Journal of the Neurological Sciences 347 (2014) 310–315

Contents lists available at ScienceDirect

Journal of the Neurological Sciences

journal homepage: www.elsevier.com/locate/jns

Influencing effect of non-motor symptom clusters on quality of life in

Parkinson's disease
Sung Reul Kim a, Hui Young So b, Eunju Choi b, Jeong Hee Kang a, Hye Young Kim a, Sun Ju Chung c,⁎
a
College of Nursing, Chonbuk National University, Republic of Korea
b
Department of Nursing, Asan Medical Center, Republic of Korea
c
Department of Neurology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: The heterogeneity of non-motor symptoms (NMSs) in patients with Parkinson's disease (PD) has been well
Received 30 May 2014 established. We investigated the effects of NMS as a cluster on the quality of life (QoL) of patients with PD. We
Received in revised form 1 October 2014 recruited 180 patients with PD and used a descriptive cross-sectional study design. To determine interrelation-
Accepted 20 October 2014
ships between non-motor symptoms, a principal component analysis with varimax rotation was performed
Available online 23 October 2014
based on the Non-Motor Symptoms Scale (NMSS). Among 180 PD patients, 172 patients (96.6%) had experienced
Keywords:
at least 2 domains of concurrent non-motor symptoms. There were two types of non-motor symptom clusters
Parkinson's disease (NMSCs). The first non-motor symptom cluster (NMSC1) consisted of mood, sleep/fatigue, attention/memory,
Non-motor symptom urinary symptoms, and miscellaneous symptoms, while the second non-motor symptom cluster (NMSC2)
Symptom cluster consisted of perceptual problems, gastrointestinal issues, and cardiovascular symptoms. The elderly PD patients
Quality of life were more often categorized as experiencing NMSC2 than NMSC1. Our subgroup cluster analysis showed that PD
Factor analysis patients with higher scoring NMS had significantly poorer QoL in both NMSC1 and NMSC2 subgroups, with
Cluster analysis subgroup-specific patterns. NMSCs also emerged differently depending on sex and the severity of PD. In conclu-
sion, PD patients with NMS may have a specific cluster pattern of NMSC. Some NMSCs may have a negative im-
pact on QoL. Understanding the clinical implications of NMSC in PD patients may provide better therapeutic
interventions.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction
Parkinson's disease (PD) is a chronic progressive neurodegenerative
disorder characterized by dopaminergic neuronal loss in the substantia
nigra and other brain regions [1]. Patients with PD not only experience
motor symptoms including tremor, rigidity, bradykinesia, and postural
instability, but also non-motor symptoms (NMSs) including anxiety,
depression, sleep disturbance, and fatigue [2].
Non-motor symptoms of PD have a substantial negative impact on
quality of life (QoL) [3–6]. Most PD patients experience NMS and suffer
from a higher frequency of NMS as the disease progresses [7]. The large-
scale multicenter PRIAMO study showed that 98.6% of PD patients expe-
rienced NMS, with a mean number of 7.8 NMS per patient (range, 0–32)
[8].
Previous studies of NMS in PD have focused mainly on single or iso-
lated NMS such as dementia, mood, fatigue, sleep disturbance, and pain
[9–14]. Although this approach has led to a better understanding of each
⁎ Corresponding author at: Department of Neurology, Asan Medical Center, University
of Ulsan College of Medicine, 88, Olympic-ro 43-gil, Songpa-gu, Seoul 138-736, Republic
of Korea. Tel.: +82 2 3010 3440; fax: +82 2 474 4691.
E-mail address: sjchung@amc.seoul.kr (S.J. Chung).
NMS, it has major limitations because most PD patients have multiple
and various NMSs.
A ‘symptom cluster’ has been defined as a group of more than two
symptoms that occur concurrently and are interrelated to each other;
they may or may not have a common etiology [15]. Each symptom in
the cluster is independent, but strongly interrelated, and one symptom
can impact another symptom through its effect on a third symptom
[15–18].
Most studies thus far have focused on a single NMS in an attempt to
identify the NMS that has the biggest impact on QoL or to assess its
effect on QoL depending on its severity. However, little information is
available on which NMSs form clusters, how clusters form depending
on patient condition, and the impact of formed clusters on patient's QoL.
Understanding the non-motor symptom cluster of PD patients can
be beneficial for the following reasons. First, it can contribute to identi-
fying the significant symptoms that patients experience but do not
report. Second, it can anticipate potential concomitant symptoms that
may be manifested in the future, thereby providing effective patient
management methods to be used over the course of time. Therefore,
investigation of this complex effect of simultaneously occurring NMS
in PD is essential for the successful management of this disease
and the improvement of QoL in PD patients [19,20]. Therefore, we

