• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Physical activity levels and cardiovascular disease risk among U.S. adults: Comparison between self-reported and objectively measured physical activity

Dissertation
Author: Jared Michael Tucker

ii TABLE OF CONTENTS INTRODUCTION 1

Dissertation Organization...................................................................................................................3 References...........................................................................................................................................5 CHAPTER 1 - COMPARISON OF SELF-REPORT AND OBJECTIVE PHYSICAL ACTIVITY MEASURES IN THE NHANES: AN EXTENDED REVIEW OF THE LITERATURE 8

Physical Activity and Health..............................................................................................................8 Physical Activity Guidelines............................................................................................................10 Physical Activity Levels in U.S. Adults...........................................................................................12 Self-Reported Physical Activity Levels.......................................................................................12 Objective Physical Activity Levels..............................................................................................14 Physical Activity Surveillance Tools................................................................................................16 Self-Report Physical Activity.......................................................................................................16 Accelerometer Physical Activity..................................................................................................18 Accelerometer Data Processing....................................................................................................21 Physical Activity in the NHANES....................................................................................................25 Self-reported Physical Activity in the NHANES.........................................................................25 Accelerometer Physical Activity in NHANES.............................................................................26 Summary...........................................................................................................................................27 References.........................................................................................................................................28 CHAPTER 2 - COMPARISON BETWEEN SELF-REPORTED AND OBJECTIVELY MEASURED PHYSICAL ACTIVITY LEVELS IN US ADULTS: COMPLIANCE WITH THE PHYSICAL ACTIVITY GUIDELINES FOR AMERICANS 37

Introduction.......................................................................................................................................37 Methods............................................................................................................................................41 Study Population..........................................................................................................................41 Measures.......................................................................................................................................42 Data Processing............................................................................................................................45 Data Analysis...............................................................................................................................48 Results...............................................................................................................................................48 Self-report vs. Accelerometer Physical Activity..........................................................................49 Physical Activity Guidelines for Americans................................................................................51 Discussion.........................................................................................................................................53 Accelerometer Physical Activity Levels in the US......................................................................53 Self-report Physical Activity Levels in the US............................................................................55 References.........................................................................................................................................58 CHAPTER 3 - EFFECTS OF ACCELEROMETER COMPLIANCE PROCESSING ON PHYSICAL ACTIVITY LEVELS IN US ADULTS 70

Introduction.......................................................................................................................................70 Methods............................................................................................................................................74 Study Population..........................................................................................................................74 Measures.......................................................................................................................................75 Demographic & Anthropometric Data.........................................................................................75 Accelerometer Physical Activity..................................................................................................76

iii Data Processing............................................................................................................................77 Data Analyses...............................................................................................................................79 Results...............................................................................................................................................80 Discussion.........................................................................................................................................84 Compliance Outcomes.................................................................................................................85 Physical Activity Outcomes.........................................................................................................86 References.........................................................................................................................................91 CHAPTER 4 - PHYSICAL ACTIVITY, THE METABOLIC SYNDROME, AND 10- YEAR RISK OF CORONARY HEART DISEASE IN US ADULTS: COMPARISON BETWEEN SELF-REPORT AND ACCELEROMETRY 100

Introduction.....................................................................................................................................100 Methods..........................................................................................................................................102 Study Sample..............................................................................................................................102 Physical Activity Measures........................................................................................................104 Clinical Measures.......................................................................................................................106 Data Analysis.............................................................................................................................107 Results.............................................................................................................................................108 Discussion.......................................................................................................................................110 References.......................................................................................................................................116 SUMMARY 124

