Link to Figures Show Figures
	Article Text Figures References
	Send Link Send HTML Article

Use of a Step Activity Monitor in Determining Outcomes

David Alan Boone, BS, CP, MPH, PhD
Kim Lisa Coleman, BA, MS

ABSTRACT

Step activity is a natural measure of mobility function that is well suited to questions of prosthetic and orthotic outcomes. Step count data are relatively easily comprehended and interpreted. The StepWatch™ Activity Monitor (SAM) is an instrument that was developed to accurately measure the impact of various interventions on the real-world activity level of individuals. It counts and records the number of steps taken in short time intervals (typically 1 minute) for up to 2 months continuously. The patient or subject is not required to do anything but wear the monitor on the leg. Many independent studies have validated the high accuracy of the SAM in populations with greatly varying gait styles and abilities.

Since the late 1970s, recording step counters have been used with prosthesis users to determine levels of activity as a measure of functional status. In a review of instruments for measuring physical activity, Warren Tryon contends, "It can be argued that the step is the preferred unit of measure [for activity monitoring] because it is a natural unit of ambulation" (p 38).¹

Commercially available pedometers are inexpensive but are limited because they yield only lump-sum measures and their displays provide feedback to subjects that may motivate non-normal behavior. In addition, pedometer accuracy has been reported to vary considerably among subjects^2–4 and to be particularly questionable for those with gait disorders.⁵ Accuracy is paramount in providing the statistical power required for measuring small but potentially important effects that are likely smaller than the error of less accurate pedometers.

Holden et al.⁶ created a device to collect long-term step activity data for subjects with lower limb amputation. They measured daily step activity with a foot switch embedded in the prosthetic foot and a storage unit attached to the tubular pylon of endoskeletal prostheses. The storage unit was read once a day, and the variability in daily step totals was used to evaluate progress in rehabilitation programs. Day⁷ used a similar device and calculated daily step totals to validate an activity questionnaire for persons with lower limb amputation. Building on the successful demonstrations of step counting as a measure suitable for prosthetics, others used daily step counts collected with a variety of devices to assess the extent of prosthetic limb use.⁸

In 1991, Prosthetics Research Study in Seattle began development of the StepWatch™ Activity Monitor (SAM; Figure 1 ). The SAM has become a reliable and exceptionally accurate instrument for the assessment of gait functionality. It was designed to overcome the limitations of previously available monitoring instruments by recording a detailed minute-byminute profile of a well-defined event for extended periods (Figure 2 ).^9,10 Step activity is more than just step counts. One should also be interested in how the counts are accumulated across time, in terms of rates or intensity of activity and patterns of activity versus inactivity. Figure 3 illustrates how an overall step count per day can be an incomplete descriptor of function. In this case, two subjects with nearly the same step count accumulated their steps in very different manners (Figure 3 ). Subject 1 added step counts intermittently, never achieved high level activity, and demonstrated little sustained activity. Subject 2 accumulated and sustained high level activity, illustrating how data may provide more useful information when recorded in relatively short time intervals. It may be this type of difference that is important to measure when, for example, describing the function of a new prosthesis designed for higher activity.

A device similar to the StepWatch has been introduced by Φssur (Reykjavik, Iceland) for use in prosthetics applications. Their Patient Activity Monitor, or PAM, is also worn on the leg and records ambulatory activity. The PAM Index calculated from the data gathered was reported to correlate well with estimated activity level used for component selection.¹¹

RATIONALE

The basic rationale behind step count-based outcome measures is that increases in activity are seen as a sign of success for physical rehabilitation. In addition, steps are a measure that is understood intuitively. It is very valuable to be able to relate the outcome measurement to easily perceived human activity.

Monitors such as the StepWatch also give us a window on real-world, continuous performance. This helps to reduce the effect of potentially unnatural performance in a controlled laboratory setting. Although the conditions of laboratory measures are more easily controlled, the act of controlling them may mask some outcomes of importance. For instance, if gait speed is controlled when measuring gait parameters, one may miss the outcome of the subject manifesting the ability to walk at higher maximum speeds occasionally. Continuous long-term monitoring captures these types of choices to use different ambulatory capabilities.

