A State-of-the-Art of Exoskeletons in Line with the WHO’s Vision on Healthy Aging: From Rehabilitation of Intrinsic Capacities to Augmentation of Functional Abilities

Abstract

The global aging population faces significant health challenges, including an increasing vulnerability to disability due to natural aging processes. Wearable lower limb exoskeletons (LLEs) have emerged as a promising solution to enhance physical function in older individuals. This systematic review synthesizes the use of LLEs in alignment with the WHO’s healthy aging vision, examining their impact on intrinsic capacities and functional abilities. We conducted a comprehensive literature search in six databases, yielding 36 relevant articles covering older adults (65+) with various health conditions, including sarcopenia, stroke, Parkinson’s Disease, osteoarthritis, and more. The interventions, spanning one to forty sessions, utilized a range of LLE technologies such as Ekso®, HAL®, Stride Management Assist®, Honda Walking Assist®, Lokomat®, Walkbot®, Healbot®, Keeogo Rehab®, EX1®, overground wearable exoskeletons, Eksoband®, powered ankle–foot orthoses, HAL® lumbar type, Human Body Posturizer®, Gait Enhancing and Motivation System®, soft robotic suits, and active pelvis orthoses. The findings revealed substantial positive outcomes across diverse health conditions. LLE training led to improvements in key performance indicators, such as the 10 Meter Walk Test, Five Times Sit-to-Stand test, Timed Up and Go test, and more. Additionally, enhancements were observed in gait quality, joint mobility, muscle strength, and balance. These improvements were accompanied by reductions in sedentary behavior, pain perception, muscle exertion, and metabolic cost while walking. While longer intervention durations can aid in the rehabilitation of intrinsic capacities, even the instantaneous augmentation of functional abilities can be observed in a single session. In summary, this review demonstrates consistent and significant enhancements in critical parameters across a broad spectrum of health conditions following LLE interventions in older adults. These findings underscore the potential of LLE in promoting healthy aging and enhancing the well-being of older adults.

Full text

Citation: Gavrila Laic, R.A.; Firouzi,
M.; Claeys, R.; Bautmans, I.; Swinnen,
E.; Beckwée, D. A State-of-the-Art of
Exoskeletons in Line with the WHO’s
Vision on Healthy Aging: From
Rehabilitation of Intrinsic Capacities
to Augmentation of Functional
Abilities. Sensors 2024,24, 2230.
https://doi.org/10.3390/s24072230
Academic Editor: Philippe Gorce
Received: 5 March 2024
Revised: 26 March 2024
Accepted: 27 March 2024
Published: 30 March 2024
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
sensors
Systematic Review
A State-of-the-Art of Exoskeletons in Line with the WHO’s
Vision on Healthy Aging: From Rehabilitation of Intrinsic
Capacities to Augmentation of Functional Abilities
Rebeca Alejandra Gavrila Laic 1, †, Mahyar Firouzi 1,2,3,4,† , Reinhard Claeys 1,3,4 , Ivan Bautmans 5,
Eva Swinnen 1,3,4,* and David Beckwée1,4,5
1Rehabilitation Research, Department of Physiotherapy, Human Physiology and Anatomy,
Vrije Universiteit Brussel, Laarbeeklaan 103, 1090 Jette, Belgium; rebecagavrila@gmail.com (R.A.G.L.);
mahyar.firouzi@vub.be (M.F.); reinhard.claeys@vub.be (R.C.); david.beckwee@vub.be (D.B.)
2Brain, Body and Cognition Research Group, Faculty of Psychology and Educational Sciences,
Vrije Universiteit Brussel, Pleinlaan 2, 1050 Elsene, Belgium
3Center for Neurosciences (C4N), Vrije Universiteit Brussel, Pleinlaan 2, 1050 Elsene, Belgium
4Brubotics (Human Robotics Research Center), Vrije Universiteit Brussel, Pleinlaan 2, 1050 Elsene, Belgium
5FRIA, Frailty in Ageing, Vrije Universiteit Brussel, Laarbeeklaan 103, 1090 Brussels, Belgium;
ivan.bautmans@vub.be
*Correspondence: eva.swinnen@vub.be
†These authors contributed equally to this work.
Abstract: The global aging population faces significant health challenges, including an increasing
vulnerability to disability due to natural aging processes. Wearable lower limb exoskeletons (LLEs)
have emerged as a promising solution to enhance physical function in older individuals. This sys-
tematic review synthesizes the use of LLEs in alignment with the WHO’s healthy aging vision,
examining their impact on intrinsic capacities and functional abilities. We conducted a comprehen-
sive literature search in six databases, yielding 36 relevant articles covering older adults (65+) with
various health conditions, including sarcopenia, stroke, Parkinson’s Disease, osteoarthritis, and
more. The interventions, spanning one to forty sessions, utilized a range of LLE technologies such as
Ekso
®
, HAL
®
, Stride Management Assist
®
, Honda Walking Assist
®
, Lokomat
®
, Walkbot
®
, Healbot
®
,
Keeogo Rehab
®
, EX1
®
, overground wearable exoskeletons, Eksoband
®
, powered ankle–foot orthoses,
HAL
®
lumbar type, Human Body Posturizer
®
, Gait Enhancing and Motivation System
®
, soft robotic
suits, and active pelvis orthoses. The findings revealed substantial positive outcomes across diverse
health conditions. LLE training led to improvements in key performance indicators, such as the
10 Meter Walk Test, Five Times Sit-to-Stand test, Timed Up and Go test, and more. Additionally,
enhancements were observed in gait quality, joint mobility, muscle strength, and balance. These im-
provements were accompanied by reductions in sedentary behavior, pain perception, muscle exertion,
and metabolic cost while walking. While longer intervention durations can aid in the rehabilitation
of intrinsic capacities, even the instantaneous augmentation of functional abilities can be observed in
a single session. In summary, this review demonstrates consistent and significant enhancements in
critical parameters across a broad spectrum of health conditions following LLE interventions in older
adults. These findings underscore the potential of LLE in promoting healthy aging and enhancing
the well-being of older adults.
Keywords: exoskeletons; assistive technology; older adults; healthy aging; intrinsic capacity; func-
tional ability
1. Background
Healthy aging, as defined by the WHO, is the process of developing and maintaining
the functional ability that enables well-being in older age [
1
]. This functional ability is
intertwined with both individual intrinsic capacity (IC) and the surrounding environment
Sensors 2024,24, 2230. https://doi.org/10.3390/s24072230 https://www.mdpi.com/journal/sensors

Sensors 2024,24, 2230 2 of 26
in which individuals reside and interact. IC helps identify the focal points for medical
assessments and treatments, while the domains related to functional ability determine
the measure of its effectiveness. Within the domain of functional ability, the “ability to be
mobile” is identified as a pivotal subdomain. It is a key factor for healthy aging, especially
if we consider that 45% of adults who are more than 75 years old have at least one physical
function difficulty. As physical losses occur with advancing age, older adults may face
challenges in performing functional mobility tasks such as walking, standing up from a
seated position, or climbing stairs [
2
]. When physical limitations hinder mobility, individu-
als are often inclined to avoid physical activity altogether. This behavior increases the risk
of diseases associated with a sedentary lifestyle and can lead to withdrawal from participa-
tion in society, ultimately impacting their independence and quality of life [
3
]. The losses
associated with declines in mobility extend beyond the individual: when older people are
not able to move around, their social networks are affected, and the community may lose
valuable contributions and might need additional resources to support older people in
their daily lives. Facilitating the ability of older people to be able to get around when and
how they choose, at an affordable cost, are important provisions of the United Nations
Convention on the rights of persons with disabilities and optional protocol [
4
]. While tradi-
tional mobility aids (e.g., crutches, canes, and walkers) offer support by unloading joints,
reducing pain, and improving balance, they come with limitations. For instance, they are
often heavy, bulky, cumbersome, and restrict upper limb movement and functionality,
hindering functional tasks requiring manual dexterity such as carrying objects, cooking,
and using one’s hands freely while walking. Additionally, they may not adequately assist
with essential functional activities like sit-to-stand transfers [
3
]. In this context, lower limb
exoskeletons (LLEs) represent a potential game-changer for healthy aging, even if there
is a significant gap in the understanding of the societal and rehabilitation implications of
integrating LLEs into the lives of the older population.
IC, a key component of healthy aging, comprises physical and mental capacities,
categorized into cognitive, psychological, locomotor, sensory, and vitality domains [
5
].
Evaluating IC in each of these areas necessitates rigorous ‘stress tests’ designed to measure
maximum capability; thus, IC needs to be differentiated with ‘performance indicators’
within the same domain. For instance, to quantify endurance as an IC, the utilization of
the 6 min walking test is considered a stress test since one is asked to walk ‘as far as
possible’ in 6 min. Conversely, when executing a 10 meter-walking test at a self-paced
walking speed, the outcomes primarily signify ‘a performance’ rather than an inherent
capacity. Of particular importance for LLE use in older adults is locomotor capacity, encom-
passing musculoskeletal aspects crucial for endurance, balance, muscle strength, function,
power, and joint functionality [
6
]. But ‘vitality capacity’ is also particularly important; for
this capacity, self-perceived fatigue and muscle fatigability have been suggested as top
biomarkers [
7
]. Most studies on LLE use in older adults lack this comprehensive assessment,
highlighting a literature gap.
Aging-related physical changes that are related to the concept of IC often impact
‘functional abilities’ such as gait function and contribute to injuries in adults aged 65 and
older [
8
,
9
]. These changes can lead to reduced physical activity, contributing to disability
and psychosocial issues such as social isolation and depression [
10
,
11
]. Conversely, regular
physical activity is associated with improved physical and cognitive functioning in older
adults, a cornerstone of healthy aging [
12
]. Consequently, LLEs could be beneficial in older
adults to bypass the decreased levels of IC, allowing for augmented functional ability [
2
].
In this context, LLEs have the potential to serve as an assistive technology, supporting older
adults during their daily life activities.
Recent research underscores the significant benefits of exercise programs for older
adults, slowing age-related changes and increasing life expectancy [
13
,
14
]. To address
the challenges of healthy aging, the focus should be on IC rather than specific chronic
diseases [
15
], as preserving physical performance, including muscle strength, power, and

