For individual use only. Duplication or distribution prohibited by law. Weight Control; Young Adults

Eat, Sleep, Work, Play: Associations of Weight Status and HealthRelated Behaviors Among Young Adult College Students Virginia Quick, PhD, RD*; Carol Byrd-Bredbenner, PhD, RD, FADA; Adrienne A. White, PhD, RD; Onikia Brown, PhD, RD*; Sarah Colby, PhD, RD*; Suzanne Shoff, PhD; Barbara Lohse, PhD, RD; Tanya Horacek, PhD, RD; Tanda Kidd, PhD, RD; Geoffrey Greene, PhD, RD

Abstract Purpose. To examine relationships of sleep, eating, and exercise behaviors; work time pressures; and sociodemographic characteristics by weight status (healthy weight [body mass index or BMI , 25] vs. overweight [BMI  25]) of young adults. Design. Cross-sectional. Setting. Nine U.S. universities. Subjects. Enrolled college students (N ¼ 1252; 18–24 years; 80% white; 59% female). Measures. Survey included the Pittsburgh Sleep Quality Index (PSQI), Three-Factor Eating Questionnaire (TFEQ), Satter Eating Competence Inventory (ecSI), National Cancer Institute Fruit/Vegetable Screener, International Physical Activity Questionnaire, Work Time Pressure items, and sociodemographic characteristics. Analysis. Chi-square and t-tests determined significant bivariate associations of sociodemographics, sleep behaviors, eating behaviors, physical activity behavior, and work time pressures with weight status (i.e., healthy vs. overweight/obese). Statistically significant bivariate associations with weight status were then entered into a multivariate logistic regression model that estimated associations with being overweight/obese. Results. Sex (female), race (nonwhite), older age, higher Global PSQI score, lower ecSI total score, and higher TFEQ Emotional Eating Scale score were significantly (p , .05) associated with overweight/obesity in bivariate analyses. Multivariate logistic regression analysis showed that sex (female; odds ratio [OR] ¼ 2.05, confidence interval [CI] ¼ 1.54–2.74), older age (OR ¼ 1.35, CI ¼ 1.21–1.50), higher Global PSQI score (OR ¼ 1.07, CI ¼ 1.01–1.13), and lower ecSI score (OR ¼ .96, CI ¼ .94–.98), were significantly (p , .05) associated with overweight/obesity. Conclusion. Findings suggest that obesity prevention interventions for college students should include an education component to emphasize the importance of overall sleep quality and improving eating competence. (Am J Health Promot 2014;29[2]:e64–e72.) Key Words: Young Adults, Health, Behaviors, Weight, Sleep, Eating, Prevention Research. Manuscript format: research; Research purpose: modeling/relationship testing; Study design: nonexperimental; Outcome measure: behavioral; Setting: school; Health focus: weight control; Strategy: skill building/behavior change; Target population age: adults; Target population circumstances: education/income level, race/ethnicity

PURPOSE Obesity is a major public health problem that is associated with many serious medical, social, and economic consequences.1,2 Overweight and obesity are of particular concern among young adults because this age group experiences the fastest weight gain, averaging 15 kg over a 15-year period.3,4 This weight increase may be in part due to young adulthood being a life stage linked to excess intake of food and alcohol, decreased physical activity, lack of time for exercise, and erratic sleep patterns.5–10 Interventions aimed at weight loss have demonstrated short-term success, but weight regain is common,11,12 thereby limiting the efficacy of weight reduction strategies as a public health remedy for the obesity epidemic. Taking a preventive approach by identifying modifiable risk factors for weight gain, however, could be beneficial in halting the progression of increasing overweight and obesity rates among

Virginia Quick, PhD, RD, is with the Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Division of Intramural Population Health Research, Bethesda, Maryland. Carol Byrd-Bredbenner, PhD, RD, FADA, is with the Department of Nutritional Sciences, Rutgers University, New Brunswick, New Jersey. Adrienne A. White, PhD, RD, is with Food Science and Human Nutrition, University of Maine, Orono, Maine; Onikia Brown, PhD, RD, is with the Department of Nutrition, Dietetics, and Hospitality Management, Auburn University, Auburn, Alabama. Sarah Colby, PhD, RD, is with the Department of Nutrition, University of Tennessee, Knoxville, Tennessee. Suzanne Shoff, PhD, is with the Department of Nutritional Sciences, University of Wisconsin-Madison, Madison, Wisconsin. Barbara Lohse, PhD, RD, is with Nutritional Sciences, The Pennsylvania State University, University Park, Pennsylvania. Tanya Horacek, PhD, RD, is with the Department of Public Health, Food Studies and Nutrition, Syracuse University, Syracuse, New York. Tanda Kidd, PhD, RD, is with the Department of Human Nutrition, Kansas State University, Manhattan, Kansas. Geoffrey Greene, PhD, RD, is with the Department of Nutrition and Food Sciences, University of Rhode Island, Kingston, Rhode Island. *When this research was conducted, Dr. Quick was with Rutgers University, Dr. Brown was with Iowa State University, and Dr. Colby was with East Carolina University. Send reprint requests to Virginia Quick, PhD, RD, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Division of Intramural Population Health Research, 6100 Executive Blvd, Room 7B13E, Bethesda, MD 20892; gingermquick@ gmail.com. This manuscript was submitted March 27, 2013; revisions were requested May 21 and July 15, 2013; the manuscript was accepted for publication July 29, 2013. Copyright Ó 2014 by American Journal of Health Promotion, Inc. 0890-1171/14/$5.00 þ 0 DOI: 10.4278/ajhp.130327-QUAN-130

