Step Watch Research Articles

Free-Living Validation and Harmonization of 10 Wearable Step Count Monitors

ABSTRACT This study aimed 1) to determine the step count accuracy of numerous wrist-, hip-, and thigh-worn consumer and research monitors (and their corresponding algorithms) compared with the StepWatch (SW) across all waking hours under free-living conditions and 2) to develop correction methods to calibrate all monitors to the SW. Forty-eight participants 28 ± 12 yr old (mean ± SD) wore monitors across two waking days. Different wrist (Apple Watch Series 2, Fitbit Alta, Garmin vivofit 3, and ActiGraph GT9X), hip (Yamax Digiwalker SW-200, Omron HJ-325, GT9X, and Fitbit Zip), and thigh (activPAL) monitors were worn across 2 d, with the exception of the SW, which was worn on both days. Monitor estimates were compared with SW to compute percent of SW steps, absolute percent error, mean difference, root-mean-square error, and Pearson correlations. Monitor-specific correction factor linear regression models were fit to estimate SW steps and evaluated using leave-one-subject-out cross validation. All monitors were significantly different from the SW (P < 0.05). Consumer wrist and hip monitors underestimated SW steps (72%–91% of SW steps per day), whereas step estimates from research monitors ranged widely (67%–189%). Mean absolute percent error across all devices were greater than 10%. After a correction method was applied, all monitor estimates were not significantly different from SW steps. Overall, some consumer monitors produced step estimates that are closer to the validated SW than research-grade monitors (and their corresponding algorithms) and could be used to measure steps for healthy adults under free-living conditions. The specific correction methods may facilitate comparisons across studies and support research efforts using consumer and research monitors for large-scale population surveillance and epidemiological studies.

Validity of Hip and Ankle Worn Actigraph Accelerometers for Measuring Steps as a Function of Gait Speed during Steady State Walking and Continuous Turning.

This study aimed to investigate the accuracy and reliability of hip and ankle worn Actigraph GT3X+ (AG) accelerometers to measure steps as a function of gait speed. Additionally, the effect of the low frequency extension filter (LFEF) on the step accuracy was determined. Thirty healthy individuals walked straight and walked with continuous turns in different gait speeds. Number of steps were recorded with a hip and ankle worn AG, and with a Stepwatch (SW) activity monitor positioned around the right ankle, which was used as a reference for step count. The percentage agreement, interclass correlation coefficients and Bland–Altmann plots were determined between the AG and the reference SW across gait speeds for the two walking conditions. The ankle worn AG with the default filter was the most sensitive for step detection at >0.6 m/s, whilst accurate step detection for gait speeds < 0.6 m/s were only observed when applying the LFEF. The hip worn AG with the default filter showed poor accuracy (12–78%) at gait speeds < 1.0 m/s whereas the accuracy increased to >87% for gait speeds < 1.0 m/s when applying the LFEF. Ankle worn AG was the most sensitive to measure steps at a vast range of gait speeds. Our results suggest that sensor placement and filter settings need to be taken into account to provide accurate estimates of step counts.

Physical Activity Monitor Accuracy for Overground Walking and Free-Living Conditions Among Pregnant Women

