Pain Fluctuations in Parkinson’s Disease and Their Association with Motor and Non-Motor Fluctuations

Study protocol approval and patient consents

We here report data on pain fluctuations from the German cohort of the prospective, observational cohort VALIDATE-PD study, which was conducted at two hospital centers in Germany (University Medicine Rostock, Movement Disorder Clinic Beelitz-Heilstätten). 27 The study was approved by the institutional review boards of both participating centers (ethic committee registry numbers A 2017-0115 for Rostock and AS 84(bB)/2018 for Beelitz-Heilstätten). All participants provided written informed consent prior to study participation.

Participants

Participants were eligible for the study if they were over 30 years old, had been diagnosed with PD according to the United Kingdom PD Society Brain Bank criteria, suffered from motor fluctuations observed by the treating physician and/or documented on part 4 of the Movement Disorder Society-revised Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) and were able to provide written informed consent. Even though there is no consensus on the definition of the advanced stage of PD, we herein define advanced PD based on the occurrence of motor fluctuations and/or dyskinesia. 32 Exclusion criteria comprised the existence of clinical signs for secondary or atypical parkinsonian syndromes, inability to complete questionnaires and/or patient diaries, lack of cooperation during the study procedures, presence of dementia (defined as scores on the Montreal Cognitive assessment [MoCA] < 21) 33 and/or relevant psychotic symptoms, ongoing treatment with advanced/invasive therapies (deep brain stimulation, subcutaneous apomorphine and levodopa-carbidopa intestinal gel) as well as the presence of miscellaneous diseases impairing the patient’s ability for consenting, participation and judgment.

Baseline assessments

The baseline assessments included cognitive screening with the MoCA, 33 clinical evaluation using the MDS-UPDRS and assessment of non-motor symptoms using the Non-Motor Symptom Scale (NMSS) and the King’s PD Pain Scale (KPPS). Furthermore, all participants were asked to fill-out the 19-item Wearing-off Questionnaire (WOQ-19), Non-Motor Symptoms Questionnaire (NMSQuest), King’s PD Pain Questionnaire, Beck Depression Inventory version 2 (BDI-II) and 39-item Parkinson’s disease Questionnaire (PDQ-39).

Diary assessments

Motor diary assessments. After inclusion into the study, all participants received detailed instructions on the PD home diary and watched a training video explaining all functional states with particular focus on the difference between tremor and dyskinesia. 34 Participants were then asked to indicate their predominant status during hour time periods for two consecutive days using the categories Asleep, “Off” (Off state), “On without dyskinesia” (On state), “On with non-troublesome dyskinesia”, and “On with troublesome dyskinesia”. While diary data for both participants and the observer was initially collected using the categories originally defined by Hauser and co-workers, 35 we eventually combined the categories “On with non-troublesome dyskinesia” and “On with troublesome dyskinesia” into the category “On with dyskinesia” (Dyskinetic state) for analysis, since the distinction between non-troublesome and troublesome dyskinesia could only be made by patients.

On both consecutive days, participants were observed by an experienced physiotherapist (A.B.), who had been trained to identify motor complications in advanced PD patients in the participating hospitals and acquired MDS certification as qualified UPDRS rater prior to the start of the study. The observer acted as single rater and independently evaluated motor states half-hourly throughout daytime (8.00 am through 6.00 pm) based on his clinical observations during a 7-meter version of the Timed Up and Go test (7m-TUGT), 36 taking into account global bradykinesia, tremor, dyskinesia and gait function. The observer was also instructed to dismiss any attempts from patients to get assistance with their own ratings. If not stated otherwise, presented data reflect results on PD Home motor diary data assessed by the participants.

Non-motor symptom diary assessment. On study day 1, participants were asked to rate eleven key NMS (psychiatric NMS: anxiety, depressive mood, inner restlessness, difficulties with concentration, hallucinations; fatigue; autonomic NMS: excessive sweating, sialorrhea, bladder urgency and dizziness; sensory NMS: pain) as present or absent during awake time using the same questions as the NMSQuest as already introduced by Ossig and colleagues. 29 During the training session, patients were instructed as to how the different functional states are defined and how the diary should be completed (placing a tick mark on a daily diary card every 60 minutes reflecting their predominant status over the prior hour period; for time asleep, the diary was completed upon awakening). Of note, if NMS were rated as “present” the NMS state in that hour was interpreted herein as NMS Off state and if NMS were rated as “absent” the state was interpreted as NMS On state to allow for an easier comparison with motor states.

