Quantification of Physical Activity and Sedentary Time in Adults with Cerebral Palsy : Medicine & Science in Sports & Exercise

Journal Logo


Quantification of Physical Activity and Sedentary Time in Adults with Cerebral Palsy


Author Information
Medicine & Science in Sports & Exercise 47(8):p 1719-1726, August 2015. | DOI: 10.1249/MSS.0000000000000589
  • Free



The purpose of this study was to determine objective and subjective quantification of habitual physical activity (HPA) and sedentary time in ambulatory and nonambulatory adults with cerebral palsy (CP).


We recruited a clinical sample of adults with CP (N = 42; 21 women; mean (SD) age, 33.5 (12.3) yr; Gross Motor Function Classification System (GMFCS) distribution: level I (n = 5), level II (n = 9), level III (n = 10), level IV (n = 11), and level V (n = 7). Objective measures of HPA and sedentary time were obtained by using ActiGraph GT3X accelerometers at both hip and wrist sites. Three previously established cut-point values distinguishing light physical activity (LPA) and moderate-to-vigorous physical activity (MVPA) were evaluated across GMFCS levels. The concurrent validity of the self-report Physical Activity Recall Assessment for People with Spinal Cord Injury (PARA-SCI) was assessed for LPA and MVPA intensities in GMFCS levels II–V.


Participants showed little reluctance to wearing accelerometers; one participant reported discomfort. Nonambulatory adults (GMFCS levels IV–V) differed from ambulatory adults (GMFCS levels I–III) for recorded activity counts (hip and wrist sites), minutes of MVPA with each cut-point value, and breaks from sedentary time (all P < 0.05). For the same measures, adults in GMFCS level III also differed from GMFCS level I (all P < 0.05). The PARA-SCI correlated significantly with accelerometer-derived minutes of MVPA per day (r = 0.396; P = 0.014) and per hour of monitoring time (r = 0.356; P = 0.027).


Our findings support the use of accelerometers to objectively measure HPA and sedentary behavior in adults with CP across the severity spectrum, regardless of cut-point implementation. The PARA-SCI is a valid tool to capture subjectively reported patterns of MVPA in adults with CP who are GMFCS levels II–V.

Cerebral palsy (CP), a neurological condition affecting the development of posture and movement (32), is the most common cause of physical disability seen in children (16). Beginning in early infancy, CP affects an individual’s functional motor ability over the lifespan (32).

Individuals with CP have been reported to be less physically active than their able-bodied peers (19,23,37). This reduction in physical activity (PA) is a concern as it has been well established that PA in healthy children and adults decreases the risk of cardiovascular health disease and other secondary health concerns (1,30). Additionally, sedentary behavior has shown to be adversely associated with various health outcomes in all age groups (17). To mitigate the possible health consequences of inactivity, promotion of increases in PA (5) and reductions in sedentary behavior (38) are recommended in clinical practice for children and adolescents with CP. However, it has been reported that individuals with physical disabilities have decreased contact with the health care system as they enter adulthood (25). As most of the research has focused primarily on children and adolescents (2,8,10,19), little is known about PA and sedentary behavior in both ambulatory and nonambulatory adults with CP.

To classify severity of CP in a meaningful and functional manner, the Gross Motor Functioning Classification System (GMFCS) has become available for worldwide use (29). The GMFCS identifies five levels ranging from “walks without restrictions” (level I) to “self-mobility is severely limited even with use of assistive technology” (level V). Although originally developed and validated for children (29), the expanded and revised version (GMFCS—E&R) has demonstrated reliability for describing gross motor ability in adolescents and adults with CP (9).

It is imperative that a practical and appropriate measure assessing PA be available before designing lifestyle intervention studies (11). Objective and subjective (i.e., self-report) measures both offer valid assessment of PA in adults, although both have distinct limitations (31). Objectively based measurement of PA and sedentary behavior, through the use of activity monitors, has been shown to be appropriate in ambulatory children and adults with CP (GMFCS levels I–III) (2,24,26,37). Most recently, the ActiGraph hip-worn accelerometer was shown to be a feasible and unobtrusive measurement tool in nonambulatory adolescents with CP (GMFCS level IV) (10). However, the conflicting evidence regarding the use of accelerometry to estimate PA in nonambulatory adult wheelchair users warrants the need for further investigation in this population (15,40). Furthermore, the use of wrist-worn accelerometers in adolescents and adults with CP has been limited (23). Past literature has also emphasized the importance of implementing population specific cut points to properly interpret accelerometry activity output (41). However, cut-point values for distinguishing various intensity levels of PA for either hip- or wrist-worn devices have not been previously validated in adults with CP.

