Schmidt, Amy L. PT, DPT1; Pennypacker, Melissa L. PT, DPT2; Thrush, Aaron H. PT, DPT3; Leiper, Carol I. PT, PhD4; Craik, Rebecca L. PT, PhD, FAPTA4
People with Parkinson disease (PD) and those with multiple sclerosis (MS) have difficulty in ambulating in the community.1 Measuring how much walking is done in home and community environments may help health care professionals understand the functional abilities and prognosis for people with chronic diseases such as PD and MS. However, there is concern expressed in the literature about the adequacy of current clinical assessment tools to measure community ambulation in these 2 populations.1–3 Identification of an adequate tool to assess ambulatory status via step count is imperative for these neurologic populations. The StepWatch Step Activity Monitor (SAM; Orthocare Innovations, Seattle, Washington) is a tool that may be valuable in this regard and is one that has received recent attention in the literature.4–8
The SAM is a pager-sized 2-dimensional accelerometer, worn just above the lateral malleolus, and is described in greater detail in the following “Methods” section. The SAM has been examined in a variety of populations from pediatric to adult.4–7,9,10 When tested with able-bodied children aged 6 to 20 years, the SAM had 99.87% accuracy when compared with observer's counted steps.4
Likewise, accuracy has been reported when testing the device with able-bodied adults with ages ranging from 15 to 65 years or older.5,9,10 In these studies, SAM accuracy was measured while walking with athletic shoes or wearing a fiberglass total contact cast,9 walking over level surfaces and stairs, and at both brisk and slow walking speeds.10 Among community-dwelling older adults, the SAM was found to be more accurate at slower (<0.8 m/s) walking speeds than a hip-worn pedometer or accelerometer.5
The gait patterns of adults with neurologic disability often present challenges for tools, which count strides, but good validity and reliability are still reported for the SAM when used in neurologic populations. Macko et al11 used the device with persons who had an ischemic stroke with residual gait deficits. The participants represented a mixed group of ambulators with some requiring assistive devices or ankle-foot orthoses and some being independent ambulators. The SAM, but not a conventional pedometer, was found to provide accurate and reliable measures of cadence and total stride counts in hemiparetic stroke patients. Mudge et al6 compared the SAM to 3-dimensional gait analysis and footswitches in people with chronic stroke (greater than 6 months) in both indoor and outdoor environments. The SAM was highly correlated to both devices, with r values of 0.896 to 0.999, and was most accurate when used on the nonparetic limb. The SAM was also found to be valid and reliable among persons with incomplete spinal cord injury at self-selected walking speed, functioning at an independent to minimal assistance level, some of whom used assistive devices.7
These studies demonstrate strong evidence for the reliability and validity of the SAM in several adult neurologic populations, but little investigation of its performance in PD and MS populations has been performed. Reliability of counting steps in PD population has been reported,12 but validity has not been investigated. While researchers have used the SAM to measure activity of persons with PD and MS,8 its validity in particular is not established.
Despite the lack of investigation in persons with PD and MS, there are many potential advantages of its use: programming ease, simplicity of use by those who wear the device, extended data collection time, and ability to review activity levels during modifiable time intervals. Establishing whether or not the SAM is a valid tool in the 2 populations of interest would benefit both researchers and clinicians. A direct quantitative measurement of ambulation over an extended time could aid in assessing the impact of impairments in neurological disorders. Therefore, the purpose of this study was to investigate the validity of the SAM in counting strides of persons with PD and MS.
A convenience sample of 20 participants diagnosed with PD and MS was recruited between March 2008 and February 2009. Participants were recruited from a community-based exercise class and a local support group for persons diagnosed with PD and MS. A total of 11 participants diagnosed with PD and 9 diagnosed with MS participated in the study. All participants in the exercise program received medical clearance to perform strenuous exercise for a period of 60 minutes, 2 times per week with appropriate rests. Inclusion criteria were a diagnosis of either PD or MS and independent ambulation with or without an assistive device. Exclusion criteria were: Mini Mental Status Examination (MMSE) score less than 20, hypertension (>140 mm Hg systolic or >90 mm Hg diastolic), bradycardia (<60 beats per minute resting), tachycardia (>100 beats per minute resting), amputation, self-reported history of stroke in the prior year, or history of hip fracture or lower-extremity joint replacement. Approval was obtained from the institutional review board at Arcadia University, and written informed consent was obtained from all participants.
