Motor Learning-based Wheelchair Propulsion Training for Older Adults

Sample Size:

Calculations resulted in a sample size of 10 participants for each of the three groups to detect a significant difference with 80% power at a significance level of 0.05 (G*Power 3.1). To account for missing data and loss to follow-up, the sample size was increased by 15% such that each of the three groups will consist of 12 participants for a total of 36 participants.

Participant honorariums:

• All participants will receive $15 for each of the three testing sessions
• Participants randomized to either of the intervention groups (practice or training) will receive $5/ visit (excluding the testing session) to partly compensate for their travel cost.

Assessment for Eligibility:

Participants interested in participating in this study will be screened for eligibility over the phone to ensure that they meet the age and weight requirements, do not have injuries to the upper limbs, and that they understand the potential time commitment. Participants will be told that they will need to fill out one questionnaire to confirm their eligibility when they come in for their first appointment. Participants will complete the Mini Mental Exam after providing informed consent to screen for cognitive impairment.

Demographics descriptive information (baseline):

Variables that are important in describing the participant population will be collected at baseline including: age; height; sex; exposure to family members or friends that use a manual wheelchair.

Physical participant information (baseline, post-intervention, and two-weeks post-intervention):

Physical variables that are required for the testing procedures or that may influence the outcomes will be measured at each point of data collection, including: weight; grip strength; upper limb joint flexibility; and self-perceived physical fitness rating.

Assessments:

Measurements will be taken before intervention (baseline), four weeks and at 6 weeks post-baseline. One trained assessor will administer all of the assessments.

Randomization:

As participants are enrolled they will be randomly assigned to either the Motor learning-based training, practice, or control groups by a randomization algorithm.

Data Collection:

After providing informed consent and demonstrating approval to exercise by a physician-signed copy of the Physical activity readiness medical examination (PARMED X) if over 69 y or the Physical activity readiness questionnaire (PAR-Q+) if under 69 y, participants will complete a demographics questionnaire including sex, height, weight, age, general fitness level, and manual wheeling experience.

After completing the questionnaires, participants will be fitted to an appropriately sized Elevation wheelchair (Instinct Mobility, Vancouver, BC). Participants will be fitted so that their elbows form an angle of approximately 100 degrees when their hands are placed at the top of the wheel while they are seated in the wheelchair. To ensure consistency across measurement points, details of the wheelchair fitting will be recorded and reproduced during post-testing to minimize bias. The instrumented measurement wheel for collecting kinetic and temporal variables (SmartWheel -Three Rivers Holdings, Mesa, AZ) will be exchanged with the right wheel of the wheelchair and an inertia-matched dummy wheel will be applied to the left side of the wheelchair. Tire pressure will be inflated to the recommended level of 100 psi.

A drag test will then be performed to determine the drag force of the wheelchair-user system on the treadmill (Max-Mobility). During the drag test, the participant will sit passively in a fitted-laboratory wheelchair, which is connected by a rope to a force transducer. At a constant speed (1.1 m/s or 4.0 km/h), the angle of the treadmill will be increased in 10 steps (by 0.5% grade), and at each angle, the drag force will be determined. These 10 force measurements will be used to calculate the power output (PO) for each angle of inclination on the treadmill. A linear regression will be used to determine PO (calculated by multiplying the drag force with velocity according to PO (W) = drag force (N) · belt velocity (m/s)). The drag test takes approximately 5 minutes to complete.

Safety straps located on the sides of the wheelchair treadmill will be attached to the front of the Elevation wheelchair (on the metal frame above the front caster wheels). Prior to the data collection, participants will have five minutes to become acquainted with the wheelchair treadmill and will gradually be brought up to the set velocity of 1.1 m/s (4.0 km/h) that will be used for data collection. Previous research has used the speed of 1.1m/s for biomechanical testing because it is the minimum speed required to cross some intersections in the United States. To ensure that all participants perform the same amount of work for the testing trials, the power output will be standardized to 0.15 W/kg at an incline of 0.7% because an incline of 0.7% produced results most similar to over ground wheeling at speeds of 4-6 km/h. Resistance will be applied to the back of the wheelchair via a pulley system to ensure that the PO is 0.15 W/kg. The oxygen cost of wheelchair propulsion is affected by the PO, therefore standardizing the PO allows us to remove differences in drag force resulting from differences in bodyweight and wheelchair configuration. Standardizing the PO also will account for variations in participants' weight from pre-testing to post-testing to one-month post-testing.

After the participant is comfortable wheeling on the treadmill, the mouthpiece and nose plug for the Parvo Medics' TrueOne 2400 metabolic measurement system will be fitted to the individual. Ten minutes of baseline (resting state) data will be collected to ensure participants are breathing normally (i.e., not hyperventilating) and their respiratory exchange ratio value is appropriate (less than 0.9). Participants will then begin wheeling with the speed gradually increasing to 1.1 m/s, once the achieved speed is reached (approximately 15 s) the five minutes of data collection will begin.

