This article was downloaded by: [Monash University Library] On: 06 October 2014, At: 11:59 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Research Quarterly for Exercise and Sport Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/urqe20

The Energy Balance Study: The Design and Baseline Results for a Longitudinal Study of Energy Balance a

a

a

a

Gregory A. Hand , Robin P. Shook , Amanda E. Paluch , Meghan Baruth , E. Patrick a

a

a

a

a

Crowley , Jason R. Jaggers , Vivek K. Prasad , Thomas G. Hurley , James R. Hebert , b

a

a

Daniel P. O'Connor , Edward Archer , Stephanie Burgess & Steven N. Blair a

a

University of South Carolina

b

University of Houston Published online: 22 Aug 2013.

To cite this article: Gregory A. Hand , Robin P. Shook , Amanda E. Paluch , Meghan Baruth , E. Patrick Crowley , Jason R. Jaggers , Vivek K. Prasad , Thomas G. Hurley , James R. Hebert , Daniel P. O'Connor , Edward Archer , Stephanie Burgess & Steven N. Blair (2013) The Energy Balance Study: The Design and Baseline Results for a Longitudinal Study of Energy Balance, Research Quarterly for Exercise and Sport, 84:3, 275-286, DOI: 10.1080/02701367.2013.816224 To link to this article: http://dx.doi.org/10.1080/02701367.2013.816224

PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Research Quarterly for Exercise and Sport, 84, 275–286, 2013 Copyright q AAHPERD ISSN 0270-1367 print/ISSN 2168-3824 online DOI: 10.1080/02701367.2013.816224

SPECIAL TOPIC

Downloaded by [Monash University Library] at 11:59 06 October 2014

The Energy Balance Study: The Design and Baseline Results for a Longitudinal Study of Energy Balance Gregory A. Hand, Robin P. Shook, Amanda E. Paluch, Meghan Baruth, E. Patrick Crowley, Jason R. Jaggers, Vivek K. Prasad, Thomas G. Hurley, and James R. Hebert University of South Carolina

Daniel P. O’Connor University of Houston

Edward Archer, Stephanie Burgess, and Steven N. Blair University of South Carolina

Purpose: The Energy Balance Study (EBS) was a comprehensive study designed to determine over a period of 12 months the associations of caloric intake and energy expenditure on changes in body weight and composition in a population of healthy men and women. Method: EBS recruited men and women aged 21 to 35 years with a body mass index between 20 and 35 kg/m2. Measurements of energy intake and multiple objective measures of energy expenditure, as well as other physiological, anthropomorphic and psychosocial measurements, were made quarterly. Resting metabolic rate and blood chemistry were measured at baseline, 6 and 12 months. Results: Four hundred and thirty (218 women and 212 men) completed all baseline measurements. There were statistically significant differences by sex uncovered for most anthropomorphic, physiological and behavioral variables. Only percent of kcals from fat and alcohol intake, as well as energy expenditure in light activity and very vigorous activity were not different. Self-reported weight change (mean ^ SD) over the previous year were 0.92 ^ 5.24 kg for women and 2 1.32 ^ 6.1 kg for men. Resting metabolic rate averages by sex were 2.88 ^ 0.35 ml/kg/min for women and 3.05 ^ 0.33 ml/kg/min for men. Conclusion: Results from EBS will inform our understanding of the impact of energy balance components as they relate to changes in body weight and composition. Initial findings suggest a satisfactory distribution of weight change to allow for robust statistical analyses. Resting metabolic rates well below the standard estimate suggest that the evaluation of the components of total energy expenditure will be impactful for our understanding of the roles of energy intake and expenditure on changes in energy utilization and storage. Keywords: body weight, obesity, physical activity, resting metabolic rate

Submitted June 5, 2013; accepted June 10, 2013. Correspondence should be addressed to Gregory A. Hand, Department of Exercise Science, University of South Carolina, 800 Sumter St., Suite 201, Columbia, SC 29203. E-mail: [email protected]

Two thirds of the US population fall into the body mass index (BMI) categories of overweight or obese (Flegal, Carroll, Kit, & Ogden, 2012). Numerous studies demonstrate a relationship between body weight and chronic diseases including cardiovascular disease, metabolic disorders

Downloaded by [Monash University Library] at 11:59 06 October 2014

276

G. A. HAND ET AL.

including diabetes, and some cancers (Haslam & James, 2005). As a result of these findings, reductions in weight and body fat have been a primary aim of national efforts to enhance the nation’s health in such policy documents as Healthy People 2020 (2013). Recommendations on achieving a reduction in body fat are varied, but the typical strategy has focused primarily on restricting caloric intake, with occasional advice to increase energy expenditure (Swinburn, Sacks, & Ravussin, 2009). Development of a theoretical framework to explain weight change and obesity has been a goal of numerous research groups for decades. The dominant framework is currently based on the concept of energy balance, which was derived from the laws of thermodynamics- that energy can be converted among forms but must be conserved. Specifically as related to body weight, an imbalance between energy consumption and expenditure will result in weight change. An imbalance from excessive consumption of energy will result in weight gain while excessive expenditure results in weight loss. The concept of “excessive” is generally thought to be defined by the relationship of intake and expenditure rather than to any particular range or threshold of kcals over time. Further, a change in the relationship of intake and expenditure is extrapolated, incorrectly, to a change in body fat rather than body mass. However, body weight regulation is a complex collection of physiologic, metabolic, environmental, behavioral, and genetic variables that control energy intake and expenditure, as well as the rate and location of energy storage (Hill, Wyatt, & Peters, 2012; Spiegelman & Flier, 2001). In addition, altering energy intake or levels of expenditure usually results in some level of reciprocal compensatory change, with the physiological regulation of weight biased towards maintaining caloric intake and protecting energy stores (Hill et al., 2012; King et al., 2007; Stubbs et al., 2004). The Energy Balance Study (EBS) was designed to determine, as precisely as possible, the impact of caloric intake and energy expenditure on changes in weight, body composition, and indices of health in a population of healthy adults aged 21 to 35 years. The goal of this report is to describe the study design and methodology for EBS, and to present a summary of baseline data. Additionally, insight will be provided as to decisions that were required in an effort to enhance the quality of the data and optimize the likelihood of significant conclusions. This observational study followed 430 individuals for 12 months with extensive baseline assessment of demographic, metabolic, anthropomorphic, and behavioral variables. Following baseline sessions, participants were evaluated quarterly. The primary study outcome variable was body weight. However, critical secondary outcomes included body composition, fat distribution and cardiometabolic variables. The primary exposures were

components of energy flux (energy intake and energy expenditure). In addition, secondary exposures will be examined including demographic and behavioral variables. Results from this study will allow a very precise examination of the extent to which total energy intake and expenditure contribute to changes in body weight and fat over a period of one year.