DESCRIPTION OF INSTRUMENT

SAM is a research-grade instrument for long-term assessment of ambulatory activity during day-to-day life. It is a small (70 X 50 X 20 mm; 38 g), waterproof, self-contained device that is worn on the ankle and records the number of strides taken every minute, for as long as 2 months between downloads.

The monitor is programmed and downloaded with a standard computer via a docking station that plugs into a USB port. The SAM and docking station communicate through an infrared link that allows the SAM to be completely sealed and impervious to tampering. The unit can be factory refurbished for battery replacement. The predicted battery life with continual use is 7 years.

The SAM provides no feedback to the subject, so it does not encourage "performance behavior."

ACCURACY

Many independent evaluations have concluded that the SAM is an accurate measure of physical activity available.^12–21 Step detection accuracy exceeds 98% both for unimpaired gait and for movement styles that traditionally have been difficult to monitor accurately, such as geriatric shuffling, hemiparetic gait, and dyskinetic gait. These findings are summarized in Table 1 . The sensitivity of the SAM is tuned to each subject by specifying the subjects' height and gait characteristics during programming. The appropriateness of the settings can be informally verified by watching a test light on the monitor blink every time a step is detected or by a formal accuracy trial that compares observers' counts to monitor counts.

Unlike most pedometers, accuracy is unaffected by soft tissue at the waist, so even extremely obese subjects can be monitored accurately.^13–19 Because the StepWatch continuously collects accurate data over long periods, statistically significant differences may be found with lower numbers of subjects than for other activity metrics, clinical measures, or questionnaire-based data. For many studies, time spent at moderate activity levels has tended to be an especially sensitive measure.

DATA

The SAM data reflect the number of strides per minute (e.g., steps taken by the leg being monitored across the entire monitoring period). Stride counts are doubled to represent the number of steps across the entire monitoring period. Visual inspection of the data plots can be used to verify whether the subject was compliant with monitoring. Days or parts of days may be excluded from the results if the subject was not compliant in wearing the monitor. By default, statistics are calculated for each full 24-hour day of data, then averaged across the included days. Intensity measures such as time at moderate activity are often more sensitive to change than overall activity level. The primary measures calculated are listed in Table 2 .

USE OF THE SAM FOR OUTCOMES RESEARCH

The SAM has been used in prosthetics and orthotics research and clinical care since 1996. Its use is steadily expanding in related fields of rehabilitation and medicine. The instrument is reliable, easy to use, and tolerated well by patients and research subjects. A wide variety of outcome studies using SAM are summarized.

LOWER LIMB AMPUTATION AND PROSTHETICS

The StepWatch initially was developed for use in lower limb prosthetics outcomes research. Early validation and use reflect this application. It continues to be used to study the effectiveness of new prosthetic components such as microprocessor- controlled knees,^22–24 shock absorbing pylons,²⁵ socket liners,^26–28 and to document funtion,^23,30–32 and track rehabilitation progress.^30,33

STROKE

As life expectancy within the United States increases, accurately assessing the functional capacity and quality of life of persons affected by conditions such as stroke becomes increasingly important. Few activity monitoring technologies are appropriate in this gait-impaired population.¹⁵ Recent studies describe effective use of the SAM in this population. ^15,21,34

DIABETES

The StepWatch has been extensively used and validated in populations with diabetes and peripheral neuropathy. ^14,35–40 It is a well-tolerated and appropriate tool for research and clinical investigations within this sector. For persons with diabetes, the SAM is typically worn on the leg with sock-like cotton Lycra cuffs, rather than standard Velcro straps.

MUSCULAR DYSTROPHY, MULTIPLE SCLEROSIS, MYELOMENINGOCELE, OTHER NEUROLOGICAL DISORDERS

Researchers at University of California at Davis Medical Center have studied the interaction between Duchenne's muscular dystrophy and physical activity for many years. Descriptive data have been published for other neurological conditions.^41–49 Studies of orthotically treated populations may also benefit from measuring activity level as an outcome of treatment.

CHALLENGES

Step count-based outcome measures, although conceptually easy to grasp, are not without their challenges. Living habits and environment play a large part in how much we ambulate in our daily lives. If one's habits or environments don't change, for instance with choosing to increase exercise, then the step activity profile is constrained to the original habits of the person being monitored. With a prosthetic or orthotic change, the habits and environment of the individual may overwhelm any biomechanical advantage given to the user. In other words, the user may simply choose not to make use of the advantage available to him/her.

