Keywords

chronic conditions, disability, exercise, measurement methods, physical activity

 

Authors

  1. Warms, Catherine PhD, RN

Abstract

Measuring the physical activity of persons with chronic and disabling conditions presents complexities related to measuring instruments, the intensity of the activity being measured, the population being measured, and individual behavior and health status. They often have limitations in mobility that do not preclude physical activity but contribute to the complexity of measuring it, such as slow or altered gait, inability to walk, and the need for assistive devices. This article reviews currently available ways to measure physical activity, describes strengths and weaknesses of various measures, and provides examples of complexities in measuring physical activity in people who move differently.

 

Article Content

REGULAR physical activity or exercise is known to provide health and fitness benefits (muscular strength, cardiorespiratory and muscular endurance, flexibility, reduced body fat, improved physical function, and decreased depression and anxiety), decrease the risk of chronic conditions (coronary heart disease, diabetes, obesity, hypertension, stroke, colorectal cancer, breast cancer, and osteoporosis), and contribute to general well-being and quality of life.1,2 These benefits apply across populations and, although even small increases in activity have been shown to be beneficial, a dose-response relationship between activity levels and lowered mortality risks is substantiated.3,4 Because of the strength of evidence supporting a dose-dependent relationship, public health officials recommend that all Americans, including people with disabilities (PWD), engage in at least 30 minutes a day of moderate aerobic activity.2,4 Although this recommendation is widely published, most people are still not obtaining the recommended amount of physical activity, and PWD report more inactivity than does the general population.5,6 Because a large proportion of Americans are reported to have a disability (19.3% of those aged 5 and older according to the 2000 census),7 promoting physical activity among PWD may have a large impact on public health. Healthy People 2010 emphasizes the importance of improving the health of PWD and lists specific goals and objectives for reducing the disparity in levels of physical activity between adults with and without disabilities.8 Thus, measuring physical activity dose (type, intensity, frequency, and duration) is important for health promotion researchers and program planners to obtain accurate estimates of physical activity among PWD and to have sensitive measures to detect change in order to determine how well a physical activity promotion program is working.

 

In the US Surgeon General's report on physical activity,1 it was noted that there were "extreme complexities" in measurement of physical activity. Measuring the activity of people with mobility-limiting chronic and disabling conditions adds another layer of complexity to the measurement problem because these conditions change the nature of physical activity itself: how major muscles are used, how much energy expenditure (EE) is required, and what types of activity can be done. Therefore, selecting the best method of measurement requires a thorough understanding of physical activity concepts, a match of conceptual understanding with measurement method capabilities and limitations, and an in-depth knowledge of the characteristics of the population. The purpose of this article is to provide a framework for thinking about and making decisions about physical activity measurement in PWD. The framework includes conceptual clarifications (understanding what is being measured), population characteristics (how a person moves and his or her physical and cognitive abilities), and measurement methods (knowing the strengths and limitations of various measures).

 

CONCEPTUAL CLARIFICATIONS

Clarity about what is being measured is essential. It is important to make distinctions between exercise, fitness, and physical activity. The definitions set forth by Caspersen et al9 are well-understood and considered to be the standard. Physical activity, the broadest of the 3 concepts, is defined as "any bodily movement produced by skeletal muscles that results in energy expenditure above basal requirements."9(p126) Physical activity can occur in any setting and is often categorized by type or purpose (eg, occupational, household, sports). Exercise is a type of physical activity that is "planned, structured and repetitive and is done to improve or maintain 1 or more of the components of physical fitness."9(p126)Fitness is "a set of attributes that people have or achieve that are either health-related (cardiorespiratory and muscular endurance, muscular strength, body composition and flexibility) or are skill-related (agility, balance, coordination, speed, power, and reaction time)."9(p126) Not all activity is exercise, but both activity and exercise can contribute to fitness. This article focuses on measurement of activity and will not discuss measurement of fitness or other outcomes of activity.

 

Physical activity has many dimensions that can be measured and quantified. These include EE, type, frequency, intensity and duration of activity, long-term patterns of activity, distance covered or steps taken during activity, and a host of outcomes including immediate responses to activity and fitness outcomes. Some measurement methods provide good data for some dimensions but not for others, and there are no methods to provide valid and reliable data for all dimensions. Physical activity and EE cannot be equated.10 This is an important consideration and point of common confusion. Physical activity is movement (and yes, energy is expended during movement), but EE for activity varies significantly from person to person, depending on age, gender, body mass, and efficiency or manner of movement. Total EE includes the EE of basal metabolism and the thermic effect of food as well as the EE of activity. Explicit definition of the dimension(s) of interest is the first step for selecting a measurement method appropriate for quantifying that dimension.