http://dx.doi.org/10.1016/j.jns.2014.10.032
0022-510X/© 2014 Elsevier B.V. All rights reserved.
 S.R. Kim et al. / Journal of the Neurological Sciences 347 (2014) 310–315 311

investigated the comprehensive features of non-motor symptom test was used to analyze differences in the frequency of non-motor
cluster (NMSC) and their effects on QoL in PD patients. symptoms across gender. To examine whether an interrelationship
between symptoms existed, a principal component analysis (PCA)
2. Patients and methods with varimax rotation was carried out for NMSS domains. In addition
to the Kaiser–Meyer–Olkin measure and Bartlett's test of sphericity,
2.1. Study design and patients inter-item correlation coefficients were examined to ensure the ade-
quacy of data for factor analysis. Cronbach's alpha was used to estimate
Descriptive cross-sectional design was used. We recruited 180 PD the internal consistency and reliability of the derived clusters.
patients between January 1 and April 30, 2013 from Asan Medical To determine subgroups of NMSC, we used a hierarchical cluster
Center, Seoul, Korea. Inclusion criteria for the present study were: analysis. Patient subgroups were stratified based on the average link-
(1) patients must have been diagnosed with PD based on the United ages between groups and the distance between factor loadings of
Kingdom Parkinson's Disease Society Brain Bank criteria [21]; respondents, which was calculated using squared Euclidian distance
(2) 20 years of age or older; and (3) no past history of other major as a measure of similarity. Proximity between patients was measured
health problems that could potentially influence the NMS, such as active using Ward's method, which is sensitive to outliers and effective
cancer and/or chronic renal disease. We also excluded patients when identifying clusters as opposed to other intergroup proximity
with atypical Parkinsonism, Parkinson-plus syndrome, or secondary measures.
Parkinsonism. Two subgroups were identified to be the optimal number of groups.
All PD patients agreed to a face-to-face interview using the Non- The subgroups of patients were compared in relation to demographic
Motor Symptoms Scale (NMSS) [7], supplemented when appropriate and clinical characteristics, non-motor symptom scores, and QoL scores
with information from a spouse or a family member. After the interview using chi-square and the Mann–Whitney U test. In addition, we also
process, we also reviewed the medical records of the patients. used the Kolmogorov–Smirnov test to analyze the normality of non-
The Institutional Review Board (IRB) of Asan Medical Center motor symptom cluster scores. A two-tailed P b 0.05 was considered
approved this study, and all subjects were required to provide in- statistically significant.
formed consent, in compliance with IRB regulation. The subjects
were given the option to voluntarily withdraw their informed
consent and their personal data was kept strictly confidential 3. Results
throughout the study.
3.1. Characteristics of subjects
2.2. Clinical measurement
The demographic and clinical characteristics of the study subjects
2.2.1. Non-Motor Symptoms Scale (NMSS) are summarized in Table 1. Of the 180 PD patients, there were 74 men
NMS was measured using the Non-Motor Symptoms Scale (NMSS) and 106 women whose mean age was 62.6 ± 10.2 years (range,
[7]. The NMSS is one of the most commonly used scales for measuring 28–81 years). The mean age at onset of PD was 55.6 ± 11.6 years
NMS, which was rated by the health professionals and obtained through (range, 22–80 years) and the mean disease duration was 7.2 ±
interview. The NMSS consists of 30 items in the following nine domains: 5.7 years (range, 0.5–24.0 years). The median Hoehn and Yahr stage
cardiovascular, sleep/fatigue, mood/cognition, perceptual problems/ was 2 (interquartile range, 2–3).
hallucinations, attention/memory, gastrointestinal tract, urinary, sexual
function, and miscellaneous. The miscellaneous domain included pain,
taste or smell, weight change, and excessive sweating. The score for
each item is based on multiplying a severity score (from 0 to 3) with a Table 1
Demographic & clinical characteristics (N = 180).
frequency score (from 1 to 4). The severity and frequency scores of 1
or higher were considered to be the presence of non-motor symptoms. Variables Mean (±SD) or N (%) Min–max
The NMSS was used with the permission of the developer. Sex, male 74 (41.1%)
Age, y 62.57 (±10.16) 28–81
2.2.2. Parkinson's Disease Questionnaire-39 (PDQ-39) Age at onset, y 55.60 (±11.64) 22–80
Quality of life was measured by Parkinson's Disease Questionnaire- Disease duration, y 7.16 (±5.68) 0.5–24
Median Schwab and England ADL, % 70 10–100
39 (PDQ-39) with the permission of the developer [22]. The PDQ-39
Duration of L-dopa treatment, year 6.73 (±5.69) 0–23
consists of 39 items within the following eight domains: mobility, L-Dopa equivalent dose, mg 652.20 (±384.28) 0–2038
activities of daily living, emotional well-being, stigma, social support, Median Hoehn and Yahr 2 0–5 (IQR 2–3)
cognition, communication, and bodily discomfort. The scores of PDQ- 0 2 (1.1%)
39 range from 0 to 100 and higher scores reflect poorer QoL. 1 18 (10%)
2 77 (42.8%)
3 67 (37.3%)
2.2.3. Other clinical characteristics 4 15 (8.3%)
In addition to the measurements described above, we obtained 5 1 (0.6%)
other clinical information about the patients, including age at onset of Motor fluctuation
PD, duration of PD, duration of levodopa therapy, daily levodopa equiv- Yes, disabling 47 (26.2%)
Yes, but not disabling 26 (14.4%)
alent dose (LED), presence of motor fluctuation and dyskinesia, the
No 107 (59.4%)
Hoehn and Yahr stage, and Schwab and England Activities of Daily Dyskinesia
Living Score. Yes, disabling 41 (22.8%)
Yes, but not disabling 31 (17.2%)
No 108 (60.0%)
2.3. Statistical analysis
Use of medication
Anti-depressants 14 (7.7%)
We used a descriptive, cross-sectional design and statistical analysis Anxiolytics 28 (15.5%)
was performed with SPSS version 20.0 (IBM SPSS Statistics, SPSS Inc., Antipsychotics 18 (10.0%)
Chicago, IL). All data were expressed as a percentage, mean ± standard Sleeping pills 51 (28.3%)