e64

American Journal of Health Promotion

November/December 2014, Vol. 29, No. 2

For individual use only. Duplication or distribution prohibited by law. young adults. Previous research has examined modifiable risk factors for obesity, such as physical activity and eating behavior13–15; however, few have considered other potentially modifiable risk factors for obesity, such as sleep behavior and biopsychosocial factors (e.g., intrapersonal approaches to eating) in young adults. Duration of sleep has decreased in the last 2 decades in all age groups in the United States,16 with growing evidence that insufficient sleep as well as poor intrapersonal approaches to eating may be related to the increasing prevalence of overweight and obesity.17,18 Researchers have reported associations between sleep duration and weight status in adult populations.19,20 However, few researchers have examined associations between sleep quality, using a standardized measure such as the Pittsburgh Sleep Quality Index, a validated indicator of overall sleep quality, and weight status in a young adult U.S. college population.21 Given the growing body of literature around the importance of sleep for health and recent action from the Centers for Disease Control and Prevention recommending adults acquire at least 7 hours of sleep per night,22 examining sleep behavior and a number of other modifiable risk factors of obesity in young adults would be of importance for prevention researchers. Biopsychosocial factors affecting food regulation and food selection include emotional eating (i.e., practice of consuming food in response to negative emotional feelings), cognitive restraint eating (i.e., limiting food intake to manage weight), uncontrolled eating (i.e., inability to control food intake especially when hungry), and eating competence.18,23 The Satter Eating Competence Model (ecSatter) defines eating competence as an approach to food- and eating-related attitudes and behaviors that has positive biopsychosocial outcomes.18 ecSatter emphasizes eating enjoyment, internal regulation of food intake, allowing body weight to be determined by lifestyle and genetic potential, using skills to regularly provide meals to oneself, and achieving dietary variety by eating primarily for pleasure rather

American Journal of Health Promotion

than for solely meeting dietary recommendations.18,24 Associations between biopsychosocial factors and weight status of young adults remain understudied. Additionally, epidemiologic data on the relationships between sleep quality, biopsychosocial factors, and healthrelated behaviors, such as lack of physical activity and poor nutrition, are lacking, with most population-based studies only describing bivariate associations of these factors with weight status.9,25 There are few studies that quantify the extent of the association of weight status with sleep quality, biopsychosocial factors, and healthrelated behaviors, with adjustments for time scarcity due to work obligations, a potential confounding variable.26 Therefore, the aims of this study were to examine the association of weight status (healthy vs. overweight) with sleep, biopsychosocial eating behaviors, and health-related behaviors (i.e., eating, exercise), and to assess the impact of adjustment for time scarcity and sociodemographic characteristics (e.g., sex, age) on the patterns of this association among young adults. Findings from this study may indicate potential obesity-related risk factors among young adults that may be important to address in future obesity prevention interventions.

METHODS The university partners in the United States Department of Agriculture Multistate Healthy Campus Research Consortium, which focuses on the health of young adults, conducted this survey. This cross-sectional survey was approved by the institutional review boards of each of the nine participating universities. Participating universities were located in northeastern (n ¼ 5), southern (n ¼ 1), and Midwestern (n ¼ 3) states. All universities were public institutions, except for one which was private. Most participating universities had an enrollment of .15,000 (n ¼ 8) and were located in rural/suburban areas (n ¼ 7). Sample Participants aged 18 to 24 years, enrolled at nine universities, with a body mass index (BMI) of 18.5 or

higher were recruited to participate through verbal announcements at student gatherings, flyers, and postings to university listservs. To maximize sample size, data from two contemporaneous studies using identical questionnaires with similar sex, age, and BMIs (p . .1), were combined. Instruments The study survey included five sections to assess weight status, sociodemographic characteristics, sleep, eating, and physical activity behaviors, and work time pressures. Weight Status. Self-reported height and weight were collected. BMI was calculated and categorized as healthy weight (BMI: 18.5 to ,25) or overweight/ obese (BMI  25) based on the Centers for Disease Control and Prevention guidelines.27 Sociodemographics. Sociodemographic data included age, sex, and race/ ethnicity. Sleep Behavior. The Pittsburgh Sleep Quality Index (PSQI)28 is a validated instrument with seven scales that assess Subjective Sleep Quality (i.e., perception of one’s own sleep quality), Sleep Latency (i.e., how long it usually takes to fall asleep per night), Sleep Duration (i.e., average number of hours of actual sleep per night), Habitual Sleep Efficiency (i.e., actual hours of sleep vs. hours spent in bed), Sleep Disturbances (i.e., factors that cause individuals to wake up in the middle of the night or early morning), Sleep Medication Use (i.e., prescribed or ‘‘over the counter’’), and Daytime Dysfunction (i.e., difficulty staying awake during the day) due to sleep quality during the past month. Each of the scales is equally weighted with score ranges from 0 to 3. Scales are then summed to generate a global index score that reflects quantitative aspects of sleep, (e.g., sleep duration and latency) and subjective aspects (e.g., restfulness of sleep). Global scores range from 0 to 21, with higher scores reflecting poorer sleep quality. A global score of 5 or higher is indicative of a poor-quality sleeper. The PSQI adequately differentiates between good and bad sleepers.29 Additionally, previous research among college students has indicated

