This section describes concepts that are important for understanding the remaining sections of the Guide. The fundamental definitions and distinctions related to both physical activity and sedentary behavior are described first, followed by summaries of physical activity and sedentary behavior recommendations. Emphasis is then placed on the unique challenges of assessing physical activity and sedentary behavior in youth, as that population is NCCOR’s focus. These sections provide the foundation for understanding how to assess behavior and movement and estimate energy expenditure in youth.
Definitions and Terminology
Physical activity can be quantified and interpreted in a variety of ways. Caspersen et al. previously described physical activity as “Any bodily movement produced by skeletal muscles that result in caloric expenditure.”15 This definition has been widely accepted but a more recent conception, developed through a consensus conference on physical activity research,6,9 provides an operational definition to avoid subjectivity and facilitate assessment: "behavior that involves human movement, resulting in physiological attributes including increased energy expenditure and improved physical fitness."
A critical element in the new definition is the labeling of physical activity as a behavior. This captures the volitional nature of physical activity and the various physiologic, psychosocial, and environmental factors that influence it. For youth, the movement captured in this behavioral definition can be categorized as either structured (i.e., repetitive, organized activity, often led by an adult and performed in physical education class) or unstructured (i.e., play, unsupervised, activity performed during recess or school breaks). Activities are also operationally characterized according to their frequency (i.e., number of movements per day), duration (i.e., recorded minutes of actual movement), intensity (i.e., associated effort to perform the movement), and type (i.e., nature of movement as being, for example, aerobic or bone-strengthening related activities). The combination of frequency, duration, and intensity is often referred to as the dose or volume of physical activity and reflects the total amount of movement performed within a specific time period. Three other important distinctions with the definition are summarized below.
First: In this new conception of physical activity, a better understanding of the context and settings where physical activity behavior occurs (e.g., home, work) as well as the purpose (e.g., recreation, occupation) is needed. Although categories can vary, four broad domains that effectively capture behaviors for both adults and youth include: (1) leisure-time physical activity (i.e., recreation, play), (2) work- or school-related physical activity, (3) home or domestic physical activity, and (4) transportation physical activity (commuting from place to place). The term “exercise” is viewed as a subcategory of leisure-time physical activity that is more structured (e.g., steady state running) and performed with a well-defined purpose in mind (e.g., improving or maintaining physical fitness). The distinctions between physical activity and exercise are more relevant for adults, but participation in sports or structured activity programs or lessons by youth can be considered analogous to “exercise” because it is also structured and purposed. Adults often report physical activity for leisure or recreation but in youth this may be captured as simply “play” or unstructured activity.
Second: The new definition of physical activity stipulates that movement needs to be of sufficient magnitude to increase energy expenditure.
Understanding Energy Expenditure Terms
Total energy expenditure (TEE) is generally divided into three components: Resting energy expenditure (REE), the thermic effect of food (TEF), and the more volitional physical activity energy expenditure (PAEE). The REE value accounts for about 50 percent to 60 percent of total energy expenditure, but PAEE is usually of more relevance because it is the most variable component of TEE and is highly susceptible to change.