deviation (SD), median (interquartile range), and range. The chi-square SD — standard deviation, IQR — interquartile range
312 S.R. Kim et al. / Journal of the Neurological Sciences 347 (2014) 310–315
3.2. Characteristics of NMS
Among the 180 PD patients, 178 patients (98.8%) experienced NMS
and 172 patients (96.56%) reported at least 2 domains of concurrent
NMS. The average number of concurrent NMS was 5.1 ± 2.1 (range,
0–9). Most frequent were symptoms related to mood/cognition
(81.7%), followed by sleep/fatigue (80.0%), miscellaneous (72.8%), and
urinary symptoms (71.7%). The frequency of sexual dysfunction was
significantly different between males (36.5%) and females (8.5%)
(P b 0.001) (Table 2, Supplementary Fig. 1).
3.3. Non-Motor Symptom Clusters (NMSC)
To ensure the adequacy of data for factor analysis, we examined
inter-item correlation coefficients by Spearman rank correlation.
Among the nine NMSS domains, sexual function (domain 8) was
excluded because the Spearman rank correlation coefficient value was
lower than 0.30, based on the guidelines for factor analysis in our symp-
tom cluster study [23]. Principal component analysis extracted compo-
nents with Eigenvalues greater than 0.90. Two NMSCs were identified
and accounted for 54.5% of total variance (Table 3, Fig. 1A).
NMSC1 consisted of sleep/fatigue, miscellaneous, mood/cognition,
attention/memory and urinary symptoms and accounted for 43.1% of
the total variance. NMSC2 consisted of perceptual problems, gastroin-
testinal issues, and cardiovascular symptoms, and accounted for 11.4%
of the total variance. NMSC emerged differently according to the sex
of patients (Supplementary Table 1, Fig. 1B, C). In male patients,
NMSC1 consisted of perceptual problems, attention/memory, sleep/
fatigue, urinary and mood/cognition-related symptoms, while NMSC2
consisted of miscellaneous, cardiovascular irregularities and gastroin-
testinal symptoms. In females, NMSC1 consisted of sleep/fatigue, mis-
cellaneous, attention/memory, urinary and mood-related symptoms,
while NMSC2 consisted of perceptual problems, gastrointestinal issues
and cardiovascular symptoms.
In addition, we analyzed NMSC according to the severity of the
symptoms. In Hoehn and Yahr stages 0–2, NMSC1 consisted of mood/
cognition, cardiovascular, sleep/fatigue and attention/memory, while
NMSC2 consisted of perceptual problems, gastrointestinal, urinary and
miscellaneous. In Hoehn and Yahr stages 3–5, NMSC1 consisted of
sleep/fatigue, attention/memory, miscellaneous, gastrointestinal, car-
diovascular and urinary, while NMSC2 consisted of perceptual problems
and mood/cognition (Table 3).
3.4. Subgroup analyses based on NMSC
Based on two NMSCs, we identified two patient subgroups by cluster
analysis. Cluster analysis is a grouping method that suggests the number
of groups into which a set of samples can be classified, with the similar-
ity measure calculated by the squared Euclidean distance. The present
study was determined to have two groups. The group with the higher
Table 2
Severity and prevalence of non-motor symptoms (N =180).
Symptoms Severity
Mean (± SD)
Cardiovascular 1.29 (±2.44)
Sleep/fatigue 5.44 (±5.92)
Mood/cognition 8.77 (±11.05)
Perceptual problems 1.08 (±2.55)
Attention/memory 4.06 (±5.51)
Gastrointestinal 3.73 (±4.73)
Urinary 7.12 (±7.74)
Sexual dysfunction 1.48 (±3.66)
Miscellaneous 5.34 (±6.02)
SD — standard deviation
score was called the higher-symptom group, and the group with the
lower score was called the low-symptom group.