 

POPULATION CHARACTERISTICS

Disability is an umbrella term used to describe the interplay of health conditions (impairments) and environments to produce activity limitations and participation restrictions in the lives of people with these conditions.6 People with disabilities, a term chosen by this population, is used in this article to indicate the broader population with any type of disabling health condition. Several characteristics of the physical activity of PWD contribute to the complexity of activity measurement. People with disabilities have lower levels of activity, move differently, may use assistive devices, and may have limitations in cognition or fine motor control. Many surveys have documented low levels of self-reported physical activity in PWD.11-14 Studies that have measured the activity of PWD compared with people without disabilities have also confirmed lower activity (decreased frequency and decreased intensity).15-17 The difficulty in measuring low activity levels is that most measurement questionnaires and some measurement devices are relatively insensitive and are unable to accurately represent this level of activity. Insensitivity of measures often results in floor effects and inability to discriminate between various levels of activity.5,18

 

One consequence of variability in "ways of moving" as well as in varying amounts of active muscle mass due to underlying pathology is that there are no existing algorithms for predicting activity EE in PWD. The 1993 Compendium of Physical Activities and its update in 200019,20 specify that data collected and norms specified apply only to "healthy, normal" adults, "without conditions that would significantly alter their metabolic or mechanical efficiency."20(pS502) Other researchers concur that the only valid energy prediction formula for a neurologically or physically impaired population would have to be developed from that population.16,21 Considering the number of disabling conditions and large variability in how these conditions affect EE, attempts to construct a Compendium of Physical Activities for such diverse populations may not be worthwhile.

 

Factors that have an impact on the accuracy of heart rate monitoring as an indicator of activity (body weight, muscle mass, temperature, medications, fatigue, and emotional stress) are particularly important for PWD. For example, people with spinal cord injury (SCI) and multiple sclerosis (MS) have varying degrees of autonomic nervous system involvement that may limit maximal heart rate in response to exercise.22 Children with cerebral palsy were found to have an "unmatched response pattern" between EE as measured by Vo2 and heart rate.21 People with disabilities frequently take multiple medications that can affect heart rate, and muscle mass is affected by many neuromuscular conditions.23

 

Another consequence of moving differently is that humans tend to compensate for difficulties in walking or propelling a chair by moving more slowly, which poses challenges for accurate EE measurement.17,24 An illustration of this is a recent study by Littlewood et al,25 which measured EE using calorimetry during rest, walking, and running in children with acquired TBI compared with same-age controls without disabilities. EEs during rest and walking were equal, but when children with TBI ran, they expended 42% more energy per step when asked to maintain a given pace. When allowed to choose their own pace, the children with TBI chose a slower pace than did controls, compensating for the higher energy costs by decreasing speed significantly. Such energy conservation strategies contribute to the problem of insensitive instruments. People with short strides or a shuffling gait have less vertical displacement of the center of gravity, making it more likely that pedometers or waist-mounted accelerometers will not register steps or activity.26 Similarly, supporting body weight with a cane, crutches, walker, or handrails reduces the energy cost of locomotion.27,28 And, using a walker decreases vertical hip movement promoting a shuffling gait that may not be detected by pedometers or accelerometers.29

 

Many physical activity researchers and interventionists consider walking to be a "quintessential" activity behavior, so much so that a measure of walking is often suggested to be an adequate representation of a person's overall activity level. This thinking suggests that people who do not walk cannot be physically active, an idea that wheelchair athletes and active wheelchair users would dispute. It is known that in ambulatory people, movement of the upper extremities accounts for only a small portion of total EE.30 Because wheelchair users have a greater portion of upper-extremity movement, quantifying upper-extremity movement is necessary for an adequate measure of physical activity. Wrist actigraphs have been shown to be valid indicators of physical activity for wheelchair users in laboratory and free-living conditions.31,32 For wheelers, wrist actigraphy is well-tolerated, does not interfere with regular activity, avoids risk of skin breakdown due to waist or hip mounting, and safeguards against underestimation of activity due to insensitive instrumentation or inappropriate placement (eg, wheeling would not be detected by a waist or hip placement).

 

MEASUREMENT METHODS

In 1985, Laporte and colleagues identified more than 30 methods of measuring physical activity reported in the literature.33 They concluded that no existing method could fulfill the criteria of being valid (measuring what it is intended to measure), reliable (consistently giving the same results under the same conditions), accurate (precise), and practical (having acceptable costs for both researchers and participants), while not affecting behavior. Twenty years later, the situation has not changed substantially. There are numerous methods to choose from, but none can meet all of the above criteria. Given this, knowledge of the strengths and weakness of each method is necessary.

