Novel Tools for the Delivery and Assessment of Exercise Programs Adapted to Individuals With Parkinson's Disease

The specific aims of this project are: a) to determine whether a technology-based personalized home exercise program delivered asynchronously and supervised remotely by a kinesiologist is more effective in improving gait than a traditional home-based exercise program, b) to assess the system acceptability and satisfaction, c) to examine and compare the impact of both approaches on several secondary outcome measures related to mobility, PD symptom, and quality of life. This study is a prospective randomized controlled single-blinded clinical trial of efficacy of a technology asynchronous assisted supervised tele-exercise program compared to an a traditional home-based exercise program with a booklet.

Interventions:

Control group (CONT): The participants (n=21) included in control group will follow a non-supervised home-based exercise program (booklet format). The exercise program will include 3 physical activity sessions per week for 12 consecutive weeks. This exercise regimen has been shown to be adequate to improve functional capacities in elderly and feasible. Each exercise session will last 40-45 minutes and includes: 1) 5 minutes of warm-up exercises; 2) 15-18 minutes of mobility and balance exercises (e.g. high knees, lateral launches, side steps); 3) 15-18 minutes of strengthening and coordination exercises (e.g. weight transfer and squats, leg extension and balance, lateral shifting, horizontal flexion and extension) and; 4) 5 minutes for cool down/stretching. This exercise protocol was developed based on general guidelines for the prescription of exercises to older adults and, validated in a previous study. The exercise booklet will be introduced and supervised for the first sessions by a trained kinesiologist. Follow-ups by phone will happen at sessions 6, 12 and 18. During these follow-up calls, if the answers are "yes" to the 3 following questions (1. Did the participant perform during the last weeks all the exercise sessions? 2. Did the participant feel safe when performing exercises? 3. Did the participant find easier to perform the exercises compared to the last contact?), then the participant will be invited, but not obligated, to increase the level of difficulty using the option indicated in the booklet.

JIN Group (n=21): The JIN group (n=21) will receive an individualized asynchronous supervised home-base exercise program using the Jintronix System, 3 times/week for 12 weeks (36 sessions). This exercise regimen has shown to be adequate to improve functional capacities in elderly. Briefly, the Jintronix Rehabilitation System (JRS) uses Microsoft Kinect cameras to track patient's movements in 3D during exercises. They are programmed as games that are visualized on a television and for which performance feedback is provided to the participant. Developed by Jintronix system designers in partnership with researchers, the system provides an interactive exergame environment to execute, specialized exercises through specific instructions and real-time feedback. The Jintronix system which includes a computer, the software and a motion capture device will be installed at participants' homes on their own television. A trained kinesiologist will install the system and supervise the exercise program. Participants will be trained to use the system with help of the kinesiologist during the visits. Each exercise session will last 40-45 minutes and includes: 1) 5 minutes of warm-up exercises; 2) 15-18 minutes of mobility and balance exercises (e.g. high knees, lateral launches, side steps); 3) 15-18 minutes of strengthening and coordination exercises (e.g. weight transfer and squats, leg extension and balance, lateral shifting, horizontal flexion and extension) and; 4) 5 minutes for cool down/stretching. Based on the Web-portal daily reports on the performance of the exercise program, the kinesiologist will do online adjustment of the degree of difficulty of the individualized exercises by increasing the number of repetitions, number of series, intensity and range of motion and provide feedback to the participant. This will ensure individualized adaptations and progress throughout the 12-week intervention.

The JIN group will use the Jintronix rehabilitation software to exercise at home whereas the CONT group will use an exercise booklet. Both groups will be free to exercise on the 3 days and times that best suit their schedule, with a mandatory minimum of one day off between sessions. In addition, to counteract human interaction bias, both exercise groups will have the same number and type of contact. The first weeks following the baseline evaluation, 3 to 4 in-person home visits are required to ensure proper and safe execution of exercise protocol (both groups) and the capacity to use the technology (JIN group only). In addition to these visits, 3 follow-up phone calls will be given to allow the kinesiologist to ensure safety and adherence. An emergency contact number will be given to participants for any questions regarding the exercise regimen during the 12-week intervention.

