Hierarchy of Dysfunction Related to Dressing Performance in Stroke Patients: A Path Analysis Study
Hierarchy of Dysfunction Related to Dressing Performance in Stroke Patients: A Path Analysis Study
Takaaki Fujita 0 1
Hirofumi Nagayama 1
Atsushi Sato 1
Yuichi Yamamoto 1
Kazuhiro Yamane 1
Koji Otsuki 1
Kenji Tsuchiya 1
Fusae Tozato 1
0 Department of Rehabilitation, Faculty of Health Sciences, Tohoku Fukushi University , Sendai, Miyagi , Japan , 2 Department of Rehabilitation Sciences, Gunma University Graduate School of Health Sciences , Maebashi, Gunma , Japan , 3 Department of Rehabilitation, Faculty of Health and Welfare, Kanagawa University of Human Services , Yokosuka, Kanagawa , Japan , 4 Department of Physical Therapy, Yachiyo Rehabilitation College , Yachiyo, Chiba , Japan , 5 Department of Rehabilitation, Northern Fukushima Medical Center , Date, Fukushima , Japan , 6 Department of Rehabilitation, Japan Community Healthcare Organization Gunma Chuo Hospital , Maebashi, Gunma , Japan
1 Editor: Johannes Boltze, Fraunhofer Research Institution of Marine Biotechnology , GERMANY
Previous reports indicated that various dysfunctions caused by stroke affect the level of independence in dressing. These dysfunctions can be hierarchical, and these effects on dressing performance can be complicated in stroke patients. However, there are no published reports focusing on the hierarchical structure of the relationships between the activities of daily living and balance function, motor and sensory functions of the affected lower limb, strength of the abdominal muscles and knee extension on the unaffected side, and visuospatial deficits. The purpose of this study was to elucidate the hierarchical and causal relationships between dressing performance and these dysfunctions in stroke patients. This retrospective study included 104 first-time stroke patients. The causal relationship between the dressing performance and age, time post stroke, balance function, motor and sensory functions of the affected lower limb, strength of the abdominal muscles and knee extension on the unaffected side, and visuospatial deficits were examined using path analysis. A hypothetical path model was created based on previous studies, and the goodness of fit between the data and model were verified. A modified path model was created that achieved an almost perfect fit to the data. Balance function and abdominal muscle strength have direct effects on dressing performance, with standardized direct effect estimates of 0.78 and 0.15, respectively. Age, motor and sensory functions of the affected lower limb, and strength of abdominal muscle and knee extension on the unaffected side have indirect effects on dressing by influencing balance function. Our results suggest that dressing performance depends strongly on balance function, and it is mainly influenced by the motor
Funding: This work was supported by the Japan
Society for the Promotion of Science (grant Number
(TF). The funders had no role in study design, data
collection and analysis, decision to publish, or
preparation of the manuscript.
function of the affected lower limb.
Competing Interests: The authors have declared
that no competing interests exist.
Various physical and perceptual dysfunctions such as motor paralysis, sensory disturbance,
muscle weakness, and visuospatial deficits can negatively impact an individual’s independence
in the activities of daily living (ADL). A stroke can cause multiple dysfunctions simultaneously,
resulting in the need for care and substantial assistance. Understanding the causal relationships
between various dysfunctions and ADL is important to facilitate effective stroke patient
rehabilitation. However, the relationships between specific deficits and ADL are not simple. For
example, it has been reported that sensory disturbances in stroke patients are associated with
independence in ADL [
], mobility [
], motor functions of the upper limb [
], and balance
]. On the other hand, other studies have reported that balance functions in stroke
patients are associated with independence in ADL [
] and the motor functions of the affected
upper limb [
]. When integrating this information, we presumed that sensory disturbances
deteriorate balance functions, which consequently reduces the independence in ADL and the
motor function of the affected upper limb. This reasoning suggests that the various
dysfunctions caused by a stroke are hierarchical, and these dysfunctions induce complex effects on
ADL. However, to the best of our knowledge, there are no published reports that investigate
the hierarchical relationships between various dysfunctions and ADL.
