This is an excerpt from Physical Activity and Obesity - 2nd Edition by Claude Bouchard & Peter Katzmarzyk.
Methods to Assess Sedentarism
To understand the relationships illustrated in figure 3.1 and to examine the consequences of behavior compensation require careful, robust, context-appropriate assessments of sedentarism (and physical activity; see chapters 4 and 5). Incomplete, improper, or inappropriate assessment procedures will serve to mask or exaggerate true relationships.
Clues for assessment methods are drawn from sedentary behaviors. Whereas a piece of exercise equipment is a cue for a method of physical activity assessment (e.g., frequency, intensity, and duration of use), a piece of equipment that promotes sedentary behavior is a cue for a method of sedentarism assessment. For example, as already discussed, televisions, computers, and video game consoles promote extended sitting and idle behavior, so screen time (frequency × duration) is an appropriate indicator of sedentarism. Indeed, screen time (or some derivative—television time, computer time, video game time) has become a common measure of sedentariness (1, 2, 4, 7, 8). The 2008 Active Healthy Kids Canada Report Card on Physical Activity for Children and Youth (www.activehealthykids.ca) identified excessive screen time as a key opportunity to reverse the physical deactivation of our children. The Report Card indicates that Canadian children and youth aged 10 to 16 years are averaging 6 h of leisure-time screen time per day based on the 2005-2006 Health Behavior in School-aged Children Survey. Such data suggest that this form of sedentary behavior is both common and extensive. Pediatric groups have established screen time guidelines (9) to inform children and their parents about healthy screen time usage. These guidelines can be used to assess the proportion of the population meeting or exceeding the guidelines at a point in time, or to monitor trends over time (7, 8).
The same methodologies we use to assess physical activity can guide our assessment of sedentarism. Table 3.1 provides a series of potential assessment methods in comparison with traditional measures of physical activity (and inactivity). Methods are categorized as direct (objective), reported (subjective), and global assessments (for a discussion on the advantages and disadvantages of the various methodologies see chapter 8). The table contrasts the different contexts for the measurement of physical activity versus sedentary behavior in an effort to provoke the conceptualization of additional assessment methodologies in these two domains. Note how physical activity measures like minutes of physical activity, minutes above thresholds, active transportation distance, energy expenditure, and outdoor time contrast with sedentarism measures like minutes of sedentary behavior, minutes below thresholds, car time, screen time, chair (sitting) time, and indoor time.
Using pedometers to assess step counts is one method to assess sedentarism. Based on extensive literature reviews (10, 11), it has been recommended that a zone-based hierarchy of step counts be used to classify physical activity behaviors among adults and children. Using this classification system, adults are categorized as sedentary if they achieve <5000 steps per day (10, 11), a threshold supported by evidence that the likelihood of being obese is increased in those categorized as sedentary using this cut point (10). Suggested thresholds for children aged 6 to 12 years are <7000 steps per day for girls and <10,000 for boys, although it is acknowledged that additional prospective, longitudinal, criterion-referenced studies are required to refine these targets.
Reilly and colleagues recently reviewed the use of direct measurements of sedentary behaviors via accelerometers (12). They concluded that objective measures of sedentary behavior using accelerometers are consistent regardless of the measurement epoch selected or the age of the child participant. Esliger and Tremblay provide an example of a comprehensive assessment of sedentary behavior derived from minute-by-minute, week-long accelerometry measurements (13). This detailed profile includes assessments of weekday-specific, weekend day-specific, and average-day sleep and sedentary behavior, as well as when (time of day) and how (sporadic, short bouts, long bouts) the sedentary behavior was accumulated. Detailed profiles of physical activity patterns are increasingly common, but equivalent profiles for sedentary behavior are rare in the research literature.
Typically, many of the indicators in table 3.1 are used to assess population-level adherence to established guidelines for physical activity (e.g., prevalence meeting physical activity guidelines; see chapters 9 and 10); although far less common, the same assessment can be made for sedentarism (e.g., prevalence meeting screen time guidelines  ). It is worth noting that the proliferation of “screens” (television, computer, video game, cell phone, portable DVD player, etc.), diversification of uses (active video-based gaming), and simultaneous multiscreen usage has and will continue to complicate the measurement of screen time.
Further research is required to explore the validity of using surrogate inhibiting or enabling measures for the assessment of sedentarism. For example, information on park space, playground availability, ownership of exercise equipment, and gym membership is often used to assess or inform physical activity levels. In a similar manner, measures such as televisions or screens per household, Internet access in the home, cable or satellite television access, and automobiles per household member may serve as useful indices of sedentarism.
Examples of Sedentarism Assessment
Rosenberg and colleagues (5) used accelerometers to investigate the ability of the International Physical Activity Questionnaire (IPAQ) to assess sedentary behavior via sitting-time questions. Correlations between sitting time and accelerometer counts <100 (indicating minimal movement) were 0.33 for the long-form and 0.34 for the short-form questionnaires. The authors further observed no relationship between sitting time and likelihood of being classified as physically inactive. They concluded that sedentary behavior should be explicitly measured in population surveillance and research instead of being defined by a lack of physical activity.
Shields and Tremblay recently reported on the sedentary behaviors of Canadian adults using a nationally representative sample (7, 8). They examined television viewing and computer use separately. Twenty-nine percent of adults aged 20 years or older reported watching television 2 h or more per day, and 15% reported leisure-time computer usage of more than 10 h per week (8). The pattern of usage varied significantly by age group, with younger adults far more likely to accumulate screen time on the computer compared to older Canadians. The opposite was true of television viewing. Differences in sociodemographic characteristics were evident but varied between television and computer usage. For example, recent immigrants were less likely than people born in Canada to be frequent television viewers but more likely to be frequent computer users.
It is likely that not all sedentary pursuits are equal contributors to obesity. Shields and Tremblay recently explored this possibility (7). In a nationally representative sample of Canadian adults (n = 42,612), they observed that among men and women the odds of being obese increased with hours of weekly television viewing, and that this relationship was independent of leisure-time physical activity and diet. When the effects of age and other confounding variables were controlled, only a modest association was observed between frequent computer use and obesity. In contrast, reading time was not associated with obesity for either sex.