YPMED-04351; No of Pages 6 Preventive Medicine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
3Q2
Kerem Shuval a,⁎, Binh T. Nguyen a, Amy L. Yaroch b, Jeffrey Drope a, Kelley Pettee Gabriel c
O
F
2
Accelerometer determined sedentary behavior and dietary quality among US adults
4 5 6
a
7
a r t i c l e
8 9
Available online xxxx
10 Q4 11 12
Keywords: Sedentary behavior Dietary quality
Intramural Research Department, American Cancer Society, Atlanta, GA 30303, USA The Gretchen Swanson Center for Nutrition, Omaha, NE 68114, USA Divisions of Epidemiology, Human Genetics and Environmental Sciences, School of Public Health, Austin Regional Campus, University of Texas Health Science Center in Houston, Austin, TX 78701, USA
i n f o
a b s t r a c t
Objective. Scant evidence exists pertaining to objectively measured sedentary time and dietary quality among adults. Therefore, we examined the relationships between sedentary time, physical activity, and dietary quality. Methods. Cross-sectional analyses of a 4,910 US adults from two cycles (2003–2006) of the National Health and Nutrition Examination Survey. The primary independent variables were sedentary time and physical activity (continuous and categorical), while the outcomes were overall dietary quality (Healthy Eating Index (HEI) 2010), fruit and vegetable scores, and empty caloric intake (kcal). Results. Multivariable analyses revealed that a 1 min increase in daily sedentary behavior was associated with a 0.2 kcal decrease in empty calories (−0.18, 95% CI = −0.34, −0.03); however, sedentary time was not significantly related to overall dietary quality (HEI) and fruit and vegetable intake. In comparison, a 1 min increase in daily moderate-to-vigorous intensity physical activity was related to a 0.1 higher HEI score (0.08, 95% CI = 0.04, 0.11), a 0.01 higher fruit score (0.01, 95% CI = 0.01, 0.02), and conversely a 1.3 kcal decrease in empty calories (−1.35, 95% CI = −2.01, −0.69). In addition, meeting physical activity guidelines was associated with a 2.8 point higher HEI score (2.82, 95% CI = 1.40, 4.25), a 0.5 point higher fruit score (0.51, 95% CI = 0.31–0.71), and 37.4 fewer empty calories (−37.43, 95% CI = −64.86, −9.10). Conclusions. Physical activity is significantly related to better overall dietary quality, while sedentary behavior is not. Findings suggest the need to promote physical activity and encourage adherence to dietary guidelines jointly, whereas sedentary behavior and overall dietary quality might need to be targeted independently. © 2015 Published by Elsevier Inc.
R
E
C
T
E
D
c
P
b
R O
1Q1
38 39 40 41 42 43 44 45 46 47 48 49 50
Sedentary behavior is a relatively recent focus of scientific exploration and the accumulating evidence indicates that excessive time spent TV viewing, commuting in a motorized vehicle, or sitting at home or on the job is linked to adverse health outcomes (Owen et al., 2010; Shuval et al., 2014). Specifically, observational studies have found that prolonged sedentary time is linked to an increased risk for obesity, type 2 diabetes mellitus, and premature death, even while considering the protective effects of physical activity (Gardiner et al., 2011; Healy et al., 2011). While the evidence on the potential deleterious effects of sedentary time is accumulating, the evidence base pertaining to physical activity and health is well established, with observed dose–response relationships between higher frequency and intensity levels of activity and reduced risk for obesity, chronic conditions (e.g. depression), and premature mortality (U.S. Department of Health and Human Services, 2008). Recent physical
N C O
37
Introduction
U
34 36
R
35 33 32
⁎ Corresponding author at: Economics and Health Policy Research, Intramural Research Department, American Cancer Society, 250 Williams Street, NE Atlanta, GA 30303, USA. Fax: +1 404 327 6450. E-mail address: [email protected] (K. Shuval).
activity guidelines reflect this evidence and underscore the importance of reducing overall sedentary time and regularly engaging in physical activity to achieve health benefits (Kushi et al., 2012; U.S. Department of Health and Human Services, 2008). With regard to diet, the 2010 Dietary Guidelines for Americans recommend a healthful diet consisting of nutrient-rich foods to enhance caloric balance for weight management (U.S. Department of Health and Human Services, 2010). A nutrient-rich diet refers to foods and beverages with high nutritional quality, such as fruits, vegetables, and whole grains. However, Americans tend to consume more energy-dense, nutrient-poor foods such as those containing an excessive amount of solid fats, added sugar, and refined grain products (U.S. Department of Health and Human Services, 2010). As with a physically active lifestyle, consuming a more healthful diet significantly reduces the risk for obesity and numerous chronic diseases (World Health Organization, 2003). Observational and interventional research often focus on these two modifiable health behaviors as distinct from one another, although these behaviors are likely inter-related and, when combined, may provide additional health benefits (Diabetes Prevention Program Research Group, 2002). Previous research has found that less than a tenth of Americans meet physical activity guidelines (Troiano et al., 2008), that ~8 h are spent in sedentary
http://dx.doi.org/10.1016/j.ypmed.2015.06.010 0091-7435/© 2015 Published by Elsevier Inc.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
72 73
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
ambulatory participants during waking hours. Participants were instructed to wear the accelerometer for 7 consecutive days on the right hip. Following the data collection period, participants returned the accelerometer by mail, and then data were downloaded and prepared for further processing (Matthews et al., 2008). For the current study, Freedson accelerometer count threshold values were used to characterize accelerometer counts as time (min/day) spent sedentary [i.e. b100 counts per minute (cpm)], light (100–1951 cpm), and in moderate to vigorous intensity physical activity (≥1952 cpm) (Freedson et al., 1998). Weekly summary averages were computed only for participants that wore the monitor ≥10 h per day on ≥4 of 7 days (Mâsse et al., 2005). In addition, daily summary averages were computed for participants. For analysis, participants were classified as meeting physical activity guidelines (yes/no) if accumulating ≥150 min a week of moderate to vigorous physical activity, and categorized into sample-specific quartiles of sedentary time (quartile cut-points: 6.8, 8.1, 9.4 h/day). Furthermore, continuous measures of sedentary time and physical activity (min/day) were examined. Accelerometer wear time was taken into account in all multivariable analyses (see Statistical analysis).
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
O
2
153
104
Methods
105
Data and participants
106 107 108 109 Q5 110 111 112 113 114 Q6 115 116 117 118 119 120 121 122 123 124 125 Q7 126
To examine the association between sedentary behavior and physical activity in relation to dietary quality, we utilized cross-sectional data from the National Health and Nutrition Examination Survey (NHANES). The NHANES, elaborately described elsewhere (Centers for Disease Control and Prevention, n.d.-a,b,c,d), is an ongoing population-based survey (including both interviews and physical examinations) which aims to determine the nutrition and health status of the US population. The NHANES sample is nationally representative of the US noninstitutional population, and utilizes a complex multistage probability sampling design obtaining data in 2-year cycles (Centers for Disease Control and Prevention, n.d.-a,b,c,d). In the current study, we examine data from two cycles (2003–2004; 2005–2006), which contain objectively measured sedentary time and physical activity (assessed via accelerometers), and detailed dietary information (gleaned from 24-h dietary recall data). The current analysis consisted of an adult sample of men and women aged ≥20 years who were not taking insulin, and not pregnant or lactating (n = 9,231). Of these, 3,509 observations were excluded due not having valid accelerometer data, and 812 more observations were omitted due to missing information on measures, leaving an analytic sample of 4,910 for the current study. The NHANES study protocols received approval from the National Center for Health Statistics Research Ethics Review Board; participants provided written informed consent prior to data collection (Centers for Disease Control and Prevention, n.d.-a,b,c,d).
127
Primary measures
Statistical analysis
191
128 129 130 131 132 Q8 133 Q9 134 135
Physical activity and sedentary behavior (independent variables) Physical activity and sedentary behavior were derived from accelerometer measurements and were processed based on publically available SAS programs developed by the National Cancer Institute (NCI) for processing NHANES accelerometer data (National Cancer Institute, n.d.-a,b). Details on the accelerometer protocol appear elsewhere (Centers for Disease Control and Prevention, n.d.-a,b, c,d). Briefly, uniaxial accelerometers (Actigraph model 7164), a small device that measures accelerations as activity counts (1-min epochs), were worn by
Descriptive statistics were utilized to describe participants’ sociodemographics and health behaviors. The association between both quartiles of sedentary time (first quartile- reference) and meeting moderate to vigorous physical activity guidelines (not meeting guidelines-reference) to dietary outcomes was examined using ordinary least squares (OLS) regression; both in partially adjusted (age, sex, BMI) and fully adjusted models (age, sex, race/ethnicity, marital status, education, PIR, health
192 193
97 98 99 100 101 102
178
Covariates included in the multivariable analysis consisted of participants’ sex, age (20–39.9, 40–59.9, ≥ 60 years), race/ethnicity (African American, Hispanic, White, other), marital status (married, widowed, divorced or separated, never married), educational attainment (less than high school, high school, some college, college and above), health status (poor, fair, good, very good, excellent), and poverty-to- income (PIR) ratio (continuous), and total caloric intake (kcal-continuous). In addition, current smoking (no/yes) was determined based on objectively measured cotinine levels; i.e. a 3.08 ng/mL cotinine level adhering to the cut-point suggested by Benowitz (Benowitz et al., 2009). Body mass index (BMI) was computed based on the standard formula (kg/m2) (World Health Organization).
179 180
R O
Covariates
P
95 96
154 155 156 157 158 Q10 159 160 161 162 163 164 165 166 167 168 Q11 169 170 Q12 171 172 173 174 175 176 177
D
93 94
Dietary quality was based on the HEI-2010, which was created by the United States Department of Agriculture to assess adherence to the primary components of the 2010 Dietary Guidelines for Americans and to determine the quality of dietary intake and patterns of consumption (Guenther et al., 2013a). Details on the components of the HEI are elaborated elsewhere (National Cancer Institute, n.d.-a,b); however, briefly the HEI score (0–100 points) was composed of the following 12 components, of which each contributed to the total dietary quality score: total fruit (5 points), whole fruit (5 points), total vegetables (5 points), greens and beans (5 points), whole grains (10 points), dairy products (10 points), total protein (5 points), seafood and proteins from plants (5 points), fatty acids (10 points), refined grains (10 points), sodium (10 points), and empty calories (20 points) (Guenther et al., 2013b). The HEI score was computed using individual 24-h dietary recall data, MyPyramid Equivalents Database (MPED) (version 2.0), CNPP MPED for whole fruit and fruit juice, and the CNPP addendum to the MPED (version 2.0B) (National Cancer Institute, n.d.-a,b). The National Cancer Institute methodology, elaborated elsewhere (Freedman et al., 2008; Guenther et al., 2013b; National Cancer Institute, n.d.-a,b), was used to calculate the HEI total score and components as well as the corresponding standard errors and confidence interval from the self-reported 24-h dietary recall data in NHANES (US Department of Agriculture). For analysis, along with the total HEI score as a primary outcome, we focused on a few specific components which are related to more healthful food choices (i.e. total vegetable score, total fruit score), and conversely on less healthful food choices: empty calories (i.e. calories (kcal) from solid fats, added sugars, and excessive alcohol intake).