Pain diary assessment. On day 2, participants were asked to rate pain in general and seven pain subdomains (pain around joints, pain related to internal organ or deep within the body (visceral pain), dyskinetic pain, painful cramps in a region during “Off” periods, generalized “Off” period pain, pain when chewing, pain due to grinding teeth, burning sensation in the mouth, burning pain in the limbs, shooting pain/pins & needles) as present or absent during awake time using similar questions as the German version of the KPPQ. 30 Pearson correlation tests comparing pain diary frequencies (expressed as percentage daily times with pain, PDT_Pain) with the corresponding frequencies as reported in King’s PD Pain Scale (KPPS) 37 were used as a measure of convergent validity of pain diary data (Table 2). The overall pain frequency and all pain subtype domain frequencies except visceral pain showed acceptable Pearson correlation coefficient (r) values higher than 0.30. 38 The item “Dyskinetic pain” was a major exception with a Pearson correlation coefficient of only – 0.06.

Global disease severity impression diaries. In addition to the motor and non-motor diaries, patients and observer rated disease severity in individual motor states hourly using 7-point versions of Patient Global Impression of Severity scale (PGI-S) and Clinical Global Impression of Severity scale (CGI-S), respectively, ranging from “normal” to “extremely ill”. For simplicity, we condensed the original 7 grades of disease severity on both scales to four categories: “normal” (containing the original grade ‘normal’), “mild” (comprising the original grades ‘borderline ill’ and ‘mildly ill’), “moderate” (containing the original grade ‘moderately ill’) and “severe” (comprising the original grades ‘profoundly ill, ‘severely ill’ and extremely ill’).

Aggregated data calculated from diary ratings. On the participant level, percentage daily times (PDT) were calculated from the diary ratings by normalizing the number of hours per day in individual motor states (PDT_Off, PDT_On and PDT_Dyskinetic) or the number of hours with pain/NMS ratings (PDT_Pain or PDT_NMS) to the total waking day hour ratings and expressed in percentage of waking day hours. The analyses of PDT_Pain for the three motor states separately resulted in PDT_Pain/Off, PDT_Pain/On and PDT_{Pain/Dyskinetic}. To analyse the pain distribution patterns within the three motor states independent of the specific motor state frequencies, the proportion of hours with pain as percentage of total hours in the respective motor state per day were calculated (percentage pain times, PPT_Off, PPT_On and PPT_Dyskinetic).

Adherence to diary assessments and asleep hours

All VALIDATE-PD study participants who had 2 days of diary datasets with at least 46 hour time periods including sleeping hours (>95% of all hour periods) and at least 21 hour time periods for observer dairy ratings between 8 : 00 and 18 : 00 (>95% of all hour periods) with simultaneous patient/observer motor diary data and NMS/pain diary data were included in the final study analyses leading to analyzable datasets from a total of 47 participants.

In total, we analyzed 2,216 hour periods with patient diary entries (98.2% of all hour periods; day 1 : 1,089 [96.5%], day 2 : 1,126 [100%]). 711 time periods (32.1% of all time periods) rated as Asleep by the participants were excluded from further analyses. 448 (29.7% of waking day periods) hour time periods were rated by the participants as as motor Off state, 749 (49.7%) as On state and 309 (20.5%) as Dyskinetic state. From observer diaries, we analyzed 1,026 hour periods during the waking day (99.2% of all hour periods between 8 am and 6 pm; day 1 : 515 [99.6%], day 2 : 511 [98.8%]) with 20 (1.9%) hour periods being excluded due to the rating “Asleep”. 279 (27.7% of all periods) hour time periods rated by observers were classified as motor Off state, 358 (35.6%) as On state and 369 (36.7%) as Dyskinetic state.

These numbers translated into 688 hour periods (92.7% of all waking day time periods) with complete simultaneous ratings of motor states in patient diaries and NMS diary data from day 1, and 728 (95.3%) hour periods with simultaneous participant-ratings and pain diary data (day 2). For observer diaries, we analyzed 498 hour periods (99.0%) with complete simultaneous ratings of motor states and NMS diary data (day 1), and 501 (99.6%) hour periods with simultaneous motor ratings and pain diary data (day 2).

