3
Long-Term Trends Affecting Physical Activity Levels

The previous chapter revealed that the majority of the U.S. population is not meeting recommended guidelines for physical activity and that a sizeable fraction characterizes itself as completely inactive or sedentary. This chapter takes a longer view to determine whether there is evidence of a growing problem, and, as the available data permit, traces trends in technology introduction and other social and economic changes over the past 50 years or longer that may help explain current inadequate levels of physical activity.

ANALYSIS APPROACH

An ideal starting point in attempting to sort out the complex links between physical activity and the built environment would be to examine long-term trends and data related to physical activity, travel behavior, and urban form. Researchers are immediately confronted, however, with the lack of direct measures and longitudinal data even on changes in physical activity levels—the primary variable of interest. Until 2001, for example, major public health surveys tracked data on leisure-time physical activity only, and reliable data were not collected until the 1980s. Trend data on physical activity at home, at work, and in transport are unavailable (see Figure 1-2 in Chapter 1, which defines the four types of physical activity of interest). Thus, it is not possible to track directly how total physical activity levels have changed over time.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 3 Long-Term Trends Affecting Physical Activity Levels The previous chapter revealed that the majority of the U.S. population is not meeting recommended guidelines for physical activity and that a sizeable fraction characterizes itself as completely inactive or sedentary. This chapter takes a longer view to determine whether there is evidence of a growing problem, and, as the available data permit, traces trends in technology introduction and other social and economic changes over the past 50 years or longer that may help explain current inadequate levels of physical activity. ANALYSIS APPROACH An ideal starting point in attempting to sort out the complex links between physical activity and the built environment would be to examine long-term trends and data related to physical activity, travel behavior, and urban form. Researchers are immediately confronted, however, with the lack of direct measures and longitudinal data even on changes in physical activity levels—the primary variable of interest. Until 2001, for example, major public health surveys tracked data on leisure-time physical activity only, and reliable data were not collected until the 1980s. Trend data on physical activity at home, at work, and in transport are unavailable (see Figure 1-2 in Chapter 1, which defines the four types of physical activity of interest). Thus, it is not possible to track directly how total physical activity levels have changed over time.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 Faced with this challenge, the committee commissioned a paper (Brownson and Boehmer 2004) to examine historical trends and societal changes affecting physical activity from which meaningful associations and shifts in behavior could be inferred over time. For example, quantitative trend data are not available on physical activity in the workplace. However, labor force data that enable occupations to be classified by activity level can be used to trace occupational changes from 1950 to 2000. From these data, one can draw inferences, at least at a gross level, about changes in physical activity levels in the workplace. Another line of inquiry is to examine time use. Time is a scarce and constrained commodity (i.e., a day has 24 hours regardless of other changes). How individuals allocate their time among a range of activities provides useful insights about their opportunities for and propensity to engage in physical activity. Data on time use are available from analyses of detailed diaries dating back to 1965 and conducted every decade since (Robinson and Godbey 1999).1 These analyses enable direct observation of changes in physical activity levels over time, such as time spent on recreational activities (e.g., active sports, walking, cycling, other fitness activities). The analyses also document changes in time use and time availability with more indirect implications for physical activity. For example, time spent on housework has declined steadily since 1965—the result of technological improvements in the home (e.g., availability of prepared foods) and increased participation of women in the workforce (Cutler et al. 2003). These data suggest a loss in physical activity for some women due to a reduction in household chores, such as housecleaning. However, the change has also freed up time that, in theory, could be used for exercise or the pursuit of other leisure-time physical activities. 1 The 1965 and 1975 surveys were conducted by the Institute for Social Research at the University of Michigan, and the 1985 and 1995 surveys by the Survey Research Center at the University of Maryland. The key methodological features of the time diary analyses are summarized by Robinson and Godbey (1999, Table 32).