E
91 92
Dietary quality and other dietary variables (dependent variables)
T
89 90
C
87 88
E
85 86
R
83 84
R
81 82
O
79 80
C
78
N
76 77
U
74 75
F
103
time daily during waking hours (Matthews et al., 2008), and that dietary quality of Americans is poor, resulting in 53.5/100 dietary quality score (based on Healthy Index Score) (Guenther et al., 2013a). Thus, despite the health benefits associated with reducing sedentary time, increasing physical activity, and consuming a healthy diet, Americans mostly do not adhere to these behaviors. The inter-relationship among multiple health behaviors (e.g. physical activity, diet, smoking, and alcohol consumption) has been previously explored (Berrigan et al., 2003; King et al., 2009; Reeves and Rafferty, 2005; Schuit et al., 2002). For example, Schuit found in a cross-sectional study that approximately 20% of their sample had at least three lifestyle risk factors (e.g. smoking, physical inactivity), whereas Reeves and Rafferty found that only 3% of US adults adhered to all the healthy lifestyle behaviors they examined: not smoking, normal weight, sufficient fruit and vegetable intake, and meeting physical activity guidelines. In fact, King observed a 7% decline (from 15% to 8%) over 18 years in adherence to healthy lifestyle behaviors (King et al., 2009). However, the relationship between sedentary behavior and dietary quality has not sufficiently been explored among adults. Specifically, systematic reviews on this topic concluded that sedentary behavior is related to unhealthy dietary intake, primarily among children and adolescents; however, the evidence for adults is inconclusive (Hobbs et al., 2014; Pearson and Biddle, 2011). Additionally, the studies in the systematic reviews mostly relied on self-reported screen time or TV viewing (Hobbs et al., 2014), which are only proxies of overall sedentary time and are subject to recall bias and subsequent measurement error (Crosby et al., 2006). Therefore, in the current study, we utilize accelerometers, an objective measure of time spent in sedentary behavior to more accurately determine the exposure. Specifically, we aimed to examine the association between both sedentary behavior and physical activity (via accelerometry) and dietary quality (Health Eating Index (HEI), 2010) among a national sample of US adults.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
181 182 183 184 185 186 187 188 189 190
194 195 196 197 198
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Discussion
250 251
Currently, a dearth of evidence exists as to whether excessive sedentary time is related to dietary quality among a national sample of US adults. Each of these health behaviors is an independent risk factor for chronic disease morbidity and mortality, thus there is a need for substantiation on whether these behaviors should be targeted independently or jointly. In the current study, we aimed to examine the association of sedentary time and physical activity (both measured objectively) with dietary quality (HEI), more healthful (fruits and vegetables) and less healthful (empty calories) dietary intake in adults. Results reveal that prolonged sedentary time was not related to overall dietary quality and more healthful food choices, whereas excessive sedentary behavior
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
252 253 254 255 256 257 258 259 260
C
223 224
E
221 222
R
219 220
R
218
N C O
212 213
U
210 211
F
249
208 209
O
247 248
Participants’ baseline characteristics are presented in Table 1. Briefly, half of the sample were women, and, on average, participants were aged 48.5 years (SE = 0.5). Most (75.7%) were non-Hispanic white, 9.1% were non-Hispanic blacks, 10.4% were Hispanic, and 59.4% had some or complete college education. In addition, 48.3% reported very good or excellent health, 26.6% were current smokers, 31.1% were of normal weight, and participants’ mean dietary quality score was 47.6/100 (SE = 0.4). Participants spent 486.3 min (SE = 2.0) per day sedentary (or 8.1 h daily), in comparison to only 23.9 min (SE = 0.6) per day participating in moderate-to-vigorous intensity physical activity. The relationship between sedentary behavior, physical activity, and dietary quality is depicted in Table 2 (partially adjusted models) and Table 3 (fully adjusted models). In the fully adjusted models, when examining these relationships continuously, more minutes per day spent in sedentary time was not significantly associated with overall dietary quality (HEI) and the fruit and vegetables scores; however, more sedentary time was significantly associated with a lower intake of empty calories (p b 0.02). Specifically, a 1-min per day increase in sedentary time was associated with a 0.2 kcal decrease in the empty calories (− 0.18, 95% CI = − 0.34, − 0.03). In comparison, a 1-min increase in moderate to vigorous physical activity per day was significantly related to a 0.1 higher HEI score (0.08, 95% CI = 0.04, 0.11), a 0.01 higher fruit score (0.01, 95% CI = 0.01, 0.02), and conversely a 1.3 kcal lower intake of empty calories (− 1.35, 95% CI = − 2.01, −0.69). When examining these relationships categorically, comparing the fourth quartile of sedentary time (≥ 9.4 h daily) to the reference group (first quartile) in relation to dietary quality resulted in nonsignificant findings (Table 3). In comparison, meeting physical activity guidelines was associated with a 2.8 point higher HEI score (2.82, 95% CI = 1.40, 4.25), a 0.5 point higher fruit score (0.51, 95% CI = 0.31–0.71), and 37.4 kcal fewer empty calories (− 37.43, 95% CI = − 64.86, − 9.10).
206 207
R O
217
205
Sex Women Age groups (years) 20–39.9 40–59.9 ≥60 Race/ethnicity Non-Hispanic White Non-Hispanic Black Hispanic Other Marital status Married Widow Divorced/separated Never married Education bHigh school High school Some college ≥College Poverty-to-income ratio (PIR)—mean (SE) BMI (kg/m2) categories Underweight (b18.5) Normal weight (18.5–24.9) Overweight (25.0–29.9) Obese (≥30.0) Self-reported health status Excellent Very good Good Fair Poor Current smokingc HEI-2010 total scored—mean (SE) Total vegetables—mean (SE) Total fruit—mean (SE) Added sugar (teaspoon)—mean (SE) Solid fat (grams)—mean (SE) Empty calories (kcal)—mean (SE) Total calories (kcal)—mean (SE) Sedentary time, minutes per daye—mean (SE) Light physical activity, minutes per daye—mean (SE) MVPA, minutes per daye—mean (SE) Meeting physical activity guidelinese
P
Results
203 204
% (n)b
Characteristics
D
216
201 202
Table 1 t1:1 Descriptive characteristics of sample (population weighted)a: NHANES 2003–2006 t1:2 (n = 4,910). t1:3
T
214 215
status, smoking, BMI, total caloric intake); all adjusting for accelerometer wear time. In addition, continuous measures of sedentary time and physical activity (min/day) were entered into separate models: (1) partially adjusted models: age, sex, BMI, accelerometer wear time; and (2) fully adjusted models: age, sex, race/ethnicity, marital status, education, PIR, health status, smoking, BMI, total caloric intake, light intensity physical activity, and accelerometer wear time. The OLS models were computed when examining the association between sedentary behavior and physical activity (categorical or continuous variables) with each dietary outcome (i.e. HEI, vegetables, fruit, and empty calories). SAS (version 9.3) statistical software was utilized to compute the HEI score and accelerometer estimates. Stata version 13 (StataCorp) was used for descriptive and analytical analyses to take into account NHANES’s complex multistage sampling design to allow for nationally representativeness. Since dietary quality was the primary outcomes (as derived from the Mobile Examination Centers (MEC) data), MEC sample weights were utilized in the analyses. An alpha of b0.05 was considered statistically significant.
E
199 200
3
50.0% (2,330) 31.6% (1,341) 43.0% (1,631) 25.4% (1,938) 75.7% (2,712) 9.1% (912) 10.4% (1,109) 4.8% (177) 68.4% (3,161) 6.3% (483) 12.2% (629) 13.1% (637) 15.1% (1,271) 25.6% (1,202) 32.3% (1,397) 27.1% (1,040) 3.20 (1.53) 1.5% (62) 31.1% (1,427) 35.3% (1,813) 32.2% (1,608) 12.3% (510) 36.0% (1,492) 36.6% (1,871) 12.8% (887) 2.3% (150) 26.6% (1,214) 47.61 (0.39) 3.11 (0.04) 2.15 (0.07) 18.96 (0.41) 434.83 (5.65) 773.56 (10.12) 2244.78 (16.39) 486.26 (1.98) 259.95 (1.46) 23.90 (0.59) 8% (365)
t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17 t1:18 t1:19 t1:20 t1:21 t1:22 t1:23 t1:24 t1:25 t1:26 t1:27 t1:28 t1:29 t1:30 t1:31 t1:32 t1:33 t1:34 t1:35 t1:36 t1:37 t1:38 t1:39 t1:40 t1:41 t1:42 t1:43 t1:44 t1:45 t1:46 t1:47 t1:48 t1:49
NHANES, National Health and Nutrition Examination Survey; SE, standard error; kcal, kilocalories; BMI, body mass index, HEI, Healthy Eating Index; MVPA, moderate to vigorous intensity physical activity. a NHANES survey weights are accounted for in the table. b The number and percentage is presented unless otherwise specified. c Current smoking was determined based on cotinine cut-point (3.08 ng/mL) suggested by Benowitz (Benowitz et al., 2009) d The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The empty calories score is 0–20 points and the total fruit and total vegetable scores are each 0–5 points. e Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minutes (cmp)), light-intensity physical activity (100–1951 cpm), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week.