Sensors 2024,24, 2230 3 of 26
endurance, is essential for a healthy and productive life among the aging population and a
key contributor to late-life mobility and independence [16].
In rehabilitation settings, LLEs have demonstrated potential for high-dosage, high-
intensity gait training, complementing conventional exercise programs, while reducing
strain on therapists [
17
]. Therefore, LLE training also has a great potential to rehabilitate IC
and augment functional abilities.
As such, LLEs have shown promise in rehabilitating patients after stroke (e.g., en-
hancing walking speed and balance [
18
]) and spinal cord injury (e.g., enhancing walk-
ing speed, walking endurance, and bone mineral density [
19
]). Additionally, they have
displayed potential in addressing age-related physical changes in functional ability of
community-dwelling older adults (e.g., improved gait kinematics and kinetics, trunk and
lower extremity muscle strength, and metabolic efficiency) [
20
]. Consequently, although
most available exoskeletons are designed to augment human performance in industrial
settings or aid in the rehabilitation of individuals with neurological conditions, their use is
expected to expand beyond these contexts to assist older people in their functional abilities
and to age in place [21].
Beyond the specific setting (e.g., industry, rehabilitation, assistive) in which they can
be employed, LLEs can be categorized in various ways based on their functional attributes
and design features. For instance, passive exoskeletons have no power sources, but rely
on kinematic forces to ensure locomotion (e.g., using springs). Active exoskeletons, on the
other hand, employ power sources to activate actuators (i.e., the devices responsible for
generating motion in a specific part or joint within the exoskeleton). These actuators can
drive a single joint (e.g., the Honda Walking Assist
®
solely assists the hip joints), while
some devices utilize multiple actuators to drive a combination of joints (e.g., the Ekso-GT
®
[Ekso Bionics, San Rafael, CA, USA] assists both hip and knee joints) [
22
]. Furthermore,
stationary exoskeletons (e.g., the treadmill-based Lokomat
®
[Hocoma, Zürich, Switzer-
land] and Walkbot
®
[Walkbot, Seoul, Republic of Korea]) offer a secure environment for
repetitive training but are confined to rehabilitative institutions [
18
]. In contrast, wearable
exoskeletons (e.g., Honda Walking Assist®[Honda, Tokyo, Japan], Ekso-GT®and Hybrid
Assistive Limb
®
[Cyberdyne, Tsukuba, Japan]) overcome this limitation, but may be less
supportive and require a minimum trunk balance. With recent technological advancements,
wearable exoskeletons designed to enhance physical functioning in aging populations [
23
]
are becoming more affordable, lighter, and less constraining [
24
]. As outlined previously,
these devices have demonstrated their potential in patients with neurological conditions,
with robot-assisted rehabilitation programs showing positive impacts on gait-related out-
comes in stroke and spinal cord injury [
18
,
25
,
26
], but also on quality of life and depressive
symptoms in a range of neurological disorders [27].
We postulate that the potential of powered exoskeletons exceeds rehabilitation and
industry settings, as they present a novel approach to enable older adults to engage in
activities with greater ease and confidence. LLEs could be employed to compensate for
diminished IC reserves (e.g., muscle strength, endurance, and movement speed) in healthy
older adults, to aid in functional mobility tasks such as level walking, uphill walking,
climbing stairs, and sit-to-stand transfers. Analogous to long-term effects reported in
neurological populations, these improvements might extend beyond mere wear periods
and can positively impact habitual physical activity and cardiorespiratory function and
facilitate high-intensity activities like hiking and keeping pace with grandchildren.
Despite promising developments in treating neurological conditions, a comprehen-
sive assessment of the effectiveness of robotic LLE-based interventions within a healthy
aging context is lacking [
28
]. Therefore, this systematic review aims to provide an updated
overview of LLE use for augmenting functional abilities as well as their impact on perfor-
mance indicators and stress tests for IC domains in older adults, both with and without
various health conditions.

Sensors 2024,24, 2230 4 of 26
2. Methods
2.1. Study Registration
The protocol for this systematic review was prospectively registered in the Inter-
national Prospective Register of Systematic Reviews (PROSPERO, registration number
[CRD42023434655]) in June 2023 and reported in accordance with the Preferred Reporting
Items for Systematic Review and Meta-Analysis Protocols (PRISMA) 2020 statement [
29
]
[Figure 1] and the Cochrane Handbook for Systematic Reviews of Interventions [30].
Sensors 2024, 24, x FOR PEER REVIEW 4 of 28
2. Methods
2.1. Study Registration
The protocol for this systematic review was prospectively registered in the Interna-
tional Prospective Register of Systematic Reviews (PROSPERO, registration number
[CRD42023434655]) in June 2023 and reported in accordance with the Preferred Reporting
Items for Systematic Review and Meta-Analysis Protocols (PRISMA) 2020 statement [29]
[Figure 1] and the Cochrane Handbook for Systematic Reviews of Interventions [30].
Figure 1. Prisma flow diagram.
2.2. Search Strategy and Study Selection Criteria
In June 2023, a systematic search was conducted across various databases and regis-
tries, including PubMed, EMBASE, Web of Science (WOS), the Cochrane Central Register
of Controlled Trials, CINAHL, PEDro, and IEEE Xplore Digital Library.
The search strategy focused on older individuals, exoskeletons, functional ability, IC,
and performance indicators, using relevant terms related to age groups and interventions.
Specific outcome-related terms were not included due to the expected limited number of
articles meeting inclusion criteria. The full search string is shown in Appendix A.
Figure 1. Prisma flow diagram.

Sensors 2024,24, 2230 5 of 26
2.2. Search Strategy and Study Selection Criteria
In June 2023, a systematic search was conducted across various databases and reg-
istries, including PubMed, EMBASE, Web of Science (WOS), the Cochrane Central Register
of Controlled Trials, CINAHL, PEDro, and IEEE Xplore Digital Library.
The search strategy focused on older individuals, exoskeletons, functional ability, IC,
and performance indicators, using relevant terms related to age groups and interventions.
Specific outcome-related terms were not included due to the expected limited number of
articles meeting inclusion criteria. The full search string is shown in Appendix A.
English-language studies were included if they involved the use of LLEs in human
participants aged
≥
50 years, with a mean age of
≥
65 years. No publication date restrictions
were imposed.
Studies that met one or more of the following criteria were excluded from the review:
not related to wearable LLEs; performed in a different setting than hospitals, universities,
rehabilitation centers, participants’ homes, and care facilities for older adults; evaluating
different outcomes than participants’ intrinsic capacities, functional independence, quality
of life, and functional abilities (i.e., the ability to meet their basic needs, to learn, grow, and
make decisions, to be mobile, to build and maintain relationships, and to contribute to
society); case report studies, research and project reports, annual or activity reports, theses,
conference proceedings, pre-prints, newsletters, technical reports, recommendations and
technical standards, patents, technical notes, presentations, field notes, laboratory research
books, academic courseware, lecture notes, and evaluations.
2.3. Data Extraction and Analysis
First, references obtained from the systematic search were entered and deduplicated
into EndNote X9 (Clarivate Analytics, Philadelphia, PA, USA). Second, titles and ab-
stracts were screened for alignment with inclusion and exclusion criteria, employing
the Rayyan QCRI web application. Selected studies’ full texts were subsequently re-
viewed for final inclusion. This selection process was conducted independently by two re-
searchers, with disagreements resolved through discussion and, if necessary, consultation
with a third reviewer.
Data extraction followed, utilizing a standardized collection form. The extracted infor-
mation encompassed the first author’s name, publication year, study design, sample size,
dropout count, patients’ clinical history, participants’ age and gender, exoskeleton speci-
fications, intervention details, measurement systems employed, and outcomes. In cases
of incomplete or unclear data, the corresponding author was contacted via e-mail for
clarification.
2.4. Data Classification
The extracted results were classified following the WHO vision on healthy aging.
Hence, two main categories were used: IC and functional ability. If studies reported on
the effects of LLE use on functional ability (i.e., mobility, ability to learn, grow and make
decisions, ability to build and maintain social relationships, ability to contribute, and ability
to meet basic needs), they were allocated to the functional ability category. If studies reported
effects of LLE use in light of measurements of IC (i.e., results of stress tests), they were
allocated to the intrinsic capacity category. However, if studies reported on performance rather
than on a specific stress test within the IC domains, they were allocated to a subcategory,
i.e., the intrinsic capacity performance indicator category. This is shown in Figure 2.
2.5. Quality Assessment
The studies’ methodological quality was assessed using the Downs and Black Scale [
31
]
[Appendix B] and the studies were classified as excellent (26–27), good (20–25), fair (15–19),
or poor (≤14).

Sensors 2024,24, 2230 6 of 26
Sensors 2024, 24, x FOR PEER REVIEW 6 of 28
Figure 2. Studies’ classification.
2.5. Quality Assessment
The studies´ methodological quality was assessed using the Downs and Black Scale
[31] [Appendix B] and the studies were classified as excellent (26–27), good (20–25), fair
(15–19), or poor (≤14).
3. Results
The database search retrieved a total of 3622 records. A total of 2917 articles were
screened by title and abstract and 792 full-text articles were assessed for eligibility, of
which 36 articles were included in the qualitative synthesis [Figure 1 and Supplementary
Material (Tables S1–S3)].
Figure 2. Studies’ classification.
3. Results
The database search retrieved a total of 3622 records. A total of 2917 articles were
screened by title and abstract and 792 full-text articles were assessed for eligibility, of which
36 articles were included in the qualitative synthesis [Figure 1and Supplementary Material
(Tables S1–S3)].