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 TRENDS IN LEISURE-TIME PHYSICAL ACTIVITY Reliable trend data on leisure-time physical activity levels for U.S. adults and adolescents have been collected since 1990.2 Data from the Behavioral Risk Factor Surveillance Survey (BRFSS) show a slight gain in meeting recommended levels of physical activity and a complementary decline in reported physical inactivity for U.S. adults from 1990 to 2000 (see Figure 3-1). In 1990, approximately 24 percent of adults met recommended physical activity levels and in 2000, about 26 percent—a compound average annual growth rate of 0.75 percent. For the same period, nearly 31 percent of adults reported they were inactive in 1990; that is, they did not engage in any leisure-time physical activity. By 2000, that figure had fallen to nearly 28 percent—a compound average annual rate of decline of 1.06 percent. The BRFSS data show improvements in physical activity levels for both men and women from 1990 to 2000 (Brownson and Boehmer 2004). However, analysis of the data by educational level and race reveals diverging trends. For example, those with less than 12 years of education showed a small but persistent decline in meeting recommended physical activity guidelines compared with those with a college education or at least some college. Non-Hispanic whites and blacks showed modest gains in meeting the guidelines, but Hispanic adults registered a slight decline.3 Trend data from the National Health Interview Survey (NHIS) for the period 1985 to 1998 show results similar to those of the BRFSS, that is, a slight improvement in the percentage of adults meeting recommended levels of physical activity. However, the NHIS data show essentially stable rates of inactive behavior (see Figure 3-2). 2 This section draws heavily on the paper by Brownson and Boehmer (2004) commissioned for this study. Although earlier data are available from the BRFSS, only even-year data for 1990–2000 were used because the surveys in these years sampled at least 43 states and the District of Columbia and are considered to be the most reliable (Brownson and Boehmer 2004). More recent data from the 2001 survey were not used because of the changes made in the questionnaire to obtain a more complete picture of physical activity levels (see Chapter 2). 3 These data are shown graphically in Figures 2 through 4 of the Brownson and Boehmer paper.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 FIGURE 3-1 Percentage of U.S. adult population meeting recommended physical activity levels or reported as inactive. (SOURCE: BRFSS 1990–2000.) FIGURE 3-2 Percentage of U.S. adult population meeting recommended physical activity levels or reporting no leisure-time physical activity. (SOURCE: NHIS 1985–1998.)

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 As noted, time-use diaries provide a longer-term perspective on changes in leisure-time activities. In 1995, survey respondents reported spending nearly 50 minutes a day in active sports, outdoor activities, walking, cycling, and other exercise for recreation—an increase of about 20 minutes a day since the 1965 survey (Cutler et al. 2003). Trend data on physical activity levels among youth are also available. The Youth Risk Behavior Survey System (YRBSS) data for 1991–2001 show that rates of vigorous activity for high school students (i.e., vigorous physical activity for 20 minutes or more at least three times a week) remained constant over the decade of the 1990s (see Figure 3-3). The percentage of students attending physical education classes daily—an indicator of physical activity levels—declined sharply during the first half of the decade, but increased gradually thereafter (Figure 3-3). FIGURE 3-3 Physical activity indicators in youths, grades 9 through 12. (SOURCE: YRBSS 1991–2001.)

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 No systematic trend data are available on changes in children’s physical activity levels. However, there has been a decline in walking and cycling to school (EPA 2003), and there is some evidence that children spend less time in play outdoors (IOM 2004). In sum, these data show that U.S. adults made modest gains in the pursuit of leisure-time physical activity over the past decade, while physical activity levels among youths appear to have remained unchanged. To obtain a more complete picture of changes in total physical activity levels, however, it is necessary to examine broader structural changes in the economy and society. TRENDS IN OTHER TYPES OF PHYSICAL ACTIVITY The twentieth century can be characterized as the century of technological change (Brownson and Boehmer 2004). The growth of white-collar jobs in the workplace, the introduction of labor-saving devices in the home, and the widespread use of the automobile as the primary form of transport have resulted in a pervasive reduction in the physical demands of daily life. Table 3-1 provides a time line of many technological innovations and supporting systems linked to reduced daily energy expenditure and increased opportunities for leisure-time sedentary activities (e.g., watching television, using the computer). The time line covers a longer period than most of the other trend data presented in this chapter, which are focused on the latter half of the twentieth century. Employment and Occupational Changes Between 1950 and 2000, the surge of women into the workforce, the continued decline in agricultural employment and manufacturing jobs, and other technological and social changes conducive to the growth of white-collar jobs brought about profound changes in physical activity levels in the workplace. The U.S. civilian labor force more than doubled from about 62 million in 1950 to about 143 million in 2000 (Brownson and Boehmer 2004). Participation by women increased by a factor of 3.6 compared with a factor of 1.7

