Sage Journals: Discover world-class research

Abstract

Background:

Pain is common in Parkinson’s disease (PD). In general and chronic pain populations, physical inactivity, poor sleep, and anxiety are associated with worse pain. However, little is known about these potential predictors of pain in PD.

Objective:

This cross-sectional observational study investigated associations between measures of physical activity, sleep, and mood with pain in people with PD.

Methods:

Pain was measured using the King’s PD Pain Scale and the Brief Pain Inventory (pain severity and interference) in 52 participants with PD. Independent variables were categorised by demographics (age, gender), disease severity (MDS-UPDRS) and duration, central sensitization (Central Sensitization Inventory), physical activity (Incidental and Planned Exercise Questionnaire), sleep (Pittsburgh Sleep Quality Index), and mood (Hospital Anxiety and Depression Scale).

Results:

Univariate regression analyses showed that increased disease severity, longer disease duration, greater central sensitization, increased physical activity, poor sleep, anxiety, and depression were associated with worse pain in one or more pain measures (p < 0.05). Multivariate regression models accounted for 56% of the variance in the King’s Pain Scale, 25% pain severity and 36% in pain interference. Poor sleep independently contributed to worse pain scores in all models (β 0.3–0.4, p < 0.05).

Conclusion:

Increased physical activity, poor sleep, anxiety, and depression are associated with worse pain scores in people with PD. For optimal management of pain in people with PD, sleep and mood may need to be addressed. Further, the nature of the relationship between physical activity and pain in PD requires further investigation.

Keywords

Parkinson’s disease pain physical activity sleep depression anxiety

INTRODUCTION

Non-motor impairments of Parkinson’s disease (PD), such as sleep disturbance, anxiety, depression and pain often precede motor impairments [1] and are greater determinants of quality of life than motor impairments [2, 3]. Pain affects up to 85% of people with PD [4], who report greater pain severity than individuals without PD [5]. Despite this high prevalence, pain is poorly understood in this population and current management is based largely on expert opinion.

The treatment of poor sleep and mood (anxiety and depression) through exercise and cognitive behavioral therapy are recommended techniques that assist in pain management in older adults [6]. However, the potential for these interventions to manage pain in people with PD is unclear. In PD, exercise (i.e., a type of physical activity that is planned, structured, repetitive and purposeful [7]) is recommended as one part of a pain management program, and there is evidence that exercise could reduce pain [8]. Additionally, increased physical activity (i.e., bodily movement produced by skeletal muscles that require energy expenditure, including exercise and incidental physical activity [7]) is associated with reduced pain in chronic pain populations [9, 10]. However, relationships between physical activity and pain in people with PD are yet to be explored.

In general and chronic pain populations, poor sleep [11], anxiety [12], and depression [13], are associated with worse pain. Similar findings have been reported in the PD population, with associations between sleep [14, 15], anxiety [3, 16], depression [3 , 17], and pain. However, previous work has predominantly focused on exploring potential predictors of sleep [18 –20] and depression [21], with very few studies exploring the potential predictors of pain. One large study [3] explored a range of potential predictors of pain in people with early PD (mean of 3 years) and found that mood (anxiety and depression combined), younger age, female gender, autonomic symptoms and motor complications were associated with increased pain severity. However, the relationships between anxiety and depression with the overall pain severity score were not assessed separately. The individual relationships between anxiety and depression with subtypes of PD pain [22] were assessed and found to be weakly correlated. Furthermore, this study did not examine the relationship between physical activity or sleep with pain. Further work investigating the relationships between physical activity, sleep and mood with pain both individually (i.e., univariate models) and in combination (i.e., multivariate models) is needed to increase understanding of the nature of pain in PD. Given physical inactivity, poor sleep and impaired mood are all potentially remediable, a greater understanding of the relationships between these factors and pain would guide future work to develop and augment interventions designed for pain management.

Studies exploring pain in people with PD have used a wide variety of pain measures [3 , 14–17]. In recent years the King’s PD Pain Scale was developed [22] and validated [23] as a researcher/clinician administered tool. It is the only PD-specific tool, and the only tool recommended by the Movement Disorders Society for measuring pain intensity in PD [23]. However, the recency of the development of this scale means that few reported studies have used it. The Brief Pain Inventory (BPI) is another commonly used pain measurement tool, which has shown good general reliability, validity and responsiveness. It has the advantage of being a short and simple, self-administered questionnaire that provides measures of both pain severity and pain interference. Though it has not been specifically validated for use in people with PD [23], it has been used in previous studies [14, 17].

Therefore, the primary aim of this study was to examine associations between physical activity, sleep and mood (anxiety and depression) with pain (King’s Pain Scale and BPI (pain severity and interference)) in people with PD, after adjusting for other factors likely to influence pain. The secondary aims were to explore potential temporal influences on these associations to see if, (1) physical activity that same day or sleep the night before predicted diarized pain severity, and (2) if diarized pain severity predicted physical activity the next day or sleep that night.

MATERIALS AND METHODS

Data for this observational cross-sectional study were obtained between 2015 and 2017 as a pre-planned adjunct to two studies investigating the presence of exercise-induced analgesia in people with PD [24]. Both studies were approved by the relevant Human Research Ethics Committee and participants gave written informed consent before testing.