 

Gold standards

The most objective and precise methods measure EE. They are frequently used as "gold standards" for comparison with other methods. These include doubly labeled water (DLW), direct and indirect calorimetry, and direct observation. DLW involves administering an oral dose of water containing specific isotopes of hydrogen and oxygen. Measurement of the rate of loss of these isotopes over time reflects the rate of CO2 production, which can be used to estimate EE over a 1- to 2-week period.34 Although DLW does not provide information on type, frequency, duration, or intensity of physical activity, its strengths are that it does not interfere with regular activity and is an extremely accurate measurement of long-term EE. However, the high cost and relatively scarce supply of the labeled water as well as special equipment and highly trained personnel required for carrying out testing make this method unsuitable for large-scale studies, and the necessity for collection of complete urine samples limits its usefulness for PWD who may have incontinence or use urinary-collection equipment.

 

Calorimetric methods determine EE by measuring heat loss from the body (direct calorimetry) or expired gases to determine VO2 (indirect calorimetry). Direct calorimetry is performed in a self-contained chamber, severely limiting the participant's ability to move about freely. Indirect calorimetry is done using a portable metabolic measurement system consisting of a face mask and portable gas analyzer carried in a backpack or on a cart. Though not as expensive as DLW, calorimetric methods require specialized equipment and trained personnel and inhibit normal activity. Calorimetry is particularly useful for determining EE for specific activities over short periods in controlled conditions but cannot be used to measure unconstrained daily physical activity in real-world conditions. Though considered to be "gold standard" measures, DLW and calorimetry are excellent indicators of EE but are not perfect indicators of physical activity.

 

Direct observation of activity is another method for validating other physical activity measures. Observation methods usually involve 1 or more human observers watching and recording activity, such as playground activities or household activity. Like the other "gold standards," this method has limitations that restrict its usefulness to small-scale studies. Observation is a good method for recording activity in a small sample in a confined space, but it is time- and personnel-intensive, depending on the quality of the rating instrument and raters, may alter the behavior of those being observed (social acceptability bias), and is not workable for measuring free-living activity.35 Although DLW, calorimetry, and activity observation methods can be and have been used in studies of people with chronic disabling conditions,15,36,37 their strengths and limitations in these populations are the same as listed above.

 

Heart rate monitors

The use of heart rate monitors as an indicator of physical activity assumes a linear relationship between activity intensity and heart rate. As mentioned previously, this method is particularly unreliable for many PWD because of differences in physiological responses.21-23,38 Methods of heart rate measurement with individual calibration (HRFlex method) and a combined HR and motion sensor method improve the accuracy of heart rate monitoring as a proxy for physical activity and are inexpensive and do not limit normal physical activity.38,39 But they involve very complex and time-consuming analyses that limit their practical use in large-scale studies.

 

Motion sensors

Motion sensors include pedometers, step counters, and a variety of accelerometry-based activity recording devices designed to provide objective measures of movement. Pedometers are relatively inexpensive, waist-mounted, electronic devices designed to measure cumulative step counts. They are simple to use and unobtrusive to wear, and several models have been shown to be very accurate (eg, Kenz Lifecorder, New Lifestyles NL-2000, Yamax Digiwalker SW-200 & SW-701).40 Limitations of pedometers are that they are unable to quantitate frequency and intensity of movement, require an ambulatory population, and require that the wearer periodically record step counts. A recent systematic review of the utility of pedometers for physical activity measurement found that their readings are strongly correlated with accelerometers (median r = 0.86) and time in observed activity (median r = 0.82), moderately correlated with different measures of EE (median r = 0.68), and weakly correlated with self-reported physical activity (median r = 0.33).41 Further, given their small size and ease of use, pedometers are feasible for population surveillance of walking activity.10

 

A major issue regarding the use of pedometers for PWD is that most pedometers are unable to accurately measure steps at slow walking speeds and are often not sensitive enough to detect steps with abnormal gait patterns.42 In a study of residents of a nursing home, a pedometer was shown to underestimate the number of steps taken by 46% to 74%, depending on gait and walking speed.29 Similarly, a study using 2 different pedometers compared to actual steps taken by ambulatory individuals with MS found significant differences between the 2 measures at low speeds.43 A relatively new ankle-mounted electronic step counter (StepWatch Activity Monitor [SAM]) is more sensitive to slow and abnormal gait patterns.21 The SAM is more sophisticated than a simple pedometer because it incorporates accelerometer technology (see the next section for further explanation). It can be programmed to account for individual gait height, cadence, and speed, and has been shown to be highly accurate and reliable in counting steps in people with stroke, MS, Parkinson's disease, and primary muscle disease.16,26,44 The ankle-mounted position is well-tolerated and unobtrusive; however, the SAM is more expensive than a simple pedometer.