Dressing is an ADL that is difficult for stroke patients to perform [
]. Previous studies have
reported that a stroke patient’s level of independence in dressing themselves is associated with
balance function [
], motor and sensory functions of the affected limbs , abdominal
muscle strength [
], and visuospatial deficits [
]. As stated above, there is a possibility that these
dysfunctions affect the level of independence in dressing not in parallel, but in a hierarchical
manner. However, the hierarchical structure of the relationships between the level of
independence in dressing and various dysfunctions has not been clarified. The findings on which
dysfunction is directly or indirectly related to the level of independence in dressing and those on
the degree to which each dysfunction affects the level of independence in dressing will be very
useful for planning and executing rehabilitation strategies. Therefore, the purpose of this study
was to elucidate the hierarchical structure and causal relationships between the level of
independence in dressing and balance function, motor and sensory functions of the affected lower
limb, strength of the abdominal muscles and knee extension on the unaffected side, and
visuospatial deficits using path analysis.
Materials and Methods
This study included 104 patients with stroke who participated in a rehabilitation program and
were discharged from a hospital convalescent rehabilitation ward from April 2011 to February
2014 (Table 1). The study inclusion criteria were: diagnosis of a first onset cerebral hemorrhage
or cerebral infarction; diagnosis of a unilateral supratentorial hemispheric lesion; being at least
4 weeks from stroke onset; and having at least a 2-week stay in the hospital convalescent
rehabilitation ward. Exclusion criteria of subarachnoid hemorrhage and missing records were
applied to all assessments mentioned below (Procedures and Data Collection). All patients
participated in a conventional in-hospital stroke rehabilitation program that included
occupational therapy, physical therapy and speech therapy. The study protocol was reviewed and
approved by the institutional ethics review board of Northern Fukushima Medical Center
(No.56) and Tohoku Fukushi University (RS141201). Participant informed consent was not
required because of the non-interventional, retrospective study design.
2 / 9
Procedures and Data Collection
This was a retrospective study by design that analyzed data from electronic medical records from
the time of discharge. The variables used in this study were selected on the basis of previous
reports (referenced in the Statistical Analysis section) where these variables were studied and
associated with dressing performance or balance function. The FIM1 instrument [
for dressing the upper and lower body were used to assess each participant’s level of
independence in dressing. The FIM1 instrument assesses the amount of assistance required in ADL and
is rated on a 7-point scale, with a higher score indicating better performance. The FIM1
instrument’s scoring criteria were as follows: a score of 7 indicates complete independence, a score of 5
indicates supervision or setup, a score of 3 indicates moderate assistance, and a score of 1
indicates total assistance [
]. The reliability of the FIM1 instrument has been confirmed in stroke
]. In this study, the lower score on FIM1 instrument for dressing the upper and
lower body was regarded as the level of independence in dressing. The Berg Balance Scale (BBS)
 was used to assess participant balance function. BBS is designed to measure balance among
the elderly. It assesses an individual’s static and dynamic balancing abilities through the
performance of functional tasks, including balance while sitting and while standing on 1 foot. It is
composed of 14 items scored on a 5-point scale ranging from 0 to 4 points. The maximum total score
is 56 points, with a higher score indicating better balance function. A BBS score of 40 points was
reported as a cut-off value for independent functional walking ability among stroke inpatients
], whereas achieving a BBS score of 44 points was the reported requirement for independent
]. The reliability and validity of this scale have been confirmed in stroke patients [
The Stroke Impairment Assessment Set (SIAS) [
] was used to assess motor and sensory
functions of the affected lower limb, strength of the abdominal muscles and knee extension on the
unaffected side, and visuospatial deficits. SIAS is used to comprehensively assess impaired
functions due to stroke. Each item is rated on a 3- or 5-point scale, with a higher score indicating
better function (S1 Table). These tests also have confirmed reliability and validity [
In this study, the total score for the motor function tests for the affected lower limb on SIAS
(the hip flexion test, knee extension test, and foot-pat test) was used to assess the motor
3 / 9
function of the affected lower limb; while the total score for the sensory function tests for the
affected lower limb on SIAS (light touch and position test) was used to assess the sensory
function of the affected lower limb.