Participants

Fifty-two participants with PD were recruited through: a research clinic; the research team’s databases of people with PD who agreed to be contacted regarding research participation; and advertisements in newsletters circulated to members of a patient support organization. Participants were included if they were: aged≥40 years; diagnosed with idiopathic PD by a neurologist; on stable PD medication for at least 2 weeks; community dwelling; and able to walk > 200 m with or without a walking aid. Participants were excluded if they had neurological impairments other than those due to PD; substantial cognitive impairment (Mini-Mental State Examination≤24); pain due to another diagnosed chronic pain condition or any other health condition that would interfere with safe conduct of the testing protocols or interpretation of results.

Measurements

Participants attended an initial measurement session to complete background information and the King’s PD Pain Scale. Participants then underwent actigraphy monitoring of physical activity and sleep for the following week, during which time they also kept a sleep and pain diary and completed 6 questionnaires. Measurements are described below, with further details and references presented in Supplementary Table 1.

Pain measures (dependent variables)

Pain was measured with the King’s PD Pain Scale and BPI. The King’s PD Pain Scale assesses the severity and frequency of pain across seven domains and is summed to provide an overall pain score. The BPI provides pain severity and interference subscores.

During the 7 days of actigraphy monitoring, participants kept a diary of pain. Pain severity from 0 to 10 (0 = no pain, 10 = worst imaginable pain) were recorded at the end of each day.

Independent variables

Independent variables were categorized into: demographics; disease severity and duration; central sensitization (defined as increased responsiveness of nociceptive neurons of the central nervous system to their normal afferent input resulting in hyperalgesia or allodynia [25]); physical activity; sleep; and mood. Demographics (age and gender), disease severity (Movement Disorder Society Unified Parkinson’s Disease Rating Scale [MDS-UPDRS] motor score), disease duration (years since diagnosis), and central sensitization (Central Sensitization Inventory questionnaire) were collected to adjust multivariate models for these variables that may influence pain.

Physical activity was assessed subjectively using the Incidental and Planned Exercise Questionnaire. The Incidental and Planned Exercise Questionnaire measures the amount of weekly planned and incidental physical activity during the past 3 months. Actigraphy was used to make objective measures of physical activity which included the number of steps taken per day and % time spent in sedentary, light, and moderate to vigorous physical activity.

Sleep was assessed using the Pittsburgh Sleep Quality Index questionnaire, Epworth Sleepiness Scale questionnaire, REM Sleep Behavior Disorder Questionnaire (REMSBDQ) and sleep actigraphy measures. The Pittsburgh Sleep Quality Index questionnaire, Epworth Sleepiness Scale questionnaire and REMSBDQ were used to assess sleep quality and disturbances, daytime sleepiness and REM sleep behavior disorder, respectively. Sleep actigraphy measures were sleep efficiency, sleep latency, wake after sleep onset and total sleep time.

Mood over the previous week was assessed using the Hospital Anxiety and Depression Scale questionnaire, which provides anxiety and depression subscores.

Actigraphy data collection and management

Actigraphy measures of physical activity and sleep were collected over 7 days from two GT3X-BT ActiGraph monitors (ActiGraph LLC, Pensacola, Florida). ActiGraph monitors are small (3.8 cm x 3.7 cm x 1.8 cm), lightweight (27 g), triaxial accelerometers.

Physical activity data were collected from a waist monitor worn on the right hip during waking hours. Wear-time validation for physical activity monitoring was performed using the Choi algorithm [26], with minimum wear time set at 540 minutes (i.e., 4 days or more of≥9 hours per day). Sleep activity data were collected from a wrist monitor worn on the participants’ less affected wrist for 24 hours per day. If both sides were equally affected, the monitor was worn on the non-dominant wrist. The minimum wear time for sleep monitoring was 4 nights.

During the 7 days of actigraphy monitoring, participants kept a diary of physical activity and sleep, which were logged at the end of the day and first thing in the morning, respectively. The sleep diary included: time into bed; time attempting to sleep; wake-up time; number of times awake during the night; time spent awake during the night and sleep quality from 0 to 10 (0 = very good, 10 = very bad). The physical activity and sleep actigraphy data were verified against each participants diary, with time attempting to sleep and wake up time used to assist in setting the sleep analysis periods.

Accelerometer data were analysed using ActiLife Version 6.13.3 software (ActiGraph LLC, Pensacola, Florida). In order to be consistent with methods used to develop the algorithms to process physical activity and sleep data, one-second epochs were reintegrated to 60-second epochs. Physical activity was analysed using the Freedson [27] cut-off points for percentage of time spent in different intensities. These cut off points are defined as: sedentary,≤100 counts per minute; light intensity 101–1951 counts per minute; moderate 1,952–5724 counts per minute and vigorous intensity≥5725 counts per minute. Sleep periods were analysed using the Cole Kripke algorithm [28].