 

Accelerometry-based monitors quantify body movement over days to weeks through the use of a piezoelectric sensor that generates electrical charges when the device changes direction or acceleration. Depending on the number and orientation of the ceramic sensors, accelerometers can measure movement in 1 (uniaxial, vertical plane only) or 3 (triaxial, omnidirectional) planes. These monitors measure motion and provide data on whether or not a person is active as well as the frequency and intensity of activity. Existing monitors vary considerably on continuums of complexity, cost, and sensitivity to motion. Most require a computer interface for programming and data management. The requirement for computer programming as well as the relatively high cost of accelerometers compared to pedometers or self-report measures limits their usefulness in large epidemiological studies.

 

Accelerometers can be worn on the waist, hip, back, or wrist, depending on the manufacturer's recommendations, population characteristics, and type of data desired. Different body sites move differently. Depending on what a person is doing, the movements of 1 body site are not necessarily correlated with those of other sites. Thus, selecting the site to place the monitor is important. Theoretically, the most accurate prediction of EE would be accomplished by using multiple accelerometers,45 but the trade-offs with interference with normal activity and discomfort for wearers must be considered. Wearing the monitor at the waist is the most common placement. In general, accelerometers tend to overestimate EE for light-intensity activities and underestimate it for high-intensity activities. Waist-mounted monitors specifically overestimate EE of walking and jogging and underestimate EE of stair-climbing, bicycling, and arm ergometry.46,47 Measurement error of waist-mounted devices is related to inability to detect arm movements as well as static work (lifting, pushing, carrying loads, and grasping).48

 

In populations with low levels of activity, triaxial accelerometers have been shown to be more sensitive than other measures in detecting variability in activity levels.17,45,49-51 Wrist actigraphs (small accelerometer units worn on the wrist initially intended for measurement of activity during sleep) are extremely sensitive to low-intensity activity. A trade-off for the greater sensitivity provided by actigraphy is the possibility of overestimating activity because of extraneous movements or environmental vibration being "counted" as physical activity. Although most devices are now constructed with filters to remove vibration artifact caused by vehicles or other sources of constant vibration, devices may sense nonrepeated artifacts (such as bumps in the road) and involuntary movements (such as muscle spasms) as activity. Given that accelerometers are the most sensitive devices for measuring movement, researchers hoping to measure the full range of physical activity in the PWD should consider using them.

 

Questionnaires and self-report log

Questionnaires and self-report logs or activity diaries are the most widely used and least expensive way of measuring physical activity and are clearly advantaged for use in large population-based or epidemiological studies. A drawback, however, is that people commonly overrate their own physical activity.34,52,53 These methods are best for measuring activities, such as programmed exercise, recreation, or sports activities, that are easily recalled (because it is known that the activities that are self-chosen and occur at a discrete time are more easily remembered than those that occur incidentally throughout the course of a day). Diaries or logs may impart a significant level of participant burden by the time required to monitor and record one's daily activities, and diaries are also subject to inaccuracy and social acceptability bias. Most existing self-report measures are not sensitive enough to measure the lower end of the physical activity continuum, often demonstrating floor effects in which the lowest score is too high for inactive respondents and not able to discriminate small but important differences in the level of activity.10,54 It is important to select the most appropriate questionnaire in relation to the specific population being investigated. For PWD, measures need to assess low-intensity and low-frequency activity as well as to be worded in such a way that alternative ways of moving about are measured. There are 3 such instruments that have been specifically constructed for measurement of physical activity in PWD. These are the Physical Activity and Disability Survey (PADS),5 the Physical Activity Scale for Individuals with Physical Disabilities (PASIPD),55 and the Human Activity Profile (HAP).56

 

The PADS is a 46-item questionnaire constructed to measure usual weekly activity. It consists of 3 subscales, measuring exercise, leisure time physical activity, and household activity. It is a semistructured interview that requires 30 to 40 minutes to administer. The questionnaire was initially validated in 103 adults with disabilities (varying conditions, all able to walk a minimum of 50 ft). Internal consistency reliability Cronbach's alpha for the subscales was reported as ranging from .67 to .95, and test-retest reliability (r) over a 1-week interval from = 0.78 to 0.99. Validity was supported by significant correlations between subscale and total scores and peak VO2.5 The authors report that the PADS is sensitive to change after an exercise intervention. However, there are no published studies in which it has been used so it is unknown if the instrument is sufficiently sensitive and specific for wide applicability.

