Background: Physical activity for persons with Parkinson Disease (PD) is recommended yet little is known about the physical activity levels in this patient population. The primary aim was to assess the feasibility of using a direct measurement and self-report measure of physical activity in patients with PD. Methods: Physical activity was recorded in 11 out-patients with mild to moderate PD. An accelerometer based sensor system (SenseWear Pro Armband?) which was worn continuously over 2 days was used to measure physical activity. Minute by minute energy expenditure and steps per day were recorded. Self-report physical activity was measured using the Short QUestionnaire to ASsess Health-enhancing physical activity (SQUASH) which assessed average weekly activity. Results: Using the accelerometer based sensor system, 83% of the day was spent in sedentary activity with the majority active time spent at a light intensity (2.7 [SD 2.0] hrs/day). Self-reported mean number of hours for activities greater than 2.0 METs was 3.4 (SD 1.5) hrs/day. Although the overall time spent in activity did not differ between the accelerometer and SQUASH, partici- pants reported a higher proportion of activities at the moderate and vigorous intensities than the accelerometer recorded. Conclusions: Measurement of physical activity is a challenge in persons with PD given the disease-related symptoms. We found that, by all accounts, a self-report measure of physical activity should be complemented with a direct measure of physical activity.
Parkinson Disease (PD) is a chronic progressive neurodegenerative disease characterized clinically by varying combinations of rest tremor, rigidity, bradykinesia, postural instability and gait disorder [
Mounting evidence suggests that exercise may play a role in altering the progression of PD though its specific role in disease modification remains speculative at this point [
Overwhelming epidemiological evidence supports the health benefits of regular physical activity in the general population such as improving muscle strength [
Little attention has been directed toward the promotion of regular physical activity in persons with PD, despite the knowledge that physical inactivity is a modifiable risk factor for many other chronic conditions such as cardiovascular disease, hypertension, diabetes mellitus, osteoporosis, obesity and depression [
A convenience sample of 11 PD outpatients was recruited between July 2009 and March 2010 at the Movement Disorders Program (Glenrose Rehabilitation Hospital in Edmonton, Alberta, Canada), which serves northern Alberta with a catchment area of approximately 1.5 million people. Inclusion criteria for the study consisted of the participant 1) having a primary diagnosis of PD as diagnosed by a movement disorders neurologist, 2) being untreated with PD medications at time of enrolment and 3) ambulating independently. All participants had Hoehn & Yahr scores stage 1 or 2. Ethics approval was obtained from the University of Alberta Health Research Ethics.
Upon providing written informed consent, participants 1) completed a battery of self-report health-related quality of life measures, and 2) wore the SenseWear Pro Armband™ to estimate energy expenditure over 48 consecutive hours. To examine physical activity in terms of the intrinsic capacity to exercise we also asked participants to complete a treadmill test to evaluate aerobic fitness.
The self-administered measures consisted of sociodemographic data, such as date of birth, gender, marital status, and education, along with medical information (ambulatory status, comorbidities, and smoking history), and disease-related information (Hoehn & Yahr stage [
A direct measure of daily physical activity was estimated from the daily energy expenditure calculated from SenseWear Pro Armband™ (Body Media, Pittsburgh, PA) (SWA) data. The SWA accumulates data from a variety of sensors (including heat flux, biaxial accelerometer, galvanic skin response, skin temperature, near-body temperature) and combines these data with demographic characteristics (gender, height, weight, handiness and smoking status) to estimate energy expenditure using algorithms provided by the manufacturer [28-30]. The SWA also provides information on the number of steps/ day. This accelerometer based sensor system does not impede normal movement and can be worn throughout the day, in virtually all environments, except in water and for all types of activities, including sleep. The SWA has been shown to have good validity under both laboratory [28,31,32] and free living conditions [
The SWA was worn on the upper arm over the triceps muscle of the least affected side. For this study, participants were instructed to wear the SWA for three consecutive days so that 48 hours of near continuous data could be recorded. Participants were instructed to remove the device when bathing, showering or swimming. Once the SWA was returned to the lab, data were downloaded to generate the total time the device was worn, steps/minute, minute by minute energy expenditure and metabolic equivalent task (MET) intensity levels. One MET is the energy cost of resting quietly, often defined in terms of oxygen uptake as 3.5 mL/ kg/min. Unlike other studies that have measured activity levels in PD using activity monitors [35,36], the intensity of activity using METs can be derived from the SWA.
Daily physical activity was also captured using a self-report measure of habitual physical activity, Short QUestionnaire to ASsess Health-enhancing physical activity (SQUASH) [
The SQUASH has been evaluated in patient populations such as total joint arthroplasty, rheumatoid arthritis and lung transplant [24,38,39]. Overall, the SQUASH has comparable reliability as other self-report physical activity measures [
For both the self-report (SQUASH) and SWA data, sedentary was defined as values below 2.0 METS. Light activity was considered 2.0 - 2.9 METS, moderate activity 3.0 - 4.9 METS and vigorous activity as any value 5.0 METS or higher for those persons over 55 years of age. For the two participants who were under 55 years of age light activity was defined as 2.0 - 3.9, moderate activity 4.0 to 6.4 and vigorous activity as MET values 6.5 or greater [24,37]. The total activity time for the SQUASH and SWA was calculated by summing the total time spent in each of the three activity categories.
