In her book, When Walking Fails, Lisa Iezzoni, MD, describes the fascination our culture has with "walking." She points our attention to famous quotes and everyday phrases, "one small step for man, one giant leap for mankind" (astronaut Neil Armstrong), "every journey begins with one step" (Chinese proverb). Our "pull yourself up by your boot straps" society and, more specifically, our "can cure that" health care system are very uncomfortable when there may not be a cure and the resulting functional limitation is either nonambulatory or significantly limited ambulation. Yet, many people face the condition of not walking every day, and have no trouble "getting around." Wheeled mobility is an alternative mobility method used by thousands of very content and highly functional people "to get things done."

For individuals with certain impairments, such as complete spinal cord injury (SCI), a wheelchair is the best method of mobility. For others who have reduced walking abilities, secondary to weakness, lack of coordination, and even shortness of breath, a wheeled mobility device may provide augmentative mobility.

As therapists, we are concerned about a person's mobility. How are they getting around? Is the method safe? Are they functional? When evaluating an ambulatory client, we determine the person's function via both objective and subjective methods. We observe a person's gait pattern. We evaluate balance, speed, and coordination. Effort is put into "gait training," with consideration given to the need for an orthotic or an ambulation aid to improve stability. If an orthotic is needed, it is custom fitted and modified by a clinical team and further gait training is provided. All of this is done because, if not done correctly, patient outcomes are poor and medical costs increase.

Do we put the same emphasis on the patient who will use a wheeled mobility device for functional mobility? What do we do to "test" wheelchair riders? Can we see an individual "limp" while propelling a wheelchair, like we can with those who are walking? Do we know when some one is "functional" in a mobility device? For instance, can the person "self-propel"? When do you recommend manual and when do you recommend power? How can we protect the upper extremities from overuse injuries? Many of the answers to these questions have been based on "clinical judgment" and not objective measurement.

Demonstrating the efficient range of stroke length—how long an arc the hand travels for a push stroke.

CLINICAL DECISION-MAKING GUIDELINES

In an article published last year, Rachel Cowan, PhD, and her colleagues¹ began to provide answers to many of these questions by providing a suggested framework to guide wheelchair-related clinical decision-making. Much of the work presented in the article is from their work in objectively measuring manual wheelchair propulsion with a device called the SmartWheel. The SmartWheel is an instrumented wheelchair wheel that measures manual propulsion, much like a gait lab measures the biomechanics of walking.

A group of researchers, clinicians, and wheelchair riders who share an interest in wheelchair biomechanics, most of whom use a SmartWheel regularly, formed the SmartWheel Users' Group (SWUG) in 2004. The SWUG is a forum in which participants share experiences, realities, and challenges related to understanding wheelchair biomechanics and advancing its clinical application. The group recognized the need to begin addressing wheeled mobility in the same way that gait analysis, orthotic fitting, and training have been prioritized. The group identified the need to collect wheelchair propulsion data from multiple centers that can be used for comparative reference. The SWUG established a standard clinical protocol for wheelchair propulsion measurement to be used for data collection, which can be compared center to center and pooled in a database for analysis. The clinical protocol calls for four conditions to measure a person propelling a chair:

1. Across tile (protocol recommendation is 10 meters):
   • Testing a low friction surface—little resistance—easier to push
2. Across carpeting (protocol recommendation is 10 meters):
   • Testing a high friction surface—greater resistance—harder to push
3. Up a ramp:
   • Testing aggressive propulsion
4. Around a "figure 8" cone pattern:
   • Testing speed and maneuverability

Patient sits in a wheelchair fitted with a SmartWheel.

The SmartWheel captures the push mechanics used by the manual wheelchair rider when operating a chair in each of these conditions. Like many measurement tools, the amount of data available from the SmartWheel can be overwhelming. To address this, the SWUG identified four measurements as most clinically relevant:

1. Speed (velocity)—How fast is the person pushing the chair, after the initial "start-up" strokes?
2. Force—How much force is the person applying to the push rim to generate the speed for propulsion?
3. Push Frequency—How often does the person contact the push rim to generate the force to achieve the speed?
4. Stroke Length—How big an arc does the hand travel from initial contact on the push rim to release of the push rim during one push stroke?

For an individual client, the SmartWheel clinical software generates both a graphical display and a summary report that shows the results of each of these key measures for each propulsion session.

