This study presents data from three independent cohorts. A total of n = 1124 patients were recruited, with n = 56 from cohort A, n = 44 from cohort B and n = 1024 from cohort C. All participants filled out semi-quantitative questionnaires to assess aspects of sleep. Moreover, 34 individuals from cohort A participated in an objective measurement of nocturnal movements using mobile devices.
The Training-PD study  was set up to evaluate the effects of different forms of training in PD patients. For this analysis, only data of the baseline assessment (cross-sectional) were used for data analysis. Fifty-six PD patients from the outpatient clinic of the department of Neurodegeneration at the University of Tübingen agreed to participate in the sleep sub-study of the Training-PD study and filled out the Parkinson’s Disease Sleep Scale (PDSS-2). Of those, 34 patients additionally agreed to wear a mobile device for the detection of nocturnal movements. Moreover, seven healthy controls took part in the assessment of nocturnal movements. The study was approved by the ethics committee of the Medical Faculty of the University of Tübingen (112/2015BO2) and written informed consent was obtained from all participants. Inclusion criteria were: (1) diagnosis of PD according to the UK brain bank criteria and (2) Hoehn and Yahr score ≤ 2.5. Healthy controls were recruited using public notices. Exclusion criteria for all study participants are comprised of: (1) the presence of relevant depressive symptoms (Beck Depression Inventory > 18 points); (2) physical status or diseases other than PD affecting physical training; (3) cognitive impairment that may interfere with study outcome (Montreal cognitive assessment, MoCA < 21); and (4) signs indicating a high risk of falls.
The aim of the ongoing, longitudinal ABC-PD study  is to assess disease progression of PD patients with and without (probable) amyloid beta pathology. One hundred PD patients selected according to their CSF Aß42 profile were recruited from the outpatient clinic of the department of Neurodegeneration at the University of Tübingen. Of those, data of 44 PD patients were included into the present data analysis. This study was approved by the same ethics committee (686/2013BO1) and written informed consent was obtained from all participants. Inclusion criteria for PD patients were: (1) diagnosis according to the UK brain bank criteria and (2) age between 50 and 85 years. Exclusion criteria were: (1) Beck Depression Inventory II ≥ 20 points); (2) history of clinically relevant strokes; (3) confirmed clinical diagnosis of possible or probable dementia according to the MDS Task Force criteria ; and (4) deep-brain stimulation surgery performed or planned.
In both studies, standardized interviews were performed, including the collection of information regarding disease duration and medication. The motor part of the Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS III) was used to assess motor function in the “On”-state. Part IV was used to rate patients’ perception on motor complications.
To improve in-house patient care, the Neurological Department of the Saarland Heilstätten GmbH Clinic Sonnenberg, Saarbrücken performed an anonymous survey in 1084 PD patients conducting the self-developed Saarbrücken PD survey. Besides sleep disturbances, the survey asked for occurrence and severity of motor and non-motor symptoms. In addition, the current medication plan was obtained from the patients. As data collection was anonymous, no ethical approval was needed (confirmed by the Ethic Commission of the State Chamber of Medicine, Saarland).
All studies were performed in accordance with the Declaration of Helsinki.
Subjective assessment of overall sleep quality and perception of nocturnal akinesia
All PD patients of the Training-PD and ABC-PD studies were asked to fill out the Parkinson’s Disease Sleep Scale (PDSS-2). The PDSS-2 comprises 15 questions on different aspects of sleep quality and impairment and is frequently used and recommended in research and clinical practice [14, 15]. A higher sum score reflects lower sleep quality. Question 9 of the PDSS-2, Did you feel uncomfortable at night because you were unable to turn around in bed or move due to immobility? was used to assess perception of nocturnal akinesia. Permission to use the PDSS-2 was obtained for both studies (https://eprovide.mapi-trust.org/).
The question Do you have problems when turning during sleep? (scoring range from 0, not true at all, to 10, absolutely true), from the Saarbrücken PD survey was used to assess perception of nocturnal akinesia in the Saarbrücken PD cohort.
Objective assessment of nocturnal movements
Participation in the assessment of nocturnal movements using a CE-certified mobile medical device (DynaPort Minimod®, McRoberts, The Netherlands; 64 × 62 × 13 mm; 100 Hz sample frequency) was optional for all study participants of the Training-PD study. The device includes a 3D accelerometer. It was worn at the lower back for 6 consecutive nights at home. Raw data of all nights were pre-analyzed by the company (https://www.mcroberts.nl/) using validated algorithms .
The following parameters of the nighttime assessment (means of all nights) were considered most meaningful for the evaluation of nocturnal akinesia, and were thus included in the analysis. Shifts were defined as small (> 30° < x ≤ 40°), medium (> 40° < x ≤ 80°) and large (> 80° < x ≤ 120°).
Duration of total sleep time: total duration of night’s rest, measured in hours.
Total movement: percentage of time of night’s rest for which movement was detected.
Movement intensity (MI) (in gram)
MI small shifts: weighted mean MI of movement duration for small shifts
MI medium shifts: weighted mean MI of movement duration for medium shifts
MI large shifts: weighted mean MI of movement duration for large shifts
MI supine periods: weighted mean MI of movement duration for supine periods
Angular velocity (degree/seconds)
Velocity small shifts: mean angular velocity of small shifts
Velocity medium shifts: mean angular velocity of medium shifts
Velocity large shifts: mean angular velocity of large shifts
To proof the validity of this method to assess sleep, we first compared all PD patients with a control group to look at PD-specific movement changes during sleep, followed by a comparison of the three different medication groups of PD patients.
The effect of Levodopa on quality of sleep and perception of nocturnal akinesia was first evaluated in the Training-PD study dataset. Second, the ABC-PD study and the Saarbrücken PD survey datasets were used for validation purposes. Third, objective assessment of nocturnal movements was analyzed in the Training-PD study. Statistics were performed with SPSS 24.0 (SPSS Inc., IBM, USA). Group comparisons were performed using the Mann–Whitney U for two-group comparisons and Kruskal–Wallis H-test for three group comparisons of non-normally distributed variables. Normally distributed variables were compared using Student’s t test or ANOVA. The Fisher’s exact test was used for comparison of the dichotomous parameters (gender). For comparisons of two groups, a p value of < 0.05, and for comparisons of three groups a p value of < 0.017 was accepted as statistically significant. Multiple linear regression using was used to identify variables predicting quality of sleep and nocturnal akinesia. Correlations were performed with the Spearman’s correlation coefficient.