This definition helps to distinguish physical activity from non-volitional forms of movement (e.g., fidgeting) and focuses attention more on larger contributions to energy expenditure. Energy expenditure is typically expressed in units of kilojoules (kJ) or kilocalories (kcal), but it is also frequently expressed as multiples of resting energy expenditure known as Metabolic Equivalent Tasks (METs).16 Resting energy expenditure is often estimated because it is challenging to measure, and the value of 3.5 ml/kg/min has been widely adopted as the oxygen consumption of a person at rest. Using standard conversions and additional assumptions, resting energy expenditure (i.e., 1 MET) has been equated to an energy cost of 1 kcal/kg/hour. Other procedures yield different estimates but the consistent adoption of MET values and methods from the Compendium of Physical Activities17 has helped to standardize outcomes. Levels of physical activity are routinely calculated using established ranges (Rest is 1.0 to 1.4, Light physical activity [LPA] is 1.5 to 2.9, Moderate physical activity [MPA] is 3.0 to 5.9, Vigorous physical activity [VPA] is 6.0+). Most physical activity research has used a combined indicator that captures both moderate physical activity and vigorous physical activity (MVPA). However, research has increasingly emphasized the importance of understanding the allocation of time spent in different intensity classifications, as they each contribute directly to overall energy expenditure and health. Considerable attention has been given to time spent in sedentary behavior because it has been shown to be independent of time spent in MVPA. By default, the time spent in LPA also has implications because it falls between these two intensities.18 More time spent in LPA can be beneficial if it corresponds with less time in sedentary behavior. However, time spent in LPA does not provide benefits that come from participation in MVPA. Distinctions of the main components and dimensions of physical activity and sedentary behavior are summarized in Figure 3, based on an established model.
Figure 3: A Model of Sedentary and Physical Activity Behaviors
Third: The new definition of physical activity specifically references its contributions to improving dimensions of physical fitness. Physical fitness has generally been defined as “the ability to carry out daily tasks with vigor and alertness, without undue fatigue and with ample energy to enjoy leisure-time pursuits and meet unforeseen emergencies.”15 It can be subdivided into performance-related fitness and health-related fitness but the latter is more relevant for the purpose of this Guide because the majority of physical activity research is focused on health-related outcomes. Another caveat with this definition is that participation in physical activity may not necessarily lead to predictable or measurable improvements in physical fitness. Improvements are influenced by baseline fitness, genetic predispositions and a number of other factors, so the emphasis should be placed on the potential of physical activity to improve dimensions of physical fitness.
Body composition is considered to be a dimension of health-related fitness and is obviously of particular relevance for research targeting childhood obesity. Based on the description above, physical activity has important implications for maintaining or improving body composition and can induce positive changes in body fat content and distribution.
Research and public health guidelines have distinguished physical activity and sedentary behavior as independent behavioral constructs and they also may have independent effects on health, although this is less established in youth.19 No universally agreed-upon consensus has yet been achieved on defining sedentary behavior for both children and adults, though concerted efforts have been made for adults. For instance, researchers in the Sedentary Behavior Research Network have come to agreement that sedentary behavior should be defined as “any waking behavior characterized by an energy expenditure ≤1.5 adult-METs while in a sitting or reclining posture.”20 The threshold of 1.5 adult-METs has been generally considered a cutpoint for identifying sedentary behavior in adults. However, different assumptions must be considered for children. Recommendations for addressing this issue have been included in Section 10.
Physical Activity and Sedentary Behavior Guidelines
Previous activity guidelines emphasized the total amount of MVPA that should be performed, and separate targets were provided for MPA and VPA. The current U.S. Physical Activity Guidelines14 provide a more flexible model for tracking physical activity levels by focusing on the total volume of physical activity performed and the construct of “MET-Minutes.” Recommendations call for individuals to obtain 500 MET-Minutes a week. However, the guidelines also emphasize the need for relative guidelines that take into account a person’s individual level of fitness. Individuals are encouraged to perform 150 minutes of moderate-intensity physical activity a week but it can be accumulated in different ways. Consistent with the MET-Minute approach, vigorous minutes are multiplied by two to reflect the higher MET costs of VPA vs. MPA (6 METs vs. 3 METs) thereby allowing a person to meet the guideline with a combination of MPA and VPA. Because individuals vary in fitness level, it is important to acknowledge that MPA and VPA may be perceived very differently in the population. The distinction between relative and absolute intensities has important implications for the different physical activity measures. For example, report-based measures capture the perceptions of physical activity while monitor-based methods capture the movement that takes place. A fit person may report performing very little physical activity but the monitor may record considerable amounts. In contrast, an unfit person may have very little absolute movement in a day but it may be moderate in intensity. The examples are generally explained in the context of adult behavior but the same implications hold for quantifying youth physical activity, as youth who are unfit and have overweight may perceive their activity as of moderate or high intensity even though the monitor may record little absolute movement. These are simple examples, but the point is that the frequently observed discrepancies between measures may not be solely due to bias or recall problems, but rather to inherent differences in reported and measured data or how physical activity intensity is expressed. Thus, monitor-based and report-based measures capture different aspects of the same underlying construct of physical activity.