Purpose: We aimed to determine the step-count validity of commonly used physical activity monitors for pregnancy overground walking and during free-living conditions. Methods: Participants (n = 39, 12–38 weeks gestational age) completed six 100-step overground walking trials (three self-selected “normal pace”, three “brisk pace”) while wearing five physical activity monitors: Omron HJ-720 (OM), New Lifestyles 2000 (NL), Fitbit Flex (FF), ActiGraph Link (AG), and Modus StepWatch (SW). For each walking trial, monitor-recorded steps and criterion-measured steps were assessed. Participants also wore all activity monitors for an extended free-living period (72 hours), with the SW used as the criterion device. Mean absolute percent error (MAPE) was calculated for overground walking and free-living protocols and compared across monitors. Results: For overground walking, the OM, NL, and SW performed well (&lt;5% MAPE) for normal and brisk pace walking trials, and also when trials were analyzed by actual speeds. The AG and FF had significantly greater MAPE for overground walking trials (11.9–14.7%). Trimester did affect device accuracy to some degree for the AG, FF, and SW, with error being lower in the third trimester compared to the second. For the free-living period, the OM, NL, AG, and FF significantly underestimated (&gt;32% MAPE) actual steps taken per day as measured by the criterion SW (M [SD] = 9,350 [3,910]). MAPE for the OM was particularly high (45.3%). Conclusion: The OM, NL, and SW monitors are valid measures for overground step-counting during pregnancy walking. However, the OM and NL significantly underestimate steps by second and third trimester pregnant women in free-living conditions.

Longitudinal trajectories of physical activity and walking performance by gross motor function classification system level for children with cerebral palsy

Purpose: This study aims to present developmental trajectories for physical activity (PA) and walking performance for children with cerebral palsy (CP).Materials and methods: Seventy-nine children with CP, 39 (49%) female, Gross Motor Functional Classification System levels I–V, and mean age 91.3 months (+/−27.7 SD) participated. Participants in levels I–V wore the Actigraph to capture PA and children in levels I-II also wore a StepWatch (SW) (n = 43) to measure walking performance. Trajectories for average PA counts/minute and number of minutes of moderate to vigorous PA were generated for levels I, II, and III/IV/V (aggregate). Single leg strides/day and average strides faster than 30 strides/min trajectories were generated for levels I–II.Results: Participants did not display plateaus in PA or walking performance based on functional level. Children in all levels showed a decrease in amount and intensity of PA from 3.0 to 12 years old, with participants in level I demonstrating the steepest decline. Children in level I decreased slightly, and level II increased slightly in both walking performance measures from 3.0 to 12 years old.Conclusions: Longitudinal curves demonstrate variations in PA and walking performance by functional level and provide prognostic information as to what changes may be anticipated for children with CP.Implications for rehabilitationLongitudinal developmental trajectories for physical activity and walking performance for children with cerebral palsy across functional levels are documented.Trajectories have potential to support collaborative intervention planning between therapists and families relative to physical activity and walking performance.

Video-Recorded Validation of Wearable Step Counters under Free-living Conditions.

The purpose of this study was to determine the accuracy of 14-step counting methods under free-living conditions. Twelve adults (mean ± SD age, 35 ± 13 yr) wore a chest harness that held a GoPro camera pointed down at the feet during all waking hours for 1 d. The GoPro continuously recorded video of all steps taken throughout the day. Simultaneously, participants wore two StepWatch (SW) devices on each ankle (all programmed with different settings), one activPAL on each thigh, four devices at the waist (Fitbit Zip, Yamax Digi-Walker SW-200, New Lifestyles NL-2000, and ActiGraph GT9X (AG)), and two devices on the dominant and nondominant wrists (Fitbit Charge and AG). The GoPro videos were downloaded to a computer and researchers counted steps using a hand tally device, which served as the criterion method. The SW devices recorded between 95.3% and 102.8% of actual steps taken throughout the day (P > 0.05). Eleven step counting methods estimated less than 100% of actual steps; Fitbit Zip, Yamax Digi-Walker SW-200, and AG with the moving average vector magnitude algorithm on both wrists recorded 71% to 91% of steps (P > 0.05), whereas the activPAL, New Lifestyles NL-2000, and AG (without low-frequency extension (no-LFE), moving average vector magnitude) worn on the hip, and Fitbit Charge recorded 69% to 84% of steps (P < 0.05). Five methods estimated more than 100% of actual steps; AG (no-LFE) on both wrists recorded 109% to 122% of steps (P > 0.05), whereas the AG (LFE) on both wrists and the hip recorded 128% to 220% of steps (P < 0.05). Across all waking hours of 1 d, step counts differ between devices. The SW, regardless of settings, was the most accurate method of counting steps.