In situations where accelerometry is not appropriate, such as water-based activities, it is vital to have a valid alternative measure of habitual PA (HPA). The Physical Activity Recall Assessment for People with Spinal Cord Injury (PARA-SCI) is a self-report measure that has been shown to be a reliable and valid tool for assessing PA in adults with SCI (22). Although the PARA-SCI has not been validated in other populations with functional limitations, its development for individuals requiring a wheelchair for their usual mobility suggests it may be a valuable tool for the assessment of PA in adults with more severe CP (GMFCS levels III–V).

In summary, there remains a fundamental knowledge gap pertaining to assessment of HPA and sedentary behavior with hip- and wrist-worn accelerometers in adults with CP, particularly in nonambulatory individuals with the most severe functional limitations (GMFCS level V). Therefore, the aim of this study was to investigate the quantification of HPA and sedentary time in adults with CP across the severity spectrum described by the GMFCS (levels I–V) through the use of hip- and wrist-worn accelerometers. Secondary objectives were to evaluate different approaches (i.e., accelerometer threshold values) used to interpret and categorize recorded accelerometer-based activity counts into distinct PA intensity levels pertaining to the hip-worn accelerometer and to assess the validity of the PARA-SCI tool compared to the direct measure of HPA by means of accelerometry. We hypothesized that objectively measured PA would decrease with increasing GMFCS level regardless of accelerometry cut points and that the PARA-SCI would be a valid assessment tool, particularly for individuals in GMFCS levels III–V.



Adults with a condition diagnosed as CP (n = 42), ranging in age from 18 to 75 yr and classified in levels I–V of the GMFCS participated in this study. Characteristics of the study population are displayed in Table 1. Participants were recruited through the Comprehensive Spasticity Management Program at the Regional Rehabilitation Centre in Hamilton, Ontario, as well as through an online posting on the Ontario Federation for Cerebral Palsy webpage (http://ofcp.ca/whats-new/) between October 2012 and March 2014. Participants met the following inclusion criteria: 1) definitive diagnosis of CP, and 2) older than 18 yr. For classification of severity, the GMFCS—E&R Self-Report Questionnaire was completed by the participant or, if necessary, their relative or caregiver. Experimental procedures were explained to the participants before obtaining written and verbal informed consent from the participant and/or their relatives as needed. Approval from the Hamilton Integrated Research Ethics Board was obtained before study commencement.

Participants’ demographic and physical characteristics.

Objective assessment of physical activity and sedentary time

Hip- and wrist-worn ActiGraph GT3X activity monitors (ActiGraph LLC, Pensacola, FL) were used to objectively assess HPA and sedentary time. These particular devices were implemented for their ability to measure HPA over an extended period of time while remaining relatively inconspicuous (10). The ActiGraph GT3X accelerometer weighs 28 g and with dimensions of 3.8 × 3.7 × 1.8 cm. This accelerometer records acceleration in three planes of motion, the vertical, anteroposterior, and mediolateral planes, and also provides activity counts as a composite vector magnitude of these three axes ranging from 0.05g to 2g in magnitude within a frequency range of 0.25–2.5 Hz. The acceleration is sampled and digitized by a 12-bit analog-to-digital converter. After passing through a digital filter that eliminates nonhuman motion, the signal is stored in user-defined intervals (i.e., epochs). Previous research has shown that epoch interval affects the estimation of HPA in children and adults, with smaller epoch lengths resulting in reduced misclassification of PA estimates (7,27). Therefore, activity was recorded in 3-s recording epochs.

