Factors Affecting Start-Up Propulsion Forces

RESNA 28th Annual Conference - Atlanta, Georgia

Rachel Cowan, MS, Michael Boninger, M.D., Jennifer Mercer, BSE, Jon Pearlman, MS, Alicia M. Koontz, PhD, Rory Cooper, PhD

Dept. PM&R, University of Pittsburgh Medical Center, PA 15261
Human Engineering Research Laboratories, Highland Drive VA Medical Center, Pittsburgh, PA


 Upper extremity pain and injury among individuals who utilize a manual wheelchair for mobility may reduce independence. High loads incurred during wheelchair propulsion have been associated with upper extremity injury. A manual wheelchair user’s weight and wheelchair configuration are related to higher loads. Twenty-two manual wheelchair users completed over-ground propulsion kinetics evaluation across three surfaces. Forces at start-up were unrelated to the combined user and wheelchair weight and axle position. Forces during stroke one and stroke two were significantly higher across a ramp compared to carpet or tile. Clinicians may wish to consider requiring their clients to transverse ramps during propulsion as a method to identify functional weaknesses only apparent during challenging environments.

Keywords: wheelchair, propulsion, force


 A spectrum of research has provided insight into the factors associated with the prevalence of upper extremity injury in manual wheelchair users [1,2,3]. Higher peak propulsion forces are associated with the prevalence of upper extremity injury [3,4]. Such forces are related to subject and wheelchair characteristics [2,3,5]. These findings are the result of research focused on steady state propulsion on a treadmill or dynamometers. It remains to be seen if such associations hold true on commonly encountered surfaces at a self-selected pace. Furthermore, the initial propulsive stroke has largely been neglected. Transitions from stationary to a self-selected pace occurs countless times daily. Clinicians may be inclined to assess this phase of propulsion in conjunction with steady state to provide a more comprehensive picture of wheelchair propulsion. The dual purposes of this study were to 1) examine the relationship between selected subject and wheelchair characteristics and start-up propulsion kinetics; 2) determine the effect of selected surfaces on start-up propulsion kinetics



Participants were recruited from the 2004 National Veterans Wheelchair Games in St. Louis, Missouri. Twenty-two individuals who utilized an ultralight manual wheelchair for their primary mode of mobility completed propulsion kinetics evaluation. Participants averaged 45.4 (+/- 8.5) years of age, 98.9 (+/- 26.9) kg, and all were male (Table 1). Diagnoses included 15 individuals with spinal cord injuries (C6-L5, complete and incomplete), 2 with leg amputations, 1 with MS, 1 with a TBI, and 2 with miscellaneous diagnosis. Wheelchair axle position averaged 4.5 (+/-2.9) cm horizontally and 14.1 (+/-3.1) cm vertically as measured from a virtual point (Figure 1.). The virtual point was the midpoint intersection of the seat and backrest plane.

Table 1. Selected subject and wheelchair characteristics. Vertical and horizontal axle positions are the horizontal and vertical distances from the virtual intersection point to the rear axle. The virtual intersection point is where midpoint of the intersection of the seat and backrest planes (Fig 1).

Subject and Wheelchair Characteristics Mean +/-SD
Subject Weight (kg) 98.9 +/-26.9
Wheelchair Weight (kg) 16.8 +/-4.1
Combined Weight (kg) 115.7 +/-28.7
Horizontal axle position (cm) 4.5 +/-2.9
Vertical axle position (cm) 14.1 +/-3.1


Participant propulsion kinetics was evaluated via a Smart Wheel mounted on the dominant side, which for all individuals was the right. Each participant propelled from a stationary position to a self-selected velocity over a sequential series of surfaces along a straight path. Three surfaces, tile, low pile carpet, and a ramp, were selected for this analysis. Pushrim kinetics was recorded for the initial ten strokes completed.

Data Analysis

The initial two strokes were selected for separate analysis as representative of start-up propulsion. Correlation analyses were conducted to determine potential relationships between subject (weight) and wheelchair (weight and horizontal and vertical axle position) characteristics and rate of rise (ROR) of peak resultant force (p=.01). Subject and wheelchair weight were combined, producing a single variable. Paired t-tests were performed to identify differences in peak resultant force between the initial two strokes and the three surfaces (p=.05).