Study workflow and monitoring: After a phone screening to ensure eligibility and obtaining an oral consent to participate, participants will follow 2 baseline visits: 1 at their home and 1 at the "Université du Québec à Montréal" (UQAM). Before the home visit, subjects will be randomized into control (CONT) or Jintronix® (JIN). During the first visit at home, participants will sign an informed consent form. Afterward, the visit will be divided in 2 parts. The first part will serve to complete validated questionnaires (Parkinson's Disease Questionnaire, Short Falls Efficacy Scale-International and Life Space Assessment). The second part is to install the wearable equipment for mobility assessment (Apple watch® and location beacons). Participants will wear the sensor for at least 7 consecutive days. Afterwards, participants will return wearable sensors and location beacons at the UQAM visit. They will arrive in the morning in "off" state (abstinence of medication 12h before appointment) to undergo the motor evaluation, which is the 6-minute walk test, and the Short Physical Performance Battery test. After these, participants will take their medication with a breakfast at the lab. While waiting to be "on" state, investigators will perform the body composition (DEXA) and muscle composition scans (Ct-scan). When the medication is fully in effect, functional capacity and 6-min walk tests will be re-assessed. This will provide us comparison measures. All evaluations will be repeated at the end of the exercise intervention using the same sequence.

Power analysis:

Sample size: Primary outcome measure is the average gait speed assessed during a 6-minute walk test. Based on previous studies, obtaining clinically significant improvement in gait speed needs to show a gain of 0.2m/sec after intervention. Data using the proposed technology yielded significant improvement (JIN:+0,2±0,12m/s vs CONT:0,05 ±0,09m/s; p=0.025). Thus, obtaining a power of 0.80, an alpha of 0.05, with an effect size of 0.5 (which is significantly smaller that the investigators previously obtained (1.37)), this project would need a total sample size of 28 participants (14 per group) to detect clinically significant differences in gait speed changes between two groups. However, considering an attrition rate of 20% (3 out of 14 in each group) and to improve chances to achieve adequate power for secondary outcomes, a sample of 21 participants per group (n=42) will be used. Even if adequate power is not achieved for the secondary outcomes' measures, sample sizes between 24 and 50 are sufficient to calculate the standard deviation of outcome measures gathered during a feasibility study, to estimate the sample size required for hypothesis testing in a randomized control trial (RCT). Thus, calculations will ensure a sufficient power analysis for the primary outcome measure and position the investigators to establish the variability on secondary measures for a larger RCT.

Statistical analyses: Descriptive statistics (means, standard deviations, ranges, normality, and proportions) will be determined for all study variables. The primary outcome, gait speed, will be assessed using a two-way mixed model analysis of variance (ANOVA) where group is the between factor and time is the within factor. Tukey post hoc or Friedman test will be used based on data characteristics. Categorical variables related to the outcome measures (i.e. clinical scales and diary) will be analyzed using χ2 or Fisher exact tests for baseline comparisons. The change in each outcome variable (before and after training) will be modeled in an unadjusted 1-factor (group) analysis of variance, to investigate the effect of interventions. Wearable sensor data will be averaged over multiple days and compared using a two-way mixed model ANOVA as described above, similar to other secondary outcome measures (i.e. DEXA, CT-Scan, etc.). Tukey post hoc or Friedman test will be used based on data characteristics. To examine changes in variability, standard deviations will be averaged over multiple days and compared using the same approach. Finally, exploratory analyses will be performed to examine 1) if the JIN intervention allows to improve on/off periods, 2) the possible effect of age, sex or medication dose on exercise adaptations.

Innovation: Patients with PD are likely to experience a decrease in their daily physical activity (PA) because of physical impairments, fatigue, and apathy. PA is a complement to pharmacological interventions to manage the inherent decline in function associated with the disease. It is well known that PA interventions induce positive effects on physical capacities and more specifically on gait, mobility, posture and balance. However, engaging older adults in PA through different modalities of exercise programs to obtain these benefits is a challenge, as more than half of older adults are inactive physically and patients with PD are even more than older adults. Supervised exercise programs are needed to motivate and engage patients with PD in becoming more physically active in structured and safe ways. Providing access to supervised exercise programs for patients is difficult as they have mobility constrains and they rely on others for their travel. Home Exercise Programs (HEPs) offer potential benefits but are often lacking close supervision and interaction with a healthcare professional. The use of a supervised in-home PA program driven and supported by an exergame such as Jintronix© has the potential to be more impactful and meaningful to patients than traditional HEPs. It can be used independently by them in their own home and do not require the presence of a professional to perform physical activity training. This allows patients to perform exercise as they wish (schedule, day, on/off period etc.) which overall will increase their adherence to the PA program and promote self-engagement and self-management of their PD. The technology provides a unique way to engage patients in PA while ensuring supervision and communication with a health care professional. The proposed works will also innovate by using an ecological mobility measurement approach based on wearable sensors to capture gait and activity metrics under real life conditions for multiples days and link this to medication use. The combination of sensors used (inertial sensor at the ankle, and interior and exterior location sensors) and their packaging into highly usable and wearable form factors is highly innovative and unique. The proposed instrumentation for the measurement approach offers a lot of potential to study the impacts of pharmacological and non-pharmacological interventions on real life functions and activities of patients. It could be of interest to other parties involved in clinical trials on interventions for PD.