1

INTRODUCTION

Physical activity (PA) is one of the most important lifestyle behaviors people can perform to improve their health (1). The literature strongly supports an inverse, dose- response relationship between PA and several diseases and health problems including coronary heart disease, stroke, type 2 diabetes, osteoporosis, depression, some cancers, and all-cause mortality (2). Furthermore, many of these reduced risks have been demonstrated across ethnicities and age groups, including children and adolescents (3-9). Due to these numerous health benefits, substantial efforts have been made to promote PA in the U.S. population. Recently, the U.S. Department of Health and Human Services released the first set of official U.S. physical activity guidelines (Physical Activity Guidelines for Americans - PAGA) (10). The PAGA provide updated evidence regarding PA and health as well as specific recommendations for how much physical activity is needed by different segments of the population. Not surprisingly, conclusions from the PAGA advisory committee confirmed the importance of PA for individuals of all ages, and suggested that greater efforts were needed to encourage PA among Americans (11). Currently, a large discrepancy exists in the PA levels of US adults depending upon whether a subjective or objective measurement method is employed (12-13). In recognition of this disparity, the PAGA committee report emphasized the need for a better understanding of these measurement tools. Furthermore, the advisory committee identified a lack of knowledge regarding how objective PA measurements relate to a variety of health outcomes, and stated that such knowledge was “…needed before such measurements can be used to inform future physical activity recommendations and policy statements.” (11). Indeed, all PA research, including epidemiological, health-related, and

2

intervention-based, hinges upon the ability to measure field-based PA accurately. However, a lack of understanding currently exists regarding how subjective and objective PA relate to each other and how these differences affect the relationship between PA and health. Habitual PA levels are inherently difficult to measure because of the need to measure long-term patterns which are often labile in nature. Currently, a wide range of assessment tools exist for measuring habitual PA including self-report questionnaires, direct observation techniques, and objective tools such as pedometers, accelerometers, heart-rate monitors, and multi-sensor devices. Each of these measures has advantages and disadvantages. Therefore, choosing the best tool for a particular research purpose depends upon several study factors such as the design, setting, sample size, budget, and demographic characteristics of the participants. Generally, objective measures of PA are considered more valid than subjective measures, though the cost and inconvenience of these tools often excludes their use in large- scale studies. Historically, most PA epidemiology research has relied on self-report techniques because they are convenient, non-invasive, and relatively inexpensive. However, self-reported techniques are prone to measurement error from issues such as misinterpretation of questions, inability to recall, and social desirability (14-17). Though these issues cannot be eliminated, a greater understanding of the limitations of self-report tools would allow researchers to more accurately interpret resultant PA. Currently, the proportion of U.S. adults meeting the new PAGA has only been assessed in one research study (18), which relied on a self-report measure. To date, only one study has examined previous national guidelines for PA using an objective, accelerometry- based measure (13). In addition, no studies are currently available which have compared the

3

PA levels of Americans measured from both self-report and accelerometer. Doing so would improve current knowledge of how physically active Americans are in regards to the PAGA, and would help facilitate a greater understanding of the relationship between PA levels when measured by self-report and accelerometer techniques. Dissertation Organization The overall purpose of this dissertation was to evaluate physical activity assessment issues and assumptions in the nationally representative National Health and Nutrition Examination Survey (NHANES). The NHANES data provides nationally representative data using both self-report and accelerometers so a fundamental goal was to understand measurement agreement between subjective and objective measures of activity. The NHANES incorporates both self-report and accelerometer PA measures for each adult subject, making it an ideal source for obtaining direct comparisons between the two measures. Study one of the proposed research assessed the proportion of the US adult population that achieved the 2008 PAGA when measured with a self-report instrument and an accelerometry-based activity monitor. Though accelerometers provide an objective measure, there are inherent limitations that preclude them from being used as a criterion measure for PA. Specifically, accelerometers measure only locomotor movements, and cannot detect changes in grade or load. Furthermore, the accelerometer data requires a series of subjective processing techniques, which may introduce error into the measure. The magnitude of this potential for error has been demonstrated by the large differences in resultant PA that can be obtained from the same data (19-20). Though significant literature has been devoted to the calibration and validation of these devices, additional work is needed to clarify the effects of these