Guidelines for sedentary behavior have been harder to establish and are less consistently endorsed. Sedentary behavior is considered a construct that is independent of physical activity and that can carry different health implications. For this reason, the Canadian Society for Exercise Physiology in collaboration with the Healthy Active Living and Obesity Research Group developed the Canadian Sedentary Behavior Guidelines for Children and Youth. These were the first guidelines to specifically address recommendations for sedentary behavior in order to improve and maintain health. The guidelines suggest that children and youth should limit recreational screen time to a maximum of two hours per day and reinforce that lower amounts of screen time can offer additional health benefits.21 Other national and international organizations, such as the Australian Department of Health, also have developed specific guidelines for children and youth while reinforcing the importance of avoiding long continuous periods of sitting time. The American Academy of Pediatrics (AAP) recommends that young children limit the amount of screen time per day to no more than one hour22 and encourages pediatricians to work with children and families to promote a lifestyle with reduced sedentary behavior.23 The focus in this Guide is on assessments of physical activity; additional details about issues with assessment of sedentary behavior are available in Section 9.
Uniqueness of Assessment in Children and Adolescents
Assessing physical activity is challenging for all populations but it is particularly difficult in children and adolescents. Children have unique behavioral patterns of physical activity, unique perceptions and cognitions related to physical activity, and distinct physiological and maturational responses and adaptations to physical activity. Most foundational work on assessing physical activity and energy expenditure has been derived in adults and the simple assumption has been that these also hold in youth. However it is clear that children are not just “little adults,”24 so special considerations are needed to evaluate and study individual physical activity behavior in this segment of the population. Three specific considerations that must be taken into account for youth assessments are outlined below:
Children are known to engage in more sporadic and intermittent activity than adults, and this has important implications when trying to capture and assess physical activity with either report-based measures or monitor-based measures.25 The patterns of physical activity also vary across childhood and throughout later stages of adolescence. For example, preschoolers go through phases of motor skill acquisition and refinement and demonstrate less refined and less efficient movement patterns than do older age groups. Elementary school children (i.e., ages 6 to 11 years) have increasingly efficient movements but highly sporadic and intermittent physical activity patterns due to the random nature of play. The transition to adolescence (i.e., ages 12 to 18 years) is typically characterized by drops in physical activity levels and a greater contribution of team sports toward total physical activity accumulated during the day. Youth do not commonly exhibit adult patterns of continuous physical activity despite the emergence of maturing physical and behavioral attributes.
The variability in movement patterns from childhood to adolescence imposes unique measurement constraints for both report-based and monitor-based measures. Preschoolers are unable to recall activity and standard energy expenditure conversions often do not account for lack of movement efficiency. Tools capturing reported physical activity and sedentary behavior are increasingly useful as youth move from elementary school to middle school (i.e., pre-adolescents). However, it is still inherently challenging to capture the sporadic movements and “play” that characterize youth activity. Activity patterns become more predictable in high school youth (i.e., adolescence) as participation in structured physical activity becomes more common. Structured physical activity is more easily reported on recall instruments and is also easier to detect and quantify with monitors so assessments become slightly easier. Nevertheless, youth participation in team sports and random forms of play are still more difficult to capture than the more structured activities common in adulthood (e.g., jogging).