Sensors 2024,24, 2230 7 of 26
3.1. Studies’ Methodological Quality
A total of 22.2% of the included studies had a poor quality [
32
–
39
], 55.5% had a fair
quality [24,40–58] and 22.2% had a good quality [59–66] [Appendix B].
3.2. Demographic and Study Characteristics
The included studies consisted of retrospective, prospective interventional, or ob-
servational research conducted in single or multi-center settings [Supplementary Mate-
rials (Tables S1–S3)]. A total of 17 studies investigated functional abilities [
33
,
34
,
36
,
42
,
46
,
47
,
49
,
51
,
53
,
54
,
56
,
57
,
59
–
61
,
63
,
65
] [Table S1 in Supplementary Material], 23 investigated
IC [
32
–
37
,
42
,
45
,
47
,
49
,
50
,
52
,
53
,
55
–
63
,
65
] [Table S2 in Supplementary Material], and 26 in-
vestigated performance indicators [
24
,
33
,
35
–
38
,
40
–
44
,
46
,
50
,
51
,
53
,
54
,
56
–
65
] [Table S3 in
Supplementary Materials] [Figure 2].
These studies covered a range of conditions, with ten investigating LLE use in pa-
tients with stroke [
24
,
45
,
46
,
51
,
54
,
56
,
59
,
63
–
65
], four focusing on patients with Parkinson’s
Disease (PD) [
38
,
58
,
60
,
61
], two examining patient cohorts with various other neurological
conditions [
50
,
52
], three centered on patients with osteoarthritis [
42
,
53
,
57
], and one each
for hip fracture [
34
], sarcopenia [
49
], and depression [
55
]. Additionally, there were 14 stud-
ies involving healthy older adults [
32
,
33
,
35
–
37
,
39
–
41
,
43
,
44
,
47
,
48
,
62
,
66
] [Supplementary
Materials (Tables S1–S3)].
3.3. LLEs to Improve Functional Ability
Seventeen studies assessed functional ability after the use of LLEs [
33
,
34
,
36
,
42
,
46
,
47
,
49
,
51
,
53
,
54
,
56
,
57
,
59
–
61
,
63
,
65
] and their results are detailed in Table S1 in the Supplemen-
tary Materials.
3.3.1. LLEs to Improve Mobility
Mobility measures were considered in 16 studies [
33
,
34
,
36
,
42
,
46
,
47
,
49
,
51
,
53
,
54
,
56
,
57
,
60
–
62
,
65
], of which two focused on patients with PD [
60
,
61
], three on patients with os-
teoarthritis [
42
,
53
,
57
], one on patients with sarcopenia [
49
], one on patients with hip
fracture [
34
], five on patients with stroke [
46
,
51
,
54
,
56
,
65
], and four on healthy older
adults [33,36,47,62] [Table S1 in Supplementary Materials].
In patients with PD, Gryfe et al. (2022) [
60
] investigated the impact of an 8-week
Keeogo Rehab
TM
[B-Temia, Saint-Augustin-de-Desmaures, Quebec, Canada] exoskeleton
intervention, demonstrating improvements in preferred gait speed and on the Freezing
of Gait Questionnaire (FoG-Q) and Unified Parkinson’s Disease Rating Scale (UPDRS)—
motor functioning sub-scale, and Parkinson’s Disease Questionnaire-39 (PDQ-39)—mobility
sub-scale (p= 0.017) post-intervention, compared to the other groups. Kawashima et al.
(2022) [
61
] studied the effects of a 3-month gait training intervention using the Stride
Management Assist
®
(SMA) [Honda, Tokyo, Japan] exoskeleton, but no significant changes
were observed in the FoG-Q and 10 Meter Walk Test (10MWT) within and between groups.
In patients with osteoarthritis, Koseki et al. (2021) [
42
] demonstrated significant im-
provement in Western Ontario and McMaster University Osteoarthritis index (WOMAC)
function scores at 8 weeks post total knee arthroplasty using the Honda Walking Assist
®
(HWA) (ES = 0.21 at T0, ES = 0.18 at T1, ES = 0.54 at T2 and ES = 2.33 at T3), in comparison
with the control group (p< 0.001) [Table S1]. Setoguchi et al. (2022) [
53
] and Yoshikawa
et al. (2018) [
57
] explored Hybrid Assistive Limb
®
(HAL) gait training post hip arthroplasty
and post total knee arthroplasty (TKA), respectively. Setoguchi et al. (2022) [
53
] reported
statistically significant within-group temporal changes in the Harris hip score function
subscore for patients who received the LLE intervention and in the control group (p< 0.05),
in the Harris hip score motion subscore for the control group (p< 0.05), in the 36-Item Short
Form Survey (SF-36) physical functioning subscore for both groups (p< 0.05), and in the
SF-36 role limitations subscore after LLE intervention (p< 0.05). Yoshikawa et al. (2018) [
57
]
showed varying between-group improvements in the WOMAC function score at differ-

Sensors 2024,24, 2230 8 of 26
ent time points, but did not consistently demonstrate statistically significant differences
between groups at the different studied timepoints.
The study on sarcopenia by Norris et al. (2007) [
49
] with powered ankle–foot orthoses
(PAFOs) revealed no statistically significant improvement in preferred walking speed.
Comparisons between walking with standard shoes, inactive PAFOs, and active PAFOs did
not yield statistically significant differences (p= 0.098, p= 0.536, and p= 0.474, respectively).
However, there was a non-significant trend observed between walking with standard shoes
and inactive PAFOs, suggesting a potential impact that did not reach statistical significance
in this small cohort of older adults.
In patients with hip fracture, Fujikawa et al. (2022) [
34
] demonstrated substantial
improvement in functional mobility with HAL
®
rehabilitation, when combined with con-
ventional rehabilitation, as indicated by the significant reduction in Five Times Sit-to-Stand
test (FTSS) scores among patients with hip fracture (p< 0.01; ES = 1.81 (95% CI = 0.93—2.66)).
Five studies on patients with stroke explored various LLE interventions, showcas-
ing improvements in functional outcomes but with mixed results in inter-group differ-
ences
[46,51,54,56,65]
. Taki (2020) [
54
] found an increase in Functional Independence Mea-
sure (FIM) motor subscores (p= 0.013) and in ambulation at a hospital ward on discharge
(p= 0.011) after gait training with HAL
®
for 3 h/day for 7 days/week. Longatelli (2021) [
46
]
showed significant improvement in the Capacity Score of patients with stroke after a 4-week
intervention consisting of 12 assisted rehabilitation sessions and 8 conventional therapy
sessions, as well as in the control group (p< 0.01). Moreover, Watanabe (2017) [
56
] found
statistically significant improvements in the FAC after 12 sessions of HAL over 4 weeks
(p= 0.026), while the differences in the LE Fugl-Meyer Assessment were not statistically
significant (p= 0.131).
Park et al. (2021) [
51
] conducted a study in which participants performed interlimb
coordinated humanoid robotic sessions with a VR/AR game, along with conventional
physical therapy 7 days/week, for 2 weeks. They found a within-group improvement
on the Fugl-Meyer Assessment Lower Extremity (FMA-LE) synergy scale flexor synergy
test (p= 0.000) and in FMA-LE synergy scale total synergy (p= 0.007) over time. Yeung
(2021) [
65
] also reported differences between groups in the Functional Ambulation Category
(FAC) over time after 30 min/weekday interventions with the Power-Assisted Ankle
Robot and Swing-Controlled Ankle Robot, together with a conventional training routine
(2 h/weekday) (increase of 1.4 [1.0, 1.9] (p< 0.001) in G1, increase of 1.4 [0.9, 2.0] (p< 0.001)
in G2, and increase of 0.9 [0.4, 1.3] (p< 0.01) in G3).
Mobility in healthy older adults was studied in four studies [
33
,
36
,
47
,
62
]. First, Ja-
yaraman et al. (2022) [
36
] conducted a study in which participants performed twelve
gait training sessions over 4–6 weeks; these authors found statistically significant im-
provements in participants’ scores on the Functional Gait Assessment (FGA) (p< 0.001),
number of sedentary bouts (>3 min) per day (p= 0.004), time spent in the sedentary bouts
(p= 0.003), and 5xSTS (p< 0.001) post-intervention. Fang (2022) [
33
] evaluated two proto-
cols involving assistance and resistance modes of an ankle exoskeleton and reported an
improvement in participants’ self-selected walking speed (1.07 vs 1.12) from T0 to T1 and
in fast walking speed (1.38 vs 1.59) over time. Martini (2019) [
47
] administered a four-week
robot-assisted gait training regimen in an active pelvis orthosis (APO) group, measuring
differences in daily steps at baseline, for which they found no significant changes [Table S1].
Finally, Lee (2022) [
62
] investigated the effects of a four-week exoskeleton exercise pro-
gram. They found significant improvements in participants’ scores on the Short Physical
Performance Battery (SPPB) over time (p< 0.01) [Table S1 in the Supplementary Material].
3.3.2. LLEs to Improve Older Persons’ Ability to Build/Maintain Social Relations
One study assessed older persons’ ability to build and/or maintain social relations in
patients with osteoarthritis after a 6-week HAL
®
gait training intervention, which showed
no significant impact on social functioning [
53
] [Table S1 in the Supplementary Materials].

Sensors 2024,24, 2230 9 of 26
3.3.3. LLEs to Improve the Ability to Meet Basic Needs
For patients with stroke, three studies investigated the effects of LLE training on their
ability to meet basic needs [
54
,
59
,
63
]. First, Rojek (2020) [
63
] investigated the effects of a
four-week Ekso GT®intervention, associated with occupational therapy and individually
tailored physical therapy, and reported statistically significant improvements over time in
the Barthel Index (p= 0.01) and Rivermead Mobility Index [Table S1]. Calabrò(2018) [
59
]
investigated the effects of an 8-week Ekso™ training intervention, associated with tran-
scranial magnetic stimulation (TMS), and also reported significant improvements in the
Rivermead Mobility Index after LLE training (ES = 0.6, p= 0.03). Finally, Taki (2020) [
54
]
investigated the effects of knee–ankle–foot orthoses (KAFO), ankle–foot orthoses (AFO),
and HAL
®
3 h per day, 7 days per week, and reported statistically significant improvements
over time in FIM total sores after gait training with HAL®(p= 0.024).
In the case of patients with PD [
60
], an intervention using the Keeogo Rehab
TM
powered knee assistance exoskeleton for eight weeks did not yield significant differences
in ADL subscores [Table S1 in the Supplementary Materials].
3.4. LLEs to Enhance IC
The impact of LLEs to enhance IC has been investigated in 15 studies [
32
–
34
,
36
,
42
,
50
,
56–63,65] and their results are detailed in Table S2 in the Supplementary Materials.
3.4.1. LLEs to Enhance Locomotor Capacity
First, locomotor capacity was taken into account in fifteen studies [
32
–
34
,
36
,
42
,
50
,
56
–
63
,
65
], of which three focused on patients with PD [
58
,
60
,
61
], one on other neurological
disorders [
50
], two in patients with osteoarthritis [
42
,
57
], four on patients with stroke [
56
,
59
,
63
,
65
], one on patients with hip fracture [
34
], and four on healthy older adults [
32
,
33
,
36
,
62
]
[Table S2 in Supplementary Materials].
In patients with PD, Yun et al. (2019) [
58
] investigated the effects of a 4-week inter-
vention using the Walkbot
®
, and reported significant improvements in the Berg Balance
Scale (BBS) scores, which were observed immediately after treatment (p= 0.004) and at the
one-month follow-up (p= 0.024). Gryfe et al. (2022) [
60
] conducted an 8-week exoskeleton
exercise intervention, and showcased notable increases in the 6 Minute Walk Test (6MWT)
for the exoskeleton group, compared to the others (p< 0.001). The study by Kawashima
et al. (2022) [
61
] investigated the effects of 10 gait training sessions with the SMA
®
exoskele-
ton for 3 months, and revealed positive outcomes, emphasizing statistically significant
improvements in the 3 Minute Walk Test (3MWT) post-intervention (p= 0.023). However,
they did not find statistically significant differences for the BBS and Functional Reach Test
(FRT) in any of the groups [Table S2].
Panizzolo et al. (2022) [
50
] extended the exploration to patients with other neurological
disorders beyond PD, utilizing the Exoband
®
[Moveowalks, Padua, Italy] in a ten-session
walking program over five weeks. Participants wore the Exoband while walking for
10 min back and forth along a 60 m corridor. They were instructed to attempt to walk
as far as possible (i.e., cover the longest possible walking distance) and were able to stop
and rest during the walking session as needed. They reported a significant increase in
the longest walking distance while wearing the Exoband
®
(p< 0.05) and a statistically
significant correlation between sessions spent walking with the Exoband
®
and meters
covered (r = 0.9126; p< 0.01).
For patients with osteoarthritis, Koseki et al. (2021) [
42
] administered 17–20 gait train-
ing sessions using the HWA
®
, from week 1 to 5 post-TKA and found statistically significant
differences in the maximal passive and active extension of the knee between groups only
at baseline (p= 0.027, ES = 1.02 and p= 0.031, ES = 0.99, respectively), but not at the other
timepoints. At one week post-TKA, there was a significant improvement in the maximum
walking speed (ES = 1.04), and at 1 and 2 weeks post-TKA, there was a significant improve-
ment in step length at maximum walking speed (ES = 1.02 and ES = 0.87, respectively).
Moreover, Yoshikawa et al. (2018) [
57
] administered 10–12 sessions of 15 min each with

