- Email: [email protected]
Weekend-weekday differences in diet among U.S. adults, 2003–2012
Annals of Epidemiology 26 (2016) 57e65
Contents lists available at ScienceDirect
Annals of Epidemiology journal homepage: www.annalsofepidemiology.org
Original article
Weekend-weekday differences in diet among U.S. adults, 2003e2012 Ruopeng An PhD * Department of Kinesiology and Community Health, College of Applied Health Sciences, University of Illinois at Urbana-Champaign, Champaign
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 July 2015 Accepted 19 October 2015 Available online 28 October 2015
Purpose: Dietary patterns differ by day of the week. This study examined weekend-weekday differences in diet among U.S. adults. Methods: Nationally representative data of 11,646 adults 18 years of age and above from the National Health and Nutrition Examination Survey 2003e2012 waves were analyzed. Individual fixed-effect regressions were performed using data from two nonconsecutive 24-hour dietary recalls. Results: Weekend diet was less healthful than weekday, with diet on Saturday the worst. Compared with weekday consumption, consumption on Saturday was associated with an increase in daily intakes of total energy by 181.04 kcal, energy from sugar-sweetened beverages 18.34 kcal, energy from alcohol 46.65 kcal, energy from discretionary foods 48.77 kcal, total fat 8.16 g, saturated fat 2.88 g, sugar 5.37 g, sodium 205.59 mg, and cholesterol 43.17 mg, a decrease in intakes of fruit by 13.90 g, vegetable 16.76 g, and fiber 0.67 g, a decrease in the Healthy Eating Index-2010 score by 2.32, and an increase in the prevalence of fast-food and full-service restaurant consumption by 10.21% and 17.79%, respectively. Weekend-weekday differences in diet varied by sex, age, race and/or ethnicity, education, income, and body weight status. Conclusions: Americans’ weekend consumption was associated with increased calorie intake and poorer diet quality. Ó 2016 Elsevier Inc. All rights reserved.
Keywords: 24-hour dietary recall Diet quality Weekend Weekday
Introduction Improving diet quality is a key health promotion strategy [1]. Since 1980, a major theme of the U.S. federal dietary guidelines has been to increase consumption of nutrient-rich foods and reduce consumption of energy-dense foods [2]. However, a large majority of the American population fails to meet these guidelines, with insufficient consumption of nutrient-rich foods and excessive discretionary calorie intake [3]. Human activities consist of many repeated patterns and habitual behaviors. People’s day-to-day lives are punctuated by work, education, and other routines. Alternations of these routines such as not working in paid employment or attending school on the weekend may lead to changes in dietary patterns [4]. Eating out tends to be more common in weekends than in weekdays, and weekend food consumption has been consistently linked to
The author has no conflict of interest to declare. * Corresponding author. 1206 South 4th Street, Tel.: þ1-217-244-0966; fax: þ1-217-333-2766. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.annepidem.2015.10.010 1047-2797/Ó 2016 Elsevier Inc. All rights reserved.
Champaign,
IL
61820.
increased discretionary calorie intake from energy-dense, nutrientpoor foods [5,6]. Previous research on weekend-weekday differences in dietary patterns covered diverse populations (e.g., children [7e13], adolescents [7,8,10,11,13,14], young adults [8,10,15e17], middle-aged and older adults [8,10,15e18], overweight/obese individuals [12,16]) residing in different countries (e.g., Canada [10], Denmark [7,13], United States [8,11,17e19], Korea [20], and other countries [9,12]). In general, these studies reported less healthful dietary intakes in weekends in comparison to weekdays. However, four issues remain to be adequately addressed. First, a large majority of studies used cross-sectional methods [7e13,15,17] and/or nonnationally representative data [12,14,16e20], with findings susceptible to confounding bias (due to failure in controlling for between-individual differences in characteristics such as food or taste preferences) and compromised in generalizability to the general population. Second, dietary outcomes were typically limited to total calories, fats and/or a few individual food items [7e9,12e20], whereas overall diet quality and source of foods (e.g., food/beverage from fast-food/full-service restaurant) were underexamined. Third, most studies grouped Friday, Saturday, and Sunday together (or Saturday and Sunday together) in analysis [7e20],
58
R. An / Annals of Epidemiology 26 (2016) 57e65
which overlooked the distinctive dietary pattern on each of these three weekend days. Finally, population heterogeneities in weekend-weekday differences in dietary patterns may be present but are largely undocumented [21]. Built on previous literature, this study examined weekendweekday differences in diet among U.S. adults using data from a nationally representative survey. To our knowledge, this study serves as the first attempt to study weekend-weekday differences in diet using a panel data approach, which used within-individual variations in diet between the two nonconsecutive 24-hour dietary recall days (one on weekday and one on weekend) and thus addressed the confounding issue due to unobservable individual characteristics like taste preferences. We hypothesized that weekend diet was less healthful than weekday diet, and the differences were heterogeneous across population subgroups. Methods Survey setting and participants Individual-level data came from the National Health and Nutrition Examination Survey (NHANES) 2003e2004, 2005e2006, 2007e2008, 2009e2010, and 2011e2012 waves. Since 1999, NHANES has been conducted continuously in 2-year cycles and focused on a variety of health and nutrition measurements [22]. A multistage probability sampling design is used to select participants representative of the civilian, noninstitutionalized U.S. population.
Control and Prevention and the National Cancer Institute [23,24]. Milk (plain or flavored) and 100% fruit juice were excluded. Alcohol includes beers and ales, cordials and liqueurs, cocktails, wines and distilled liquors, consistent with the USDA Food and Nutrient Database for Dietary Studies food and/or beverage categorization. In the NHANES 2011e2012 wave, SSB and alcohol consist of 48 and 33 reported beverage items, respectively. The number of reported items in each beverage category differed only slightly across survey waves. Discretionary food consumption The discretionary food category identifies energy-dense, nutrient-poor food products that do not belong to the main food groups or necessarily contain essential nutrients that the human body requires but may add diversity [25]. Foods in this category may be consumed “sometimes in small amounts by those who are physically active, but are not a necessary part of the diet” [25]. Following Bleich et al. (2014) [26], specific food items in the discretionary food category include cookies, pies, ice cream, confectionery, chocolate, other desserts (e.g., custards, puddings, mousse, gelatin dessert), sweet rolls, waffles, cakes, pastries (e.g., crepes, cream puffs, strudels, croissants, muffins, sweet breads), biscuits, hush puppies, chips, popcorn, pretzels, party mixes, and fries. In the NHANES 2011e2012 wave, the discretionary food category consists of 661 reported food items. The number of reported items in the discretionary food category differed only slightly across survey waves.
Dietary interview Calorie intake Except for the NHANES 1999e2000 wave where all respondents were asked to complete a single 24-hour dietary recall interview, all subsequent waves incorporated two dietary recalls, with the first collected in-person and the second by telephone 3 to 10 days later. After the dietary interview, the calorie and nutrient contents of each reported food and/or beverage item were systematically coded with the U.S. Department of Agriculture (USDA) Food and Nutrient Database for Dietary Studies. Access restrictions apply to the day 2 dietary recall data collected in the NHANES 2001e2002 wave, whereas dietary data for both recall days are released to the public for all subsequent waves. Among the 23,865 U.S. adults, 18 years of age and above who participated in the NHANES 2003e2012 waves, 1352 (5.7%) who were pregnant, lactating, and/or on a special diet to lose weight at the time of interview were excluded. Of the remaining 22,513 adults who had complete dietary measures, only those whose two dietary recall days occupied a weekday and a weekend contributed to the individual fixed-effect estimation. Therefore, the effective sample sizes (number of individuals) were 11,646 for all adults, and 5812, 5834, 5524, 2457, 2930, 8883, 2763, 5478, 5315, 3444, 3417, 3932, 3423, 3885, and 3960 for male adults, female adults, white adults, African American adults, Hispanic adults, adults 18e64 years of age, adults 65 years of age and above, adults with high school or lower education, adults with college and above education, adults of low household income level, adults of middle household income level, adults of high household income level, normal weight adults, overweight adults, and obese adults, respectively. Sugar-sweetened beverage and alcohol consumption Sugar-sweetened beverage (SSB) includes sodas, fruit drinks, energy drinks, sports drinks, and sweetened bottled waters, consistent with definitions reported by the Centers for Disease
In the NHANES dietary interview data, calorie derived from each consumed food or beverage item was recorded based on the quantity of food and/or beverage reported and the corresponding energy contents. We calculated total calorie intake, calorie intake from SSB and alcohol, and calorie intake from discretionary foods for each survey participant on a dietary recall day. Diet quality The Healthy Eating Index (HEI)-2010 was developed by the USDA as a measure of dietary quality in accordance with the 2010 Dietary Guidelines for Americans [2,27]. It consists of 12 components: total fruit, whole fruit, total vegetables, greens and beans, whole grains, diary, total protein foods, seafood and plant proteins, fatty acids, refined grains, sodium, and empty calories (calories from solid fats, alcohol, and added sugars). With a maximum score of 100, a higher HEI-2010 score reflects closer adherence to the Federal dietary guidelines. Details of the HEI-2010 are available elsewhere [27]. We calculated each NHANES participant’s HEI-2010 score on either 24-hour dietary recall day using the MyPyramid Equivalents Database and following the procedures established by the USDA and National Cancer Institute [28e31]. Fast-food and full-service restaurant consumption NHANES dietary interviews asked about source of each food and/or beverage item consumed on a recall day. Following An (2015) and Powell et al. (2012), fast-food and full-service restaurant consumption are defined as consumption of any food or beverage item on a given day from a fast-food restaurant (fast-food or pizza outlet) or a full-service restaurant (restaurant with waiter/waitress or bar/tavern/lounge) [32,33].
R. An / Annals of Epidemiology 26 (2016) 57e65
Other outcome variables Other outcome variables included daily quantities consumed of fruit, vegetable, total fat, saturated fat, sugar, sodium, cholesterol, and dietary fiber. Individual characteristics The following individual characteristics were reported for U.S. adults 18 years of age and above: sex, age (stratified into two age groups: 18e64 years of age and 65 years of age and above), race and/or ethnicity (non-Hispanic white, non-Hispanic African American, non-Hispanic other race or multi-race, and Hispanic), education (high school and below and college and above), household income (income to poverty ratio (IPR) < 130%, 130% IPR < 300%, and PIR 300%), and body weight status. The NHANES participants’ body height and weight were measured by stadiometer and digital scale in the MEC. Body mass index (BMI) is defined by weight in kilograms divided by height in meters squared. Body weight status was classified into underweight (BMI 18.5 kg/m2), normal weight (18.5 kg/m2 BMI < 25 kg/m2), overweight (25 kg/m2 BMI < 30 kg/m2), and obesity (BMI 30 kg/m2) based on the international classification of adult BMI values. Statistical analysis Individual fixed-effect regressions were performed based on data from day 1 and day 2 dietary interviews that provided two observations per participant. The 13 continuous outcome variables are daily consumptions of total energy (kcal), energy from SSB (kcal), energy from alcohol (kcal), energy from discretionary foods (kcal), fruit (g), vegetable (g), total fat (g), saturated fat (g), sugar (g), sodium (mg), cholesterol (mg), and fiber (g), and the HEI-2010 score for overall diet quality on a dietary recall day. The two dichotomous outcome variables are any fast-food or full-service restaurant consumption on a dietary recall day. The independent variables are three dichotomous variables for a dietary recall day of Friday, Saturday, and Sunday, with MondayeThursday as their common reference group. Compared to the conventional pooled cross-sectional regression, individual fixed-effect regression is preferred because it uses and only uses within-individual variations in dietary recall days (weekend vs. weekday) to identify the effect of weekend consumption on dietary behavior and intake, thus removing potential omitted variable bias due to differences in individual characteristics such as gender, race and/or ethnicity, eating habits, and taste preferences. Given the short time interval of 3e10 days between the two nonconsecutive dietary interviews, the possible confounders from individual characteristics that vary only in the longer term (e.g., age, education attainment, income level, and body weight status) are also removed in individual fixed-effect estimates. Linear individual fixed-effect regressions were estimated for continuous outcome variables, and logistic individual fixed-effect regressions were estimated for dichotomous outcome variables. Average treatment effects were calculated for logistic individual fixed-effect regressions to translate estimates into probabilities (prevalences). In sensitivity analysis, we used six dichotomous independent variables for a dietary recall day of Tuesday, Wednesday, Thursday, Friday, Saturday, and Sunday, with Monday as their common reference group, to replace the three dichotomous independent variables for a dietary recall day of Friday, Saturday, and Sunday and re-estimated all regressions. The estimated coefficients for weekdays (Tuesday, Wednesday, and Thursday) were mostly nonsignificant at P < .05, indicating the absence of a significant
59
difference in diet between Monday and the other weekdays. We therefore decided to exclusively report estimates from regressions with three dichotomous independent variables for a dietary recall day of Friday, Saturday, and Sunday. Individual fixed-effect regressions were estimated for the overall adult sample and subsamples stratified by sex, race and/or ethnicity (non-Hispanic other race or multirace excluded due to insufficient sample size), age group, education attainment, income level, and body weight status (underweight excluded due to insufficient sample size). To test for population heterogeneities in weekend-weekday differences in dietary intake, we included an interaction term between a stratification variable (e.g., sex, income level) and the weekend indicator in the regression (besides the main effect), so that difference across population subgroups would be indicated by a statistically significant coefficient for the interaction. Owing to the large number of statistical tests conducted (15 outcome variables 6 stratification variables ¼ 90 tests in total), we chose P < .001 as the criterion for statistical significance in between-subgroup comparisons. NHANES 2003e2012 multiyear complex survey design was accounted for in both descriptive statistics and regression analyses. All statistical procedures were performed in Stata 14.0, SE version (StataCorp, College Station, TX). NHANES was approved by the NCHS Research Ethics Review Board. This study used NHANES de-identified public data and was deemed exempt by the University of Illinois at UrbanaeChampaign Institutional Review Board. Results Table 1 reports individual characteristics of the study sample during 2003e2012. Nearly one-fifth (19.79%) of them had an IPR less than 130%. Approximately two-thirds (65.18%) were overweight or obese (BMI 25), and about a third (32.88%) were obese (BMI 30). Tables 2 and 3 report the estimated changes in diet on Friday, Saturday, and Sunday in comparison to weekdays in the whole
Table 1 Individual characteristics of adult study participants, NHANES 2003e2012 Individual characteristics Sex Male Female Race/ethnicity White, non-Hispanic African American, non-Hispanic Other race/multirace, non-Hispanic Hispanic Age group (y) 18e64 65 Education High school education and below College education and above Income to poverty ratio (IPR) IPR < 130% 130% IPR < 300% IPR 300% Body weight status Underweight (BMI < 18.5) Normal weight (18.5 BMI < 25) Overweight (25 BMI < 30) Obese (BMI 30)
Proportion (%) 48.52 (47.86e49.17) 51.48 (50.83e52.41) 70.80 11.08 5.63 12.49
(67.78e73.83) (9.43e12.72) (4.95e6.32) (10.47e14.50)
82.31 (81.34e83.28) 17.69 (16.72e18.66) 40.27 (38.38e42.17) 56.15 (54.28e58.02) 19.79 (18.35e21.23) 27.69 (26.42e28.97) 46.70 (44.65e48.75) 1.86 30.87 33.30 32.88
(1.61e2.12) (29.75e32.00) (32.30e34.30) (31.79e33.97)
Individual-level data (n ¼ 11,646) came from NHANES 2003e2012 waves. Descriptive statistics count for NHANES multiyear complex survey design. Proportions may not add to 100% due to missing values in some individual characteristics. 95% confidence intervals are in parentheses.