Data analysis

Descriptive statistics were calculated for all participant characteristics and all variables were presented on histograms and visually inspected for data normality. The associations between physical activity, sleep and mood (independent variables) and pain (King’s PD Pain Scale, BPI) were explored with linear regression. Results from the univariate regression were used to develop multivariate regression models for each pain outcome, where independent variables from the univariate analyses with p≤0.1 were eligible for the multivariate model. One independent variable from each domain was permitted in multivariate models, with a maximum of 5 variables per model to ensure 10 observations per variable to avoid overspecification [29]. When more than one independent variable was eligible, the one with the lowest p-value was selected. Where two independent variables from different domains were highly correlated (r > 0.7), the variable with the lowest p-value was included.

Post-hoc univariate linear regressions were conducted to examine if variables utilized in the King’s PD Pain Scale multivariate model were associated with the different King’s PD Pain Scale domains. Domains where more than half of participants scored > 0 were assessed.

Exploratory univariate linear regression was conducted with the daily diary data to determine if light physical activity the same day and sleep efficiency the night before predicted the worst reported pain severity (dependent variable), and to determine if the worst reported pain severity (independent variable) predicted light physical activity the next day and sleep efficiency that night. The day with worst pain severity was defined as the first day with the highest pain severity rating where physical activity/sleep data was available for that day, the day prior and the day after.

Data were analyzed using the Statistical Package for the Social Sciences (version 22; IBM SPSS statistics). For all analyses, significance was set at p < 0.05.

RESULTS

Participants’ characteristics, pain scores and independent variables are presented in Table 1. Participants had mild to moderate disease severity and 49 of the 52 participants (94%) were on levodopa medication. On the King’s PD Pain Scale domains, musculoskeletal and nocturnal pain scores were the highest. The BPI pain severity score (out of 10) indicated 33 participants had none to mild pain (0–3), 18 had moderate pain (4–6) and 1 participant had severe pain (≥7). Thirty-four participants (65%) reported scores suggestive of sleep impairments on the REMSBDQ and the Pittsburgh Sleep Quality Index questionnaire. Hospital Anxiety and Depression Scale questionnaire results suggested that eight (15%) and seven (13%) participants had symptoms of anxiety and depression, respectively.

Table 1

Descriptive data for participant characteristics, pain measures and independent variables (n = 52)

Demographic Variable	Mean (SD) and/or N (%)	Range (min–max)
Age (y)	67.8 (7.8)	46–79
Male gender, n	36 (69.2%)
Disease duration (y)^*	7.8 (6.5)	0.25–38.0
MDS-UPDRS III (0 –132)^*	33.7 (12.8)	7–61
Pain
King’s PD Pain Scale total (0 –168)^*	15.3 (12.3)	0–64
Domain 1: Musculoskeletal pain (0 –12)	4.3 (3.2)	0–12
	42 (80.8%)
Domain 2: Chronic pain (0 –24)	1.7 (2.9)	0–12
	18 (34.6%)
Domain 3: Fluctuation-related pain (0 –36)	2.7 (5.6)	0–24
	16 (30.8%)
Domain 4: Nocturnal pain (0 –24)	4.4 (4.6)	0–20
	34 (65.4%)
Domain 5: Oro-facial pain (0 –36)	0.3 (1.4)	0–8
	4 (7.7%)
Domain 6: Discoloration; edema/swelling (0 –24)	0.8 (2.2)	0–8
	7 (13.5%)
Domain 7: Radicular pain (0 –12)	1.2 (2.6)	0–12
	15 (28.8%)
BPI severity (0 –10)^*	2.5 (1.8)	0.0–7.5
BPI interference (0 –10)^*	2.1 (2.0)	0.0–8.4
Central Sensitization
CSI (0 –100)^*	33.3 (10)	10–54
Physical Activity
IPEQ (activity h/week)	28.7 (15.7)	6.0–68.7
Avg % time sedentary/day^*∧	69.7 (11.0)	41.0–95.5
Avg % time light activity/day^∧	28.1 (10.8)	4.4–56.8
Avg % time MVPA/day^∧	2.3 (2.8)	0.0–13.8
Avg. total steps/day^∧	5365.6 (2708.5)	294–12567
Sleep
PSQI (0 –21)^*	7.5 (3.6)	1–18
ESS (0 –24)^*	8.1 (4.4)	0–18
REMSBDQ (0 –13)^*	6.2 (3.5)	1–13
Avg. Sleep efficiency/day (%)^∧	87.2 (10.1)	51–98
Avg. Sleep latency/day (min)^*∧	5.3 (6.2)	0.1–31.3
Avg. Wake after sleep onset/day (min)^*∧	51.4 (42.6)	5.1–196.0
Avg. total sleep time/day (min)^∧	383.1 (68.3)	188.9–509.5
Mood
HADS anxiety (0 –21)^*	4.5 (3.3)	0–14
HADS depression (0 –21)^*	4.7 (3.1)	0–13
7 day diary data (n = 51)
Worst Pain severity (0 –10)^*	3.5 (2.6)	0–9
% time light activity that same day^∧	29.4 (13.6)	3.2–68.8
% time light activity next day^∧	28.3 (12.7)	4.6–68.5
Sleep efficiency night before (%)^*∧	89.5 (9.1)	57.6–100
Sleep efficiency that night (%)^*∧	88.7 (9.4)	62.5–99.7

^*High score is worse, ^∧measured using actigraphy. SD, standard deviation; MDS-UPDRS, Movement Disorders Society sponsored Unified Parkinson’s Disease Rating Scale; BPI, Brief Pain Inventory; CSI, Central Sensitization Inventory; IPEQ, Incidental and Planned Exercise Questionnaire; Avg., average; No., number; MVPA, moderate to vigorous physical activity; PSQI, Pittsburgh Sleep Quality Index; ESS, Epworth Sleepiness Scale; REMSBDQ, REM Sleep Behavior Disorder Questionnaire; HADS, Hospital Anxiety and Depression Scale.