Accelerometer data were uploaded onto the ActiLife software (version 6.9.0; ActiGraph LLC, Pensacola, FL), where accelerometer wear time was visually inspected to ensure the recorded information properly matched the activity output. Periods of non wear time, as indicated by participants’ logbooks, were removed from analysis. Consecutive epochs of zero counts or any activity less than 100 counts per minute (cpm) during wear time as indicated by the participants’ logbooks were registered as sedentary time (13,21). Accelerometer data were only included in the analysis if a minimum wear time of 5 h on four consecutive days was met, as this criterion has been recommended for best practice and was previously used to maximize participant inclusion (10,33,39).

Self-report assessment of physical activity

The PARA-SCI was used as a self-report measure of HPA. After reviewing other self-report measures of HPA with the study team and consulting with an adult with CP, the PARA-SCI was chosen for its ability to detect both activities of daily living and leisure time PA, particularly in individuals who require walking aids or wheelchair assistance (GMFCS levels II–V). This self-report measure is completed during a standardized structured interview (22). Participants are asked to recall all PA performed over the past 3 d, in eight periods from their morning to evening routine, and to rate the intensity of each activity using validated definitions of mild/light, moderate, and heavy/vigorous intensity activities (22). Activities are reported as an average number of minutes per day (22). The PARA-SCI has been previously shown to have adequate test-retest reliability, construct, and criterion validity in people with SCI (18,22,35).


Study participants were instructed to wear both the hip- and wrist-worn ActiGraph accelerometers for seven consecutive days to gain an accurate representation of HPA and sedentary time (36,39). The accelerometers were removed when participants engaged in any water activities, such as swimming or bathing, to ensure the devices were not damaged. Participants were also instructed to complete a logbook intended to track all times and reasons for removal and replacement of the accelerometers. Accompanying the logbook was an instruction sheet illustrating proper accelerometer use and wear. Contact information for the study coordinator was also included should the participant have any questions or concerns regarding how to wear the device. Upon completion of the 7 d, logbooks and accelerometers were obtained from the participants and downloaded for further analysis. Difficulties and barriers associated with the use of either hip- or wrist-worn accelerometers were discussed with the study coordinator upon completion of the activity assessment to determine the feasibility of accelerometer use in adults with CP. If logbooks lacked sufficient detail, wear time was verified by cross-referencing the logbook with the raw activity output using the ActiLife software (10). After the accelerometer assessment period, the participants were asked to complete the PARA-SCI. From the completed questionnaire, total minutes of mild/light PA (LPA) and moderate-to-vigorous PA (MVPA; the sum of moderate and heavy intensities) per day as well as per hour of wake time were determined.

Activity derived from the accelerometers was first assessed using mean vertical axis counts per minute of wear time and mean vector magnitude counts per minute, before implementing distinct cut points to distinguish various activity intensities. Sedentary time was defined as less than 100 cpm (13,21) and was normalized to actual monitoring time to account for differences in time spent wearing the accelerometer (i.e., min·h−1). Additionally, breaks from sedentary time and the number of breaks from sedentary time relative to total sedentary time were investigated.

LPA and MVPA threshold values

Accelerometer cut points need to be age and population specific (34). To our knowledge, there are no existing accelerometer cut points for adults with CP. Therefore, three distinct sets of cut points distinguishing LPA and MVPA were chosen owing to their previous calibration with the ActiGraph accelerometer in a cohort of able-bodied adults (6,34) or in a cohort of adolescents with CP (2,3). This study sought to evaluate how distinct threshold values affect the observed trend of HPA across the GMFCS levels. Chosen values included the Freedson cut points calibrated in able-bodied adults during treadmill exercises (LPA, 100–759 cpm; and MVPA, ≥1952 cpm) (6), the Evenson cut points recently validated in ambulatory children and adolescents with CP (GMFCS levels I–III) (LPA, 101–2295; and MVPA ≥2296) (2,3), and the Swartz cut points calibrated in able-bodied adults during six different habitual or recreational activities (LPA, 100–573; and MVPA, ≥574) (34).