November/December 2014, Vol. 29, No. 2

e65

For individual use only. Duplication or distribution prohibited by law. the PSQI is a reliable measure (Cronbach a ¼ .73).9 Eating Behavior. Biopsychosocial eating behaviors were assessed by using two instruments: Three-Factor Eating Questionnaire (TEFQ)23,30 and Satter Eating Competence Inventory (ecSI).18,31 The National Cancer Institute Daily Fruit and Vegetable Screener served as an indicator of healthy eating behaviors.32 The TFEQ (18 items) is a reliable and validated instrument with 3 scales (i.e., Cognitive Restraint, Uncontrolled Eating, Emotional Eating) that measure cognitive and behavioral components of eating behaviors.23,30 The Cognitive Restraint Scale (6 items) assesses intentions to limit food intake for weight management purposes. The Uncontrolled Eating Scale (9 items) assesses uncontrolled eating behaviors, and the Emotional Eating Scale (3 items) assesses how emotions influence the urge to eat. All items are on a 4point Likert scale and scale scores are reported as percentages with a range of 0% (low) to 100% (high).23 The ecSI, a valid and reliable measure, contains 4 scales (i.e., Eating Attitudes [5 items], Food Acceptance [3 items], Internal Regulation [3 items], and Contextual Skills [5 items]) measuring eating competence. The Eating Attitudes Scale measures the degree to which an individual has a positive and flexible orientation towards eating. The Food Acceptance Scale measures an individual’s inclination to try new foods and learn one’s own unique food preferences. The Internal Regulation Scale measures an individual’s awareness of and responsiveness to cycles of hunger, appetite, and satiety. The Contextual Skills Scale measures an individual’s ability to organize meals and ensure they are eaten and engage with eating.18,31 Respondents select from 5 response options (never, rarely, sometimes, often, always), which are scored on a 4point scale from 0 (never/rarely) to 3 (always), then summed for a total score and 4 scale scores. The National Cancer Institute Daily Fruit and Vegetable Screener is a validated 19-item instrument.32 It was used to measure fruit and vegetable intake during the previous month (cups per day).

e66

American Journal of Health Promotion

Physical Activity Behavior. A physical activity index was created to estimate physical activity level by using items from the validated, reliable International Physical Activity Questionnaire (IPAQ).33 The number of days per week (0 to 7 days) individuals engaged in vigorous activities (e.g., heavy lifting, digging, aerobics), moderate activities (e.g., carrying light loads, bicycling at a regular pace), walking for at least 10 minutes at a time, and strength training (e.g., lifting weights, sit-ups) were used to create the physical activity index score as follows: (No. Days of Vigorous Activities per Week 3 3) þ (No. Days of Moderate Activities 3 2) þ (Days of Walking 10 Minutes at a Time) þ (Days of Strength Training) ¼ Physical Activity Index Score. This scoring method is a streamlined and enhanced version of the IPAQ Short Form Scoring Methodology. Streamlining permits calculations without using metabolic equivalents (i.e., METS per minute per week), which can sometimes be misreported. Enhancements accounted for relative intensity of activity; that is, vigorous activity was weighted higher than other types of activities and moderate activity was weighted higher than walking and strength training. The physical activity index score could range from 0 to 49. Scores were categorized into three levels of physical activity: sedentary (score 0 to ,20), moderate (score 20 to ,30), and high (score 30). Work Time Pressure. The Work Time Pressure Scale was used as a proxy for time scarcity. This measure was examined by an expert review panel (n ¼ 9) and found to have good content validity. This scale score was calculated by summing the hours of paid employment per week (0 ¼ less than 1 hour; 1 ¼ 1 to 10 hours; 2 ¼ more than10 hours) and student credit load (1 ¼ less than full time; 2 ¼ full time; 3 ¼ more than full time) for a possible score range of 1 to 5, higher scores reflecting greater time constraints. Data Analysis Chi-square tests for categorical variables and t-tests for continuous variables were performed to determine significant bivariate associations among sociodemographics, sleep behaviors, eating behaviors, physical ac-

tivity behavior, and time scarcity with weight status (i.e., healthy vs. overweight/obese). Variables found to be statistically significant in bivariate associations with weight status and weakly correlated (r , .20) with each other, using Spearman rank correlations, were entered into a multivariate logistic regression model that estimated associations with being overweight/ obese. Beta estimates, beta standard errors, odds ratios (ORs), and 95% confidence intervals (CI) for each independent variable in the model predicting overweight/obese status were computed. Probability level was set at p , .05. Analyses were performed with SPSS software version 21.0 (IBM corporation, Chicago, Illinois).