4

processing decisions and to help minimize potential error. Specifically, little has been done to establish the influence of wear-time compliance decisions on accelerometer outcomes. Therefore, the second study of the dissertation assessed the effects of accelerometer compliance processing on resulting physical activity levels in the US adult population, and determined the relationship between these PA outcomes and adiposity among US adults. By comparing the potential range in accelerometer PA to a known correlated variable, it was possible to provide insight into which accelerometer results were most highly correlated with the measure of interest, and potentially most representative of true PA levels. Ultimately, this study provided recommendations regarding accelerometer data processing, which if utilized, will improve the comparability of accelerometer study results in the future. As documented in the PAGA, the link between self-reported PA and health-related outcomes has been well established (11). However, the available literature regarding the link between objective PA and health is sparse. Furthermore, no studies have assessed the relationship between health risks and PA assessed both subjectively and objectively. In doing so, the relationship that these assessment tools have on health outcomes can be compared to help differentiate which technique more accurately quantifies PA. Hence, the purpose of the third study of this dissertation was to assess the relationship between meeting the PAGA, when measured by self-report and accelerometry, with coronary heart disease (CHD) risk, the metabolic syndrome, and its individual risk factors among US adults. A secondary purpose of this study was to examine whether the relationship between objectively measured PA and CHD risk was affected by whether the activity was accumulated in sustained bouts of 10-min or more. This research provides new and important knowledge regarding self-reported and accelerometry-based PA assessment techniques. In addition, this

5

is the first study to assess the relationship between objectively measured PA and CHD risk among a representative sample of US adults. Therefore, this study also provides valuable insight into the relationship between objective PA and important health factors, and provides evidence regarding the benefits that achieving the PAGA can provide for the cardiovascular health of Americans. References 1. United States Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, The President's Council on Physical Fitness and Sports. 1996.

2. Bauman AE. Updating the evidence that physical activity is good for health: an epidemiological review 2000-2003. J Sci Med Sport. 2004 Apr;7(1 Suppl):6-19.

3. Manini TM, Everhart JE, Patel KV, Schoeller DA, Colbert LH, Visser M, Tylavsky F, Bauer DC, Goodpaster BH, Harris TB. Daily activity energy expenditure and mortality among older adults 49. JAMA 2006 Jul 12;296(2):171-9.

4. Crespo CJ, Palmieri MR, Perdomo RP, Mcgee DL, Smit E, Sempos CT, Lee IM, Sorlie PD. The relationship of physical activity and body weight with all-cause mortality: results from the Puerto Rico Heart Health Program 12. Ann. Epidemiol. 2002 Nov;12(8):543-52.

5. Fujita K, Takahashi H, Miura C, Ohkubo T, Sato Y, Ugajin T, Kurashima K, Tsubono Y, Tsuji I, Fukao A, et al. Walking and mortality in Japan: the Miyagi Cohort Study 20. J. Epidemiol. 2004 Feb;14 Suppl 1:S26-S32.

6. Lam TH, Ho SY, Hedley AJ, Mak KH, Leung GM. Leisure time physical activity and mortality in Hong Kong: case-control study of all adult deaths in 1998 40. Ann.Epidemiol. 2004 Jul;14(6):391-8.

7. Andersen LB, Harro M, Sardinha LB, Froberg K, Ekelund U, Brage S, Anderssen SA. Physical activity and clustered cardiovascular risk in children: a cross-sectional study (The European Youth Heart Study). Lancet 2006 Jul 22;368(9532):299-304.

8.

Imperatore G, Cheng YJ, Williams DE, Fulton J, Gregg EW. Physical activity, cardiovascular fitness, and insulin sensitivity among U.S. adolescents: the National Health and Nutrition Examination Survey, 1999-2002. Diabetes Care 2006

6

Jul;29(7):1567-72. 9. Gidding SS, Barton BA, Dorgan JA, Kimm SY, Kwiterovich PO, Lasser NL, Robson AM, Stevens VJ, Van HL, Simons-Morton DG. Higher self-reported physical activity is associated with lower systolic blood pressure: the Dietary Intervention Study in Childhood (DISC). Pediatrics 2006 Dec;118(6):2388-93. 10. GovTrack.us. S. 2748--110th Congress (2008): Physical Activity Guidelines for Americans Act of 2008, GovTrack.us (database of federal legislation). Available at: http://www.govtrack.us/congress/bill.xpd?bill=s110-2748&tab=analysis. Accessed April 20, 2009. 11. US Department of Health and Human Services. Physical Activity Guidelines Advisory Committee final report. Available at: http://www.health.gov/PAGuidelines/Report/Default.aspx. Accessed April 20, 2009. 12. Prevalence of regular physical activity among adults--United States, 2001 and 2005. Centers for Disease Control and Prevention (CDC). MMWR Morb Mortal Wkly Rep. 2007 Nov 23;56(46):1209-12. 13. Troiano RP, Berrigan D, Dodd KW, Mâsse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008 Jan;40(1):181-8. 14. Sallis JF, Saelens BE. Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport 2000;71:1 – 14.