In a clinic situation, without a SmartWheel, there are relatively easy ways to measure or observe three out of four of these key measures:

1. Speed—mark out a distance (10 meters or 33 feet) and use a stop watch to time the person. The speed is simply a division of distance/time expressed as either meters/second or feet/second
2. Push Frequency—over the same marked out distance, simply count how many times a person contacts the push rim.
3. Stroke Length—the length of the arc from initial contact on the push rim (which is a 360º circle) until release. The length of the arc can be expressed in number of degrees—zero being the spot of initial contact—or expressed as points on the clock at contact and release—eg, on at 10 and off at 2.

Observing and measuring force, however, is not nearly as easy. While we can see when a person is likely propelling with excessive or minimal force, it is difficult, if not impossible, to quantify by observation alone. When propelling, there are actually several directions of forces working together to create (or obstruct) forward movement. The key summary measure that the SmartWheel is able to calculate is the "average peak resultant force," which is the overall average of peak (maximum) forces exerted on the push rim.

The velocity test measures how fast a person gets from point A to point B.

For clinical purposes, if there is not enough force to push at walking speed, then the person will not be functional in a manual chair. At the other end of the scale, if the person is "pushing too hard," it can result in upper extremity (UE) injury and pain. Earlier lab studies have shown that those wheelchair riders who generate high force on the push rim during propulsion are more likely to have UE impairments, including pain, rotator cuff tears, and carpal tunnel syndrome. UE compromise is likely to worsen with long-term wheelchair use.

Clinically, the goal for safe, effective, manual wheelchair propulsion is to use long, smooth strokes with as low a force as possible to generate speeds that are, at the very least, as fast as walking speed.

The clinical practice guideline from the Paralyzed Veterans of America (PVA), titled "Preservation of Upper Limb Function Following Spinal Cord Injury: A Clinical Practice Guideline for Health-Care Professionals," published in 2005, outlines 35 recommendations for clinical practice to reduce or prevent UE impairments for persons with SCI. These recommendations are based on a rigorous review of published evidence. The guideline has two main themes: Reduce force and reduce repetition (frequency).

Hence, in the area of equipment selection and training, the guideline emphasizes the need to reduce both force and frequency when propelling a manual chair. Clearly, with any attempt to "reduce" force and frequency, both must be measured to gauge effectiveness of clinical interventions. The SmartWheel provides an objective measure of the effectiveness of a variety of interventions intended to reduce force and frequency. Interventions may include changing the setup of the wheelchair (eg, moving the rear wheel forward), propulsion training to use long, smooth strokes, or moving someone from a standard, fixed axle chair to an ultralight adjustable chair.

Importantly, objective measures of wheelchair propulsion can also help determine whether an individual has sufficient strength and coordination to self-propel at a functional speed (eg, 1.3 meters/second, a typical walking speed) and/or at a minimally safe speed (1.06 m/s, the speed required to cross a street with a timed light). If not, then the prescription of a power wheelchair may be warranted.

In their article, Cowan and her colleagues also introduce the database that has been developed as a result of the collection of multicenter data from participants in the SWUG, all using the clinical protocol to collect data. The database is now able to generate normative data about "desired" manual propulsion mechanics (eg, typical propulsion frequencies, forces, and speeds).

Now, as a result of the availability of the database, an individual's performance can be plotted on a graph (see Figure 1 below) to establish this person's performance relative to all the other users in the database (the green semicircular area and the diagonal line). The data—collected for an individual client—either Average Speed versus Push Frequency or Average Speed versus Force (normalized for the person's body weight), is plotted on the graph.

Figure 1. Copyright © 2008 Three Rivers Holdings LLC. The data presented here is for informational purposes only.

The graphs were created because it is critical to view push frequency and force in the context of how fast the person in the wheelchair is moving. For example, a client might have low forces and a low push frequency, which if viewed in the absence of speed might seem good. However, if that person is moving at below a functional speed (velocity), that is not acceptable. So the graphs enable you, the therapist, to rapidly assess your client compared to the database, while considering speed.

There are two horizontal dotted lines on each graph, representing two levels of functional pushing speed referred to above. The lower dotted line represents a minimum functional speed—defined as the speed needed to cross an intersection. The higher line on the graph represents the functional speed needed to match the normative walking speed of an adult. These threshold lines remind clinicians that if a person cannot push at least at a minimum functional speed (eg, the speed it takes to safely cross an intersection), then perhaps power mobility should be considered. Pushing below the threshold speed, regardless of frequency or force, will not be functional.