Statistics

Statistical analysis was performed using IBM SPSS Statistics software version 27 (IBM Corporation, New York, USA). Values are provided as numbers (percentages) or median (interquartile range, IQR), as appropriate. Boxplots are shown with a central mark at the median, bottom, and top edges of the boxes at 25^th and 75^th percentiles, respectively, whiskers out to the most extreme points within 1.5 times the interquartile range, and outliers scoring more than 1.5×IQR but at most 3×IQR outside the quartiles. Pairwise exclusion was used for missing values. For statistical comparisons, we used Kruskal-Wallis tests or Friedman tests with post-hoc Wilcoxon signed-rank tests with adjustment for multiple comparisons as indicated. p-values<0.05 were considered to be statistically significant. All diary sets were included for analysis. Individual time periods were excluded from analysis if there was no response or more than one response on either diary, or if the patient indicated they were asleep on one diary but not the other. For all analyses except for the 24-hours distribution analyses, hours marked as asleep or switches from or towards asleep were omitted from the calculations.

Association of pain with demographic/clinical data and motor and non-motor diary data. Assessments on the hour time level were analyzed using χ² tests comparing pain and no-pain hours or all three motor states with post-hoc pair-wise comparisons χ² tests test with Bonferroni adjustment.

To cluster pain and the other NMS we used hierarchical clustering of simultaneously occurring NMS using average linkage between groups as the agglomeration rule and size difference as the (dis)similarity measure using IBM SPSS Statistics software version 27.39–41 The size difference quantifies the difference in the number of observations present on one item but absent on the other. In the context of binary data as in the present cluster analyses, size difference is an index of asymmetry that ranges from 0 to 1: A value of 0 indicates perfect symmetry (both variables are equally distributed), while a value closer to 1 indicates greater asymmetry (one variable dominates over the other). The formula to calculate size difference is as follows: size difference=(|b-c|)/n, with n representing the total number of observations and b and c represent the diagonal cells corresponding to observations present on one item but absent on the other and (|b-c|) the absolute difference between the two diagonal cells.

On the participant level, Friedman tests with post-hoc Wilcoxon signed-rank tests with Bonferroni correction for multiple comparisons were applied to compare pain times between the three motor stwtes. Pearson’s correlation tests were used for correlations of PDTs and PPTs and demographic and clinical data in all motor states or in individual motor states. Pearson’s correlation coefficient |r| < 0.3 was considered a weak, |r| = 0.3– 0.59 a moderate and |r|≥0.6 a strong correlation.

Temporal connection between pain and motor Off episodes. For analyzing the temporal connection between pain and Off episodes as rated by PD Home diaries or clinical observers, Off episodes (defined herein as an motor Off state of at least 1 hour duration following a motor On period of at least 2 hours) from all participants were synchronized by summation of all Off state periods using the first hour of the motor Off period as the trigger event (start of Off episode). The Off state ratings were then cross-classified with pain diary ratings by putting them into 2×2 contingency tables for each hour motor Off state interval. All diary sets were included for analysis. Individual time periods were excluded from analysis if there was no response or more than one response on either diary.

Association of pain diary data with hr-QoL. Pearson’s correlation tests and multiple linear regression modelling was used to test the association of pain times from pain diary ratings and PDQ-39 as a measure of hr-QoL. Since hr-QoL had been reported to potentially correlate with various demographic and clinical parameters,42–45 we adjusted the correlations of pain diary data and PDQ-39 for the variables age, sex, symptom duration, MDS-UPDRS part III motor score as a measure of disease severity, and included BDI-2 and MoCA as independent covariates into the models. Multicollinearity of candidate variables were excluded by Pearson correlation tests (|r|<0.5).

Demographic and clinical data

We screened 55 participants for eligibility of whom 47 (85%) were successfully included into the pain study according to inclusion and exclusion criteria. Seven participants were excluded since they reached Montreal Cognitive Assessment (MoCA) scores below 21 points at screening (n = 2) or were not able to sufficiently adhere to motor (n = 2) or pain/NMS assessments (n = 3). One participant declined further participation due to undisclosed reasons after the screening visit. Demographic and clinical characteristics of the cohort are displayed in Table 1, further details in Supplementary Table 1). In brief, complete datasets were available from 24 male (51%) and 23 female (49%) PD patients with a median age of 65 years. Participants had been diagnosed with PD about 10 years prior to the study and were suffering from motor fluctuations since 61 months. During structured interviews, patients reported a wide range of motor fluctuations, in particular wearing-off (94%), nocturnal off (85%) and peak-dose dyskinesia (77%). Clinical scores and antiparkinsonian medication were representative for a patient population suffering from advanced PD (Table 1). 14 (30%) patients had joint or spinal osteoarticular degeneration (see Supplementary Table 1 for concomitant diseases). For data on diary adherence and data quantity, please refer to the Participants and Methods section.