t1:50 t1:51 t1:52 t1:53 t1:54 t1:55 t1:56 t1:57 t1:58 t1:59 t1:60 t1:61 t1:62 t1:63 t1:64 t1:65 t1:66
was significantly and inversely associated with less healthful dietary intake. Engaging in physical activity, in comparison, was significantly related to more healthful food choices (fruit intake), and decreased consumption of less healthful foods. Our results pertaining to objectively measured sedentary time and diet among adults are unique and have rarely been explored, since previous studies have primarily relied on reported sedentary time predominantly among children and adolescents. Indeed, a recent systematic review by Hobbs et al. noted that while excessive sedentary behavior
261
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
262 263 264 265 266 267 268 269
4 t2:1 t2:2
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Table 2 Association between sedentary time, physical activitya and dietary qualityb: partially adjusted modelsc: NHANES 2003–2006.
t2:3 t2:4 t2:5 t2:6 t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14
HEI-2010
Vegetables
Mean (SE) 95% CI Continuous independent variables Sedentary time (min/day)d 0.00 (0.00) −0.00, 0.01 Physical activity (min/day)d 0.13 (0.02)** 0.10, 0.17 Categorical independent variables Sedentary time—Quartiles (first quartile reference)e Second quartile 0.95 (0.61) −0.25, 2.14 Third quartile 1.88 (0.62)** 0.67, 3.09 Fourth quartile 1.51 (0.66)* 0.22, 2.81 Physical activity (not meeting guidelines-reference)e Meeting guidelines 5.14 (0.73)** 3.70, 6.58
Fruit
Empty Calories
Mean (SE)
95% CI
Mean (SE)
95% CI
Mean (SE)
95% CI
0.00 (0.00) 0.00 (0.00)
−0.00, 0.00 −0.00, 0.01
0.00 (0.00) 0.02 (0.00)**
−0.00, 0.00 0.02, 0.03
−0.51 (0.13)** −2.05** (0.56)
−0.77, −0.25 −3.15, −0.96
0.08 (0.07) 0.17 (0.07)* 0.07 (0.08)
−0.06, 0.22 0.03, 0.31 −0.08, 0.22
0.09 (0.09) 0.17 (0.09) 0.11 (0.09)
−0.08, 0.26 −0.00, 0.34 −0.69, 0.30
−25.50 (19.47) −77.62 (19.61)** −109.34 (21.09)**
−63.68, 12.67 −116.08, −39.16 −150.68, −68.00
0.04 (0.08)
−0.13, 0.20
0.78 (0.10)**
0.58, 0.99
−85.97 (23.32)**
−131.68, −40.25
NHANES, National Health and Nutrition Examination Survey; SE, standard error; HEI, Healthy Eating Index. *p b .05, **p b .01. a Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minute (cmp)), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week. b The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The total fruit and total vegetable scores are each 0–5 points, whereas empty calories are kcals from solid fats, added sugars, and excessive alcohol intake. c Separate OLS regression models were constructed for the continuous and the categorical sedentary time and physical activity measures in relation to each dietary outcome. d The continuous sedentary time (minutes per day) and physical activity (minutes per day in moderate to vigorous intensity physical activity) variables were entered jointly into each regression model while adjusting for age, sex, BMI, light intensity physical activity, and accelerometer wear time. e The categorical sedentary time (quartiles) and physical activity (meeting guidelines) variables were entered jointly into each regression model while adjusting for age, sex, BMI, and accelerometer wear time.
270 271
282 283
(primarily TV viewing) was related to lower fruit and vegetable intake and greater consumption of energy-dense, nutrient-poor foods among children and adolescents, the evidence for adults was insufficient to determine a clear relationship (Hobbs et al., 2014). TV viewing has been used as a proxy for sedentary behavior in numerous studies (Shuval et al., 2013, 2014); however, it is not a direct measure of sitting and individuals might “multi-task” (i.e. perform other activities) while watching TV (Marsh et al., 2013). Thus, the link between TV viewing and a less healthful diet found previously might stem from exposure to advertisements of energy-dense foods and ‘distracted eating’ rather than the activity of sitting itself (Giese et al., 2014; Marsh et al., 2013). ‘Distracted eating’ refers to the dominance of the automatic system (over the reflective one) when making decisions pertaining to eating during TV viewing; that is, people are more likely to make emotional and
t3:1 t3:2
Table 3 Association between sedentary time, physical activitya and dietary qualityb: fully adjusted modelsc: NHANES 2003–2006.
t3:15 t3:16 t3:17 t3:18 t3:19 t3:20 t3:21 t3:22 t3:23 t3:24 t3:25 t3:26 t3:27
O
P
R O
t3:3
HEI-2010
N
C
Mean (SE) 95% CI Continuous independent variables Sedentary time (min/day)d 0.00 (0.00) −0.00, 0.01 Physical Activity (min/day)d 0.08** (0.02) 0.04, 0.11 Categorical independent variables Sedentary time (first quartile reference)e Second quartile 0.85 (0.60)** −1.10, 1.27 Third quartile 0.63 (0.62) −0.58, 1.84 Fourth quartile 0.07 (0.67) −1.25, 1.39 Physical activity (not meeting guidelines-reference)e Meeting guidelines 2.82 (0.73)** 1.40, 4.25
U
t3:4 t3:5 t3:6 t3:7 t3:8 t3:9 t3:10 t3:11 t3:12 t3:13 t3:14
D
E
T
C
280 281
E
278 279
R
276 277
uncontrolled decisions about eating and are less exposed to cues that would limit food intake when distracted by screen time (Kahneman, 2003; Marsh et al., 2013; Shaikh et al., 2008). However, in the current study, the targeted exposure was not limited to TV viewing, but rather focused on accumulated sedentary time across all waking hours and its relationship with dietary intake. Notably, our results point to a null association between sedentary time and overall dietary quality and a significant and inverse association between prolonged sedentary time and the consumption of less healthful foods (i.e. empty calories). While the underlying mechanism explaining this finding is unclear and warrants further investigation, it is quite plausible that the activity of sitting itself (which involves low energy expenditure) might enable the reflective system to be more dominant, resulting in more rational decisions (Kahneman, 2003; Shaikh et al., 2008). Consequently, individuals
R
274 275
O
272 273
F
t2:15 t2:16 t2:17 t2:18 t2:19 t2:20 t2:21 t2:22 t2:23 t2:24 t2:25 t2:26
Vegetables
Fruit
Empty Calories
Mean (SE)
95% CI
Mean (SE)
95% CI
Mean (SE)
95% CI
0.00 (0.00) 0.00 (0.00)
−0.00, 0.00 −0.00, 0.00
0.00 (0.00) 0.01** (0.00)
−0.00, 0.00 0.01, 0.02
−0.18 (0.08)* −1.35 (0.34)**
−0.34, −0.03 −2.01, −0.69
0.01 (0.07) 0.09 (0.07) −0.02 (0.08)
−0.13, 0.15 −0.05, 0.23 −0.18, 0.13
0.02 (0.09) 0.06 (0.09) −0.03 (0.10)
−0.15, 0.19 −0.11, 0.23 −0.21, 0.16
−3.41 (11.62) −23.52* (11.91) −25.39 (12.97)
−26.21, 19.39 −46.87, −0.18 −50.81, 0.04
−0.06 (0.08)
−0.22, 0.11
0.51 (0.10)**
0.31, 0.71
−37.43** (13.99)
−64.86, −9.10
NHANES, National Health and Nutrition Examination Survey; SE, standard error; HEI, healthy eating index. *p b .05, **p b .01. a Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minutes (cmp)), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week. b The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The total fruit and total vegetable scores are each 0–5 points, whereas empty calories are kcals from solid fats, added sugars and excessive alcohol intake. c Separate OLS regression models were constructed for the continuous and the categorical sedentary time and physical activity measures in relation to each dietary outcome. d The continuous sedentary time (minutes per day) and physical activity (minutes per day in moderate to vigorous intensity physical activity) variables were entered jointly into each multivariable model while adjusting for age, sex, race/ethnicity, marital status, education, PIR, self-reported health status, current smoking, BMI, caloric intake, light intensity physical activity, and accelerometer wear time. e The categorical sedentary time (quartiles) and physical activity (meeting guidelines) variables were entered jointly into each multivariable model while adjusting for age, sex, race/ethnicity, marital status, education, PIR, self-reported health status, current smoking, BMI, caloric intake, and accelerometer wear time.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
284 285 286 287 288 289 290 291 292 293 294 295 296 297
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
N C O
R
R
E
C
Conclusions
348
Study findings reveal that excessive sedentary time was not associated with overall dietary quality. In comparison, engaging in health-promoting physical activity was significantly related to consuming a diet of higher quality consisting of more fruits and fewer empty calories. These findings underscore the need to design intervention programs focusing on increasing and maintaining physical activity levels while encouraging adherence to dietary guidelines among adults. In comparison, reducing and breaking up sedentary time might need to be targeted independently.