Univariate linear regression analyses (Table 2) revealed that increased disease severity, longer disease duration, worse central sensitization (Central Sensitization Inventory questionnaire) scores, increased physical activity scores (Incidental and Planned Exercise Questionnaire, and % time light), less time sedentary (% time sedentary), poor sleep (REMSBDQ, Pittsburgh Sleep Quality Index questionnaire) and mood (Hospital Anxiety and Depression Scale questionnaire, both anxiety and depression subscores) were significantly (p < 0.05) associated with one or more pain variables.

Table 2

Univariate associations between independent variables and measures of pain

Independent variables	Adjusted R²	Unstandardized B (95% CI)	p
King’s PD Pain total
Demographics
Age (y)	–0.02	–0.06 (–0.5–0.39)	0.8
Gender (male)	–0.02	–0.49 (–8.0–7.02)	0.9
Disease severity and duration
Disease duration (y)^*	0.24	0.96 (0.49–1.42)	<0.001
MDS-UPDRS III (0 –132)^*	–0.01	0.07 (–0.2–0.34)	0.6
Central Sensitization
CSI (0 –100)^*	0.08	0.38 (0.05–0.71)	0.03
Physical Activity
IPEQ	0.15	0.31 (0.11–0.52)	0.003
Avg % time sedentary*	0.06	–0.32 (–0.63 - –0.17)	0.04
Avg % time light	0.08	0.37 (0.06–0.67)	0.02
Avg % time MVPA	–0.01	–0.45 (–1.71–0.81)	0.48
Avg total steps	0.001	0.001 (–0.001–0.002)	0.31
Sleep
PSQI (0 –21)^*	0.31	1.91 (1.12–2.71)	<0.001
ESS (0 –24)^*	–0.01	0.22 (–0.57–1.01)	0.58
REMSBDQ (0 –13)^*	0.18	1.56 (0.65–2.46)	0.001
Avg sleep efficiency (%)	0.05	–0.31 (–0.65–0.03)	0.07
Avg sleep latency (min)^*	–0.02	0.04 (–0.55–0.6)	0.89
Avg wake after sleep onset (min)^*	0.05	0.07 (–0.0–0.15)	0.07
Avg total sleep time (min)	0.04	–0.04 (–0.09–0.01)	0.1
Mood
HADS Anxiety (0 –21)^*	0.09	1.26 (0.24–2.27)	0.02
HADS Depression (0 –21)^*	0.02	0.77 (–0.35–1.9)	0.17
BPI severity
Demographics
Age (y)	0.03	–0.05 (–0.12–0.01)	0.12
Gender (male)	–0.02	0.29 (–0.83–1.41)	0.61
Disease severity and duration
Disease duration (y)^*	0.15	0.12 (0.04–0.19)	0.003
MDS-UPDRS III (0 –132)^*	0.003	0.02 (–0.02–0.06)	0.28
Central Sensitization
CSI (0 –100)^*	0.09	0.06 (0.01–0.11)	0.02
Physical Activity
IPEQ	0.03	0.03 (–0.01–0.06)	0.10
Avg % time sedentary*	0.06	–0.32 (–0.63 - –0.17)	0.04
Avg % time light	0.03	0.04 (–0.01–0.09)	0.11
Avg % time moderate	0.01	–0.11 (–0.3–0.08)	0.27
Avg % time MVPA	0.01	–0.11 (–0.29–0.08)	0.26
Avg total steps	–0.02	0.00004 (–0.0002–0.0002)	0.67
Sleep
PSQI (0 –21)^*	0.17	0.22 (0.09–0.35)	0.001
ESS (0 –24)^*	–0.02	0.03 (–0.09–0.14)	0.65
REMSBDQ (0 –13)^*	0.18	0.23 (0.1–0.37)	0.001
Avg sleep efficiency (%)	0.01	–0.03 (–0.08–0.02)	0.23
Avg sleep latency (min)^*	0.003	0.05 (–0.04–0.13)	0.29
Avg wake after sleep onset (min)^*	–0.01	0.01 (–0.01–0.02)	0.44
Avg total sleep time (min)	0.01	–0.01 (–0.01–0.003)	0.21
Mood
HADS Anxiety (0 –21)^*	0.07	0.17 (0.01–0.32)	0.04
HADS Depression (0 –21)^*	0.04	0.14 (–0.03–0.31)	0.1
BPI interference
Demographics
Age (y)	0.03	–0.06 (–0.13–0.02)	0.12
Gender (male)	–0.02	0.31 (–0.89–1.51)	0.61
Disease severity and duration
Disease duration (y)^*	0.06	0.08 (0.001–0.17)	0.05
MDS-UPDRS III (0–132)^*	0.08	0.05 (0.01–0.09)	0.02
Central Sensitization
CSI (0 –100)^*	0.08	0.06 (0.01–0.12)	0.02
Physical Activity
IPEQ (h/week)	–0.02	0.01 (–0.03–0.04)	0.67
Avg % time sedentary*	0.03	–0.04 (–0.09–0.01)	0.1
Avg % time light	0.05	0.05 (–0.001–0.1)	0.06
Avg % time MVPA	–0.003	–0.09 (–0.29–0.11)	0.36
Avg total steps	–0.003	0.0001 (–0.0001–0.0003)	0.37
Sleep
PSQI (0 –21)^*	0.22	0.27 (0.13–0.4)	<0.001
ESS (0 –24)^*	0.004	0.07 (–0.06–0.19)	0.28
REMSBDQ (0 –13)^*	0.05	0.15 (–0.01–0.3)	0.07
Avg sleep efficiency (%)	0.05	–0.05 (–0.11–0.001)	0.06
Avg sleep latency (min)^*	–0.01	0.04 (–0.05–0.13)	0.43
Avg wake after sleep onset (min)^*	0.04	0.01 (–0.002–0.02)	0.08
Avg total sleep time (min)	–0.003	–0.004 (–0.01–0.004)	0.36
Mood
HADS Anxiety (0 –21)^*	0.23	0.3 (0.15–0.45)	<0.001
HADS Depression (0 –21)^*	0.08	0.2 (0.03–0.38)	0.02