We step to move from place to place. It would seem that a decrease in steps means that we are not interacting as much with our environment. However, if there were an increased stride length because of an intervention, there would be a concomitant decrease in the number of steps required to cover the same distance. Likewise, increased speed should decrease the duration of ambulation. How treatment affects gait parameters should always be considered in conjunction with measuring step counts.

CONCLUSIONS

Step activity is a natural measure of mobility function that is well suited to questions of prosthetic and orthotic outcomes. Step count data are relatively easily comprehended and interpreted. These data can be used in concert with other measures to facilitate or validate interpretation of results.

Step activity monitoring also can be used as an effective measure of compliance with use of a prosthesis or orthosis and for verification of the level of activity of the user. The user should be clear about whether the goal is measuring or motivating activity because one can do either with the data. The SAM can measure without providing feedback to the user minimizing test bias. However, the ability to give feedback to the wearer is a powerful tool that could be used to improve outcomes through motivation.

Step activity measures should be made for at least 1 to 2 weeks to get a true representation of day-to-day variability and weekly routine of the subject. Patterns of activity such as the duration of moderate or high level activity as a percentage of total activity are important indicators of changes in functional status and generally are more powerful than are overall step counts. Within research and clinical practice, accuracy is the paramount requirement for measurment of step counts both in formal studies and for individual assessments. High levels of accuracy assure that subtle but functionally important changes in activity profiles are not obscured by error of the measurement.

Correspondence to: Kim Coleman, CYMA Corporation, Suite 100, 6405 218th Street SW, Mountlake Terrace, WA 98043; e-mail: .

DAVID ALAN BOONE, BS, CP, MPH, PhD, is associate professor in the Rehabilitation Engineering Centre, Department of Health Technology and Informatics, The Hong Kong Polytechnic University.

KIM LISA COLEMAN, BA, MS, is president of CYMA Corporation, Mountlake Terrace, WA.

References:

