but are shaped by technological design, psychosocial factors, and rehabilitation intensity. Importantly, rigid exoskeletons often incorporate male-biased anthropometrics, which may disadvantage female users. By contrast, adaptive technologies such as exergaming²³ and immersive VR²⁶ appear more gender neutral. The underrepresentation of women (mean enrolment 39.9%) and the complete exclusion of non-binary individuals limit the generalizability of these findings.

In conclusion, this scoping review demonstrates that sex related differences in technology-based stroke rehabilitation are domain-specific and influenced by biological factors, technological design biases, and sociocultural variables. These findings do not support a sex only stratification model but rather highlight the need for multifactorial, inclusive, and equity-oriented approaches in future research and clinical practice. Standardized reporting of sex and gender disaggregated data, and inclusion of non-binary individuals are strongly recommended.

Clinical messages

• Responses to technology-based stroke rehabilitation may differ by sex and gender, particularly when device characteristics, functional goals, and social context are taken into account.

• Women may achieve greater gains with robotic gait training, functional electrical stimulation, soft robotic gloves, and virtual reality balance interventions.

• More inclusive device design and routine sex and gender disaggregated reporting are essential for clinically applicable evidence.

Supplemental Material

sj-docx-1-cre-10.1177_02692155261462424 - Supplemental material for Sex and gender differences in technology-based rehabilitation for people with stroke: A scoping review

Supplemental material, sj-docx-1-cre-10.1177_02692155261462424 for Sex and gender differences in technology-based rehabilitation for people with stroke: A scoping review by Romina Rahmaniharedasht, Martina Andò, Alex Lando, Nicola Ferri, Valentin Konika, Maria Rosaria Leo, Roberto Piperno, Andrea Turolla, Paolo Pillastrini and Marco Tramontano in Clinical Rehabilitation

Footnotes

ORCID iDs

Romina Rahmaniharedasht

Alex Lando

Paolo Pillastrini

Marco Tramontano

Author contributions

Romina Rahmaniharedasht: Conceptualization; Methodology; Data curation; Formal analysis; Investigation; Writing original draft; Writing-review & editing; Project administration.

Martina Andò: Methodology; Screening in blind review; Writing-review and editing; Approval of the final manuscript.

Alex Lando: Methodology; Screening in blind review; Writing-review and editing; Approval of the final manuscript.

Nicola Ferri: Methodology; Search strategy development; Writing-review and editing; Formal analysis; Data interpretation; Approval of the final manuscript.

Valentin Konika: Writing-review and editing; Approval of the final manuscript.

Maria Rosaria Leo: Writing-review and editing; Approval of the final manuscript.

Roberto Piperno: Conceptualization; Writing-review and editing; Approval of the final manuscript.

Andrea Turolla: Writing-review and editing; Data analysis and interpretation; Approval of the final manuscript.

Paolo Pillastrini: Supervision; Data analysis and interpretation; Writing-review and editing; Approval of the final manuscript.

Marco Tramontano: Supervision; Conceptualization; Methodology; Search strategy development; Data analysis and interpretation; Project administration; Writing-review and editing; Approval of the final manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Data are available from the corresponding author upon reasonable request.

Supplemental Material

Supplemental material for this article is available online.

References

Feigin

, et al. Global, regional, and national burden of stroke and its risk factors, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Neurol Oct. 2021; 20: 795–820.

Feigin

, et al. Pragmatic solutions to reduce the global burden of stroke: a World Stroke Organization–Lancet Neurology Commission. Amsterdam, Netherlands: Elsevier Ltd, 2023. doi: https://doi.org/10.1016/S1474-4422(23)00277-6.

Feigin

, et al. World stroke organization: global stroke fact sheet 2025. Int. J. Stroke Feb. 2025; 20: 132–144.

Morone

, et al. Robot-assisted gait training for stroke patients: Current state of the art and perspectives of robotics. Dove Medical Press Ltd, May 16, 2017, doi: 10.2147/NDT.S114102.

Morone

, et al. Who may have durable benefit from robotic gait training?: a 2-year follow-up randomized controlled trial in patients with subacute stroke. Stroke Apr. 2012; 43: 1140–1142.

Calabrò

, et al. Robotic neurorehabilitation in patients with chronic stroke: psychological well-being beyond motor improvement. Int J Rehabil Res Aug. 2015; 38: 219–225.

Iosa

. Seven capital devices for the future of stroke rehabilitation. Stroke Res Treat 2012; 2012: Article ID 187965, 9 pages. doi: https://doi.org/10.1155/2012/187965 https://doi.org/10.1155/2012/187965.

