ABSTRACT
Objectives : This study aimed to investigate whether lower limb
phase angle (LL-PA) at admission predicts functional outcomes in patients
undergoing rehabilitation.
Methods : In this retrospective study, we measured the LL-PA of 256
patients (mean age 74.09±12.86 years) admitted to a convalescent rehabilitation
ward upon arrival. The primary outcomes at discharge were motor Functional
Independence Measure (FIM) scores and walking independence. The secondary
outcomes included a 10-m walking speed and Timed Up-and-Go test result.
Results : The LL-PA independently predicted discharge motor FIM
scores (β=0.323, P <0.001) and walking independence (odds ratio=3.302, 95%
confidence interval: 1.714–6.360). The sex-specific cut-off value for predicting
walking independence was 3.050° for men (sensitivity, 0.804; specificity, 0.853)
and 2.650° for women (sensitivity, 0.769; specificity, 0.812).
Conclusions : LL-PA at admission is a significant predictor of
functional outcomes in patients undergoing rehabilitation, with potential
utility in early prognostic assessments.
INTRODUCTION
Predicting functional outcomes in older adult patients in convalescent rehabilitation
hospitals is crucial for developing effective intervention strategies and for
discharge planning.1)
Traditionally, factors such as age, disease severity, and Functional Independence
Measure (FIM) scores at admission have been used as prognostic indicators. However,
these indicators do not adequately reflect the body composition and muscle quality
of patients.
Recently, phase angle (PA) measurements using bioelectrical impedance analysis (BIA)
has gained attention as a novel assessment tool.2,3) PA reflects cell membrane integrity and function,
with lower values associated with decreased muscle mass, malnutrition, and
functional decline.4,5) Notably, lower-limb PA (LL-PA) has shown strong
correlations between walking ability and lower-limb muscle strength,6,7) indicating its potential
utility in the functional assessment of older adults.
However, in patients undergoing rehabilitation, several important problems regarding
the clinical significance of LL-PA remain unresolved. First, the relationship
between LL-PA values at admission and functional independence or walking ability at
discharge remains unclear. Second, the utility of LL-PA as an independent predictor
compared with conventional prognostic factors is uncertain. Third, sex-specific
evaluation criteria for LL-PA have not been established. Therefore, to address these
problems in the present study, we aimed to analyze the relationship between LL-PA
and functional indicators in patients admitted to a convalescent rehabilitation
hospital and to verify the utility of LL-PA as a functional prognostic
indicator.
MATERIALS AND METHODS
Study Design and Participants
This retrospective observational study was conducted in the convalescent
rehabilitation ward of Tamus Urayasu Hospital. For patients admitted between
April 2022 and March 2023, the following inclusion criteria were used: (1)
first-time admission to a convalescent rehabilitation ward, (2) ability to
maintain a stable standing position (defined as the ability to stand
independently for at least 30 s without support) during BIA measurements, (3)
completion of all outcome measurements at both admission and discharge, and (4)
BIA measurements conducted within 4 weeks of admission. Of a total of 1516
patients, 1260 patients were excluded. First, 15 patients using pacemakers were
excluded because of safety concerns. Among the remaining 1501 patients, 412
patients with severe cognitive impairment [Mini-Mental State Examination (MMSE)
<10 points] were excluded because we considered that cognitive impairment,
consciousness disorder, or severe higher brain dysfunction may have
significantly affected the reliability of measurements and rehabilitation
outcomes. Of the 1089 remaining patients, 247 were excluded because they were
unable to maintain the required standing position for BIA measurements. Among
the 842 eligible patients, 294 were excluded because their initial body
composition measurements were not conducted within 4 weeks of admission.
Finally, 292 patients were excluded because of missing outcome data, resulting
in a final study population of 256 patients. All BIA measurements were performed
in the morning before any rehabilitation activities to ensure standardized
conditions.
A total of 256 patients (146 men and 110 women) were included in the analysis
(Fig. 1). Primary diagnoses
included cerebrovascular disease (n=114, 44.5%), orthopedic conditions (n=76,
29.7%), and disuse syndrome (n=66, 25.8%). This study was approved by the Ethics
Committee of Tamus Urayasu Hospital (approval number: 2023001). Because this was
a non-invasive retrospective observational study using only medical record data,
we employed an opt-out methodology instead of obtaining written informed
consent. Information about the study was publicly disclosed on our institution’s
website and in the hospital, providing patients with the opportunity to refuse
participation. This procedure was conducted in accordance with the Japanese
Ethical Guidelines for Medical and Health Research Involving Human Subjects.