METHODS Description of Participants The EBS recruited 430 healthy women and men aged 21 to 35 years with a BMI of 20– 35 kg/m2. This age distribution was selected to include individuals ranging from young adults to early middle age, and is a population at risk for decreasing metabolic rate and increasing weight and percent body fat (Ogden, Carroll, Kit, & Flegal, 2012; Sheehan, DuBrava, DeChello, & Fang, 2003). Recruitment consisted of targeting participants across four demographic cells: females aged $ 21 to , 28 years, males aged $ 21 to , 28 years, females aged $ 28 to 35, and males aged $ 28 to 35. Inclusion criteria also required access at all times to a telephone for receiving dietary recall interviews and no plans to move from the catchment area in the next 15 months. Exclusion criteria were designed to select a broad group of healthy individuals with no major acute or chronic conditions who had not made any large changes in health behaviors in the previous months. Therefore, the exclusion criteria included use of medications to lose weight, a change in smoking status in the previous 6 months, or planned weight loss surgery. Further, individuals were excluded for resting blood pressure (BP) exceeding 150 mmHg systolic and/or 90 mmHg diastolic, an ambulatory blood glucose level of greater than 145 mg/dl, or those currently diagnosed with, or taking medications for a major chronic health condition. Individuals with a history of major depression, anxiety disorder, or panic disorder were excluded, as were those taking selective serotonin inhibitors for any reason. Pregnancy and planned births in the previous 12 months were exclusion criteria. Finally, while exclusion criteria were minimal, there were concerns associated with transient and cyclical changes in body water and potential changes in appetite associated with contraceptive medications. However, due to the prevalence of contraceptive usage in the targeted age group, the exclusion was limited to women planning to begin or to stop birth control during the 12 month observational period, and temporary exclusion for those who had begun or changed their birth control regimen in the previous 3 months. All study protocols were approved by the University of South Carolina Institutional Review Board.

ENERGY BALANCE METHODS AND BASELINE RESULTS

Participant Flow and Timeline Flow of contact with participants is described in Figure 1. The study followed participants for a period of 12 months, and included three baseline measurement sessions and follow-up measurement sessions at 3-month intervals. Initial contact by participants was via a brief online screening procedure that addressed general demographic and health questions. This initial online screening was followed by a telephone screen.

Downloaded by [Monash University Library] at 11:59 06 October 2014

Orientation Session All participants who were determined to be eligible initially were invited to an orientation session that included a 25-min presentation describing the study and all associated procedures and expectations for participation in the study. The orientation session included measurement of height and weight for verifying BMI. This session began a series of three additional baseline measurement sessions that were performed over the following 3 weeks. Measurements All measurements were obtained by trained research staff who had demonstrated competency and were certified in the

277

specific measurement technique. Measurement procedures and timeframes were consistent across baseline and followup sessions. However, the 12-month activity measurement period was increased to 14 days so as to coincide with the entire doubly labeled water (DLW) collection period. Baseline 1 began with an extensive review of medical history and demographic information. The primary purpose of the baseline 1 session was completion of an extensive battery of demographic, psychometric, and activity recall questionnaires. Participants also received portion size estimation training to prepare them for interviewer administered dietary recall interviews conducted later in the study. Baseline 2 included measurements of resting BP, height and weight, waist and hip circumference, body composition via dual x-ray absorptiometry (DXA) full body scan, as well as a maximal fitness test using a modified Bruce protocol with 12-lead electrocardiogram (ECG) and BP measurements. Baseline 3 was conducted while the participant was in a 12-hr fasting state and 24-hr abstention of physical activity. Measurements included height and weight, waist and hip circumferences, resting metabolic rate, and a blood draw. In addition, the participants were issued a SenseWear Mini Armband (BodyMedia Inc., Pittsburgh, PA) and an ActivPal (PAL Technologies Ltd., Glasgow, UK) physical activity monitor for a 10-day measurement period. During the activity measurement period, the participants received three interviewer-administered dietary recalls on random days that included two weekdays and one weekend day. These surveys included extensive questioning about dietary intake over the previous 24-hr period and the surveys were conducted by licensed dieticians who had extensive experience in conducting dietary assessments. Follow-up measurements were completed every three months for a total of one year. Table 1 describes the measurements completed at each follow-up session. Motivating Participants and Incentives Participants in the EBS received $500 for completing the study. Additionally, upon completion of the EBS participants received a one-on-one counseling session with a member of the study staff during which they received much of the information collected on them. An accompanying report included information on the participant’s measured energy intake, energy expenditure, body weight, body composition, resting metabolic rate, and cardiorespiratory fitness levels. This information was not provided to participants until completion of the study. Outcome and Exposure Measurements

FIGURE 1 Flow chart of participant progress from initial contact through final baseline session.

A comprehensive health screening was completed online during the week following the orientation session. Briefly,

278

G. A. HAND ET AL. TABLE 1 Laboratory Measures Administered in the Energy Balance Study Baseline 1

Downloaded by [Monash University Library] at 11:59 06 October 2014

Physical activity and dietary assessment Body composition assessment Resting metabolic rate Blood chemistry Fitness test Doubly-labeled water (n ¼ 200)

Baseline 2

Baseline 3

3M

6M

9M

12M

15M

18M

21M

24M

X

X X

X X X X

X X

X X X X

X X

X X X X

X X

X X X X

X X X X

X X

information regarding personal and family health, personal health habits and demographic information were collected from each participant. In addition, multiple psychometric instruments including the Profiles of Mood State (POMS) (McNair, Lorr, & Droppleman, 1992) and Perceived Stress Scale (PSS) (Pbert, Doerfler, & DeCosimo, 1992) were administered, along with assessment of the subjects’ physical activity using the International Physical Activity Questionnaire- Long Form (IPAQ) (Craig et al., 2003). These instruments have proven valid and reliable in previous studies. Table 2 provides a complete list of questionnaires. Anthropometry All anthropomorphic measurements were performed with the participant dressed in surgical scrubs and in bare feet. BMI (kg/m2) was calculated from the average of three height and weight measurements using a traditional stadiometer and electronic scale and recorded to the nearest 0.1 centimeter and 0.1 kg, respectively. Body composition was measured using a Lunar fan-beam dual X-ray absorptiometry (DXA) scanner (GE Healthcare model 8743, Waukesha, WI). Total fat and lean tissue mass, as well as torso, arm and leg tissue composition were determined. Bone mineral density and content also were recorded. Hip and waist circumferences were measured with a calibrated,

spring-loaded tape measure. Waist circumference was determined at the point midway between the costal margin and iliac crest in the mid-axillary line approximately 2 inches above the umbilicus. Hip circumference was measured at the widest point around the greater trochanter. Circumferences recorded were the average of three measurements and were rounded to the nearest 0.1 cm. Measurement of Energy Intake (Dietary Recalls) Estimates of energy and nutrient intakes were calculated from a self-report of food consumption. This information was collected using multiple, telephone-administered, 24-hr recall interviews (24 HR). Although this method is not devoid of measurement error, it is considered an imperfect “gold standard” for estimating current dietary intake (Novotny et al., 2001). The Nutrient Data System for Research software (NDSR Version 2012), licensed from the Nutrition Coordinating Center (NCC) at the University of Minnesota, was utilized to conduct the dietary interviews. NDSR is considered the state-of-the-art research software for conducting 24 HR. Its food database includes over 19,000 foods, is updated yearly, and provides nutrient composition information for over 120 nutrients. The quality of a 24 HR depends both on the ability of the subject to remember which foods were consumed (Novotny et al.,