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 TABLE 3-1 Twentieth-Century Technological Innovations and Supporting Systems Time Interval Work Home Production/Food Transport and Land Use Communications 1900–1925   1901: vacuum cleaner invented 1923: frozen food invented circa 1925: first electric washer, automatic washer, and automatic dryer 1900: modern escalator invented 1903: Wright Brothers invent the first engined airplane 1904: invention of the tractor 1906: first Mack trucks built 1908: Henry Ford improves the assembly line, and the first Model T is sold 1916: first Federal-Aid Road Act 1923: traffic signal invented 1916: first radio tuners that receive different stations invented 1923: television cathode-ray tube invented 1926–1950   1946: microwave oven invented 1949: cake mix invented 1930: jet engine invented 1940: first freeway in California from Pasadena to Los Angeles 1949: Levittown 1927: first successful talking motion picture 1939: first scheduled television broadcasts 1949: network television starts in the United States 1951–1975 1950: first automatic elevators 1958: photocopier invented 1962: introduction of first industrial robot 1972: word processor invented 1954: firstMcDonald’s; first TV dinner introduced 1964: permanent press fabric invented 1971: food processor invented 1952: first jet airliner forcommercial passenger service 1956: Federal-Aid Highway Act and beginning of the Interstate highway system 1963: first people mover introduced in the United States 1951: computers first sold commercially 1955: firstwireless TV remote invented 1958: integrated circuit invented 1959: microchip invented 1968: first computer with integrated circuits 1971: microprocessor invented; video-cassette recorder invented 1976–2000       1976: Apple home computer invented 1981: first IBM PC sold 1990: World Wide Web/Internet protocol and language created SOURCES: Twentieth-Century Inventions 1900–1999, History of Transportation, History of Communication (inventors.about.com, accessed June 6, 2004); Bruno 1993.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 for men, presumably with a large increase in lower-activity, white-collar jobs. Agricultural employment, typically a high-activity occupation, continued to decline from 12 percent of the labor force in 1950 to 2 percent in 2000. In nonagricultural establishments, the number of employees engaged in manufacturing fell sharply from 30 to 13 percent between 1950 and 2000, while those in the service sector—with a higher fraction of less physically demanding white-collar jobs—grew from about two-fifths to nearly four-fifths of civilian employment over the same period (BLS 2004a). Occupational data from the U.S. census categorized by activity level for this same time period show the results of these major structural changes.4 The share of the eligible labor force in low-activity occupations nearly doubled from 1950 to 2000, with the majority of that shift taking place in the first 20 years (see Figure 3-4). Today, approximately one-quarter of the eligible labor force, or 58.2 million people, is employed in low-activity occupations (Brownson and Boehmer 2004). The proportion of high-activity occupations remained relatively stable at 16 to 17 percent of the eligible labor force over this period, but then declined to about 14 percent from 1990 to 2000 (Figure 3-4). Today, about 31 million people are employed in high-activity occupations. It is not possible to characterize the occupations of the remaining 59 percent of the eligible labor force or to disaggregate the data by gender or other demographic variables. Nevertheless, the available data show major shifts in the tails of the distribution, which suggest a generally downward trend in physical activity levels in the workplace. In 1950, approximately 30 percent more of the labor force was engaged in high-activity than in low-activity occupations. By 2000, roughly twice as many persons were employed in low-activity than in high-activity occupations (Brownson and Boehmer 2004). 4 Brownson and Boehmer took occupational data from the U.S. census that had been recoded by researchers at the University of Minnesota Population Center to enhance compatibility in job classifications across years, and categorized the data by activity level on the basis of occupational descriptions contained in the 1988–1994 National Health and Nutrition Examination Survey III database. For more detail, see Brownson and Boehmer 2004.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 FIGURE 3-4 Occupations classified by activity level, percent of eligible labor force at least 16 years old, 1950–2000. (SOURCE: Brownson and Boehmer 2004, Figure 8.) Changes in Household Activities The sharp increase in women in the labor force, along with the introduction of labor-saving technology improvements in the home, has resulted in major changes in the time and energy devoted to household production. These changes in turn have important implications for physical activity levels. Foremost among these changes is the decline in time spent on housework and other moderate-level activities in the home. Longitudinal data from time diaries show that for women, time spent on household activities, including housework (e.g., housecleaning, laundry, meal preparation and cleanup), shopping, and child care, fell by nearly one-third from 1965 to 1995, from about 40 to about 27 hours per week (Robinson and Godbey 1999). Although the trend for men is in the opposite direction, overall the data indicate a net reduction in time devoted to household and family care, with the decline in housework being the dominant explanatory factor (Robinson and Godbey 1999). Thus, physical