 

The PASIPDS is also recently developed and is not widely tested. It is a 13-item paper-based questionnaire on which participants record number of days per week and hours per day of participation in recreational activities (including exercise), household activities, occupational activities, and inactivity over the past 7 days. It was evaluated in a sample of 227 men and 145 women with a wide range of disabilities.55 Construct validity was supported by factor analysis and differentiation between groups. Five factors were identified (home repair/lawn/garden, housework, light exercise/sport/recreation, vigorous exercise/sport/recreation, and occupation/transportation) accounting for 63% of the variance in the total score. The total scale score was able to significantly differentiate those with excellent health from those with poor health and those reporting no activity versus those reporting moderate and extreme activity.55 The PASIPDS is disability-sensitive (includes examples of wheeling activities and adapted sports and recreation). One limitation is that it uses MET values established for the general population to calculate estimated EE (although there is really no other choice currently); another is that it has not been validated by an external criterion measure.

 

The HAP is an older, more established instrument originally developed for use in chronic-obstructive pulmonary disease rehabilitation programs to rapidly survey common human activities, especially activities with very low EE. It consists of 94 items listing common activities in order from those requiring the least to most EE. Respondents are asked to mark after each listed activity in 1 of 3 columns ("still doing this," "stopped doing this," or "have never done this"). The HAP produces a number of scoring possibilities including maximum activity (the most strenuous activity a person is still able to perform), adjusted or "usual" activity, activity classification (compared to same-age peers), etc. It is very sensitive to impairment and has been validated in samples with chronic pain, chronic-obstructive pulmonary disease, myocardial infarction, osteoarthritis, and renal disease.56,57 It has also been used in studies of people with MS, post-polio fatigue, and stroke.58-60 Test-retest reliability is good to excellent (r = 0.79-0.96).56,57 Compared with other activity questionnaires in people with end-stage renal disease, the HAP was the most highly correlated with accelerometer measures accounting for 44% of the variance.50 In the study, combining the PASE (the parent measure to the PASIPDS) and the HAP scores accounted for 75% of the variance. A critique of the HAP is that it is really more of a measure of functional ability than actual physical activity.17 It asks about current performance or ability to perform activities but does not quantitate actual frequency or duration of physical activity. It may be best used in conjunction with another measure of physical activity to more fully explain the relationship between functional ability and physical activity but it is not a good indicator of actual physical activity by itself. Table 1 presents a summary of the various methods for measuring physical activity and their strengths and limitations in people with chronic disabling conditions.

  
Table 1 - Click to enlarge in new windowTable 1. Measurement methods, strengths, and weaknesses

DISCUSSION AND CONCLUSIONS

The question of how to select the best method for a particular study requires application of the framework proposed in this article. First, conceptual clarity is needed. What is the purpose of the study? Is it a survey of the types of sports and recreational activities in which people participate? Or is it a clinical trial of a program to promote more time spent in physical activity? Is it important to know actual EE? Are you hoping to change behavior or improve fitness? If research questions are formulated, it is important to clarify the aspect of activity or inactivity that corresponds with the posed questions.28 Specifying the dimension of interest defines what should be measured, which usually narrows choices for how to measure it.

 

The next step is to ask, what are the characteristics of the population of interest? How does the disabling condition affect the way the individual moves? Does the impairment change physical activity options, intensity, frequency, or duration? Is measurement of EE a realistic possibility? How does the use of assistive devices interact with the measurement method? It is necessary to keep in mind that people with chronic mobility-limiting conditions might be offended at being asked repeatedly about how many flights of stairs they can climb or how far they can walk and, as a result, may reject participating in studies using such measures. Also, someone who needs to sit frequently may find wearing a waist-mounted activity monitor very uncomfortable. Paper diaries may not be workable options for people with difficulty writing or limitations in cognition or memory. These are some of the practical issues that should be considered along with the scientific issues.

 

Finally, it is important to understand how various measuring techniques work and their strengths and limitations. Questions to consider include the following: How much error can be tolerated? How many participants will be in the study? What is reasonable to expect of these participants? What is your budget? What kind of data do you need and are able to analyze? Are there existing measures that are valid, reliable, and sensitive? Some compromises will be necessary given that there are no "perfect" measures.