N
All 11646
Male 5812
Female 5834
White 5524
Black 2457
Hispanic 2930
18e64 y 8883
65 y 2763
130.71*** (92.53e168.89) 181.74*** (128.14e235.35) 95.11*** (58.20e132.03)
96.86** (25.83e167.89) 156.24** (62.14e250.34) 16.03 (41.39 to 73.45)
78.69 (7.82 to 165.19) 207.58*** (128.13e287.03) 113.11** (42.35e183.87)
118.75*** (81.02e156.48) 182.94*** (138.50e227.39) 94.91*** (59.28e130.55)
53.45* (3.86e103.04) 59.99 (10.19 to 130.18) 48.74* (5.82e91.65)
13.46** (4.47e22.45) 19.68*** (8.98e30.39) 3.73 (3.72 to 11.18)
11.45 (3.98 to 26.89) 13.82 (7.92 to 35.56) 7.99 (6.71 to 22.70)
21.56** (8.67e34.45) 15.54 (1.49 to 32.58) 24.42* (5.87e42.98)
13.87** (5.52e22.21) 18.19*** (8.77e27.61) 9.59* (2.36e16.83)
6.94 (0.60 to 14.48) 6.56 (4.87 to 17.99) 2.98 (9.85 to 3.89)
25.21*** (13.99e36.42) 42.61*** (30.60e54.62) 18.69*** (9.82e27.55)
17.79** (7.46e28.13) 51.98*** (37.36e66.59) 19.68** (18.98e34.55)
23.56* (1.14e45.98) 74.11*** (53.16e95.05) 26.13*** (16.27e36.00)
24.80*** (14.51e35.10) 50.46*** (40.15e60.77) 19.99*** (11.29e28.69)
3.15 (9.51 to 3.22) 0.34 (10.13 to 10.82) 7.41* (1.05e13.78)
38.53*** (19.89e57.16) 52.80** (22.30e83.30) 67.37*** (47.40e87.34)
5.45 (26.79 to 37.69) 12.46 (21.89 to 46.81) 26.96 (0.42 to 54.34)
3.11 (40.17 to 33.94) 41.89* (7.85e75.94) 12.72 (13.81 to 39.25)
25.44** (8.27e42.61) 43.39** (18.17e68.60) 49.99*** (30.14e56.08)
35.46* (3.13e67.78) 54.68* (12.99e96.37) 69.53*** (44.56e94.50)
4.65 (12.16 to 2.86) 9.44 (19.55 to 0.68) 9.91* (17.62 to 2.21)
0.85 (9.76 to 8.07) 12.69* (23.58 to 1.80) 9.06* (16.97 to 1.15)
9.30 (18.93 to 0.33) 31.91*** (45.28 to 18.55) 11.40* (20.14 to 2.65)
4.98 (11.22 to 1.27) 14.99*** (22.84 to 7.14) 9.05* (16.42 to 1.68)
2.83 (15.21 to 9.54) 3.17 (17.25 to 10.90) 15.29** (23.76 to 6.83)
8.12 (17.79 to 1.55) 19.67** (31.53 to 7.81) 8.49* (16.94 to 0.04)
11.68** (20.35 to 3.01) 9.06 (18.41 to 0.29) 6.62 (3.72 to 16.96)
11.71 (26.22 to 2.80) 7.87 (23.34 to 7.60) 3.29 (14.13 to 7.55)
9.13* (17.11 to 1.15) 16.56** (26.36 to 6.75) 3.92 (12.14 to 4.31)
12.83 (28.64 to 2.98) 21.13 (45.40 to 3.15) 9.22 (19.61 to 1.17)
6.45*** (4.58e8.32) 8.30*** (5.46e11.14) 4.84*** (2.86e6.82)
3.62* (0.41e6.83) 5.52* (1.11e9.94) 1.13 (3.92 to 1.66)
4.17* (0.38e7.95) 8.83*** (4.87e12.79) 5.50** (2.07e8.93)
5.94*** (4.27e7.60) 8.43*** (6.24e10.63) 4.71*** (2.87e6.54)
3.45* (0.82e6.08) 3.70 (0.49 to 7.89) 2.89* (0.60e5.18)
1.89*** (1.16e2.61) 2.87*** (1.83e3.92)
0.94 (0.11 to 1.99) 2.01** (0.53e3.50)
1.11 (0.34 to 2.57) 3.02*** (1.49e4.55)
1.75*** (1.12e2.39) 3.00*** (2.17e3.83)
0.94* (0.10e1.79) 1.03 (0.55 to 2.60)
R. An / Annals of Epidemiology 26 (2016) 57e65
Change in total energy consumption (kcal) 148.12*** 90.94*** Friday 118.05*** (86.92e149.19) (90.54e205.69) (60.66e121.22) Saturday 181.04*** 206.60*** 155.31*** (141.98e220.10) (146.46e266.73) (106.52e204.10) 98.23*** 75.96*** Sunday 88.37*** (58.32e118.42) (56.72e139.74) (40.42e111.49) Change in energy from sugar-sweetened beverage consumption (kcal) Friday 14.00*** 19.27** 9.28** (7.15e20.86) (6.88e31.67) (3.17e15.40) Saturday 18.34*** 22.44** 14.46** (10.07e26.62) (9.62e35.25) (5.52e23.40) Sunday 7.38* 10.99* 3.72 (1.61 to 13.16) (2.46 to 19.51) (2.63 to 10.07) Change in energy from alcohol consumption (kcal) Friday 22.04*** 27.91*** 17.05*** (13.32e30.77) (12.62e43.19) (11.15e22.95) Saturday 46.65*** 65.64*** 27.70*** (37.89e55.41) (51.57e79.71) (19.06e36.34) Sunday 17.82*** 24.87*** 10.45** (10.61e25.04) (14.00e35.74) (3.97e16.94) Change in energy from discretionary food consumption (kcal) Friday 29.81*** 38.31*** 21.83* (15.17e44.45) (17.66e58.97) (2.95e40.71) Saturday 48.77*** 39.07* 58.58*** (26.55e70.99) (2.24e75.90) (32.07e85.10) Sunday 55.07*** 54.56*** 55.06*** (38.49e71.65) (29.91e79.20) (35.70e74.42) Change in fruit consumption (g) Friday 4.69 6.02 3.49 (10.23 to 0.85) (13.32 to 1.29) (11.40 to 4.43) Saturday 13.90*** 11.66* 16.46** (20.92 to 6.89) (22.55 to 0.77) (25.51 to 7.40) Sunday 10.35** 13.38** 7.72* (16.67 to 4.04) (22.82 to 3.94) (14.35 to 1.09) Change in vegetable consumption (g) Friday 9.39* 7.53 11.21* (16.46 to 2.32) (17.66 to 2.60) (20.70 to 1.72) Saturday 16.76*** 22.20** 11.37* (25.71 to 7.81) (34.63 to 9.77) (21.98 to 0.76) Sunday 4.92 5.91 4.20 (11.66 to 1.81) (15.46 to 3.65) (13.92 to 5.52) Change in total fat consumption (g) Friday 5.77*** 6.80*** 4.82*** (4.33e7.20) (4.46e9.15) (2.98e6.67) Saturday 8.16*** 8.43*** 7.89*** (6.14e10.18) (5.54e11.32) (5.34e10.45) Sunday 4.39*** 4.85*** 3.88*** (2.78e6.00) (2.64e6.21) (1.83e5.92) Change in saturated fat consumption (g) Friday 1.70*** 1.83*** 1.59*** (1.15e2.26) (1.00e2.66) (0.91e2.26) Saturday 2.88*** 3.04*** 2.71*** (2.13e3.62) (1.99e4.15) (1.87e3.55)
60
Table 2 Estimated changes in daily diet among U.S. adults on Friday, Saturday, and Sunday (Monday-Thursday average as reference) by gender, race and/or ethnicity, and age group
Sunday
1.36*** (0.65e2.07)
1.76*** (1.06e2.46)
0.27 (1.21 to 0.68)
1.67** (0.53e2.81)
1.64*** (0.99e2.29)
1.25** (0.48e2.02)
2.47 (0.06 to 5.00) 5.71** (1.44e9.98) 3.40* (0.75e6.05)
3.39* (0.10e6.67) 7.17** (2.76e11.58) 3.25* (0.44e6.06)
3.19 (2.38 to 8.76) 2.51 (4.15 to 9.17) 3.63 (1.21 to 8.46)
0.79 (4.78 to 6.37) 1.15 (4.48 to 6.79) 3.67 (0.99 to 8.34)
1.91 (1.28 to 5.11) 4.65* (1.13e8.17) 3.27* (0.39e6.15)
3.15 (0.97 to 7.28) 2.91 (3.02 to 8.84) 1.90 (1.29 to 5.09)
151.42*** (75.51e227.33) 250.25*** (163.81e336.70) 145.88*** (79.60e212.16)
197.00*** (115.54e278.45) 192.81** (77.61e308.01) 121.06** (45.57e196.56)
118.02 (29.97 to 266.00) 104.77 (86.29 to 295.83) 72.55 (190.47 to 45.38)
60.09 (56.60 to 176.78) 311.42*** (188.55e434.29) 222.38*** (114.64e330.12)
168.89*** (101.51e236.25) 211.40*** (114.53e308.27) 138.83*** (69.69e207.98)
88.29 (15.98 to 192.56) 41.59 (73.70 to 156.87) 70.42 (28.22 to 169.07)
16.30** (6.41e26.19) 40.21*** (29.82e50.59) 44.53*** (35.50e53.55)
17.96** (5.22e30.70) 43.05*** (29.01e57.10) 57.09*** (46.88e67.30)
11.54 (8.19 to 31.27) 29.46* (5.62e53.29) 26.98** (10.28e43.69)
8.85 (12.15 to 29.84) 55.15*** (34.89e75.41) 32.99** (14.37e51.61)
17.64** (6.18e29.09) 48.42*** (35.88e60.97) 50.70*** (40.67e60.73)
15.21 (1.59 to 32.01) 3.51 (24.14 to 17.12) 47.86*** (32.86e62.86)
0.45* (0.84 to 0.06) 0.50 (1.04 to 0.04) 0.55* (1.01 to 0.09)
0.10 (0.55 to 0.35) 0.54 (1.12 to 0.03) 0.58** (1.00 to 0.15)
0.35 (1.00 to 0.29) 0.81* (1.53 to 0.09) 0.75** (1.30 to 0.21)
0.05 (0.62 to 0.71) 1.13* (2.11 to 0.15) 0.53 (1.31 to 0.26)
0.24 (0.63 to 0.15) 0.82** (1.31 to 0.33) 0.45* (0.87 to 0.03)
0.07 (0.83 to 0.68) 0.12 (0.96 to 0.72) 1.03*** (1.55 to 0.50)
1.20** (1.89 to 0.51) 2.64*** (3.61 to 1.66) 1.88*** (2.44 to 1.33)
1.28*** (1.88 to 0.68) 1.91*** (2.88 to 0.93) 1.93*** (2.49 to 1.37)
0.72 (1.52 to 0.07) 1.84** (3.00 to 0.67) 0.93 (1.87 to 0.02)
1.10** (1.91 to 0.28) 3.49*** (4.69 to 2.30) 2.09*** (3.05 to 1.14)
1.32*** (1.87 to 0.78) 2.42*** (3.27 to 1.57) 1.46*** (2.04 to 0.89)
0.86 (1.89 to 0.16) 1.51 (3.13 to 0.11) 3.09*** (3.82 to 2.35)
11.65*** (9.01e14.28%) 12.26*** (9.26e15.25) 2.80* (5.54 to 0.06)
9.27*** (6.33e12.21) 6.87*** (3.50e10.24) 2.67 (5.48 to 0.15)
14.31*** (10.67e17.96) 12.86*** (8.38e17.34) 9.78*** (13.83 to 5.73)
13.63*** (10.02e17.25) 12.66*** (9.03e16.29) 6.48** (2.80e10.17)
11.85*** (9.83e13.86) 10.60*** (8.39e12.82) 1.84 (3.97 to 0.29)
10.34*** (5.11e15.57) 6.36 (0.59 to 13.30) 0.75 (5.20 to 3.70)
16.02*** (13.20e18.83) 19.89*** (16.90e22.88) 9.97*** (7.17e12.77)
16.56*** (13.64e19.49) 16.27*** (12.89e19.65) 5.67*** (2.91e8.44)
15.79*** (10.97e20.62) 16.75*** (11.17e22.33) 8.77*** (3.91e13.62)
14.53*** (10.43e18.62) 20.04*** (16.25e23.83) 13.41*** (9.50e17.32)
15.58*** (13.34e17.82) 18.64*** (16.35e20.92) 8.55*** (6.24e10.86)
12.96*** (7.63e18.28) 10.95** (3.54e18.35) 6.15** (2.09e10.21)
R. An / Annals of Epidemiology 26 (2016) 57e65
1.58*** 1.79*** (1.02e2.14) (0.92e2.65) Change in sugar consumption (g) Friday 2.74* 3.00 (0.09e5.39) (1.65 to 7.64) Saturday 5.37** 4.90 (2.33e8.41) (0.33 to 10.13) Sunday 3.21** 2.78 (0.99e5.43) (0.37 to 5.93) Change in sodium consumption (mg) Friday 166.79*** 182.14** (105.17e228.40) (69.69e294.59) Saturday 205.59*** 158.82* (123.35e287.84) (22.52e295.11) Sunday 127.96*** 105.05* (65.98e189.94) (5.66e204.41) Change in cholesterol consumption (mg) Friday 17.09** 18.00* (6.98e27.19) (1.35e34.65) Saturday 43.17*** 45.99*** (32.74e53.60) (28.50e63.49) Sunday 50.05*** 55.21*** (41.24e58.87) (40.09e70.34) Change in fiber consumption (g) Friday 0.15 0.17 (0.48 to 0.18) (0.42 to 0.77) Saturday 0.67** 0.83* (1.11 to 0.23) (1.46 to 0.19) Sunday 0.54** 0.55* (0.90 to 0.19) (1.07 to 0.03) Change in Healthy Eating Index-2010 score (out of 100) Friday 1.24*** 1.28** (1.71 to 0.77) (1.96 to 0.61) Saturday 2.32*** 2.02*** (3.06 to 1.57) (2.84 to 1.19) Sunday 1.80*** 1.74*** (2.23 to 1.36) (2.35 to 1.13) Change in prevalence of fast-food restaurant consumption (%) Friday 11.64*** 11.56*** (9.76e13.53) (8.87e14.25) Saturday 10.21*** 8.30*** (8.10e12.32) (5.33e11.27) Sunday 1.64 0.54 (3.55 to 0.28) (3.22 to 2.15) Change in prevalence of full-service restaurant consumption (%) Friday 15.08*** 14.03*** (13.00e17.16) (10.96e17.10) Saturday 17.79*** 15.62*** (15.58e19.99) (12.38e18.85) Sunday 7.91*** 5.91*** (5.90e9.92) (3.04e8.78)
Individual-level data came from NHANES 2003e2012 waves. Individual fixed-effect models were used to estimate the changes in daily diet among U.S. adults on Friday, Saturday, and Sunday in comparison to that on weekdays, accounting for NHANES multiyear complex survey design. 95% confidence intervals are in parentheses. *.01 P < .05; **.001 P < .01; and P < .001.