^*High score is worse. CI, confidence interval; MDS-UPDRS, Movement Disorders Society sponsored Unified Parkinson’s Disease Rating Scale; CSI, Central Sensitization Inventory; IPEQ, Incidental and Planned Exercise Questionnaire; Avg., average; No., number; MVPA, Moderate to vigorous physical activity; PSQI, Pittsburgh Sleep Quality Index; ESS, Epworth Sleepiness Scale; REMSBDQ, REM Sleep Behavior Disorder Questionnaire; HADS, Hospital Anxiety and Depression Scale; BPI, Brief Pain Inventory.

Multivariate models for each pain variable are presented in Table 3. Poor sleep measured by Pittsburgh Sleep Quality Index questionnaire made an independent contribution to pain in all models. Increased disease duration made an independent contribution to two models: King’s PD Pain Scale and BPI severity; while increased physical activity (Incidental and Planned Exercise Questionnaire) and increased anxiety made independent contributions in one model each (King’s PD Pain Scale and BPI interference respectively). The adjusted R² was 0.56 for the King’s PD Pain Scale model, 0.25 for the BPI severity model and 0.36 for the BPI pain interference model, with the King’s PD Pain Scale model including independent contributions from disease duration, Incidental and Planned Exercise Questionnaire and Pittsburgh Sleep Quality Index questionnaire.

Table 3

Multivariate independent variables of pain

Independent variables	Unstandardized B (95% CI)	Standardized β	p
King’s PD Pain Scale total (Adjusted R² = 0.56, p < 0.001)
Disease severity and duration
Disease duration (y)^*	0.61 (0.2–1.01)	0.32	0.004
Central Sensitization
CSI (0–100)^*	–0.01 (–0.3–0.27)	–0.01	0.93
Physical Activity
IPEQ (h/week)	0.19 (0.03–0.36)	0.25	0.03
Sleep
PSQI (0–21)^*	1.48 (0.73–2.23)	0.44	<0.001
Mood
HADS Anxiety (0–21)^*	0.6 (–0.24–1.44)	0.16	0.16
BPI Severity (Adjusted R² = 0.25, p = 0.002)
Disease severity and duration
Disease duration (y)^*	0.09 (0.01–0.161)	0.3	0.03
Central Sensitization
CSI (0–100)^*	0.02 (–0.04–0.08)	0.11	0.47
Physical Activity
IPEQ (h/week)	0.01 (–0.02–0.04)	0.07	0.58
Sleep
PSQI (0–21)^*	0.16 (0.02–0.29)	0.31	0.03
Mood
HADS Anxiety (0–21)^*	0.05 (–0.11–0.21)	0.08	0.56
BPI Interference (Adjusted R² = 0.36, p < 0.001)
Disease severity and duration
MDS-UPDRS III (0–132)^*	0.03 (–0.004–0.07)	0.22	0.08
Central Sensitization
CSI (0–100)^*	–0.001 (–0.05–0.05)	–0.01	0.96
Physical Activity
Avg % time light	0.03 (–0.02–0.07)	0.14	0.25
Sleep
PSQI (0–21)^*	0.19 (0.05–0.33)	0.35	0.01
Mood
HADS Anxiety (0–21)^*	0.18 (0.01–0.34)	0.29	0.04

The REMSBDQ and the PSQI had identical p values in the univariate analyses for the BPI severity score, therefore for consistency between models, the PSQI was included in the multivariate model. *high score is worse; CI, confidence interval; CSI, Central Sensitization Inventory; IPEQ, Incidental and Planned Exercise Questionnaire; PSQI, Pittsburgh Sleep Quality Index; HADS, Hospital Anxiety and Depression Scale; BPI, Brief Pain Inventory; MDS-UPDRS, Movement Disorders Society sponsored Unified Parkinson’s Disease Rating Scale.