Measurements
Body Composition Assessment
Body composition was measured using a multifrequency bioelectrical impedance
analyzer (TANITA 780A; TANITA, Tokyo, Japan). The patients achieved stable
medical conditions and maintained a standing position. Their measurements
were taken under standardized conditions: in the morning after a minimum of
2 h of fasting, post-urination, at room temperature 22–26 °C, and when
wearing light hospital clothing. The patients stood barefoot on the
electrode plates while holding hand electrodes. LL-PA was calculated from
the reactance and resistance values obtained at 50 kHz.
Functional Assessment
FIM was assessed at admission and discharge and consisted of 13 motor items
(91 points) and 5 cognitive items (35 points), totaling 18 items (126
points).8) The motor FIM score was primarily used in
the analysis. Walking independence at discharge was defined as a FIM walking
score of 6 or above (independent indoor walking), whereas a score of 5 or
below was considered dependent.
Physical function was evaluated using the 10-m walk and the Timed Up-and-Go
(TUG) test. The 10-m walk test was performed at a comfortable speed
following Bohannon’s method.7) The TUG test measures the time taken to
stand up from a chair, walk 3 m, turn around, walk back, and sit down,
according to Podsiadlo and Richardson’s protocol.9)
Nutritional Status and Cognitive Function Assessment
Nutritional status was evaluated using the Controlling Nutritional Status
(CONUT) score, which was calculated using serum albumin, total cholesterol,
and lymphocyte count. The CONUT score ranges from 0 to 12, with higher
scores indicating poorer nutritional status.10) Cognitive function was
assessed using MMSE, which evaluates orientation, memory, calculation,
language function, and other cognitive domains using a 30-point
scale.11)
Rehabilitation Program
Our convalescent rehabilitation ward provided standardized rehabilitation
programs 7 days a week. Comprehensive rehabilitation, including physical,
occupational, and speech therapy, was provided at an average of 7.57 ± 0.70
units (approximately 2.5 h) per day, with each unit lasting 20 min. Physical
therapy included basic movement gait, balance training, and
muscle-strengthening exercises. Occupational therapy focuses on the
activities of daily living, upper limb function, and cognitive training.
Speech therapy, including swallowing rehabilitation, articulation, and
higher brain function training, was provided when necessary.
Statistical Analysis
Statistical analyses were performed using the Statistical Package for Social
Sciences version 27.0 (IBM, Armonk, NY, USA), with a statistical
significance level of 5%. Dummy variables were created for cerebrovascular
disease and disuse syndrome for the primary diagnosis, with orthopedic
conditions as the reference. Pearson’s correlation coefficients were used to
examine the relationships between LL-PA and various assessment items for men
and women.12)
Multiple regression analysis was performed using the discharge motor FIM
score as a dependent variable and LL-PA as an independent variable,
adjusting for age, sex, body mass index, primary diagnosis, length of stay,
admission motor FIM score, and MMSE using the forced entry method.13) Logistic
regression analysis was performed using walking independence at discharge as
the dependent variable with the same covariates.14) Finally, receiver operating
characteristic (ROC) curve analysis was performed separately for men and
women to calculate the area under the curve (AUC) and optimal cut-off values
to evaluate the predictive ability of LL-PA for walking
independence.15)
RESULTS
Participant Characteristics
This study included 256 participants with a mean age of 74.09 ± 12.86 years. The
primary diagnoses were cerebrovascular disease (n=114, 44.5%), orthopedic
conditions (n=76, 29.7.0%), and disuse syndrome (n=66, 25.8%). Disease
distribution differed by sex, with men (n=146) showing a predominance of
cerebrovascular disease (n= 83, 56.8%), followed by disuse syndrome (n=41,
28.1%) and orthopedic conditions (n=22, 15.1%). In contrast, women (n=110)
showed a higher prevalence of orthopedic conditions (n= 54, 49.1%), followed by
cerebrovascular disease (n=31, 28.2%) and disuse syndrome (n=25, 22.7%). Among
patients with disuse syndrome, the underlying conditions in men included
COVID-19 (n=14), pneumonia (n=12), septic shock (n=4), post-cardiovascular
surgery (n=4), and others (n=7). In women, these conditions included pneumonia
(n=9), COVID-19 (n=6), septic shock (n=4), post-cardiovascular surgery (n=2),
and others (n=4).