The two graphs in Figure 1 illustrate the data from four test sessions. In the graph on the left side, data from sessions 1 and 3 (data points displayed as a circle) are acceptable for functional speed and push frequency. Acceptable is good; however, there still may be opportunities for improvement that you as a therapist might identify. For example, data point 3 is just about functional velocity so interventions to increase speed could be examined.

Sessions 2 and 4 are not acceptable, as the speed achieved falls below minimal functional threshold. These data points are displayed as arrows. The direction of the arrow provides a visual prompt on which direction the results should move in subsequent trials. For example, data in the arrow marked by "4" indicates the clinician may want to consider an intervention that increases the person's speed, to move up above the dotted line. The frequency of push rim contacts is acceptable, but the speed is too slow.

The graph on the right side shows "Speed versus Force" (normalized for weight). For persons who push at a speed above the top line, it is important to know how much force they are applying to the push rims to achieve that speed. In this example, data point 3 is illustrating speed just above the minimum threshold, but using a higher level of force than people in the database (outside the norm). The clinician is prompted to consider an intervention to reduce the force needed to achieve that speed, using less force and thus protecting the UE over time. Ideally, data points should be located in the area of the upper left side of the graph—illustrating high speed with as low a force and as low a push frequency as possible to achieve that speed. Data to the right side of the graph illustrate pushing too hard and/or pushing too often.

In their article, Cowan and her colleagues present a flowchart that links the plotted data with the possible outcomes from clinical interventions (see Figure 2 below).

As we noted earlier, clinical interventions can include a wide range of activities including:

1. Improving the user's posture when sitting in the chair;
2. Changing the setup of an existing chair;
   1. Adjusting the rear wheel forward;
   2. Changing the vertical position of the wheel;
3. Introducing a completely new chair—fully adjusted and optimally configured;
4. Providing training on stroke mechanics, seeking a long smooth push; and
5. Strength training.

Figure 2. Clinician Decision-Making Flow Chart*

*Adapted from Cowan RE, Boninger ML, Sawatzky BJ, Mazoyer BD, Cooper RA. "Preliminary Outcomes of the SmartWheel Users' Group Database: A Proposed Framework for Clinicians to Objectively Evaluate Manual Wheelchair Propulsion." In Press in the Archives of Physical Medicine and Rehabilitation.

INTERVENTION AND FOLLOW-UP

With each intervention and follow-up assessment, the SmartWheel data provides objective measures to understand the impact of the intervention. As is indicated in the flow chart:

• The first goal is to increase speed (velocity) to above the threshold speed;
• The second goal is to reduce the force;
• The third goal is to keep the force low using an acceptable frequency.

When the resultant data is plotted in the previously described graphs, the clinician has objective data indicating that the interventions have achieved the goal of manual wheelchair propulsion at a functional speed without excess force or frequency. This data is invaluable in patient education and justification for clinical treatment time, as well as for justification for new or modified equipment or training.

Use of a wheelchair can no longer be thought of as a failure to achieve upright ambulation. For many, wheelchair use is the most appropriate mobility option; for them, the wheelchair is an enabling technology that enhances independence and quality of life. However, research has shown that pushing a wheelchair can compromise long-term preservation of the upper extremity. To minimize this known risk, persons who rely on a manual wheelchair need to have access to knowledgeable clinicians to assist in equipment selection, setup, and training.

We need to continue to generate clinically relevant evidence to support and advance our clinical practices. New high-tech measurement devices such as the SmartWheel provide valuable objective information, but alone they are just tools. The real value is in the data generated by the tool and the ability of the clinician to use the data to guide important clinical interventions. A well-t rained, well-equipped wheelchair rider will quite quickly gain the mobility needed to "get things done"—even if they are not able to walk.

Jean L. Minkel, PT, ATP, directs the firm Minkel Consulting, New Windsor, NY. The firm provides consultation services to professionals in the field of assistive technology. For more information, Minkel can be reached at .

REFERENCE

1. Cowan RE, Boninger ML, Sawatzky BJ, Mazoyer BD, Cooper RA. Preliminary outcomes of the SmartWheel Users' Group database: a proposed framework for clinicians to objectively evaluate manual wheelchair propulsion. Arch Phys Med Rehabil. 2008;89:260-8.