Temporal association of pain types with motor symptoms

To assess pain in every hour of the waking day, we used the NMS diary as reported earlier plus pain question (to compare the occurrence of pain with that of other NMS) 29 on study day 1 and the pain diary using questions from the German version of the King’s PD Pain Questionnaire (KPPQ)12,30, 12,30 on study day 2 to determine the various pain types. Analyzing these pain diary data on the time level, participants experienced pain on the two study days in 34.9% of waking hours independent of the motor state (day 1 using the NMS diary pain item: 27.5%; day 2 using the pain diary: 42.0%; Fig. 2a). Using the pain diary assessment on study day 2, the median number of pain types (max. 9 pain items) in each hour was 1 (IQR: 1– 2; range: 1– 6) with “Musculoskeletal pain” being by far the most frequent type of pain (29.6% of waking hours), followed by “Shooting pain/pins & needles” (10.5%), “Generalized Off pain” (10.3%) and “Off period dystonia in a specific region” (7.8%; Table 2).

OPEN IN VIEWER

Fig. 2

Temporal association of pain and its various types with simultaneous PD home motor diary states as rated by the participants. (a) Proportions of hourly diary ratings of pain by participants with respect to simultaneous PD home motor diary ratings (motor Off state, motor On state, Dyskinetic state). Note that participants reported the presence of pain in 34.9% of all ratings (dotted line as reference). Values are percentages. p-values above the diagram are from post-hoc χ² tests with Bonferroni adjustment (p-value from χ² test comparing all three motor states:<0.001). (b) Proportion of hourly participant diary ratings of the various pain types with respect to simultaneous PD home motor diary ratings (motor Off state, motor On state, Dyskinetic state). Values are percentages. ^*p < 0.05 and ^***p < 0.001 are from χ² tests comparing all three motor states, while ^$/#/§ represents p < 0.05, ^$$/##/§§p < 0.01 and ^{$$$/###/§§§}p < 0.001 for pair-wise comparisons as displayed in upper left corner applying χ² tests test with Bonferroni adjustment. Data are based on 1,506 simultaneous hourly ratings (67.9% of all time periods and 98.2% of waking hours on 2 consecutive days) in (a) and 764 simultaneous hourly ratings (67.7% of all time periods and 100% of waking hours on 1 day) in (b) by 47 participants.

We then analyzed pain frequencies with respect to motor states from participant PD Home diary data (Fig. 2a; for motor state frequencies, refer to Participants and Methods section). Pain was more frequent on motor Off state as compared to motor ON and Dyskinetic state, while we did not detect any differences between On and Dyskinetic state. When analyzing the various pain types, most types showed significant differences in their frequency between the three different motor states except “Dyskinetic pain” and “Pain when chewing” (Fig. 2b). Pain frequency patterns were similar as compared to overall pain with more frequent pain reported in motor Off state as compared to On and Dyskinetic state with “Visceral pain (deep in body, inner organs)” as the only exception (Fig. 2b): This pain type was least frequent in motor On state and more common in Dyskinetic than in motor Off state. Comparing pain and its types with motor states as rated by clinical observers revealed similar pain type patterns (Supplementary Figure 1).

Pain occurrence showed a clear circadian rhythm peaking in the early morning period with up to 62% of hours with pain. This pattern closely mirrored periods with motor Off but not Dyskinetic periods (see Supplementary Text for details). In addition to the comparison of pain and motor diary data from hour periods, we also assessed the timing of pain occurrence during Off episodes 31 : We did not detect any association of pain occurrence with the beginning and the further course of Off episodes (Supplementary Text).