61
N
Low education 5478
High education 5315
Low income 3444
Middle income 3417
High income 3932
Normal weight 3423
Overweight 3885
Obese 3960
105.40** (38.51e172.28) 255.18*** (184.98e325.37) 36.16 (18.48e90.81)
143.34*** (92.83e193.86) 185.00*** (119.49e250.50) 127.73*** (80.85e174.61)
142.07*** (90.92e193.22) 152.05*** (87.72e216.37) 44.97 (13.89e103.83)
111.94*** (59.84e164.04) 217.68*** (150.85e284.52) 80.48** (23.69e137.27)
107.13*** (49.77e164.48) 185.94*** (115.67e256.22) 133.01*** (83.82e182.21)
4.04 (8.94 to 17.02) 13.99 (2.40 to 30.37) 0.29 (e11.08 to 11.66)
12.56* (1.90e23.22) 19.52** (5.74e33.31) 7.06 (e0.88 to 14.99)
15.93* (3.70e28.15) 19.66** (5.62e33.70) 1.09 (e8.35 to 10.53)
19.23*** (9.51e28.96) 19.69* (4.82e34.56) 12.87** (3.33e22.41)
8.05 (3.94 to 20.04) 16.82* (2.13e31.50) 8.09 (e1.30 to 17.47)
17.75* (2.26e33.24) 56.67*** (38.78e74.56) 12.23** (3.70e20.75)
27.81*** (14.16e41.46) 47.30*** (32.97e61.63) 23.04*** (10.56e35.52)
35.25*** (21.21e49.29) 54.44*** (40.64e68.25) 17.57** (5.63e29.51)
24.37** (10.25e38.49) 47.54*** (31.37e63.70) 18.65** (8.12e29.18)
9.31 (e2.29 to 20.90) 41.28*** (27.14e55.42) 18.72** (7.82e29.63)
22.96 (e10.31 to 56.22) 71.67** (28.53e114.81) 34.35* (4.82e63.87)
41.26** (18.60e63.91) 50.20* (9.54e90.86) 79.20*** (52.79e105.60)
36.79* (6.06e67.52) 24.03 (e10.93 to 58.99) 31.73* (3.51e59.96)
35.89* (8.51e63.27) 78.05*** (38.63e117.48) 60.29*** (22.16e55.69)
17.29 (e9.48 to 44.05) 49.37** (13.86e84.88) 72.35*** (46.98e97.72)
4.55 (e5.34 to 14.44) e11.92 (e24.40 to 0.55) e8.51* (e15.26 to e1.75)
e8.33 (e17.60 to 0.93) e16.04* (e28.20 to e3.89) e11.32* (e22.27 to e0.37)
e7.37 (e20.85 to 6.10) e20.64*** (e31.37 to e9.90) e17.54*** (e26.95 to e8.13)
e3.13 (e13.85 to 7.58) e16.94* (e30.73 to e3.16) e7.96 (e16.37e0.44)
e3.18 (e9.84 to 3.49) e5.13 (e16.70 to 6.45) e7.88 (e19.06e3.31)
e6.80 (e18.14 to 4.54) e15.67* (e28.08 to e3.26) e0.68 (e10.15 to 8.78)
e10.94 (e22.94 to 1.05) e19.94* (e35.14 to e4.74) e9.49 (e21.76 to 2.77)
e0.39 (e11.34 to 10.56) e27.06*** (e41.15 to e12.96) 1.68 (e10.17 to 13.53)
e21.27*** (e32.76 to e9.78) e13.21 (e31.44e5.02) e12.83* (e23.50 to e2.16)
e8.98 (e20.79 to 2.83) e13.17 (e27.14 to 0.80) e6.24 (e17.98 to 5.50)
5.43*** (2.50e8.35) 10.55*** (6.79e14.31) 1.65 (e1.40e4.70)
7.05*** (4.63e9.47) 8.75*** (5.31e12.19) 6.86*** (4.31e9.41)
5.34*** (2.59e8.09) 6.55*** (3.26e9.85) 1.90 (e0.87e4.67)
4.50*** (2.16e6.83) 9.61*** (6.23e13.00) 4.01* (0.90-7.12)
7.38*** (4.49e10.28) 8.64*** (5.00e12.27) 7.07*** (4.83e9.31)
1.14* (0.11e2.16) 3.80*** (2.32e5.27)
2.34*** (1.49e3.19) 3.09*** (1.86e4.31)
1.31** (0.34e2.28) 2.21*** (1.01e3.42)
1.64** (0.64e2.65) 3.57*** (2.29e4.85)
2.19*** (1.14e3.24) 2.86*** (1.60e4.13)
R. An / Annals of Epidemiology 26 (2016) 57e65
Change in total energy consumption (kcal) 133.15*** 103.76** Friday 101.28*** (46.20e156.36) (96.33e169.96) (32.65e174.87) Saturday 183.37*** 180.10*** 87.08* (118.21e248.53) (127.64e232.57) (15.75e158.41) 86.99*** 57.45 Sunday 84.61*** (41.79e127.43) (46.29e127.70) (10.34e125.23) Change in energy from sugar-sweetened beverage consumption (kcal) Friday 20.02*** 9.79* 32.19** (9.56e30.49) (0.66e18.91) (14.08e50.30) Saturday 17.67** 17.79** 12.54 (6.86e28.49) (5.30e30.27) (2.43 to 27.51) Sunday 6.68 6.68 16.17* (3.85 to 17.20) (e1.26 to 14.62) (1.81e30.54) Change in energy from alcohol consumption (kcal) Friday 13.15 28.62*** 19.32** (e1.44 to 27.74) (18.74e38.50) (8.01e30.63) Saturday 41.43*** 50.85*** 38.32*** (29.40e53.45) (37.01e64.69) (23.33e53.32) Sunday 14.38** 20.69*** 8.33 (5.13e23.62) (11.46e29.92) (e1.87e18.54) Change in energy from discretionary food consumption (kcal) Friday 11.18 39.22*** 19.39 (e13.35 to 35.72) (19.58e58.85) (e12.94 to 51.72) Saturday 38.28** 58.02*** 6.50 (10.45e66.10) (27.28e88.77) (e30.41 to 43.41) Sunday 43.67** 60.00*** 22.75 (37.30e63.25) (37.99e82.01) (e1.59 to 47.08) Change in fruit consumption (g) Friday e6.66 e3.52 e5.06 (e15.46 to 2.14) (e11.36 to 4.32) (e13.91 to 3.79) Saturday e14.02** e15.50** e13.90** (e23.88 to e4.15) (e26.11 to e4.89) (e24.10 to e3.70) Sunday e7.48* e12.57** e11.65** (e14.53 to e0.42) (e21.57 to e3.57) (e20.15 to e3.16) Change in vegetable consumption (g) Friday e13.10* e7.35 e4.11 (e23.83 to e2.37) (e16.33 to 1.63) (e14.96 to 6.75) Saturday e9.81 e22.72** e7.28 (e21.84 to 2.22) (e35.29 to e10.14) (e19.72e5.17) Sunday e1.34 e7.70 2.91 (e9.18 to 6.49) (e17.49 to 2.09) (e9.78 to 15.60) Change in total fat consumption (g) Friday 5.21*** 6.29*** 4.30** (2.93e7.49) (4.32e8.26) (1.32e7.29) Saturday 8.80*** 7.62*** 3.94* (5.84e11.76) (4.83e10.41) (0.61e7.26) Sunday 3.41** 4.86*** 2.31 (1.07e5.76) (2.76e6.97) (e0.65e5.27) Change in saturated fat consumption (g) Friday 1.30** 2.05*** 1.26* (0.39e2.20) (1.34e2.77) (0.22e2.29) Saturday 2.97*** 2.82*** 1.32* (1.93e4.00) (1.78e3.87) (0.20e2.44)
62
Table 3 Estimated changes in daily diet among U.S. adults on Friday, Saturday, and Sunday (Monday-Thursday average as reference) by education, income and body weight status
Sunday
1.10* (0.21e1.99)
0.48 (e0.79 to 1.76)
2.42*** (1.51e3.32)
0.63 (e0.22 to 1.48)
1.75** (0.66e2.84)
2.40*** (1.53e3.26)
5.95 (e0.16 to 12.06) e0.89 (e7.36 to 5.57) 4.87 (e0.25 to 9.99)
3.04 (e2.46 to 8.54) 6.65* (0.33e12.96) 0.86 (e3.36 to 5.07)
1.86 (e1.35 to 5.08) 6.95** (1.97e11.94) 3.63* (0.39e6.87)
5.71* (1.05e10.38) 4.13 (e1.35 to 9.62) 0.97 (e2.79 to 4.72)
3.69 (e0.15 to 7.53) 8.22** (2.45e13.99) 3.47* (0.20e6.74)
0.19 (e4.22 to 4.60) 4.11 (e0.69 to 8.92) 4.95** (1.26e8.65)
101.71 (e38.88 to 242.29) 38.88 (e119.51 to 197.28) 69.38 (e52.34 to 191.10)
144.07* (26.11e262.04) 295.98*** (170.19e421.77) 85.14 (e40.44 to 210.72)
222.54*** (106.17e338.91) 224.89** (81.60e368.18) 179.87*** (96.01e263.74)
158.72** (50.86e266.58) 45.79 (e97.82 to 189.39) 85.70 (e12.72 to 184.12)
167.35** (58.16e276.55) 314.90*** (181.25e448.55) 108.10* (6.72e209.47)
173.16** (54.75e291.56) 251.12** (108.20e394.03) 169.97** (62.74e277.20)
11.16 (e5.25 to 27.58) 41.60*** (22.47e60.74) 22.98** (7.05e38.91)
16.88 (e1.33 to 35.10) 44.03*** (24.91e63.16) 44.91*** (28.83e60.99)
19.85** (5.71e33.99) 44.74*** (25.41e64.07) 64.12*** (49.58e78.66)
14.14 (e2.70 to 30.97) 38.32*** (19.42e57.23) 47.56*** (34.29e60.83)
18.13* (0.90e35.36) 46.49*** (28.24e64.73) 53.23*** (36.89e69.56)
18.30* (2.57e34.03) 47.18*** (30.87e63.49) 49.26*** (32.64e65.89)
e0.28 (e0.96 to 0.40) e0.75* (e1.48 to e0.02) e0.41 (e1.09 to 0.27)
0.14 (e0.60 to 0.87) 0.04 (e0.81 to 0.88) e0.87** (e1.35 to e0.39)
e0.23 (e0.80 to 0.34) e1.05** (e1.71 to e0.39) e0.39 (e1.00 to 0.23)
e0.37 (e0.95 to 0.21) e1.31*** (e1.96 to e0.67) e0.93* (e1.64 to e0.23)
e0.09 (e0.73 to 0.55) e0.79 (e1.67 to 0.09) e0.74* (e1.37 to e0.11)
0.04 (e0.54 to 0.63) 0.01 (e0.85 to 0.86) e0.16 (e0.80 to 0.48)
e0.76 (e1.74 to 0.23) e1.76** (e2.90 to e0.61) e1.36** (e2.12 to e0.59)
0.06 (e0.83 to 0.95) e1.84** (e2.92 to e0.76) e1.26** (e2.08 to e0.45)
e2.09*** (e2.80 to e1.37) e2.75*** (e3.84 to e1.67) e2.09*** (e2.82 to e1.37)
e1.52** (e2.37 to e0.66) e2.12** (e3.28 to e0.96) e1.42** (e2.28 to e0.55)
e1.22** (e2.06 to e0.38) e2.75*** (4.01 to 1.50) e2.22*** (e2.94 to e1.51)
e0.96* (e1.81 to e0.11) 2.22*** (3.32 to 1.12) e1.78*** (e2.51 to 1.06)
11.36*** (7.79e14.93) 8.66*** (4.81e12.51) 0.55 (4.05 to 2.94)
9.53*** (6.06e13.01) 11.67*** (7.77e15.58) 0.38 (3.99 to 3.24)
12.40*** (9.25e15.56) 9.57*** (5.90e13.23) 4.21* (7.47 to 0.95)
10.56*** (6.91e14.22) 6.83** (2.79e10.87) 3.13 (6.67 to 0.31)
11.64*** (8.31e14.97) 10.16*** (6.49e13.84) 1.19 (2.18 to 4.56)
12.44*** (9.37e15.50) 12.45*** (8.94e15.96) 3.02 (6.23 to 0.19)
11.65*** (6.85e16.44) 15.05*** (10.42e19.68) 9.58*** (5.33e13.82)
16.17*** (12.37e19.97) 18.92*** (14.77e23.06) 5.98** (2.26e9.70)
15.20*** (12.10e18.31) 17.10*** (13.63e20.58) 8.10*** (5.00e11.21)
16.16*** (12.34e19.99) 18.56*** (14.60e22.52) 5.47** (1.81e9.13)
15.61*** (12.05e19.16) 17.56*** (13.79e21.33) 8.47*** (5.11e11.84)
13.70*** (10.12e17.28) 17.55*** (13.63e21.47) 9.11*** (5.56e12.66)
R. An / Annals of Epidemiology 26 (2016) 57e65
0.85* 2.04*** (0.00e1.71) (1.30e2.78) Change in sugar consumption (g) Friday 2.63 2.77 (e1.04 to 6.30) (e0.67 to 6.22) Saturday 3.46 6.67** (e1.35 to 8.27) (1.98e11.36) Sunday 3.60* 2.54 (0.19e7.00) (e0.56 to 5.64) Change in sodium consumption (mg) Friday 155.68** 181.80*** (59.26e252.09) (97.16e266.45) Saturday 232.56*** 183.68** (119.24e345.87) (72.66e294.70) Sunday 159.74** 111.74** (55.54e263.95) (39.03e184.44) Change in cholesterol consumption (mg) Friday 17.48* 18.72** (3.81e31.14) (6.09e31.35) Saturday 43.08*** 43.46*** (26.21e59.94) (29.17e57.76) Sunday 45.74*** 54.24*** (32.13e59.34) (41.68e66.80) Change in fiber consumption (g) Friday 0.10 e0.33 (e0.40 to 0.59) (e0.79 to 0.12) Saturday e0.29 e0.93** (e0.98 to 0.40) (e1.57 to e0.30) Sunday e0.13 e0.79** (e0.57 to 0.30) (e1.31 to e0.27) Change in Healthy Eating Index-2010 score (out of 100) Friday e0.21 e1.98*** (e1.01 to 0.59) (e2.65 to e1.32) Saturday e2.17*** e2.55*** (e3.30 to e1.04) (e3.48 to e1.61) Sunday e0.99** e2.36*** (e1.56 to e0.42) (e3.00 to e1.71) Change in prevalence of fast-food restaurant consumption (%) Friday 12.79*** 10.34*** (9.95e15.64) (7.59e13.09) Saturday 10.93*** 8.74*** (7.79e14.06) (5.59e11.90) Sunday 2.72 2.43 (5.59 to 0.15) (5.27 to 0.41) Change in prevalence of full-service restaurant consumption (%) Friday 15.73*** 14.39*** (12.41e19.06) (11.54e17.24) Saturday 17.34*** 17.72*** (13.89e20.80) (14.60e20.85) Sunday 7.75*** 7.73*** (4.61e10.88) (4.96e10.51)
Individual-level data came from NHANES 2003e2012 waves. Individual fixed-effect models were used to estimate the changes in daily diet among U.S. adults on Friday, Saturday, and Sunday in comparison to that on weekdays, accounting for NHANES multiyear complex survey design. 95% confidence intervals are in parentheses. *.01 P < .05; **.001 P < .01; and ***P < .001.