TABLE 2 List of Questionnaires Administered in the Energy Balance Study Questionnaire Demographics Medical History Pittsburgh Sleep Quality Index Perceived Stress Scale Eating disorders Three-Factor Eating Questionnaire Control of Eating Questionnaire Body Image Profile of Mood State Social desirability-Social approval International Physical Activity Questionnaire (Long form) Physical Activity Life events (Past 3 months)

Baseline

3m

6m

9m

X X X X X X X X X X X X

X X

X X

X X X

12 m

15 m

18 m

21 m

24 m

X

X

X X X X X X X X

X X X X X X X X

X X

X X

X X X

X X

X X

Downloaded by [Monash University Library] at 11:59 06 October 2014

ENERGY BALANCE METHODS AND BASELINE RESULTS

2001), as well as the skill of the interviewer in eliciting complete and accurate information. In this study, the 24 HRs were collected by a team of experienced (. 6 years using NDSR), registered dietitians specifically trained in using the NCC protocol. This protocol employs the multipass approach which utilizes prompting to reduce omissions, and standardizes the interview methodology across interviewers (Dwyer, Ellwood, Leader, Moshfegh, & Johnson, 2001). Portion estimation by the participant is facilitated with the use of a validated, 2-dimensional, food portion visual (FPV) that is an integral part of the NDSR software (Posner et al., 1992), and which we have used successfully in multiple studies in adults and adolescents. Prior to data collection, study participants undergo a brief training (10 – 15 min) on how to use the FPV to estimate portion sizes of commonly eaten foods. The training incorporates life-sized plates, glasses and utensils and food models, in a hands-on experiential interchange (Wilcox, Sharpe, Parra-Medina, Granner, & Hutto, 2011). Interviews are assigned on randomly selected, nonconsecutive days, and cold calls are made to the study subject to minimize preparation/rehearsing that could bias recall (Hebert et al., 2002). The sampling window was set at 10 days to be adequately large to allow multiple attempts on multiple days to maximize, as much as possible the likelihood of completing an interview. The sampling window was 14 days during the doubly labeled water (DLW) administration period (additional discussion of DLW below) to coincide with the period of energy expenditure estimation.

Measurements of Energy Expenditure Accelerometer device. While criterion measures of energy expenditure have been DLW and indirect calorimetry, these techniques are inappropriate for a large freeliving population due to the required cost and extensive expertise and equipment. We have chosen, therefore, to use the SenseWear Mini Armband to measure multiple components of energy expenditure over 10-day periods on a quarterly time table and administering DLW to a subset of the participants at the 12-month measurement period. This portable, multi-sensor device, worn on the upper arm, incorporates tri-axial accelerometry, heat flux, galvanic skin response, skin temperature, and near-body ambient temperature. These measures are entered in combination with demographic information into an algorithm to estimate energy expenditure, activity, and sleep. The armband has been shown to be a valid device to measure energy expenditure and activity (St Onge, Mignault, Allison, & Rabasa-Lhoret, 2007; Welk, McClain, Eisenmann, & Wickel, 2007). Concurrent validation of the armband data collected during this study will occur by correlating a subset of the study participants (n ¼ 200) with the results of the 12-month energy expenditure measurement with DLW.

279

Participants were trained for approximately 20 min on the care and use of the armband activity monitor. The individuals started wearing the monitor immediately and were asked to continue use of the monitor except during periods when the monitor could get wet. For most individuals, this only included periods of showering or bathing. The participants wore the armband for 10 days and recorded their activities during any period of that time that the armband was not worn. Participants were instructed to record when they removed and replaced the Armband, and provide detail for the reason the Armband was removed. These non-wear periods of time were then added to the Armband on-body recording estimates to allow for an entire 24-hr estimate of energy expenditure. Periods of non-wear time were filled based on matching the activity with the corresponding MET value according to the 2011 Compendium of Physical Activities (Ainsworth et al., 2011). Energy expenditure for such periods was calculated as the MET value (MET minutes) times the individual’s measured resting metabolic rate (RMR). The use of the individual RMR should provide a conservative estimate of energy expenditure during non-wear periods as compared to the commonly used RMR value of 3.5 ml/kg/min. Participants were deemed compliant if they completed 7 days of wear (including two weekend days) with at least 21 h of verifiable time on each of the days. Inclinometer. The activPAL is a small posture inclinometer that adheres to the anterior thigh, and can determine time spent lying/sitting, standing and stepping over a continuous 7- to 8-day period. This device can be used to estimate daily energy expenditure and physical activity and primarily has been included in this study to determine static work levels during the activity monitoring periods that occur at baseline and at quarterly follow up periods. The activPAL has been shown to be a valid measure of sedentary behavior (Kozey-Keadle, Libertine, Lyden, Staudenmayer, & Freedson, 2011), and long periods of sedentary behavior have been linked to morbidity and mortality independent of physical activity (Owen, Healy, Matthews, & Dunstan, 2010). Differentiating sedentary behaviors of sitting/lying from standing postures can provide insight into slight differences in energy expenditure that could affect energy balance. Participants were provided the activPAL monitor and sheets of adhesive tape for the 10-day measurement period coinciding with the armband period. Compliance was determined as described for the armband. Doubly labeled water (DLW). DLW is used to estimate carbon dioxide production using isotope dilution to determine total energy expenditure over a period of two weeks. DLW is a mixture of stable isotope-labeled water (2H18 2 O) administered orally in a body-weight adjusted dose. The isotopes equilibrate with body water in several hours,

Downloaded by [Monash University Library] at 11:59 06 October 2014

280

G. A. HAND ET AL.

and urine samples are collected and analyzed to determine the elimination rates over time. A standard procedure was chosen that includes collecting 2 urine samples, at least 60 min apart, on day 1 following the dosing, then on day 8 and day 14. Each urine sample is analyzed by isotope ration mass spectrometry using automated analyzers for 2H2 and 18 O (GasBench, Finnigan) (Delany, Schoeller, Hoyt, Askew, & Sharp, 1989). The difference between elimination rates of the two isotopes is proportional to carbon dioxide production, which is used to calculate total energy expenditure (TEE). Combined with measurement of resting energy expenditure (REE), activity energy expenditure (AEE) can be calculated as AEE ¼ TEE-(REE þ 0.1TEE) (Speakman, 1998). Because of budgetary concerns and participant burden, DLW was administered to a subset (n ¼ 200) of the study participants evenly distributed across the four age/sex cells previously described. Resting metabolic rate (RMR). RMR was determined for each participant using standard indirect calorimetry with a ventilated hood (True One 2400, Parvo Medics, Sandy, UT). Subjects arrived at the clinical exercise research center for a morning session in a 12-hr dietary fasting state and 24 hr after the last bout of exercise. RMR was measured over a 30 min period with data collection beginning after a 15 min supine resting period. RMR was calculated from O2 consumption and CO2 production as measured continuously during the testing period with a constant airflow rate into the ventilated hood. Airflow rate was based approximately on body weight with the goal of maintaining the fraction of end tidal CO2 between 1.0 and 1.2% (L/min). The flow rate was set to not exceed 33 L/min for a person weighing 68 kg or 40 L/min for a 91 kg person. Participants remained quiet and still through the entire RMR procedure. The room was maintained in low light and noise was kept at a minimum. Participants remained awake during the measurement period with continuous monitoring by a staff member. Fitness Assessment A trained exercise physiologist prepared eligible subjects for the graded exercise test (GXT), and a standard 12-lead ECG was performed. Subjects sat quietly for examination of the real-time resting ECG, heart rate (HR), and BP. Once cleared for participation, the exercise physiologist administered the GXT using a Modified Bruce protocol on a motorized Trackmaster treadmill (Full Vision, Inc., Newton, KS). All subjects exercised to volitional fatigue, followed by continued walking at a slow pace until HR and BP returned to near baseline levels. Heart rate, BP, RPE and treadmill total time were recorded at each stage of the protocol. Criteria for a maximal test included 2 of the following variables: a heart rate plateau or VO2 plateau with increasing workload, a respiratory quotient (RQ) $ 1.15,