The SAM is a 2-dimensional accelerometer worn around either ankle above the lateral malleolus. It is a microprocessor-controlled device, which records stride count per interval and patterns of activity during extended monitoring periods. Data are downloaded to a host computer via a docking station and documented in spreadsheet format showing stride breakdown in minute or second intervals. The SAM is programmed to the gait pattern of a participant by altering sensitivity and cadence values. Changing the cadence limits how quickly strides can be detected and should be set to the highest value without missing strides. Changing the sensitivity value determines how much movement is necessary for a stride to be detected. Higher-sensitivity numbers indicate that the SAM is less sensitive to movement.8,13
The GaitMat II (GM; E.Q., Inc, Chalfont, Pennsylvania) is a system used to measure spatial and temporal gait parameters. Data are collected via pressure sensors located on the 3.87 m mat. When using the GM, participants are able to walk across the mat in either direction, and data is collected regarding gait characteristics. The GM is considered the gold standard in gait analysis research and has been reported to be accurate and reliable.14,15 (Figures 1 and 2)
All participants underwent a thorough neurological examination to assess their disability. Participants with PD were evaluated using 2 scales: Unified Parkinson Disease Rating Scale (UPDRS) and the Modified Hoehn and Yahr Staging (H and Y).
The UPDRS is a widely used tool in PD populations and has good interrater reliability (r = 0.98).16 For our purposes, only the motor subscore (range, 0-56) and activities of daily living subscore (range, 0-52) were used. Higher scores indicate higher levels of disease severity. H and Y scores range from 0 to 5, with a score of zero indicating no disease and a score of 5 indicating a person that is wheelchair-bound or bedridden. Both PD scales were completed before participants participated in the walking protocol.
Participants with MS were evaluated using 2 scales: the Expanded Disability Status Scale (EDSS) and Modified Fatigue Impact Scale (MFIS). The EDSS is extensively used as a neurological assessment tool to describe MS severity and is composed of stages from the Kurtzke Functional Systems Score. Scores range from 0, which indicates normal neurological examination, to 10, indicating death secondary to MS.17 The MFIS assesses how much fatigue a person feels during the day because of their MS. This scale has been reported to be sensitive for detecting the amount of fatigue experienced by MS patients.18 Scores range from 0 to 84, with higher numbers indicating higher levels of fatigue. Both MS scales were completed before participants participated in the walking protocol.
Each participant was observed while walking approximately 15 m. Then, advanced programming options were used to directly set the sensitivity and cadence as per SAM manufacturer guidelines based on observed gait characteristics. For 14 of the 20 participants (70%), the initial calibrations were grossly appropriate on the basis of visual stride counts compared to SAM stride counts over approximately 15 m. These participants advanced to the GM II stage, as described in the next paragraph. Six of the 20 participants (30%) required that multiple trials of gross comparison between visual and SAM counts be performed. Interestingly, our first 3 participants required multiple trials, which suggests a short learning curve to refine this skill, and the remaining 3 participants, needing multiple visual trials, were patients with greater disease severity and a more shuffling gait.
Each participant then donned the device and was instructed to walk at their usual speed for 3 separate passes over the GM. The total number of strides counted by the SAM was recorded, as was the total number of strides counted by the GM. If 90% or greater agreement was achieved, the data from this set of 3 trials was kept. If agreement was less than 90%, the sensitivity and cadence settings were altered to better match the participant's gait. StepWatch Step Activity Monitor settings were altered until 90% or more agreement was achieved between the devices, with the final data set recorded for statistical use in the validation analysis.11,19
The Pearson correlation coefficient (r) was used to assess concurrent validity of the SAM with the GM II as the gold standard. The Pearson correlation coefficient was used to quantify covariation. The following specific comparisons were made to assess concurrent validity: the number of strides counted in 3 passes over the GM II as counted by (1) the SAM, and (2) the GM II. Mean stride count by the SAM and GM was also calculated.
Twenty participants consented and were eligible for participation according to inclusion and exclusion criteria, of whom 11 were persons with PD and 9 were persons with MS. Characteristics are described in Tables 1 and 2.
As previously described, initial SAM calibrations were manually set on the basis of observed gait characteristics. If 90% agreement was not attained between the stride counts of the SAM and the GM II, alterations were made to the settings until this threshold was achieved. In our population of 20 participants, 4 participants required at least 1 alteration to the initial SAM calibration (2 participants required 1 alteration, 1 required 2 alterations, and 1 required 3 alterations). Three of the 4 participants who required alterations were in the lower 50th percentile on their respective disease severity scales.
The Pearson correlation coefficients for MS and PD, respectively, were 0.99 and 1.0. Mean strides (95% CI) taken by each participant were 15.55 (13.43-17.67) and 15.85 (13.59-18.11) as counted by the SAM and the GM, respectively.
In this study, we investigated the validity of the SAM to measure stride count in persons with PD and MS. A tool identified to accurately count strides over an extended period would benefit clinicians and researchers to measure strides and potentially quantify activity level and community ambulation. The emergence of the SAM as a tool to count strides over an extended time period could fulfill this need, but validity has not been tested in PD and MS population prior to this study.