 

The question remains: How can the inherent complexities of physical activity measurement in people with chronic disabling conditions best be addressed? Following the decision-making framework set forth in this article is one potentially helpful approach. Another suggestion is to consider using multiple methods of measurement.10,38,50 Wareham and Rennie state that "precision can be improved by assessing exposure using combined methods provided that they have uncorrelated error."38(pS32) Applying a similar approach to physical activity measurement in people with chronic and disabling conditions, it may be appropriate to combine a measure of movement (accelerometry), a measure of habitual activity (questionnaire), and an indicator of fitness (6-minute walk or wheel) to more clearly delineate the amount of activity, the regularity of activity, and the effects or benefits of activity. Research regarding the best combination of measures is warranted.

 

Improving physical activity measurement for populations with chronic and disabling conditions will facilitate adaptation of physical activity programs known to be effective in other populations for use with this specific population with known differences in amount and types of physical activity options. Such adaptations have been recommended by public health officials to address known health disparities related to lack of access to effective programs.8 Selection of appropriate sensitive, valid, and reliable indicators of physical activity will not guarantee program effectiveness, but it will provide better indications of the actual physical activity dose required to enhance health and fitness in the PWD.

 

REFERENCES

 

1. US Surgeon General. Surgeon General's report on physical activity and health. JAMA. 1996;276(7):522. [Context Link]

 

2. NIH Consensus Development Panel. Physical activity and cardiovascular health. JAMA. 1996;276(7):241-246. [Context Link]

 

3. Sparling PB, Owen N, Lambert EV, Haskell WL. Promoting physical activity: the new imperative for public health. Health Education Research. 2000;15(3):367-376. [Context Link]

 

4. Pate RR, Pratt M, Blair SN, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA. 1995;273:402-407. [Context Link]

 

5. Rimmer JH, Riley BB, Rubin SS. A new measure for assessing the physical activity behaviors of persons with disabilities and chronic health conditions: the Physical Activity and Disability Survey. American Journal of Health Promotion. 2001;16(1):34-45. [Context Link]

 

6. Heath GW, Fentem PH. Physical activity among persons with disabilities-a public health perspective. Exercise and Sport Sciences Reviews. 1997;25:195-234. [Context Link]

 

7. Waldrop J, Stern S. Disability status: 2000. US Census Bureau, US Dept of Commerce. Available at: http://www.census.gov/hhes/www/disability.html. Accessed June 1, 2005. [Context Link]

 

8. US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA: USDHHS, Centers for Disease Control and Prevention; 1996. [Context Link]

 

9. Caspersen CJ, Powell KE, Christenson GM. Physical activity, exercise, and physical fitness: definitions and distinctions for health-related research. Public Health Reports. 1985;100(2):126-131. [Context Link]

 

10. Tudor-Locke CE, Myers AM. Challenges and opportunities for measuring physical activity in sedentary adults. Sports Medicine. 2001;3(2):91-100. [Context Link]

 

11. Hogan A, McClellan L, Bauman A. Health promotion needs of young people with disabilities: a population study. Disability and Rehabilitation. 2000;22(8):352-357. [Context Link]

 

12. LaPlante M, Carlson D. Disability Statistics Report (7). Washington, DC: US Dept of Education, National Institute of Disability and Rehabilitation Research; 1992. [Context Link]

 

13. Washburn R, Hedrick BN. Descriptive epidemiology of physical activity in university graduates with locomotor disabilities. International Journal of Rehabilitation Research. 1997;20(3):275-287. [Context Link]

 

14. Coyle CP, Kinney WB. Leisure characteristics of adults with physical disabilities. Therapeutic Recreation Journal. 1990;24:64-73. [Context Link]

 

15. Monroe MB, Tataranni PA, Pratley R, Manore MM, Skinner JS, Ravussin E. Lower daily energy expenditure as measured by a respiratory chamber in subjects with spinal cord injury compared with control subjects. The American Journal of Clinical Nutrition. 1998;68:1223-1227. [Context Link]

 

16. Busse ME, Pearson OR, VanDeusen R, Wiles CM. Quantified measurement of activity provides insight into motor function in neurological disease. J Neurol Neurosurg Psychiatr. 2004;75:884-888. [Context Link]

 

17. Ng AV, Kent-Braun JA. Quantitation of lower physical activity in persons with multiple sclerosis. Medicine and Science in Sports and Exercise. 1997;29(4):517-523. [Context Link]

 

18. Dearwater SR, LaPorte RE, Cauley JA, Brenes G. Assessment of physical activity in inactive populations. Medicine and Science in Sports and Exercise. 1985;17(6):651-655. [Context Link]

 

19. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of physical activities: classification of energy costs of human physical activities. Medicine and Science in Sports and Exercise. 1993;25(1):71-80. [Context Link]