Sensors 2024,24, 2230 10 of 26
HAL
®
, over 4 weeks, along with conventional physical therapy. They found that passive
knee extension ROM differences among groups were statistically significant at 2 weeks
following TKA (p= 0.034) and 4 weeks post-TKA (p= 0.006), and that there was a significant
between-group difference in maximal walking speed 4 and 5 weeks post-TKA (p= 0.006
and p= 0.027, respectively), and in step length at maximum walking speed in weeks 2, 4,
and 5 post-TKA (p= 0.016, p= 0.001 and p= 0.003, respectively). However, the differences
in active knee extension ROM among groups were only statistically significant at weeks 2
and 3 post-TKA (p= 0.005 and p= 0.048).
The locomotor capacity of patients with stroke was explored in studies by Calabrò
et al. (2018) [
59
], Rojek et al. (2020) [
65
], Watanabe et al. (2017) [
59
,
63
], and Yeung
et al. (2021) [
65
], which used the Ekso
®
[
65
], HAL
®
[
56
], and various robotic-assisted
trainings [
65
]. Calabròet al. (2018) [
59
] reported significant improvements in exoskeleton-
assisted gait training (EGT)-induced functional outcomes, measured via the Timed Up
and Go (TUG) test at 8 weeks post-gait training (ES = 0.5, p< 0.02), while Rojek et al.
(2020) [
63
] reported significant changes in both balance and functional status after EGT,
together with a slight and insignificant trend towards reducing the total load distribution
on the feet, particularly on the uninvolved limb. Similarly, Yeung et al. (2021) [
65
], found
significant improvements in BBS scores post-intervention in the whole group (increase of
18.8 [13.1, 24.4] (p< 0.001) in G1, increase of 12.6 [6.2, 18.9] (p< 0.01) in G2, and increase of
14.4 [9.4, 19.3] (p< 0.001) in G3).
Watanabe et al. (2017) [
56
], however, did not find statistically significant differences
among groups in maximal walking speed (p= 0.975), 6MWT (p= 0.810), and TUG (p= 0.413)
after 12 HAL sessions.
Fujikawa et al. (2022) [
34
] conducted the only study investigating locomotor capacity
in patients with hip fracture, implementing conventional rehabilitation alongside HAL
®
rehabilitation and reporting a reduction in TUG over time in all participants.
Four studies investigated locomotor capacity in healthy older adults [
32
,
33
,
36
,
62
].
First, Jayaraman et al. (2022) [
36
] conducted a study in which participants used the Gait
Enhancing and Motivating System (GEMS-H) during 12 gait training sessions (30 min
each) over a period of 4–6 weeks and saw an improvements in the 10MWT (p= 0.001),
6MWT (p< 0.001) and BBS scores (p< 0.001). Aprigliano et al. (2019) [
32
] administered one
session and 14 experimental trials with the APO. They reported that the assistive approach
effectively enhanced balance recovery in the sagittal plane for both perturbation paradigms.
However, it did not demonstrate effectiveness in maintaining stability in the frontal plane.
Fang et al. (2022) [
33
] conducted a study in which participants used a dual-mode ankle
exoskeleton for ankle assistance and resistance; these authors found that the resistance
protocol produced a 35% increase in 6MWT distance (m), an increase of 18% (right side) and
43% (left side) in plantar flexor strength, and an increase in fast walking speed (m/s) over
time. Lee et al. (2022) [
62
] administered a 4-week intervention using EX1
®
and reported
significant improvements in BBS scores (p< 0.01), TUG (p< 0.01), and FRT (p< 0.01) for all
the groups, with associated changes in muscle strength.
3.4.2. LLEs to Enhance Vitality Capacity
In addition to examining locomotor capacity, several studies delved into vitality
capacity across diverse patient groups [Table S2 in Supplementary Material].
Kawashima et al. (2022) [
61
] examined vitality in patients with PD, using the SMA
®
exoskeleton for 3 months, revealing a significant reduction in energy expenditure as mea-
sured by the Physiological Cost Index (PCI) during the 3MWT after the SMA intervention
(p= 0.046).
In patients with osteoarthritis, Koseki et al. (2021) [
42
] conducted 17–20 gait training
sessions with the HWA
®
, reporting no significant changes in knee extension and flexion
torque 1 to 5 weeks after TKA. Conversely, Yoshikawa et al. (2018) [
57
], using HAL
®
,
observed significant changes in knee extension torque 5 weeks post-intervention (p= 0.014).

Sensors 2024,24, 2230 11 of 26
Lefeber et al. (2018) [
45
] focused on the immediate effect of walking with a Lokomat
®
across three conditions in patients with stroke, and reported significant differences in the
net oxygen consumption (p= 0.037), net respiratory exchange ratio (p= 0.047), and the net
oxygen cost (p= 0.037) after 6 min of walking based on the implemented level of assistance.
3.4.3. LLEs to Enhance Psychological Capacity
Psychological capacity was assessed in studies involving patients with PD [
58
,
60
],
osteoarthritis [
53
], depression [
55
], sarcopenia [
49
], and healthy older adults [
62
] [Table S2
in Supplementary Materials].
Two studies in patients with PD investigated the effects of LLEs on psychological
capacity [
58
]. Yun et al. (2019) [
58
] administered a 4-week intervention using the Walkbot
®
and did not find significant changes in the Korean version of the Falls Efficacy Scale (KFES).
Gryfe et al. (2022) [
60
] administered an 8-week intervention with the Keeogo Rehab
TM
and
did not find significant decreases in the Activities-Specific Balance Confidence (ABC) test
for any group at any timepoint.
Verrusio et al. (2018) [
55
] investigated psychological capacity in patients with depres-
sion using the Human Body Posturizer
®
(HBP) [Posturizer, Italy] (3 sessions/week for
6 months) and observed a reduction in the Geriatric Depression Scale (p< 0.01).
Norris et al. (2006) [
49
] explored vitality capacity in patients with sarcopenia, uti-
lizing the EXO3
®
exoskeleton. Significant improvements in the SF-36 vitality sub-scale
were reported.
Lee (2022) [
62
] investigated psychological capacity in healthy older adults, using the
EX1
®
for 4 weeks, reporting improvements in the Geriatric Depression Scale Short Form
(GDS-SF) (p< 0.05).
3.4.4. LLEs to Enhance Cognitive Capacity
In patients with PD, Yun et al. (2019) [
58
] assessed cognitive capacity with Walkbot
®
interventions over 4 weeks. They reported a tendency towards an increase in dual-task
interference in gait velocity, although differences were not statistically significant, but not
in dual-task physical aspects [Table S2].
3.5. LLEs to Improve Performance Indicators
The impact of lower limb exoskeletons (LLEs) on various performance indicators has
been the subject of extensive research, with multiple studies shedding light on their effects
in different populations [
24
,
38
,
42
,
46
,
51
,
53
,
54
,
56
–
59
,
61
,
63
–
65
] [Table S3 in Supplementary
Materials].
3.5.1. LLEs to Improve Locomotor Performance Indicators
Fifteen studies investigated locomotor performance indicators [
24
,
38
,
42
,
46
,
51
,
53
,
54
,
56
–
59,61,63–65] [Table S3 in Supplementary Material].
In the domain of locomotor performance for patients with PD, significant findings
emerged. Romanato et al. (2022) [
38
] observed notable improvements in muscle forces
during various walking phases after 4 weeks of EksoGT
®
training (p< 0.05). Yun et al.
(2020) [
58
], using the Walkbot
®
for 4 weeks of gait training, reported a significant increase
in the 10MWT comfortable gait speed over time (p= 0.041). Kawashima et al. (2022) [
61
],
in a 3-month randomized controlled trial (RCT) with the SMA
®
exoskeleton, reported
significant enhancements in walking speed, step length, and ranges of motion.
Studies focusing on interventions for patients with osteoarthritis also showed positive
outcomes [
42
,
53
,
57
]. Setoguchi et al. (2022) [
53
] conducted a study in which participants
used the HAL
®
; these authors reported significant improvements in hip extension and
range of motion over time (p< 0.05). Koseki et al. (2021) [
42
] conducted a study in which
participants performed 17–20 sessions using the HWA
®
after TKA, and demonstrated
notable changes in self-selected walking speed (p= 0.022) and step length (p= 0.032), and
Yoshikawa et al. (2018) [
57
] also found significant differences in self-selected walking speed