and a rate of perceived exertion (RPE) $ 17 using the Borg scale of perceived exertion. The Modified Bruce GXT begins at a speed of 1.7 mph at 0% grade for 2 min then progresses to 1.7 mph at 5% grade for 3 min. After this stage, the protocol is identical to that of the Bruce Protocol. The Modified Bruce Protocol was used due to its lower initial intensity for this generally deconditioned population. Previous research has shown a high correlation between the Modified Bruce and Bruce protocols for HR responses (r ¼ 0.97) and peak VO2 measurements (r ¼ 0.72) (McInnis & Balady, 1994). Blood Lipids/Glucose Blood draws were completed following a 12-hr fast as described previously (Sieverdes et al., 2011). Participants were asked to refrain from taking aspirin and other antiinflammatory medications for 48 h prior to the blood draw. Twenty milliliters of whole venous blood were drawn into standard vacutainer tubes by trained personnel. Of the collected amount, half of the sample was centrifuged for separation of serum assays while the other half was centrifuged for separation of plasma. All components are stored at 280 degrees for future analyses. Statistical Analysis and Power All frequencies and distributions were calculated using IBM SPSS Statistics version V.19. All results are described as mean ^ SD unless indicated. Statistical power of .954 was determined using an average effect size of 2 ^ 15 kg change in body weight, n ¼ 400 and alpha set at 0.05.

RESULTS Participant Flow A total of 1831 automatic, online screens resulted in followup telephone calls to 866 potential participants [Figure 1]. Based on the inclusion and exclusion criteria used during the telephone screening process, individuals identified as initially eligible were invited to the orientation sessions. Four hundred and seventy four individuals attended orientation sessions, with 446 providing informed consent and returning for the first baseline session. Four hundred and thirty participants completed all baseline sessions and are currently in the follow-up stages of the study. Therefore, data from 430 participants (218 women and 212 men) are included in the baseline dataset. Demographics, Anthropometrics, and Resting Metabolic Rate The mean age was 27.7 ^ 3.8 years with a racial/ethnic makeup of predominantly white/European Americans

ENERGY BALANCE METHODS AND BASELINE RESULTS TABLE 3A Demographic Variables

Downloaded by [Monash University Library] at 11:59 06 October 2014

Variable Age Race, n (%) White/European Black/African Hispanic/Latino Asian Native American Other Education, n (%) HS Graduate/GED Some College College (4 þ years) Income ($), n (%) 0– 9,999 10,000–19,999 20,000–29,999 30,000–39,999 40,000–49,999 50,000–59,999 60,000–69,999 70,000–79,999 80,000 þ Employment, n (%) Employed for wages Self-employed Out of work (,1 year) Homemaker Student Unable to work Marital Status, n (%) Married Divorced Separated Never married Unmarried couple Children, age , 18 yrs, n (%) 0 1 2 3 4

All Participants (n ¼ 430)

Women (n ¼ 218)

Men (n ¼ 212)

27.7 ^ 3.8

27.8 ^ 3.7

27.5 ^ 3.8

286 (66.5) 54 (12.6) 13 (3) 46 (10.7) 15 (3.5) 16 (3.7)

142 (65.1) 38 17.4) 8 (3.7) 16 (7.3) 8 (3.7) 6 (2.8)

144 (67.9) 16 (7.5) 5 (2.4) 30 (14.2) 7 (3.3) 10 (4.7)

4 (0.9) 66 (15.3) 360 (83.7)

1 (0.5) 21 (9.6) 196 (89.9)

3 (1.4) 45 (21.2) 164 (77.4)

281

41.4 ^ 5.6 kg and 60.9 ^ 8.1 kg for women and men, respectively. Resting metabolic rate for women was 2.88 ^ .36 ml/kg/min and 3.04 ^ .33 ml/kg/min for men [Figure 3]. Energy Intake (Random 24-hr Dietary Recalls)

24 (5.6) 47 (10.9) 75 (17.4) 75 (17.4) 54 (12.6) 32 (7.4) 34 (7.9) 19 (4.4) 68 (15.8)

13 23 45 40 22 21 14 10 28

(6) (10.6) (20.6) (18.3) (10.1) (9.6) (6.4) (4.6) (12.8)

11 (5.2) 24 (11.3) 30 (14.2) 35 (16.5) 32 (15.1) 11 (5.2) 20 (9.4) 9 (4.2) 40 (18.9)

227 (52.8) 9 (2.1) 2 (0.5) 2 (0.5) 189 (44) 1 (0.2)

122 (56) 3 (1.4) 0 2 (0.9) 90 (41.3) 1 (0.5)

105 (49.5) 6 (2.8) 2 (0.9) 0 99 (46.7) 0

140 (32.6) 7 (1.6) 1 (0.2) 224 (52.1) 58 (13.5)

64 (29.4) 4 (1.8) 0 114 (52.3) 36 (16.5)

76 (35.8) 3 (1.4) 1 (0.5) 110 (51.9) 22 (10.4)

366 (85.1) 35 (8.1) 20 (4.7) 4 (0.9) 4 (0.9)

189 (86.7) 15 (6.9) 9 (4.1) 2 (0.9) 2 (0.9)

177 (83.5) 20 (9.4) 11 (5.2) 2 (0.9) 2 (0.9)

(66.5%) followed by black/African Americans (12.6%) and Asian/Asian Americans (10.7%) [Table 3a]. A majority of participants have four or more years of college education (83.7%) and have an annual income of $30,000 or more (65.5%). Almost half are students (46.7%) and most have not been married (52.1%) and have no children (85.1%). The average height and weight of participants resulted in a BMI of 25.4 ^ 3.8 [Table 3b]. Weight change over the previous year, calculated as current fasting body weight minus self-reported body weight from one year previously, was 0.92 ^ 5.24 kg for women and 21.32 ^ 6.09 for men [Figure 2]. Average percent body fat was 35 ^ 8.1 for women and 21 ^ 8.4 for men with lean mass totaling