More than 26 (50%) participants scored > 0 in the musculoskeletal and nocturnal pain domains of the King’s PD Pain Scale. Post-hoc univariate linear regression analyses (Supplementary Table 2) revealed increased disease duration was significantly associated with musculoskeletal and nocturnal pain (p < 0.05). Increased physical activity (Incidental and Planned Exercise Questionnaire) was associated with musculoskeletal pain (unstandardized β 0.08 (95% CI 0.03–0.13)). Central sensitization and poor sleep (Pittsburgh Sleep Quality Index questionnaire) were associated with nocturnal pain (unstandardized β 0.16 (95% CI 0.04–0.29); 0.62 (0.31–0.94)).

Exploratory analysis on the diary data revealed reduced sleep efficiency predicted worst pain the next day (unstandardized β–0.09 (95% CI –0.17–0.01)) (Supplementary Table 3). However, pain severity on the worst day did not predict sleep efficiency that night (unstandardized β –0.80 (95% CI –1.80–0.20)) (Supplementary Table 4). There were no associations between physical activity with the worst pain severity.

DISCUSSION

This study explored the associations between physical activity, sleep and mood with pain in people with PD using pain assessment tools recommended for use in the PD population [23]. The results of this study suggest that increased light physical activity and reduced sedentary time, poor sleep, anxiety and depression are associated with worse pain in people with PD, with the association between poor sleep and pain being the most consistent. There was some variability in the results (e.g., increased physical activity was associated with the King’s PD Pain Scale total and the musculoskeletal pain domain, but not with the BPI pain severity and interference scores). This variability in results may be, in part, because the different pain measures used evaluate different aspects of the pain experience. The multivariate model for the King’s PD Pain Scale explained 56% of the variability in the pain scale total with increased physical activity, poor sleep and longer disease duration making independent contributions. The seven-day diary results suggested that reduced sleep efficiency the night before predicted the worst pain severity the next day. There were no associations between physical activity and pain in the diary data.

Exercise is an important intervention for people with PD, being effective for maintaining mobility, balance and muscle strength and reducing falls [8]. Our study found increased physical activity and reduced sedentary time was associated with increased King’s PD Pain Scale scores. Furthermore, more hours of physical activity per week (Incidental and Planned Exercise Questionnaire) made an independent contribution to increased pain in the King’s PD Pain multivariate model and was associated with increased scores on the musculoskeletal pain domain. This finding is unexpected given evidence that exercise may be beneficial in pain management in people with PD [8], and with exercise leading to acute reductions in pain sensitivity [24]. While it is characteristic for people with PD to perform low levels of activity [30] and while our finding only relates to light physical activity, further research could seek to include more people who perform moderate to higher intensities of activity to explore any association between higher intensity activity and pain. In some chronic pain populations, a U-shaped association has been found between activity and pain, whereby inactivity and excessive activity are both associated with worse pain [31]; this could be possible in people with PD. Another explanation is that some more active participants were not following exercise programs that were appropriate for their pain and were therefore exacerbating their pain. Qualitative research exploring the relationship between physical activity and pain in people with PD would assist in understanding this relationship. As exercise is an important evidence-based intervention for people with PD, these results suggest that therapists should ask about pain and carefully monitor people with ongoing pain to ensure necessary adjustments to exercise programs are made if required.

Sleep disturbances occur in up to 95% of people with PD [32], with around 30% experiencing rapid eye movement sleep behavior disorder [33]. Poor sleep, measured by the Pittsburgh Sleep Quality Index questionnaire and REMSBDQ, was consistently associated with greater pain. Additionally, analyses of King’s PD Pain Scale domains found poor sleep on the Pittsburgh Sleep Quality Index questionnaire was associated with nocturnal pain. These findings concur with previous studies [14, 15]. Notably however, in the present study, the actigraphy sleep measures did not demonstrate significant associations with pain. While the reason for this is unclear, one contributing factor could be because the Pittsburgh Sleep Quality Index questionnaire reflects an overall subjective composite score that combines all the components of an individual’s sleep-wake experience. Contrastingly, the actigraph-derived sleep measures tease out sleep into its smaller components, which individually did not demonstrate an association with pain compared to an overall composite sleep score [34]. Therefore, further study with larger numbers of participants could help to clarify any associations with individual sleep actigraphy measures with pain.

Our exploratory analysis found reduced sleep efficiency predicted the worst pain severity the following day, however, the worst pain severity did not predict poor sleep efficiency that night. Similar findings have been reported in other pain populations [11]. These results, taken together with the consistent association between poor sleep and pain in our main analysis, suggests that improving sleep may reduce pain. There is some evidence to support this in other pain populations where sleep interventions such as cognitive behavioral therapy and pharmacological therapies can improve pain symptoms [35]. Therefore, the potential for interventions that improve sleep to also reduce pain in people with PD warrants investigation.