63
64
R. An / Annals of Epidemiology 26 (2016) 57e65
study sample and by gender, race and/or ethnicity, age group, education, income, and body weight status using individual fixedeffect models. Diet on weekends was less healthful than that on weekdays, and among the three weekend days, diet quality on Saturday was the poorest. Compared with weekday average consumption, consumption on Saturday was associated with an increase in daily intakes of total energy by 181.04 kcal (95% confidence interval ¼ 141.98e220.10), energy from SSB by 18.34 kcal (10.07e26.62), energy from alcohol by 46.65 kcal (37.89e55.41), energy from discretionary foods by 48.77 kcal (26.55e70.99), total fat by 8.16 g (6.14e10.18), saturated fat by 2.88 g (2.13e3.62), sugar by 5.37 g (2.33e8.41), sodium by 205.59 mg (123.35e287.84), and cholesterol by 43.17 mg (32.74e53.60), and a decrease in intakes of fruit by 13.90 g (20.92 to 6.89), vegetable by 16.76 g (25.71 to 7.81), and fiber by 0.67 g (1.11 to 0.23). Compared with weekday average, the estimated reduction in HEI-2010 score was 2.32 (3.06 to 1.57) on Saturday. The prevalence of fast-food restaurant consumption was 10.21 percentage point (8.10e12.32) higher and that of full-service restaurant consumption 17.79 percentage point (15.58e19.99) higher on Saturday than on weekdays. In general, weekend consumption was found to be associated with higher intakes of total energy and certain foods/nutrients that warrant significant reduction by dietary guidelines and lower intakes of fruit/vegetable and fiber among all population subgroups under examination. However, substantial heterogeneities in the relationship between diet and day of the week (weekends vs. weekdays) were present across these subgroups. Compared with their female counterparts, weekend-weekday differences in intakes of total energy, energy from SSB and alcohol, total fat, saturated fat, and cholesterol were larger among male adults. Compared with older adults 65 years of age and above, weekendweekday differences in intakes of total energy, energy from SSB, alcohol and discretionary foods, total fat, saturated fat, sodium and cholesterol, and the prevalence of full-service restaurant consumption were larger among young and middle-aged adults 18e64 years of age. Compared with those with high school education and below, weekend-weekday differences in intakes of energy from alcohol and discretionary foods, cholesterol, and fiber were larger among those with college education and above. Compared with people with lower income, weekend-weekday differences in cholesterol intake and overall diet quality measured by the HEI-2010 were larger among those at the highest income level. Compared with their racial and/or ethnic counterparts, weekend-weekday difference in energy intake from discretionary foods was larger among non-Hispanic whites; whereas weekend-weekday difference in fruit intake was larger among Hispanics. All the above reported between-subgroup comparisons had a P value less than .001.
Discussion This study assessed weekend-weekday differences in diet among U.S. adults using data from NHANES 2003e2012 waves. It confirmed findings from previous literature on the increased discretionary calorie intake during weekend days [7e20]. Weekend consumption was found to be associated with higher intake of total energy and certain foods and/or nutrients that warrant substantial reduction in accordance with dietary guidelines and lower intake of fruit and/or vegetable and fiber. Diet quality on Saturday tended to be the poorest among the three weekend days. Considerable heterogeneities in the relationship between diet and day of the week were present across population subgroups by sex, age, race and/or ethnicity, education, income, and body weight status.
The elevated prevalence of fast-food and full-service restaurant consumption may contribute to the poorer diet quality on Saturdays. An (2015) reported both fast-food and full-service restaurant use to be associated with substantial increase in daily total energy consumption and intakes of total fat, saturated fat, cholesterol, and sodium [32]. Based on data from the USDA’s Nationwide Food Consumption Survey, U.S. adults were found to consume more eggs, bacon, meat, and alcoholic beverages but less cereal on weekends than on weekdays [15]. Coinciding with the less healthful eating style, reduced physical activity level and increased sedentary behavior tend to be more prevalent on weekends, which jointly lead to weight gain [4,12,18]. Gender and age differences in the relationship between diet and day of the week have been documented [15]. Thompson et al. (1986) [15] found weekend-weekday differences in diet to be more pronounced in women than in men and in younger adults than in older adults. The underlying determinants of these population heterogeneities remain to be understood but could partially correlate with differences in work pattern, social role, food preference, diet habit, and affordability. The strength of this study included use of nationally representative data, a comprehensive set of measures on dietary intake and dining behavior, stratified estimates on each of the three weekend days (Friday, Saturday, and Sunday) and various population subgroups, and a fixed-effect identification strategy. Nevertheless, a few limitations of this study should be noted. Dietary intakes in NHANES were self-reported and subject to measurement error and social desirability bias [34]. Individual fixed-effect estimator eliminated confounding bias from unobservable factors that remained constant within-participant between the two dietary interviews but could not control for more transient factors like daily variations in physical activity, appetite, or emotions. Compared with pooled cross-sectional estimators, the individual fixed-effect estimator is underpowered (less precision and larger standard error) because only a subsample consisting of individuals whose two dietary recall days occupied a weekday and a weekend contributed to the effect estimation, whereas those with both recalls on weekdays or weekends did not. In conclusion, American’s weekend consumption was associated with increased calorie intake and poorer diet quality. Policy interventions and public health campaigns are warranted to improve awareness, change social norms, and promote healthful eating on the weekend.
References [1] U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2020. http://www.healthypeople.gov/ 2020/topics-objectives/topic/nutrition-and-weight-status. Accessed October 1, 2015. [2] U.S. Department of Agriculture, U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2010. Washington, DC: U.S. Government Printing Office; 2010. [3] Krebs-Smith SM, Guenther PM, Subar AF, Kirkpatrick SI, Dodd KW. Americans do not meet federal dietary recommendations. J Nutr 2010;140(10):1832e8. [4] McCarthy S. Weekly patterns, diet quality and energy balance. Physiol Behav 2014;134(1):55e9. [5] Orfanos P, Naska A, Trichopoulos D, Slimani N, Ferrari P, van Bakel M, et al. Eating out of home and its correlates in 10 European countries. The European Prospective Investigation into Cancer and Nutrition (EPIC) study. Public Health Nutr 2007;10(12):1515e25. [6] Yang PH, Black JL, Barr SI, Vatanparast H. Examining differences in nutrient intake and dietary quality on weekdays versus weekend days in Canada. Appl Physiol Nutr Metab 2014;39(12):1413e7. [7] Rothausen BW, Matthiessen J, Hoppe C, Brockhoff PB, Andersen LF, Tetens I. Differences in Danish children’s diet quality on weekdays v. weekend days. Public Health Nutr 2012;15(9):1653e60. [8] Haines PS, Hama MY, Guilkey DK, Popkin BM. Weekend eating in the United States is linked with greater energy, fat, and alcohol intake. Obes Res 2003;11(8):945e9.
R. An / Annals of Epidemiology 26 (2016) 57e65 [9] Svensson A, Larsson C, Eiben G, Lanfer A, Pala V, Hebestreit A, et al. European children’s sugar intake on weekdays versus weekends: the IDEFICS study. Eur J Clin Nutr 2014;68(7):822e8. [10] Yang PHW, Black JL, Barr SI, Vatanparast H. A temporal analysis of Canadian dietary choices using the Canadian Community Health Survey Cycle 2.2: does nutrient intake and diet quality vary on weekends versus weekdays. FASEB J 2013;27:354.5. [11] Hanson KL, Olson CM. School meals participation and weekday dietary quality were associated after controlling for weekend eating among U.S. school children aged 6 to 17 years. J Nutr 2013;143(5):714e21. [12] Hart CN, Raynor HA, Osterholt KM, Jelalian E, Wing RR. Eating and activity habits of overweight children on weekdays and weekends. Int J Pediatr Obes 2011;6(5-6):467e72. [13] Rothausen BW, Matthiessen J, Andersen LF, Brockhoff PB, Tetens I. Dietary patterns on weekdays and weekend days in 4-14-year-old Danish children. Br J Nutr 2013;109(9):1704e13. [14] Lachapelle D, Gamache C, Brodeur JM, Sévigny J. Frequency of food consumption in children: weekdays versus weekends. J Can Dent Assoc 1989;55(1):61e5. [15] Thompson FE, Larkin FA, Brown MB. Weekend-weekday differences in reported dietary intake: The nationwide food consumption survey, 1977-78. Nutr Res 1986;6(6):647e62. [16] Rhodes DG, Cleveland LE, Murayi T, Moshfegh AJ. The effect of weekend eating on nutrient intakes and dietary patterns. FASEB J 2007;21:835.2. [17] Hartline-Grafton HL, Rose D, Johnson CC, Rice JC, Webber LS. The influence of weekday eating patterns on energy intake and BMI among female elementary school personnel. Obesity (Silver Spring) 2010;18(4):736e42. [18] Racette SB, Weiss EP, Schechtman KB, Steger-May K, Villareal DT, Obert KA, et al. Influence of weekend lifestyle patterns on body weight. Obesity (Silver Spring) 2008;16(8):1826e30. [19] Cullen KW, Lara KM, de Moor C. Children’s dietary fat intake and fat practices vary by meal and day. J Am Diet Assoc 2002;102(12):1773e8. [20] Kim MJ. Comparison of food and nutrient intake between weekday and weekend for elementary and middle school students by gender in Busan and some parts of Kyungsangnamdo. J Nutr Health 2013;46(4):332e45. [21] Faith MS, Kral TVE, Institute of Medicine Committee on Assessing Interactions Among Social, Behavioral, and Genetic Factors in Health. Social environmental and genetic influences on obesity and obesity-promoting behaviors: fostering research integration. In: Hernandez LM, Blazer DG,
[22]
[23] [24]
[25] [26]
[27]
[28]
[29] [30]
[31]
[32]
[33] [34]
65
editors. Genes, Behavior, and the Social Environment: Moving Beyond the Nature/Nurture Debate. Washington, DC: National Academies Press; 2006 http://www.ncbi.nlm.nih.gov/books/NBK19935/. Accessed October 1, 2015. Centers for Disease Control and Prevention, National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD: U.S: Department of Health and Human Services; 2014 http:// www.cdc.gov/nchs/nhanes.htm. Accessed October 1, 2015. Ogden CL, Kit BK, Carroll MD, Park S. Consumption of sugar drinks in the United States, 2005-2008. NCHS Data Brief 2011;71:1e8. U.S. National Cancer Institute. Sources of beverage intakes among the US population, 2005-06. http://riskfactor.cancer.gov/diet/foodsources/beverages/. Accessed October 1, 2015. National Health and Medical Research Council. Australian Dietary Guidelines. Canberra, Australia: NHMRC; 2013. Bleich SN, Wolfson JA, Vine S, Wang YC. Diet-beverage consumption and caloric intake among US adults, overall and by body weight. Am J Public Health 2014;104(3):e72e8. Guenther PM, Casavale KO, Reedy J, Kirkpatrick SI, Hiza HA, Kuczynski KJ, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet 2013;113(4):569e80. Bowman SA, Friday JE, Moshfegh AJ. MyPyramid Equivalents Database, 2.0 for USDA Survey Foods, 2003-2004. Beltsville, MD: U.S. Department of Agriculture; 2008. National Cancer Institute. HEI tools for researchers. http://appliedresearch. cancer.gov/hei/tools.html. Accessed October 1, 2015. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. Healthy Eating Index. http://www.cnpp.usda.gov/healthyeatingindex. Accessed October 1, 2015. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. Archived projects. http://www.cnpp.usda.gov/OtherProjects.htm. Accessed October 1, 2015. An R. Fast-Food and full-service restaurant consumption and daily energy and nutrient intakes in U.S. adults. Eur J Clin Nutr 2015. http://dx.doi.org/10.1038/ ejcn.2015.104 [Epub ahead of print]. Powell LM, Nguyen BT, Han E. Energy intake from restaurants: demographics and socioeconomics, 2003-2008. Am J Prev Med 2012;43(5):498e504. Larson MR. Social desirability and self-reported weight and height. Int J Obes Relat Metab Disord 2000;24(5):663e5.