 

20. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of physical activities: an update of activity codes and MET intensities. Medicine and Science in Sports and Exercise. 2000;32(9):S471-S480. [Context Link]

 

21. Keefer DJ, Tseh W, Caputo JL, Apperson K, McGreal S, Morgan DW. Comparison of direct and indirect measures of walking energy expenditure in children with hemiplegic cerebral palsy. Developmental Medicine and Child Neurology. 2004;46(5):320-324. [Context Link]

 

22. Gater D, Ugalde V. Physiologic foundations for exercise prescription in tetraplegia. In: Shankar K, ed. Exercise Prescription. Philadelphia: Hanley and Belfus Inc; 1998:199-246. [Context Link]

 

23. Durstine JL, Moore GE. ACSM's Exercise Management for Persons With Chronic Disease and Disabilities. 2nd ed. Champaign, IL: Human Kinetics; 2003. [Context Link]

 

24. Hunter J. Energy costs of wheelchair propulsion by elderly and disabled people. International Journal of Rehabilitation Research. 1987;10(4):50-54. [Context Link]

 

25. Littlewood RA, Davies PSW, Cleghorn GJ, Grote RH. Physical activity cost in children following an acquired brain injury-a comparative study. Clinical Nutrition. 2004;23:99-104. [Context Link]

 

26. Haeuber E, Shaughnessy M, Forrester LW, Coleman KL, Macko RF. Accelerometer monitoring of home- and community-based ambulatory activity after stroke. Archives of Physical Medicine and Rehabilitation. 2004;85:1997-2001. [Context Link]

 

27. Green MA, Foster C. Effect of magnitude of handrail support on prediction of oxygen uptake during treadmill testing. Medicine and Science in Sports and Exercise. 1991;23:166. [Context Link]

 

28. Bussmann JBJ. Techniques for measurement and assessment of mobility in rehabilitation: a theoretical approach. Clinical Rehabilitation. 1998;12:455-464. [Context Link]

 

29. Cyarto EV, Myers AM, Tudor-Locke C. Pedometer accuracy in nursing home and community-dwelling older adults. Medicine and Science in Sports and Exercise. 2004;36(2):205-209. [Context Link]

 

30. Kumahara H, Tanaka H, Schutz Y. Daily physical activity assessment: what is the importance of upper limb movements vs whole body movements? International Journal of Obesity. 2004;28:1105-1110. [Context Link]

 

31. Washburn RA, Copay AG. Assessing physical activity during wheelchair pushing: validity of a portable accelerometer. Adapted Physical Activity Quarterly. 1999;16:290-299. [Context Link]

 

32. Warms CA, Belza BL. Actigraphy as a measure of physical activity for wheelchair users with spinal cord injury. Nursing Research. 2004;53(2):136-143. [Context Link]

 

33. LaPorte RE, Montoye HJ, Caspersen CJ. Assessment of physical activity in epidemiologic research: problems and prospects. Public Health Reports. 1985;100(2):131-146. [Context Link]

 

34. Ainslie PN, Reilly T, Westerterp KR. Estimating human energy expenditure: a review of techniques with particular reference to doubly labeled water. Sports Medicine. 2003;33(9):683-698. [Context Link]

 

35. Fernhall B, Unnithan VB. Physical activity, metabolic issues, and assessment. Physical Medicine and Rehabilitation Clinics of North America. 2002;13:925-947. [Context Link]

 

36. Mollinger LA, Spurr GB, ElGhatit AZ, et al. Daily energy expenditure and basal metabolic rates of patients with spinal cord injury. Archives of Physical Medicine and Rehabilitation. 1985;66:420-426. [Context Link]

 

37. Cooper R, Quatrano L, Axelson P, et al. Research on physical activity and health among people with disabilities: a consensus statement. Journal of Rehabilitation Research and Development. 1999;36(2):142-154. [Context Link]

 

38. Wareham NJ, Rennie KL. The assessment of physical activity in individuals and populations: why try to be more precise about how physical activity is assessed? International Journal of Obesity and Related Metabolic Disorders. 1998;22(suppl 2):S30-S38. [Context Link]

 

39. Strath SJ, Bassett DR, Thompson DL, Swartz AM. Validity of the simultaneous heart rate-motion sensor technique for measuring energy expenditure. Medicine and Science in Sports and Exercise. 2002;34(5):888-894. [Context Link]

 

40. Schneider PL, Crouter SE, Bassett DR. Pedometer measures of free-living physical activity: comparison of 13 models. Medicine and Science in Sports and Exercise. 2004;36(2):331-335. [Context Link]