In MNSC1, there were 26 patients (higher-symptom group) with a
high NMSS score and 154 patients (lower-symptom group) with a low
NMSS score. In NMSC2, there were 38 patients (higher-symptom
group) with a high NMSS score and 142 patients (lower-symptom
group) with a low NMSS score.
The demographic and clinical characteristics between these two
subgroups, according to NMSCs, are presented in Table 4.
In NMSC1, the higher-symptom group had generally experienced
longer PD duration compared to the lower-symptom group (P =
0.002). The Schwab and England Activities of Daily Living Scores were
significantly lower in the higher-symptom group, compared to the
lower-symptom group (P = 0.003). The duration of levodopa therapy
was significantly longer in the subgroup with higher NMSS scores com-
pared to those with lower NMSS scores (P = 0.002). The levodopa
equivalent dose was significantly higher in the higher-symptom group
compared to the lower-symptom group (P = 0.025). In addition, the
Hoehn and Yahr stage was higher in the higher-symptom group com-
pared to the lower-symptom group (P = 0.009). Differences in QoL
scores between the two subgroups within NMSC1 are presented in
Table 4. The higher-symptom group of NMSC1 had significantly poorer
QoL in all domains than the lower-symptom group.
In NMSC2, the mean age (P = 0.019) and the levodopa dose (P =
0.033) were significantly higher in the higher-symptom group, com-
pared to the lower-symptom group. The duration of PD (P = 0.019)
and the duration of levodopa therapy (P = 0.013) were significantly
longer in the higher-symptom group compared to the lower-symptom
group. The Schwab and England Activities of Daily Living Scores were
significantly lower in the higher-symptom group compared to the
lower-symptom group (P = 0.037). However, there were no significant
differences in Hoehn and Yahr stage and the presence of motor fluctua-
tion and dyskinesia between the higher-symptom group and the lower-
symptom group. The higher-symptom group within NMSC2 had signif-
icantly poorer QoL in all domains, except for stigma and bodily
discomfort.
The sub-analyses were performed according to Hoehn and Yahr
stage (0–2 vs. 3–5) (Supplementary Tables 2 and 3). In NMSC1, the
higher-symptom group with Hoehn and Yahr stage ≤ 2 showed longer
PD duration and longer duration of levodopa therapy compared to the
lower-symptom group (P = 0.018 and P = 0.015, respectively). They
also showed significantly poorer QoL in all domains, except for stigma
and communication, than those with lower NMSS scores. In NMSC2,
NMSS scores of PD patients with Hoehn and Yahr stage ≤ 2 were corre-
lated with QoL in activities of daily living, stigma, and cognition do-
mains, but they were not correlated with QoL in mobility, emotional
well-being, social support, and communication domains.
In the subgroup of PD patients with Hoehn and Yahr stage ≥ 3 and
higher NMSS scores, Schwab and England Activities of Daily Living
Scores were significantly lower compared with those with lower
Prevalence of non-motor symptoms
Total n (%) Male n (%) Female n (%) P value
(n = 180) (n = 74) (n = 106)
67 (37.2%) 28 (37.8%) 39 (36.8%) 1.000
144 (80.0%) 61 (82.4%) 83 (78.3%) 0.572
147 (81.7%) 65 (87.8%) 82 (77.4%) 0.074
43 (23.9%) 13 (17.6%) 30 (28.3%) 0.112
106 (58.9%) 42 (56.8%) 64 (60.4%) 0.647
115 (63.9%) 49 (66.2%) 66 (62.3%) 0.638
129 (71.7%) 56 (75.7%) 73 (68.9%) 0.401
36 (20.0%) 27 (36.5%) 9 (8.5%) b0.001
131 (72.8%) 53 (71.6%) 78 (73.6%) 0.865
 S.R. Kim et al. / Journal of the Neurological Sciences 347 (2014) 310–315 313