Contents lists available at ScienceDirect
Annals of Epidemiology journal homepage: www.annalsofepidemiology.org
Original article
Weekend-weekday differences in diet among U.S. adults, 2003e2012 Ruopeng An PhD * Department of Kinesiology and Community Health, College of Applied Health Sciences, University of Illinois at Urbana-Champaign, Champaign
a r t i c l e i n f o
a b s t r a c t
Article history: Received 16 July 2015 Accepted 19 October 2015 Available online 28 October 2015
Purpose: Dietary patterns differ by day of the week. This study examined weekend-weekday differences in diet among U.S. adults. Methods: Nationally representative data of 11,646 adults 18 years of age and above from the National Health and Nutrition Examination Survey 2003e2012 waves were analyzed. Individual fixed-effect regressions were performed using data from two nonconsecutive 24-hour dietary recalls. Results: Weekend diet was less healthful than weekday, with diet on Saturday the worst. Compared with weekday consumption, consumption on Saturday was associated with an increase in daily intakes of total energy by 181.04 kcal, energy from sugar-sweetened beverages 18.34 kcal, energy from alcohol 46.65 kcal, energy from discretionary foods 48.77 kcal, total fat 8.16 g, saturated fat 2.88 g, sugar 5.37 g, sodium 205.59 mg, and cholesterol 43.17 mg, a decrease in intakes of fruit by 13.90 g, vegetable 16.76 g, and fiber 0.67 g, a decrease in the Healthy Eating Index-2010 score by 2.32, and an increase in the prevalence of fast-food and full-service restaurant consumption by 10.21% and 17.79%, respectively. Weekend-weekday differences in diet varied by sex, age, race and/or ethnicity, education, income, and body weight status. Conclusions: Americans’ weekend consumption was associated with increased calorie intake and poorer diet quality. Ó 2016 Elsevier Inc. All rights reserved.
Keywords: 24-hour dietary recall Diet quality Weekend Weekday
Introduction Improving diet quality is a key health promotion strategy [1]. Since 1980, a major theme of the U.S. federal dietary guidelines has been to increase consumption of nutrient-rich foods and reduce consumption of energy-dense foods [2]. However, a large majority of the American population fails to meet these guidelines, with insufficient consumption of nutrient-rich foods and excessive discretionary calorie intake [3]. Human activities consist of many repeated patterns and habitual behaviors. People’s day-to-day lives are punctuated by work, education, and other routines. Alternations of these routines such as not working in paid employment or attending school on the weekend may lead to changes in dietary patterns [4]. Eating out tends to be more common in weekends than in weekdays, and weekend food consumption has been consistently linked to
The author has no conflict of interest to declare. * Corresponding author. 1206 South 4th Street, Tel.: þ1-217-244-0966; fax: þ1-217-333-2766. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.annepidem.2015.10.010 1047-2797/Ó 2016 Elsevier Inc. All rights reserved.
Champaign,
IL
61820.
increased discretionary calorie intake from energy-dense, nutrientpoor foods [5,6]. Previous research on weekend-weekday differences in dietary patterns covered diverse populations (e.g., children [7e13], adolescents [7,8,10,11,13,14], young adults [8,10,15e17], middle-aged and older adults [8,10,15e18], overweight/obese individuals [12,16]) residing in different countries (e.g., Canada [10], Denmark [7,13], United States [8,11,17e19], Korea [20], and other countries [9,12]). In general, these studies reported less healthful dietary intakes in weekends in comparison to weekdays. However, four issues remain to be adequately addressed. First, a large majority of studies used cross-sectional methods [7e13,15,17] and/or nonnationally representative data [12,14,16e20], with findings susceptible to confounding bias (due to failure in controlling for between-individual differences in characteristics such as food or taste preferences) and compromised in generalizability to the general population. Second, dietary outcomes were typically limited to total calories, fats and/or a few individual food items [7e9,12e20], whereas overall diet quality and source of foods (e.g., food/beverage from fast-food/full-service restaurant) were underexamined. Third, most studies grouped Friday, Saturday, and Sunday together (or Saturday and Sunday together) in analysis [7e20],
58
R. An / Annals of Epidemiology 26 (2016) 57e65
which overlooked the distinctive dietary pattern on each of these three weekend days. Finally, population heterogeneities in weekend-weekday differences in dietary patterns may be present but are largely undocumented [21]. Built on previous literature, this study examined weekendweekday differences in diet among U.S. adults using data from a nationally representative survey. To our knowledge, this study serves as the first attempt to study weekend-weekday differences in diet using a panel data approach, which used within-individual variations in diet between the two nonconsecutive 24-hour dietary recall days (one on weekday and one on weekend) and thus addressed the confounding issue due to unobservable individual characteristics like taste preferences. We hypothesized that weekend diet was less healthful than weekday diet, and the differences were heterogeneous across population subgroups. Methods Survey setting and participants Individual-level data came from the National Health and Nutrition Examination Survey (NHANES) 2003e2004, 2005e2006, 2007e2008, 2009e2010, and 2011e2012 waves. Since 1999, NHANES has been conducted continuously in 2-year cycles and focused on a variety of health and nutrition measurements [22]. A multistage probability sampling design is used to select participants representative of the civilian, noninstitutionalized U.S. population.
Control and Prevention and the National Cancer Institute [23,24]. Milk (plain or flavored) and 100% fruit juice were excluded. Alcohol includes beers and ales, cordials and liqueurs, cocktails, wines and distilled liquors, consistent with the USDA Food and Nutrient Database for Dietary Studies food and/or beverage categorization. In the NHANES 2011e2012 wave, SSB and alcohol consist of 48 and 33 reported beverage items, respectively. The number of reported items in each beverage category differed only slightly across survey waves. Discretionary food consumption The discretionary food category identifies energy-dense, nutrient-poor food products that do not belong to the main food groups or necessarily contain essential nutrients that the human body requires but may add diversity [25]. Foods in this category may be consumed “sometimes in small amounts by those who are physically active, but are not a necessary part of the diet” [25]. Following Bleich et al. (2014) [26], specific food items in the discretionary food category include cookies, pies, ice cream, confectionery, chocolate, other desserts (e.g., custards, puddings, mousse, gelatin dessert), sweet rolls, waffles, cakes, pastries (e.g., crepes, cream puffs, strudels, croissants, muffins, sweet breads), biscuits, hush puppies, chips, popcorn, pretzels, party mixes, and fries. In the NHANES 2011e2012 wave, the discretionary food category consists of 661 reported food items. The number of reported items in the discretionary food category differed only slightly across survey waves.
Dietary interview Calorie intake Except for the NHANES 1999e2000 wave where all respondents were asked to complete a single 24-hour dietary recall interview, all subsequent waves incorporated two dietary recalls, with the first collected in-person and the second by telephone 3 to 10 days later. After the dietary interview, the calorie and nutrient contents of each reported food and/or beverage item were systematically coded with the U.S. Department of Agriculture (USDA) Food and Nutrient Database for Dietary Studies. Access restrictions apply to the day 2 dietary recall data collected in the NHANES 2001e2002 wave, whereas dietary data for both recall days are released to the public for all subsequent waves. Among the 23,865 U.S. adults, 18 years of age and above who participated in the NHANES 2003e2012 waves, 1352 (5.7%) who were pregnant, lactating, and/or on a special diet to lose weight at the time of interview were excluded. Of the remaining 22,513 adults who had complete dietary measures, only those whose two dietary recall days occupied a weekday and a weekend contributed to the individual fixed-effect estimation. Therefore, the effective sample sizes (number of individuals) were 11,646 for all adults, and 5812, 5834, 5524, 2457, 2930, 8883, 2763, 5478, 5315, 3444, 3417, 3932, 3423, 3885, and 3960 for male adults, female adults, white adults, African American adults, Hispanic adults, adults 18e64 years of age, adults 65 years of age and above, adults with high school or lower education, adults with college and above education, adults of low household income level, adults of middle household income level, adults of high household income level, normal weight adults, overweight adults, and obese adults, respectively. Sugar-sweetened beverage and alcohol consumption Sugar-sweetened beverage (SSB) includes sodas, fruit drinks, energy drinks, sports drinks, and sweetened bottled waters, consistent with definitions reported by the Centers for Disease
In the NHANES dietary interview data, calorie derived from each consumed food or beverage item was recorded based on the quantity of food and/or beverage reported and the corresponding energy contents. We calculated total calorie intake, calorie intake from SSB and alcohol, and calorie intake from discretionary foods for each survey participant on a dietary recall day. Diet quality The Healthy Eating Index (HEI)-2010 was developed by the USDA as a measure of dietary quality in accordance with the 2010 Dietary Guidelines for Americans [2,27]. It consists of 12 components: total fruit, whole fruit, total vegetables, greens and beans, whole grains, diary, total protein foods, seafood and plant proteins, fatty acids, refined grains, sodium, and empty calories (calories from solid fats, alcohol, and added sugars). With a maximum score of 100, a higher HEI-2010 score reflects closer adherence to the Federal dietary guidelines. Details of the HEI-2010 are available elsewhere [27]. We calculated each NHANES participant’s HEI-2010 score on either 24-hour dietary recall day using the MyPyramid Equivalents Database and following the procedures established by the USDA and National Cancer Institute [28e31]. Fast-food and full-service restaurant consumption NHANES dietary interviews asked about source of each food and/or beverage item consumed on a recall day. Following An (2015) and Powell et al. (2012), fast-food and full-service restaurant consumption are defined as consumption of any food or beverage item on a given day from a fast-food restaurant (fast-food or pizza outlet) or a full-service restaurant (restaurant with waiter/waitress or bar/tavern/lounge) [32,33].
R. An / Annals of Epidemiology 26 (2016) 57e65
Other outcome variables Other outcome variables included daily quantities consumed of fruit, vegetable, total fat, saturated fat, sugar, sodium, cholesterol, and dietary fiber. Individual characteristics The following individual characteristics were reported for U.S. adults 18 years of age and above: sex, age (stratified into two age groups: 18e64 years of age and 65 years of age and above), race and/or ethnicity (non-Hispanic white, non-Hispanic African American, non-Hispanic other race or multi-race, and Hispanic), education (high school and below and college and above), household income (income to poverty ratio (IPR) < 130%, 130% IPR < 300%, and PIR 300%), and body weight status. The NHANES participants’ body height and weight were measured by stadiometer and digital scale in the MEC. Body mass index (BMI) is defined by weight in kilograms divided by height in meters squared. Body weight status was classified into underweight (BMI 18.5 kg/m2), normal weight (18.5 kg/m2 BMI < 25 kg/m2), overweight (25 kg/m2 BMI < 30 kg/m2), and obesity (BMI 30 kg/m2) based on the international classification of adult BMI values. Statistical analysis Individual fixed-effect regressions were performed based on data from day 1 and day 2 dietary interviews that provided two observations per participant. The 13 continuous outcome variables are daily consumptions of total energy (kcal), energy from SSB (kcal), energy from alcohol (kcal), energy from discretionary foods (kcal), fruit (g), vegetable (g), total fat (g), saturated fat (g), sugar (g), sodium (mg), cholesterol (mg), and fiber (g), and the HEI-2010 score for overall diet quality on a dietary recall day. The two dichotomous outcome variables are any fast-food or full-service restaurant consumption on a dietary recall day. The independent variables are three dichotomous variables for a dietary recall day of Friday, Saturday, and Sunday, with MondayeThursday as their common reference group. Compared to the conventional pooled cross-sectional regression, individual fixed-effect regression is preferred because it uses and only uses within-individual variations in dietary recall days (weekend vs. weekday) to identify the effect of weekend consumption on dietary behavior and intake, thus removing potential omitted variable bias due to differences in individual characteristics such as gender, race and/or ethnicity, eating habits, and taste preferences. Given the short time interval of 3e10 days between the two nonconsecutive dietary interviews, the possible confounders from individual characteristics that vary only in the longer term (e.g., age, education attainment, income level, and body weight status) are also removed in individual fixed-effect estimates. Linear individual fixed-effect regressions were estimated for continuous outcome variables, and logistic individual fixed-effect regressions were estimated for dichotomous outcome variables. Average treatment effects were calculated for logistic individual fixed-effect regressions to translate estimates into probabilities (prevalences). In sensitivity analysis, we used six dichotomous independent variables for a dietary recall day of Tuesday, Wednesday, Thursday, Friday, Saturday, and Sunday, with Monday as their common reference group, to replace the three dichotomous independent variables for a dietary recall day of Friday, Saturday, and Sunday and re-estimated all regressions. The estimated coefficients for weekdays (Tuesday, Wednesday, and Thursday) were mostly nonsignificant at P < .05, indicating the absence of a significant
59
difference in diet between Monday and the other weekdays. We therefore decided to exclusively report estimates from regressions with three dichotomous independent variables for a dietary recall day of Friday, Saturday, and Sunday. Individual fixed-effect regressions were estimated for the overall adult sample and subsamples stratified by sex, race and/or ethnicity (non-Hispanic other race or multirace excluded due to insufficient sample size), age group, education attainment, income level, and body weight status (underweight excluded due to insufficient sample size). To test for population heterogeneities in weekend-weekday differences in dietary intake, we included an interaction term between a stratification variable (e.g., sex, income level) and the weekend indicator in the regression (besides the main effect), so that difference across population subgroups would be indicated by a statistically significant coefficient for the interaction. Owing to the large number of statistical tests conducted (15 outcome variables 6 stratification variables ¼ 90 tests in total), we chose P < .001 as the criterion for statistical significance in between-subgroup comparisons. NHANES 2003e2012 multiyear complex survey design was accounted for in both descriptive statistics and regression analyses. All statistical procedures were performed in Stata 14.0, SE version (StataCorp, College Station, TX). NHANES was approved by the NCHS Research Ethics Review Board. This study used NHANES de-identified public data and was deemed exempt by the University of Illinois at UrbanaeChampaign Institutional Review Board. Results Table 1 reports individual characteristics of the study sample during 2003e2012. Nearly one-fifth (19.79%) of them had an IPR less than 130%. Approximately two-thirds (65.18%) were overweight or obese (BMI 25), and about a third (32.88%) were obese (BMI 30). Tables 2 and 3 report the estimated changes in diet on Friday, Saturday, and Sunday in comparison to weekdays in the whole
Table 1 Individual characteristics of adult study participants, NHANES 2003e2012 Individual characteristics Sex Male Female Race/ethnicity White, non-Hispanic African American, non-Hispanic Other race/multirace, non-Hispanic Hispanic Age group (y) 18e64 65 Education High school education and below College education and above Income to poverty ratio (IPR) IPR < 130% 130% IPR < 300% IPR 300% Body weight status Underweight (BMI < 18.5) Normal weight (18.5 BMI < 25) Overweight (25 BMI < 30) Obese (BMI 30)
Proportion (%) 48.52 (47.86e49.17) 51.48 (50.83e52.41) 70.80 11.08 5.63 12.49
(67.78e73.83) (9.43e12.72) (4.95e6.32) (10.47e14.50)
82.31 (81.34e83.28) 17.69 (16.72e18.66) 40.27 (38.38e42.17) 56.15 (54.28e58.02) 19.79 (18.35e21.23) 27.69 (26.42e28.97) 46.70 (44.65e48.75) 1.86 30.87 33.30 32.88
(1.61e2.12) (29.75e32.00) (32.30e34.30) (31.79e33.97)
Individual-level data (n ¼ 11,646) came from NHANES 2003e2012 waves. Descriptive statistics count for NHANES multiyear complex survey design. Proportions may not add to 100% due to missing values in some individual characteristics. 95% confidence intervals are in parentheses.