Path analysis based on structural equation modeling was performed to clarify the
hierarchical structure of the relation and causal association between the level of independence in
dressing and various dysfunctions. A hypothetical path model was created on the basis of previous
], which reported that the level of independence in dressing for stroke patients is
associated with: balance function, motor and sensory function of affected side lower limb, the
strength of the abdominal muscles, and unilateral spatial neglect (Fig 1). Therefore, we
established a hypothesis describing which of these 5 functions has a direct effect on the level of
independence in dressing in addition to the age and time post stroke. In addition, previous studies
have reported that the balance function associates with motor function and sensory function of
affected side lower limb [
], the strength of abdominal muscle [
] and knee extension of
unaffected side [
], and unilateral spatial neglect [
], in stroke patients. Therefore, we set
up a hypothesis in which these functions have direct effects on balance functions and indirect
effects on dressing independence in addition to age and time post stroke.
As indicators of goodness of fit, we used the chi-squared test, Goodness Fit Index (GFI),
Adjusted Goodness Fit Index (AGFI), Comparative Fit Index (CFI), and Root Mean Square
Error of Approximation (RMSEA). The criteria for a good fit between the study data and
the model were a GFI of at least 0.95, an AGFI of at least 0.90, a CFI of at least 0.95, and an
Fig 1. Path diagram of the relationship between the level of independence in dressing and physical and visuospatial dysfunctions. Abbreviations:
L/L, Lower limb.
4 / 9
Fig 2. Modified model of the path diagram with the results of parameter estimates. A unidirectional arrow indicates a standardized path coefficient. A
bidirectional arrow indicates a correlation coefficient. Abbreviations: L/L, Lower limb; SIAS, Stroke impairment assessment set; BBS, Berg balance scale.
*P < 0.05, ** P < 0.01.
RMSEA of less than 0.07 [
]. SPSS Amos 23.0 statistical software (IBM, Armonk, New York,
USA) was used for all analyses. The level of significance was set at <5%.
The assessment outcomes are presented in Table 1. The FIM1 instrument mean ± standard
deviation score for dressing (1–7 points) was 4.3 ± 1.8 points. The indices of goodness of fit
using the path model described in Fig 1 were χ2 = 0.326 (P = 0.568), GFI = 0.999, AGFI = 0.968,
CFI = 1.000, RMSEA = 0.000, with all indices fulfilling the established criteria. However,
nonsignificant path coefficients were also included in this model. Therefore, a modified path model
that eliminated the non-significant paths was created (Fig 2). The indices of goodness of fit in
the modified model presented in Fig 2 were χ2 = 4.760 (P = 0.575), GFI = 0.988, AGFI = 0.943,
CFI = 1.000, and RMSEA = 0.000. All path coefficients were significant in the modified model,
which accounted for 79.9% of the variance in FIM1 instrument score for dressing, and for
78.8% of the variance in BBS.
The standardized estimated direct effects of the BBS and the strength of abdominal muscles
from SIAS on the FIM1 instrument score for dressing were 0.78 and 0.15, respectively. The
standardized estimated direct effects of the motor and sensory functions of the affected side
lower limb, strength of the abdominal muscles from SIAS, strength of knee extension on the
unaffected side, and individual age on the BBS score were 0.46, 0.13, 0.22, 0.23, and −0.22,
respectively. The standardized estimated total effects of the BBS score, motor and sensory
functions of the affected side lower limb, strength of the abdominal muscles from SIAS, strength of
knee extension on the unaffected side, and individual age on the FIM1 instrument score for
dressing were 0.78, 0.36, 0.10, 0.32, 0.18, and −0.17, respectively.
5 / 9
This study demonstrated 2 major findings. First, the balance function and the strength of the
abdominal muscles have direct effects on the level of independence in dressing in stroke
patients. Second, the motor and sensory functions of the affected lower limb, the strength of
the abdominal muscles and knee extension on the unaffected side, and age have indirect effects
on dressing independence by mediating balance functions in stroke patients.