Statistical analyses

Statistical Package for the Social Sciences (SPSS, version 20, Chicago, IL) was used for data analysis. Statistical significance was set at P ≤ 0.05 for all analyses. Descriptive statistics were used to evaluate baseline characteristics of each participant, describe CP classification, GMFCS levels, and calculate the proportion of participants meeting the PA guidelines for adults of 150 min or more of MVPA each week (42). Data were tested for normality using the Shapiro–Wilk test and, when necessary, transformation of the data using the base 10 logarithms was undertaken. Duration and intensity of PA and sedentary behavior (vertical axis counts per minute, vector magnitude counts per minute, sedentary time, breaks from sedentary time, number of breaks per hour of sedentary time, LPA, and MVPA) were compared between the five GMFCS levels using a one-way analysis of variance (ANOVA). Post hoc analysis was completed with Tukey honestly significant differences test when homogeneity of variances was assumed and Dunnett T3 test when homogeneity of variances was not assumed (significant results for the Levene test of equality of error variances). The Kruskal–Wallis nonparametric ANOVA was undertaken if the data still failed to achieve normality upon transformation. The relationship between PARA-SCI and accelerometer-derived LPA and MVPA was examined using a one-tailed Pearson correlation test. As the PARA-SCI was designed and validated for nonambulatory individuals (22), participants who displayed the ability to walk without limitation (i.e., GMFCS level I, n = 5) were excluded from the analysis.


Power analysis

A post hoc evaluation of achieved statistical power using the analysis program G*Power (Macintosh Version (4) revealed the sample size and calculated power were sufficient (>0.8) to perform one-way ANOVA between groups for HPA and sedentary assessment (data not shown).


All participants were able to meet the minimum required accelerometer wear time of 5 h or longer on four or more consecutive days. The mean wear time for hip- and wrist-worn accelerometers was 6.83 d with a monitoring period ranging from 422.9 to 934.7 min·d−1 (mean (SD), 682.4 (127.1) and 681.5 (128), hip- and wrist-worn accelerometers, respectively). The mean accelerometer monitoring time for each GMFCS level is displayed in Table 2. Only one participant reported discomfort with the hip-worn accelerometer throughout the study period, which resulted in a decreased wear time. Therefore, this participant’s hip-worn accelerometer data from only days 1 to 4 were included in the analysis.

Monitoring time and sedentary behavior across GMFCS levels.

Physical activity and sedentary time

Wrist-worn accelerometry data failed to achieve normality both before and after transformation. Therefore, the Kruskal–Wallis nonparametric ANOVA was performed to analyze the wrist-worn accelerometry output. Analysis of both hip- and wrist-worn accelerometers revealed that the mean vertical axis counts per minute and vector magnitude counts per minute of adults classified at GMFCS levels IV and V differed significantly from those classified at GMFCS levels I, II, and III. Additionally, the activity counts of adults classified at GMFCS level III were significantly different from those classified at level I (Fig. 1).

Mean and SD of the vertical axis and vector magnitude activity counts per minute (CPM) from both hip- and wrist-worn accelerometers across GMFCS levels I–V. A, Different from B and C; B, different from A and C; C, different from A and B (all P < 0.05). X, Kruskal–Wallis nonparametric test completed.

The participants spent a mean (SD) of 10.5 (2.0) h·d−1 in sedentary time with a range of 37.5 to 59.9 (55.8 (4.3)) min of sedentary time per hour of monitoring time. Adults classified in GMFCS levels IV and V displayed greater sedentary time and had a lower frequency of breaks interrupting sedentary time than those classified in GMFCS levels I, II, and III (Table 2).

MVPA cut points

As was reported for vertical axis counts per minute and vector magnitude counts per minute, adults classified at GMFCS levels IV and V presented significantly lower levels of MVPA (both minutes per day and minutes per hour) than those at levels I, II, and III (Table 3). Similarly, participants classified at level III presented with lower MVPA compared to those classified at levels I and II in all comparisons except Swartz MVPA minutes per hour (Table 3). Furthermore, adults in GMFCS levels IV–V had lower LPA than those at levels I–II. No significant difference was seen between GMFCS levels I and II or between GMFCS IV and V for any activity intensity or implemented cut-point value (Table 3).

Minutes of activity per day and hour of monitoring time across GMFCS levels.