N
All 11646
Male 5812
Female 5834
White 5524
Black 2457
Hispanic 2930
18e64 y 8883
65 y 2763
130.71*** (92.53e168.89) 181.74*** (128.14e235.35) 95.11*** (58.20e132.03)
96.86** (25.83e167.89) 156.24** (62.14e250.34) 16.03 (41.39 to 73.45)
78.69 (7.82 to 165.19) 207.58*** (128.13e287.03) 113.11** (42.35e183.87)
118.75*** (81.02e156.48) 182.94*** (138.50e227.39) 94.91*** (59.28e130.55)
53.45* (3.86e103.04) 59.99 (10.19 to 130.18) 48.74* (5.82e91.65)
13.46** (4.47e22.45) 19.68*** (8.98e30.39) 3.73 (3.72 to 11.18)
11.45 (3.98 to 26.89) 13.82 (7.92 to 35.56) 7.99 (6.71 to 22.70)
21.56** (8.67e34.45) 15.54 (1.49 to 32.58) 24.42* (5.87e42.98)
13.87** (5.52e22.21) 18.19*** (8.77e27.61) 9.59* (2.36e16.83)
6.94 (0.60 to 14.48) 6.56 (4.87 to 17.99) 2.98 (9.85 to 3.89)
25.21*** (13.99e36.42) 42.61*** (30.60e54.62) 18.69*** (9.82e27.55)
17.79** (7.46e28.13) 51.98*** (37.36e66.59) 19.68** (18.98e34.55)
23.56* (1.14e45.98) 74.11*** (53.16e95.05) 26.13*** (16.27e36.00)
24.80*** (14.51e35.10) 50.46*** (40.15e60.77) 19.99*** (11.29e28.69)
3.15 (9.51 to 3.22) 0.34 (10.13 to 10.82) 7.41* (1.05e13.78)
38.53*** (19.89e57.16) 52.80** (22.30e83.30) 67.37*** (47.40e87.34)
5.45 (26.79 to 37.69) 12.46 (21.89 to 46.81) 26.96 (0.42 to 54.34)
3.11 (40.17 to 33.94) 41.89* (7.85e75.94) 12.72 (13.81 to 39.25)
25.44** (8.27e42.61) 43.39** (18.17e68.60) 49.99*** (30.14e56.08)
35.46* (3.13e67.78) 54.68* (12.99e96.37) 69.53*** (44.56e94.50)
4.65 (12.16 to 2.86) 9.44 (19.55 to 0.68) 9.91* (17.62 to 2.21)
0.85 (9.76 to 8.07) 12.69* (23.58 to 1.80) 9.06* (16.97 to 1.15)
9.30 (18.93 to 0.33) 31.91*** (45.28 to 18.55) 11.40* (20.14 to 2.65)
4.98 (11.22 to 1.27) 14.99*** (22.84 to 7.14) 9.05* (16.42 to 1.68)
2.83 (15.21 to 9.54) 3.17 (17.25 to 10.90) 15.29** (23.76 to 6.83)
8.12 (17.79 to 1.55) 19.67** (31.53 to 7.81) 8.49* (16.94 to 0.04)
11.68** (20.35 to 3.01) 9.06 (18.41 to 0.29) 6.62 (3.72 to 16.96)
11.71 (26.22 to 2.80) 7.87 (23.34 to 7.60) 3.29 (14.13 to 7.55)
9.13* (17.11 to 1.15) 16.56** (26.36 to 6.75) 3.92 (12.14 to 4.31)
12.83 (28.64 to 2.98) 21.13 (45.40 to 3.15) 9.22 (19.61 to 1.17)
6.45*** (4.58e8.32) 8.30*** (5.46e11.14) 4.84*** (2.86e6.82)
3.62* (0.41e6.83) 5.52* (1.11e9.94) 1.13 (3.92 to 1.66)
4.17* (0.38e7.95) 8.83*** (4.87e12.79) 5.50** (2.07e8.93)
5.94*** (4.27e7.60) 8.43*** (6.24e10.63) 4.71*** (2.87e6.54)
3.45* (0.82e6.08) 3.70 (0.49 to 7.89) 2.89* (0.60e5.18)
1.89*** (1.16e2.61) 2.87*** (1.83e3.92)
0.94 (0.11 to 1.99) 2.01** (0.53e3.50)
1.11 (0.34 to 2.57) 3.02*** (1.49e4.55)
1.75*** (1.12e2.39) 3.00*** (2.17e3.83)
0.94* (0.10e1.79) 1.03 (0.55 to 2.60)
R. An / Annals of Epidemiology 26 (2016) 57e65
Change in total energy consumption (kcal) 148.12*** 90.94*** Friday 118.05*** (86.92e149.19) (90.54e205.69) (60.66e121.22) Saturday 181.04*** 206.60*** 155.31*** (141.98e220.10) (146.46e266.73) (106.52e204.10) 98.23*** 75.96*** Sunday 88.37*** (58.32e118.42) (56.72e139.74) (40.42e111.49) Change in energy from sugar-sweetened beverage consumption (kcal) Friday 14.00*** 19.27** 9.28** (7.15e20.86) (6.88e31.67) (3.17e15.40) Saturday 18.34*** 22.44** 14.46** (10.07e26.62) (9.62e35.25) (5.52e23.40) Sunday 7.38* 10.99* 3.72 (1.61 to 13.16) (2.46 to 19.51) (2.63 to 10.07) Change in energy from alcohol consumption (kcal) Friday 22.04*** 27.91*** 17.05*** (13.32e30.77) (12.62e43.19) (11.15e22.95) Saturday 46.65*** 65.64*** 27.70*** (37.89e55.41) (51.57e79.71) (19.06e36.34) Sunday 17.82*** 24.87*** 10.45** (10.61e25.04) (14.00e35.74) (3.97e16.94) Change in energy from discretionary food consumption (kcal) Friday 29.81*** 38.31*** 21.83* (15.17e44.45) (17.66e58.97) (2.95e40.71) Saturday 48.77*** 39.07* 58.58*** (26.55e70.99) (2.24e75.90) (32.07e85.10) Sunday 55.07*** 54.56*** 55.06*** (38.49e71.65) (29.91e79.20) (35.70e74.42) Change in fruit consumption (g) Friday 4.69 6.02 3.49 (10.23 to 0.85) (13.32 to 1.29) (11.40 to 4.43) Saturday 13.90*** 11.66* 16.46** (20.92 to 6.89) (22.55 to 0.77) (25.51 to 7.40) Sunday 10.35** 13.38** 7.72* (16.67 to 4.04) (22.82 to 3.94) (14.35 to 1.09) Change in vegetable consumption (g) Friday 9.39* 7.53 11.21* (16.46 to 2.32) (17.66 to 2.60) (20.70 to 1.72) Saturday 16.76*** 22.20** 11.37* (25.71 to 7.81) (34.63 to 9.77) (21.98 to 0.76) Sunday 4.92 5.91 4.20 (11.66 to 1.81) (15.46 to 3.65) (13.92 to 5.52) Change in total fat consumption (g) Friday 5.77*** 6.80*** 4.82*** (4.33e7.20) (4.46e9.15) (2.98e6.67) Saturday 8.16*** 8.43*** 7.89*** (6.14e10.18) (5.54e11.32) (5.34e10.45) Sunday 4.39*** 4.85*** 3.88*** (2.78e6.00) (2.64e6.21) (1.83e5.92) Change in saturated fat consumption (g) Friday 1.70*** 1.83*** 1.59*** (1.15e2.26) (1.00e2.66) (0.91e2.26) Saturday 2.88*** 3.04*** 2.71*** (2.13e3.62) (1.99e4.15) (1.87e3.55)
60
Table 2 Estimated changes in daily diet among U.S. adults on Friday, Saturday, and Sunday (Monday-Thursday average as reference) by gender, race and/or ethnicity, and age group
Sunday
1.36*** (0.65e2.07)
1.76*** (1.06e2.46)
0.27 (1.21 to 0.68)
1.67** (0.53e2.81)
1.64*** (0.99e2.29)
1.25** (0.48e2.02)
2.47 (0.06 to 5.00) 5.71** (1.44e9.98) 3.40* (0.75e6.05)
3.39* (0.10e6.67) 7.17** (2.76e11.58) 3.25* (0.44e6.06)
3.19 (2.38 to 8.76) 2.51 (4.15 to 9.17) 3.63 (1.21 to 8.46)
0.79 (4.78 to 6.37) 1.15 (4.48 to 6.79) 3.67 (0.99 to 8.34)
1.91 (1.28 to 5.11) 4.65* (1.13e8.17) 3.27* (0.39e6.15)
3.15 (0.97 to 7.28) 2.91 (3.02 to 8.84) 1.90 (1.29 to 5.09)
151.42*** (75.51e227.33) 250.25*** (163.81e336.70) 145.88*** (79.60e212.16)
197.00*** (115.54e278.45) 192.81** (77.61e308.01) 121.06** (45.57e196.56)
118.02 (29.97 to 266.00) 104.77 (86.29 to 295.83) 72.55 (190.47 to 45.38)
60.09 (56.60 to 176.78) 311.42*** (188.55e434.29) 222.38*** (114.64e330.12)
168.89*** (101.51e236.25) 211.40*** (114.53e308.27) 138.83*** (69.69e207.98)
88.29 (15.98 to 192.56) 41.59 (73.70 to 156.87) 70.42 (28.22 to 169.07)
16.30** (6.41e26.19) 40.21*** (29.82e50.59) 44.53*** (35.50e53.55)
17.96** (5.22e30.70) 43.05*** (29.01e57.10) 57.09*** (46.88e67.30)
11.54 (8.19 to 31.27) 29.46* (5.62e53.29) 26.98** (10.28e43.69)
8.85 (12.15 to 29.84) 55.15*** (34.89e75.41) 32.99** (14.37e51.61)
17.64** (6.18e29.09) 48.42*** (35.88e60.97) 50.70*** (40.67e60.73)
15.21 (1.59 to 32.01) 3.51 (24.14 to 17.12) 47.86*** (32.86e62.86)
0.45* (0.84 to 0.06) 0.50 (1.04 to 0.04) 0.55* (1.01 to 0.09)
0.10 (0.55 to 0.35) 0.54 (1.12 to 0.03) 0.58** (1.00 to 0.15)
0.35 (1.00 to 0.29) 0.81* (1.53 to 0.09) 0.75** (1.30 to 0.21)
0.05 (0.62 to 0.71) 1.13* (2.11 to 0.15) 0.53 (1.31 to 0.26)
0.24 (0.63 to 0.15) 0.82** (1.31 to 0.33) 0.45* (0.87 to 0.03)
0.07 (0.83 to 0.68) 0.12 (0.96 to 0.72) 1.03*** (1.55 to 0.50)
1.20** (1.89 to 0.51) 2.64*** (3.61 to 1.66) 1.88*** (2.44 to 1.33)
1.28*** (1.88 to 0.68) 1.91*** (2.88 to 0.93) 1.93*** (2.49 to 1.37)
0.72 (1.52 to 0.07) 1.84** (3.00 to 0.67) 0.93 (1.87 to 0.02)
1.10** (1.91 to 0.28) 3.49*** (4.69 to 2.30) 2.09*** (3.05 to 1.14)
1.32*** (1.87 to 0.78) 2.42*** (3.27 to 1.57) 1.46*** (2.04 to 0.89)
0.86 (1.89 to 0.16) 1.51 (3.13 to 0.11) 3.09*** (3.82 to 2.35)
11.65*** (9.01e14.28%) 12.26*** (9.26e15.25) 2.80* (5.54 to 0.06)
9.27*** (6.33e12.21) 6.87*** (3.50e10.24) 2.67 (5.48 to 0.15)
14.31*** (10.67e17.96) 12.86*** (8.38e17.34) 9.78*** (13.83 to 5.73)
13.63*** (10.02e17.25) 12.66*** (9.03e16.29) 6.48** (2.80e10.17)
11.85*** (9.83e13.86) 10.60*** (8.39e12.82) 1.84 (3.97 to 0.29)
10.34*** (5.11e15.57) 6.36 (0.59 to 13.30) 0.75 (5.20 to 3.70)
16.02*** (13.20e18.83) 19.89*** (16.90e22.88) 9.97*** (7.17e12.77)
16.56*** (13.64e19.49) 16.27*** (12.89e19.65) 5.67*** (2.91e8.44)
15.79*** (10.97e20.62) 16.75*** (11.17e22.33) 8.77*** (3.91e13.62)
14.53*** (10.43e18.62) 20.04*** (16.25e23.83) 13.41*** (9.50e17.32)
15.58*** (13.34e17.82) 18.64*** (16.35e20.92) 8.55*** (6.24e10.86)
12.96*** (7.63e18.28) 10.95** (3.54e18.35) 6.15** (2.09e10.21)
R. An / Annals of Epidemiology 26 (2016) 57e65
1.58*** 1.79*** (1.02e2.14) (0.92e2.65) Change in sugar consumption (g) Friday 2.74* 3.00 (0.09e5.39) (1.65 to 7.64) Saturday 5.37** 4.90 (2.33e8.41) (0.33 to 10.13) Sunday 3.21** 2.78 (0.99e5.43) (0.37 to 5.93) Change in sodium consumption (mg) Friday 166.79*** 182.14** (105.17e228.40) (69.69e294.59) Saturday 205.59*** 158.82* (123.35e287.84) (22.52e295.11) Sunday 127.96*** 105.05* (65.98e189.94) (5.66e204.41) Change in cholesterol consumption (mg) Friday 17.09** 18.00* (6.98e27.19) (1.35e34.65) Saturday 43.17*** 45.99*** (32.74e53.60) (28.50e63.49) Sunday 50.05*** 55.21*** (41.24e58.87) (40.09e70.34) Change in fiber consumption (g) Friday 0.15 0.17 (0.48 to 0.18) (0.42 to 0.77) Saturday 0.67** 0.83* (1.11 to 0.23) (1.46 to 0.19) Sunday 0.54** 0.55* (0.90 to 0.19) (1.07 to 0.03) Change in Healthy Eating Index-2010 score (out of 100) Friday 1.24*** 1.28** (1.71 to 0.77) (1.96 to 0.61) Saturday 2.32*** 2.02*** (3.06 to 1.57) (2.84 to 1.19) Sunday 1.80*** 1.74*** (2.23 to 1.36) (2.35 to 1.13) Change in prevalence of fast-food restaurant consumption (%) Friday 11.64*** 11.56*** (9.76e13.53) (8.87e14.25) Saturday 10.21*** 8.30*** (8.10e12.32) (5.33e11.27) Sunday 1.64 0.54 (3.55 to 0.28) (3.22 to 2.15) Change in prevalence of full-service restaurant consumption (%) Friday 15.08*** 14.03*** (13.00e17.16) (10.96e17.10) Saturday 17.79*** 15.62*** (15.58e19.99) (12.38e18.85) Sunday 7.91*** 5.91*** (5.90e9.92) (3.04e8.78)
Individual-level data came from NHANES 2003e2012 waves. Individual fixed-effect models were used to estimate the changes in daily diet among U.S. adults on Friday, Saturday, and Sunday in comparison to that on weekdays, accounting for NHANES multiyear complex survey design. 95% confidence intervals are in parentheses. *.01 P < .05; **.001 P < .01; and P < .001.