Yang

Gao

Zhou

, et al. Effects of gait adaptation training on augmented reality treadmill for patients with stroke in community ambulation. Int J Qual Health Care Feb. 2024; 36: mzae008.

Evron

Schwartz

Sajina

, et al. A digital exercise and augmented reality training system improved mobility among stroke patients: a randomized control trial. Technol Health Care 2024; 32: 89–101.

10.

Colamarino

, et al.

A scoping review of technology-based approaches for upper limb motor rehabilitation after stroke: are we really targeting severe impairment?

J Clin Med Sep. 2024; 13: 5414.

11.

Clayton

Tannenbaum

. Reporting sex, gender, or both in clinical research? JAMA 2016; 316: 1863–1864.

12.

Elgendy

Gad

Mahmoud

, et al. Acute stroke during pregnancy and puerperium. J Am Coll Cardiol Jan. 2020; 75: 180–190.

13.

Liu

, et al. Stroke and cerebrovascular disease in pregnancy: incidence, temporal trends, and risk factors. Stroke 2019; 50: 13–20.

14.

Reeves

, et al. Sex differences in stroke: epidemiology, clinical presentation, medical care, and outcomes. Lancet Neurol 2008; 7: 915–926.

15.

Howard

, et al. Sex and race differences in the association of incident ischemic stroke with risk factors. JAMA Neurol Feb. 2019; 76: 179–186.

16.

Kuo

Donelan

. Dynamic principles of gait and their clinical implications. Phys Ther Feb. 2010; 90: 157–174.

17.

, et al. Rehabilitation of motor function after stroke: A bibliometric analysis of global research from 2004 to 2022. Front Aging Neurosci Nov. 2022; 14. doi: 10.3389/fnagi.2022.1024163.

18.

“rayyan,”. Available: https://rayyan.ai/cite.

19.

Morone

Iosa

Marinozzi

, et al. Effectiveness of robotic assisted walking therapy: the role of age and sex,” in 2014 IEEE International Conference on Rehabilitation Robotics (ICORR), 2014, 1–3.

20.

Morone

Masiero

Coiro

, et al. Clinical features of patients who might benefit more from walking robotic training. Restor Neurol Neurosci 2018; 36: 293–300.

21.

Lamberti

Manfredini

Lissom

, et al. Beneficial effects of robot-assisted gait training on functional recovery in women after stroke: a cohort study. Medicina 2021; 57: 1200.

22.

Mayouf

Al-Jubouri

. The effect of soft robotic glove on the flexion and extension of hand for stroke patients: a clinical trial. Georgian Med News 2025; 361: 104–108.

23.

Tollár

Nagy

Csutorás

, et al. High frequency and intensity rehabilitation in 641 subacute ischemic stroke patients. Arch Phys Med Rehabil 2021; 102: 9–18.

24.

Franceschini

Goffredo

Pournajaf

, et al. Predictors of activities of daily living outcomes after upper limb robot-assisted therapy in subacute stroke patients. PLoS ONE 2018; 13: e0193235.

25.

Pignolo

Servidio

Basta

, et al. The route of motor recovery in stroke patients driven by exoskeleton-robot-assisted therapy: a path-analysis. Med Sci 2021; 9: 64.

26.

Shurupova

Aizenshtein

Trofimova

, et al. Clinical and anamnestic data that affect the outcome of rehabilitation on virtual reality in patients with stroke. Zhurnal Nevrologii i Psikhiatrii im. S.S. Korsakova 2021; 121: 33–40.

27.

Sota

Uchiyama

Ochi

, et al. Examination of factors related to the effect of improving gait speed with functional electrical stimulation intervention for stroke patients. PM&R 2018; 10: 1142–1150.

28.

Guzik

Drużbicki

Wolan-Nieroda

, et al. Analysis of the association between selected factors and outcomes of treadmill gait training with biofeedback in patients after stroke. J Back Musculoskelet Rehabil 2019; 32: 23–30.

29.

Hawe

Cluff

Dowlatshahi

, et al. Assessment of sex differences in recovery of motor and sensory impairments poststroke. Neurorehabil Neural Repair 2020; 34: 746–757.

30.

MacDonald

Hall

Bell

, et al. Sex differences in the outcomes of adults admitted to inpatient rehabilitation after stroke. PM&R 2021. doi: https://doi.org/10.1002/pmrj.12660.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.02 MB

0.00 MB