PARA-SCI validity

The Pearson correlation coefficients describing the relationship between accelerometer-derived and self-reported LPA and MVPA are presented in Figure 2. Accelerometer-derived LPA and MVPA were determined using the Freedson cut points. These cut-point values have been frequently used in adolescent and adult cohorts (6,12,21,36) and yield activity values that fall between the liberal Swartz-derived and conservative Evenson-derived values (Table 3). A significant correlation between self-report and accelerometry-based measures was found for MVPA, in both minutes per day and minutes per hour of wake/monitoring time (Figs. 2A, B); however, there was no correlation between the two measures for LPA (Figs. 2C, D).

Pearson correlation coefficients between PARA-SCI and accelerometer-derived activity measures (log transformed). A. Relationship between PARA-SCI minutes per hour of wake time and accelerometer activity minutes per hour of monitoring time for MVPA. B. Relationship between PARA-SCI minutes per day and accelerometer activity minutes per day for MVPA. C. Relationship between PARA-SCI minutes per hour of wake time and accelerometer activity minutes per hour of monitoring time for LPA. D. Relationship between PARA-SCI minutes per day and accelerometer activity minutes per day for LPA. PARA-SCI: Physical Activity Recall Assessment for people with Spinal Cord Injury; MVPA: moderate-to-vigorous physical activity; LPA: light physical activity.


We have demonstrated that the use of an accelerometer worn at both the hip and wrist sites is feasible for the assessment of HPA and sedentary behavior in adults with CP who are both ambulatory and nonambulatory participants, including adults classified in GMFCS level V. These results complement recently published research evaluating the use of accelerometers to assess HPA in children and adolescents with CP (2,10) as well as the use of activity monitors in adults with CP (23,24,37). Our results add novel data supporting for the use of accelerometers in adults with the largest degree of functional limitations (i.e., GMFCS level V).

In the present study, the data analysis of wrist-worn accelerometers presented several complications. The wrist-derived activity counts not only failed to achieve normality, even upon data transformation, but the clinical interpretation of these data was restricted owing to the limited knowledge base surrounding the use of wrist-worn accelerometers. The extra time required to analyze the wrist-derived data and the cost of the wrist-worn accelerometers may outweigh their added benefit to the activity measurement accuracy (34). For these reasons, statistical comparisons between hip- and wrist-worn accelerometers were not completed.

Consistent with previously published studies, participants in all GMFCS levels reported highly sedentary lifestyles; individuals in GMFCS levels III–V displayed increased sedentary time (23,26). Additionally, participants with more severe CP (GMFCS levels III–V) also reported decreased breaks from sedentary time. Previous research has shown that increased breaks from sedentary behavior are beneficially associated with metabolic risk factors in able-bodied adults, independent of total time in MVPA or sedentary time (12,28). These results highlight the need for intervention studies and community programs that emphasize integration of all forms of PA into daily life, further adding to the discussion regarding the health implications of HPA and sedentary time in adults with CP (38).

In all manners of data interpretation, common was the differentiation of adults classified in GMFCS levels I–II from those classified in levels III–V. To our knowledge, this was the first study to implement and evaluate several different accelerometer cut points in adults with CP. From this study, it was found that individuals with greater physical impairment (GMFCS levels IV–V) produced less activity counts and participated in less MVPA than their ambulatory peers (GMFCS levels I–III), regardless of activity cut-point implementation. When the Freedson or the Evenson MVPA cut points were used (3,6), 17% of the study participants (GMFCS level I, n = 4; GMFCS level II, n = 2; GMFCS level III, n = 1) were able to meet the PA guidelines of 150 min or more of MVPA (42) on a weekly basis. This number increased to 36% (GMFCS level I, n = 5; GMFCS level II, n = 8; GMFCS level III, n = 2) with the implementation of the Swartz MVPA cut point (34). Owing to the limited number of participants within each GMFCS level, we were unable to determine the agreement between cut points. Since accelerometer cut points specific to the adult CP population have yet to be validated, the clinical implications of the current findings are unknown.

Previous research has reported a higher oxygen cost for walking in individuals with CP compared to reference values, suggesting that these individuals, particularly those with more severe CP, may have increased physical strain and energy expenditure compared to able-bodied persons (20). Additionally, to provide insight into how this increased physical strain affects the health of adults with CP, future research assessing cardiovascular and metabolic health across all levels of the GMFCS is warranted. In previous studies, accelerometer cut points were developed using calorimetry, the criterion standard measurement to establish the metabolic equivalent for various activities (3,6,34). Inclusion of a form of calorimetry to establish metabolic costs for associated cut points in adults with CP was outside the scope of the present study but would be valuable in future research.

