Association between non-motor symptoms and health-related quality of life for patients with Parkinson’s disease: Evidence from a smartphone-based survey

This study aims to explore the association between the total number of non-motor symptoms and the Parkinson’s disease (PD) patients’ health-related quality of life quantitatively, using data from one of the largest smartphone-powered studies and a simplied version of 8-item Parkinson’s disease Quality of life Questionnaire (PDQ-8). The data used for analysis constitutes one part of a dataset derived from the project named ‘100 for Parkinson’s’, an initiative of uMotif in conjunction with the Cure Parkinson’s Trust, Parkinson’s UK and the European Parkinson’s Disease Association. 1246 patients were included in the baseline survey. The 30-item Non-Motor Symptom (NMS-30) Questionnaire and PDQ-8 Summary Index (PDQ-8 SI) were used to measure health-related quality of life and a generalized linear model was used to analyze the association of the non-motor symptoms and quality of life.

Considering the fact that motor characteristic of patients with PD (with and without limited ability to work) having different quality of life, data analysis was divided into two group of participants: patients with limited ability to work and without limited ability to work. Differences in variables between PDQ-8 questions and motor characteristic were compared using Chi-square tests. The generalized linear model (GLM) was employed to analyze the in uence of total number of non-motor symptoms on the healthrelated quality of life for people with PD and explore the potential determinants of quality of life quantitatively when controlling for confounding factors (including socio-economic factors, lifestyle factors and psychological symptoms prior PD diagnosis). The PDQ-8 SI score is the outcome variable; the sum of NMS-30 questions is the main determinant. The statistical analyses were performed using STATA statistical software version 12.0 (StataCorp LP, College station 77845, USA), related gures were made using Excel 2016. All p values were 2-sided and values of less than 0.05 were considered statistically signi cant.

Basic characteristics
The basic demographic features of 1246 patients with PD were displayed in Table 1. As for socioeconomic characteristics, there were 577 females and 615 males. The majority of patients' education were university/college/equivalent (72.64%). For the Life-style characteristics, 62.87% of patients never smoked tobacco and 15.33% of patients never consume drinks containing alcohol. 87.67% of patients live with others (including their wife, partners, children, relatives and hired carer). In the psychological symptoms prior PD diagnosis characteristics, from the sample, 261 (21.62%) patients suffered from depression prior to PD diagnosis, 225 (18.6%) had anxiety prior to PD diagnosis, and 677 (56.09%) had limited ability to work.  Figure 1 showed neuropsychiatric symptoms, sleep disorders, autonomic symptoms and gastrointestinal symptoms were less frequent in the patients without limitations to work than the patients with limited ability to work. The key symptoms were urgency (77.4%), feeling sad or blue (71.05%) and di culty getting to sleep or staying asleep (70.90%) respectively for the PD patients with limited ability to work, whereas the urgency (62.45%), di culty getting to sleep or staying asleep (59.81%) and getting up regularly at night to pass urine (55.09%) were the most frequent symptoms for the PD patients without limited ability to work. The distribution of PDQ-8 questions between patients with and without limit ability to work were shown in Table 2  The association of NMS-30 and PDQ-8 SI For patients with limited ability to work (Table 3)   For patients without limited ability to work -after controlling for the socio-economic determinants (Table 4), life-style and psychological symptoms prior to PD diagnosis determinants in model 3 -where the total sum of NMS increases by 1 point, the PDQ-8 SI score will increase by 1.56 points (95%CI: 1.37, 1.76; P < 0.001). In other words, where the non-motor symptoms are more, the health-related quality of life of PD patients without limited ability to work was worse. The GLM also indicated that depression prior to PD diagnosis (β = 4.06, 95%CI: 1.03, 7.09; P = 0.009) was also the potential determinant of worse quality of life than no depression prior to PD diagnosis; while age of 60-69 (β=-2.84, 95%CI: -4.85, -0.82; P = 0.006) was the potential determinant of better health-related quality of life of PD patients, compared to age ≤ 59 and education of up to 16 years old, respectively.

Discussion
This is the rst report of a comprehensive exploration of the association of total number of non-motor symptoms and the quality of life for patients with PD quantitatively using a large smartphone-based prospective study to capture data in the UK. Non-motor symptoms have been recognized as an important part of PD, and appropriate therapy can considerably improve quality of life [25]. Many researchers have explored how some items of the NMS independently have impact on the health-related quality of life, such as depression, sleep, mood and attention [13,26], less studied are the associations of the total number of non-motor symptoms and health-related quality of life. In this study, we explored the relationship between total number of non-motor symptoms and health-related quality of life for patients with PD and found that where the sum of NMS-30 questions is higher, PD patients' health-related quality of life of is worse. The NMS could be related to degeneration of different brain structures and could exert a stronger in uence on severe disability, impaired health-related quality of life and shortened life expectancy, which should be taken in account regarding treatment options [7].
Our study reveals that almost every patient with PD presented with more than one symptom. The relative NMS occurrence, as assessed by NMS-30, agrees with several other studies, showing sleep (di culty getting to sleep or staying asleep), mood/cognition (sad or blue), memory/attention and urination (urgency, getting up regularly at night to pass urine) as the most prevalent symptoms both for PD patients with and without limited ability to work [27][28][29]. However, the absolute occurrence of the most frequent symptoms is lower than Paulo Bugalho's study (Sleep/fatigue, affect/cognition, attention/memory were above 80%) [30], but higher than in most studies that used this same scale [28,29], in which above 60% are rare, while in our sample sleep disturbances reached 70.90%, and urgency reached around 80%. This difference of results could be caused by cultural, geographic or pattern of health care particularities in our sample and deserves further investigation to verify.
Depression in PD is believed to be a common complication ranging in prevalence between 2.7% and 70% but is under self-reported [31]. Depression can precede the development of PD and has a major impact on a patient's quality of life [31][32][33][34]. In our study, depression is one of the most prevalent symptoms both for patients with and without limited ability to work. Depression prior PD diagnosis was also a potential determinant of worse quality of life.
Age and education were found to be correlated with health-related quality of life of PD patients signi cantly. It is well known that age is the greatest risk factor for the development of PD.

Limitations
Our study presents some limitations. Firstly, the determinants of health-related quality of life are limited by the pre-speci ed questions in the surveys. There could be some potential unobserved confounding factors (such as individual exercise, occupation, income level or marital status) we did not control for in the generalised linear model. Secondly, this study conducted a brief question on whether patients' motor symptoms limited their ability to work and divided the characteristic of motor symptoms into with or without limited ability to work, but lacked a detailed evaluation of Parkinson's patients' motor symptoms.
More detailed study are needed to verify the association between non-motor symptoms and healthrelated quality of life for patients with Parkinson's disease after controlling for their motor symptoms. A nal limitation is our study is an observational study and reports the association between non-motor symptoms and health-related quality of life for patients with PD based on a quantitative study, more evidence based on qualitative studies and randomized controlled trials are needed to clarify the causal association in future research.

Conclusions
Non-motor symptoms are a key determinant of health-related quality of life and societal cost of PD and are often less appreciated than motor symptoms but are important sources of disability for PD patients.
In this study, we nd a revered relationship between total numbers of non-motor symptoms with healthrelated quality of life for PD patients; the more the total number of non-motor symptoms, the worse of the health-related quality of life. Improved awareness and understanding of the relationship between NMS and quality of life will encourage a multidisciplinary approach to patient care and facilitate the provision of more comprehensive education for patients and caregivers.  Other symptoms