The intrinsic capacity to exercise was assessed using a submaximal exercise test on a treadmill. Participants were cleared to perform the exercise test by their neurologist and were screened using the Physical Activity Readiness Questionnaire [
During the five minute rest period prior to testing and throughout the exercise test, VO2 was recorded using a mobile metabolic cart (Oxycon Mobile, Jaegar Germany). Respiratory gases were recorded breath by breath using a mouth piece with nasal clips worn by the participant. Data were recorded for each breath and averaged over each one minute interval. Heart rate was also recorded for each one minute interval (Polar Electro, Finland) while blood pressure and Borg Scale of Perceived Exertion, a 15 point scale that quantifies the degree of exertion from none to exhaustion [
Health-related quality of life was assessed with the HUI3, a generic preference-based measure [
Energy expenditure was examined both as a continuous and categorical variable. For ease of interpretability, categorical data are presented. These categories were based on the Ainsworth’s Compendium of Physical Activities [
Descriptive statistics were used for all variables collected. The mean and standard deviation (SD) were reported for normally distributed data and the median and interquartile range (IQR) reported for data that were not normally distributed. To assess the significance of associations between categorical variables, c2 tests were performed. Independent sample t-tests were used for normally distributed continuous variables. The agreement between the SWA and SQUASH times spent at the three activity levels: light, moderate and vigorous was visually inspected using Bland-Altman method [
All statistical testing was performed with two-tailed tests and at a 0.05 level of significance unless otherwise stated. Statistical analyses were performed using the SPSSâ software version 15.0.1 for Windows, SPSS Inc.
Eleven participants with mild to moderate PD between the ages of 48 and 71 years and who had a disease duration of 4.0 (SD 2.6) years were recruited. All participants were from the community and 8 were living with a spouse. All except one person were currently non-smokers. Participants walked independently: 3 participants could only walk 1 to 5 blocks, 6 reported the ability to walk more than 10 blocks. The major limitation to walking was fatigue as reported by participants (n = 5). All participants were categorized at a high level of mobility confidence by the ABC score (mean 90 SD 8). The mean number of comorbid conditions was 4.1 (SD 1.7) with mental health problems (n = 6) and chronic back pain (n = 5) being the most frequently cited conditions. The mean overall HUI3 score was 0.66 (SD 0.20). Dexterity (0.83) and pain (0.87) attributes had the lowest scores (
The estimated VO2max achieved from the treadmill test was 31.2 (SD 8.2) mL/kg/min with a range from 18 to 49 mL/kg/min. The mean Borg exertion rating was 14 (SD 2). Four participants’ VO2 max scores exceeded the 50th percentile for their age and gender.
Daily Physical ActivityThe participants wore the SWA an average of 23.3 (SD 0.7) hrs per day over the two days. Data from the SWA dual accelerometers recorded that participants averaged 5458 (SD 4416) steps/day.
When energy expenditures (METs) from the SWA were categorized, participants were sedentary for 19.4 (SD 3.8) hours/day or 83% of the total time recorded. Participants reported performing light intensity work an average of 2.7 hrs (SD 2.0) hours/day or 79% of the time when they were active. Moderate activities accounted for 19% of their active time or a mean 1.1 (SD 1.8) hours/ day. The least amount of time when active was spent performing vigorous activities (2%) with an average of 0.1 (SD 0.2) hours/day. The majority of participants (n = 7) reported no change in their routine activities when wearing the SWA; however, one participant reported increased activity level while wearing the SWA and another reported less activity while wearing the armband.
According to the SQUASH, the mean daily number of hours spent at any activity level greater than 2.0 METs
was 3.4 hrs (SD 1.5). All participants reported activities in household (mean 0.9 [SD 0.5] hours/day) and leisure activities (mean 1.7 [SD 1.2] hours /day) with 8 reporting walking activities. Participants reported that 48% of this time was spent performing light activities, 38% at moderate intensity activities, and only 14% spent on vigorous activities.
The Bland-Altman plot (
To our knowledge there have been few studies that have examined physical activity level in persons with PD [19,20]. Because of the limited literature in this area, we conducted a feasibility study to identify measurement issues of physical activity in patients with PD. Overall, the measurement of physical activity in PD has been either self-report or direct measure [19,47] but rarely both types of measures have been used [
Inactivity in this group was also exemplified by the number of daily steps. The number of steps recorded in these participants was typical of older adults and of those living with chronic conditions that average between 3500 and 5500 steps/day [
In spite of the inactivity, it is likely these participants were a relatively healthy cohort with minimal physical involvement as reflected by the Hoehn & Yahr scores (stage 1 or 2) and the UPDRS motor scores which indicated mild to moderate disability. Based on our exclusion criteria it was unlikely that participants had symptomatic cardiac disease. While others have recognized that the fear of falling may limit activity [
The poor agreement seen between the SWA and the SQUASH suggests that a self-report measure of physical activity alone will not provide adequate detail of physical activity including sedentary activity. This agreement may have been affected by the SWA in that it has not been validated in PD, although participants in this study had mild PD. It was unlikely that tremor affected the readings from the SWA because only two participants had a tremor score greater than 0 on the UPDRS motor score for the least-affect upper limb side. These two participants had UPDRS tremor scores of 1—“slight or infrequently present” on the least affected side. The poor agreement between the two measures may be due, in part, to recall bias associated with the SQUASH values. It has been proposed by others that more intense activities are easier to recall than light ones [
As with many chronic conditions, exercise is recommended for persons with PD [
We wish to thank Dr. Martin Stevens for his assistance with the analysis of the SQUASH data.
Dr. Allyson Jones received salary support from the Alberta Heritage Foundation for Medical Research and the Canadian Institutes of Health Research.
ABC scale: Activities-Specific Balance Confidence Scale HR: Heart Rate HUI3: Health Utilities Index Mark 3 ICC: Intra Class Correlation MET: Metabolic Equivalent Task PD: Parkinson Disease SD: Standard Deviation SWA: SenseWear Pro Armband™ SQUASH: Short QUestionnaire to ASsess Health-Enhancing Physical Activity UPDRS: Unified Parkinson’s Disease Rating Scale