349 350 351 352 353 354 355 356
U
347
357 Q15 Conflict of interest 358 The author has no conflict of interest.
References
359
Benowitz, N.L., Bernert, J.T., Caraballo, R.S., Holiday, D.B., Wang, J., 2009. Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different racial/ethnic groups in the United States between 1999 and 2004. Am. J. Epidemiol. 169, 236–248. Berrigan, D., Dodd, K., Troiano, R.P., Krebs-Smith, S.M., Barbash, R.B., 2003. Patterns of health behavior in U.S. adults. Prev. Med. (Baltim.) 36, 615–623. Centers for Disease Control and Prevention, a. National Health and Nutrition Examination Survey [WWW Document] URL, http://www.cdc.gov/nchs/nhanes.htm (accessed 5.1.15). Centers for Disease Control and Prevention, b. Key Concepts About the NHANES Survey Design [WWW Document] URL http://www.cdc.gov/nchs/tutorials/ dietary/SurveyOrientation/SurveyDesign/Info1.htm. Centers for Disease Control and Prevention, c. National Health and Nutrition Examination Survey—Research Ethics Review Board Approval [WWW Document] URL, http:// www.cdc.gov/nchs/nhanes/irba98.htm (accessed 4.12.14). Centers for Disease Control and Prevention, d. NHANES 2003–2004 Physical Activity Monitor Data Documentation, Codebook, and Frequencies [WWW Document] URL, http:// www.cdc.gov/nchs/nhanes/nhanes2003-2004/PAXRAW_C.htm (accessed 1.8.14). Crosby, R.A., DiClemente, R.J., Salazar, L.F., 2006. Research Methods in Health Promotion. Jossey-Bass. Diabetes Prevention Program Research Group, 2002. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N. Engl. J. Med. 346, 393–403. Freedman, L.S., Guenther, P.M., Krebs-smith, S.M., Kott, P.S., 2008. A population’s mean Healthy Eating Index-2005 scores are best estimated by the score of the population ratio when one 24-hour recall is available. J. Nutr. 138, 1725–1729. Freedson, P.S., Melanson, E., Sirard, J., 1998. Calibration of the Computer Science and Applications, Inc. accelerometer. Med. Sci. Sports Exerc. 30, 777–781. Gardiner, P.A., Healy, G.N., Eakin, E.G., et al., 2011. Associations between television viewing time and overall sitting time with the metabolic syndrome in older men and women: the Australian Diabetes, Obesity and Lifestyle study. J. Am. Geriatr. Soc. 59, 788–796. Giese, H., König, L., Tăut, D., et al., 2014. Exploring the association between television advertising of healthy and unhealthy foods, self-control, and food intake in three European countries. Appl. Psychol. Health. (Well Being Epub ahead). Gillman, M.W., Pinto, B.M., Tennstedt, S., Glanz, K., Marcus, B., Friedman, R.H., 2001. Relationships of physical activity with dietary behaviors among adults. Prev. Med. (Baltim.) 32, 295–301. Guenther, P.M., Casavale, K.O., Kirkpatrick, S.I., et al., 2013a. Diet Quality of Americans in 2001-02 and 2007-08 as Measured by the Healthy Eating Index-2010. Guenther, P.M., Casavale, K.O., Reedy, J., et al., 2013b. Update of the Healthy Eating Index: HEI-2010. J. Acad. Nutr. Diet. 113, 569–580. Haskell, W.L., 2012. Physical activity by self-report: A brief history and future issues. J. Phys. Act. Health 9, 5–10. Healy, G.N., Matthews, C.E., Dunstan, D.W., Winkler, E. a H., Owen, N., 2011. Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06. Eur. Heart J. 32, 590–597. Hobbs, M., Pearson, N., Foster, P.J., Biddle, S.J.H., 2014. Sedentary behaviour and diet across the lifespan: An updated systematic review. Br. J. Sports Med. 1–11. Joseph, R.J., Alonso-Alonso, M., Bond, D.S., Pascual-Leone, A., Blackburn, G.L., 2011. The neurocognitive connection between physical activity and eating behaviour. Obes. Rev. 12, 800–812. Kahneman, D., 2003. A perspective on judgment and choice: Mapping bounded rationality. Am. Psychol. 58, 697–720. King, D.E., Mainous, A.G., Carnemolla, M., Everett, C.J., 2009. Adherence to healthy lifestyle habits in US adults, 1988-2006. Am. J. Med. 122, 528–534. Kushi, L.H., Doyle, C., McCullough, M., et al., 2012. American Cancer Society Guidelines on Nutrition and Physical Activity for Cancer Prevention Reducing the Risk of Cancer with Healthy Food Choices and Physical Activity. CA Cancer J. Clin. 62, 30–67. Loprinzi, P.D., Smit, E., Mahoney, S., 2014. Physical activity and dietary behavior in US adults and their combined influence on health. Mayo Clin. Proc. 89, 190–198. Ma, Y., Olendzki, B.C., Pagoto, S.L., et al., 2010. Number of 24-hour diet recalls needed to estimate energy intake. Ann. Epidemiol. 19, 553–559. Marsh, S., Mjurchu, C.N., Maddison, R., 2013. The non-advertising effects of screen-based sedentary activities on acute eating behaviours in children, adolescents, and young adults. A systematic review. Appetite 71, 259–273. Mâsse, L., Fuemmeler, B., Anderson, C., et al., 2005. Accelerometer data reduction: A comparison of four reduction algorithms on select outcome variables. Med. Sci. Sports Exerc. 37, S544–S554. Matthews, C.E., Chen, K.Y., Freedson, P.S., et al., 2008. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am. J. Epidemiol. 167, 875–881. National Cancer Institute, a. SAS Programs for Analyzing NHANES 2003-2004 Accelerometer Data [WWW Document] URL, http://appliedresearch.cancer.gov/nhanes_pam/ (accessed 3.10.14a). National Cancer Institute, b. Healthy Eating Index [WWW Document] URL http:// appliedresearch.cancer.gov/hei/. Owen, N., Sparling, P.B., Healy, G.N., Dunstan, D.W., Matthews, C.E., 2010. Sedentary behavior: Emerging evidence for a new health risk. Mayo Clin. Proc. 85, 1138–1141. Pearson, N., Biddle, S.J.H., 2011. Sedentary behavior and dietary intake in children, adolescents, and adults. A systematic review. Am. J. Prev. Med. 41, 178–188. Reeves, M.J., Rafferty, A.P., 2005. Healthy lifestyle characteristics among adults in the United States, 2000. Arch. Intern. Med. 165, 854–857. Schuit, A.J., van Loon, A.J.M., Tijhuis, M., Ocke, M.C., 2002. Clustering of lifestyle risk factors in a general adult population. Prev. Med. (Baltim.) 35, 219–224.
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 Q14 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
O
R O
P
D
E
T
might be more cognizant of the fact that they are predominantly sedentary throughout the day and avoid less healthful foods to 300 achieve health benefits. This supposition, however, necessitates fur301 ther exploration. 302 In comparison, our findings on physical activity and diet confirm a 303 number of previous studies that have primarily observed that engaging 304 in physical activity is associated with dietary intake and quality, and that 305 there is a reinforcing inter-relationship between the two (Joseph et al., 306 2011). Specifically, Gillman, in a cross-sectional survey of ethnically 307 diverse adults, found that higher amounts of self-reported physical 308 activity were correlated with an overall higher quality diet (Gillman 309 et al., 2001). Similar to the present study, Loprinzi observed a significant 310 relationship between objectively measured physical activity and dietary 311 quality; however, they did not explore other dietary outcomes (e.g. fruit 312 and vegetables, empty calories) and did not use the latest HEI-2010, 313 which is a more accurate representation of adherence to the 2010 314 Dietary Guidelines (Loprinzi et al., 2014; U.S. Department of Health 315 and Human Services, 2010). Thus, the current findings, coupled with 316 previous evidence, emphasize the inter-relationship between physical 317 activity and healthy eating. A review by Joseph suggests that engaging 318 in physical activity could affect one’s ability to self-regulate eating 319 behavior and minimize hedonic hunger through increased self-efficacy, 320 and enhanced executive function as well as physiologically increasing 321 the sensitivity of satiety signaling (Joseph et al., 2011). 322 The current study has several strengths and limitations. The primary 323 strengths include the novelty of the study question, the national sample, 324 and the objective measures (i.e. accelerometry) utilized to determine 325 the exposure. Specifically, while some studies have explored the rela326 tionship between physical activity and diet, none (to our knowledge) 327 have examined the relationship between objectively measured seden328 tary time and dietary intake among adults. Although accelerometers 329 are an objective measure of activity, they do not capture all forms of 330 physical activity (e.g. cycling, swimming), and are not able to discern 331 the specific type sedentary behavior performed. Furthermore, monitors 332 were worn for 4–7 days and thus might not be representative of longer333 term habitual activity levels. Thus, self-reported activity coupled with 334 accelerometer-determined activity might provide a more comprehen335 sive measure of physical activity (Haskell, 2012). Similarly, nutrition 336 information was based on 24-h dietary recall data, which is subject to 337 over- or under-estimation of portion size and might not be indicative 338 of longer-term dietary behavior (Crosby et al., 2006; Ma et al., 2010). 339 Specifically, the use of the HEI in the present study to determine dietary 340 quality on the individual level might generate measurement error. 341 Therefore, adjustment for energy intake is suggested to reduce bias 342 when examining the relationship between individuals’ dietary intake 343 Q13 and other health related outcomes (National Cancer Institute, n.d.-a,b); 344 an approach we adhered to in the current study. Finally, the study design 345 is cross-sectional, therefore temporality, an important criterion in 346 determining causality, cannot be established.
F
298 299
5
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
6
444 445 446 447 448 449 450 451 452 453 454 455 456
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Shaikh, A.R., Yaroch, A.L., Nebeling, L., Yeh, M.-C., Resnicow, K., 2008. Psychosocial predictors of fruit and vegetable consumption in adults a review of the literature. Am. J. Prev. Med. 34, 535–543. Shuval, K., Gabriel, K.P., Leonard, T., 2013. TV viewing and BMI by race/ethnicity and socio-economic status. PLoS One 8, e63579. Shuval, K., Finley, C.E., Barlow, C.E., Gabriel, K.P., Leonard, D., Kohl, H.W., 2014. Sedentary behavior, cardiorespiratory fitness, physical activity, and cardiometabolic risk in men: The Cooper Center longitudinal study. Mayo Clin. Proc. 88, 1052–1062. Troiano, R.P., Berrigan, D., Dodd, K.W., Mâsse, L.C., Tilert, T., McDowell, M., 2008. Physical activity in the United States measured by accelerometer. Med. Sci. Sports Exerc. 40, 181–188. U.S. Department of Health and Human Services, 2008. Physical Activity Guidelines for Americans.
U.S. Department of Health and Human Services, 2010. Dietary Guidelines for Americans, 2010 [WWW Document] URL, http://www.health.gov/dietaryguidelines/dga2010/ DietaryGuidelines2010.pdf (accessed 1.12.14). US Department of Agriculture, d. Healthy Eating Index-2010 [WWW Document] URL http:// www.cnpp.usda.gov/sites/default/files/healthy_eating_index/Readme_HEI2010_ NHANES0708_PopulationScore.pdf. World Health Organization, 2003. Diet , nutrition and the prevention of prevention of chronic disease (Geneva). World Health Organization, d. WHO Global Database on Body Mass Index [WWW Document] URL http://apps.who.int/bmi/index.jsp.