RESULTS Of the original 1294 participants, 42 were excluded from the total dataanalytic sample (n ¼ 1252) owing to missing sociodemographic information, 33 were deleted owing to missing survey measures, and 9 had a BMI lower than the cutoff to be considered ‘‘healthy’’ (i.e., ,18.5). Those excluded did not differ from those included for analyses on any factor, except BMI. By definition, the excluded group had a lower BMI (21.2 6 3.7 compared to 23.6 6 3.7 in the sample retained for analysis, p , .001). All measures were normally distributed except the National Cancer Institute Fruit and Vegetable Screener and the PSQI subscales. Thus, these variables were dichotomized into meeting recommendations or not (e.g., ,5 cups and 5 cups of fruits and vegetables per day). As shown in Table 1, participants were mostly white (80%), female (59%), and approximately 19 years of age. Approximately one-third were overweight/obese, with a greater proportion of males being overweight/ obese (32% vs. 22% females, p , .001). In addition, a greater proportion of nonwhites were overweight/obese (31% vs. 24.9% white, p ¼ .04). Nearly half (49%) reported less than 7 hours of sleep per night and had a mean Global PSQI score indicative of poor sleep quality (score 5 indicator of poor sleep quality) (Table 2). Addi-

November/December 2014, Vol. 29, No. 2

For individual use only. Duplication or distribution prohibited by law.

Table 1 Sociodemographic Characteristics by Weight Status (N ¼ 1252)

Characteristic

Nonoverweight (n ¼ 924) No. (%) or Mean 6 SD

Overweight or Obese (n ¼ 328) No. (%) or Mean 6 SD

350 (68.0) 574 (77.9)

165 (32.0) 163 (22.1)

15.44

,0.001

171 (68.7) 753 (75.1) 19.0 6 1.1 21.9 6 1.6

78 (31.3) 250 (24.9) 19.5 6 1.3 28.5 6 3.8

4.23

0.044

5.57 30.24

,0.001 ,0.001

Sociodemographic characteristics Sex Male Female Race/ethnicity Nonwhite White Age, y Body mass index

t-test or v2 (df ¼ 1)*

p

* t-tests were conducted for continuous variables and v2 tests were conducted for categorical variables. df indicates degrees of freedom.

tionally, 10% reported sleep disturbances, 6% reported using sleep medications, and 64% reported daytime dysfunctions at least four times or more in the past month. Almost half (46%) had low or very low physical activity index scores and only one-fifth of participants consumed five or more cups of fruits and vegetables per day (Table 3). Work time pressures were moderate. Bivariate associations among weight status (healthy weight vs. overweight/ obese) and sleep were significant (p , .05). That is, compared to healthy weight participants, short sleep duration and sleep disturbances were significantly more prevalent in the overweight/obese participants. Sleep latency and Global PSQI mean score also were significantly higher in the overweight/obese group than the healthy weight group. Significant (p , .05) bivariate associations among weight status (healthy weight vs. overweight/obese) and all ecSI (i.e., Eating Attitudes, Food Acceptance, Internal Regulation, Contextual Skills) and TFEQ (i.e., Cognitive Restraint, Uncontrolled Eating, Emotional Eating) subscales also occurred (see Table 3). That is, overweight/obese participants (n ¼ 328) were less likely to be eating competent, and more likely to have greater Cognitive Restraint, Uncontrolled Eating, and Emotional Eating than healthy weight participants (n ¼ 924). A multivariate logistic regression analysis with overweight/obese status

American Journal of Health Promotion

as the dependent variable is presented in Table 4. Variables found to be statistically significant in bivariate associations with weight status (i.e., sex, race, age, Global PSQI, Eating Competence total score, and TFEQ scales) were entered as independent variables. To reduce the chance of multicollinearity, only the TFEQ scale with the highest Spearman rank correlation with overweight status was included in the model (i.e., Emotional Eating). Additionally, only the Eating Competence total score and Global PSQI score were entered into the model, rather than significant subscales from each of these measures, to reduce multicollinearity effects. Based on findings from the multivariate logistic analysis, sex (female) (OR ¼ 2.05; CI: 1.54–2.74), age (OR ¼ 1.35; CI: 1.21–1.51), Global PSQI (OR ¼ 1.07; CI: 1.01–1.13), and Eating Competence total score (OR ¼ .96; CI: .94–.98) were each significantly (p , .05) associated with overweight/obese status. Significant sex interactions were only found with Emotional Eating; Eating Competence approached a significant sex interaction (p ¼ .06).