Sensors 2024,24, 2230 12 of 26
(p= 0.022-p= 0.030) and step length (p= 0.002-p= 0.011) after participants performed 10–12
HAL®training sessions.
Nine studies investigated locomotor performance indicators in patients with stroke [
24
,
46
,
51
,
54
,
56
,
59
,
63
–
65
]. Calabròet al. (2018) [
59
] conducted a study in which participants
used Ekso
®
for 8 weeks, and reported significant improvements in walking speed during
the 10MWT (ES = 0.9, p< 0.001), hip and knee muscle activation (ES = 0.8, p= 0.001), gait
quality index (ES = 0.9, p< 0.001), step cadence (ES = 0.9, p< 0.001), and stance/swing
ratio in the affected limb (ES = 0.8, p= 0.008). Also, they reported EGT-induced reductions
in gait cycle duration in the affected and unaffected limb, and stance/swing ratio in the
unaffected limb (ES = 0.9, p< 0.001).
Longatelli et al. (2021) [
46
] found selective improvements in patients’ muscular
activation strategies, especially in the semitendinosus muscle, and Rojek et al. (2020) [
63
]
observed that patients increased their walking time and steps during LLE gait therapy;
both studies administered a 4-week Ekso
®
intervention. Taki et al. (2020) [
53
] conducted a
study in which participants used HAL
®
3 h/day for 7 days/week and found no significant
differences in Br-stage between groups. Watanabe et al. (2017) [
56
] investigated the effects
of 12 sessions with HAL
®
and they reported no statistically significant differences after
training. Also, Park et al. (2021) [
50
] reported significant improvements in the hip and knee
angles and active forces of patients after a 2-week intervention using the Walkbot®.
Firouzi et al. (2022) [
24
] conducted a study assessing the immediate effects of the
HWA
®
during three walking conditions: normal walking at a self-selected comfortable
speed (I1), unassisted walking with HWA
®
(I2), and optimal assisted walking with HWA
®
(I3). Each condition involved walking three times on a 5 m walkway, totaling 40–60 min.
Comparisons between conditions revealed that walking speed increased in most patients in
I1 vs I3, showed mixed results in I1 vs I2, and uniformly increased in I2 vs I3. Stride lengths
and velocities generally increased across interventions for both paretic and non-paretic
limbs. The paretic swing phase increased in I1 vs I3, while the non-paretic swing phase
increased in I2 vs I3. Paretic and non-paretic stance phases and double support phases
exhibited mixed results across interventions. Moreover, percentage changes in various
parameters demonstrated individual-specific improvements or declines.
Son et al. (2021) [
64
], administering 10 Healbot T sessions, reported significant in-
creases in self-selected speed for both pelvis-off and pelvis-on groups, and notable im-
provements in muscle activity, stride length, cadence, and walking speed. Finally, Yeung
et al. (2021) [
65
] conducted a study in which participants utilized the Power-Assisted
Ankle Robot and Swing-Controlled Ankle Robot for 30 min/weekday, together with a
conventional training routine (2 h/weekday), and reported significant increases in self-
selected walking speed (10 MWT) over time for all groups (increase of 0.32 [0.18, 0.46]
(p< 0.001) at G1, increase of 0.17 [0.09, 0.25] (p< 0.01) at G2, and increase of 0.17 [0.06, 0.29]
(p< 0.01) at G3).
Eight studies investigated locomotor performance indicators in healthy older adults [
33
,
35–37,41,44,62].
Jayaraman et al. (2022) [
36
] conducted a study in which participants used GEMS-H
®
in twelve gait training sessions (30 min each) over a period of 4–6 weeks; these authors
reported an improvement in 10MWT self-selected gait speed (p= 0.001).
Lee et al. (2017) [
43
] conducted a study in which participants performed one session
with the GEMS
®
at their own comfortable speed; these authors found statistically significant
increases in gait speed, cadence, stride length, step width, and single support time with
robot assistance (p< 0.01), and they also found a reduction in muscle activation due to
hip assistance. Lee et al. (2017) [
44
], using the same protocol, reported an increase in
gait speed, stride length, cadence, and single support time with robot assistance, together
with reduced muscle activity in the rectus femoris and medial gastrocnemius throughout
the terminal stance phase, and the medial gastrocnemius throughout pre-swing phase.
This was associated with an increase in the maximum force and peak pressure.

Sensors 2024,24, 2230 13 of 26
Jin et al. (2017) [
37
] conducted a study in which participants used a soft wearable
robotic suit built in-house for 2 days (four 6-min treadmill trials, at fixed preferred walking
speed)); these authors reported an increase in maximum hip angle (p< 0.05), maximum
vertical displacement of center of mass (COM) (p< 0.01), maximum vertical position of
knee (p< 0.001), maximum vertical position of ankle (p= NS), maximum vertical position
of toe (p< 0.001), stride duration (p< 0.001), and walk ratio (p< 0.001) with the robotic suit
powered on. Additionally, Jin et al. (2019) [
41
] conducted a study in which participants
performed a 6-week training intervention using a soft robotic suit; these authors reported
an average increase in the maximum hip angle, maximum knee angle, maximum ankle
angle, and walk ratio (p= 0.0052) with the device powered on. When the device was
powered off, there was a reduction of the maximum hip flexion (p= 0.0411), maximum
knee flexion (p= 0.0350), and maximum ankle dorsiflexion (p= 0.0085).
Fang et al. (2022) [
33
] conducted a study in which participants used a dual-mode
ankle exoskeleton during two visits; using soleus imaging electromyography (iEMG), these
authors observed changes in the minimum soleus variance ratio and stance phase with the
ankle resistance protocol. Galle et al. (2022) [
35
] conducted a study in which participants
performed four trials using bilateral ankle–foot exoskeletons at a fixed speed; these authors
reported statistically significant changes in step length and walking conditions with the
exoskeleton. Finally, Lee et al. (2022) [
62
] conducted a study in which participants used
the EX1
®
for 4 weeks; these authors reported significant changes in 10MWT self-selected
velocity over time.
3.5.2. LLEs to Improve Vitality Performance Indicators
Panizzolo et al. (2022) [
50
] investigated vitality performance indicators in other
neurological conditions beyond PD. They conducted a study in which participants used
the Exoband
®
in 10 walking sessions (10-min each) for 5 consecutive weeks, and reported
a rate of perceived exertion (RPE) difference over the 10 sessions (<0.05), and a negative
correlation between the sessions spent walking with Exoband®and RPE (p< 0.01).
Setoguchi et al. (2022) [
53
] investigated vitality performance indicators in a study in
which particpants performed three HAL sessions/week for 6 weeks, but these authors only
reported statistically significant within-group changes for the SF-36 vitality category in the
control group (p< 0.05).
Norris et al. (2006) [
49
] conducted a study in which patients with sarcopenia used
a PAFO for ankle plantarflexion assistance; these authors observed a lower metabolic
cost of transport and metabolic energy per stride at preferred walking speeds with the
PAFOs active.
In healthy older adults, vitality performance indicators were assessed in six
studies [33,35,37,47,62].
First, Jin et al. (2017) [
37
] conducted a study in which participants performed four
6-min treadmill walking trials at a preferred walking speed using a soft wearable robotic
suit built in-house; these authors performed two sets of measurements in 2 days, and
reported a lower energy expenditure and energy efficiency (p< 0.05) with the robotic suit
powered on. Martini et al. (2019) [
47
] conducted a study in which participants used the
APO at a fixed walking speed for 4 weeks; these authors observed significant reductions in
the metabolic cost of transport (p< 0.01), and increases in oxygen uptake rate (p= 0.0024)
and metabolic power (p= 0.011) post-intervention.
Fang et al. (2022) [
33
] conducted a study in which participants used a dual-mode
ankle exoskeleton at a fixed walking speed and observed that four out of five participants
experienced a reduction (up to 19%) in metabolic power during assisted walking compared
to their baseline, especially in participants with higher baseline metabolic power, while the
metabolic power decreased 9% over time using the resistance training protocol.
Galle et al. (2022) [
35
] conducted a study in which participants used bilateral ankle–
foot exoskeletons in four walking trials at a fixed walking speed; these authors reported
statistically significant changes in the net metabolic power (I1 > I4 (p= 0.05, ES = 0.91)) and

Sensors 2024,24, 2230 14 of 26
an increased perceived fatigue in the legs given by the visual analogue scale (VAS) over
time (p= 0.05). Lee et al. (2022) [
62
] administered a 4-week intervention using the EX1
®
and
reported reductions in the participants’ net cardiopulmonary metabolic cost (p< 0.05) post-
intervention. Finally, Lee et al. (2017) [
43
] investigated the immediate effect of using the
GEMS
®
at the participants’ own comfortable speed across three conditions. They reported
an oxygen consumption per unit mass that was about 7% lower while using full robot
assistance vs no robot assistance at self-selected speeds (p< 0.05); the EEm (kcal/min) at the
participants’ own comfortable speed was 6.6% lower while using full robot assistance vs
no robot assistance at self-selected speeds (p< 0.05) [Table S3 in Supplementary Materials].
3.5.3. LLEs to Improve Psychological Performance Indicators
The effects of LLEs on psychological performance indicators were also explored
in specific populations. Gryfe et al. (2022) [
60
] investigated them in patients with PD,
using Keeogo Rehab
TM
for 8 weeks, and revealed a significant decrease in the PDQ-39
emotional well-being sub-scale, but did not find statistically significant changes over
time in the Hospital Anxiety and Depression Scale (HADS) anxiety score, the HADS
depression score, the UPDRS mentation sub-scale, and the PDQ-39 stigma sub-scale. Carral
et al. (2022) [
40
], however, delved into the psychological aspects in healthy older adults
using AUTONOMYO
®
[Autonomyo, Switzerland], highlighting participants’ perceptions
of enhanced autonomy while also perceiving that their usage would alleviate the sense
of burden they might impose on their support network. However, there was a degree of
ambivalence among participants, influenced by their personal experiences of the aging process
and their perceptions of human–machine interactions [Table S3 in Supplementary Materials].
Roggeman et al. (2022) [
52
] conducted a study in which older patients used the HWA
®
while performing 30 min of walking; these authors reported the following median scores
using the Intrinsic Motivation Inventory (IMI): 43 for interest/enjoyment, 36 for perceived
competence, 19 for effort/importance, 6 for pressure/tension, 45 for value/usefulness, and
32 for relatedness.
Setoguchi et al. (2022) [
53
] assessed psychological indicators in patients with os-
teoarthritis using the HAL
®
during three sessions per week for 6 weeks in total, reporting
significant improvements in the SF-36 role emotional scores (p< 0.05), but not statistically
significant changes for the SF-36 mental health scores.
3.5.4. LLEs to Improve Cognitive Performance Indicators
Gryfe et al. (2022) [
60
] conducted a study in which patients with PD used Keeogo
Rehab
TM
for 8 weeks; these authors reported significant increases in the SCales for Out-
comes in PArkinson’s disease-COGnition (SCOPA-COG) (p= 0.003) and memory and
learning (p= 0.001) over time. Moreover, Taki (2020) [
54
] conducted a study in which
participants used the HAL
®
3 h/day for 7 days/week; these authors found statistically
significant differences over time in the FIM cognitive subscore (p= 0.008) [Table S3 in
Supplementary Materials].
3.5.5. Sensory Performance Indicators (“Symptom-Based”)
Gryfe et al. (2022) [
60
] investigated sensory performance indicators (“symptom-
based”) in patients with PD. These researchers administered the Keeogo Rehab
TM
for
8 weeks of aerobic strength and functional mobility exercises and reported no statisti-
cally significant changes in the PDQ-39 communication sub-scale and the PDQ-39 bodily
discomfort sub-scale over time.
Three studies investigated sensory performance indicators (“symptom-based”) in
patients with osteoarthritis [
42
,
53
,
57
]. First, Koseki et al. (2021) [
42
] conducted a study in
which participants performed 17–20 sessions using the HWA
®
, but these authors did not
find any statistically significant changes in the WOMAC-p for any group at any timepoint.
Setoguchi et al. (2022) [
53
] conducted a study in which participants used the HAL
®
for 6 weeks; these authors found within-group changes in the Harris hip score pain sub-