This study found that anxiety and depression were associated with worse pain, consistent with previous research which identified associations between poor emotional functioning [17], anxiety [3, 16], and depression [3, 14] with pain. In the present study, anxiety appeared to be more strongly associated with pain than depression. Additionally, multivariate analyses showed that increased anxiety independently contributed to increased pain interference, but not to pain severity. This result suggests that anxiety may influence the ability of people with PD and pain to perform daily activities more than it influences pain severity. Therefore, it remains to be determined if treating anxiety and depression reduces pain in people with PD.

This study has several limitations. First, the study is relatively small and executed many analyses, increasing the risk of false positive findings (type 1 error) being identified. However, consistent findings across different analyses increases the confidence in our results. Second, as participants in this study had mild to moderate disease, relatively mild pain, and relatively normal scores for anxiety and depression, further research is needed to explore these associations in people with worse disease severity, more severe pain, and impaired mood. Finally, the data from actigraph monitors in people with PD can be influenced by involuntary movements such as tremor and dyskinesia which may have influenced the actigraph-related results from this study.

In conclusion, increased physical activity, poor sleep, anxiety and depression and are associated with worse pain scores in people with PD. The seven-day diary results suggest that poor quality sleep may predict pain the following day. Further research is required to better understand the relationship between physical activity and pain and to assess if treating altered sleep and mood will improve pain symptoms in people with PD.

CONFLICT OF INTEREST

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Footnotes

ACKNOWLEDGMENTS

This research was funded by a Parkinson’s NSW Bendigo Bank Parkinson’s Research Grant.

The supplementary material is available in the electronic version of this article: .

References

Chaudhuri

, Healy

, Schapira

AHV

(2006) Non-motor symptoms of Parkinson’s disease: Diagnosis and management. Lancet Neurol5, 235–245.

Martinez-Martin

, Rodriguez-Blazquez

, Kurtis

, Chaudhuri

(2011) The impact of non-motor symptoms on health-related quality of life of patients with Parkinson’s disease. Mov Disord26, 399–406.

Silverdale

, Kobylecki

, Kass-Iliyya

, Martinez-Martin

, Lawton

, Cotterill

, Chaudhuri

, Morris

, Baig

, Williams

, Hubbard

, Hu

, Grosset

(2018) A detailed clinical study of pain in 1957 participants with early/moderate Parkinson’s disease. Parkinsonism Relat Disord56, 27–32.

Beiske

, Loge

, Ronningen

, Svensson

(2009) Pain in Parkinson’s disease: Prevalence and characteristics. Pain141, 173–177.

Rana

, Qureshi

ARM

, Shamli Oghli

, Saqib

, Mohammed

, Sarfraz

, Rana

(2018) Decreased sleep quality in Parkinson’s patients is associated with higher anxiety and depression prevalence and severity, and correlates with pain intensity and quality. Neurol Res40, 696–701.

Reid

, Eccleston

, Pillemer

(2015) Management of chronic pain in older adults. BMJ350, h532.

Caspersen

, Powell

, Christenson

(1985) Physical activity, exercise, and physical fitness: Definitions and distinctions for health-related research. Public Health Rep100, 126–131.

Allen

, Moloney

, van Vliet

, Canning

(2015) The rationale for exercise in the management of pain in Parkinson’s disease. J Parkinsons Dis5, 229–239.

Geneen

, Moore

, Clarke

, Martin

, Colvin

, Smith

(2017) Physical activity and exercise for chronic pain in adults: An overview of Cochrane Reviews. Cochrane Database Syst Rev4, CD011279.

10.

Landmark

, Romundstad

, Borchgrevink

, Kaasa

, Dale

(2011) Associations between recreational exercise and chronic pain in the general population: Evidence from the HUNT 3 study. Pain152, 2241–2247.

11.

Finan

, Goodin

, Smith

(2013) The association of sleep and pain: An update and a path forward. J Pain14, 1539–1552.

12.

McWilliams

, Goodwin

, Cox

(2004) Depression and anxiety associated with three pain conditions: Results from a nationally representative sample. Pain111, 77–83.

13.

Bair

, Robinson

, Katon

, Kroenke

(2003) Depression and pain comorbidity: A literature review. Arch Intern Med163, 2433–2445.

14.

, Mao

, Ma

, Zhang

, Wang

, Li

, Huang

, Liu

(2018) Pain correlates with sleep disturbances in Parkinson’s disease patients. Pain Pract18, 29–37.

15.

Rana

, Khan

, Galange

, Khan

, Yousuf

(2012) Effects of pain on activities of daily living and functioning in Parkinson’s disease patients. Transl Neurosci3, 328–333.

16.

Buhmann

, Wrobel

, Grashorn

, Fruendt

, Wesemann

, Diedrich

, Bingel

(2017) Pain in Parkinson disease: A cross-sectional survey of its prevalence, specifics, and therapy. J Neurol264, 758–769.

17.

Roh

, Kim

, Jang

, Seo

, Lee

, Kim

, Oh

, Park

, Lee

, Koh

(2009) The relationship of pain and health-related quality of life in Korean patients with Parkinson’s disease. Acta Neurol Scand119, 397–403.

18.