61
N
Low education 5478
High education 5315
Low income 3444
Middle income 3417
High income 3932
Normal weight 3423
Overweight 3885
Obese 3960
105.40** (38.51e172.28) 255.18*** (184.98e325.37) 36.16 (18.48e90.81)
143.34*** (92.83e193.86) 185.00*** (119.49e250.50) 127.73*** (80.85e174.61)
142.07*** (90.92e193.22) 152.05*** (87.72e216.37) 44.97 (13.89e103.83)
111.94*** (59.84e164.04) 217.68*** (150.85e284.52) 80.48** (23.69e137.27)
107.13*** (49.77e164.48) 185.94*** (115.67e256.22) 133.01*** (83.82e182.21)
4.04 (8.94 to 17.02) 13.99 (2.40 to 30.37) 0.29 (e11.08 to 11.66)
12.56* (1.90e23.22) 19.52** (5.74e33.31) 7.06 (e0.88 to 14.99)
15.93* (3.70e28.15) 19.66** (5.62e33.70) 1.09 (e8.35 to 10.53)
19.23*** (9.51e28.96) 19.69* (4.82e34.56) 12.87** (3.33e22.41)
8.05 (3.94 to 20.04) 16.82* (2.13e31.50) 8.09 (e1.30 to 17.47)
17.75* (2.26e33.24) 56.67*** (38.78e74.56) 12.23** (3.70e20.75)
27.81*** (14.16e41.46) 47.30*** (32.97e61.63) 23.04*** (10.56e35.52)
35.25*** (21.21e49.29) 54.44*** (40.64e68.25) 17.57** (5.63e29.51)
24.37** (10.25e38.49) 47.54*** (31.37e63.70) 18.65** (8.12e29.18)
9.31 (e2.29 to 20.90) 41.28*** (27.14e55.42) 18.72** (7.82e29.63)
22.96 (e10.31 to 56.22) 71.67** (28.53e114.81) 34.35* (4.82e63.87)
41.26** (18.60e63.91) 50.20* (9.54e90.86) 79.20*** (52.79e105.60)
36.79* (6.06e67.52) 24.03 (e10.93 to 58.99) 31.73* (3.51e59.96)
35.89* (8.51e63.27) 78.05*** (38.63e117.48) 60.29*** (22.16e55.69)
17.29 (e9.48 to 44.05) 49.37** (13.86e84.88) 72.35*** (46.98e97.72)
4.55 (e5.34 to 14.44) e11.92 (e24.40 to 0.55) e8.51* (e15.26 to e1.75)
e8.33 (e17.60 to 0.93) e16.04* (e28.20 to e3.89) e11.32* (e22.27 to e0.37)
e7.37 (e20.85 to 6.10) e20.64*** (e31.37 to e9.90) e17.54*** (e26.95 to e8.13)
e3.13 (e13.85 to 7.58) e16.94* (e30.73 to e3.16) e7.96 (e16.37e0.44)
e3.18 (e9.84 to 3.49) e5.13 (e16.70 to 6.45) e7.88 (e19.06e3.31)
e6.80 (e18.14 to 4.54) e15.67* (e28.08 to e3.26) e0.68 (e10.15 to 8.78)
e10.94 (e22.94 to 1.05) e19.94* (e35.14 to e4.74) e9.49 (e21.76 to 2.77)
e0.39 (e11.34 to 10.56) e27.06*** (e41.15 to e12.96) 1.68 (e10.17 to 13.53)
e21.27*** (e32.76 to e9.78) e13.21 (e31.44e5.02) e12.83* (e23.50 to e2.16)
e8.98 (e20.79 to 2.83) e13.17 (e27.14 to 0.80) e6.24 (e17.98 to 5.50)
5.43*** (2.50e8.35) 10.55*** (6.79e14.31) 1.65 (e1.40e4.70)
7.05*** (4.63e9.47) 8.75*** (5.31e12.19) 6.86*** (4.31e9.41)
5.34*** (2.59e8.09) 6.55*** (3.26e9.85) 1.90 (e0.87e4.67)
4.50*** (2.16e6.83) 9.61*** (6.23e13.00) 4.01* (0.90-7.12)
7.38*** (4.49e10.28) 8.64*** (5.00e12.27) 7.07*** (4.83e9.31)
1.14* (0.11e2.16) 3.80*** (2.32e5.27)
2.34*** (1.49e3.19) 3.09*** (1.86e4.31)
1.31** (0.34e2.28) 2.21*** (1.01e3.42)
1.64** (0.64e2.65) 3.57*** (2.29e4.85)
2.19*** (1.14e3.24) 2.86*** (1.60e4.13)
R. An / Annals of Epidemiology 26 (2016) 57e65
Change in total energy consumption (kcal) 133.15*** 103.76** Friday 101.28*** (46.20e156.36) (96.33e169.96) (32.65e174.87) Saturday 183.37*** 180.10*** 87.08* (118.21e248.53) (127.64e232.57) (15.75e158.41) 86.99*** 57.45 Sunday 84.61*** (41.79e127.43) (46.29e127.70) (10.34e125.23) Change in energy from sugar-sweetened beverage consumption (kcal) Friday 20.02*** 9.79* 32.19** (9.56e30.49) (0.66e18.91) (14.08e50.30) Saturday 17.67** 17.79** 12.54 (6.86e28.49) (5.30e30.27) (2.43 to 27.51) Sunday 6.68 6.68 16.17* (3.85 to 17.20) (e1.26 to 14.62) (1.81e30.54) Change in energy from alcohol consumption (kcal) Friday 13.15 28.62*** 19.32** (e1.44 to 27.74) (18.74e38.50) (8.01e30.63) Saturday 41.43*** 50.85*** 38.32*** (29.40e53.45) (37.01e64.69) (23.33e53.32) Sunday 14.38** 20.69*** 8.33 (5.13e23.62) (11.46e29.92) (e1.87e18.54) Change in energy from discretionary food consumption (kcal) Friday 11.18 39.22*** 19.39 (e13.35 to 35.72) (19.58e58.85) (e12.94 to 51.72) Saturday 38.28** 58.02*** 6.50 (10.45e66.10) (27.28e88.77) (e30.41 to 43.41) Sunday 43.67** 60.00*** 22.75 (37.30e63.25) (37.99e82.01) (e1.59 to 47.08) Change in fruit consumption (g) Friday e6.66 e3.52 e5.06 (e15.46 to 2.14) (e11.36 to 4.32) (e13.91 to 3.79) Saturday e14.02** e15.50** e13.90** (e23.88 to e4.15) (e26.11 to e4.89) (e24.10 to e3.70) Sunday e7.48* e12.57** e11.65** (e14.53 to e0.42) (e21.57 to e3.57) (e20.15 to e3.16) Change in vegetable consumption (g) Friday e13.10* e7.35 e4.11 (e23.83 to e2.37) (e16.33 to 1.63) (e14.96 to 6.75) Saturday e9.81 e22.72** e7.28 (e21.84 to 2.22) (e35.29 to e10.14) (e19.72e5.17) Sunday e1.34 e7.70 2.91 (e9.18 to 6.49) (e17.49 to 2.09) (e9.78 to 15.60) Change in total fat consumption (g) Friday 5.21*** 6.29*** 4.30** (2.93e7.49) (4.32e8.26) (1.32e7.29) Saturday 8.80*** 7.62*** 3.94* (5.84e11.76) (4.83e10.41) (0.61e7.26) Sunday 3.41** 4.86*** 2.31 (1.07e5.76) (2.76e6.97) (e0.65e5.27) Change in saturated fat consumption (g) Friday 1.30** 2.05*** 1.26* (0.39e2.20) (1.34e2.77) (0.22e2.29) Saturday 2.97*** 2.82*** 1.32* (1.93e4.00) (1.78e3.87) (0.20e2.44)
62
Table 3 Estimated changes in daily diet among U.S. adults on Friday, Saturday, and Sunday (Monday-Thursday average as reference) by education, income and body weight status
Sunday
1.10* (0.21e1.99)
0.48 (e0.79 to 1.76)
2.42*** (1.51e3.32)
0.63 (e0.22 to 1.48)
1.75** (0.66e2.84)
2.40*** (1.53e3.26)
5.95 (e0.16 to 12.06) e0.89 (e7.36 to 5.57) 4.87 (e0.25 to 9.99)
3.04 (e2.46 to 8.54) 6.65* (0.33e12.96) 0.86 (e3.36 to 5.07)
1.86 (e1.35 to 5.08) 6.95** (1.97e11.94) 3.63* (0.39e6.87)
5.71* (1.05e10.38) 4.13 (e1.35 to 9.62) 0.97 (e2.79 to 4.72)
3.69 (e0.15 to 7.53) 8.22** (2.45e13.99) 3.47* (0.20e6.74)
0.19 (e4.22 to 4.60) 4.11 (e0.69 to 8.92) 4.95** (1.26e8.65)
101.71 (e38.88 to 242.29) 38.88 (e119.51 to 197.28) 69.38 (e52.34 to 191.10)
144.07* (26.11e262.04) 295.98*** (170.19e421.77) 85.14 (e40.44 to 210.72)
222.54*** (106.17e338.91) 224.89** (81.60e368.18) 179.87*** (96.01e263.74)
158.72** (50.86e266.58) 45.79 (e97.82 to 189.39) 85.70 (e12.72 to 184.12)
167.35** (58.16e276.55) 314.90*** (181.25e448.55) 108.10* (6.72e209.47)
173.16** (54.75e291.56) 251.12** (108.20e394.03) 169.97** (62.74e277.20)
11.16 (e5.25 to 27.58) 41.60*** (22.47e60.74) 22.98** (7.05e38.91)
16.88 (e1.33 to 35.10) 44.03*** (24.91e63.16) 44.91*** (28.83e60.99)
19.85** (5.71e33.99) 44.74*** (25.41e64.07) 64.12*** (49.58e78.66)
14.14 (e2.70 to 30.97) 38.32*** (19.42e57.23) 47.56*** (34.29e60.83)
18.13* (0.90e35.36) 46.49*** (28.24e64.73) 53.23*** (36.89e69.56)
18.30* (2.57e34.03) 47.18*** (30.87e63.49) 49.26*** (32.64e65.89)
e0.28 (e0.96 to 0.40) e0.75* (e1.48 to e0.02) e0.41 (e1.09 to 0.27)
0.14 (e0.60 to 0.87) 0.04 (e0.81 to 0.88) e0.87** (e1.35 to e0.39)
e0.23 (e0.80 to 0.34) e1.05** (e1.71 to e0.39) e0.39 (e1.00 to 0.23)
e0.37 (e0.95 to 0.21) e1.31*** (e1.96 to e0.67) e0.93* (e1.64 to e0.23)
e0.09 (e0.73 to 0.55) e0.79 (e1.67 to 0.09) e0.74* (e1.37 to e0.11)
0.04 (e0.54 to 0.63) 0.01 (e0.85 to 0.86) e0.16 (e0.80 to 0.48)
e0.76 (e1.74 to 0.23) e1.76** (e2.90 to e0.61) e1.36** (e2.12 to e0.59)
0.06 (e0.83 to 0.95) e1.84** (e2.92 to e0.76) e1.26** (e2.08 to e0.45)
e2.09*** (e2.80 to e1.37) e2.75*** (e3.84 to e1.67) e2.09*** (e2.82 to e1.37)
e1.52** (e2.37 to e0.66) e2.12** (e3.28 to e0.96) e1.42** (e2.28 to e0.55)
e1.22** (e2.06 to e0.38) e2.75*** (4.01 to 1.50) e2.22*** (e2.94 to e1.51)
e0.96* (e1.81 to e0.11) 2.22*** (3.32 to 1.12) e1.78*** (e2.51 to 1.06)
11.36*** (7.79e14.93) 8.66*** (4.81e12.51) 0.55 (4.05 to 2.94)
9.53*** (6.06e13.01) 11.67*** (7.77e15.58) 0.38 (3.99 to 3.24)
12.40*** (9.25e15.56) 9.57*** (5.90e13.23) 4.21* (7.47 to 0.95)
10.56*** (6.91e14.22) 6.83** (2.79e10.87) 3.13 (6.67 to 0.31)
11.64*** (8.31e14.97) 10.16*** (6.49e13.84) 1.19 (2.18 to 4.56)
12.44*** (9.37e15.50) 12.45*** (8.94e15.96) 3.02 (6.23 to 0.19)
11.65*** (6.85e16.44) 15.05*** (10.42e19.68) 9.58*** (5.33e13.82)
16.17*** (12.37e19.97) 18.92*** (14.77e23.06) 5.98** (2.26e9.70)
15.20*** (12.10e18.31) 17.10*** (13.63e20.58) 8.10*** (5.00e11.21)
16.16*** (12.34e19.99) 18.56*** (14.60e22.52) 5.47** (1.81e9.13)
15.61*** (12.05e19.16) 17.56*** (13.79e21.33) 8.47*** (5.11e11.84)
13.70*** (10.12e17.28) 17.55*** (13.63e21.47) 9.11*** (5.56e12.66)
R. An / Annals of Epidemiology 26 (2016) 57e65
0.85* 2.04*** (0.00e1.71) (1.30e2.78) Change in sugar consumption (g) Friday 2.63 2.77 (e1.04 to 6.30) (e0.67 to 6.22) Saturday 3.46 6.67** (e1.35 to 8.27) (1.98e11.36) Sunday 3.60* 2.54 (0.19e7.00) (e0.56 to 5.64) Change in sodium consumption (mg) Friday 155.68** 181.80*** (59.26e252.09) (97.16e266.45) Saturday 232.56*** 183.68** (119.24e345.87) (72.66e294.70) Sunday 159.74** 111.74** (55.54e263.95) (39.03e184.44) Change in cholesterol consumption (mg) Friday 17.48* 18.72** (3.81e31.14) (6.09e31.35) Saturday 43.08*** 43.46*** (26.21e59.94) (29.17e57.76) Sunday 45.74*** 54.24*** (32.13e59.34) (41.68e66.80) Change in fiber consumption (g) Friday 0.10 e0.33 (e0.40 to 0.59) (e0.79 to 0.12) Saturday e0.29 e0.93** (e0.98 to 0.40) (e1.57 to e0.30) Sunday e0.13 e0.79** (e0.57 to 0.30) (e1.31 to e0.27) Change in Healthy Eating Index-2010 score (out of 100) Friday e0.21 e1.98*** (e1.01 to 0.59) (e2.65 to e1.32) Saturday e2.17*** e2.55*** (e3.30 to e1.04) (e3.48 to e1.61) Sunday e0.99** e2.36*** (e1.56 to e0.42) (e3.00 to e1.71) Change in prevalence of fast-food restaurant consumption (%) Friday 12.79*** 10.34*** (9.95e15.64) (7.59e13.09) Saturday 10.93*** 8.74*** (7.79e14.06) (5.59e11.90) Sunday 2.72 2.43 (5.59 to 0.15) (5.27 to 0.41) Change in prevalence of full-service restaurant consumption (%) Friday 15.73*** 14.39*** (12.41e19.06) (11.54e17.24) Saturday 17.34*** 17.72*** (13.89e20.80) (14.60e20.85) Sunday 7.75*** 7.73*** (4.61e10.88) (4.96e10.51)
Individual-level data came from NHANES 2003e2012 waves. Individual fixed-effect models were used to estimate the changes in daily diet among U.S. adults on Friday, Saturday, and Sunday in comparison to that on weekdays, accounting for NHANES multiyear complex survey design. 95% confidence intervals are in parentheses. *.01 P < .05; **.001 P < .01; and ***P < .001.