During the baseline dietary assessment period, approximately 75% of participants completed three dietary assessments, and more than 97% of participants completed at least two dietary recall interviews, the average call time was 18.5 min. As described in Table 3b, energy intake averaged 1792 ^ 471 kcals/day for women and 2407 ^ 818 kcals/day for men, with the largest component of those calories coming from carbohydrates in both groups. Energy Expenditure (Activity Monitors) The compliance target for armband wear time was 7 complete days of wear ($21 h/day of recorded time) including two weekend days. Participants achieved an average wear time of 10.0 ^ 0.9 days of wear with 23.4 ^ 1.3 h/day. As described under Methods, journal entries were used to fill gaps in wear periods (e.g. during activities such as swimming). Total daily energy expenditure averaged 2389 ^ 515 and 3115 ^ 420 for women and men, respectively [Table 3b]. Approximately half of the calories expended by both groups were metabolized during sedentary activity. Fitness Testing Average peak oxygen consumption (mL O2/kg body weight/ min) was 33.0 ^ 7.9 for women and 44.5 ^ 9.4 for men, with peak respiratory exchange ratios (RER) of 1.21 ^ 0.11 and 1.25 ^ .09 indicating that high work levels were achieved. In addition, participants achieved maximal rates of perceived exertion (RPE) of 17.4 ^ 6.3 and 17.4 ^ 1.9 for women and men, respectively. Finally, BPs (systolic/ diastolic/mmHg) at maximal exertion were 166 ^ 22/75 ^ 9 and 188 ^ 20/79 ^ 10 for women and men, respectively.

DISCUSSION Multifactorial Nature of Weight Change and Obesity While in practice its true that increased weight must be due to excessive intake of calories relative to expenditure, obesity and weight change are multifactorial with significant evidence to support an influence of genetics, environmental factors, physiology, behavior, and interactions of these variables. To further complicate the understanding of weight regulation and obesity, the three primary components of the energy balance model are interactive. For example, numerous studies have shown a

282

G. A. HAND ET AL. TABLE 3B Anthropomorphic, Physiological, and Behavioral Variables

Downloaded by [Monash University Library] at 11:59 06 October 2014

Variable Height (cm) Weight (kg) BMI Waist Circ (cm) Hip Circ (cm Body fat (%) Total fat mass (kg) Total lean mass (kg) Wt 1 year ago (kg) VO2 max (ml/kg/min) RER max RMR L/min RMR (ml/kg/min) Energy Intake (kcal/day) Fat (g/day) Carbohydrates (g/day) Protein (g/day) Alcohol (g/day) Fat kcals (%) Carbohydrates kcals(%) Protein kcals (%) Alcohol kcals (%) Energy Expenditure (kcal/day) Sedentary EE (kcal/day) Light activity EE (kcal/day) Mod activity EE (kcal/day) Vigorous act EE (kcal/day) Very vig act EE (kcal/day) All MVPA (kcal/day)

All Participants (n ¼ 430) Mean ^ SD

Women (n ¼ 218) Mean ^ SD

Men (n ¼ 212) Mean ^ SD

Cohen’s d Between Sexes

p-value Between Sexes

171.8 ^ 9.4 75.1 ^ 13.9 25.4 ^ 3.8 79.9 ^ 10.2 100.7 ^ 7.9 28.0 ^ 10.8 21.2 ^ 9.8 51.0 ^ 11.9 74.9 ^ 14.6 38.7 ^ 10.4 1.24 ^ 0.11 0.22 ^ 0.04 2.96 ^ 0.36 2094 ^ 732 79.9 ^ 36.5 244.3 ^ 86.1 88.5 ^ 39.6 2.9 ^ 5.1 33 ^ 8 47 ^ 10 17 ^ 5 3^5 2747 ^ 515 1459 ^ 272 614 ^ 196 598 ^ 333 50 ^ 62 28 ^ 78 676 ^ 395

165.2 ^ 6.5 69.0 ^ 12.7 25.3 ^ 4.3 76.1 ^ 9.7 101.6 ^ 8.7 35.0 ^ 8.1 24.9 ^ 9.4 41.4 ^ 5.6 68 ^ 11.5 33.0 ^ 7.9 1.2 ^ 0.1 0.2 ^ 0.03 2.88 ^ 0.36 1792 ^ 471 67.6 ^ 25.5 219.0 ^ 61.6 72.6 ^ 22.5 7.2 ^ 12.7 32 ^ 7 49 ^ 9 17 ^ 4 3^4 2387 ^ 301 1282 ^ 172 608 ^ 192 440 ^ 212 36 ^ 41 22 ^ 76 498 ^ 264

178.5 ^ 7.1 81.3 ^ 12.5 25.5 ^ 3.3 83.7 ^ 9.1 99.9 ^ 7 21.0 ^ 8.4 17.5 ^ 8.9 60.9 ^ 8.1 81.6 ^ 14 44.5 ^ 9.4 1.3 ^ 0.1 0.25 ^ 0.03 3.04 ^ 0.33 2407 ^ 818 92.6 ^ 41.6 270.6 ^ 99.3 104.9 ^ 46.3 14.9 ^ 40.9 33 ^ 8 45 ^ 11 18 ^ 6 3^6 3118 ^ 417 1639 ^ 235 620 ^ 201 759 ^ 356 66 ^ 75 34 ^ 79 858 ^ 423

1.95 0.98 0.05 0.81 0.26 1.7 0.81 2.8 1.06 1.32 1.0 1.67 0.46 0.94 0.72 0.62 0.89 0.25 0.13 0.4 0.2 0.0 2.0 1.69 0.06 1.1 0.50 0.16 1.02

,.001 ,.001 ,.001 ,.001 .04 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 ,.001 .01 .28 .001 .001 .07 ,.001 ,.001 .53 ,.001 ,.001 .10 ,.001

Note: BMI ¼ body mass index; circ ¼ circumference; wt ¼ weight; RER ¼ respiratory exchange ratio; RMR ¼ resting metabolic rate; EE ¼ energy expenditure; Mod ¼ moderate activity; vig ¼ vigorous; MVPA ¼ moderate and vigorous physical activity

compensatory response to reduced energy intake resulting in reduced physical activity and caloric expenditure (King, Hopkins, Caudwell, Stubbs, & Blundell, 2008; Whybrow et al., 2008). Further, some studies suggest that increasing physical activity will result in a higher level of energy intake by altering levels of satiety, with resulting changes in eating behavior (Stubbs et al., 2004). But there is some evidence that creating a negative energy balance by different means will evoke differential and complex responses. One example is the work of Hubert et al. (Hubert, King, & Blundell, 1998) indicating that creating a short-term energy deficit by restricting caloric intake to a low-calorie breakfast results in increased food cravings and eating behavior. By contrast, increased physical activity in conjunction with either a lowcalorie or high-calorie breakfast had no significant effect on either variable. Specific to energy expenditure, the predominant component of calorie usage is through resting metabolic rate, which accounts for 50–80% of total daily caloric expenditure. However, the role of RMR on changes in body weight or body fatness is not entirely clear. Studies report low RMRs associated with weight gain (Ravussin

et al., 1988) while others find no relationship between RMR and either body mass or fatness (Katzmarzyk, Perusse, Tremblay, & Bouchard, 2000; Marra, Scalfi, Covino, Esposito-Del, & Contaldo, 1998). Clearly, there is significant uncertainty in both the compensatory effects of various methods of caloric restriction and the effect of RMR on changes in weight and body composition. To further complicate these relationships, most studies use a standardized RMR estimate of 3.5 ml O2/kg body weight/min. Results from the present study (Figure 3) verifies findings from previous work (Byrne, Hills, Hunter, Weinsier, & Schutz, 2005) illustrating that this standardized quantity significantly overestimates the average resting energy expenditure. Therefore, typical weight loss calculators are both overestimating the effect of both RMR and the multiples of RMR used to estimate the energy expenditure of activities. Excessive Caloric Intake or Caloric Expenditure The prevailing view among many investigators is that excessive caloric intake is the predominant characteristic of