Another independent challenge is capturing the representative nature of their behavior (i.e., assessment of habitual physical activity and sedentary behavior patterns). Monitor-based methods may assess behavior over multiple days or full weeks while report-based methods typically involve recall over time (e.g., previous seven days) or estimates of “typical” behavior. Attention should be given to determining the time frame needed to obtain reliable indicators of actual behavior because it has important implications for research with youth. In addition to variability in overall behavior, it is important to consider inherent variability within a day (e.g., school-based physical activity vs. home-based physical activity), across days (e.g., days with physical education vs. days with no physical education), between days (e.g., school-day vs. non-school day), and across seasons (e.g., winter vs. summer activity patterns). Therefore, it is important to consider the definition of time frame as the period of time of interest and account for this variability when characterizing the physical activity behavior even though in most scenarios, if not all, the typical behavior is of most interest. Details on how to determine variability (i.e., reliability) of physical activity behavior are provided in Section 4.
Perceptions and Cognitions
Children’s cognitions and perceptions (e.g., knowledge of physical activity) also must be considered when using report-based tools. When using these measures, ambiguous terms like “physical activity” and “moderate intensity” can generate confusion as children display a limited understanding of the concept of physical activity and have difficulties reporting the intensity of the activities in which they engage. These challenges become clear when children are asked to indicate how many bouts of moderate or vigorous physical activity they performed in the previous day or past week. A more prominent concern is related to the limited ability of children to provide details of past physical activity events with retrospective recall instruments (e.g., previous week, previous month). The recall of physical activity requires complex processes that can lead to inaccurate reports of this behavior in all populations, and particularly in youth.26 The accuracy of reporting is highly dependent on the appropriate use and selection of episodic memories, which are associated with the child’s capacity to remember a specific event within a particular place and moment in time. The memory of the episode is like re-experiencing it (e.g., when describing events such as school graduation, a person can clearly remember the setting and even feel the same sensations or emotions of that day). Episodic memory refers to experiences of daily living, such as eating breakfast or exercising, and can be replaced by generic memories (i.e., memories of general events or patterns of events) that are used when individual memories or episodic memories are not available. For example, short-term or specific physical activity recall questionnaires (e.g., previous day, previous week, number of exercise bouts) are examples of instruments that refer to episodic memories.27 This can be problematic considering that the “natural” intermittent patterns in youth behavior makes these events quite common and therefore harder to recall or report with sufficient level of detail.
Physiological and Metabolic Responses
Another distinction with assessing physical activity and sedentary behavior in youth is that standard physiologic adaptations and relationships do not always hold when applied to youth. Perhaps the most critical distinction is the difference in metabolic cost of physical activity as a result of aging or growth. The amount of activity performed based on absolute intensity (e.g., use of MET values) assumes a standardized resting state of 3.5 ml/kg/min, a value established based on adult values of resting energy expenditure. Resting energy expenditure is primarily determined by body composition and more particularly by muscle mass, but other important predictors include age, sex, and body fat. These factors lead to error when using the standardized value of 3.5 ml/kg/min in adults but more significant errors (and systematic bias) when applied in youth populations.24, 29-32 The error is introduced due to known differences in resting energy expenditure for youth. For example, the resting energy expenditure in a 13-year-old child can be approximated as 4.2 ml/kg/min or 1.2 adult METs.33 If actual child resting energy expenditure values are used, light intensity would be described as activities eliciting up to 6.3 ml/kg/min (i.e., 1.5 times above their actual resting state) rather than up to 5.3 ml/kg/min (i.e., 1.5 METs x adult resting energy expenditure of 3.5 ml/kg/min). Failure to consider this difference leads to systematic over-estimation of children’s physical activity intensity and a misclassification of performed activities.b Error is further compounded due to additional variability associated with differences in lean body mass in children classified as normal weight and children classified as overweight or obese.c
It is important to note that the concepts of METs were not intended to take into account inter-individual differences or effects of different body composition and fitness levels.34 Detailed coverage of error due to estimation of METs is beyond the scope of the Guide but readers should consider the implications of these issues when processing and interpreting physical activity data. New methods to refine and standardize MET values for coding youth physical activity behaviors are described in the NCCOR Youth Compendium of Physical Activities website. Recommendations for standardizing youth MET outcomes are provided in Section 10.