Sensors 2024,24, 2230 15 of 26
score and in the Harris hip score bodily pain subscore (p< 0.05). Finally, Yoshikawa et al.
(2018) [
57
] also conducted a study in which participants used the HAL
®
for 4 weeks; these
authors found statistically significant changes in the WOMAC-p (p= 0.021) over time
[Table S3 in Supplementary Materials].
4. Discussion
In this systematic review, we aimed to shed light on the potential of LLEs in the
context of healthy aging. We therefore comprehensively assessed the impact of LLEs
on the intrinsic capacities, functional abilities, and physical performance of older adults,
in alignment with the WHO’s healthy aging vision. We reviewed 36 studies, encompassing
healthy individuals and individuals with a spectrum of health conditions, including stroke,
PD, osteoarthritis, hip fracture, sarcopenia, and depression.
Our findings collectively reveal consistently positive outcomes in various intrinsic
capacities crucial to healthy aging, manifesting after as few as one to forty LLE sessions.
Across all health conditions and in healthy individuals, LLE interventions showed notable
improvements on stress tests, measuring maximum capability across the subdomains of
locomotor capacity, vitality capacity, and psychological capacity. Specifically, maximum
walking speed, step length at maximum walking speed, and maximum walking distance
were increased following LLE training. Apart from these locomotor improvements, signifi-
cant enhancements in energy expenditure, perceived vitality, and depressive symptoms
were also reported.
Furthermore, significant improvements were observed in a wide range of performance
indicators spanning the same subdomains. Locomotor enhancements encompassed im-
proved muscle strength, muscle activation, joint angles, stance and swing phases, single
and double support phases, stride and step lengths, number of steps, cadence, and self-
paced gait speed. Regarding the other subdomains, significant improvements in energy
expenditure, energy efficiency, muscle exertion, pain perception, and perceived vitality
were also reported. Interestingly, while psychological performance indicators revealed
modest or no changes in most populations, improvements in depressive symptoms were
reported in those with depression and in healthy older adults.
The positive effects of LLEs extend beyond IC, with notable improvements in func-
tional abilities, encompassing mobility and the ability to meet basic needs. For instance,
individuals with PD showed improvements in their preferred gait speed, overall motor
functioning, mobility, and severity of freezing of gait, suggesting a potential impact on
slowing PD progression and enhancing functional mobility. Similarly, patients post-TKA,
hip fracture, and stroke also showed improvements in various functional mobility measures.
Finally, although positive effects of LLEs on functional outcomes are expected in individu-
als with impairments due to physical limitations, a key takeaway from this review is that
healthy older adults also experienced positive effects of LLE training on their intrinsic ca-
pacities and functional abilities, including lower extremity function, physical performance,
walking speed, overall sedentary behavior, and functional mobility. These findings align
with previous work by Federici and colleagues (2015) [
67
], reinforcing the positive effects
of LLEs in patients with neurological conditions, and extending their potential benefits
for healthy older adults. Although the majority of existing exoskeletons are designed to
assist in the rehabilitation of neurological populations or to augment human performance
in industrial environments, their implications exceed physical functioning. Hence, LLEs
should be considered as pivotal instruments in helping older adults maintain functional
abilities, promoting independent living, and engaging in active lives as they age. Therefore,
future research should also focus on evaluating LLEs in realistic home environments by
implementing relevant activities of daily living to gain a deeper understanding of their
potential benefits beyond a standard lab setting.
When considering the potential of LLEs for assessing, monitoring, and promoting
older individuals’ health, it is imperative to contextualize these findings within the broader
framework of IC, which has been introduced and discussed as a marker of healthy aging

Sensors 2024,24, 2230 16 of 26
by the WHO [
68
,
69
]. Unlike the traditional disease-centered paradigm, which often fails
to adequately address the complex and heterogeneous needs of older individuals, IC of-
fers a comprehensive perspective, potentially aiding health monitoring through various
technologies [
70
], where LLEs could play a significant role. While IC gains traction as
a standard for measuring and monitoring older adults’ health [
70
,
71
], its integration in
research and clinical practice remains limited. For instance, although most studies included
in this review postdated the WHO’s introduction of the concept of IC as part of its World
Report on Aging and Health [
1
], there remains considerable heterogeneity in the outcome
measures employed. This variability can be attributed to the lack of standardized criteria
for assessing IC subdimensions, as well as for quantifying IC as a global measure [
69
].
Although there is no consensus on assessing IC dimensions or establishing a global IC score,
the WHO has already provided outcome measure recommendations for each dimension,
as reviewed by Lopez-Ortiz et al. (2022) [69].
In light of these considerations, future research endeavors should adhere to the WHO’s
recommendations for assessing IC subdimensions and prioritize interventions and inno-
vative concepts to optimize IC in older adults. Modern technologies, like LLEs, have
the potential to target specific IC subdimensions, and enable the development of inter-
ventions that could effectively maintain functional abilities or reverse functional loss
throughout life [72].
In the context of neurological rehabilitation, LLE training benefits are founded on the
assumption that task repetition would enhance motor learning and increase functional
recovery [
73
]. Although improvements in gait parameters have been documented during
a single LLE session post-stroke [
24
], assumedly, more than one session would be neces-
sary to observe the substantial, lasting benefits of task repetition. Outside the context of
neurological rehabilitation, however, this principle may be less applicable. For instance,
analogous to how some exoskeletons can instantaneously, but artificially augment the
user’s capabilities to levels exceeding that of normal human performance (e.g., allowing
the user to carry heavy loads with minimal effort [
74
]), LLEs could be employed in healthy
older adults to augment muscle strength and endurance, increase movement speed, and
improve gait patterns (e.g., by increasing stride lengths and reducing asymmetries) [
2
].
Such improvements are not only confined to periods when users wear the exoskeletons
but may extend to high-intensity activities, such as hiking or keeping pace with grand-
children [
33
]. Consequently, LLEs may offer an effective solution to address aging-related
challenges by increasing habitual levels of physical activity and securing independent daily
life in the long-term [2,75].
Nonetheless, it remains challenging to pinpoint the threshold of fatigue, effort, pain,
or fall risk that would prompt individuals to opt for exoskeleton-assisted mobility in their
daily lives [
2
]. A crucial aspect to consider in this regard is the adherence to user-centric
design principles, encompassing factors such as the user-friendliness, safety, and comfort
of LLEs [
3
]. While many prototypes and even some commercial LLEs currently fall short
in these areas, the field and technologies are rapidly evolving, paving the way for more
widespread use in the foreseeable future.
Apart from current limitations related to the design of LLEs, this review exposes
several other gaps in the current understanding of LLE technologies within the context of
healthy aging. For one, further research is urgently required to elucidate the relationship
between the mechanisms of action of diverse LLEs and the targeted outcomes. Additionally,
identifying the most effective LLE type, dosage, and intervention protocols for older adults
of various health statuses is essential. Our review encompassed multiple studies utilizing a
wide range of LLEs, different intervention durations, and a variety of outcome measures.
For instance, various studies reported improvements in a wide range of parameters af-
ter interventions lasting two [
51
], four [
37
,
46
,
47
,
56
–
58
,
62
–
64
], five [
42
,
50
], six [
36
,
41
,
53
],
and eight weeks [
59
,
60
]. Interestingly, Galle et al. (2022) [
35
], Lefeber et al. (2018) [
45
],
Firouzi et al. (2022) [24]
, and Lee et al. (2017) [
43
] reported improvements even after a
single session. Conversely, Kawashima et al. (2022) [
61
] did not observe improvements

Sensors 2024,24, 2230 17 of 26
after a three-month intervention and Watanabe et al. (2017) [
56
] did not find a significant
impact on maximal walking speed and functional mobility after 12 sessions.
Evidently, the included studies exhibited high heterogeneity in the health conditions
studied, reported outcome measures, and intervention durations, ranging from one to
40 sessions. While this heterogeneity poses challenges in interpreting and synthesizing the
reported findings, this review employed the WHO’s framework for the measurement of
healthy aging as a guiding structure to descriptively categorize the included studies [
28
].
To this end, it is critical to distinguish between the rehabilitation of IC (e.g., muscle strength,
balance, endurance) and the augmentation of functional abilities (e.g., walking, stairclimbing,
sit-to-stand transfers), when considering LLE interventions.
Similar to conventional rehabilitation strategies, the LLE-focused rehabilitation of IC
takes time. Extensive training programs are necessary to induce long-lasting, clinically rele-
vant improvements in IC and functional abilities. This is supported by the observed trend
indicating that longer intervention durations tend to yield more significant improvements
across various performance indicators, including locomotor function, vitality, psychological
well-being, cognitive capacity, and sensory symptoms. Studies with longer intervention
periods, particularly those exceeding four weeks and involving three sessions per week,
consistently report more substantial enhancements in functional mobility, gait parameters,
psychological measures, and cognitive function.
On the other hand, immediate effects can be expected regarding the augmentation
of functional abilities through LLEs. Analogous to how impaired vision can be instanta-
neously augmented—but not intrinsically enhanced—by corrective glasses, LLEs have
the potential to augment older individuals’ functional abilities. This is further evidenced
by the immediate benefits observed in specifically targeted parameters in single-session
trials [24,43,45].
There is currently no global consensus on measurement methods for IC and functional
ability, hindering widespread application. To enable comparison between exoskeleton
types and outcome measures, standardized measurement protocols are required to allow
better comparisons between groups. Future studies may reveal differential benefits based
on factors like locomotion capacity, requiring tailored training approaches. In clinical
practice, optimizing IC trajectory through screening, in-depth assessment, personalized
care plans, and community engagement is crucial. LLE manufacturers should consider
parameters related to older patients’ functional abilities and intrinsic capacities to design
efficient exoskeletons. Additionally, involving older adults in the development process,
through co-design methodologies, could enhance the effectiveness and acceptance of such
technologies [76].
Unfortunately, our study encountered limitations in making outcome comparisons
due to the substantial variability in study characteristics and participant profiles among
the included studies. The limited number of eligible studies also reflected the scarcity of
the literature on lower limb exoskeleton use in older adults, compounded by insufficient
details regarding participant heterogeneity, comorbidities, and clinical management. Future
investigations should aim to address these variables to mitigate potential biases. Moreover,
we were not able to study differences in outcomes by training duration and frequency,
mainly due to the high heterogeneity among studies. However, future studies should focus
on further investigating the association between training effects and intervention frequency
and duration in older adults.
In summary, the findings reviewed here underscore the potential of LLEs to enhance
IC and support the implementation of LLEs to augment functional abilities in older adults,
regardless of their health status. While numerous gait rehabilitation exoskeletons are
available for clinical settings, the range of exoskeletons designed to address IC and func-
tional ability in daily life for older adults is limited. Therefore, we believe that exoskeleton
manufacturers and clinicians should work together towards the development of better
exoskeletons, which could effectively support independent living, and enhance the overall
well-being of older adults.