1. Tryon WW. Activity Measurement in Psychology and Medicine. New York and London: Plenum Press; 1991.
2. Washburn R, Chin MK, Montoye HJ. Accuracy of pedometer in walking and running. Res Q Exerc Sport 1980;51 (4):695–702.
3. Gayle R, Montoye HJ, Philpot J. Accuracy of pedometers for measuring distance walked. Res Q 1977;48:632–636.
4. Shepherd EF, Toloza E, McClung CD, et al. Step Activity Monitor: increased accuracy in quantifying ambulatory activity. J Orthop Res 1999;17 (5):703–708.
5. Kochersberger G, McConnell E, Kuchibhatla MN, et al. The reliability, validity, and stability of a measure of physical activity in the elderly. Arch Phys Med Rehabil 1996;77 (8):793–795.
6. Holden J, Fernie GR, Soto M. An assessment of a system to monitor the activity of patients in a rehabilitation programme. Prosthet Orthot Int 1979;3:99–102.
7. Day HJB. The assessment and description of amputee activity. Prosthet Orthot Int 1981;5:23–28.
8. Stewart CPU. Walking activity in primary lower limb amputees in the first 6 months after surgery. Saudi J. Disabil Rehabil 1998;4:106–112.
9. Smith DG, Joseph AW, Boone DA, et al. Gait Activity Monitor [Patent]. Prosthetics Research Study, USA. Patent No. 5. 1996; 485:402.
10. Coleman KL, Smith DG, Boone DA, et al. Step Activity Monitor: long-term, continuous recording of ambulatory function. J Rehabil Res Dev 1999;36:8–18.
11. Karason G. PAM Index: objective activity measurement. Poster presented at AOPA Annual Meeting, Fort Lauderdale, September 2004.
12. Coleman KL, Boone DA, Smith DG, et al. Quantification of prosthetic outcome: FlexWalk-II foot vs. SACH foot. Proceedings 1st National Meeting Rehabilitation Research and Development Service, Washington, DC, October 1-3, 1998, p. 197.
13. Shepherd EF, Toloza E, McClung CD, et al. Step Activity Monitor: increased accuracy in quantifying ambulatory activity. J Orthop Res 1999;17 (5):703–708.
14. Hartsell H, Fitzpatrick D, Brand R, et al. Accuracy of a customdesigned activity monitor: indications for diabetic foot ulcer healing. J Rehabil Res Devel 2002;39 (3):395–400.
15. Macko MF, Haeuber E, Shaughnessy M, et al. Microprocessorbased ambulatory activity monitoring in stroke patients. Med Sci Sports Exer 2002;34 (3):394–399.
16. Resnick B, Nahm ES, Orwig D, et al. Measurement of activity in older adults: reliability and validity of the Step Activity Monitor. J Nurs Measure 2001;9 (3):275–290.
17. Brandes M, Rosenbaum D. Correlations between the Step Activity Monitor and the DynaPort ADL-monitor. Clin Biomech 2004;19 (1):91–94.
18. Bjornson K, Song K, Coleman K, et al. Application of the Step Activity Monitor (SAM) to define normal activity levels in children. Develop Med Child Neurol 2000;42 (Suppl 83):23.
19. Foster RC, Lanningham-Foster LM, Manohar C, et al. Precision and accuracy of an ankle-worn accelerometer-based pedometer in step counting and energy expenditure. Prev Med 2005;41 (3-4):778–783.
20. Karabulut M, Crouter SE, Bassett DR Jr . Comparison of two waist-mounted and two ankle-mounted electronic pedometers. Eur J Appl Physiol 2005;95 (4):335–343.
21. Haeuber E, Shaughnessy M, Forrester LW, et al. Accelerometer monitoring of home- and community-based ambulatory activity after stroke. Arch Phys Med Rehabil 2004;85 (12):1997–2001.
22. Kaufman KR, Padgett D, Brey RH, et al. How does the OttoBock C-Leg compare to the Mauch SNS? Annual Meeting of the Gait and Clinical Movement Analysis Society, Portland, OR, April 6-9, 2005.
23. Orendurff MS. Design, delivery and assessment: from benchtop science to outcomes. Proceedings of the 11th World Congress of the International Society for Prosthetics & Orthotics, Hong Kong, August 1-6, 2004, p 78.
24. Kaufman K, Padgett D, Brey RH, et al. Objective comparison of the OttoBock C-Leg to the Mauch SNS. Proceedings of the 11th World Congress of the International Society for Prosthetics & Orthotics, Hong Kong, August 1-6, 2004, p 157.
25. Coleman KL, Boone DA, Laing LS, et al. Quantification of prosthetic outcomes: elastomeric gel liner with locking pin suspension versus polyethylene foam liner with neoprene sleeve suspension, J Rehabil Res Devel 2004;41 (4):591–602.
26. Berge JS, Klute GK, Czerniecki JM. Mobility and comfort for unilateral transtibial amputees using a shock-absorbing pylon. Journal of Proceedings for the 2004 Annual Meeting for the American Academy of Orthotists and Prosthetists, New Orleans, LA, February 25-28, 2004.
27. Smith DG, Coleman KL, Boone DA. Improved activity following elective transtibial amputation for pain: outcome assessment by long-term activity monitoring (a case report). In: Soede M, Verbout AJ, Swart M, eds. IXth World Congress of the International Society for Prosthetics and Orthotics, Amsterdam, The Netherlands, June 28-July 3, 1998, pp 728-730.