15. Warnecke RB, Johnson TP, Cha´vez N, Sudman S, O’Rourke DP, Lacey L, et al. improving question wording in surveys of culturally diverse populations. Ann Epidemiol 1997;7:334– 42. 16. Motl RW, McAuley E, DiStefano C. Is social desirability associated with self- reported physical activity? Prev Med. 2005 Jun;40(6):735-9.

17. Baranowski, T. Validity and reliability of self-report of physical activity: An information processing perspective. Research Quarterly. 1988;59:314-327.

18. Prevalence of self-reported physically active adults--United States, 2007. Centers for Disease Control and Prevention (CDC). MMWR Morb Mortal Wkly Rep. 2008 Dec 5;57(48):1297-300. 19. Guinhouya CB, Hubert H, Soubrier S, Vilhelm C, Lemdani M, Durocher A. Moderate-to-vigorous physical activity among children: discrepancies in accelerometry-based cut-off points. Obesity (Silver Spring). 2006 May;14(5):774-7

7

20. Masse, LC, Fuemmeler BF, Anderson CB, et al. Accelerometer data reduction: a comparison of four reduction algorithms on select outcome variables. Med. Sci. Sports Exerc. 37:S544–S554, 2005.

8

CHAPTER 1

COMPARISON OF SELF-REPORT AND OBJECTIVE PHYSICAL ACTIVITY MEASURES IN THE NHANES: AN EXTENDED REVIEW OF THE LITERATURE

This research will examine physical activity assessment issues in the NHANES surveillance system. The literature review will provide background information and research summaries to guide the development of this work. The first section will briefly review research on the health benefits of physical activity and provide a summary of the recently established physical activity guidelines. The guidelines serve as the basis for public health surveillance so the second section will review current physical activity surveillance tools and provide an overview of the NHANES project. The final section will review physical activity assessment issues and describe the advantages and disadvantages of the different measurement tools. Special emphasis will be placed on accelerometer and self-report instruments since these will be the focus of the proposed research. The separate sections on these two techniques will review the available literature and provide an empirical rationale for the proposed comparisons and studies to be conducted in this research. Physical Activity and Health Being physically active is one of the most important ways people can improve their health. According to the 2008 Physical Activity Guidelines for Americans, there is strong evidence to support an inverse association between physical activity (PA) and several diseases and health problems including coronary heart disease, stroke, type 2 diabetes, osteoporosis, depression, some cancers, and all-cause mortality (1). Of particular interest is the inverse association between PA and cardiovascular disease (CVD), the number one killer in the US (2). In short, numerous health benefits have been linked with PA among a wide-

9

range of populations including both youth and adults, males and females, and across ethnicities. In addition, these findings have been shown using several different types and indicators of physical activity including aerobic fitness, structured exercise, and free-living activity performed during work, home, or leisure-time. Many of the early studies examining the health benefits of PA have used aerobic fitness as a proxy measure for PA behavior (3). Physical fitness is typically assessed by maximal or submaximal exercise testing, which makes it a useful indicator of physical activity since it is relatively objective, accurate, and easy to measure. However, it is important that physical fitness not be confused with physical activity since fitness is also influenced by several other factors including age, sex, health status, and genetics (4). A review by Blair, et al. compared the health impact of fitness and activity on all-cause mortality (5). Each of the nine fitness-based studies included in the review found at least a 50% lower mortality rate among those in the highest fitness group compared to those in the lowest fitness group, with some showing a three- to four-fold difference in risk. Most of the 49 physical activity studies reviewed also showed an inverse dose-response relationship between PA and all-cause mortality, though the results were more variable. When comparing the two measures, the relationship between fitness and mortality was stronger and more consistent than the PA and mortality relationship. However, the authors concluded that this was likely due to the fact that fitness was objectively, and therefore, more accurately measured than the self-reported PA levels. When examining the health effects of physical activity, researchers have typically used self-report PA tools due to their convenience and low cost, features that are especially important in epidemiological studies employing large sample sizes. A review conducted by