Downloaded by [Monash University Library] at 11:59 06 October 2014

ENERGY BALANCE METHODS AND BASELINE RESULTS

weight gain and obesity with energy expenditure perhaps playing a small role. This contention is based on information gleaned from national data sets suggesting that (a) the increase in the food supply is adequate to explain the change in the population’s body weight (Swinburn et al., 2009), and (b) that physical activity energy expenditure has not declined over the last few decades and, therefore cannot account for the change in body weight (Westerterp & Speakman, 2008). The implication for this work is that eating behavior changes over decades have caused the change in the population’s weight and, therefore efforts to address obesity should focus on controlling caloric intake. Recent work, however suggests that there has been a significant reduction in energy expenditure over the last several decades. This reduction could be due to increased occupational automation, labor-saving devices associated with housework, and life style choices for leisure time. Mathematical modeling was used to predict weight change since 1960 with predictions based on the over 100 kcals/day reduction estimated in occupational energy expenditure (Church et al., 2011). The results of this modeling demonstrate that the progressively decreasing average

FIGURE 2 Histograms illustrating weight change over the previous year for women and men. Change was calculated as baseline body weight measured on-site minus self-reported weight for previous12-month time point.

283

occupational energy expenditure of over 100 kcals/day predicts closely the increase in body weight over the same period of time. An evaluation of the time spent in household activities by women over the last four decades indicates that time associated with household chores was reduced by at least seven hr/week with a concomitant decrease in energy expenditure of over 265 kcals/day (Archer et al., 2013). As with the results of the occupational energy expenditure study, this dramatic drop in energy expenditure by the population over decades can explain a significant amount, if not all of the increased weight over the period of years. Triangulation of Results As with all of behavioral science, there are limitations to the validity and reliability of measurements for energy intake and expenditure. Each type of measurement technique has its own set of strengths and limitations. These are amplified in studies such as the EB Study, which uses both objective and self-report measures. A commonly stated concern with determining dietary habits is underreporting of food consumption. The standard methods of measurement (journal, frequency questionnaires, and random recalls) all

FIGURE 3 Histogram illustrating the measured resting metabolic rates for women and men. Horizontal lines indicate the mean RMR.

Downloaded by [Monash University Library] at 11:59 06 October 2014

284

G. A. HAND ET AL.

have challenges associated with validity and subjectspecific bias (Kipnis et al., 2002). The choice of multiple, random 24HR dietary recalls was based on their ability to capture food items that may not be listed on a food frequency questionnaire. A single 24HR provides an unstable estimate of current dietary behavior for an individual due to large day-to-day variability. With a sufficient number of interviews per subject the 24HR is the assessment method with the lowest overall measurement error. In this study, three 24HR, which is considered the minimum number of days to reliably estimate energy and macronutrient intake (Hebert, Hurley, Chiriboga, & Barone, 1998), were collected at each measurement period. Interviews are scheduled so that dietary data can be collected from two weekdays and one weekend day to help balance known differences in food intake patterns (Basiotis, Welsh, Cronin, Kelsay, & Mertz, 1987). The use of objective measures of movement through development of portable accelerometry devices has revolutionized the study of physical activity in free-living individuals. In the EBS measurement of activity, with concomitant extrapolation of energy expenditure, was performed with an armband monitor that estimates energy expenditure from a regression model of accelerometry combined with numerous physiological measurements. As presented by Chen et al. (Chen, Janz, Zhu, & Brychta, 2012), physiological monitors can enhance the sensitivity and accuracy of energy expenditure estimates. This is especially important when applying monitoring to a relatively large population of individuals who participate in a wide range of activities that may not be suited to basic accelerometry. While the measurement of the primary components of energy balance has shortcomings, it is a strength of the EBS that each component can be measured or estimated through multiple mechanisms over an extended period of time. For example, the determination of energy expenditure will be measured by armband activity monitors, these monitors will be validated through measurement of expenditure using doubly labeled water, and energy expenditure will be estimated from physical activity self-report using the IPAQ. In addition, energy expenditure can be calculated from energy storage estimates arising from accurate measurements of body composition/weight change and calculations of energy intake (energy expenditure ¼ energy intake – energy storage). Conversely, the energy balance equation stated above can be used to estimate energy intake. Additionally, the accuracy of energy intake measurements will be enhanced through multiple dietary recalls administered quarterly and repeated over the study period.

(and fat mass) depend on changes in energy intake and expenditure as well as body weight (and fat mass) at the start of the study period. Consequently, the EB study will use transitional models (also known as dynamic models) as a general analytical framework (Diggle, Heagerty, Liang, & Zeger, 2002; Heagerty & Comstock, 2013), operationalized using linear mixed models software. The parameters of interest in these models represent the relation between the predictors (i.e., intake and expenditure) and the outcome (i.e., body weight) controlled for the prior outcome (i.e., previous time period’s body weight). Transition models may also be constructed to allow for the outcome to vary by prior values of the predictors. Such time-varying covariates are in contrast to time-invariant covariates, such as sex and race, and represent dynamic effects over time. For example, change in body weight over a given time period may depend on changes in energy intake and expenditure from one time period to the next, which can be modeled by including intake or expenditure for both the current and prior time periods in the model. This flexible modeling approach will allow for testing how changes from typical energy intake and expenditure are related to changes in body weight. Covariates will be added to the model to determine how variables such as demographics, behaviors, and psychological states affect the associations between changes in energy intake and energy expenditure to changes in body weight. These models can accommodate nonlinear relations (e.g., heavier individuals having different weight changes over time compared to lighter individuals) as well as interactions among covariates and energy intake or expenditure. The EBS has multiple outcomes, including body weight, body composition (fat mass and fat-free mass, or percent body fat), and cardiometabolic variables. The EBS also includes a number of different methods for estimating energy intake and energy expenditure. Using the above analytical approach and various combinations of outcome, energy intake, and energy expenditure variables in a series of models provides a way to make use of the full data. While metrics and magnitudes of coefficients will vary among these disparate analyses, the general trends and associations between changes in energy intake and expenditure and body weight and other outcomes should be consistent across all models and analyses. This triangulation of evidence by using different measures and methods to investigate and test the same phenomenon provides a way to evaluate the stability and robustness of the results and conclusions.