Temporal association of pain with other non-motor symptoms

We then compared no-pain and pain hours of study day 1 with respect to the co-occurrence of other NMS (Fig. 3). When analyzing all motor states together, we detected significantly increased frequencies of “Depressive mood”, “Fatigue”, “Inner restlessness” and “Excessive sweating” but not of the remaining NMS in pain as compared to no-pain hours (Fig. 3a). When analyzing the association of pain and other NMS for the three motor states separately, we did not detect any differences between pain and no-pain hours in motor Off state, but only increased frequencies of “Fatigue”, “Inner restlessness”, “Concentration/attention” and “Excessive sweating” in motor On state and only increased frequency of “Fatigue” in Dyskinetic state (Fig. 3b– d). Cluster analyses largely confirmed the association of pain with “Fatigue” and “Inner restlessness” when analyzing all motor states together or the three motor states separately (Supplementary Text). Together, pain was particularly associated with “Fatigue” and “Inner restlessness” than other fluctuating NMS on the hour level.

OPEN IN VIEWER

Fig. 3

Temporal association of pain with other non-motor symptoms on the hour level from pain and non-motor diary ratings by the participants. Proportions of hourly diary ratings of various non-motor symptoms by participants with respect to simultaneous pain diary ratings by the participants. Upper left diagram (a) represents data for all motor states, while the other diagrams display data for motor Off state (b), motor On state (c) and Dyskinetic state (d) as assessed by PD home diary. Values are percentages. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001 from χ² tests comparing pain and no-pain hours. Data are based on 1,089 simultaneous hourly ratings (68.1% of all time periods and 96.5% of waking hours on 1 day) by 47 participants.

Impact of pain on global disease severity impression

Global disease severity as perceived by participants was assessed by simultaneous hourly ratings of patient global impression of severity (PGI-S; Fig. 4). PGI-S ratings displayed higher disease burden in pain hours as compared to no-pain hours when analyzing all motor states (Fig. 4a). These differences are essentially driven by higher disease burden in pain as compared to no-pain hours exclusively in motor On and Dyskinetic state hours (Fig. 4b– d). In contrast, hourly ratings of clinical global impression of severity by the clinical observer (CGI-S) did not show relevant differences between pain and no-pain hours (Supplementary Figure 2).

OPEN IN VIEWER

Fig. 4

Perception of disease severity by participants on the hour level with respect to pain diary data and motor states. Perception of disease severity as rated by participants on the Patient Global Impression of Severity (PGI-S) on the hour level. Illustration of the distribution of PGI-S ratings with respect to simultaneous pain ratings using the levels “Normal” (deep green color), “Mild” (bright green color), “Moderate” (yellow color) and “Severe” (orange color). The analyses were performed for all motor states (left diagram) and for motor Off state, motor On state and Dyskinetic state (from left to right diagram) as assessed by simultaneous PD home motor diary ratings. All PGI-S values were originally assessed with 7 severity grades, but then condensed to four levels to enhance clarity of the figure (please refer to Methods for more details). Values are percentages. p-values above the diagrams are from χ² tests comparing all PGI-S scores in pain versus no-pain hours. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001 are from pair-wise comparisons of PGI-S score applying χ² tests with Bonferroni adjustment. Data are based on 1,506 simultaneous hourly ratings (67.9% of all time periods and 99.2% of waking hours on 2 consecutive days) by 47 participants.

Pain fluctuations on the participant level

To analyze pain fluctuations on the participant level, we first analyzed the percentage daily time with pain for all motor states (PDT_Pain; Fig. 5a): The median PDT_Pain was 35.3% (IQR: 6.7– 69.2%; for PDTs of the various types of pain, refer to Table 2). We then compared PDT_Pain separately for each motor state, which revealed significant differences between motor Off or On state when compared to Dyskinetic state, but not between motor Off and On state (Fig. 5a). Of note, the analyses of PDTs_Pain during the different motor states did not only depend on the pain times per day but also on the PDTs spent in the different motor states and therefore do not allow the direct comparison of the pain patterns during the different motor states.