10

Lee et al. examined the relation between PA and all-cause mortality, and specifically assessed the role of volume and intensity on this relationship (6). Of the 38 PA papers reviewed, 26 assessed leisure-time PA, three assessed occupational PA, and nine assessed both leisure-time and occupational PA. The authors concluded that a dose-response trend was evident between the volume of PA (usually represented by energy expenditure) and all-cause mortality. This relationship was clear among both men and women, and in middle-aged and older adults. When examining the intensity of PA, there was a trend towards vigorous intensity activity being more beneficial than less-vigorous PA, though there were not enough studies to confirm this trend. However, the authors did suggest that it is logical to assume that higher intensities of activity will lead to a greater volume of PA if performed for a similar duration. A review examining the PA and CVD relationship showed similar results (7). In short, the authors concluded that PA was negatively associated with CVD in a dose-response fashion. The review also separately examined coronary heart disease (CHD) and stroke, the primary manifestations of CVD. Results showed strong support for an inverse relationship between PA and CHD, though the results between PA and stroke were mixed. Thus, it appears that one of the primary ways PA reduces risk of CVD, and all-cause mortality, is through its effect on CHD. These findings have been confirmed in both middle-aged adults (8, 9) and older adults as well (10). Physical Activity Guidelines In recent decades the abundance of research documenting the health benefits of PA (and the detrimental effects of physical inactivity) has led health promotion agencies to recommend habitual PA as part of a healthy lifestyle. In 1995, the Centers for Disease

11

Control (CDC) and the American College of Sports Medicine (ACSM) issued the following PA recommendation to the public: “Every US adult should accumulate 30 minutes or more of moderate-intensity physical activity on most, preferably all, days of the week” (11). Since then, other PA guidelines have been issued with focuses on promoting energy balance as well as general health (12, 13). Recently, the initial CDC/ACSM recommendations were modified in order to adapt to progressions in the relevant literature. This update, published in 2007, established new aerobic PA guidelines for healthy adults (ages 18-65) which included a minimum of 30 min of moderate activity 5 d/wk, 20 min of vigorous activity 3 d/wk, or a combination of the two (14). This updated recommendation emphasized the added benefits associated with vigorous PA and provided clarification for how to distinguish moderate and vigorous intensities. In 2008, the U.S. Department of Health and Human Services issued the Physical Activity Guidelines for Americans (PAGA) in order to provide guidance for PA to health professionals, policy makers, and the public (15). Similar to previous guidelines, the PAGA were developed by a committee of experts in the field who based their report on the available evidence regarding PA and health (16). Though the PAGA provides aerobic and strength training recommendations for children and adolescents, adults, and older adults, the current review will focus on the aerobic PA guidelines for adults. To obtain significant health benefits the PAGA recommends that adults obtain at least 150 min/wk of moderate-intensity aerobic activity, 75 min/wk of vigorous-intensity PA, or a combination of moderate and vigorous PA. Though similar to the 2007 ACSM/AHA guidelines, the PAGA gives adults more flexibility to accumulate weekly PA in various ways as long as it is performed in at least 10-min intervals. In addition, the PAGA increased the vigorous PA guideline from 60