WHAT DOES THIS ARTICLE ADD? Analytical Approach The primary aim of the EBS is to investigate how energy intake and energy expenditure are related to changes in body weight (and fat mass) over time. Changes in body weight

The EBS is designed to evaluate the impact of energy intake and expenditure on weight change in a group of healthy women and men over a 12-month period. This multicomponent, observational study includes multiple variables

Downloaded by [Monash University Library] at 11:59 06 October 2014

ENERGY BALANCE METHODS AND BASELINE RESULTS

associated with energy intake, expenditure, and storage. These combinations of variables are designed to “triangulate” on the exposures that are the major determinants of weight change and body composition. Such triangulation will allow us to cross-validate results, and enhance our confidence in the credibility of our analyses. Initial analyses indicate that there is significant weight change among the observed population and that the RMR for most men and women are below the standard estimate of 3.5 ml/kg/min. The study is powered statistically to evaluate small changes in the four variables of primary interest (weight, fatness, energy intake and energy expenditure) using quarterly measurements for over 400 participants. These precise measurements, administered multiple times over a broad range of variables will inform our understanding of energy balance and its impact on energy storage. Findings from this study will enhance our understanding of the primary determinants of weight gain and obesity.

ACKNOWLEDGMENTS The authors wish to thank the study participants and our project consultants (Drs. Allison, Blundell, Church, Hamilton, Hill, Jakicic, Katzmarzyk, Thomas, and Welk). Funding for this project was provided through an unrestricted grant from The Coca-Cola Company. The sponsor played no role in the study design, collection, analysis and interpretation of data, or preparation and submission of this manuscript.

REFERENCES Ainsworth, B. E., Haskell, W. L., Herrmann, S. D., Meckes, N., Bassett, D. R. Jr, Tudor-Locke, C., . . . Leon, A. S. (2011). 2011 Compendium of Physical Activities: A second update of codes and MET values. Medicine and Science in Sports and Exercise, 43, 1575– 1581. doi: 10.1249/ MSS.0b013e31821ece12 Archer, E., Shook, R. P., Thomas, D. M., Church, T. S., Katzmarzyk, P. T., Hebert, J. R., & . . . Blair, S. N. (2013). 45-year trends in women’s use of time and household management energy expenditure. PLoS.One, 8, e56620. doi: 10.1371/journal.pone.0056620 Basiotis, P. P., Welsh, S. O., Cronin, F. J., Kelsay, J. L., & Mertz, W. (1987). Number of days of food intake records required to estimate individual and group nutrient intakes with defined confidence. Journal of Nutrition, 117, 1638 – 1641. Retrieved from http:// ernaehrungsdenkwerkstatt.ed/fileadmin/user_upload/EDWtest/Text Elemente/PHN-Texte/Erhebungsmethoden/Tage_Anzahl_Food_intake_ Basiotis_JNU_1987.pdfin Byrne, N. M., Hills, A. P., Hunter, G. R., Weinsier, R. L., & Schutz, Y. (2005). Metabolic equivalent: one size does not fit all. Journal of Applied Physiology, 99, 1112–1119. doi: 10.1152/japplphysiol.00023.2004 Chen, K. Y., Janz, K. F., Zhu, W., & Brychta, R. J. (2012). Redefining the roles of sensors in objective physical activity monitoring. Medicine and Science in Sports and Exercise, 44, S13 – S23. doi: 10.1249/ MSS.0b013e3182399bc8 Church, T. S., Thomas, D. M., Tudor-Locke, C., Katzmarzyk, P. T., Earnest, C. P., Rodarte, R. Q., & . . . Bouchard, C. (2011). Trends over 5 decades in U.S. occupation-related physical activity and their associations with obesity. PLoS.One, 6, e19657. doi: 10.1371/journal. pone.0019657

285

Craig, C. L., Marshall, A. L., Sjostrom, M., Bauman, A. E., Booth, M. L., Ainsworth, B. E., . . . Oja, P. (2003). International physical activity questionnaire: 12-country reliability and validity. Medicine and Science in Sports and Exercise, 35, 1381 – 1395. doi: 10.1249/01. MSS.0000078924.61453.FB Delany, J. P., Schoeller, D. A., Hoyt, R. W., Askew, E. W., & Sharp, M. A. (1989). Field use of D2 18O to measure energy expenditure of soldiers at different energy intakes. Journal of Applied Physiology, 67, 1922–1929. Retrieved from http://jap.physiology.org/content/67/5/1922.long Diggle, P., Heagerty, P. L., Liang, K. Y., & Zeger, S. (2002). Analysis of Longitudinal Data. New York, NY: Oxford University Press. ISBN-10: 0198524846. Dwyer, J., Ellwood, K., Leader, N. P., Moshfegh, A. J., & Johnson, C. L. (2001). Integration of the Continuing Survey of Food Intakes by Individuals and the National Health And Nutrition Examination Survey. Journal of the American Dietetic Association, 101, 1142– 1143. doi: 10.1016/S0002-8223(01)00279-6 Flegal, K. M., Carroll, M. D., Kit, B. K., & Ogden, C. L. (2012). Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999– 2010. Journal of the American Medical Association, 307, 491–497. doi: 10.1001/jama.2012.39 Haslam, D. W., & James, W. P. (2005). Obesity. Lancet, 366(9492), 1197– 1209. Retrieved from http://ncbi.nlm.nih.gov/pubmed/16198769 Heagerty, P. J., & Comstock, B. A. (2013). Exploration of lagged associations using longitudinal data. Biometrics, 69, 197 –205. doi: 10.1111/j.1541-0420.2012.01812.x Hebert, J. R., Ebbeling, C. B., Matthews, C. E., Hurley, T. G., Ma, Y., Druker, S., et al., (2002). Systematic errors in middle-aged women’s estimates of energy intake: Comparing three self-report measures to total energy expenditure from doubly labeled water. Annals of Epidemiology, 12, 577 –586. doi: 10.1016/S1047-2797(01)00297-6 Hebert, J. R., Hurley, T. G., Chiriboga, D. E., & Barone, J. (1998). A comparison of selected nutrient intakes derived from three diet assessment methods used in a low-fat maintenance trial. Public Health Nutrition, 1, 207–214. doi: 10.1079/PHN19980032 Hill, J. O., Wyatt, H. R., & Peters, J. C. (2012). Energy balance and obesity. Circulation, 126, 126 –132. doi: 10.1161/CIRCULATIONAHA.111. 087213 Hubert, P., King, N. A., & Blundell, J. E. (1998). Uncoupling the effects of energy expenditure and energy intake: Appetite response to short-term energy deficit induced by meal omission and physical activity. Appetite, 31, 9 –19. doi: 10.1006/appe.1997.0148 Katzmarzyk, P. T., Perusse, L., Tremblay, A., & Bouchard, C. (2000). No association between resting metabolic rate or respiratory exchange ratio and subsequent changes in body mass and fatness: 5-1/2 year follow-up of the Quebec family study. European Journal of Clinical Nutrition, 54, 610–614. doi: 10.1038/sj/ejcn.1601053 King, N. A., Caudwell, P., Hopkins, M., Byrne, N. M., Colley, R., Hills, A. P., & . . . Blundell, J. E. (2007). Metabolic and behavioral compensatory responses to exercise interventions: barriers to weight loss. Obesity.(Silver.Spring), 15, 1373–1383. doi: 10.1038/oby.2007. 164 King, N. A., Hopkins, M., Caudwell, P., Stubbs, R. J., & Blundell, J. E. (2008). Individual variability following 12 weeks of supervised exercise: identification and characterization of compensation for exercise-induced weight loss. International Journal of Obesity (Lond), 32, 177–184. doi: 10.1038/sj.ijo.0803712 Kipnis, V., Midthune, D., Freedman, L., Bingham, S., Day, N. E., Riboli, E., & . . . Carroll, R. J. (2002). Bias in dietary-report instruments and its implications for nutritional epidemiology. Public Health Nutrition, 5, 915–923. doi: 10.1079/PHN2002383 Kozey-Keadle, S., Libertine, A., Lyden, K., Staudenmayer, J., & Freedson, P. S. (2011). Validation of wearable monitors for assessing sedentary behavior. Medicine and Science in Sports and Exercise, 43, 1561– 1567. doi: 10.1249/MSS.0b013e3182ce174