OPEN IN VIEWER

Fig. 5

Percentage daily times with pain (PDT_Pain) with respect to motor states and percentage pain time during specific motor states (PPT). (a) Percentage daily times with pain (PDT_Pain) from pain diary ratings for all motor states (white box) and with respect to simultaneous PD home motor diary ratings (motor Off state [blue box], motor On state [green box], Dyskinetic state [red box]). (b) Percentage motor state time with pain for each motor state normalized to motor state hours (percentage pain time during the specific motor states; PPT_Motorstate) for the three motor states from simultaneous PD home motor diary ratings (motor Off state, On state, Dyskinetic state). Boxplots are shown with a central mark at the median, bottom, and top edges of the boxes at 25^th and 75^th percentiles, respectively, whiskers out to the most extreme points within 1.5 times the interquartile range, and outliers scoring more than 1.5×IQR but at most 3×IQR outside the quartiles. Displayed p-values are from Friedman tests with post-hoc Wilcoxon signed-rank tests with Bonferroni correction for multiple comparisons. (c) 3D correlation analysis of PPT_Motorstate for the three motor states. Solid yellow line represents the regression line and dotted yellow lines display the crossing of the regression line at PPT_Off = 0% and at PPT_Off = 100% for easy interpretation of the 3D course of the regression line. Data are based on 1,506 simultaneous hourly ratings (67.9% of all time periods and 98.2% of waking hours on 2 consecutive days) by 47 participants.

To get better insights into pain distribution patterns within the three motor states independent of the specific motor state frequencies, we thus calculated the proportions of hours with pain as percentage of total hours in the respective motor state (percentage pain times, PPTs_Motorstate; Fig. 5b). PPT_Off was significantly higher as compared to PPT_On and PPT_Dyskinetic, while the latter two states did not show any differences. We found significant moderate to strong correlations of PPTs of the three motor states among each other with r ranging from 0.383 (PPT_Off versus PPT_Dyskinetic) through 0.579 (PPT_Off versus PPT_On) to 0.696 (PPT_On versus PPT_Dyskinetic; p < 0.05; Fig. 5c) showing interdependences of pain times across the three motor states.

Association of pain fluctuations with non-motor fluctuations on the participant level

When comparing the PDT_Pain and PDTs with other NMS from diary assessments at study day 1, we solely detected a significant correlation between PDT_Pain and PDT_Fatigue with r = 0.427 (p = 0.003, Pearson correlation test) when analysing all motor states (Supplementary Table 2). This association was also exclusively observed when correlating PDTs_Pain with PDTs_Fatigue or the corresponding PPTs for the three motor states separately (r values ranging from 0.345 to 0.530, P < 0.05; Supplementary Table 2).

Association of pain with health-related quality of life (hr-QoL)

We detected significant mild-to-moderate correlations of PDT_Pain/On and PDT_{Pain/Dyskinetic} but not PDT_Pain or PDT_Pain/Off with PDQ-39 sum scores as a measure of hr-QoL with Pearson correlation coefficients between 0.281 and 0.381 (Supplementary Table 4). Similar patterns of correlations were observed for pain time in the three motor states when expressed as PPTs (Supplementary Table 4). These significant correlations persisted after the deletion of subscore 8 (“Bodily pain/discomfort”) from PDQ-39 sum scores (r values: 0.261– 0.320; Supplementary Table 4). These correlations survived adjustment for potential demographic and clinical confounders by multivariate linear regression modelling (see Supplementary Text for details).

Analyses of pain frequency with other demographic and clinical characteristics did not reveal any associations of pain frequency parameters, neither with sex nor with other demographic or clinical parameters (regardless whether times of pain were expressed as PDTs or PPTs; Supplementary Table 3). Of note, osteoarthritis as a concomitant condition had also no influences on pain diary measures (p≥0.05; Mann-Whitney U-test).

Pain fluctuations on the cohort level

Using the pain diaries as the pain assessment tool, 89% of all participants experienced at least one hour of pain during the two days of pain assessment (Table 2). The chord diagram in Fig. 6 displays the many-to-many relationship between participants with pain in motor Off, On and/or Dyskinetic state: 32% of participants complained about pain in all motor states, while 28% complained about pain in Off and On state, but only 6% in On and Dyskinetic state. Complaints of pain in only one of the three motor states was rather infrequent and only 2% of participants complained about pain only in Dyskinetic state.

OPEN IN VIEWER

Fig. 6

Proportion of participants with Parkinson’s disease reporting pain with respect to motor states. The chord diagram shows the many-to-many relationship between participants with pain as reported in pain diary with respect to motor Off, motor On and/or Dyskinetic state as rated by participants using the PD home diary. Participants were classified as having pain in the various motor states if they reported at least one hour of pain in the pain diary in the respective motor state as simultaneously rated in the PD Home motor diary. Note that only 11% of participants reported no pain at all. Data are based on 1,506 simultaneous hourly ratings (67.9% of all time periods and 99.2% of waking hours on 2 consecutive days) by 47 participants.