12

min/wk to 75 min/wk. To understand why the guideline for vigorous activity was changed it is important to recognize how the advisory committee established the min/wk values. When developing the PAGA, one of the key conclusions of the review committee was that the health benefits that result from PA are based, primarily, on the total energy expenditure (EE) during these activities. The committee also found that the amount of EE necessary to achieve many of the health benefits was in the range of 500-1000 MET-min/wk. Thus, the current PAGA guidelines were adapted from this MET-min estimate in order to simplify the recommendation for the general public. Using 3 METs as the threshold for moderate activity and 6 METs and above for vigorous, the 500 MET-min per week minimum was adapted into a min/wk estimate for moderate or vigorous PA. After conversion, 500 MET-min per week represented 167 min/wk of moderate PA (500/3) and 83 min/wk of vigorous PA (500/6). To simplify the guidelines, these estimates were rounded to 150 min/wk and 75 min/wk of moderate and vigorous PA, respectively. Since the moderate PA threshold (3 METs) is half of the vigorous threshold (6 METs) the PAGA can be achieved with half as many minutes of vigorous PA as moderate PA (i.e., both will produce the same EE in MET-min). Physical Activity Levels in U.S. Adults Self-Reported Physical Activity Levels Despite the numerous health benefits that can be achieved with habitual PA, more than half of Americans fall short of current recommendations (17). In addition, nearly a quarter of US adults are completely sedentary resulting in substantial and unnecessary increases in health risks to these populations (18). Accurate assessments of PA are needed in order to understand which demographics are at the greatest risk, and to evaluate the impact of

13

current health promotion efforts. Until recently, epidemiological evidence for PA levels has relied exclusively on self-report assessment tools. Such studies often report segments of PA based on why it was performed, such as for leisure, occupational, transportation, or household tasks. Most epidemiological PA studies have examined leisure-time activity exclusively (6), though some have looked at activities from multiple lifestyle contexts (19). Researchers may favor leisure-time PA since it is the most modifiable form of activity for most adults. However, other forms of PA should also be considered, as they can have a positive effect on health (19) and can also influence leisure-time PA. For example, one study found that leisure-time walking made up only half of total walking for adults (20). Another study showed that half of adults who report no leisure-time PA also report an hour or more of hard PA at work (21). Thus, it is important that all forms of PA be considered when assessing total PA levels. One of the primary methods for tracking PA levels among US adults is through the Behavioral Risk Factor Surveillance System (BRFSS) survey. The BRFSS uses a random- digit-dialed telephone survey designed to assess PA performed in bouts of at least 10-min outside of work (i.e., during household work, transportation, and leisure-time). Recent estimates from the BRFSS suggest that the prevalence of Americans who were regularly active increased from 43% in 2001 to 46.7% in 2005 (17). For this study regular PA was defined as meeting the 30 min/d on 5d/wk guideline. Another report based on part of the same dataset (2001-2003) showed similar trends in activity levels (45.3% in 2001 and 45.9% in 2003) (22).

14

To assess levels of physical inactivity outside of work, the BRFSS uses the classification of reporting no activity of 10-min or more of moderate or vigorous PA. According to this definition, the prevalence of lifestyle inactivity has remained relatively stable overall (16.0% in 2001 and 15.6% in 2003), with significant decreases in 14 states and territories, and significant increases in 5 states and territories (22). Recently, one study assessed the prevalence of US adults who met the PAGA (150 min/wk) according to data from 2007 (23). To assess a combination of moderate and vigorous PA, the number of vigorous minutes was multiplied by two before being combined with moderate minutes. The authors found that 64.5% of adults achieved the PAGA (68.9% of men and 60.4% of women). When compared to the ACSM/AHA guidelines only 48.8% of Americans (50.7% of men and 47.0% of women) achieved the guidelines. The authors suggested two reasons why there was difference between the PA prevalence rates. First, the PAGA was easier to achieve because it did not have a daily frequency and duration requirement (i.e., 30 minutes of moderate activity, 5 days per week, or 20 minutes of vigorous activity, 3 days per week). In addition, the PAGA criteria allowed respondents to attain the guidelines through a combination of moderate and vigorous (multiplied by two) activity. Objective Physical Activity Levels In 2008, nationally representative data of objectively measured PA levels were released (24). Using accelerometer data from National Health and Nutrition Examination Survey (NHANES) 2003-2004, Troiano et al. (24) found PA trends with similarities to self- report data (i.e., males were more active than females, PA declined with age, etc.). However, actual PA levels were drastically different than had been previously reported. When