DISCUSSION Findings from this study describe relationships of weight status with sleep and health-related behaviors (i.e., eating and physical activity), and sociodemographic characteristics. Poor sleep quality and low eating compe-

tence were associated with overweight/ obese status even after multivariable adjustment for sociodemographic characteristics. Thus, among college students, modifiable risk factors, such as inadequate sleep quality and low eating competence, may impact net energy balance leading to weight gain and obesity. Few researchers have examined the relationship of weight status and sleep quality, especially among college students. However, a link between sleep duration and obesity in adolescents and adults may exist.17,34–36 For instance in a 13-year prospective cohort study of young adults, authors found significant associations between short sleep duration with obesity and BMI.17 Average change rate of weight gain tended to be negatively associated with average change rate of sleep duration. In the current study, overall sleep quality included sleep duration along with other sleep-related behaviors such as sleep disturbances and sleep latency, which in the bivariate analysis were significantly associated with weight status; thus, sleep quality, in particular sleep disturbances and sleep latency, may be just as important as the total hours slept at night in predicting overweight/obesity and should be considered in future sleep research. Because this study was crosssectional, causal impact of sleep quality on overweight/obesity cannot be determined. Inversely, being overweight/obese may influence sleep quality.34,37,38 Additionally, the rela-

November/December 2014, Vol. 29, No. 2

e67

For individual use only. Duplication or distribution prohibited by law.

Table 2 Sleep Behaviors and Quality* by Weight Status of College Students (N ¼ 1252)

Measure Sleep Behaviors Subjective Sleep Quality Very or fairly good Very or fairly bad Sleep Latency (0 to 3) Sleep Duration .7 h per night 7 h per night Habitual Sleep Efficiency 75% efficient ,75% efficient Sleep Disturbance ,4 times in past month 4 times in past month Sleep Medication Use 0 times in last month 1 time in last month Daytime Dysfunction ,4 times in past month 4 times in past month Bedtime‡ ,1 A.M. 1 A.M. Wake-up time ,8:30 A.M. 8:30 A.M. Global PSQI score

Nonoverweight (n ¼ 924) No. (%) or Mean 6 SD

Overweight or Obese (n ¼ 328) No. (%) or Mean 6 SD

782 (84.6) 142 (15.4) 1.0 6 0.8

264 (80.5) 64 (19.5) 1.2 6 0.8

498 (53.4) 426 (46.1)

t-test or v2 (df ¼ 1)†

p

3.02

0.084

2.66

0.008

135 (41.2) 193 (58.8)

15.71

,0.001

889 (96.2) 35 (3.8)

303 (92.4) 25 (7.6)

7.80

0.010

843 (91.2) 81 (8.8)

290 (88.4) 38 (11.6)

3.24

0.047

877 (94.9) 47 (5.1)

303 (92.4) 25 (7.6)

2.87

0.098

328 (35.5) 596 (64.5)

127 (38.7) 201 (61.3)

1.09

0.316

627 (67.9) 297 (32.1)

215 (65.5) 113 (34.5)

0.59

0.452

480 (51.9) 444 (48.1) 5.2 6 2.4

170 (51.8) 158 (48.2) 5.9 6 2.7

0.001

1.00

4.19

,0.001

* Measured by the Pittsburgh Sleep Quality Index (PSQI). † t-tests were conducted for continuous variables and v2 tests were conducted for categorical variables. df indicates degrees of freedom. ‡ Item not included in the PSQI.

tionship between sleep quality and overweight/obesity may be bidirectional, whereby sleep quality and being overweight/obese may have reciprocal effects on each other. However, based on previous longitudinal research, evidence is strong of an association between lack of sleep and weight gain in children,39 adolescents,35,40 and adults17,20,36; this association may be the case with the college students in this study. Regardless of the directionality of this relationship, health care providers of young adults should evaluate sleep patterns and recommend regular, sufficient sleep as a component of weight management regimens.41 Eating competence was significantly inversely associated with overweight/obesity. Other researchers have found a similar association in

e68

American Journal of Health Promotion

that competent eaters have lower BMIs and fewer cardiovascular risk factors than less competent eaters.42,43 Additionally, mean eating competence scores of participants in this study were similar to those previously reported,31,44 including college students.45 Thus, this study’s findings support the need for obesity prevention programs among college students to incorporate eating competence concepts that have been difficult for young adults to grasp/ accept,46 such as providing predictable opportunities for eating and trusting internal processes for choosing food and regulating food intake. Eating competence is associated with better sleep quality,25 dietary quality,47 and physical activity48; thus, eating competence may be reflective of an overall approach to good self-

care. A recent 10-week online nutrition and physical activity intervention aimed at developing eating competence demonstrated promising results at improving fruit and vegetable intake among young adults enrolled in college.49 Emotional eating was not independently associated with overweight/ obesity in the regression model, but a significant sex interaction was found. Females in this study who also were more likely to be overweight/obese may have had greater emotional eating than males. This finding was anticipated given that eating for emotional reasons is exacerbated by weight reduction dieting and disordered eating, both of which are more prevalent among females than males.50,51 Additionally, females in this study were more likely to be overweight/obese,

November/December 2014, Vol. 29, No. 2

For individual use only. Duplication or distribution prohibited by law.