Admission LL-PA values were significantly higher in men than in women (male, 3.8
± 0.9°; female, 3.2 ± 0.8°; P <0.001). The detailed baseline characteristics
of the patients are presented in Table
1.
Table 1. Baseline characteristics of study participants
| Characteristic |
Total (n=256) |
Male (n=146) |
Female (n=110) |
P value |
| Demographic data |
|
|
|
|
| Age, years |
74.09±12.86 |
70.42±13.48 |
78.85±10.84 |
<0.001 |
| MMSE |
26.22±4.24 |
26.47±4.44 |
25.90±3.96 |
0.752 |
| Primary diagnosis |
|
|
|
|
| Orthopedic conditions |
76 (29.7) |
22 (15.1) |
54 (49.1) |
<0.001 |
| Cerebrovascular disease |
114 (44.5) |
83 (56.8) |
31 (28.2) |
<0.001 |
| Disuse syndrome |
66 (25.8) |
41 (28.1) |
25 (22.7) |
0.042 |
| FIM scores at admission |
|
|
|
|
| Transfer |
3.84±1.11 |
3.89±1.25 |
3.77±0.89 |
0.382 |
| Walking |
1.67±1.42 |
1.75±1.53 |
1.56±1.27 |
0.298 |
| Stairs |
1.02±0.12 |
1.03±0.16 |
1.00±0.00 |
0.064 |
| FIM scores at discharge |
|
|
|
|
| Transfer |
6.37±0.98 |
6.53±0.94 |
5.89±1.40 |
<0.001 |
| Walking |
6.01±1.21 |
6.12±1.37 |
5.82±1.40 |
0.258 |
| Stairs |
5.49±1.37 |
5.65±1.37 |
5.29±1.36 |
<0.001 |
| Physical performance |
|
|
|
|
| Walking speed, m/s |
0.719±0.278 |
0.767±0.284 |
0.659±0.263 |
0.003 |
| TUG, s |
20.73±10.91 |
19.46±12.49 |
22.39±8.41 |
0.038 |
| Discharge walking speed, m/s |
0.987±0.29 |
1.031±0.30 |
0.932±0.28 |
0.009 |
| Discharge TUG, s |
13.63±8.67 |
13.28±10.39 |
14.08±5.44 |
0.047 |
| Clinical parameters |
|
|
|
|
| CONUT score |
2.53±1.71 |
2.59±1.81 |
2.45±1.57 |
0.298 |
| Length of stay, days |
81.15±29.02 |
83.23±30.01 |
78.49±27.43 |
0.215 |
Values are presented as mean ± standard deviation or number
(percentage).
SMI, Skeletal Muscle Mass Index.
P values were calculated using independent t-tests for
continuous variables and chi-square tests for categorical variables.
The nutritional status assessed by CONUT score showed varying distributions
across disease categories and for sex (Table 2). In male patients with cerebrovascular disease, orthopedic
conditions, and disuse syndrome, the majority showed normal to mild nutritional
impairment (CONUT score 0–4). Similarly, female patients predominantly exhibited
normal to mild nutritional impairment across all disease categories. Only one
male cerebrovascular disease patient showed severe nutritional impairment (CONUT
score ≥9).
Table 2. Distribution of CONUT scores by sex and disease category
| Sex |
Disease category |
Normal (0–1) |
Mild (2–4) |
Moderate (5–8) |
Severe (≥9) |
| Male |
|
|
|
|
|
Cerebrovascular disease (n=83) |
8 |
14 |
2 |
1 |
|
Orthopedic conditions (n=22) |
3 |
4 |
3 |
0 |
|
Disuse syndrome (n=41) |
6 |
8 |
4 |
0 |
| Female |
|
|
|
|
|
Cerebrovascular disease (n=31) |
4 |
11 |
1 |
0 |
|
Orthopedic conditions (n=54) |
6 |
12 |
2 |
0 |
|
Disuse syndrome (n=25) |
3 |
6 |
2 |
0 |
Values are presented as number of patients.