Are clinic-based walking measures associated with community walking activity in children with cerebral palsy?

To examine the relationship between clinic-based walking capacity measures and community-based walking activity in ambulatory children with cerebral palsy (CP). A secondary analysis of a cross-sectional cohort was employed at tertiary care children's hospital; n= 128, ages 2-9 years, Gross Motor Function Classification System (GMFCS) I-III. Walking capacity was captured with 1- and 6-minute walk tests (1MWT, 6MWT), Gross Motor Function Measure-walk/run/jump score (GMFM-E), and Activity Scale for Kids performance version (ASKp-30). Walking activity performance in the community was quantified by StepWatch (SW). Moderate correlations were documented for 6MWT to SW outputs of walking level, moderate high intensity, 60-minute peak and peak activity index (r= 0.55-0.58, p< 0.01). GMFM-E correlated with all SW outputs (r= 0.55-0.69, p< 0.01) except 1-minute peak walking rate. Per regression modeling, GMFM-E was associated with walking level and intensity (p< 0.02) and 6MWT related to high intensity walking (p< 0.4, R=2 0.28-0.48). 6MWT and GMFM-E have the strongest associations with level, amount and intensity of walking in daily life. Results suggest that the 6MWT and GMFM-E can be employed to estimate community walking activity in ambulatory children with CP. Future studies should focus on environmental and personal factors that influence community walking performance.

Developing a Clinical Protocol for Habitual Physical Activity Monitoring in Youth With Cerebral Palsy.

The StepWatch (SW) has been used to monitor physical activity (PA) in youth with cerebral palsy; however, there is no standard collection protocol. The goal was to develop such a protocol. Data were examined from patients who wore the SW for 8 to 14 days. The Spearman-Brown prediction formula determined the minimum number of days for reliable PA. Weekdays were compared to weekends and 10- and 60-second collection intervals were examined. The PA data were collected from 98 youth with cerebral palsy. Results showed 3 days would provide reliable representation of PA, participants took significantly more steps during school days compared with weekends, and there were no differences between collection intervals. We recommend setting the SW for 7 days at 10-second collection rate. Data should be analyzed if at least 3 days of data are present. Weekdays and weekend days should be noted, and both included when possible.

Abstract 13860: Patient and Partner (P+P) Intervention to Improve Outcomes After and Implantable Cardioverter Defibrillator (ICD): Three Month Physical Health Outcomes

Introduction: Partners of patients who get an ICD play a pivotal role in the patient’s adjustment over the initial 3 months. The purpose of this study was to compare 2 intervention programs after an initial ICD: an intervention for the patient-only (P) was compared to an intervention involving both the patient and their intimate partner (P+P). Hypothesis: The P+P intervention will demonstrate better patient reported outcomes than the (P)-only intervention. Methods: A prospective randomized control trial (RCT) comparing 2 interventions after an initial ICD on patient and partner outcomes, was conducted with 301 intimate couples in the Pacific NW. The self-efficacy based nurse telephone intervention was delivered over the 1 st 12 weeks following ICD implant for primary (60%) or secondary (40%) prevention. Patient physical health outcomes were measured at baseline hospital discharge and 3 months later using the Patient Concerns Assessment (PCA), SF-36 PCS, total steps taken on a Step Watch physical activity monitor, and the Sudden Cardiac Arrest Self-Efficacy (SCA-SE) scale. Structural equation modeling was used to test the hypothesized effects of the intervention and the mediating effects of self-efficacy on patient physical health. Results: ICD patients (N=150) in P-only were (mean±SD) 65.0±12.3 years old, 90.1% Caucasian, 70.2% were male, with Charlson scores of 2.26±1.4, EF% 33.8±13.5, &amp; household income $30-49,000 (23%). ICD patients (N=151) in P+P were 63.3±12.5 years old, 92% Caucasian, 77.3% were male, with Charlson scores of 2.2±1.5, EF% 34.4±15.2. From baseline to 3 months, PCA symptoms were reduced in both groups p=0.84, and SF-36 physical health (p=0.62) and total steps/day (p=0.26) improved in both groups. The direct effect of the intervention on patient self-efficacy at 3 months was significant ( B =0.11, p=0.03). Improvements in patient self-efficacy were associated with a reduction in 3 month physical symptoms ( B =-0.19, p&lt;0.001), and improved physical health ( B =0.16, p&lt;0.001), but not total steps/day ( B =0.07, p=0.07). Fit indicies met standards for model fit: CFI=0.90, RMSEA=0.08. Conclusions: The intervention arm involving the partner (P+P) resulted in higher levels of patient physical health after an initial ICD.