Figure 1. Virtual point for wheelchair measurements (Click image for larger view)
This figure shows where the virtual point exists in a generalized wheelchair.  The virtual point is the intersection point of the backrest and seat planes.

Combined subject and wheelchair weight were not correlated to ROR of peak resultant force for any of the three surfaces or two strokes. Peak resultant force for both strokes was significantly higher for the ramp when compared to the tile and low pile carpet (Table 2). Tile and low pile carpet peak resultant force were statistically similar. For all surfaces, peak resultant force was statistically similar across stroke one and stroke two.

Table 2 . Peak resultant force for strokes 1 and 2 across 3 surfaces. Asterisk denotes statistical significance p=.01 of between surface comparisons.
Peak Resultant Force (N) Mean (SD)
  Stroke 1 Stroke 2
Tile 106.31(33.5) 103.42(32.2)
Carpet 107.8(35.9) 104.0(32.1)
Ramp 129.6*(32.2) 136.3*(22.8)


 Previous research has indicated a relationship between peak propulsion forces and subject and wheelchair characteristics. Large propulsive forces are associated with the prevalence of upper extremity injury in manual wheelchair users [4]. User weight has demonstrated a positive correlation with propulsion forces during steady state propulsion on dynamometers. Additionally, axle position has proven to be directly correlated to the rate of rise of the resultant force, which may be related to upper extremity injury [2]. Contrary to previous research, the results of this analysis indicate weight was not correlated with peak resultant force during start-up propulsion. Axle position was also not correlated with ROR during start-up. It appears the traditional associations between weight and propulsion forces and wheelchair configuration and propulsion forces fail to hold true during start-up propulsion in over-ground testing.

A relatively small sample size limited the power of this analysis. Weight and axle position are related to rolling resistance and therefore kinetics. The relationship between weight, axle position, and propulsion kinetics may have been found with a larger sample size. The variability of participant’s diagnosis increases the ability to generalize these results, but could also mask potential relationships. Previous research has generally focused on a homogeneous group selected to reduce the variance within the group.

External resistance to propulsion has an observable effect on peak resultant force. Peak resultant force during start-up was similar in trials over low-pile carpet and tile, but was significantly higher on a ramp. Clinicians should evaluate all clients as they transverse a ramp, which by requiring greater propulsion forces may facilitate the identification of functional deficits. Such deficits could be minimized through alterations in wheelchair configuration. The protocol utilized in this study is easily reproducible in a clinic with the use of a single Smart Wheel mounted on the client’s dominant side. Such a protocol could provide a standardized, objective method by which to evaluate propulsion kinetics.


This study was supported by the VA RR+D Center (F2181C).

Reference List

  1. Boninger, M. L., Dicianno, B. E., Cooper R.A., Towers, J. D., Koontz, A. M., & Souza, A. L. (2003). Shoulder MRI Abnormalities, Wheelchair Propulsion, and Gender. Archives of Physical Medicine and Rehabilitation, 84, 1615-1620.
  2. Boninger, M. L., Baldwin, M. A., Cooper, R. A., Koontz, A. M., & Chan, L. (2000). Manual Wheelchair Pushrim Biomechanics and Axle Position. Archives of Physical Medicine and Rehabilitation, 81.
  3. Boninger, M. L., Towers, J. D., Cooper R.A., Dicianno, B. E., & Munin, M. C. (2001). Shoulder Imaging abnormalities in Individuals with Paraplegia. Journal of Rehabilitation R & D, 38, 401-408.
  4. Boninger, M. L., Cooper, R. A., Baldwin, M. A., Shimada, S. D., & Koontz, A. (1999). Wheelchair pushrim kinetics: body weight and median nerve function. Archives of Physical Medicine and Rehabilitation, 80, 910-915.
  5. Kotajarvi, B. R., Sabick, M. B., An, K. N., Zhao, K. D., Kaufman, K. R., & Basford, J. R. (2004). The effect of seat position on wheelchair propulsion biomechanics. Journal of Rehabilitation R & D, 41, 403-414.