CONUT scores were categorized as normal (0–1), mild (2–4), moderate
(5–8), and severe (≥9).
Correlation analysis between admission LL-PA and functional measures revealed
significant associations (Table 3).
In men, LL-PA showed moderate positive correlations with motor FIM (r=0.429, P
<0.001), transfer (r=0.381, P <0.001), walking (r=0.414, P <0.001), and
stairs (r=0.496, P <0.001). In women, stronger correlations were observed,
particularly with motor FIM (r=0.659, P <0.001), and there were moderate
correlations with transfer (r=0.454, P <0.001), walking (r=0.478, P
<0.001), and stairs (r=0.474, P <0.001). Regarding physical function
assessments, significant positive correlations were found between LL-PA and 10-m
walking speed in both sexes, with females showing a stronger correlation (male,
r=0.573; female, r=0.712; P <0.001). In contrast, significant negative
correlations were observed for the TUG test (male, r=−0.347; female, r=−0.558; P
<0.001). The stronger correlations observed in women suggest that LL-PA might
be a more sensitive predictor of functional performance in female patients.
Table 3. Correlation coefficients between LL-PA and functional outcomes by
sex
| Variable |
Male (n=146) |
|
Female (n=110) |
| r |
P value |
|
r |
P value |
| Functional measures |
|
|
|
|
|
| Motor FIM |
0.429 |
<0.001 |
|
0.659 |
<0.001 |
| FIM transfer |
0.381 |
<0.001 |
|
0.454 |
<0.001 |
| FIM walking |
0.414 |
<0.001 |
|
0.478 |
<0.001 |
| FIM stairs |
0.496 |
<0.001 |
|
0.474 |
<0.001 |
| Physical performance |
|
|
|
|
|
| 10-m walking speed |
0.573 |
<0.001 |
|
0.712 |
<0.001 |
| TUG test |
−0.347 |
<0.001 |
|
−0.558 |
<0.001 |
| Clinical parameters |
|
|
|
|
|
| CONUT score |
−0.411 |
<0.001 |
|
−0.284 |
<0.03 |
| SMI |
0.551 |
<0.001 |
|
−0.133 |
0.164 |
Values are presented as Pearson correlation coefficients (r).
SMI, Skeletal Muscle Mass Index.
Multiple regression analysis was performed with the discharge motor FIM score as
the dependent variable. The results showed that LL-PA remained a significant
independent predictor (β=0.323, P <0.001) after adjusting for confounding
factors. Other significant predictors included MMSE (β=0.427, P <0.001),
cerebrovascular disease (β=−0.221, P <0.001), and disuse syndrome (β=−0.117,
P=0.034). The motor FIM score also showed a significant association (β=0.134,
P=0.01). Age, sex, and length of stay did not show significant associations with
the discharge motor FIM score (Table
4).
Table 4. Multiple regression analysis for predicting discharge motor FIM
score
| Variable |
B |
SE |
β |
t |
P value |
VIF |
| (Constant) |
45.07 |
7.118 |
- |
6.332 |
<0.001 |
- |
| Clinical measures |
|
|
|
|
|
|
| Motor FIM |
0.131 |
0.05 |
0.134 |
2.607 |
0.01 |
1.367 |
| MMSE |
1.055 |
0.121 |
0.427 |
8.731 |
<0.001 |
1.231 |
| Disease categories |
|
|
|
|
|
|
| Cerebrovascular disease |
−4.642 |
1.355 |
−0.221 |
−3.427 |
<0.001 |
2.133 |
| Disuse syndrome |
−2.797 |
1.311 |
−0.117 |
−2.134 |
0.034 |
1.548 |
| Other variables |
|
|
|
|
|
|
| Length of stay |
−0.003 |
0.019 |
−0.009 |
−0.179 |
0.858 |
1.386 |
| Age |
−0.062 |
0.051 |
−0.077 |
−1.224 |
0.222 |
2.046 |
| Sex |
0.707 |
1.065 |
0.034 |
0.664 |
0.507 |
1.309 |
| LL-PA |
3.418 |
0.753 |
0.323 |
4.539 |
<0.001 |
2.599 |
VIF, Variance Inflation Factor.