Self-efficacy Mediates the Relationship between Balance/Walking Performance, Activity, and Participation after Stroke

Background: Many outcome measures (OM) that assess individuals’ ability or beliefs in their ability to perform tasks exist to evaluate activity and participation after stroke; however, the relationship between various OM and activity/participation is unclear.Objective: The purpose of this study was to explore the relationships between different OM and activity and participation in people after stroke.Methods: Fifty-nine subjects post stroke participated in an assessment including self-selected walking speed, 6 minute walk test, Timed “Up and Go” test, Berg Balance Scale, Functional Gait Assessment, Walk 12, and Activity-specific Balance Confidence Scale. Step Watch Activity Monitoring (SAM) was used as a measure of activity and Stroke Impact Scale-Participation (SIS-P) as a measure of participation. Exploratory Factor Analysis was performed including all measures except SAM and SIS-P. Two factors were extracted and termed performance based (PB) and self-efficacy (SE). A path analysis assessed the role of SE as a mediator in the relationships of PB and SAM/SIS-P.Results: In the path analysis, PB significantly predicts SE (p < 0.001, b = 0.44), but not SAM or SIS-P (p > 0.05, b = 0.25, and b = 0.11, respectively). SE significantly predicts both SAM and SIS-P (p < 0.001, b = 0.46, and b = 0.59, respectively). The indirect effects of PB on SAM and SIS-P were significant (p < 0.001; b = 0.20, and b = 0.26, respectively).Conclusion: These results suggest that SE mediates the relationship between PB and activity and participation after stroke, reinforcing that improving activity and participation is more complicated than only targeting performance. Clinicians should administer SE and PB measures to determine the most accurate view of patients after stroke and seek to improve SE through interventions.

Validity of Activity Monitor Step Detection Is Related to Movement Patterns.

There is a need to examine step-counting accuracy of activity monitors during different types of movements. The purpose of this study was to compare activity monitor and manually counted steps during treadmill and simulated free-living activities and to compare the activity monitor steps to the StepWatch (SW) in a natural setting. Fifteen participants performed laboratory-based treadmill (2.4, 4.8, 7.2 and 9.7 km/h) and simulated free-living activities (eg, cleaning room) while wearing an activPAL, Omron HJ720-ITC, Yamax Digi- Walker SW-200, 2 ActiGraph GT3Xs (1 in "low-frequency extension" [AGLFE] and 1 in "normal-frequency" mode), an ActiGraph 7164, and a SW. Participants also wore monitors for 1-day in their free-living environment. Linear mixed models identified differences between activity monitor steps and the criterion in the laboratory/free-living settings. Most monitors performed poorly during treadmill walking at 2.4 km/h. Cleaning a room had the largest errors of all simulated free-living activities. The accuracy was highest for forward/rhythmic movements for all monitors. In the free-living environment, the AGLFE had the largest discrepancy with the SW. This study highlights the need to verify step-counting accuracy of activity monitors with activities that include different movement types/directions. This is important to understand the origin of errors in step-counting during free-living conditions.

Assessment of objective ambulation in lower extremity sarcoma patients with a continuous activity monitor: rationale and validation.