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 activity associated with housework is on the decline, at least for women. What is less clear is how women, and to a lesser extent men, are using the time thus made available, a topic discussed in a subsequent section. Other changes in household structure (e.g., increasing numbers of single-person households, activity patterns and sharing of dual-worker households) are also likely to affect the time allocated to physical activity. Changes in Travel Behavior Personal transport in the twentieth century has been dominated by the introduction and growth of automobile travel. In 2001, respondents to the household interview for the National Household Travel Survey (NHTS) reported that, for the first time, the number of personal vehicles per household (1.9) exceeded the mean number of reported drivers per household (1.8) (BTS 2003). In 1969, there were 1.2 reported personal vehicles per household and 1.6 reported licensed drivers per household (Hu and Young 1999). According to the U.S. census, the proportion of households owning more than one vehicle in 2000 was more than double that reported in 1960, a reflection of both the increased disposable personal income and the preferences of the U.S. population (Brownson and Boehmer 2004). Not surprisingly, increased vehicle ownership and improvements in highway infrastructure, among other factors, have been associated with a sharp increase in personal travel, although the dominant direction of causality is not clear. The 2001 NHTS reported about 4 trillion person miles of travel, an average of about 14, 500 miles per person annually (BTS 2003). In 1969, 1.4 trillion person miles of travel was reported, for an annual per person average of about 7,100 (DOT 2001). The vast majority of trips are made by passenger vehicle, and this has been true for decades. In 1995, respondents to the Nationwide Personal Transportation Survey (NPTS)—the precursor to the current NHTS—reported making approximately 87 percent of daily trips for all purposes in a personal vehicle; in 1977, the equiv-

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 FIGURE 3-5 Percentage of trips by transport mode for U.S. daily travel, all trip purposes, 1977–1995. “Other” includes primarily school bus trips, as well as trips by taxicab, ferry, airplane, and helicopter. (SOURCE: Pucher and Renne 2003, 51.) alent number was 84 percent (Pucher and Renne 2003) (see Figure 3-5).5 The journey-to-work data from the U.S. census, which provide comparable data for a longer period, show increasing reliance on the automobile for commutes. In 1960, roughly two-thirds of such trips were made by car; by 2000, this share had grown to more than four-fifths (Pucher and Renne 2003) (see Figure 3-6). For all trips, the average amount of time spent daily in driving reported by all drivers has increased steadily in recent years—in part because of increased travel and in part because of greater road congestion. Comparable data for 1990–2001 alone show a growth in 5 Data from the 1969 survey were not included because walking and bicycle trips were not sampled, so the shares of motorized travel were artificially inflated. Data from the 2001 survey were not included because of a change in sampling methods that captures previously unreported walking trips (Pucher and Renne 2003).

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 Jobs have followed population to the suburbs. In 1950, about 70 percent of jobs were located in central cities; by 1990, that figure had fallen to 45 percent (Mieszkowski and Mills 1993). Furthermore, since World War II, employment—and, to a lesser extent, population—has grown more rapidly in small and less dense MSAs.11 This trend is referred to as deconcentration and has been attributed primarily to the costs of congestion—both higher living costs for households and higher production costs for firms (Carlino 2000). The result has been a more uniform spatial distribution of employment and population both within and across MSAs, although the largest and densest MSAs still account for the highest share of total population and employment (Carlino 2000). Metropolitan areas can also be characterized by spatial clustering in central cities with respect to both race and income (Berube and Tiffany 2004; NRC 1999). Minority and poor populations live disproportionately in central cities rather than in suburbs, a situation reflecting racial as well as economic segregation (NRC 1999).12 The concentration of the poor in the ghettos and barrios of central cities magnifies the social ills that accompany poverty and has exacerbated the flight of middle- and higher-income populations to the suburbs, further magnifying the concentration effect (Jargowsky 2003). High-poverty neighborhoods typically exhibit a cycle of disinvestment and decay—gradually declining investments in housing, commerce, and infrastructure; reductions in public services (e.g., garbage pickup, bus service); loss of established institutions (e.g., banks and supermarkets); and loss of population. Between 1970 and 1990, both the number and share of people living in high-poverty neighborhoods (i.e., neighborhoods where the poverty rate is 40 percent or higher) rose sharply in many MSAs. With the exception of the Hispanic population, however, the incidence of those living in high-poverty neighborhoods declined by nearly 11 Density is measured as employment or population per square mile. 12 The residential racial segregation of blacks is not simply a by-product of economic segregation. Massey and Denton (1993) found that high-income blacks live in areas nearly as segregated as those populated by low-income blacks, while the segregation of Hispanics and Asians falls steadily as income rises.