U
N
C
O
R
R
E
C
T
E
D
P
R O
O
F
467
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
457 458 459 460 461 462 463 Q3 464 465 466
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
3Q2
Kerem Shuval a,⁎, Binh T. Nguyen a, Amy L. Yaroch b, Jeffrey Drope a, Kelley Pettee Gabriel c
O
F
2
Accelerometer determined sedentary behavior and dietary quality among US adults
4 5 6
a
7
a r t i c l e
8 9
Available online xxxx
10 Q4 11 12
Keywords: Sedentary behavior Dietary quality
Intramural Research Department, American Cancer Society, Atlanta, GA 30303, USA The Gretchen Swanson Center for Nutrition, Omaha, NE 68114, USA Divisions of Epidemiology, Human Genetics and Environmental Sciences, School of Public Health, Austin Regional Campus, University of Texas Health Science Center in Houston, Austin, TX 78701, USA
i n f o
a b s t r a c t
Objective. Scant evidence exists pertaining to objectively measured sedentary time and dietary quality among adults. Therefore, we examined the relationships between sedentary time, physical activity, and dietary quality. Methods. Cross-sectional analyses of a 4,910 US adults from two cycles (2003–2006) of the National Health and Nutrition Examination Survey. The primary independent variables were sedentary time and physical activity (continuous and categorical), while the outcomes were overall dietary quality (Healthy Eating Index (HEI) 2010), fruit and vegetable scores, and empty caloric intake (kcal). Results. Multivariable analyses revealed that a 1 min increase in daily sedentary behavior was associated with a 0.2 kcal decrease in empty calories (−0.18, 95% CI = −0.34, −0.03); however, sedentary time was not significantly related to overall dietary quality (HEI) and fruit and vegetable intake. In comparison, a 1 min increase in daily moderate-to-vigorous intensity physical activity was related to a 0.1 higher HEI score (0.08, 95% CI = 0.04, 0.11), a 0.01 higher fruit score (0.01, 95% CI = 0.01, 0.02), and conversely a 1.3 kcal decrease in empty calories (−1.35, 95% CI = −2.01, −0.69). In addition, meeting physical activity guidelines was associated with a 2.8 point higher HEI score (2.82, 95% CI = 1.40, 4.25), a 0.5 point higher fruit score (0.51, 95% CI = 0.31–0.71), and 37.4 fewer empty calories (−37.43, 95% CI = −64.86, −9.10). Conclusions. Physical activity is significantly related to better overall dietary quality, while sedentary behavior is not. Findings suggest the need to promote physical activity and encourage adherence to dietary guidelines jointly, whereas sedentary behavior and overall dietary quality might need to be targeted independently. © 2015 Published by Elsevier Inc.
R
E
C
T
E
D
c
P
b
R O
1Q1
38 39 40 41 42 43 44 45 46 47 48 49 50
Sedentary behavior is a relatively recent focus of scientific exploration and the accumulating evidence indicates that excessive time spent TV viewing, commuting in a motorized vehicle, or sitting at home or on the job is linked to adverse health outcomes (Owen et al., 2010; Shuval et al., 2014). Specifically, observational studies have found that prolonged sedentary time is linked to an increased risk for obesity, type 2 diabetes mellitus, and premature death, even while considering the protective effects of physical activity (Gardiner et al., 2011; Healy et al., 2011). While the evidence on the potential deleterious effects of sedentary time is accumulating, the evidence base pertaining to physical activity and health is well established, with observed dose–response relationships between higher frequency and intensity levels of activity and reduced risk for obesity, chronic conditions (e.g. depression), and premature mortality (U.S. Department of Health and Human Services, 2008). Recent physical
N C O
37
Introduction
U
34 36
R
35 33 32
⁎ Corresponding author at: Economics and Health Policy Research, Intramural Research Department, American Cancer Society, 250 Williams Street, NE Atlanta, GA 30303, USA. Fax: +1 404 327 6450. E-mail address: [email protected] (K. Shuval).
activity guidelines reflect this evidence and underscore the importance of reducing overall sedentary time and regularly engaging in physical activity to achieve health benefits (Kushi et al., 2012; U.S. Department of Health and Human Services, 2008). With regard to diet, the 2010 Dietary Guidelines for Americans recommend a healthful diet consisting of nutrient-rich foods to enhance caloric balance for weight management (U.S. Department of Health and Human Services, 2010). A nutrient-rich diet refers to foods and beverages with high nutritional quality, such as fruits, vegetables, and whole grains. However, Americans tend to consume more energy-dense, nutrient-poor foods such as those containing an excessive amount of solid fats, added sugar, and refined grain products (U.S. Department of Health and Human Services, 2010). As with a physically active lifestyle, consuming a more healthful diet significantly reduces the risk for obesity and numerous chronic diseases (World Health Organization, 2003). Observational and interventional research often focus on these two modifiable health behaviors as distinct from one another, although these behaviors are likely inter-related and, when combined, may provide additional health benefits (Diabetes Prevention Program Research Group, 2002). Previous research has found that less than a tenth of Americans meet physical activity guidelines (Troiano et al., 2008), that ~8 h are spent in sedentary
http://dx.doi.org/10.1016/j.ypmed.2015.06.010 0091-7435/© 2015 Published by Elsevier Inc.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71
72 73
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
ambulatory participants during waking hours. Participants were instructed to wear the accelerometer for 7 consecutive days on the right hip. Following the data collection period, participants returned the accelerometer by mail, and then data were downloaded and prepared for further processing (Matthews et al., 2008). For the current study, Freedson accelerometer count threshold values were used to characterize accelerometer counts as time (min/day) spent sedentary [i.e. b100 counts per minute (cpm)], light (100–1951 cpm), and in moderate to vigorous intensity physical activity (≥1952 cpm) (Freedson et al., 1998). Weekly summary averages were computed only for participants that wore the monitor ≥10 h per day on ≥4 of 7 days (Mâsse et al., 2005). In addition, daily summary averages were computed for participants. For analysis, participants were classified as meeting physical activity guidelines (yes/no) if accumulating ≥150 min a week of moderate to vigorous physical activity, and categorized into sample-specific quartiles of sedentary time (quartile cut-points: 6.8, 8.1, 9.4 h/day). Furthermore, continuous measures of sedentary time and physical activity (min/day) were examined. Accelerometer wear time was taken into account in all multivariable analyses (see Statistical analysis).
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
O
2
153
104
Methods
105
Data and participants
106 107 108 109 Q5 110 111 112 113 114 Q6 115 116 117 118 119 120 121 122 123 124 125 Q7 126
To examine the association between sedentary behavior and physical activity in relation to dietary quality, we utilized cross-sectional data from the National Health and Nutrition Examination Survey (NHANES). The NHANES, elaborately described elsewhere (Centers for Disease Control and Prevention, n.d.-a,b,c,d), is an ongoing population-based survey (including both interviews and physical examinations) which aims to determine the nutrition and health status of the US population. The NHANES sample is nationally representative of the US noninstitutional population, and utilizes a complex multistage probability sampling design obtaining data in 2-year cycles (Centers for Disease Control and Prevention, n.d.-a,b,c,d). In the current study, we examine data from two cycles (2003–2004; 2005–2006), which contain objectively measured sedentary time and physical activity (assessed via accelerometers), and detailed dietary information (gleaned from 24-h dietary recall data). The current analysis consisted of an adult sample of men and women aged ≥20 years who were not taking insulin, and not pregnant or lactating (n = 9,231). Of these, 3,509 observations were excluded due not having valid accelerometer data, and 812 more observations were omitted due to missing information on measures, leaving an analytic sample of 4,910 for the current study. The NHANES study protocols received approval from the National Center for Health Statistics Research Ethics Review Board; participants provided written informed consent prior to data collection (Centers for Disease Control and Prevention, n.d.-a,b,c,d).
127
Primary measures
Statistical analysis
191
128 129 130 131 132 Q8 133 Q9 134 135
Physical activity and sedentary behavior (independent variables) Physical activity and sedentary behavior were derived from accelerometer measurements and were processed based on publically available SAS programs developed by the National Cancer Institute (NCI) for processing NHANES accelerometer data (National Cancer Institute, n.d.-a,b). Details on the accelerometer protocol appear elsewhere (Centers for Disease Control and Prevention, n.d.-a,b, c,d). Briefly, uniaxial accelerometers (Actigraph model 7164), a small device that measures accelerations as activity counts (1-min epochs), were worn by
Descriptive statistics were utilized to describe participants’ sociodemographics and health behaviors. The association between both quartiles of sedentary time (first quartile- reference) and meeting moderate to vigorous physical activity guidelines (not meeting guidelines-reference) to dietary outcomes was examined using ordinary least squares (OLS) regression; both in partially adjusted (age, sex, BMI) and fully adjusted models (age, sex, race/ethnicity, marital status, education, PIR, health
192 193
97 98 99 100 101 102
178
Covariates included in the multivariable analysis consisted of participants’ sex, age (20–39.9, 40–59.9, ≥ 60 years), race/ethnicity (African American, Hispanic, White, other), marital status (married, widowed, divorced or separated, never married), educational attainment (less than high school, high school, some college, college and above), health status (poor, fair, good, very good, excellent), and poverty-to- income (PIR) ratio (continuous), and total caloric intake (kcal-continuous). In addition, current smoking (no/yes) was determined based on objectively measured cotinine levels; i.e. a 3.08 ng/mL cotinine level adhering to the cut-point suggested by Benowitz (Benowitz et al., 2009). Body mass index (BMI) was computed based on the standard formula (kg/m2) (World Health Organization).
179 180
R O
Covariates
P
95 96
154 155 156 157 158 Q10 159 160 161 162 163 164 165 166 167 168 Q11 169 170 Q12 171 172 173 174 175 176 177
D
93 94
Dietary quality was based on the HEI-2010, which was created by the United States Department of Agriculture to assess adherence to the primary components of the 2010 Dietary Guidelines for Americans and to determine the quality of dietary intake and patterns of consumption (Guenther et al., 2013a). Details on the components of the HEI are elaborated elsewhere (National Cancer Institute, n.d.-a,b); however, briefly the HEI score (0–100 points) was composed of the following 12 components, of which each contributed to the total dietary quality score: total fruit (5 points), whole fruit (5 points), total vegetables (5 points), greens and beans (5 points), whole grains (10 points), dairy products (10 points), total protein (5 points), seafood and proteins from plants (5 points), fatty acids (10 points), refined grains (10 points), sodium (10 points), and empty calories (20 points) (Guenther et al., 2013b). The HEI score was computed using individual 24-h dietary recall data, MyPyramid Equivalents Database (MPED) (version 2.0), CNPP MPED for whole fruit and fruit juice, and the CNPP addendum to the MPED (version 2.0B) (National Cancer Institute, n.d.-a,b). The National Cancer Institute methodology, elaborated elsewhere (Freedman et al., 2008; Guenther et al., 2013b; National Cancer Institute, n.d.-a,b), was used to calculate the HEI total score and components as well as the corresponding standard errors and confidence interval from the self-reported 24-h dietary recall data in NHANES (US Department of Agriculture). For analysis, along with the total HEI score as a primary outcome, we focused on a few specific components which are related to more healthful food choices (i.e. total vegetable score, total fruit score), and conversely on less healthful food choices: empty calories (i.e. calories (kcal) from solid fats, added sugars, and excessive alcohol intake).