Previous studies have reported that the level of independence in dressing relates to the
balance function [
], motor and sensory functions of the affected limb , strength of the
abdominal muscles [
], and degree of visuospatial deficits [
]. Therefore, we established a
hypothesis by which all these functions are related to the level of independence in dressing.
From the initial model, we created a modified path model that eliminated the non-significant
paths. This modified model was validated by the goodness of fit between the model and data,
and all path coefficients were significant in the modified model. The model could account for
more than 79.9% of the variance in the FIM1 instrument score for dressing.
In particular, the impact of balance function on dressing performance was very strong. This
result is in agreement with a previous study that reported a close relationship between the level
of independence in dressing and balance function [
]. Because dressing involves gross
movement of the limbs and a shift in the center of gravity when sitting or standing, the level of
independence in dressing can depend on balance function. In addition, the strength of the
abdominal muscles contributes to trunk stability, and trunk stability is important for
maintaining antigravity postures such as sitting and standing as well as smooth execution of limb
]. The strength of the abdominal muscles may contribute to the level of
independence in dressing by stabilizing the trunk. On the other hand, our study revealed that
motor and sensory functions of the affected lower limb, strength of knee extension on the
unaffected side, and age also contribute to the level of independence in dressing, but indirectly
through their influence on balance function. Considering their impact on balance function, the
strongest effects were from the motor function of the affected lower limb, followed in
decreasing order by the strength of knee extension on the unaffected side, the strength of the
abdominal muscles, and the sensory function of the affected lower limb.
These findings suggest that among stroke patients, dressing performance strongly depends
on the balance function. Consequently, to improve dressing performance, training that focuses
on improving balance function should be recommended. In our experience, a basic practice
that improves dressing performance is repetitive dressing training because it facilitates the
learning of compensatory movement techniques involving the unaffected upper limb and also
improves balance functions. However, previous studies have shown that the amount of balance
required for independent dressing with and without supervision or setup from caregivers was
marked by BBS scores of 32 and 44 points, respectively [
]. The present findings indicate that
high-level balancing function is required for dressing independence. Therefore, with a focus on
improving balance function, repetitive training with dressing may not be sufficient, and it may
also be necessary to add balance-specific exercises to achieve effective improvement in dressing
In addition, our results suggested that the exercises that improve motor and sensory
functions of the affected lower limb, strength of the abdominal muscles, and strength of knee
extension on the unaffected side can improve the ability to balance, thus improving dressing
performance. In particular, our results suggest that the motor function of the affected lower
limb and the strength of the abdominal muscles are very important because they have large
effects on the balance function. Previous studies have shown that trunk muscle strength has
deteriorated in patients with stroke when compared with the strength in healthy age-matched
6 / 9
], and the greatest impairment is in forward flexion strength [
]. In addition, the
strength of abdominal muscles or trunk flexion has been associated with dressing performance
] and balance [
] in previous reports. Therefore, it is possible that the exercises focused on
not only the lower limb motor function but also on strengthening the abdominal muscles will
effectively improve individual balance function and dressing performance. In addition, both
the present study and a previous report [
] have indicated that the sensory functions of the
affected lower limb can impact balance function, although this effect was somewhat limited
compared with impact of altered motor functions. It was recently reported that proprioception
training alone or in combination with imagery exercises improved stroke patient perception of
joint positioning and ability to balance [
]. For stroke patients with severe sensory
disturbances, proprioception exercises may be an effective means of improving balance.
There are some limitations in this study. One of the limitations is that no psycho-social
variables were addressed. In addition, a stratified analysis based on stroke severity and affected side
was not performed. Moreover, the stroke patients in this study had a large range of
demographics and stroke characteristics, and this study involved only a small sample size. Additional
research is warranted to address these limitations and validate the study findings.