Sex was coded as male=1 and female=0. Disease categories were coded as
presence=1 and absence=0, with orthopedic conditions as references.
Model statistics: R2=0.519, adjusted R2=0.504,
F=33.334 (P <0.001), and Durbin-Watson=1.244.
Logistic regression analysis revealed that LL-PA was an independent predictor of
walking independence at discharge [odds ratio (OR)=3.302, 95% confidence
interval (CI): 1.714–6.360; P <0.001). Other significant predictors included
MMSE (OR=1.197, 95% CI: 1.096–1.307; P <0.001) and motor FIM (OR=1.067, 95%
CI: 1.011–1.126; P=0.017). ROC curve analysis showed good predictive ability for
walking independence in both sexes (male, AUC=0.820, 95% CI: 0.731–0.909;
female, AUC=0.835, 95% CI: 0.749–0.920). The optimal cut-off value was 3.050°
for men (sensitivity, 0.804; specificity, 0.853) and 2.650° for women
(sensitivity, 0.769; specificity, 0.812) (Table 5 and Fig. 2).
Table 5. Logistic regression analysis for predicting walking independence at
discharge
| Variable |
OR |
95% CI |
P value |
| Clinical measures |
|
|
|
| MMSE |
1.197 |
1.096–1.307 |
<0.001 |
| Motor FIM |
1.067 |
1.011–1.126 |
0.017 |
| Disease categories |
|
|
|
| Cerebrovascular disease |
0.394 |
0.136–1.142 |
0.086 |
| Disuse syndrome |
0.644 |
0.259–1.598 |
0.343 |
| Other variables |
|
|
|
| Length of stay |
1.008 |
0.992–1.023 |
0.321 |
| Sex |
1.380 |
0.637–2.988 |
0.414 |
| Age |
0.985 |
0.945–1.028 |
0.495 |
| LL-PA |
3.302 |
1.714–6.360 |
<0.001 |
The model was adjusted for age, sex, MMSE score, cerebrovascular disease,
disuse syndrome, length of hospital stay, motor FIM score, and LL-PA.
Sex was coded as male=1 and female=0. Disease categories were coded as
presence=1 and absence=0, with orthopedic conditions as references.
Walking independence was defined as FIM walking score ≥6.
DISCUSSION
In this study, we demonstrated that LL-PA level measured at admission is a useful
prognostic indicator of functional outcomes in patients admitted to convalescent
rehabilitation hospitals. The non-invasive and nature of BIA measurement and the
ability to implement BIA at the bedside offers practical advantages in the
rehabilitation setting. Notably, LL-PA remained an independent predictor after
adjusting for conventional prognostic factors, including age and admission motor FIM
scores.
The relationship between LL-PA and functional independence at discharge revealed
several significant findings. Multiple regression analysis showed that LL-PA was a
significant independent predictor of discharge motor FIM scores (β=0.323, P
<0.001), along with MMSE and disease categories. The timing of LL-PA measurement
within 4 weeks of admission was chosen to ensure stable medical conditions while
providing early prognostic information. Correlation analysis demonstrated
moderate-to-strong positive correlations between the LL-PA and FIM subscores,
particularly in transfer, walking, and stair climbing. These findings support those
of previous studies by Norman et al.3) and Yamada et al.4) regarding the relationship between PA and muscle
function.
Regarding walking ability, significant correlations were observed with the 10-m
walking speed and TUG test results. The correlation with the 10-m walking speed was
notable, showing moderate-to-strong positive correlations in both sexes (male,
r=0.573; female, r=0.712; P <0.001), consistent with the findings of Beaudart et
al.5)
Interestingly, the relationship between LL-PA and skeletal muscle mass index was
sex-dependent. A moderate positive correlation was observed in male participants
(r=0.551, P <0.001), whereas no significant correlation was observed in female
participants. This sex-specific difference in the relationship between LL-PA and
muscle mass suggests that LL-PA may provide information about muscle quality that is
distinct from quantitative measures of muscle mass. According to Lukaski et
al.,16) LL-PA
reflects not only muscle mass but also cellular integrity and muscle quality. This
finding emphasizes that LL-PA could serve as a unique indicator of muscle function,
particularly in rehabilitation settings where both quantitative and qualitative
aspects of muscle tissue may be relevant to functional recovery. The non-invasive
nature of BIA measurements further supports its practical utility in clinical
settings.