Sensors 2024,24, 2230 18 of 26
5. Conclusions
This review revealed consistent and remarkable improvements in various key param-
eters across all studied health conditions following LLE training. These improvements
encompassed functional abilities, IC, and performance indicators, leading to improvements
in QoL. Although longer intervention durations tend to yield more substantial improve-
ments across various indicators and aid in the rehabilitation of IC (e.g., muscle strength,
balance, endurance), even the instantaneous augmentation of functional abilities (e.g., walk-
ing, stairclimbing, sit-to-stand transfers) can be observed in a single session. These findings
underscore the potential of LLEs in promoting healthy aging and enhancing the well-being
of older adults.
Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/s24072230/s1. Table S1: Summary of studies which used ex-
oskeletons and assessed functional ability; Table S2: Summary of studies which used exoskeletons
and assessed intrinsic capacity; Table S3: Summary of studies which used exoskeletons and assessed
performance indicators.
Author Contributions: Conceptualization: R.A.G.L., E.S., I.B. and D.B.; methodology: R.A.G.L., M.F.,
R.C., E.S., I.B. and D.B.; formal analysis: R.A.G.L., M.F. and R.C.; investigation: R.A.G.L., M.F. and
R.C.; resources: M.F., E.S., I.B., D.B.; data curation: R.A.G.L., M.F. and R.C.; writing—original draft
preparation: R.A.G.L., M.F. and R.C.; writing—review and editing: R.A.G.L., M.F., R.C., E.S., I.B. and
D.B.; visualization: R.A.G.L., M.F. and R.C.; supervision: E.S., I.B. and D.B.; project administration:
E.S., I.B. and D.B.; funding acquisition: M.F., E.S., I.B. and D.B.; All authors have read and agreed to
the published version of the manuscript.
Funding: This work was supported by the strategic basic research project RevalExo (grant no.
S001024N) funded by the Research Foundation—Flanders (FWO), and was partly funded by Inter-
disciplinary Research Program funding from the Vrije Universiteit Brussel (IRP3, IRP12 and IRP22).
Mahyar Firouzi is a Fundamental Research fellow funded by the Research Foundation Flanders
(FWO, award number 11G9622N).
Institutional Review Board Statement: Ethical review and approval were waived for this study, since
systematic reviews generally do not need ethics committee or institutional review board approval.
Informed Consent Statement: Not applicable.
Data Availability Statement: No new data were created or analyzed in this study. Data sharing is
not applicable to this article.
Conflicts of Interest: The authors declare no conflicts of interest.
Appendix A. Systematic Search’s Full String
PubMed
Concept 1: older adults
“Aged”[Mesh] OR “oldest old”[tiab] OR quinquagenarian*[tiab] OR sexagenarian*[tiab]
OR septuagenarian*[tiab] OR octogenarian*[tiab] OR nonagenarian*[tiab] OR frail*[tiab]
OR “functionally impair*” [tiab] OR “older adult*” [tiab] OR senior*[tiab] OR retire*[tiab]
OR “Geriatrics”[Mesh:NoExp] OR geriatric*[tiab] OR ((50[tiab] OR 55[tiab] OR 65[tiab]
OR 70[tiab] OR 75[tiab] OR 80[tiab] OR 85[tiab] OR 90[tiab] OR 95[tiab] OR 100[tiab])
AND (age[tiab] OR aged[tiab] OR ages[tiab] OR old[tiab] OR year*[tiab])) OR Aged[tiab]
OR “advanced age*”[tiab] OR “advancing years” [tiab] OR ageing[tiab] OR aging[tiab]
OR elder*[tiab] OR gerontology[MeSH] OR gerontolog*[tiab] OR “old adult*” [tiab] OR
“old age*” [tiab] OR “rest home*”[tiab] OR “very old”[tiab] OR “Nursing Homes”[Mesh]
OR “nursing home*”[tiab] OR “Health Services for the Aged”[Mesh] OR “Homes for the
Aged”[Mesh] OR “Housing for the Elderly”[Mesh] OR “senior center*” [tiab]
Concept 2: exoskeletons
“Exoskeleton Device”[Mesh] OR “exoskeleton*”[tiab] OR “Exoskeleton Device*”[tiab] OR
“robotic exoskeleton*”[tiab]

Sensors 2024,24, 2230 19 of 26
Embase
Concept 1: older adults
‘aged’/exp OR ‘aged’:ti,ab,kw OR ‘elderl*’:ti,ab,kw OR ‘geriatric care’/exp OR ‘geriatric
nursing’/exp OR ‘home for the aged’/exp OR ‘old age assistance’:ti,ab,kw OR ‘aged hos-
pital patient’/exp OR ‘senior center’/exp OR ‘geriatric rehabilitation’/exp OR ‘nursing
home’/exp OR ‘nursing home*’:ti,ab,kw OR ‘skilled nursing facilit*’:ti,ab,kw OR ‘nursing
home patient’/exp OR ‘frail elderly’/exp OR ‘geriatric’/exp OR ‘geriatric*’:ti,ab,kw OR
((50 OR 55 OR 65 OR 70 OR 75 OR 80 OR 85 OR 90 OR 95 OR 100) NEAR/3 (age OR ages OR
aged OR old OR year*)):ti,ab,kw OR ‘oldest old’:ti,ab,kw OR ‘quinquagenarian*’:ti,ab,kw
OR ‘sexagenarian*’:ti,ab,kw OR ‘septuagenarian*’:ti,ab,kw OR ‘octogenarian*’:ti,ab,kw OR
‘nonagenarian*’:ti,ab,kw OR ‘frail*’:ti,ab,kw OR ‘functionally impair*’:ti,ab,kw OR ‘older
adult*’:ti,ab,kw OR ‘senior*’:ti,ab,kw OR ‘retire*’:ti,ab,kw OR ‘advanced age*’:ti,ab,kw OR
‘advancing years’:ti,ab,kw OR ‘ageing’:ti,ab,kw OR ‘aging’:ti,ab,kw OR ‘gerontolog*’:ti,ab,kw
OR ‘old adult*’:ti,ab,kw OR ‘old age*’:ti,ab,kw OR ‘rest home*’:ti,ab,kw OR ‘very old’:ti,ab,kw
Concept 2: exoskeletons
‘exoskeleton’/de OR ‘exoskeleton*’:ti,ab,kw OR ‘Exoskeleton Device*’:ti,ab,kw OR ‘robotic
exoskeleton*’:ti,ab,kw
WOS
Concept 1: older adults
“aged” OR “elderl*” OR “oldest old” OR ‘’quinquagenarian*” OR ‘’sexagenarian*” OR
‘’septuagenarian*” OR ‘’octogenarian*” OR ‘’nonagenarian*” OR ‘’ frail*” OR “functionally
impair*” OR “older adult*” OR ‘’senior*” OR ‘’retire*” OR ‘’geriatric*” OR “advanced
age*”OR “advancing years” OR ‘’ageing” OR ‘’aging” OR ‘’gerontolog*” OR “old adult*”
OR “old age*” OR “rest home*” OR “very old” OR ‘’nursing home*”OR “old age assistance“
OR ‘’skilled nursing facilit*” OR “50” OR “55” OR “65” OR “70” OR “75” OR “80” OR “85”
OR “90” OR “95” OR “100”
Concept 2: exoskeletons
“exoskeleton*” OR “Exoskeleton Device*” OR “robotic exoskeleton*”
Cochrane
Concept 1: older adults
#1: [mh “Aged”]
#2: [mh “Health Services for the Aged”]
#3: [mh “Senior Centers”]
#4: [mh “Geriatrics”]
#5: [mh “Gerontology”]
#6: [mh “Housing for the Elderly”]
#7: [mh “Nursing Homes”]
#8: (“oldest old” OR quinquagenarian* OR sexagenarian* OR septuagenarian* OR octo-
genarian* OR nonagenarian* OR frail* OR (functionally NEXT impair*) OR (older NEXT
adult*) OR senior* OR retire* OR geriatric* OR
((50 OR 5565 OR 70 OR 75 OR 80 OR 85 OR 90 OR 95 OR 100) NEAR/3 (age OR aged OR
ages OR old OR year*))
OR Aged
OR (advanced NEXT age*) OR “advancing years” OR ageing OR aging OR elder* OR
gerontolog* OR (old NEXT adult*) OR (old NEXT age*) OR (rest NEXT home*) OR “very
old” OR (nursing NEXT home*) OR ‘’geriatr*”):ti,ab,kw
#9: #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8
Concept 2: traumatic brain injury
#10: [mh “Exoskeleton Device”]
#11: exoskeleton NEXT device OR robotic NEXT exoskeleton* OR exoskeleton:ti,ab,kw
#12: #10 OR #11
#13: #9 AND #12