Table 3
Factor loadings from the principal component analysis of non-motor symptoms (N = 180).

Non-motor symptom Total (n = 180) HY (0–2) (n = 100) HY (3–5) (n = 80)

Factor loadings Non-motor symptom Factor loadings Non-motor symptom Factor loadings

Component 1 Component 2 Component 1 Component 2 Component 1 Component 2

Sleep/fatigue 0.756 Mood/cognition 0.866 Sleep/fatigue 0.822

Miscellaneous 0.753 Cardiovascular 0.722 Attention/memory 0.711
Mood/cognition 0.691 Sleep/fatigue 0.665 Miscellaneous 0.664
Attention/memory 0.629 Attention/memory 0.629 Gastrointestinal 0.612
Urinary 0.511 Perceptual problems 0.870 Cardiovascular 0.552
Perceptual problems 0.841 Gastrointestinal 0.758 Urinary 0.467
Gastrointestinal 0.677 Urinary 0.604 Perceptual problems 0.888
Cardiovascular 0.501 Miscellaneous 0.333 Mood/cognition 0.570
Cronbach's α 0.861 0.637 Cronbach's α 0.659 0.603 Cronbach's α 0.715 0.228
Eigenvalues 3.450 0.910 Eigenvalues 3.549 1.030 Eigenvalues 3.098 1.027
Explained variance 2.649 1.711 Explained variance 2.433 2.147 Explained variance 2.686 1.438
Explained % 43.12% 11.38% Explained % 44.37% 12.88% Explained % 38.72% 12.83%
Cumulative % 43.12% 54.50% Cumulative % 44.37% 57.24% Cumulative % 38.72% 51.55%

NMSS scores in NMSC1 (P b 0.001). They had significantly poorer QoL in
all domains than those with lower NMSS scores. In PD patients with
Hoehn and Yahr stage ≥ 3, the higher NMSS scores of NMSC1 were as-
sociated with motor fluctuation (P = 0.022), while the higher NMSS
scores of NMSC2 were correlated with poorer QoL in all domains, except
stigma.
4. Discussion
The present study suggests that PD patients may have a specific
NMSC and NMSCs have a negative effect to their QoL. Our results have
clinical implications, as we suggest that NMS of PD patients should be
assessed and managed based on the clusters of complex NMS in PD.
Our results showed that the frequency of NMS in PD was 98.8%.
Consistent with previous reports [8], 96.6% (n = 172) of PD patients
in the study reported symptoms in at least two domains of NMS, with
an average number of 5.1 per PD patient. The frequencies of NMS in
each domain in the present study were higher than those in previous
studies [5,8]. The frequency of mood-related symptoms was as high
as 81.7%, although a small proportion of subjects were known to be tak-
ing anti-depressants (7.7%) or anxiolytics (15.5%) during the study.
These results indicate that mood-related symptoms might be largely

Fig. 1. Non-motor symptom cluster distribution: Component plots in rotated space. NMSC — non-motor symptom cluster, A — total, B — male, C — female.
314 S.R. Kim et al. / Journal of the Neurological Sciences 347 (2014) 310–315

Table 4
Comparison of clinical characteristics and quality of life between higher-symptom and lower-symptom groups according to symptom cluster.

Non-motor symptom cluster 1 Non-motor symptom cluster 2

Higher-symptom group Lower-symptom group P Higher-symptom group Lower-symptom group P

(n = 26) (n = 154) (n = 38) (n = 142)