63
64
R. An / Annals of Epidemiology 26 (2016) 57e65
study sample and by gender, race and/or ethnicity, age group, education, income, and body weight status using individual fixedeffect models. Diet on weekends was less healthful than that on weekdays, and among the three weekend days, diet quality on Saturday was the poorest. Compared with weekday average consumption, consumption on Saturday was associated with an increase in daily intakes of total energy by 181.04 kcal (95% confidence interval ¼ 141.98e220.10), energy from SSB by 18.34 kcal (10.07e26.62), energy from alcohol by 46.65 kcal (37.89e55.41), energy from discretionary foods by 48.77 kcal (26.55e70.99), total fat by 8.16 g (6.14e10.18), saturated fat by 2.88 g (2.13e3.62), sugar by 5.37 g (2.33e8.41), sodium by 205.59 mg (123.35e287.84), and cholesterol by 43.17 mg (32.74e53.60), and a decrease in intakes of fruit by 13.90 g (20.92 to 6.89), vegetable by 16.76 g (25.71 to 7.81), and fiber by 0.67 g (1.11 to 0.23). Compared with weekday average, the estimated reduction in HEI-2010 score was 2.32 (3.06 to 1.57) on Saturday. The prevalence of fast-food restaurant consumption was 10.21 percentage point (8.10e12.32) higher and that of full-service restaurant consumption 17.79 percentage point (15.58e19.99) higher on Saturday than on weekdays. In general, weekend consumption was found to be associated with higher intakes of total energy and certain foods/nutrients that warrant significant reduction by dietary guidelines and lower intakes of fruit/vegetable and fiber among all population subgroups under examination. However, substantial heterogeneities in the relationship between diet and day of the week (weekends vs. weekdays) were present across these subgroups. Compared with their female counterparts, weekend-weekday differences in intakes of total energy, energy from SSB and alcohol, total fat, saturated fat, and cholesterol were larger among male adults. Compared with older adults 65 years of age and above, weekendweekday differences in intakes of total energy, energy from SSB, alcohol and discretionary foods, total fat, saturated fat, sodium and cholesterol, and the prevalence of full-service restaurant consumption were larger among young and middle-aged adults 18e64 years of age. Compared with those with high school education and below, weekend-weekday differences in intakes of energy from alcohol and discretionary foods, cholesterol, and fiber were larger among those with college education and above. Compared with people with lower income, weekend-weekday differences in cholesterol intake and overall diet quality measured by the HEI-2010 were larger among those at the highest income level. Compared with their racial and/or ethnic counterparts, weekend-weekday difference in energy intake from discretionary foods was larger among non-Hispanic whites; whereas weekend-weekday difference in fruit intake was larger among Hispanics. All the above reported between-subgroup comparisons had a P value less than .001.
Discussion This study assessed weekend-weekday differences in diet among U.S. adults using data from NHANES 2003e2012 waves. It confirmed findings from previous literature on the increased discretionary calorie intake during weekend days [7e20]. Weekend consumption was found to be associated with higher intake of total energy and certain foods and/or nutrients that warrant substantial reduction in accordance with dietary guidelines and lower intake of fruit and/or vegetable and fiber. Diet quality on Saturday tended to be the poorest among the three weekend days. Considerable heterogeneities in the relationship between diet and day of the week were present across population subgroups by sex, age, race and/or ethnicity, education, income, and body weight status.
The elevated prevalence of fast-food and full-service restaurant consumption may contribute to the poorer diet quality on Saturdays. An (2015) reported both fast-food and full-service restaurant use to be associated with substantial increase in daily total energy consumption and intakes of total fat, saturated fat, cholesterol, and sodium [32]. Based on data from the USDA’s Nationwide Food Consumption Survey, U.S. adults were found to consume more eggs, bacon, meat, and alcoholic beverages but less cereal on weekends than on weekdays [15]. Coinciding with the less healthful eating style, reduced physical activity level and increased sedentary behavior tend to be more prevalent on weekends, which jointly lead to weight gain [4,12,18]. Gender and age differences in the relationship between diet and day of the week have been documented [15]. Thompson et al. (1986) [15] found weekend-weekday differences in diet to be more pronounced in women than in men and in younger adults than in older adults. The underlying determinants of these population heterogeneities remain to be understood but could partially correlate with differences in work pattern, social role, food preference, diet habit, and affordability. The strength of this study included use of nationally representative data, a comprehensive set of measures on dietary intake and dining behavior, stratified estimates on each of the three weekend days (Friday, Saturday, and Sunday) and various population subgroups, and a fixed-effect identification strategy. Nevertheless, a few limitations of this study should be noted. Dietary intakes in NHANES were self-reported and subject to measurement error and social desirability bias [34]. Individual fixed-effect estimator eliminated confounding bias from unobservable factors that remained constant within-participant between the two dietary interviews but could not control for more transient factors like daily variations in physical activity, appetite, or emotions. Compared with pooled cross-sectional estimators, the individual fixed-effect estimator is underpowered (less precision and larger standard error) because only a subsample consisting of individuals whose two dietary recall days occupied a weekday and a weekend contributed to the effect estimation, whereas those with both recalls on weekdays or weekends did not. In conclusion, American’s weekend consumption was associated with increased calorie intake and poorer diet quality. Policy interventions and public health campaigns are warranted to improve awareness, change social norms, and promote healthful eating on the weekend.
References [1] U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2020. http://www.healthypeople.gov/ 2020/topics-objectives/topic/nutrition-and-weight-status. Accessed October 1, 2015. [2] U.S. Department of Agriculture, U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2010. Washington, DC: U.S. Government Printing Office; 2010. [3] Krebs-Smith SM, Guenther PM, Subar AF, Kirkpatrick SI, Dodd KW. Americans do not meet federal dietary recommendations. J Nutr 2010;140(10):1832e8. [4] McCarthy S. Weekly patterns, diet quality and energy balance. Physiol Behav 2014;134(1):55e9. [5] Orfanos P, Naska A, Trichopoulos D, Slimani N, Ferrari P, van Bakel M, et al. Eating out of home and its correlates in 10 European countries. The European Prospective Investigation into Cancer and Nutrition (EPIC) study. Public Health Nutr 2007;10(12):1515e25. [6] Yang PH, Black JL, Barr SI, Vatanparast H. Examining differences in nutrient intake and dietary quality on weekdays versus weekend days in Canada. Appl Physiol Nutr Metab 2014;39(12):1413e7. [7] Rothausen BW, Matthiessen J, Hoppe C, Brockhoff PB, Andersen LF, Tetens I. Differences in Danish children’s diet quality on weekdays v. weekend days. Public Health Nutr 2012;15(9):1653e60. [8] Haines PS, Hama MY, Guilkey DK, Popkin BM. Weekend eating in the United States is linked with greater energy, fat, and alcohol intake. Obes Res 2003;11(8):945e9.
R. An / Annals of Epidemiology 26 (2016) 57e65 [9] Svensson A, Larsson C, Eiben G, Lanfer A, Pala V, Hebestreit A, et al. European children’s sugar intake on weekdays versus weekends: the IDEFICS study. Eur J Clin Nutr 2014;68(7):822e8. [10] Yang PHW, Black JL, Barr SI, Vatanparast H. A temporal analysis of Canadian dietary choices using the Canadian Community Health Survey Cycle 2.2: does nutrient intake and diet quality vary on weekends versus weekdays. FASEB J 2013;27:354.5. [11] Hanson KL, Olson CM. School meals participation and weekday dietary quality were associated after controlling for weekend eating among U.S. school children aged 6 to 17 years. J Nutr 2013;143(5):714e21. [12] Hart CN, Raynor HA, Osterholt KM, Jelalian E, Wing RR. Eating and activity habits of overweight children on weekdays and weekends. Int J Pediatr Obes 2011;6(5-6):467e72. [13] Rothausen BW, Matthiessen J, Andersen LF, Brockhoff PB, Tetens I. Dietary patterns on weekdays and weekend days in 4-14-year-old Danish children. Br J Nutr 2013;109(9):1704e13. [14] Lachapelle D, Gamache C, Brodeur JM, Sévigny J. Frequency of food consumption in children: weekdays versus weekends. J Can Dent Assoc 1989;55(1):61e5. [15] Thompson FE, Larkin FA, Brown MB. Weekend-weekday differences in reported dietary intake: The nationwide food consumption survey, 1977-78. Nutr Res 1986;6(6):647e62. [16] Rhodes DG, Cleveland LE, Murayi T, Moshfegh AJ. The effect of weekend eating on nutrient intakes and dietary patterns. FASEB J 2007;21:835.2. [17] Hartline-Grafton HL, Rose D, Johnson CC, Rice JC, Webber LS. The influence of weekday eating patterns on energy intake and BMI among female elementary school personnel. Obesity (Silver Spring) 2010;18(4):736e42. [18] Racette SB, Weiss EP, Schechtman KB, Steger-May K, Villareal DT, Obert KA, et al. Influence of weekend lifestyle patterns on body weight. Obesity (Silver Spring) 2008;16(8):1826e30. [19] Cullen KW, Lara KM, de Moor C. Children’s dietary fat intake and fat practices vary by meal and day. J Am Diet Assoc 2002;102(12):1773e8. [20] Kim MJ. Comparison of food and nutrient intake between weekday and weekend for elementary and middle school students by gender in Busan and some parts of Kyungsangnamdo. J Nutr Health 2013;46(4):332e45. [21] Faith MS, Kral TVE, Institute of Medicine Committee on Assessing Interactions Among Social, Behavioral, and Genetic Factors in Health. Social environmental and genetic influences on obesity and obesity-promoting behaviors: fostering research integration. In: Hernandez LM, Blazer DG,
[22]
[23] [24]
[25] [26]
[27]
[28]
[29] [30]
[31]
[32]
[33] [34]
65
editors. Genes, Behavior, and the Social Environment: Moving Beyond the Nature/Nurture Debate. Washington, DC: National Academies Press; 2006 http://www.ncbi.nlm.nih.gov/books/NBK19935/. Accessed October 1, 2015. Centers for Disease Control and Prevention, National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD: U.S: Department of Health and Human Services; 2014 http:// www.cdc.gov/nchs/nhanes.htm. Accessed October 1, 2015. Ogden CL, Kit BK, Carroll MD, Park S. Consumption of sugar drinks in the United States, 2005-2008. NCHS Data Brief 2011;71:1e8. U.S. National Cancer Institute. Sources of beverage intakes among the US population, 2005-06. http://riskfactor.cancer.gov/diet/foodsources/beverages/. Accessed October 1, 2015. National Health and Medical Research Council. Australian Dietary Guidelines. Canberra, Australia: NHMRC; 2013. Bleich SN, Wolfson JA, Vine S, Wang YC. Diet-beverage consumption and caloric intake among US adults, overall and by body weight. Am J Public Health 2014;104(3):e72e8. Guenther PM, Casavale KO, Reedy J, Kirkpatrick SI, Hiza HA, Kuczynski KJ, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet 2013;113(4):569e80. Bowman SA, Friday JE, Moshfegh AJ. MyPyramid Equivalents Database, 2.0 for USDA Survey Foods, 2003-2004. Beltsville, MD: U.S. Department of Agriculture; 2008. National Cancer Institute. HEI tools for researchers. http://appliedresearch. cancer.gov/hei/tools.html. Accessed October 1, 2015. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. Healthy Eating Index. http://www.cnpp.usda.gov/healthyeatingindex. Accessed October 1, 2015. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion. Archived projects. http://www.cnpp.usda.gov/OtherProjects.htm. Accessed October 1, 2015. An R. Fast-Food and full-service restaurant consumption and daily energy and nutrient intakes in U.S. adults. Eur J Clin Nutr 2015. http://dx.doi.org/10.1038/ ejcn.2015.104 [Epub ahead of print]. Powell LM, Nguyen BT, Han E. Energy intake from restaurants: demographics and socioeconomics, 2003-2008. Am J Prev Med 2012;43(5):498e504. Larson MR. Social desirability and self-reported weight and height. Int J Obes Relat Metab Disord 2000;24(5):663e5.