Sensors 2024,24, 2230 20 of 26
Cinahl
Concept 1: older adults
(MH “Aged+”) OR (MH “Health Services for the Aged”) OR (MH ‘’Aged, Hospitalized”)
OR (MH “Senior Centers”) OR (MR” Rehabilitation, Geriatric”) OR (MH “Housing for
the Elderly”) OR (MH “Gerontologic Nursing+”) OR (MH “Gerontologic Care”) OR (MH
“Nursing Homes+”) OR (MH “Nursing Home Patients”) OR (MH “Frail Elderly”)
OR TI(“aged” OR “elderl*” OR ‘’oldest old” OR ‘’ quinquagenarian *” OR ‘’sexagenar-
ian*” OR ‘’septuagenarian*” OR ‘’octogenarian*” OR ‘’nonagenarian*” OR ‘’ frail*” OR
“functionally impair*” OR “older adult*” OR ‘’senior*” OR ‘’retire*” OR ‘’geriatric*” OR
“advanced age*”OR “advancing years” OR ‘’ageing” OR ‘’aging” OR ‘’gerontolog*” OR
“old adult*” OR “old age*” OR “rest home*” OR “very old” OR ‘’nursing home*” OR ‘’old
age assistance” OR ‘’skilled nursing facilit*”
OR ((50 OR 55 OR 65 OR 70 OR 75 OR 80 OR 85 OR 90 OR 95 OR 100) N3 (age OR ages OR
aged OR old OR year*)))
OR AB(“aged” OR “elderl*” OR ‘’oldest old” OR ‘’ quinquagenarian *” OR ‘’sexagenar-
ian*” OR ‘’septuagenarian*” OR ‘’octogenarian*” OR ‘’nonagenarian*” OR ‘’ frail*” OR
“functionally impair*” OR “older adult*” OR ‘’senior*” OR ‘’retire*” OR ‘’geriatric*” OR
“advanced age*”OR “advancing years” OR ‘’ageing” OR ‘’aging” OR ‘’gerontolog*” OR
“old adult*” OR “old age*” OR “rest home*” OR “very old” OR ‘’nursing home*” OR ‘’old
age assistance” OR ‘’skilled nursing facilit*”
OR ((50 OR 55 OR 65 OR 70 OR 75 OR 80 OR 85 OR 90 OR 95 OR 100) N3 (age OR ages OR
aged OR old OR year*)))
Concept 2: exoskeletons
(MH “Exoskeleton Device”) OR TI(“exoskeleton*” OR “Exoskeleton Device*” OR “robotic
exoskeleton*”)OR AB(“exoskeleton*” OR “Exoskeleton Device*” OR “robotic exoskele-
ton*”)
PEDro
Concept 1: older adults
old*quinquagenarian*sexagenarian*septuagenarian*octogenarian*nonagenarian*frail* im-
pair*seni*retire*geriatric*age*aging*elder*gerontolog*adult*rest*nursing*senior*
Concept 2: exoskeletons
exoskeleton*
IEEE Xplore Digital Library
Concept 1: older adults
oldest old OR quinquagenarian OR sexagenarian OR septuagenarian OR octogenarian OR
nonagenarian OR frail OR functionally impair* OR older adult* OR senior OR retire* OR
geriatric OR Aged OR advanced age* OR advancing years OR ageing OR aging OR elder
OR gerontology* OR old adult OR old age* OR rest home OR very old OR nursing home*
OR senior center
Concept 2: exoskeletons
exoskeleton OR Exoskeleton Device OR robotic exoskeleton

Sensors 2024,24, 2230 21 of 26
Appendix B. Studies’ Quality Assessment Based on the Downs and Blacks Scale [31]
Author (Year) Reporting External
Validity Internal Validity-Bias Internal Validity—Confounding
(Selection Bias) P T Quality
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Aprigliano (2019) [32] 1 1 1 1 0 1 1 0 0 1 0 0 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 12 Poor
Calabrò(2018) [59] 1 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 25 Good
Carral (2022) [40] 1 1 1 1 0 1 0 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 1 15 Fair
Fang (2022) [33] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 12 Poor
Firouzi (2022) [24] 1 1 1 1 0 1 1 1 0 0 1 1 1 0 0 0 1 1 1 1 0 0 0 0 0 1 0 15 Fair
Fujikawa (2022) [34] 1 1 1 2 0 1 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 12 Poor
Galle (2022) [35] 1 1 0 1 0 1 1 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 11 Poor
Gryfe (2022) [60] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 21 Good
Jayaraman (2022) [36] 1 1 0 1 0 1 1 0 0 1 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 12 Poor
Jin (2017) [37] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 0 0 14 Poor
Jin (2019) [41] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 0 16 Fair
Kawashima (2022) [61] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 1 1 20 Good
Koseki (2021) [42] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 1 1 18 Fair
Lee (2017) [43] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 0 15 Fair
Lee (2017) [44] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 0 0 17 Fair
Lee (2022) [62] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 1 1 21 Good
Lefeber (2018) [45] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 1 16 Fair
Longatelli (2021) [46] 1 1 1 1 0 1 1 0 0 1 1 1 1 1 0 0 1 1 1 1 1 1 0 0 0 0 1 18 Fair
Martini (2019) [47] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 0 1 19 Fair
Monaco (2017) [48] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 0 1 0 0 0 0 0 15 Fair
Norris (2007) [49] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 1 17 Fair

Sensors 2024,24, 2230 22 of 26
Author (Year) Reporting External
Validity Internal Validity-Bias Internal Validity—Confounding
(Selection Bias) P T Quality
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Panizzolo (2022) [50] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 0 15 Fair
Park (2021) [51] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 1 0 1 1 1 1 1 1 1 1 0 0 1 19 Fair
Roggeman (2022) [52] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 1 17 Fair
Rojek (2020) [63] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 21 Good
Romanato (2022) [38] 1 1 1 1 0 1 0 0 0 0 1 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 11 Poor
Setoguchi (2022) [53] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 0 1 18 Fair
Shore (2022) [39] 1 1 0 1 0 1 0 0 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 0 0 1 13 Poor
Son (2021) [64] 1 1 1 1 0 1 0 0 0 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 0 1 1 20 Good
Taki (2020) [54] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 1 1 18 Fair
Verrusio (2018) [55] 1 1 1 1 0 1 1 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 0 1 18 Fair
Watanabe (2017) [56] 1 1 0 1 0 1 1 0 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 0 1 1 19 Fair
Yeung (2021) [65] 1 1 1 1 1 1 1 1 0 0 1 1 1 0 1 0 1 1 1 1 1 1 1 1 1 0 1 22 Good
Yoshikawa (2018) [57] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 1 17 Fair
Yoshimoto (2022) [66] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 0 6 22 Good
Yun (2020) [58] 1 1 1 1 0 1 1 0 0 1 1 1 1 0 0 0 1 1 1 1 1 1 0 0 0 1 1 18 Fair
P = power; T = tot.

Sensors 2024,24, 2230 23 of 26
Downs and Blacks Scale questions (Downs and Black, 1998):
1. Is the hypothesis/aim/objective of the study clearly described?
2. Are the main outcomes to be measured clearly described in the Introduction or Methods
section?
3. Are the characteristics of the patients included in the study clearly described?
4. Are the interventions of interest clearly described?
5. Are the distributions of principal confounders in each group of subjects to be compared
clearly described?
6. Are the main findings of the study clearly described?
7. Does the study provide estimates of the random variability in the data for the main
outcomes?
8. Have all important adverse events that may be a consequence of the intervention been
reported?
9. Have the characteristics of patients lost to follow-up been described?
10. Have actual probability values been reported (e.g., 0.035 rather than <0.05) for the main
outcomes except where the probability value is less than 0.001?
11. Were the subjects asked to participate in the study representative of the entire population
from which they were recruited?
12. Were those subjects who were prepared to participate representative of the entire
population from which they were recruited?
13. Were the staff, places, and facilities where the patients were treated, representative of
the treatment the majority of patients receive?
14. Was an attempt made to blind study subjects to the intervention they have received?
15. Was an attempt made to blind those measuring the main outcomes of the intervention?
16. If any of the results of the study were based on “data dredging”, was this made clear?
17. In trials and cohort studies, do the analyses adjust for different lengths of follow-up
of patients, or in case-control studies, is the time period between the intervention and
outcome the same for cases and controls?
18. Were the statistical tests used to assess the main outcomes appropriate?
19. Was compliance with the intervention/s reliable?
20. Were the main outcome measures used accurate (valid and reliable)?
21. Were the patients in different intervention groups (trials and cohort studies) or were the
cases and controls (case-control studies) recruited from the same population?
22. Were study subjects in different intervention groups (trials and cohort studies) or were
the cases and controls (case-control studies) recruited over the same period of time?
23. Were study subjects randomized to intervention groups?
24. Was the randomized intervention assignment concealed from both patients and health
care staff until recruitment was complete and irrevocable?
25. Was there adequate adjustment for confounding in the analyses from which the main
findings were drawn?
26. Were losses of patients to follow-up taken into account?
27. Did the study have sufficient power to detect a clinically important effect where the
probability value for a difference being due to chance is less than 5%?
All questions were scored on the following scale: yes = 1, unable to determine = 0, and
no = 0.
References
1. WHO. World Report on Ageing and Health; WHO Press: Geneva, Switzerland, 2015; p. 260.
2.
Grimmer, M.; Riener, R.; Walsh, C.J.; Seyfarth, A. Mobility related physical and functional losses due to aging and disease—A
motivation for lower limb exoskeletons. J. Neuroeng. Rehabil. 2019,16, 2. [CrossRef] [PubMed]
3.
Jung, M.M.; Ludden, G.D.S. Potential of Exoskeleton Technology to Assist Older Adults with Daily Living. In Proceedings of
the Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems, Montreal, QC, Canada, 21–26
April 2018; pp. 1–6.
4. United Nations. Convention on the rights of persons with disabilities. Eur. J. Health Law 2007,14, 281–298.

Sensors 2024,24, 2230 24 of 26
5.
Beard, J.R.; Jotheeswaran, A.T.; Cesari, M.; Araujo de Carvalho, I. The structure and predictive value of intrinsic capacity in a
longitudinal study of ageing. BMJ Open 2019,9, e026119. [CrossRef] [PubMed]
6.
Veronese, N.; Honvo, G.; Amuthavalli Thiyagarajan, J.; Rizzoli, R.; Cooper, C.; Bruyère, O.; Mikton, C.; Sumi, Y.; Diaz, T.; Reginster,
J.Y.; et al. Attributes and definitions of locomotor capacity in older people: A World Health Organisation (WHO) locomotor
capacity working group meeting report. Aging Clin. Exp. Res. 2022,34, 481–483. [CrossRef] [PubMed]
7.
Bautmans, I.; Knoop, V.; Amuthavalli Thiyagarajan, J.; Maier, A.B.; Beard, J.R.; Freiberger, E.; Belsky, D.; Aubertin-Leheudre, M.;
Mikton, C.; Cesari, M.; et al. WHO working definition of vitality capacity for healthy longevity monitoring. Lancet Healthy Longev.
2022,3, e789–e796. [CrossRef] [PubMed]
8.
Gusi, N.; Adsuar, J.C.; Corzo, H.; del Pozo-Cruz, B.; Olivares, P.R.; Parraca, J.A. Balance training reduces fear of falling and
improves dynamic balance and isometric strength in institutionalised older people: A randomised trial. J. Physiother. 2012,
58