In addition to patient reported outcome measures, accelerometers may provide useful information on the outcome of sarcoma patients treated with limb salvage. The StepWatch (SW) Activity Monitor (SAM) is a two-dimensional accelerometer worn on the ankle that records an objective measure of walking performance. The purpose of this study was to validate the SW in a cross-sectional population of adult patients with lower extremity sarcoma treated with limb salvage. The main outcome was correlation of total steps with the Toronto Extremity Salvage Score (TESS). In a sample of 29 patients, a mean of 12 days of SW data was collected per patient (range 6–16), with 2767 average total steps (S.D. 1867; range 406–7437). There was a moderate positive correlation between total steps and TESS (r = 0.56, P = 0.002). Patients with osseous tumors walked significantly less than those with soft tissue sarcoma (1882 versus 3715, P < 0.01). This study supports the validity of the SAM as an activity monitor for the objective assessment of real world physical function in sarcoma patients.

StepWatch stride counting: Accuracy, precision, and prediction of energy expenditure in children

Accuracy and precision of the StepWatch (SW) was tested in typically developing youth (TDY) with the prediction of oxygen consumption derived from stride counts. Ten TDY (5 boys, 5 girls) with an average age 14.1 years (SD=2.2) enrolled. Participants underwent treadmill cardiopulmonary exercise testing (CPET) wearing two SW devices at 1, 2, 3, and 4 miles per hour (mph). Average sign-corrected disagreement for strides counted between the 2 monitors was 0.4 (0.52), 0.2 (0.42), 0.1 (0.32), and 0.1 (0.32) for 1, 2, 3, and 4 mph, respectively. The ratio of SW counts to observed manual counts averaged 100.03% with ICC=0.995. Oxygen consumption equations were derived from resting and walking VO(2), age, gender, and stride counts. Excellent accuracy and precision was documented for treadmill walking speeds up to 4 mph while VO(2) per number of strides taken appears to decrease with age and is lower in girls.

The effect of intense physical therapy for children with cerebral palsy

To determine the effect of an intense physical therapy intervention on gross motor function, community walking and participation in children with cerebral palsy (CP). A single group design was used with two pre-test and two post-test measures. Subjects were 17 ambulatory children with CP who participated in an intense intervention (i.e., four hours per day, five days per week, three weeks), a modified version of the TheraSuit protocol. Gross motor function measure (GMFM-66), Step watch activity monitor (SAM), Canadian occupational performance measure (COPM) and pediatric outcomes data collection instrument (PODCI) were tested twice at baseline, immediately following the intervention, and three months later. Immediately following the intervention, GMFM-66, COPM and PODCI scores improved significantly (p < 0.001). At three months, improvements remained for GMFM-66 and COPM (p < 0.01). Walking amount or intensity (SAM) did not improve. Participants improved gross motor skills and participation but not community ambulation following this intense physical therapy intervention.

Daily ambulatory activity monitoring in patients with peripheral artery disease

Background: Peripheral artery disease (PAD) is characterized by arterial lesions of the lower extremities. Patients with PAD present ambulatory dysfunction due to intermittent claudication, limiting daily physical activity and increasing morbidity. Therefore, objectively measured daily ambulation is a clinically relevant outcome in these patients.Objective: To quantify daily ambulation with a step activity monitor (Step Watch 3™; Cyma Inc., Mountlake Terrace, WA, USA) in patients with PAD.Major findings: Patients with claudication ambulate less each day than controls, as they take 1081 fewer strides (1 stride=2 steps) and spend 48 fewer minutes ambulating each day. The cadence obtained during endurance ambulation (maximum cadences for 60, 30, and 20 continuous minutes of ambulation) and short bouts of ambulation (maximum cadence for continuous minutes of ambulation and for the minute having the single highest cadence value each day) is also lower in patients. Demographic factors and comorbid conditions, particularly sex, dyslipidemia, abdominal obesity, and the number of components of metabolic syndrome, affect the ambulatory activity patterns of patients with claudication.Conclusions: Patients with claudication perform a lower amount of daily ambulatory activities and lower intense physical activities than controls, which is impaired further by comorbid conditions. The clinical implication is that patients with claudication should not only be advised to walk more during each day, but to do so at a cadence that is higher than their self-selected slower cadence. This approach will increase total daily exercise volume by increasing both the duration as well as the intensity of ambulation.