The relationship between LL-PA and nutritional status, as assessed by CONUT score,
revealed important sex-specific patterns. Male participants showed a moderate
negative correlation with CONUT scores (r=−0.411, P <0.001), whereas female
participants showed a weak negative correlation (r=−0.284, P <0.03). The varying
distributions of CONUT scores across disease categories and sex, with most patients
showing normal to mild nutritional impairment, suggests that LL-PA might provide
complementary information to traditional nutritional assessment tools. According to
Cruz-Jentoft et al.,17)
these sex differences may reflect variations in body composition and the
relationship between muscle function and nutritional status. This comprehensive
evaluation approach could enhance the accuracy of functional outcome predictions in
rehabilitation settings.
The clinical significance of our findings is particularly evident in predicting
walking independence. Logistic regression analysis revealed that LL-PA was an
independent predictor of walking independence at discharge (OR=3.302, 95% CI:
1.714–6.360; P <0.001), along with MMSE and motor FIM scores. This indicates that
for each degree rise in admission LL-PA, the likelihood of achieving walking
independence at discharge increases by approximately 3.3-fold. Furthermore, the
sex-specific cut-off values derived from ROC curve analysis (male, 3.3050°; female,
2.650°) demonstrated high sensitivity (male, 0.804; female, 0.769) and high
specificity (male, 0.853; female, 0.812), indicating practical clinical
applicability for early prognostic assessment.
Several limitations should be considered when interpreting the results of our study.
First, our study was limited to patients who could maintain a standing position for
BIA measurements, which might have excluded more severely impaired patients. This
requirement for standing measurement could have led to selection bias, potentially
limiting the generalizability of our findings to patients with severe mobility
impairments. Second, although early functional assessment is crucial for
rehabilitation planning, the timing of LL-PA measurements within 4 weeks of
admission was chosen to ensure medical stability and reliable measurements,
considering that patients’ physical conditions typically stabilize during this
period. Third, we did not examine the longitudinal changes in LL-PA during
rehabilitation or their relationship with functional recovery. Therefore, the effect
of rehabilitation interventions on LL-PA and subsequent functional outcomes remains
uncertain. Fourth, nutritional status was assessed using the CONUT score; however,
other nutritional indicators, as well as the effects of medications and
comorbidities, could not be fully evaluated. Finally, we focused on discharge
outcomes without long-term follow-up data, which limited our understanding of the
long-term prognostic value of LL-PA.
CONCLUSION
We demonstrated that LL-PA level measured at admission is a useful prognostic
indicator of functional outcomes in patients admitted to convalescent rehabilitation
hospitals. The non-invasive and straightforward nature of LL-PA measurement and its
predictive ability, particularly when considering sex-specific differences,
highlight its potential application in clinical practice. Although our findings are
promising, they should be interpreted within the context of our study limitations,
particularly regarding the requirement for standing measurements. LL-PA, when
combined with conventional assessment tools such as MMSE and motor FIM, may enhance
accurate prognosis prediction and individualized rehabilitation planning. Future
studies should investigate the longitudinal changes in LL-PA during rehabilitation
interventions and their relationship with long-term outcomes in diverse
rehabilitation settings.
ACKNOWLEDGMENTS
We express our sincere gratitude to all patients who participated in this study and
the rehabilitation staff who assisted with data collection. We also thank Editage
(www.editage.jp)for English language editing.
CONFLICTS OF INTEREST
The authors declare no conflict of interest.