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 one-quarter during the 1990s, as did the concentration of poverty. Blacks and Native Americans showed the largest declines on both measures in central cities and rural areas, respectively. Nevertheless, in 2000, blacks remained the single largest group living in high-poverty neighborhoods, and both blacks and Native Americans exhibited the highest concentrated poverty rates (Jargowsky 2003). Spatial concentration by income and race has been a constant feature of the built environment, but the location of these groups has shifted over time. After World War II, policies of urban renewal, central city revitalization, and gentrification resulted in the displacement of poor populations mainly within central cities, but often from the central core. This process of dispersion has continued, most recently with the movement of many minority groups to the older suburbs (U.S. Bureau of the Census 2002). In fact, the inner-ring suburbs were the only geographic areas that did not show a decline in the number of high-poverty neighborhoods between 1990 and 2000, and many experienced increases in poverty over the decade (Jargowsky 2003). What do these trends imply for travel, particularly by non-motorized modes? First, geographic characterization of the spatial dimensions of the built environment according to central cities, suburbs, and nonmetropolitan areas falls short of capturing the complexity of urban settings (e.g., ghetto neighborhoods in inner cities, inner suburbs, “edge cities,”13 exurban areas) and the ways in which these differences may affect residents’ propensity to be physically active. For example, the concentration of development in edge cities may be sufficiently compact to support public transit and encourage walking and cycling to some destinations. In contrast, large residential suburban developments without sidewalks or bicycle trails and with cul-de-sac street layouts may make driving the only reasonable alternative for most trips. Second, suburbanization of the population should decrease the accessibility, that is, the proximity and convenience, of many 13 The term “edge cities,” coined by Washington Post journalist and author Joel Garreau in 1991, refers to suburban cities, typically located near major freeway intersections.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 destinations, thereby increasing the reliance on more time-saving automobile travel for many trip purposes. Data from the 2001 American Housing Survey suggest that a sizeable fraction of the U.S. population still lives in settings with destinations that could be reached by nonmotorized modes. For example, nearly two-thirds of survey respondents reported having satisfactory neighborhood shopping within 1 mile of their home. Fifty-five percent reported having access to public transit, and among U.S. residents with children ?13 years old, nearly 57 percent had a public elementary school within 1 mile of their residence. However, when the data are analyzed by geographic characteristics—central cities and suburban areas within MSAs and areas outside of MSAs—more densely populated central cities exhibit higher levels of access, which offer their residents greater opportunities for non-motorized travel (see Table 3-3). Unfortunately, these data are available only for the 1997, 1999, and 2001 surveys, which makes any meaningful trend analysis impossible. Furthermore, the level of detail is insufficient to indicate the characteristics of particular locations that might encourage walking or cycling or taking transit to accessible destinations. TABLE 3-3 Selected Access Measures for Neighborhood-Occupied Housing Units by Geographic Area Selected Access Measure In MSAs (%) Outside MSAs (%) Central Cities Suburbs Housing units with public elementary school <1 milea 72 54 40 Housing units with public transportationb 82 52 23 Housing units with shopping <1 mileb 77 62 41 NOTE: MSA = metropolitan statistical area. a This measure is based on the number of households with children aged 0 to 13—9.2 million in central cities, 16.6 million in suburbs, and 5.6 million outside MSAs. b This measure is based on the total number of occupied housing units—31.7 million in central cities, 53.6 million in suburbs, and 20.9 million outside MSAs. SOURCE: American Housing Survey, 2001: Neighborhood-Occupied Units, Table 2-8, pp. 58–63.