Clinical characteristics
Sex, male 8 (30.8%) 66 (42.9%) 0.173 16 (42.1%) 58 (40.8%) 0.888
Age, y 65.27 ± 9.91 62.11 ± 10.16 0.070 66.18 ± 7.61 61.60 ± 10.66 0.019
Age at onset, y 56.27 ± 9.78 55.49 ± 11.95 0.711 56.74 ± 10.18 55.30 ± 12.01 0.522
Disease duration, y 9.96 ± 5.26 6.69 ± 5.62 0.002 9.36 ± 6.64 6.57 ± 5.26 0.019
Schwab and England ADL, % 59.60 ± 22.45 73.40 ± 14.78 0.003 64.74 ± 20.89 73.29 ± 14.96 0.037
Duration of L-dopa treatment, y 9.58 ± 5.25 6.25 ± 5.63 0.002 9.00 ± 6.52 6.12 ± 5.31 0.013
L-dopa equivalent dose, mg 848.44 ± 493.30 618.81 ± 353.83 0.025 767.43 ± 418.10 621.15 ± 370.30 0.033
Hoehn and Yahr stage, median 3 (IQR 2–4) 2 (IQR 2–3) 3 (IQR 2–3) 2 (IQR 2–3)
0 0 2 0.009 0 2 0.143
1 2 16 3 15
2 7 70 12 65
3 10 57 16 51
4 6 9 6 9
5 1 0 1 0
Motor fluctuation 0.055 0.234
Yes, disabling 10 97 14 33
Yes, but not disabling 5 21 5 21
No 11 36 19 88
Dyskinesia 0.131 0.527
Yes, disabling 9 32 11 30
Yes, but not disabling 6 25 7 24
No 11 97 20 88

Non-motor symptom clusters

Non-motor symptom cluster 1 79.57 ± 20.08 22.47 ± 15.87 b0.001 52.73 ± 30.75 24.83 ± 21.11 b0.001
Non-motor symptom cluster 2 15.07 ± 11.64 4.58 ± 5.34 b0.001 16.55 ± 8.90 3.30 ± 3.78 b0.001

Quality of life
Mobility 65.29 ± 27.10 30.25 ± 26.86 b0.001 51.12 ± 33.69 31.08 ± 26.92 0.001
Activities of daily living 64.10 ± 25.82 26.13 ± 24.00 b0.001 48.03 ± 31.95 27.23 ± 24.72 0.001
Emotional well-being 62.98 ± 21.55 27.14 ± 19.72 b0.001 43.31 ± 26.84 29.37 ± 21.83 0.001
Stigma 40.87 ± 27.74 26.38 ± 22.20 0.003 33.06 ± 22.18 27.24 ± 23.84 0.177
Social support 44.23 ± 27.57 15.20 ± 22.20 b0.001 29.60 ± 30.55 16.67 ± 22.63 0.019
Cognition 45.43 ± 16.91 23.62 ± 18.65 b0.001 38.65 ± 19.32 23.59 ± 18.90 b0.001
Communication 44.55 ± 31.44 16.94 ± 21.68 b0.001 33.55 ± 25.37 17.55 ± 24.13 b0.001
Bodily discomfort 49.36 ± 23.32 26.03 ± 20.90 b0.001 34.87 ± 27.93 27.93 ± 21.18 0.151
PDQ39-SI 55.77 ± 17.47 25.63 ± 16.10 b0.001 42.05 ± 21.91 26.76 ± 17.41 b0.001

IQR—interquartile range.
Non-motor symptom cluster 1; sleep/fatigue, miscellaneous, mood/cognition, attention/memory and urinary.
Non-motor symptom cluster 2; perceptual problems, gastrointestinal and cardiovascular.