Table 3 Eating and Physical Activity Behaviors, and Work Time Pressures by Weight Status (N ¼ 1252)

Measure (Possible Score Range) Eating Behaviors Three Factor Eating Questionnaire23 Cognitive Restraint (0 to 100) Uncontrolled Eating (0 to 100) Emotional Eating (0 to 100) Satter Eating Competence Inventory18 Eating Attitudes (0 to 15) Food Acceptance (0 to 9) Internal Regulation (0 to 9) Contextual Skills (0 to 15) Eating Competence Total score (0 to 48) NCI Fruit and Vegetable Screener32 ,5 cups per day 5 cups per day Physical Activity Behaviors33 Physical Activity Index† Sedentary Moderate to high Work Time Pressure‡ (1 to 5)

No. of Items (Cronbach a)

Nonoverweight (n ¼ 924) No. (%) or Mean 6 SD

Overweight or Obese (n ¼ 328) No. (%) or Mean 6 SD

t-test or v2 (df ¼ 1)*

p

6 (0.82) 9 (0.84) 3 (0.87)

42.4 6 21.8 42.1 6 19.3 36.7 6 41.6

46.9 6 19.0 44.8 6 20.1 43.8 6 29.3

3.52 2.15 2.84

,0.001 0.032 0.005

5 3 3 5 16

(0.84) (0.73) (0.71) (0.73) (0.81)

11.1 6 3.3 5.3 6 2.3 6.8 6 1.8 8.4 6 3.2 31.5 6 7.1

10.4 6 3.2 4.8 6 2.3 6.4 6 1.8 7.8 6 3.4 29.35 6 7.1

3.17 3.44 3.31 2.87 4.74

0.002 0.001 0.001 0.004 ,0.001

n/a

765 (82.8) 159 (17.2)

276 (84.1) 52 (15.9)

0.32

0.607

n/a

420 (45.5) 504 (54.5) 3.0 6 0.7

156 (47.6) 172 (52.4) 3.0 6 0.8

0.43

0.520

0.97

0.923

n/a 2

* t-tests were conducted for continuous variables and v tests were conducted for categorical variables. df indicates degrees of freedom; and n/a indicates not applicable. † Physical Activity Index score derived from the International Physical Activity Questionnaire ¼ (No. Days of Vigorous Activities per Week 3 3) þ (No. Days of Moderate Activities 3 2) þ (No. Days of Walking 10 Minutes at a Time) þ (No. Days of Strength Training). Physical Activity Index score was then categorized into three levels of physical activity (sedentary [,20], moderate [,30], and high [40]). ‡ Work Time Pressure is a proxy for Time Scarcity and was calculated by summing the score for the number of employment hours per week (,1 hour, 1 to 10 hours, or 10 hours, scored 0 to 2, respectively) and student credit load (less than full-credit load, full-credit load, overload, scored 1 to 3, respectively).

Table 4 Multivariate Logistic Regression Predicting Overweight/Obesity

Predictor* Constant Sex (female) Race (nonwhite) Age Eating Competence Total Score† Emotional Eating‡ Global PSQI§ Test* Goodness-of-fit test Hosmer and Lemeshow

b

SE b

Wald v2

df

p

eb (OR)

95% CI

6.46 0.72 0.27 0.30 0.05 0.01 0.06

1.14 0.15 0.16 0.06 0.01 0.00 0.03

32.05 24.11 2.70 27.93 20.58 7.18 5.51

1 1 1 1 1 1 1

,0.001 ,0.001 0.100 ,0.001 ,0.001 0.007 0.019

NA 2.05 1.31 1.35 0.96 1.01 1.07

NA 1.54, 2.74 0.95, 1.81 1.21, 1.51 0.94, 0.98 1.00, 1.01 1.01, 1.13

12.22

8

0.142

* Cox and Snell R2 ¼ 0.075. df indicates degrees of freedom; SE, standard error; OR, odds ratio; CI, confidence interval; and NA, not applicable. † Satter Eating Competence Inventory. ‡ Measured by Three Factor Eating Questionnaire. § Measured by the Pittsburgh Sleep Quality Index.