Comparison Of Two Waist-mounted And Two Anklemounted Electronic Pedometers

PURPOSE The purpose of this study was to compare two ankle-mounted pedometers (Step Watch 3 [SW-3] and Activity Monitoring Pod 331 [AMP]) and two waist-mounted pedometers (New Lifestyles NL-2000 [NL] and Digiwalker SW-701 [DW-701]) under controlled and free-living conditions. METHODS In part I, 10 males and 10 females walked on a motor driven treadmill at various speeds (27, 40, 54, 67, 80, and 107 m·min−1). An investigator determined the actual steps with a hand counter. In part II, participants performed leg swinging, heel tapping, stationary cycling, and car driving (6.4 km on city streets). In part III, a subset of 15 participants volunteered to wear all four pedometers for a 24-hr period. RESULTS The SW-3 displayed values that were within 1% of actual steps during treadmill walking at all speeds. The other devices underestimated steps at slow speeds but all gave mean values that were within ± 3% of actual steps at 80 m·min−1 and above. The SW-3 registered some steps during heel tapping, leg swinging, and cycling, while the AMP was only responsive to leg swinging. During car driving none of the devices recorded more than 8 steps, on average. Compared to the SW-3, over 24 hours the AMP and DW-701 recorded 18% and 15% fewer steps, respectively (P <0.05), while the NL recorded 11% less (NSD). CONCLUSIONS The SW-3 has superior validity during slow treadmill walking (<67 m·min−1). Although it is more likely to detect certain types of non-ambulatory activities, this is probably not a major source of error. Over 24 hours, the SW-3 gave higher estimates of steps day−1 versus the DW-701 and AMP. This suggests that the SW-3 could be a useful criterion for validating other pedometers during 24-hr free-living activity.

Accelerometer monitoring of home- and community-based ambulatory activity after stroke

Haeuber E, Shaughnessy M, Forrester LW, Coleman KL, Macko RF. Accelerometer monitoring of home- and community-based ambulatory activity after stroke. Arch Phys Med Rehabil 2004;85:1997-2001. Objectives To investigate the utility of a novel microprocessor-linked Step Watch Activity Monitor (SAM) to quantify ambulatory activity after stroke and to evaluate the validity and reliability of conventional accelerometers to measure free-living physical activity in this population. Design Cross-sectional with repeated measures of 2 separate 48-hour recordings in 17 persons wearing an ankle-mounted SAM and Caltrac, a hip-mounted mechanical accelerometer. Setting Home and community. Participants Seventeen subjects with chronic hemiparetic gait after stroke. Interventions Not applicable. Main outcome measures The SAM derived stride counts per day and Caltrac estimated the daily caloric expenditure of physical activity. Results SAM data revealed that stroke patients had a mean strides per day ± standard deviation of 3035±1944 and demonstrated a broad range of daily activity profiles (400–6472 strides). SAM test-retest reliability was high across separate monitoring periods ( r=.96, P<.001). Although Caltrac also revealed a broad range of daily activity calories (346±217kcal/d; range, 83–1222kcal/d), reliability was poor ( r=.044, P=not significant) and Caltrac accounted for only 64% of the ambulatory activity quantified by the SAM. Conclusions Microprocessor-linked accelerometer monitoring, but not conventional accelerometers, are accurate and highly reliable for quantifying ambulatory activity levels in stroke patients. These findings support the utility of personal status monitoring of ambulatory activity as an outcomes instrument and metric in programs to increase physical activity and cardiovascular health after stroke.