15

categorized according to the 2007 ACSM/AHA guidelines, results showed that less than 5% of adults obtained enough moderate or vigorous PA to reach the guideline. Total minutes of MVPA ranged from 23.6 min/d for young adults (aged 20-29) to 5.4 min/d for those in the oldest category (aged 70+). Furthermore, when considered in bouts of at least 10 min, PA levels ranged from a mere 7.4 min/d to 2.2 min/d from the youngest to oldest age group, respectively. Using the same accelerometer data (NHANES 2003-2004), Metzger et al. categorized US adults by their daily activity patterns using latent class analysis (25). Results showed 5 classes of activity patterns, with 34% of the adult population falling into the least active class (5.4 min/d MVPA). The second-lowest PA class obtained an average of 21.0 min/d MVPA and made up 45% of adults. Thus, the majority of adults (79%) were classified into categories that were well-below recommended PA levels. Even fewer adults were found to be active when the classifications were developed from data showing only PA occurring in 10- min bouts. The divergent results observed for self-report and accelerometer-based assessments raises concerns regarding the validity of one, or potentially both, of these tools. According to both PA assessments, the US adult population has substantial potential for improved health through improvements in currently inadequate PA levels. Indeed, the need for additional PA promotion efforts is especially apparent when considering the accelerometer-based results. However, for appropriate PA promotional efforts and public policies to be made, a clearer understanding of true PA levels in the US is needed. Thus, further investigation is warranted in order to identify reasons for these divergent PA findings, and if possible, to improve the accuracy of these tools.

16

Physical Activity Surveillance Tools As indicated previously, current activity levels in the US vary widely depending upon the PA assessment tools used. Specifically, PA results were substantially higher when using self-report measures when compared to accelerometer measures. In order to determine what factors may lead to these discrepancies it is important to understand how these instruments work and the potential sources of error inherent to each. Self-Report Physical Activity Self-report tools are the most commonly used method for assessing PA, especially in large-scale research, due to acceptability, cost, convenience, and low interference with physical activity behavior (26, 27). In addition, self-report PA measures provide the benefit of being able to capture the frequency, duration, intensity and type of PA reported. Often, the self-reported activities are categorized into the domain in which they are performed (i.e., leisure-time, occupational, domestic, and transportation), allowing researchers to characterize how PA is accumulated throughout the day. Typically, self-reported PA is measured using a series of questions presented in the form of a questionnaire or interview. Such questionnaires require respondents to report recent or typical PA behaviors, usually over a set period of time. Most self-report tools use either a time-based or an activity-based recall strategy. Time-based measures help facilitate recall of past PA by dividing the day into separate periods, after which activities performed during each period are reported. Activity-based recalls involve recalling the specific types of PA that are typically performed, and then estimating the frequency and duration of each activity over a given period of time. Both time-based and activity-based instruments usually also involve reporting the intensity (e.g., moderate, vigorous) that corresponds to each reported PA.

17

Though self-report tools offer some advantages over other measures, they also have several limitations which may introduce bias into the estimate. Discrepancies between self- reported and actual PA may result from some respondents who are unable to accurately recall all PA performed over a given period, which could lead to underestimation. Conversely, some respondents may tend to overestimate their PA, since higher activity levels are socially desirable (28). For example, one study examining this link found that social desirability was the strongest predictor of physical activity frequency in 423 adults (18-50 yrs old) (29). However, other research has shown that social desirability only accounted for 1-3% of the variance in self-reported PA levels (r = 0.11-0.17) (30). Misinterpretation of survey questions may also lead to biased results. According to Baranowski et al, recalling physical activity is a complex cognitive task, which requires respondents to understand ambiguous terms such as “physical activity”, “leisure-time”, and “moderate-intensity” (31). Unfortunately, such terms are often interpreted differently depending on education, culture, experience, or other personal factors. Another limitation of PA surveys is that they often fail to capture all domains of activity (28). Leisure-time activity has been most commonly promoted form of PA, so a number of have focused exclusively on this domain. The increased emphasis on lifestyle activity has led to the development of instruments with an emphasis on total PA levels. Instruments with this focus have items that attempt to capture activity in different domains. For example, the BRFSS captured only leisure-time activity from 1984-2000 (32). Starting in 2001, the BRFSS incorporated a new set of questions which assessed PA from leisure-time, domestic, and transportation. Other national surveys also capture these three domains

Full document contains 132 pages