E
91 92
Dietary quality and other dietary variables (dependent variables)
T
89 90
C
87 88
E
85 86
R
83 84
R
81 82
O
79 80
C
78
N
76 77
U
74 75
F
103
time daily during waking hours (Matthews et al., 2008), and that dietary quality of Americans is poor, resulting in 53.5/100 dietary quality score (based on Healthy Index Score) (Guenther et al., 2013a). Thus, despite the health benefits associated with reducing sedentary time, increasing physical activity, and consuming a healthy diet, Americans mostly do not adhere to these behaviors. The inter-relationship among multiple health behaviors (e.g. physical activity, diet, smoking, and alcohol consumption) has been previously explored (Berrigan et al., 2003; King et al., 2009; Reeves and Rafferty, 2005; Schuit et al., 2002). For example, Schuit found in a cross-sectional study that approximately 20% of their sample had at least three lifestyle risk factors (e.g. smoking, physical inactivity), whereas Reeves and Rafferty found that only 3% of US adults adhered to all the healthy lifestyle behaviors they examined: not smoking, normal weight, sufficient fruit and vegetable intake, and meeting physical activity guidelines. In fact, King observed a 7% decline (from 15% to 8%) over 18 years in adherence to healthy lifestyle behaviors (King et al., 2009). However, the relationship between sedentary behavior and dietary quality has not sufficiently been explored among adults. Specifically, systematic reviews on this topic concluded that sedentary behavior is related to unhealthy dietary intake, primarily among children and adolescents; however, the evidence for adults is inconclusive (Hobbs et al., 2014; Pearson and Biddle, 2011). Additionally, the studies in the systematic reviews mostly relied on self-reported screen time or TV viewing (Hobbs et al., 2014), which are only proxies of overall sedentary time and are subject to recall bias and subsequent measurement error (Crosby et al., 2006). Therefore, in the current study, we utilize accelerometers, an objective measure of time spent in sedentary behavior to more accurately determine the exposure. Specifically, we aimed to examine the association between both sedentary behavior and physical activity (via accelerometry) and dietary quality (Health Eating Index (HEI), 2010) among a national sample of US adults.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
181 182 183 184 185 186 187 188 189 190
194 195 196 197 198
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Discussion
250 251
Currently, a dearth of evidence exists as to whether excessive sedentary time is related to dietary quality among a national sample of US adults. Each of these health behaviors is an independent risk factor for chronic disease morbidity and mortality, thus there is a need for substantiation on whether these behaviors should be targeted independently or jointly. In the current study, we aimed to examine the association of sedentary time and physical activity (both measured objectively) with dietary quality (HEI), more healthful (fruits and vegetables) and less healthful (empty calories) dietary intake in adults. Results reveal that prolonged sedentary time was not related to overall dietary quality and more healthful food choices, whereas excessive sedentary behavior
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246
252 253 254 255 256 257 258 259 260
C
223 224
E
221 222
R
219 220
R
218
N C O
212 213
U
210 211
F
249
208 209
O
247 248
Participants’ baseline characteristics are presented in Table 1. Briefly, half of the sample were women, and, on average, participants were aged 48.5 years (SE = 0.5). Most (75.7%) were non-Hispanic white, 9.1% were non-Hispanic blacks, 10.4% were Hispanic, and 59.4% had some or complete college education. In addition, 48.3% reported very good or excellent health, 26.6% were current smokers, 31.1% were of normal weight, and participants’ mean dietary quality score was 47.6/100 (SE = 0.4). Participants spent 486.3 min (SE = 2.0) per day sedentary (or 8.1 h daily), in comparison to only 23.9 min (SE = 0.6) per day participating in moderate-to-vigorous intensity physical activity. The relationship between sedentary behavior, physical activity, and dietary quality is depicted in Table 2 (partially adjusted models) and Table 3 (fully adjusted models). In the fully adjusted models, when examining these relationships continuously, more minutes per day spent in sedentary time was not significantly associated with overall dietary quality (HEI) and the fruit and vegetables scores; however, more sedentary time was significantly associated with a lower intake of empty calories (p b 0.02). Specifically, a 1-min per day increase in sedentary time was associated with a 0.2 kcal decrease in the empty calories (− 0.18, 95% CI = − 0.34, − 0.03). In comparison, a 1-min increase in moderate to vigorous physical activity per day was significantly related to a 0.1 higher HEI score (0.08, 95% CI = 0.04, 0.11), a 0.01 higher fruit score (0.01, 95% CI = 0.01, 0.02), and conversely a 1.3 kcal lower intake of empty calories (− 1.35, 95% CI = − 2.01, −0.69). When examining these relationships categorically, comparing the fourth quartile of sedentary time (≥ 9.4 h daily) to the reference group (first quartile) in relation to dietary quality resulted in nonsignificant findings (Table 3). In comparison, meeting physical activity guidelines was associated with a 2.8 point higher HEI score (2.82, 95% CI = 1.40, 4.25), a 0.5 point higher fruit score (0.51, 95% CI = 0.31–0.71), and 37.4 kcal fewer empty calories (− 37.43, 95% CI = − 64.86, − 9.10).
206 207
R O
217
205
Sex Women Age groups (years) 20–39.9 40–59.9 ≥60 Race/ethnicity Non-Hispanic White Non-Hispanic Black Hispanic Other Marital status Married Widow Divorced/separated Never married Education bHigh school High school Some college ≥College Poverty-to-income ratio (PIR)—mean (SE) BMI (kg/m2) categories Underweight (b18.5) Normal weight (18.5–24.9) Overweight (25.0–29.9) Obese (≥30.0) Self-reported health status Excellent Very good Good Fair Poor Current smokingc HEI-2010 total scored—mean (SE) Total vegetables—mean (SE) Total fruit—mean (SE) Added sugar (teaspoon)—mean (SE) Solid fat (grams)—mean (SE) Empty calories (kcal)—mean (SE) Total calories (kcal)—mean (SE) Sedentary time, minutes per daye—mean (SE) Light physical activity, minutes per daye—mean (SE) MVPA, minutes per daye—mean (SE) Meeting physical activity guidelinese
P
Results
203 204
% (n)b
Characteristics
D
216
201 202
Table 1 t1:1 Descriptive characteristics of sample (population weighted)a: NHANES 2003–2006 t1:2 (n = 4,910). t1:3
T
214 215
status, smoking, BMI, total caloric intake); all adjusting for accelerometer wear time. In addition, continuous measures of sedentary time and physical activity (min/day) were entered into separate models: (1) partially adjusted models: age, sex, BMI, accelerometer wear time; and (2) fully adjusted models: age, sex, race/ethnicity, marital status, education, PIR, health status, smoking, BMI, total caloric intake, light intensity physical activity, and accelerometer wear time. The OLS models were computed when examining the association between sedentary behavior and physical activity (categorical or continuous variables) with each dietary outcome (i.e. HEI, vegetables, fruit, and empty calories). SAS (version 9.3) statistical software was utilized to compute the HEI score and accelerometer estimates. Stata version 13 (StataCorp) was used for descriptive and analytical analyses to take into account NHANES’s complex multistage sampling design to allow for nationally representativeness. Since dietary quality was the primary outcomes (as derived from the Mobile Examination Centers (MEC) data), MEC sample weights were utilized in the analyses. An alpha of b0.05 was considered statistically significant.
E
199 200
3
50.0% (2,330) 31.6% (1,341) 43.0% (1,631) 25.4% (1,938) 75.7% (2,712) 9.1% (912) 10.4% (1,109) 4.8% (177) 68.4% (3,161) 6.3% (483) 12.2% (629) 13.1% (637) 15.1% (1,271) 25.6% (1,202) 32.3% (1,397) 27.1% (1,040) 3.20 (1.53) 1.5% (62) 31.1% (1,427) 35.3% (1,813) 32.2% (1,608) 12.3% (510) 36.0% (1,492) 36.6% (1,871) 12.8% (887) 2.3% (150) 26.6% (1,214) 47.61 (0.39) 3.11 (0.04) 2.15 (0.07) 18.96 (0.41) 434.83 (5.65) 773.56 (10.12) 2244.78 (16.39) 486.26 (1.98) 259.95 (1.46) 23.90 (0.59) 8% (365)
t1:4 t1:5 t1:6 t1:7 t1:8 t1:9 t1:10 t1:11 t1:12 t1:13 t1:14 t1:15 t1:16 t1:17 t1:18 t1:19 t1:20 t1:21 t1:22 t1:23 t1:24 t1:25 t1:26 t1:27 t1:28 t1:29 t1:30 t1:31 t1:32 t1:33 t1:34 t1:35 t1:36 t1:37 t1:38 t1:39 t1:40 t1:41 t1:42 t1:43 t1:44 t1:45 t1:46 t1:47 t1:48 t1:49
NHANES, National Health and Nutrition Examination Survey; SE, standard error; kcal, kilocalories; BMI, body mass index, HEI, Healthy Eating Index; MVPA, moderate to vigorous intensity physical activity. a NHANES survey weights are accounted for in the table. b The number and percentage is presented unless otherwise specified. c Current smoking was determined based on cotinine cut-point (3.08 ng/mL) suggested by Benowitz (Benowitz et al., 2009) d The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The empty calories score is 0–20 points and the total fruit and total vegetable scores are each 0–5 points. e Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minutes (cmp)), light-intensity physical activity (100–1951 cpm), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week.
t1:50 t1:51 t1:52 t1:53 t1:54 t1:55 t1:56 t1:57 t1:58 t1:59 t1:60 t1:61 t1:62 t1:63 t1:64 t1:65 t1:66
was significantly and inversely associated with less healthful dietary intake. Engaging in physical activity, in comparison, was significantly related to more healthful food choices (fruit intake), and decreased consumption of less healthful foods. Our results pertaining to objectively measured sedentary time and diet among adults are unique and have rarely been explored, since previous studies have primarily relied on reported sedentary time predominantly among children and adolescents. Indeed, a recent systematic review by Hobbs et al. noted that while excessive sedentary behavior
261
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
262 263 264 265 266 267 268 269
4 t2:1 t2:2
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Table 2 Association between sedentary time, physical activitya and dietary qualityb: partially adjusted modelsc: NHANES 2003–2006.