As has been previously observed in other cohorts, self-reported LPA and MVPA were much greater than that detected using accelerometer measures in GMFCS levels II–V (31). This observation may be due to accelerometers failing to capture activities such as swimming, transferring, or wheeling over surfaces that would require increased effort. Previous research has shown that self-reports both overestimate and underestimate levels of PA compared to direct measures; hence, correlations between the two measurement types have largely been low to moderate (31). The current analyses, however, determined a moderate positive correlation between accelerometer and PARA-SCI measurement of MVPA. This finding contributes to the literature on the measurement of HPA in people with CP by demonstrating the concurrent validity of the PARA-SCI in this population. The low and nonsignificant correlation for LPA between self-report and accelerometer-based measures in the current study is consistent with the original report evaluating the criterion validity of the PARA-SCI and is likely due to the comparative ease of recall of MVPA versus LPA (22). Although the PARA-SCI may not be a valid measure of LPA, it is recommended that participants report all activity when using the PARA-SCI, including mild/light activity, as it may assist in facilitating recall (22). These findings demonstrate that the PARA-SCI is an appropriate tool to measure MVPA in adults with CP in GMFCS levels II–V. However, a limitation of the PARA-SCI questionnaire is that because it was developed among wheelchair users with SCI, it may not be a valid measure of HPA for those in GMFCS level I. Given the relatively small sample, we were unable to assess the validity of the PARA-SCI among those in GMFCS level I. As this is the first study to evaluate the validity of PARA-SCI in adults with CP, further investigation is necessary to determine reliability.

There are noteworthy limitations that accompany the use of an accelerometer to measure levels of HPA in adults with CP. It is unknown whether the hip-worn accelerometers accurately captured all activity performed by nonambulatory individuals, such as those who are able to self-propel their wheelchair, which may result in an underestimation of activity levels in these adults simply due to decreased vertical displacement compared to ambulatory peers. The wrist-worn accelerometers were unable to effectively mitigate this inherent limitation owing to restrictions associated with analysis. Measured levels of PA may have been further underestimated for some participants owing to accelerometer sensitivity to water. In other studies, swimming is noted as a preferred activity for persons with CP (8,14), and this form of PA would not have been captured with the accelerometer. Nevertheless, the ability of the accelerometer to assess HPA in all other community and home settings far outweighs the limitation of missing water-related PA.


In the current study, hip-worn accelerometers provided a valid assessment of both HPA and sedentary behavior in the adult CP population across the complete range of the GMFCS. Regardless of the implementation of previously determined cut points, the results remained consistent: those in GMFCS levels I–II had significantly greater levels of MVPA, spent less time in a sedentary state, and had greater breaks in sedentary time compared to those in GMFCS levels III–V. Moreover, the PARA-SCI was positively correlated with accelerometer assessment of MVPA, but not LPA, in adults with CP with GMFCS levels II–V. This suggests that the PARA-SCI may be a valid tool to capture MVPA in this population. Future research focusing on the calibration of cut points specific to the adult CP population as well as assessment of the cardiovascular health of adults with CP across the GMFCS spectrum is warranted.

This project is part of the larger Stay-FIT study and was funded by the Ontario Federation for Cerebral Palsy. Dr. Jan Willem Gorter holds the Scotiabank Chair in Child Health Research (2013–2017).

The authors thank Dr. Todd Bentley and staff at the Comprehensive Spasticity Management Program for their assistance in participant recruitment. The authors likewise thank Alison McFadden for assisting in data collection and analysis of the PARA-SCI questionnaire and Joyce Obeid for her assistance with accelerometer data preparation. A special thank you is due to all participating adults and their parents or caregivers for their engagement and patience throughout this study.

The authors have no financial conflict of interest with the monitor manufacturers and have received no research funding from these companies. The results of the present study do not constitute endorsement of the products described in this article by the authors or the American College of Sports Medicine.