American Journal of Health Promotion

November/December 2014, Vol. 29, No. 2

e69

For individual use only. Duplication or distribution prohibited by law. and researchers have reported that in certain emotional situations obese people eat more than normal-weight individuals.52 Our study findings suggest that among female college students, obesity prevention programs should consider incorporating strategies to help them cope with emotional eating. However, a recent qualitative study of emotional eaters in college students found little interest of wanting to change,53 thus these interventions may need to be tailored for Stage of Change.54 Many college students struggle with the challenges of balancing work and school. Although some evidence indicates that time scarcity may be associated with obesity and adverse health behaviors that contribute to obesity (e.g., physical inactivity, unhealthy eating habits),55 participants’ weight status did not appear to be affected by time scarcity in this study. However, just 6% of study participants reported working more than 10 hours per week. To clarify the effect of time scarcity on weight status, future research should include a more heterogeneous sample with regard to work time pressures. Interestingly, low fruit and vegetable intake and low physical activity levels were not each independently associated with overweight/obesity. Perhaps these health-related behaviors, generally presumed to promote weight gain and obesity, were not significant owing to self-reporting errors of these behaviors. Indeed, self-reports of physical activity56 and diet57 are subject to many sources of error owing to social desirability bias (i.e., tendency of respondents to give socially favorable answers).57 However, self-reports of physical activity and diet are commonly used in large-scale epidemiologic studies because of the efficiency and utility of this approach and should be considered a suitable measure as used in our study. With regard to fruits and vegetables, it may be that this single ‘‘protective’’ indicator of dietary intake is insufficient and that a broader look at dietary factors, including those that can contribute to weight gain (e.g., foods high in sugar and/or fat, such as desserts, fried snack foods, and fast foods), may be needed to better

e70

American Journal of Health Promotion

understand the dietary components associated with weight status. Additionally, the physical activity index created for this study may have underestimated physical activity level. Future research is needed to clarify this study’s findings, using more objective measures of physical activity (e.g., accelerometer) and fruit and vegetable consumption (e.g., 24-hour recall methodology). The limitations and strength of this study are worth noting. As with all human research of this type, study participants self-selected themselves to participate. In addition, study participants were mostly white, so findings may not be generalizable to the entire U.S. college population. There also may have been biases in self-reports of height and weight; however, previous research has found high correlations between objective and subjective measures of height and weight among young adults.58 Socioeconomic status before and during college was not assessed, which may influence overall findings.59 This cross-sectional study also does not permit the determination of temporality in the associations among weight status and health-related behaviors of college students. Despite these limitations, the study design included valid, reliable measures, a large sample size from nine unrelated institutions of higher education, and robust statistical procedures. The findings from this study may help to inform future longitudinal studies and weight management interventions for young adults. In conclusion, poor sleep quality and low eating competence were factors associated with overweight/ obesity among young adults enrolled in college. These findings have important public health implications for preventing weight gain and reducing the prevalence of overweight/obesity in college students, a group at increased risk for weight gain.3,60,61 In addition to healthy diet62 and adequate physical activity,63–65 findings from this study suggest that college campus obesity prevention programs may benefit from nutrition education that emphasizes eating competence24 and by addressing good sleep hygiene.

SO WHAT? Implications for Health Promotion Practitioners and Researchers What is already known on this topic? Previous obesity research and prevention efforts among college students have focused mainly on improving eating and physical activity behaviors. Recent literature suggests an association between sleep duration and weight. However, few researchers have quantified the extent of the association of weight status with sleep quality and other health-related behaviors, such as eating and physical activity, among college students, a population at risk for weight gain. What does this article add? This article contributes to the obesity prevention literature by examining associations between weight status and obesity-related risk factors of college students that may warrant addressing in future obesity prevention efforts. What are the implications for health promotion practice or research? Findings from this study suggest that in addition to promoting healthy eating and physical activity behaviors, health care providers of young adults should evaluate sleep patterns and recommend regular, sufficient sleep as a component of weight management regimens. Additionally, healthy weight management promotion materials that emphasize the importance of both adequate amounts of sleep and sleep quality along with improving eating competence may benefit college students on campus.

Acknowledgments Research was supported in part by the intramural research program of the National Institutes of Health (NIH), Eunice Kennedy Shriver National Institute of Child Health and Human Development; USDA National Research Initiative 2005-35215-154121541; NIH Grant M01RR10732; Kansas, Maine, New Jersey, Iowa, Rhode Island, and Wisconsin Agricultural Experiment Stations; Syracuse University; and East Carolina University.

References

1. Yach D, Stuckler D, Brownell K. Epidemiologic and economic consequences of the global epidemics of obesity and diabetes. Nat Med. 2006;12:62– 66. 2. Bray G. Medical consequences of obesity. J Clin Endocrinol Metab. 2004;89(6):2583– 2589.