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 Finally, the geographic concentration of the poor in central cities generates a host of social ills that accompany poverty—drug trafficking, violent crime, economic (poor access to suburban jobs) and social isolation, limited provision of public services, and poorly maintained infrastructure—that are likely to discourage poor populations from engaging in physical activity except for necessary trips. The effect on physical activity levels of the recent move of poor and minority populations to the inner suburbs is likely to be mixed. The inner suburbs of older cities are apt to look much like their downtowns, with sidewalks and transit service. This may not be true, however, in newer cities, where the inner suburbs may offer less in the way of transit services and physical facilities (e.g., sidewalks). In both cases, crime and public safety are likely to be salient concerns. CHANGES IN TIME USE AND SEDENTARY ACTIVITIES A comparison of time use in 1995 and 1965 that combines the results for women and men (Cutler et al. 2003) reveals some gains in free time due to a decline in housework (discussed previously) and, to a lesser extent, declines in eating and personal care time, which could be used in more physically active endeavors (see Figure 3-8). Part of the freed-up time was in fact used for increased recreation—active sports, outdoor activities, walking, hiking, and other exercise. The majority, however, was spent on more television watching and additional hours of sleep (Figure 3-8).14 Sleeping continues to claim the largest share of available daily time—about one-third on average. Television watching accounts for about another 10 percent of available daily time and has grown to be the dominant leisure-time activity. A recent analysis of one of the time-use diary surveys from the 1990s used the data on activity type to estimate daily energy 14 Additional daily hours spent sleeping, however, peaked in 1975 and have remained relatively constant over the next two decades.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 FIGURE 3-8 Time use, 1965–1995 (ages 18–64). (SOURCE: Cutler et al. 2003.) expenditure.15 The results show a picture of daily life in which sedentary and low-intensity activities predominate (Dong et al. 2004). Excluding sleeping, which accounts for nearly one-fifth of the overall energy expenditure of the population, the activities that account for 50 percent of waking-hour energy expenditure, in order of priority, are driving a car, office work, watching television or a movie, taking care of children, sitting, eating, and cleaning house (see Table 3-4). With the exception of taking care of children and cleaning house, these activities are of very light intensity. 15 Data from the National Human Activity Pattern Survey conducted in 1992–1994 were used to estimate and rank the energy expenditure for each activity. Survey respondents reported activities in their own words for a 24-hour period, including the location and duration of the activity. Activities were recoded into 255 categories, which were then assigned appropriate metabolic expenditure values using the Ainsworth compendium and update, respectively (Ainsworth et al. 1993; Ainsworth et al. 2000). A score was created for each activity by multiplying the duration and intensity for each individual and summing across individuals, and then each score was ranked by its contribution to total population energy expenditure. More detail on the calculation of energy expenditure is given by Dong et al. (2004).

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 TABLE 3-4 Ranking of Activities That Account for 50 Percent of Daily Energy Expenditure in the United States Rank Activity Description MET Percent of Total Score Cumulative Percentage (1) Sleeping, napping 0.9 (19.1) — 1 Driving car 2.3 10.9 10.9 2 Job: office work, typing 1.5 9.2 20.1 3 Watching TV/movie, home or theater 1.0 8.6 28.7 4 Taking care of child (feeding, bathing, dressing) 3.0 8.4 37.1 5 Activities performed while sitting quietly 1.3 5.8 42.9 6 Eating (sitting) 1.5 5.3 48.2 7 Cleaning house, general 3.0 3.9 52.1 NOTE: MET = metabolic equivalent (see Chapter 2 for a definition). SOURCE: Dong et al. 2004. Leisure-time, high-intensity activities account for less than 3 percent of total waking energy expenditure in this sample population. In August 2004, the Bureau of Labor Statistics released the results of the first American Time-Use Survey (ATUS). A monthly survey conducted by the U.S. Bureau of the Census, the ATUS will provide a consistent and continuous source of nationally representative daily time-use data that can readily be combined with demographic and employment data, as well as data on energy expenditure.16 On an “average day” in 2003, persons in the United States aged 15 and older reported that they slept about 8.6 hours, engaged in leisure and sports activities for 5.1 hours, worked for 3.7 hours, and spent 1.8 hours doing household activities.17 The remaining 4.8 hours was 16 The ATUS was administered to an outrotated panel of the Current Population Survey, thereby providing demographic and labor force information. Data collection began in January 2003, and the ATUS estimates for that year are based on interviews of about 21,000 individuals. The survey was administered to one member of a household (15 years or older), who provided information on activities lasting 5 minutes or longer in the preceding 24-hour period (BLS 2004b). 17 An average day encompasses both weekdays and weekends and is computed on the basis of all responses from a given population, including respondents who did not engage in a particular activity on their diary day. The activities cited are primary activities, that is, those identified by respondents as their main activity (BLS 2004b).