S1 Table. Stroke Impairment Assessment Set’s scoring criteria (Excerpt of the item used by
The use of the FIM1 instrument to collect data for this research study was authorized and
conducted in accordance with the term of a special purpose license granted to the licensee by the
Uniform Data System for Medical Rehabilitation (UDSMR). Licensee has not been trained by
UDSMR in the use of the FIM1 instrument, and the patient data collected during the course
of this research study has not been submitted to or processed by UDSMR. No implication is
intended that such data has been or will be subjected to UDSMR’s standard data processing
procedures or that it is otherwise comparable to data processed by UDSMR.
Conceived and designed the experiments: TF FT. Performed the experiments: TF AS YY KY
KO. Analyzed the data: TF HN KT. Contributed reagents/materials/analysis tools: TF. Wrote
the paper: TF.
7 / 9
8 / 9
1. Tyson SF , Crow JL , Connell L , Winward C , Hillier S . Sensory impairments of the lower limb after stroke: a pooled analysis of individual patient data . Top Stroke Rehabil 2013 ; 20 : 441 - 9 . doi: 10 .1310/tsr2005- 441 PMID: 24091286
2. Yanase Y , Murakami M. Relationship between sensory disturbance and performance in upper limb among patients with hemiparesis . J Jpn Phys Ther Assoc 2010 ; 37 : 397 - 402 .
3. Mercier L , Audet T , Hebert R , Rochette A , Dubois MF . Impact of motor, cognitive, and perceptual disorders on ability to perform activities of daily living after stroke . Stroke 2001 ; 32 : 2602 - 8 . PMID: 11692024
4. Wee SK , Hughes AM , Warner MB , Brown S , Cranny A , Mazomenos EB , et al. Effect of trunk support on upper extremity function in people with chronic stroke and people who are healthy . Phys Ther 2015 ; 95 : 1163 - 71 . doi: 10 .2522/ptj.20140487 PMID: 25721122
5. Edmans JA. , Lincoln NB . The relationship between perceptual deficits after stroke and independence in activities of daily living . Br J Occup Ther 1990 ; 53 : 139 - 42 .
6. Fujita T , Sato A , Yamamoto Y , Tsuchiya K , Otsuki K , Tozato F. Motor function cut-off values for independent dressing in stroke patients . Am J Occup Ther 70 :in press.
7. Fujita T , Sato A , Yamamoto Y , Yamane K , Otsuki K , Tsuchiya K , et al. Relationship between dressing and motor functions in stroke patients: A study with partial correlation analysis . J Phys Ther Sci 2015 ; 27 : 3771 - 3774 . doi: 10 .1589/jpts.27.3771 PMID: 26834349
8. Walker MF , Lincoln NB . Factors influencing dressing performance after stroke . J Neurol Neurosurg Psychiatry 1991 ; 54 : 699 - 701 . PMID: 1940941
9. Fujita T , Sato A , Togashi Y , Kasahara R , Ohashi T , Yamamoto Y. Contribution of abdominal muscle strength for various activities of daily living in stroke patients with mild paralysis . J Phys Ther Sci 2015 ; 27 : 815 - 8 . doi: 10 .1589/jpts.27.815 PMID: 25931737
10. Saito Y , Toshima M , Nori K , Kimura N. Factors affecting upper-body dressing ability in post-stroke patients: a functional assessment . J Jpn Occup Ther Assoc 2012 ; 31 : 134 - 40 .
11. The Data Management Service of the Uniform Data System for Medical Rehabilitation and the Center for functional Assessment Research. Guide for the Uniform Data Set for Medical Rehabilitation, Version 3 .0. State University of New York at Buffalo, NY, 1990
12. Friche J , Unsworth C , Worrell F. Reliability of the functional independence measure with occupational therapists . Aust Occup Ther J 1993 ; 40 : 7 - 15 .
13. Segal ME , Schall RR . Determining the functional/health status and its relation to disability in stroke survivors . Stroke 1994 ; 25 : 2391 - 7 . PMID: 7974578
14. Berg K , Wood-Dauphinee S , Williams JI , Gayton D . Measuring Balance in the elderly: Preliminary development of an instrument . Physiother Can 1989 ; 41 : 304 - 11 .