underestimated in a clinical practice, suggesting that mood-related
symptoms should be carefully examined and managed.
Interestingly, the frequency of sexual dysfunction was significantly
greater in male (36.5%) than in female (8.5%) patients. This finding
was consistent with previous results [24–26] and must be taken into
account in a clinical setting.
Previous studies reported that the frequency of NMS increased along
with the disease duration and severity of PD [8,27]. Also, medications
used in PD patients can trigger, worsen or even be the primary cause
of NMS [7,28]. In the present study, the scores of NMSC are correlated
not only with the duration of PD, but also with the duration of levodopa
treatment and levodopa equivalent dose, suggesting a possible role of
anti-PD medications in the development of NMS.
The elderly PD patients were more often categorized as experiencing
NMSC2 than NMSC1. The disease duration of higher-symptom group
was not significantly different between patients with NMSC1 and
those with NMSC2 (9.96 years vs. 9.36 years). The Hoehn and Yahr
stage was associated with NMSC1 scores. These findings suggested
that each NMSC is associated with specific features of NMS in PD
patients, reflecting the complex features of NMSC.
In the current study, perceptual problems did not cluster with mood,
sleep/fatigue, and attention/memory. These findings are not in keeping
with previous studies which have demonstrated that perceptual
problems, sleep impairment, depression, and cognitive disturbances
are interconnected in PD [29]. In addition, a sub-analysis of the
PRIAMO study demonstrated that psychosis in early stages of PD is asso-
ciated with cognitive decline and depression at 2-year follow-up [30].
The discrepancy between the current study and previous reports may
be due to the difference of questionnaire structures for perceptual prob-
lems, which might lead to underestimation of these symptoms. The
NMSS questionnaire may be not sensitive enough to detect mild percep-
tual problems because there is just one question about hallucination,
one question about delusion, and one question about diplopia. Howev-
er, in the present study, the sub-analyses according to the severity of
motor symptoms showed that the perceptual problem was clustered
with mood in PD patients with Hoehn and Yahr stage ≥ 3, but not in
those with Hoehn and Yahr stage ≤ 2. These findings suggest that the
features of NMSC may change over the longitudinal course of PD.
In previous studies, domains of cardiovascular and perceptual prob-
lems in NMSS were not correlated with QoL [5,6]. However, our results
showed that cardiovascular symptoms and perceptual problems influ-
enced QoL through the formation of a NMSC with gastrointestinal
symptoms. These results suggest that the effects of NMS on QoL related
to PD need to be examined as a cluster rather than individually.
In the present study, the higher-symptom group in NMSC1 had
significantly poorer QoL in all domains. The higher-symptom group
in NMSC2 also had poorer QoL, except for in relation to stigma and
bodily discomfort. These findings suggest that sleep/fatigue, miscella-
neous symptoms, mood/cognition, attention/memory, and urinary
symptoms can affect all domains of QoL, while perceptual problems,
 S.R. Kim et al. / Journal of the Neurological Sciences 347 (2014) 310–315
gastrointestinal, and cardiovascular symptoms can affect QoL associated
with mobility, activities of daily living, emotional wellbeing, social
support, cognition, and communication.
The present study showed sex-specific patterns of NMSC, although
the number of patients in each gender group was too small to analyze
using principal component analysis. These sex-specific NMSC features
may be different among different cultures and regions and need to be
further investigated.
This study has some limitations. First, it was conducted in a single,
tertiary referral hospital, where PD patients with severe and complex
symptoms tend to visit. This may have led to the higher frequency of
observed NMS. Second, this study was conducted as a descriptive,
cross-sectional study and did not identify longitudinal changes in
NMSC associated with PD. Finally, our study could not address patho-
physiology, progression, and outcome of NMSC. Further studies are
required to explore these issues.
5. Conclusions
PD patients with NMS may have a specific cluster pattern of NMSC.
Some NMSCs may have a negative impact on QoL. Understanding the
clinical implications of NMSC in PD patients may provide better thera-
peutic interventions.
Conflict of interest
The authors declare that they have no competing interests.
Acknowledgments
This paper was supported by research funds of Chonbuk National
University in 2012.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jns.2014.10.032.
References
[1] Fearnley JM, Lees AJ. Ageing and Parkinson's disease: substantia nigra regional
selectivity. Brain 1991;114:2283–301.
[2] Adler CH. Nonmotor complication in Parkinson's disease. Mov Disord 2005;20:23–9.