 

41. Tudor-Locke CE, Williams JE, Reis JP, Pluto D. Utility of pedometers for assessing physical activity: construct validity. Sports Medicine. 2004;34(5):281-291. [Context Link]

 

42. Steele BG, Belza B, Cain K, Warms C, Coppersmith J, Howard J. Bodies in motion: monitoring daily activity and exercise with motion sensors in people with chronic pulmonary disease. Journal of Rehabilitation Research and Development. 2003;40(5):45-58. [Context Link]

 

43. Motl R, McAuley E, Snook EM, Scott JA. Accuracy of two electronic pedometers for measuring steps under controlled conditions among ambulatory individuals with MS. Multiple Sclerosis. 2005;11:343-345. [Context Link]

 

44. Macko RF, Haeuber M, Shaughnessy M, et al, Micro-processor-based ambulatory activity monitoring in stroke patients. Medicine and Science in Sports and Exercise. 2002;34:394-399. [Context Link]

 

45. Westerterp KR. Physical activity assessment with accelerometers. International Journal of Obesity and Related Metabolic Disorders. 1999;23(suppl 3):S45-S49. [Context Link]

 

46. Campbell KL, Crocker PRE, Mckenzie DC. Field evaluation of energy expenditure in women using tritrac accelerometers. Medicine and Science in Sports and Exercise. 2002;34(10):1667-1674. [Context Link]

 

47. Bassett DR, Ainsworth BE, Swartz AM, Strath SJ, O'Brien WL, King GA. Validity of four motion sensors in measuring moderate intensity physical activity. Medicine and Science in Sports and Exercise. 2000;32(9):S471-S480. [Context Link]

 

48. Kumahara H, Schutz Y, Ayabe M, et al. The use of unaxial accelerometry for the assessment of physical-activity-related energy expenditure: a validation study against whole-body indirect calorimetry. The British Journal of Nutrition. 2004;28:1105-1110. [Context Link]

 

49. Sugimoto A, Hara Y, Findley TW, Yonemoto K. A useful method for measuring daily physical activity by a three-direction monitor. Scandinavian Journal of Rehabilitation Medicine. 1997;29(1):37-42. [Context Link]

 

50. Johansen KL, Painter P, Kent-Braun JA, et al. Validation of questionnaires to estimate physical activity and functioning in end-stage renal disease. Kidney International. 2001;59:1121-1127. [Context Link]

 

51. Steele BG, Holt L, Belza B, Ferris S, Lakshminaryan S, Buchner DM. Quantitating physical activity in COPD using a triaxial accelerometer. Chest. 2000;117(5):1359-1367. [Context Link]

 

52. Leenders NY, Sherman WM, Nagaraja HN. Comparisons of four methods of estimating physical activity in adult women. Medicine and Science in Sports and Exercise. 2000;32:S1320-S1326. [Context Link]

 

53. Montoye HJ, Kemper HCG, Saris WHM, Washburn RA. Measuring Physical Activity and Energy Expenditure. Champaign, IL: Human Kinetics; 1996. [Context Link]

 

54. Washburn RA, Janney CA, Fenster JR. The validity of objective physical activity monitoring in older individuals. Research Quarterly for Exercise and Sport. 1999;61(1):114-117. [Context Link]

 

55. Washburn RA, Zhu W, McAuley E, Frogley M, Figoni SF. The Physical Activity Scale for Individuals with Physical Disabilities: development and evaluation. Archives of Physical Medicine and Rehabilitation. 2002;83:193-200. [Context Link]

 

56. Daughton DM, Fix AJ. Human Activity Profile: Professional Manual. Odessa, FL: Psychological Assessment Resources Inc; 1988. [Context Link]

 

57. Bennell KL, Hinman RS, Crossley, et al. Is the Human Activity Profile a useful measure in people with knee osteoarthritis? Journal of Rehabilitation Research and Development. 2004;41(4):621-630. [Context Link]

 

58. Stuifbergen AK. Physical activity and perceived health status in persons with multiple sclerosis. The Journal of Neuroscience Nursing. 1997;29(4):238-243. [Context Link]

 

59. Packer T, Martins I, Krefting L, Brouwer B. Activity and post-polio fatigue. Orthopedics. 1991;14:1223-1226. [Context Link]

 

60. Teixeira-Salmela L, Olney S, Nadeaus S, Brouwer B. Muscle strengthening and physical conditioning to reduce impairment and disability in chronic stroke survivors. Archives of Physical Medicine and Rehabilitation. 1999;80:1211-1218. [Context Link]