REFERENCES
- 1. Mehrholz J, Thomas S, Elsner B: Treadmill
training and body weight support for walking after stroke. Cochrane Libr
2017;2017:CD002840. PMID:28815562,
DOI:10.1002/14651858.CD002840.pub4
- 2. Kyle U, Bosaeus I, De Lorenzo AD,
Deurenberg P, Elia M, Gómez JM, Heitmann BL, Kent-Smith L, Melchior JC, Pirlich
M, Scharfetter H, Schols AM, Pichard C, Composition of the ESPEN Working Group:
Bioelectrical impedance analysis—part I: review of principles and methods. Clin
Nutr 2004;23:1226–1243. PMID:15380917,
DOI:10.1016/j.clnu.2004.06.004
- 3. Norman K, Stobäus N, Pirlich M,
Bosy-Westphal A: Bioelectrical phase angle and impedance vector
analysis—clinical relevance and applicability of impedance parameters. Clin Nutr
2012;31:854–861. PMID:22698802, DOI:10.1016/j.clnu.2012.05.008
- 4. Yamada M, Kimura Y, Ishiyama D, Nishio N,
Otobe Y, Tanaka T, Ohji S, Koyama S, Sato A, Suzuki M, Ogawa H, Ichikawa T, Ito
D, Arai H: Phase angle is a useful indicator for muscle function in older
adults. J Nutr Health Aging 2019;23:251–255. PMID:30820513,
DOI:10.1007/s12603-018-1151-0
- 5. Beaudart C, Rizzoli R, Bruyère O,
Reginster JY, Biver E: Sarcopenia: burden and challenges for public health. Arch
Public Health 2014;72:45. PMID:25810912,
DOI:10.1186/2049-3258-72-45
- 6. Janssen I, Heymsfield SB, Wang ZM, Ross
R: Skeletal muscle mass and distribution in 468 male and female aged 18–88 yr. J
Appl Physiol 2000;89:81–88.
- 7. Bohannon RW: Reference values for the
timed up and go test: a descriptive meta-analysis. J Geriatr Phys Ther
2006;29:64–68. PMID:16914068,
DOI:10.1519/00139143-200608000-00004
- 8. Ottenbacher KJ, Hsu Y, Granger CV,
Fiedler RC: The reliability of the functional independence measure: a
quantitative review. Arch Phys Med Rehabil 1996;77:1226–1232. PMID:8976303,
DOI:10.1016/S0003-9993(96)90184-7
- 9. Podsiadlo D, Richardson S: The timed “Up
& Go”: a test of basic functional mobility for frail elderly persons. J Am
Geriatr Soc 1991;39:142–148. PMID:1991946,
DOI:10.1111/j.1532-5415.1991.tb01616.x
- 10. Ignacio de Ulíbarri J, González-Madroño
A, de Villar NG, González P, González B, Mancha A, Rodríguez F, Fernández G:
CONUT: a tool for controlling nutritional status. First validation in a hospital
population. Nutr Hosp 2005;20:38–45. PMID:15762418
- 11. Folstein MF, Folstein SE, McHugh PR:
“Mini-mental state”. A practical method for grading the cognitive state of
patients for the clinician. J Psychiatr Res 1975;12:189–198. PMID:1202204,
DOI:10.1016/0022-3956(75)90026-6
- 12. Mukaka MM: Statistics corner: a guide to
appropriate use of correlation coefficient in medical research. Malawi Med J
2012;24:69–71. PMID:23638278
- 13. Cohen J, Cohen P, West SG, Aiken LS: Applied
multiple regression/correlation analysis for the behavioral sciences. 3rd ed.
Mahwah: Lawrence Erlbaum Associates; 2003.
- 14. Hosmer DW, Jr, Lemeshow S, Sturdivant RX:
Applied logistic regression. 3rd ed. Hoboken: Wiley; 2013.
- 15. Hanley JA, McNeil BJ: The meaning and use
of the area under a receiver operating characteristic (ROC) curve. Radiology
1982;143:29–36. PMID:7063747,
DOI:10.1148/radiology.143.1.7063747
- 16. Lukaski HC, Kyle UG, Kondrup J:
Assessment of adult malnutrition and prognosis with bioelectrical impedance
analysis: phase angle and impedance ratio. Curr Opin Clin Nutr Metab Care
2017;20:330–339. PMID:28548972,
DOI:10.1097/MCO.0000000000000387
- 17. Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie
Y, Bruyère O, Cederholm T, Cooper C, Landi F, Rolland Y, Sayer AA, Schneider SM,
Sieber CC, Topinkova E, Vandewoude M, Visser M, Zamboni M, Writing Group for the
European Working Group on Sarcopenia in Older People 2 (EWGSOP2), and the
Extended Group for EWGSOP2: Sarcopenia: revised European consensus on definition
and diagnosis. Age Ageing 2019;48:16–31. PMID:30312372,
DOI:10.1093/ageing/afy169