[3] Rahman S, Griffin HJ, Quinn NP, Jahanshahi M. Quality of life in Parkinson's disease:
the relative importance of the symptoms. Mov Disord 2008;23:1428–34.
[4] Gallagher DA, Lees AJ, Schrag A. What are the most important nonmotor symptoms
in patients with Parkinson's disease and are we missing them? Mov Disord 2010;
25:2493–500.
[5] Martinez-Martin P, Rodriguez-Blazquez C, Kurtis MM, Chaudhuri KR. On behalf of
the NMSS validation group. The impact of non-motor symptoms on health-related
quality of life of patients with Parkinson's disease. Mov Disord 2011;26:399–406.
315
[6] Li H, Zhang M, Chen L, Zhang J, Pei Z, Hu A, et al. Nonmotor symptoms are indepen-
dently associated with impaired health-related quality of life in Chinese patients
with Parkinson's disease. Mov Disord 2010;25:2740–6.
[7] Chaudhuri KR, Martinez-Martin P, Brown RG, Sethi K, Stocchi F, Odin P, et al. The
metric properties of a novel non-motor symptoms scale for Parkinson's disease:
results from an international pilot study. Mov Disord 2007;22:1901–11.
[8] Barone P, Antonini A, Colosimo C, Marconi R, Morgante L, Avarello TP, et al. The
priamo study: a multicenter assessment of nonmotor symptoms and their impact
on quality of life in Parkinson's disease. Mov Disord 2009;24:1641–9.
[9] Duncan GW, Firbank MJ, O'Brien JT, Burn DJ. Magnetic resonance imaging: a
biomarker for cognitive impairment in Parkinson's disease? Mov Disord 2013;28:
425–38.
[10] Horvath J, Herrmann FR, Burkhard PR, Bouras C, Kövari E. Neuropathology of demen-
tia in a large cohort of patients with Parkinson's disease. Parkinsonism Relat Disord
2013;19:864–8.
[11] Schrag A, Barone P, Brown RG, Leentjens AF, McDonald WM, Starkstein S, et al.
Depression rating scales in Parkinson's disease: critique and recommendations.
Mov Disord 2007;22:1077–92.
[12] Friedman JH, Alves G, Hagell P, Marinus J, Marsh L, Martinez-Martin P, et al. Fa-
tigue rating scales critique and recommendations by the movement disorders
society task force on rating scales for Parkinson's disease. Mov Disord 2010;25:
805–22.
[13] Peeraully T, Yong M-H, Chokroverty S, Tan E-K. Sleep and Parkinson's disease: a
review of case–control polysomnography studies. Mov Disord 2012;27:1729–37.
[14] Ha AD, Jankovic J. Pain in Parkinson's disease. Mov Disord 2012;27:485–91.
[15] Kim HJ, McGuire DB, Tulman L, Barsevick AM. Symptom clusters: concept analysis
and clinical implications for cancer nursing. Cancer Nurs 2005;28:270–82.
[16] Barsevick AM, Whitmer K, Nail LM, Beck SL, Dudley WN. Symptom cluster research:
conceptual, design, measurement, and analysis issues. J Pain Symptom Manage
2006;31:85–95.
[17] Dodd MJ, Miaskowski C, Paul SM. Symptom clusters and their effect on the functional
status of patients with cancer. Oncol Nurs Forum 2001;28:465–70.
[18] Kim SH, Lee R, Lee KS. Symptom clusters in patients with breast cancer. Korean J
Adult Nurs 2009;21:705–17.
[19] Ferreira KA, Kimura M, Teixeira MJ, Mendoza TR, da Nóbrega JC, Graziani SR, et al.
Impact of cancer-related symptom synergisms on health-related quality of life and
performance status. J Pain Symptom Manage 2008;35:604–16.
[20] Ryu E, Kim K, Cho MS, Kwon IG, Kim HS, Fu MR. Symptom clusters and quality of life
in Korean patients with hepatocellular carcinoma. Cancer Nurs 2010;33:3–10.
[21] Gibb WR, Lees AJ. The relevance of the Lewy body to the pathogenesis of idiopathic
Parkinson's disease. J Neurol Neurosurg Psychiatry 1988;51:745–52.
[22] Peto V, Jenkinson C, Fitzpatrick R, Greenhall R. The development and validation of a
short measure of functioning and well being for individuals with Parkinson's
disease. Qual Life Res 1995;4:241–8.
[23] Skerman HM, Yates PM, Battistutta D. Multivariate methods to identify cancer-
related symptom clusters. Res Nurs Health 2009;32:345–60.
[24] Martinez-Martin P, Falup-Pecurariu C, Rodriguez-Blazquez C, van Hilten B, Odin P,
Chaudhuri KR, et al. Gender differences in non-motor symptoms in Parkinson's
disease. Parkinsonism Relat Disord 2009;15(Suppl. 2):66–7.
[25] Solla P, Cannas A, Ibba FC, Loi F, Corona M, Orofino G, et al. Gender differences in
motor and non-motor symptoms among Sardinian patients with Parkinson's
disease. J Neurol Sci 2012;323:33–9.
[26] Guo X, Song W, Chen K, Chen X, Zheng Z, Cao B, et al. Gender and onset age-related
features of non-motor symptoms of patients with Parkinson's disease: a study from
Southwest China. Parkinsonism Relat Disord 2013;19:961–5.
[27] Guo X, Song W, Chen K, Chen X, Zheng Z, Cao B, et al. Disease duration-related
differences in non-motor symptoms: a study of 616 Chinese Parkinson's disease
patients. J Neurol Sci 2013;330:32–7.
[28] Lim S-Y, Lang AE. The nonmotor symptoms of Parkinson's disease: an overview. Mov
Disord 2010;25(Suppl. 1):123–30.
[29] Aarsland D, Taylor J-P, Weintraub D. Psychiatric issues in cognitive impairment. Mov
Disord 2014;29(5):651–62.
[30] Morgante L, Colosimo C, Antonini A, Marconi R, Meco G, Pederzoli M. Psychosis
associated to Parkinson's disease in the early stages: relevance of cognitive decline
and depression. J Neurol Neurosurg Psychiatry 2012;83(1):76–82.