1. Blair SN, Kohl HW, Barlow CE, Paffenbarger RS, Gibbons LW, Macera CA. Changes in physical fitness and all-cause mortality: a prospective study of healthy and unhealthy men. J Am Med Assoc. 1995; 273 (14): 1093–8.
2. Clanchy KM, Tweedy SM, Boyd RN, Trost SG. Validation of accelerometry in ambulatory children and adolescents with cerebral palsy. Eur J Appl Physiol. 2011; 111: 2951–9.
3. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG. Calibration of two objective measures of physical activity for children. J Sports Sci. 2008; 26 (14): 1557–65.
4. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007; 39 (2): 175–91.
5. Fowler EG, Kolobe THA, Damiano DL, et al. Promotion of physical fitness and prevention of secondary conditions for children with cerebral palsy: section on pediatrics research summit proceedings. Phys Ther. 2007; 87 (11): 1495–510.
6. Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications, Inc. accelerometer. Med Sci Sports Exerc. 1998; 30: 777–81.
7. Gabriel KP, McClain JJ, Schmid KK, et al. Issues in accelerometer methodology: the role of epoch length on estimates of physical activity and relationships with health outcomes in overweight, post-menopausal women. Int J Behav Nutr Phys Act. 2010; 7: 50.
8. Gorter JW, Currie SJ. Aquatic exercise programs for children and adolescents with cerebral palsy: what do we know and where do we go? Int J Pediatr. 2011; 2011: 712165.
9. Gorter JW, Slaman J, Bartlett D, Van den Berg-Emons RJ. Reliability of the gross motor function classification system expanded and revised (GMFCS-ER) when used with adolescents and young adults with cerebral palsy. Dev Med Child Neurol. 2011; 53 (5): 42–3.
10. Gorter JW, Noorduyn SG, Obeid J, Timmons BW. Accelerometry: a feasible method to quantify physical activity in ambulatory and nonambulatory adolescents with cerebral palsy. Int J Pediatr. 2012; 2012:1–6.
11. Gorter JW, Timmons BW. Measurement of habitual physical activity and sedentary behavior of youth with cerebral palsy: work in progress. Dev Med Child Neurol. 2014; 56 (9): 911.
12. Healy GN, Dunstan DW, Salmon J, et al. Breaks in sedentary time: beneficial associations with metabolic risk. Diabetes Care. 2008; 31 (4): 661–6.
13. Healy GN, Clark BK, Winkler EAH, Gardiner PA, Brown WJ, Matthews CE. Measurement of adults’ sedentary time in population-based studies. Am J Prev Med. 2011; 41 (2): 216–27.
14. Heller T, Ying GS, Rimmer JH, Marks BA. Determinants of exercise in adults with cerebral palsy. Public Health Nurs. 2002; 19: 223–31.
15. Hiremath SV, Ding D. Evaluation of activity monitors in manual wheelchair users with paraplegia. J Spinal Cord Med. 2011; 34 (1): 110–7.
16. Hirtz D, Thurman DJ, Gwinn-Hardy K, Mohamed M, Chaudhuri AR, Zalutsky R. How common are the ‘common’ neurologic disorders? Neurology. 2007; 68: 326–37.
17. Katzmarzyk PT, Church TS, Craig CL, Bouchard C. Sitting time and mortality from all causes, cardiovascular disease, and cancer. Med Sci Sports Exerc. 2009; 41: 998–1005.
18. Latimer AE, Ginis KAM, Craven BC, Hicks AL. The physical activity recall assessment for people with spinal cord injury: validity. Med Sci Sports Exerc. 2006; 38 (2): 208–16.
19. Maher CA, Williams MT, Olds T, Lane AE. Physical and sedentary activity in adolescents with cerebral palsy. Dev Med Child Neurol. 2007; 49 (6): 450–7.
20. Maltais DB, Pierrynowski MR, Galae VA, Bar-Or O. Physical activity level is associated with the O2 cost of walking in cerebral palsy. Med Sci Sports Exerc. 2005; 37 (3): 347–53.
21. Matthews CE, Chen KY, Freedson PS, et al. Amount of time spend in sedentary behaviors in the United States, 2003–2004. Am J Epidemiol. 2008; 167 (7): 875–81.