Kurtis

, Rodriguez-Blazquez

, Martinez-Martin

(2013) Relationship between sleep disorders and other non-motor symptoms in Parkinson’s disease. Parkinsonism Relat Disord19, 1152–1155.

19.

Rana

, Qureshi

ARM

, Haris

, Danish

, Furqan

, Shaikh

, Sarfraz

, Rana

(2018) Negative impact of severity of pain on mood, social life and general activity in Parkinson’s disease. Neurol Res40, 1054–1059.

20.

Rolinski

, Szewczyk-Krolikowski

, R Tomlinson

, Nithi

, Talbot

, Ben-Shlomo

, Hu

(2014) REM sleep behaviour disorder is associated with worse quality of life and other non-motor features in early Parkinson’s disease. J Neurol Neurosurg Psychiatry85, 560–566.

21.

Rana

, Qureshi

ARM

, Rahman

, Mohammed

, Sarfraz

, Rana

(2017) Disability from pain directly correlated with depression in Parkinson’s disease. Clin Neurol Neurosurg160, 1–4.

22.

Chaudhuri

, Rizos

, Trenkwalder

, Rascol

, Pal

, Martino

, Carroll

, Paviour

, Falup-Pecurariu

, Kessel

, Silverdale

, Todorova

, Sauerbier

, Odin

, Antonini

, Martinez-Martin

(2015) King’s Parkinson’s disease pain scale, the first scale for pain in PD: An international validation. Mov Disord30, 1623–1631.

23.

Perez-Lloret

, Ciampi de Andrade

, Lyons

, Rodríguez-Blázquez

, Chaudhuri

, Deuschl

, Cruccu

, Sampaio

, Goetz

, Schrag

, Martinez-Martin

, Stebbins

, Development tMotMCoRS (2016) Rating scales for pain in Parkinson’s disease: Critique and recommendations. Mov Disord Clin Pract3, 527–537.

24.

Nguy

, Barry

, Moloney

, Hassett

, Canning

, Lewis

SJG

, Allen

(2019) Exercise-induced hypoalgesia is present in people with Parkinson’s disease: Two observational cross-sectional studies. Eur J Pain23, 1329–1339.

25.

International Association for the Study of Pain, IASP Terminology, https://www.iasp-pain.org/Education/Content.aspx?ItemNumber=1698#Centralsensitization, Accessed 3 March.

26.

Choi

, Liu

, Matthews

, Buchowski

(2011) Validation of accelerometer wear and nonwear time classification algorithm. Med Sci Sports Exerc43, 357–364.

27.

Freedson

, Melanson

, Sirard

(1998) Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc30, 777–781.

28.

Cole

, Kripke

, Gruen

, Mullaney

, Gillin

(1992) Automatic sleep/wake identification from wrist activity. Sleep15, 461–469.

29.

Babyak

(2004) What you see may not be what you get: A brief, nontechnical introduction to overfitting in regression-type models. Psychosom Med66, 411–421.

30.

van Nimwegen

, Speelman

, Hofman-van Rossum

EJM

, Overeem

, Deeg

DJH

, Borm

, van der Horst

MHL

, Bloem

, Munneke

(2011) Physical inactivity in Parkinson’s disease. J Neurol258, 2214–2221.

31.

Heneweer

, Vanhees

, Picavet

(2009) Physical activity and lowback pain:AU-shaped relation?Pain143, 21–25.

32.

Lees

, Blackburn

, Campbell

(1988) The nighttime problems of Parkinson’s disease. Clin Neuropharmacol11, 512–519.

33.

Barone

, Antonini

, Colosimo

, Marconi

, Morgante

, Avarello

, Bottacchi

, Cannas

, Ceravolo

, Cicarelli

, Gaglio

, Giglia

, Iemolo

, Manfredi

, Meco

, Nicoletti

, Pederzoli

, Petrone

, Pisani

, Pontieri

, Quatrale

, Ramat

, Scala

, Volpe

, Zappulla

, Bentivoglio

, Stocchi

, Trianni

, Dotto

(2009) The PRIAMO study: A multicenter assessment of nonmotor symptoms and their impact on quality of life in Parkinson’s disease. Mov Disord24, 1641–1649.

34.

Buysse

, Hall

, Strollo

, Kamarck

, Owens

, Lee

, Reis

, Matthews

(2008) Relationships between the Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), and clinical/polysomnographic measures in a community sample. J Clin Sleep Med4, 563–571.

35.

KKN

, Ferreira

, Pinheiro

, Aquino Silva

, Miller

, Grunstein

, Simic

(2019) Sleep interventions for osteoarthritis and spinal pain: A systematic review and meta-analysis of randomized controlled trials. Osteoarthritis Cartilage27, 196–218.

The Associations Between Physical Activity,Sleep,and Mood with Pain in People with Parkinson’s Disease: An Observational Cross-Sectional Study

Abstract

Background:

Objective:

Methods:

Results:

Conclusion:

Keywords

INTRODUCTION

MATERIALS AND METHODS

Participants

Measurements

Pain measures (dependent variables)

Independent variables

Actigraphy data collection and management

Data analysis

RESULTS

DISCUSSION

CONFLICT OF INTEREST

Footnotes

ACKNOWLEDGMENTS

References