15. Usuda S , Araya K , Umehara K , Endo M , Shimizu T , Endo F : Construct validity of functional balance scale in stroke inpatients . J Phys Ther Sci 1998 ; 10 : 53 - 56 .
16. Berg K , Wood-Dauphinee S , Williams JI . The Balance Scale: reliability assessment with elderly residents and patients with an acute stroke . Scand J Rehabil Med 1995 ; 27 : 27 - 36 . PMID: 7792547
17. Chino N , Sonoda S , Domen K , Saito E , Kimura A . Stroke impairment assessment set (SIAS) -a new evaluation instrument for stroke patients . Jpn J Rehabil Med 1994 ; 31 : 119 - 25 .
18. Domen K. Reliability and validity of stroke impairment assessment set (SIAS) (1): items of affected-side motor function, motor tone, deep tendon reflex, and unaffected-side function . Jpn J Rehabil Med 1995 ; 32 : 113 - 22 .
19. Sonoda . S. Reliability and validity of stroke impairment assessment set (SIAS) (2): the items comprise the trunk, higher cortical function, and sensory function and effectiveness as outcome predictor . Jpn J Rehabil Med 1995 ; 32 : 123 - 32 .
20. Tyson SF , Hanley M , Chillala J , Selley A , Tallis RC . Balance disability after stroke . Phys Ther 2006 ; 86 : 30 - 8 . PMID: 16386060
21. Karatas M , Cetin N , Bayramoglu M , Dilek A . Trunk muscle strength in relation to balance and functional disability in unihemispheric sroke patients . Am J Phys Med Rehabil 2004 ; 83 : 81 - 7 . PMID: 14758293
22. Olawale OA , Ajiboye AO , Maduagwu V . Relationship between Dynamic Balance and the Muscular Strength of the Non-paretic Lower Limbs of Patients with Post-stroke Hemiplegia . Niger J Health Biomed Sci 2003 ; 2 : 16 - 20 .
23. Bonan IV , Hubeaux K , Gellez-Leman MC , Guichard JP , Vicaut E , Yelnik AP . Influence of subjective visual vertical misperception on balance recovery after stroke . J Neurol Neurosurg Psychiatry 2007 ; 78 : 49 - 55 . PMID: 17012343
24. Nijboer TC , Ten Brink AF , van der Stoep N , Visser-Meily JM . Neglecting posture: differences in balance impairments between peripersonal and extrapersonal neglect . Neuroreport 2014 ; 25 : 1381 - 5 . doi: 10 . 1097/WNR.0000000000000277 PMID: 25340562
25. Hooper D , Coughlan J , Mullen M. Structural Equation Modelling: Guidelines for Determining Model Fit . EJBRM 2008 ; 6 : 53 - 60 .
26. Hodges PW , Richardson CA. Relationship between limb movement speed and associated contraction of the trunk muscles . Ergonomics 1997 ; 40 : 1220 - 30 . PMID: 9375536
27. Tokuno CD , Cresswell AG , Thorstensson A , Carpenter MG . Recruitment order of the abdominal muscles varies with postural task . Scand J Med Sci Sports 2013 ; 23 : 349 - 54 . doi: 10 .1111/j.1600- 0838 . 2011 . 01394 . x PMID : 22092690
28. Liao CF , Liaw LJ , Wang RY , Su FC , Hsu AT . Relationship between trunk stability during voluntary limb and trunk movements and clinical measurements of patients with chronic stroke . J Phys Ther Sci 2015 ; 27 : 2201 - 6 . doi: 10 .1589/jpts.27.2201 PMID: 26311954
29. Bohannon RW , Cassidy D , Walsh S : Trunk muscle strength is impaired multidirectionally after stroke . Clin Rehabil 1995 ; 9 : 47 - 51 .
30. Lee H , Kim H , Ahn M , You Y : Effects of proprioception training with exercise imagery on balance ability of stroke . J Phys Ther Sci 2015 ; 27 : 1 - 4 . doi: 10 .1589/jpts.27.1 PMID: 25642023