22. Martin Ginis KA, Latimer AE, Hicks AL, Craven BC. Development and evaluation of an activity measure for people with spinal cord injury. Med Sci Sports Exerc. 2005; 37 (7): 1099–111.
23. Nieuwenhuijsen C, Van der Slot WM, Beelen A, et al. Inactive lifestyle in adults with bilateral spastic cerebral palsy. J Rehabil Med. 2009; 41: 375–81.
24. Nieuwenhuijsen C, Van der Slot WMA, Dallmeijer AJ, et al. Physical fitness, everyday physical activity, and fatigue in ambulatory adults with bilateral spastic cerebral palsy. Scand J Med Sci Sports. 2011; 21: 535–42.
25. Ng SY, Dinesh SK, Tay SK, Lee EH. Decreased access to health care and social isolation among young adults with cerebral palsy after leaving school. J Orthop Surg. 2003; 11: 80–9.
26. Nooijen CFJ, Slaman J, Stam HJ, Roebroeck ME, Van den Berg-Emons RJ. Inactive and sedentary lifestyles amongst ambulatory adolescents and young adults with cerebral palsy. J Neuroeng Rehabil. 2014; 11: 49.
27. Obeid J, Nguyen T, Gabel L, Timmons BW. Physical activity in Ontario preschoolers: prevalence and measurement issues. Appl Physiol Nutr Metab. 2011; 36: 291–7.
28. Owen N, Healy GN, Matthews CE, Dunstan DW. Too much sitting: the population health science of sedentary behavior. Exerc Sport Sci Rev. 2010; 38 (3): 105–13.
29. Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997; 39: 214–23.
30. Pedersen BK, Saltin B. Evidence for prescribing exercise as a therapy in chronic disease. Scand J Med Sci Sports. 2006; 16: 3–63.
31. Prince SA, Adamo KB, Hamel ME, Hardt J, Gorber SC, Tremblay M. A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008; 5: 56.
32. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M. A report: the definition and classification of cerebral palsy. Dev Med Child Neurol. 2007; 49: 8–14.
33. Strath SJ, Pfeiffer KA, Whitt-Glover MC. Accelerometer use with children, older adults, and adults with functional limitations. Med Sci Sports Exerc. 2012; 41 (1): S77–85.
34. Swartz AM, Strath SJ, Bassett DR, O’Brien WL, King GA, Ainsworth BE. Estimation of energy expenditure using CSA accelerometers at hip and wrist sites. Med Sci Sports Exerc. 2000; 32 (9): 450–6.
35. Tanhoffer RA, Tanhoffer AI, Raymond J, Hills AP, Davis GM. Comparison of methods to assess energy expenditure and physical activity in people with spinal cord injury. J Spinal Cord Med. 2012; 35 (1): 35–45.
36. Trost SG, Pate RR, Freedson PS, Sallis JF, Taylor WC. Using objective physical activity measures with youth: how many days of monitoring are needed? Med Sci Sports Exerc. 2000; 32 (2): 426–31.
37. Van der Slot WM, Roebroeck ME, Landkroon AP, Terburg M, Van den Berg-Emons RJ, Stam HJ. Everyday physical activity and community participation of adults with hemiplegic cerebral palsy. Disabil Rehabil. 2007; 29: 179–89.
38. Verschuren O, Darrah J, Novak I, Ketelaar M, Wiart L. Health enhancing physical activity in children with cerebral palsy: more of the same is not enough. Phys Ther. 2014; 94 (2): 297–305.
39. Ward DS, Evenson KR, Vaughn A, Rodgers AB, Troiano RP. Accelerometer use in physical activity: best practices and research recommendations. Med Sci Sports Exerc. 2005; 37 (11): S582–8.
40. Warms CA, Belza BL. Actigraphy as a measure of physical activity for wheelchair users with spinal cord injury. Nurs Res. 2004; 53 (2): 136–43.
41. Welk GJ. Use of accelerometry-based activity monitors to assess physical activity. In: Welk GJ, ed. Physical Activity Assessments for Health-Related Research. Champaign, IL: Human Kinetics, 2002: 125–41.
42. World Health Organization. Global Recommendations on Physical Activity for Health. Geneva, Switzerland: World Health Organization; 2010.


© 2015 American College of Sports Medicine