t2:3 t2:4 t2:5 t2:6 t2:7 t2:8 t2:9 t2:10 t2:11 t2:12 t2:13 t2:14
HEI-2010
Vegetables
Mean (SE) 95% CI Continuous independent variables Sedentary time (min/day)d 0.00 (0.00) −0.00, 0.01 Physical activity (min/day)d 0.13 (0.02)** 0.10, 0.17 Categorical independent variables Sedentary time—Quartiles (first quartile reference)e Second quartile 0.95 (0.61) −0.25, 2.14 Third quartile 1.88 (0.62)** 0.67, 3.09 Fourth quartile 1.51 (0.66)* 0.22, 2.81 Physical activity (not meeting guidelines-reference)e Meeting guidelines 5.14 (0.73)** 3.70, 6.58
Fruit
Empty Calories
Mean (SE)
95% CI
Mean (SE)
95% CI
Mean (SE)
95% CI
0.00 (0.00) 0.00 (0.00)
−0.00, 0.00 −0.00, 0.01
0.00 (0.00) 0.02 (0.00)**
−0.00, 0.00 0.02, 0.03
−0.51 (0.13)** −2.05** (0.56)
−0.77, −0.25 −3.15, −0.96
0.08 (0.07) 0.17 (0.07)* 0.07 (0.08)
−0.06, 0.22 0.03, 0.31 −0.08, 0.22
0.09 (0.09) 0.17 (0.09) 0.11 (0.09)
−0.08, 0.26 −0.00, 0.34 −0.69, 0.30
−25.50 (19.47) −77.62 (19.61)** −109.34 (21.09)**
−63.68, 12.67 −116.08, −39.16 −150.68, −68.00
0.04 (0.08)
−0.13, 0.20
0.78 (0.10)**
0.58, 0.99
−85.97 (23.32)**
−131.68, −40.25
NHANES, National Health and Nutrition Examination Survey; SE, standard error; HEI, Healthy Eating Index. *p b .05, **p b .01. a Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minute (cmp)), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week. b The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The total fruit and total vegetable scores are each 0–5 points, whereas empty calories are kcals from solid fats, added sugars, and excessive alcohol intake. c Separate OLS regression models were constructed for the continuous and the categorical sedentary time and physical activity measures in relation to each dietary outcome. d The continuous sedentary time (minutes per day) and physical activity (minutes per day in moderate to vigorous intensity physical activity) variables were entered jointly into each regression model while adjusting for age, sex, BMI, light intensity physical activity, and accelerometer wear time. e The categorical sedentary time (quartiles) and physical activity (meeting guidelines) variables were entered jointly into each regression model while adjusting for age, sex, BMI, and accelerometer wear time.
270 271
282 283
(primarily TV viewing) was related to lower fruit and vegetable intake and greater consumption of energy-dense, nutrient-poor foods among children and adolescents, the evidence for adults was insufficient to determine a clear relationship (Hobbs et al., 2014). TV viewing has been used as a proxy for sedentary behavior in numerous studies (Shuval et al., 2013, 2014); however, it is not a direct measure of sitting and individuals might “multi-task” (i.e. perform other activities) while watching TV (Marsh et al., 2013). Thus, the link between TV viewing and a less healthful diet found previously might stem from exposure to advertisements of energy-dense foods and ‘distracted eating’ rather than the activity of sitting itself (Giese et al., 2014; Marsh et al., 2013). ‘Distracted eating’ refers to the dominance of the automatic system (over the reflective one) when making decisions pertaining to eating during TV viewing; that is, people are more likely to make emotional and
t3:1 t3:2
Table 3 Association between sedentary time, physical activitya and dietary qualityb: fully adjusted modelsc: NHANES 2003–2006.
t3:15 t3:16 t3:17 t3:18 t3:19 t3:20 t3:21 t3:22 t3:23 t3:24 t3:25 t3:26 t3:27
O
P
R O
t3:3
HEI-2010
N
C
Mean (SE) 95% CI Continuous independent variables Sedentary time (min/day)d 0.00 (0.00) −0.00, 0.01 Physical Activity (min/day)d 0.08** (0.02) 0.04, 0.11 Categorical independent variables Sedentary time (first quartile reference)e Second quartile 0.85 (0.60)** −1.10, 1.27 Third quartile 0.63 (0.62) −0.58, 1.84 Fourth quartile 0.07 (0.67) −1.25, 1.39 Physical activity (not meeting guidelines-reference)e Meeting guidelines 2.82 (0.73)** 1.40, 4.25
U
t3:4 t3:5 t3:6 t3:7 t3:8 t3:9 t3:10 t3:11 t3:12 t3:13 t3:14
D
E
T
C
280 281
E
278 279
R
276 277
uncontrolled decisions about eating and are less exposed to cues that would limit food intake when distracted by screen time (Kahneman, 2003; Marsh et al., 2013; Shaikh et al., 2008). However, in the current study, the targeted exposure was not limited to TV viewing, but rather focused on accumulated sedentary time across all waking hours and its relationship with dietary intake. Notably, our results point to a null association between sedentary time and overall dietary quality and a significant and inverse association between prolonged sedentary time and the consumption of less healthful foods (i.e. empty calories). While the underlying mechanism explaining this finding is unclear and warrants further investigation, it is quite plausible that the activity of sitting itself (which involves low energy expenditure) might enable the reflective system to be more dominant, resulting in more rational decisions (Kahneman, 2003; Shaikh et al., 2008). Consequently, individuals
R
274 275
O
272 273
F
t2:15 t2:16 t2:17 t2:18 t2:19 t2:20 t2:21 t2:22 t2:23 t2:24 t2:25 t2:26
Vegetables
Fruit
Empty Calories
Mean (SE)
95% CI
Mean (SE)
95% CI
Mean (SE)
95% CI
0.00 (0.00) 0.00 (0.00)
−0.00, 0.00 −0.00, 0.00
0.00 (0.00) 0.01** (0.00)
−0.00, 0.00 0.01, 0.02
−0.18 (0.08)* −1.35 (0.34)**
−0.34, −0.03 −2.01, −0.69
0.01 (0.07) 0.09 (0.07) −0.02 (0.08)
−0.13, 0.15 −0.05, 0.23 −0.18, 0.13
0.02 (0.09) 0.06 (0.09) −0.03 (0.10)
−0.15, 0.19 −0.11, 0.23 −0.21, 0.16
−3.41 (11.62) −23.52* (11.91) −25.39 (12.97)
−26.21, 19.39 −46.87, −0.18 −50.81, 0.04
−0.06 (0.08)
−0.22, 0.11
0.51 (0.10)**
0.31, 0.71
−37.43** (13.99)
−64.86, −9.10
NHANES, National Health and Nutrition Examination Survey; SE, standard error; HEI, healthy eating index. *p b .05, **p b .01. a Information on sedentary time and physical activity were derived from accelerometers. Continuous sedentary time and physical activity are in minutes per day. The intensity levels are based on Freedson’s cut-points; i.e. sedentary time (0–99 counts per minutes (cmp)), and moderate to vigorous intensity physical activity (≥1,952 cpm). Quartile cut-points for sedentary time are: 6.8, 8.1, 9.4 h per day. Individuals were regarded as meeting physical activity guidelines if engaging in moderate to vigorous intensity physical activity 150 min or more per week. b The Healthy Eating Index-2010 score (0–100) assesses adherence to the primary components of the 2010 Dietary Guidelines for Americans. The total fruit and total vegetable scores are each 0–5 points, whereas empty calories are kcals from solid fats, added sugars and excessive alcohol intake. c Separate OLS regression models were constructed for the continuous and the categorical sedentary time and physical activity measures in relation to each dietary outcome. d The continuous sedentary time (minutes per day) and physical activity (minutes per day in moderate to vigorous intensity physical activity) variables were entered jointly into each multivariable model while adjusting for age, sex, race/ethnicity, marital status, education, PIR, self-reported health status, current smoking, BMI, caloric intake, light intensity physical activity, and accelerometer wear time. e The categorical sedentary time (quartiles) and physical activity (meeting guidelines) variables were entered jointly into each multivariable model while adjusting for age, sex, race/ethnicity, marital status, education, PIR, self-reported health status, current smoking, BMI, caloric intake, and accelerometer wear time.
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
284 285 286 287 288 289 290 291 292 293 294 295 296 297
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
N C O
R
R
E
C
Conclusions
348
Study findings reveal that excessive sedentary time was not associated with overall dietary quality. In comparison, engaging in health-promoting physical activity was significantly related to consuming a diet of higher quality consisting of more fruits and fewer empty calories. These findings underscore the need to design intervention programs focusing on increasing and maintaining physical activity levels while encouraging adherence to dietary guidelines among adults. In comparison, reducing and breaking up sedentary time might need to be targeted independently.