28. Coleman KL, Smith DG, Joseph AW, et al Gait Activity Monitor: a device for objectively evaluating the functional outcome of prosthetic, orthotic or medical treatment. Abstracts of the International Society for Prosthetics and Orthotics 8th World Congress, Melbourne, Australia, April 3-7, 1995, p 74.
29. Coleman KL, Boone DA, Smith DG, et al Functional outcome assessment using objective measures of patient satisfaction and gait activity: a comparison of 2 feet. In: Soede M, Verbout AJ, Swart M, eds. IXth World Congress of the International Society for Prosthetics and Orthotics, Amsterdam, The Netherlands, June 28-July 3, 1998, pp 690-692.
30. Coleman K, Boone D, Smith DG, et al Relative influence of changing the prosthetic foot vs. the socket suspension system on ambulatory activity and patient satisfaction. Proceedings of the 10th World Congress of the International Society for Prosthetics and Orthotics, Glasgow, Scotland, July 2-6, 2001.
31. Coleman K. Distinctive differences in duration and rate profiles of ambulation between populations with transtibial amputation, with diabetes, and without impairment. Proceedings of the 10th World Congress of the International Society for Prosthetics and Orthotics, Glasgow, Scotland, July 2-6, 2001.
32. Coleman K, Boone D, Smith DG, et al. Crossover trial comparing Alpha liner with Pelite liner for transtibial prostheses using ambulatory activity and questionnaire responses. Proceedings of the 10th World Congress of the International Society for Prosthetics and Orthotics, Glasgow, Scotland, July 2-6, 2001.
33. Coleman K, Boone D, Smith DG, et al. Progress following elective transtibial amputation: one year of continuous ambulatory monitoring. A Case Study. Proceedings of the 10th World Congress of the International Society for Prosthetics and Orthotics. Glasgow, Scotland, July 2-8, 2001.
34. Shaughnessy M, Michael KM, Sorkin JD, et al. Steps after stroke: capturing ambulatory recovery. Stroke 2005;36: 1305–1307.
35. Del Aguila M.. Assessment of physical activity in patients with diabetes [dissertation]. University of Washington, 1998.
36. Saltzman CL, Zimmerman MB, Holdsworth RL, et al. Effect of initial weight-bearing in a total contact cast on healing of diabetic foot ulcer. J Bone Joint Surg 2004;86-A (12): 2714–2719.
37. Lott DJ, Bohnert KJ, Klaesner JW, et al. Amount and intensity of activity during off-loading treatment and the healing of neuropathic plantar ulcers. J Orthop Sports Phys Ther 2004;34 (1):A –24.
38. Smith DG, Domholdt E, Coleman KL, et al. Ambulatory activity in men with diabetes: relationship between self-reported and real-world performance-based measures. J Rehabil Res Devel 2004;41 (4):571–580.
39. Maluf K. Preventing ulcer recurrence in patients with diabetes: experimental application of the physical stress theory [dissertation]. Movement Science, Washington University, 2002.
40. Maluf KS, Mueller MJ. Novel Award 2002. Comparison of physical activity and cumulative plantar tissue stress among subjects with and without diabetes mellitus and a history of recurrent plantar ulcers. Clin Biomech 2003;18:567–575.
41. McDonald CM, Widman LM, Walsh DD, et al. Use of step activity monitoring for continuous physical activity assessment in boys with Duchenne muscular dystrophy. Arch Phys Med Rehabil 2005;86 (4):802–808.
42. McDonald CM, Walsh DD, Widman LA, et al. Quantitative assessment of community physical activity levels in disabled children with the Step Activity Monitor. Arch Phys Med Rehabil 2000;81:1273.
43. Widman L, Walsh DD, Walsh SA, et al. Use of the Step Activity Monitor for continuous objective physical activity assessment in children with Duchenne muscular dystrophy. Devel Med Child Neurol 1999;41 (Suppl 80):20.
44. Pearson OR, Busse ME, van Deursen R, et al. Quantification of walking mobility in multiple sclerosis using an ambulatory activity monitor: a pilot study. J Neurol Neurosurg Psych 2003;74 (10):1450–1451.
45. Busse ME, Pearson OR, Van Deursen R, et al. Quantified measurement of activity provides insight into motor function and recovery in neurological disease. J Neurol Neurosurg Psych 2004;75 (6):884–888.
46. Prim L, Novak R, Witka J, et al. Ambulatory activity and oxygen consumption: is there a relationship in patients with myelomeningocele. Develop Med Child Neurol 2002;44 (Suppl 91):19.
47. Pearson OR, Busse ME, Van Deursen RWM, et al. Quantification of walking mobility in neurological disorders. Q J Med 2004;97:463–475
48. Lawthom C, Van Deursen RMW, Wiles CM. An improved method of quantifying activity levels in neurological patients. J Neurol Neurosurg Psych 2002;72:140.
49. McDonald CM. Physical activity, health impairments, and disability in neuromuscular disease. Am J Phys Med Rehabil 2002;81 (11 Suppl):S108–S120.