Downloaded by [Monash University Library] at 11:59 06 October 2014

286

G. A. HAND ET AL.

Marra, M., Scalfi, L., Covino, A., Esposito-Del, P. A., & Contaldo, F. (1998). Fasting respiratory quotient as a predictor of weight changes in non-obese women. International Journal of Obesity and Related Metabolic Disorders, 22, 601 –603. doi: 10.1038/sj.ijo.0800612 McInnis, K. J., & Balady, G. J. (1994). Comparison of submaximal exercise responses using the Bruce vs modified Bruce protocols. Medicine and Science in Sports and Exercise, 26, 103–107. doi: 10.1249/00005768199401000-00017 McNair, D. M., Lorr, M., & Droppleman, L. F. (1992). EdITS Manual for the Profile of Moods States. San Diego, CA: EdITS/Educational and Industrial Testing Service. Retrieved from http://www.parqol.com/page.cfm?id¼91 Novotny, J. A., Rumpler, W. V., Judd, J. T., Riddick, P. H., Rhodes, D., McDowell, M., & Briefel, R. (2001). Diet interviews of subject pairs: how different persons recall eating the same foods. Journal of the American Dietetic Association, 101, 1189–1193. doi: 10.1016/S00028223(01)00291-7 Ogden, C. L., Carroll, M. D., Kit, B. K., & Flegal, K. M. (2012). Prevalence of obesity in the United States. NCHS.Data Brief., 1–8. doi: 10.1001/ jama.2012.40 Owen, N., Healy, G. N., Matthews, C. E., & Dunstan, D. W. (2010). Too much sitting: the population health science of sedentary behavior. Exercise and Sport Science Reviews, 38, 105–113. doi: 10.1097/ JES.0b013e3181e373a2 Pbert, L., Doerfler, L. A., & DeCosimo, D. (1992). An evaluation of the perceived stress scale in two clinical populations. Journal of Psychopathology and Behavioral Assessment, 14, 363 – 375. doi: 10.1007/BF00960780 Posner, B. M., Smigelski, C., Duggal, A., Morgan, J. L., Cobb, J., & Cupples, L. A. (1992). Validation of two-dimensional models for estimation of portion size in nutrition research. Journal of the American Dietetic Association, 92, 738 –741. Retrieved from http://journals.lww. com/epidem/pages/articleviewer.aspx?year¼1992&issue¼030008 Ravussin, E., Lillioja, S., Knowler, W. C., Christin, L., Freymond, D., Abbott, W. G., . . . Bogardus, C. (1988). Reduced rate of energy expenditure as a risk factor for body-weight gain. New England Journal of Medicine, 318, 467 –472. doi: 10.1056/NEJM198802253180802 Sheehan, T. J., DuBrava, S., DeChello, L. M., & Fang, Z. (2003). Rates of weight change for black and white Americans over a twenty year period. International Journal of Obesity and Related Metabolic Disorders, 27, 498–504. doi: 10.1038/sj.ijo.0802263 Sieverdes, J. C., Sui, X., Hand, G. A., Barry, V. W., Wilcox, S., Meriwether, R. A., . . . Blair, S. N. (2011). Cardiometabolic results from

an armband-based weight loss trial. Diabetes Metabolic Syndrome and Obesity, 4, 187– 194. doi: 10.2147/DMSO.S18649 Speakman, J. R. (1998). The history and theory of the doubly labeled water technique. American Journal of Clinical Nutrition, 68, 932S–938S. doi: 10.1098/rstb.1998.0312 Spiegelman, B. M., & Flier, J. S. (2001). Obesity and the regulation of energy balance. Cell, 104, 531–543. doi: 10.1016/S0092-8674(01)00240-9 St Onge, M., Mignault, D., Allison, D. B., & Rabasa-Lhoret, R. (2007). Evaluation of a portable device to measure daily energy expenditure in free-living adults. American Journal of Clinical Nutrition, 85, 742–749. Retrieved from http://ajcn.nutrition.org/content/85//3/742.long Stubbs, R. J., Hughes, D. A., Johnstone, A. M., Whybrow, S., Horgan, G. W., King, N., & Blundell, J. (2004). Rate and extent of compensatory changes in energy intake and expenditure in response to altered exercise and diet composition in humans. American Journal of Physiology (Regulatory Integrative and Comparitive Physiology), 286, R350–R358. doi: 10.1152/ajpregu.00196.2003 Swinburn, B., Sacks, G., & Ravussin, E. (2009). Increased food energy supply is more than sufficient to explain the US epidemic of obesity. American Journal of Clinical Nutrition, 90, 1453– 1456. doi: 10.3945/ ajcn.2009.28595 U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. (2013). Healthy People 2020. Washington, DC: Author. Retrieved from www.healthypeople.gov/2020.aspx Welk, G. J., McClain, J. J., Eisenmann, J. C., & Wickel, E. E. (2007). Field validation of the MTI Actigraph and BodyMedia armband monitor using the IDEEA monitor. Obesity.(Silver.Spring), 15, 918–928. doi: 10.1038/ oby.2007.624 Westerterp, K. R., & Speakman, J. R. (2008). Physical activity energy expenditure has not declined since the 1980s and matches energy expenditures of wild mammals. International Journal of Obesity (Lond), 32, 1256–1263. doi: 10.1038/ijo.2008.74 Whybrow, S., Hughes, D. A., Ritz, P., Johnstone, A. M., Horgan, G. W., King, N., et al., (2008). The effect of an incremental increase in exercise on appetite, eating behaviour and energy balance in lean men and women feeding ad libitum. British Journal of Nutrition, 100, 1109–1115. doi: 10.1017/S0007114508968240 Wilcox, S., Sharpe, P. A., Parra-Medina, D., Granner, M., & Hutto, B. (2011). A randomized trial of a diet and exercise intervention for overweight and obese women from economically disadvantaged neighborhoods: Sisters Taking Action for Real Success (STARS). Contemporary Clinical Trials, 32, 931–945. doi: 10.1016/j.cct.2011.08.003