349 350 351 352 353 354 355 356
U
347
357 Q15 Conflict of interest 358 The author has no conflict of interest.
References
359
Benowitz, N.L., Bernert, J.T., Caraballo, R.S., Holiday, D.B., Wang, J., 2009. Optimal serum cotinine levels for distinguishing cigarette smokers and nonsmokers within different racial/ethnic groups in the United States between 1999 and 2004. Am. J. Epidemiol. 169, 236–248. Berrigan, D., Dodd, K., Troiano, R.P., Krebs-Smith, S.M., Barbash, R.B., 2003. Patterns of health behavior in U.S. adults. Prev. Med. (Baltim.) 36, 615–623. Centers for Disease Control and Prevention, a. National Health and Nutrition Examination Survey [WWW Document] URL, http://www.cdc.gov/nchs/nhanes.htm (accessed 5.1.15). Centers for Disease Control and Prevention, b. Key Concepts About the NHANES Survey Design [WWW Document] URL http://www.cdc.gov/nchs/tutorials/ dietary/SurveyOrientation/SurveyDesign/Info1.htm. Centers for Disease Control and Prevention, c. National Health and Nutrition Examination Survey—Research Ethics Review Board Approval [WWW Document] URL, http:// www.cdc.gov/nchs/nhanes/irba98.htm (accessed 4.12.14). Centers for Disease Control and Prevention, d. NHANES 2003–2004 Physical Activity Monitor Data Documentation, Codebook, and Frequencies [WWW Document] URL, http:// www.cdc.gov/nchs/nhanes/nhanes2003-2004/PAXRAW_C.htm (accessed 1.8.14). Crosby, R.A., DiClemente, R.J., Salazar, L.F., 2006. Research Methods in Health Promotion. Jossey-Bass. Diabetes Prevention Program Research Group, 2002. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N. Engl. J. Med. 346, 393–403. Freedman, L.S., Guenther, P.M., Krebs-smith, S.M., Kott, P.S., 2008. A population’s mean Healthy Eating Index-2005 scores are best estimated by the score of the population ratio when one 24-hour recall is available. J. Nutr. 138, 1725–1729. Freedson, P.S., Melanson, E., Sirard, J., 1998. Calibration of the Computer Science and Applications, Inc. accelerometer. Med. Sci. Sports Exerc. 30, 777–781. Gardiner, P.A., Healy, G.N., Eakin, E.G., et al., 2011. Associations between television viewing time and overall sitting time with the metabolic syndrome in older men and women: the Australian Diabetes, Obesity and Lifestyle study. J. Am. Geriatr. Soc. 59, 788–796. Giese, H., König, L., Tăut, D., et al., 2014. Exploring the association between television advertising of healthy and unhealthy foods, self-control, and food intake in three European countries. Appl. Psychol. Health. (Well Being Epub ahead). Gillman, M.W., Pinto, B.M., Tennstedt, S., Glanz, K., Marcus, B., Friedman, R.H., 2001. Relationships of physical activity with dietary behaviors among adults. Prev. Med. (Baltim.) 32, 295–301. Guenther, P.M., Casavale, K.O., Kirkpatrick, S.I., et al., 2013a. Diet Quality of Americans in 2001-02 and 2007-08 as Measured by the Healthy Eating Index-2010. Guenther, P.M., Casavale, K.O., Reedy, J., et al., 2013b. Update of the Healthy Eating Index: HEI-2010. J. Acad. Nutr. Diet. 113, 569–580. Haskell, W.L., 2012. Physical activity by self-report: A brief history and future issues. J. Phys. Act. Health 9, 5–10. Healy, G.N., Matthews, C.E., Dunstan, D.W., Winkler, E. a H., Owen, N., 2011. Sedentary time and cardio-metabolic biomarkers in US adults: NHANES 2003-06. Eur. Heart J. 32, 590–597. Hobbs, M., Pearson, N., Foster, P.J., Biddle, S.J.H., 2014. Sedentary behaviour and diet across the lifespan: An updated systematic review. Br. J. Sports Med. 1–11. Joseph, R.J., Alonso-Alonso, M., Bond, D.S., Pascual-Leone, A., Blackburn, G.L., 2011. The neurocognitive connection between physical activity and eating behaviour. Obes. Rev. 12, 800–812. Kahneman, D., 2003. A perspective on judgment and choice: Mapping bounded rationality. Am. Psychol. 58, 697–720. King, D.E., Mainous, A.G., Carnemolla, M., Everett, C.J., 2009. Adherence to healthy lifestyle habits in US adults, 1988-2006. Am. J. Med. 122, 528–534. Kushi, L.H., Doyle, C., McCullough, M., et al., 2012. American Cancer Society Guidelines on Nutrition and Physical Activity for Cancer Prevention Reducing the Risk of Cancer with Healthy Food Choices and Physical Activity. CA Cancer J. Clin. 62, 30–67. Loprinzi, P.D., Smit, E., Mahoney, S., 2014. Physical activity and dietary behavior in US adults and their combined influence on health. Mayo Clin. Proc. 89, 190–198. Ma, Y., Olendzki, B.C., Pagoto, S.L., et al., 2010. Number of 24-hour diet recalls needed to estimate energy intake. Ann. Epidemiol. 19, 553–559. Marsh, S., Mjurchu, C.N., Maddison, R., 2013. The non-advertising effects of screen-based sedentary activities on acute eating behaviours in children, adolescents, and young adults. A systematic review. Appetite 71, 259–273. Mâsse, L., Fuemmeler, B., Anderson, C., et al., 2005. Accelerometer data reduction: A comparison of four reduction algorithms on select outcome variables. Med. Sci. Sports Exerc. 37, S544–S554. Matthews, C.E., Chen, K.Y., Freedson, P.S., et al., 2008. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am. J. Epidemiol. 167, 875–881. National Cancer Institute, a. SAS Programs for Analyzing NHANES 2003-2004 Accelerometer Data [WWW Document] URL, http://appliedresearch.cancer.gov/nhanes_pam/ (accessed 3.10.14a). National Cancer Institute, b. Healthy Eating Index [WWW Document] URL http:// appliedresearch.cancer.gov/hei/. Owen, N., Sparling, P.B., Healy, G.N., Dunstan, D.W., Matthews, C.E., 2010. Sedentary behavior: Emerging evidence for a new health risk. Mayo Clin. Proc. 85, 1138–1141. Pearson, N., Biddle, S.J.H., 2011. Sedentary behavior and dietary intake in children, adolescents, and adults. A systematic review. Am. J. Prev. Med. 41, 178–188. Reeves, M.J., Rafferty, A.P., 2005. Healthy lifestyle characteristics among adults in the United States, 2000. Arch. Intern. Med. 165, 854–857. Schuit, A.J., van Loon, A.J.M., Tijhuis, M., Ocke, M.C., 2002. Clustering of lifestyle risk factors in a general adult population. Prev. Med. (Baltim.) 35, 219–224.
360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 Q14 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
O
R O
P
D
E
T
might be more cognizant of the fact that they are predominantly sedentary throughout the day and avoid less healthful foods to 300 achieve health benefits. This supposition, however, necessitates fur301 ther exploration. 302 In comparison, our findings on physical activity and diet confirm a 303 number of previous studies that have primarily observed that engaging 304 in physical activity is associated with dietary intake and quality, and that 305 there is a reinforcing inter-relationship between the two (Joseph et al., 306 2011). Specifically, Gillman, in a cross-sectional survey of ethnically 307 diverse adults, found that higher amounts of self-reported physical 308 activity were correlated with an overall higher quality diet (Gillman 309 et al., 2001). Similar to the present study, Loprinzi observed a significant 310 relationship between objectively measured physical activity and dietary 311 quality; however, they did not explore other dietary outcomes (e.g. fruit 312 and vegetables, empty calories) and did not use the latest HEI-2010, 313 which is a more accurate representation of adherence to the 2010 314 Dietary Guidelines (Loprinzi et al., 2014; U.S. Department of Health 315 and Human Services, 2010). Thus, the current findings, coupled with 316 previous evidence, emphasize the inter-relationship between physical 317 activity and healthy eating. A review by Joseph suggests that engaging 318 in physical activity could affect one’s ability to self-regulate eating 319 behavior and minimize hedonic hunger through increased self-efficacy, 320 and enhanced executive function as well as physiologically increasing 321 the sensitivity of satiety signaling (Joseph et al., 2011). 322 The current study has several strengths and limitations. The primary 323 strengths include the novelty of the study question, the national sample, 324 and the objective measures (i.e. accelerometry) utilized to determine 325 the exposure. Specifically, while some studies have explored the rela326 tionship between physical activity and diet, none (to our knowledge) 327 have examined the relationship between objectively measured seden328 tary time and dietary intake among adults. Although accelerometers 329 are an objective measure of activity, they do not capture all forms of 330 physical activity (e.g. cycling, swimming), and are not able to discern 331 the specific type sedentary behavior performed. Furthermore, monitors 332 were worn for 4–7 days and thus might not be representative of longer333 term habitual activity levels. Thus, self-reported activity coupled with 334 accelerometer-determined activity might provide a more comprehen335 sive measure of physical activity (Haskell, 2012). Similarly, nutrition 336 information was based on 24-h dietary recall data, which is subject to 337 over- or under-estimation of portion size and might not be indicative 338 of longer-term dietary behavior (Crosby et al., 2006; Ma et al., 2010). 339 Specifically, the use of the HEI in the present study to determine dietary 340 quality on the individual level might generate measurement error. 341 Therefore, adjustment for energy intake is suggested to reduce bias 342 when examining the relationship between individuals’ dietary intake 343 Q13 and other health related outcomes (National Cancer Institute, n.d.-a,b); 344 an approach we adhered to in the current study. Finally, the study design 345 is cross-sectional, therefore temporality, an important criterion in 346 determining causality, cannot be established.
F
298 299
5
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
6
444 445 446 447 448 449 450 451 452 453 454 455 456
K. Shuval et al. / Preventive Medicine xxx (2015) xxx–xxx
Shaikh, A.R., Yaroch, A.L., Nebeling, L., Yeh, M.-C., Resnicow, K., 2008. Psychosocial predictors of fruit and vegetable consumption in adults a review of the literature. Am. J. Prev. Med. 34, 535–543. Shuval, K., Gabriel, K.P., Leonard, T., 2013. TV viewing and BMI by race/ethnicity and socio-economic status. PLoS One 8, e63579. Shuval, K., Finley, C.E., Barlow, C.E., Gabriel, K.P., Leonard, D., Kohl, H.W., 2014. Sedentary behavior, cardiorespiratory fitness, physical activity, and cardiometabolic risk in men: The Cooper Center longitudinal study. Mayo Clin. Proc. 88, 1052–1062. Troiano, R.P., Berrigan, D., Dodd, K.W., Mâsse, L.C., Tilert, T., McDowell, M., 2008. Physical activity in the United States measured by accelerometer. Med. Sci. Sports Exerc. 40, 181–188. U.S. Department of Health and Human Services, 2008. Physical Activity Guidelines for Americans.
U.S. Department of Health and Human Services, 2010. Dietary Guidelines for Americans, 2010 [WWW Document] URL, http://www.health.gov/dietaryguidelines/dga2010/ DietaryGuidelines2010.pdf (accessed 1.12.14). US Department of Agriculture, d. Healthy Eating Index-2010 [WWW Document] URL http:// www.cnpp.usda.gov/sites/default/files/healthy_eating_index/Readme_HEI2010_ NHANES0708_PopulationScore.pdf. World Health Organization, 2003. Diet , nutrition and the prevention of prevention of chronic disease (Geneva). World Health Organization, d. WHO Global Database on Body Mass Index [WWW Document] URL http://apps.who.int/bmi/index.jsp.
U
N
C
O
R
R
E
C
T
E
D
P
R O
O
F
467
Please cite this article as: Shuval, K., et al., Accelerometer determined sedentary behavior and dietary quality among US adults, Prev. Med. (2015), http://dx.doi.org/10.1016/j.ypmed.2015.06.010
457 458 459 460 461 462 463 Q3 464 465 466