On the basis of our investigation, we conclude that the SAM is a valid tool for counting steps in persons with PD and MS. Our results indicate that the SAM can be accurately calibrated by clinicians who have familiarized themselves with the device. While we used the GM II to help identify an optimal sensitivity and cadence setting, clinicians can use visual stride counts to modify advanced settings until more than 90% agreement is achieved.11,19 The participants in this study represented a diverse sample of persons with PD and MS in respect to disease severity and functional mobility.
Limitations of our study include the small sample size, which limits the strength of the results. Additional research should be done with a larger sample size to strengthen our results. Our research examined SAM utilization in an indoor environment and on a short, linear course. The validity of the SAM in PD and MS populations on nonlinear courses, various terrain, and stairs would be beneficial to assess validity in home and community settings.
If further research confirms validity in larger samples and various settings, then 2 potentially relevant questions emerge: (1) Can the SAM predict ambulation and activity in persons with PD and MS? and (2) What is the clinical meaningfulness of quantifying step count in these disease populations?
Funding was received from the Gerri and Dan Aaron Scholarship.
All authors have made substantive contributions to this piece. Approval by the institutional review board of Arcadia University was obtained before commencing work with human participants.
1. Klingbeil H, Baer HR, Wilson PE, Aging with a disability. Arch Phys Med Rehabil. 2004;85(7):S68–S73.
2. Morris ME, Locomotor training in people with Parkinson disease. Phys Ther. 2006;86:1426–1435.
3. Kos D, Nagels G, D'Hooghe MB, et al., Measuring activity patterns using actigraphy in multiple sclerosis. Chronobiol Int. 2007;24:345–356.
4. McDonald CM, Widman L, Abresch T, Walsh SA, Walsh DD, Utility of a step activity monitor for the measurement of daily ambulatory activity in children. Arch Phys Med Rehabil. 2005;86:793–801.
5. Storti KL, Pettee KK, Brach JS, et al., Gait speed and step-count monitor accuracy in community-dwelling older adults. Med Sci Sports Exerc. 2008;40(1):59–64.
6. Mudge S, Stott NS, Walt SE, Criterion validity of the StepWatch activity monitor as a measure of walking activity in patients after stroke. Arch Phys Med Rehabil. 2007;88:1710–1715.
7. Bowden MG, Behrman AL, Step activity Monitor: accuracy and test-retest reliability in persons with incomplete spinal cord injury. J Rehabil Res Dev. 2007;44(3):355–362.
8. Busse ME, Pearson OR, Deursen RV, Wiles CM, Quantified measurement of activity provides insight into motor function and recovery in neurological disease. J Neurol Neurosurg Psychiatry. 2004;75:884–888.
9. Hartsell H, Fitzpatrick D, Brand R, Frantz R, Saltzman C, Accuracy of a custom designed activity monitor: indications for diabetic foot ulcer healing. J Rehabil Res Dev. 2002;39(3):395–400.
10. Shepherd EF, Toloza E, McClung CD, et al., Step activity monitor: increased accuracy in quantifying ambulatory activity. J Orthop Res. 1999;17:703–708.
11. Macko RF, Haeuer E, Shaughnessy M, et al., Microprocessor based ambulatory activity monitoring in stroke patients. Med Sci Sports Exerc. 2002;34:3:394–399.
12. Skidmore FM, Mackman CA, Pav B, Daily ambulatory activity levels in idiopathic Parkinson disease. J Rehabil Res Dev. 2008;45(9):1343–1348.
14. Barker SP, Craik RL, Freedman W, Hillstrom H, Herrmann N, Accuracy, reliability, and validity of a spatiotemporal gait analysis system. Med Eng Phys. 2006;28:460–467.
15. Huang WW, VanSwearingen JM, Brach JS, Gait variability in older adults: observational rating validated by comparison with a computerized walkway gold standard. Phys Ther. 2008;88:1146–1153.
16. O'Sullivan SB, Parkinson's disease. In: O'Sullivan SB, Schmitz TJ. Physical Rehabilitation Assessment and Treatment. 5th ed. Philadelphia, PA: FA Davis; 2007:871–872.
17. Kurtzke JF, Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology. 1983;33(11):1444–1452.
18. Fisk JD, Pontefract A, Ritvo PG, Archibald CJ, Murray TJ. The impact of fatigue on patients with multiple sclerosis. Can J Neurol Sci. 1994;21(1):9–14.
19. Behrman AL, Lawless-Dixon AR, Davis SB, Locomotor training progression and outcomes after incomplete spinal cord injury. Phys Ther. 2005;85:1356–1371.
community ambulation; multiple sclerosis; Parkinson disease; validity; step activity monitor
Copyright © 2011 the Section on Geriatrics of the American Physical Therapy Association