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 spent on such activities as eating and drinking, attending school, and shopping (BLS 2004b). These results confirm the findings of the earlier 1995 time-use survey regarding sedentary use of free time. According to the ATUS, the population at large, on average, has approximately 5 free hours available on an average day and spends approximately half of this time watching television. Only 18 minutes on average is spent on sports, exercise, and recreation (BLS 2004b). Time spent on transportation is not identified separately in the ATUS but is included with the appropriate activity. To estimate time spent on travel, particularly on active travel, that is, on walking, cycling, and accessing public transit, a detailed analysis of the 1995 NPTS was conducted by one of the committee members. The results show that, on average, adults (persons 18 years and older) spend 64 minutes per day traveling by all modes of transport. Of that time, an average of 3 minutes is spent on active travel. The committee recognizes that these results are likely to undercount active travel—detailed analysis of the 2001 NHTS and future surveys should provide better estimates of nonmotorized travel. Nevertheless, the results suggest that active travel represents a small fraction of the total time spent in transportation. REFERENCES Abbreviations BLS Bureau of Labor Statistics BTS Bureau of Transportation Statistics DOT U.S. Department of Transportation EPA U.S. Environmental Protection Agency IOM Institute of Medicine NRC National Research Council TRB Transportation Research Board Ainsworth, B. E., W. L. Haskell, A. S. Leon, D. R. Jacobs, Jr., H. J. Montoye, J. F. Sallis, and R. S. Paffenbarger, Jr. 1993. Compendium of Physical Activities: Classification of Energy Costs of Human Physical Activities. Medicine and Science in Sports and Exercise, Vol. 25, pp. 71–80.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 Ainsworth, B. E., W. L. Haskell, M. C. Whitt, M. L. Irwin, A. M. Swartz, S. J. Strath, W. L. O’Brien, D. R. Bassett, Jr., K. H. Schmitz, P. O. Emplaincourt, D. R. Jacobs, Jr., and A. S. Leon. 2000. Compendium of Physical Activities: An Update of Activity Codes and MET Intensities. Medicine and Science in Sports and Exercise, Vol. 32, pp. S498–S504. Berube, A., and T. Tiffany. 2004. The Shape of the Curve: Household Income Distributions in U.S. Cities, 1979–1999. Living Census Series, Metropolitan Policy Program, Brookings Institution, Aug. BLS. 2004a. Employment and Earnings. Table B-1: Employees on Non-Farm Payrolls by Industry Sector and Selected Industry Detail. data.bls.gov/servlet/SurveyOutputServlet. Accessed March 19, 2004. BLS. 2004b. Time-Use Survey—First Results Announced by BLS and Technical Note. News. U.S. Department of Labor, Washington, D.C., Sept. 14. Brownson, R. C., and T. K. Boehmer. 2004. Patterns and Trends in Physical Activity, Occupation, Transportation, Land Use, and Sedentary Behaviors. Department of Community Health and Prevention Research Center, School of Public Health, St. Louis University. Prepared for the Committee on Physical Activity, Health, Transportation, and Land Use, June 25. Bruno, L. C. 1993. On the Move: A Chronology of Advances in Transportation. Gale Research, Inc., Detroit, Mich. BTS. 2003. NHTS 2001 Highlights Report. BTS03-05. U.S. Department of Transportation, Washington, D.C. Carlino, G. A. 2000. From Centralization to Deconcentration: People and Jobs Spread Out. Business Review, Federal Reserve Bank of Philadelphia, Nov.–Dec., pp. 15–27. Cutler, D. M., E. L. Glaser, and J. M. Shapiro. 2003. Why Have Americans Become More Obese? Journal of Economic Perspectives, Vol. 17, No. 3, Summer, pp. 93–118. Dong, L., G. Block, and S. Mandel. 2004. Activities Contributing to Total Energy Expenditure in the United States: Results from the NHAPS Study. International Journal of Behavioral Nutrition and Physical Activity, Vol. 1, No. 4, Feb. 12. DOT. 2001. Summary Statistics on Demographic Characteristics and Total Travel 1969, 1977, 1983, 1990, and 1995 NPTS, and 2001 NHTS. nhts.ornl.gov/2001/html_files/trends_verb6.shtml. Accessed March 9, 2004. EPA. 2003. Travel and Environmental Implications of School Siting. EPA-231-R-03-004. Washington, D.C. Federal Register. 2000. Standards for Defining Metropolitan and Micropolitan Statistical Areas. Notice of Decision. Office of Management and Budget. Vol. 65, No. 249, Dec. 27, pp. 82227–82238. Hu, P. S., and J. R. Young. 1999. Summary of Travel Trends, 1995 Nationwide Personal Transportation Survey. Federal Highway Administration, U.S. Department of Transportation, Dec.