Kinetic and spatial-temporal characteristic:

Data from the SmartWheel will be collected at 240 Hz using proprietary software and filtered using a 4th order low-pass Butterworth filter with a cut-off frequency of 20 Hz. Variables collected with the SmartWheel™ will include three dimensional forces, total force (√(Fx2+Fy2+Fz2)), tangential force (force acting on forward propulsion), cadence (cycles/s), push angle, and velocity.

Wheeling pattern Two Optotrak 3020 (NDI, Waterloo, ON) position sensors will be used to record sagittal-plane 2D kinematic data of the position of the 3rd metacarpophalangeal joint. The marker placed on the 3rd metacarpophalangeal joint will be used to define the wheeling strategy for each subject. Kinematic data will be collected at a sampling frequency of 200 Hz. All data will be filtered using a fourth order, 7 Hz low-pass Butterworth filter.

Data analysis:

Kinetic and kinematic wheeling data will be synchronized by an external trigger pulse, divided into cycles based on initiation of force applied to the push rim, and normalized in time to 100% of the cycle. Data from wheeling cycles that occur within the final 5th minute will be averaged for each participant.

Hand trajectory profiles of wheeling cycles that occur within the final 5th minute will be used to categorize the wheeling pattern into one of four categories. Two investigators, blinded to the participants group allocation will independently categorize the wheeling strategies according to their definitions. Wheeling strategies will be compared between two investigators to ensure consistency.

Motor learning-based training intervention:

Motor learning-based training will be administered by 1 trainer. She has over 6 years of experience studying wheelchair propulsion and has experience providing other ML-based training relating to manual and power wheelchair skills. Three trained volunteers will assist with pre and post-testing as well as the one-month retention testing on a rotating schedule. Volunteers will undergo 3 hours of training on the testing protocol. Participants will be asked to participate in 6 x 20 minute one-on-one training sessions over a 3 week period (2 sessions/ week). Six sessions have been selected because improvements in biomechanical variables have been observed in as little as 12 minutes of training, however improvement in movement efficiency of other continuous skills has been observed within 6 sessions.

The motor learning-based training will focus on some of the core concepts in the field of motor learning. The motor learning concepts will include variable practice with respect to speed, feedback (both intrinsic and extrinsic but focusing on extrinsic), feedback focusing on three main variables (wheeling pattern, speed of hand when contacting push rim, push angle), sequential learning (specifically relating to wheeling pattern), and mental imagery (during the rest breaks).

Practice intervention:

The practice intervention will be administered by 1 trainer. She has over 6 years of manual wheelchair propulsion research experience. Participants will be asked to participate in 6 x 20 minute one-on-one training sessions over a 3 week period (2 sessions/ week). Six sessions have been selected because improvements in biomechanical variables have been observed in as little as 12 minutes of training, however improvement in movement efficiency of other continuous skills has been observed within 6 sessions.

Although general practice offers experience through exposure and the repetition obtained from practice can be considered as an element of motor learning, it is important to determine if there is an added benefit to one-on-one training with an emphasis on motor learning principles over what an individual could perform on their own. It is anticipated that some 'motor learning' will occur during the practice intervention, however it is postulated that it will occur to a lesser extent.

Motor learning-based training and practice scheduling and equipment:

For this study, the training will be held in Vancouver at the Blusson Spinal Cord Centre at the Human Mobility Laboratory. The Blusson Spinal Cord Centre is a community-based facility that has a spine medical clinic as well as several research labs.

Control group:

The control group will not receive any training but will be tested at the same time points as the intervention groups.

Other:

Participants will be provided with a brief overview of the practice or motor learning-based training, depending on which of the two groups they are assigned to, after completing the pre-testing. Participants will be informed that there is a 'motor learning-based training' and 'control' group, however the fact that we are exploring two different types of training will not be disclosed so that the practice group does not feel that their training is inferior to the motor learning-based training group.

Upon completion of the practice or motor learning-based training, participants will be asked to complete an open-ended exit questionnaire about what they liked and disliked about the training and whether they thought it would be useful for a new wheelchair user.

Data synthesis and analysis:

Descriptive Statistics will be synthesized for demographic and physical information.

The effectiveness of the motor learning-based training on the primary outcome (gross mechanical efficiency as well as the secondary outcomes will be estimated using a 3 x 3 mixed factorial repeated measures Analysis of Variance (ANOVA). An adjusted p-value will be used to correct for multiple comparisons between baseline characteristics. A chi-square test will be used to determine differences in the categorical variable (propulsion technique) pre and post testing based on training exposure. Significance will be alpha = 0.05.