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 IOM. 2004. Preventing Childhood Obesity: Health in the Balance. National Academies Press, Washington, D.C. Jargowsky, P. A. 2003. Stunning Progress, Hidden Problems: The Dramatic Decline of Concentrated Poverty in the 1990s. Center on Urban and Metropolitan Policy, Brookings Institution, May. Massey, D., and N. Denton. 1993. American Apartheid: Segregation and the Making of the Underclass. Harvard University Press, Cambridge, Mass. Mieszkowski, P., and E. S. Mills. 1993. The Causes of Metropolitan Suburbanization. Journal of Economic Perspectives, Vol. 7, No. 3, pp. 137–147. NRC. 1999. Governance and Opportunity in Metropolitan America. A. Altshuler, W. Morrill, H. Wolman, and F. Mitchell (eds.), National Academy Press, Washington, D.C. Pucher, J., and J. I. Renne. 2003. Socioeconomics of Urban Travel: Evidence from the 2001 NHTS. Transportation Quarterly, Vol. 57, No. 3, Summer, pp. 49–77. Robinson, J. P., and G. Godbey. 1999. Time for Life: The Surprising Ways Americans Use Their Time, 2nd ed. Pennsylvania State University Press, University Park. TRB. 2002. Special Report 269: The Relative Risks of School Travel: A National Perspective and Guidance for Local Community Risk Assessment. National Research Council, Washington, D.C. U.S. Bureau of the Census. 2002. Racial and Ethnic Residential Segregation in the United States: 1980–2000. U.S. Government Printing Office, Washington, D.C.

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 SUMMARY Contextual Factors Affecting Physical Activity Recent national surveys report that Americans walk, and to a lesser extent cycle, primarily for exercise and recreation. However, reported levels of both activities fall short of recommended daily guidelines (i.e., 30 minutes per day of moderate-level physical activity on 5 or more days per week), a result confirmed by the public health surveys reviewed in Chapter 2. The barriers to meeting adequate physical activity levels include personal reasons (disabilities and other health impairments), concerns for safety and security, and time constraints and environmental impediments (long distances between destinations, limited travel choices). From the perspective of environmental barriers, it is important to distinguish among different population groups and their geographic locations. Impediments to walking, cycling, and other forms of physical activity are likely to differ greatly among an inner-city neighborhood, a typical suburban development, and a remote rural community. Interventions to encourage greater physical activity should be tailored to reflect these differences, and the target populations should be segmented accordingly. It is also important to distinguish among different types of physical activity in addressing environmental barriers. Americans appear to be interested and engaged in walking and cycling for recreation and, to a lesser extent, for local shopping. Interventions should reinforce these behaviors and provide opportunities for those who want to be physically active. Moreover, while the convenience and mobility of the car for commuting and regional shopping trips are not easily matched by walking or cycling, census data indicate that

OCR for page 59
Does the Built Environment Influence Physical Activity? Examining the Evidence - Special Report 282 many Americans have convenient access to satisfactory neighborhood shopping, schools, and public transit, which provides numerous opportunities for using nonmotorized travel. Opportunities to modify the built environment to make it more conducive to physical activity are numerous, but the ease or difficulty of such changes depends on the intervention. For example, overturning long-standing zoning and land use ordinances to increase development density and mixed land uses is likely to face formidable barriers that cannot easily be overcome. Bringing investment back to inner-city neighborhoods and creating safe environments with desirable destinations conducive to walking are long-term processes. More flexible and targeted approaches—such as context-sensitive design, special overlay districts, traffic calming measures, and community policing—are more likely to win support and can be implemented more rapidly. Construction of new buildings and developments also offers promising opportunities for creating more activity-friendly environments. To design effective policies and interventions, however, will require a more complete understanding of how the built environment facilitates or constrains physical activity, a topic investigated in the following two chapters.