November/December 2014, Vol. 29, No. 2

For individual use only. Duplication or distribution prohibited by law. 3. Williamson D, Kahn H, Remington P, Anda R. The 10-year incidence of overweight and major weight gain in US adults. Arch Intern Med. 1990;150:665–672. 4. Truesdale K, Stevens J, Lewis C, et al. Changes in risk factors for cardiovascular disease by baseline weight in young adults who maintain or gain weight over 15 years: the CARDIA study. Int J Obes. 2006;30: 1397–1407. 5. Guo S, Huang C, Maynard L, et al. Body mass index during childhood, adolescence and young adulthood in relation to adult overweight and adiposity: the Fels Longitudinal Study. Int J Obes. 2000;24:1628–1635. 6. Stamatakis K, Brownson R. Sleep duration and obesity-related risk factors in the rural Midwest. Prev Med. 2008;46:439–444. 7. Baranowski T, Cullen K, Basen-Engquist K, et al. Transitions out of high school: time of increased cancer risk? Prev Med. 1997; 26:694–703. 8. Bray S, Born H. Transition to university and vigorous physical activity: implications for health and well-being. J Am Coll Health. 2004;52:181–188. 9. Lund H, Reider B, Whiting A, Richard J. Sleep patterns and predictors of disturbed sleep in a large population of college students. J Adolesc Health. 2010;46:124–132. 10. Strong K, Parks S, Anderson E, et al. Weight gain prevention: identifying theory-based targets for health behavior change in young adults. J Am Diet Assoc. 2008;108:1708–1715. 11. Jeffery R, Drewnowski A, Epstein L, et al. Long-term maintenance of weight loss: current studies. Health Psychol. 2000; 19(suppl):5–16. 12. Franz M, VanWormer J, Crain L, et al. Weight-loss outcomes: a systematic review and meta-analysis of weight-loss clinical trials with a minimum 1-year follow-up. J Am Diet Assoc. 2007;107:1755–1767. 13. Neumark-Sztainer D, Wall M, Story M, Standish A. Dieting and unhealthy weight control behaviors during adolescence: associations with 10-year changes in body mass index. J Adolesc Health. 2012;50:80– 86. 14. Haines J, Neumark-Sztainer D, Wall M, Story M. Personal, behavioral, and environmental risk and protective factors for adolescent overweight. Obesity. 2007;15: 2748–2760. 15. Haines J, Kleinman K, Rifas-Shiman S, et al. Examination of shared risk and protective factors for overweight and disordered eating among adolescents. Arch Pediatr Adolesc Med. 2010;164:336–343. 16. Center for Disease Control and Prevention. QuickStats. MMWR. 2005;54: 933. 17. Hasler G, Buysse D, Klaghofer R, et al. The association between short sleep duration and obesity in young adults: a 13-year prospective study. Sleep. 2004;27:661–666. 18. Satter E. Eating competence: definition and evidence for the Satter Eating Competence model. J Nutr Educ Behav. 2007;39(S5):S142–S153.

American Journal of Health Promotion

19. Stranges S, Cappuccio F, Kandala N-B, et al. Cross-sectional versus prospective associations of sleep duration with changes in relative weight and body fat distribution: the Whitehall II study. Am J Epidemiol. 2008;167:321–329. 20. Vioque J, Torres A, Quiles J. Time spent watching television, sleep duration, and obesity in adults living in Valencia, Spain. Int J Obes Relat Metab Disord. 2000;24:1683– 1688. 21. Mathews D. Assessing Sleep Quality in Young Adult College Students, Aged 18–24 in Relation to Quality of Life and Anthropometrics [master’s thesis]. Orono, Me: University of Maine; 2010. 22. Sleep and sleep disorders. Washington, DC: Center for Disease Control and Prevention. Available at: http://www.cdc. gov/features/sleep/. Accessed March 1, 2013. 23. Karlsson J, Persson L-O, Sjostrom L, Sullivan M. Psychometric properties and factor structure of the Three-Factor Eating Questionnaire (TFEQ) in obese men and women: results from the Swedish Obese Subjects (SOS) study. Int J Obes. 2000;24: 1715–1725. 24. Satter E. Satter eating competence: nutrition education with the Satter eating competence model. J Nutr Educ Behav. 2007;39(suppl 5):189–194. 25. Shoff S, Nuss E, Horacek T, et al. Sleep quality is associated with eating behavior in 18–24 year old college students. J Nutr Educ Behav. 2009;41:S8–S9. 26. Brunner E, Chandola T, Marmot M. Prospective effect of job strain on general and central obesity in the Whitehall II study. Am J Epidemiol. 2007;165:828–837. 27. About BMI for adults. Washington, DC: Center for Disease Control and Prevention. Available at: http://www.cdc. gov/healthyweight/assessing/bmi/ adult_bmi/index.html. Accessed November 10, 2011. 28. Buysee D, Reynolds C, Monk T, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatr Res. 1989;28:193– 213. 29. Grandner G, Kripke D, Yoon I, Youngstedt S. Criterion validity of the Pittsburgh Sleep Quality Index: investigation in a nonclinical sample. Sleep Biol Rhythms. 2006;4: 129–136. 30. de Lauzon B, Romon M, Deschamps V, et al. The Three-Factor Eating Questionnaire-R18 is able to distinguish among different eating patterns in a general population. J Nutr Educ Behav. 2004;134:2372–2380. 31. Lohse B, Satter E, Horacek T, et al. Measuring eating competence: psychometeric properties and validity of the ecSatter Inventory. J Nutr Educ Behav. 2007;39(S5):S145–S166. 32. Thompson F, Subar A, Smith A, et al. Fruit and vegetable assessment: performance of 2 new short instruments and a food frequency questionnaire. J Am Diet Assoc. 2002;102:1764–1772.

33. Booth M. Assessment of physical activity: an international perspective. Res Q Exerc Sport. 2000;71:s114–s120. 34. Patel S, Hu F. Short sleep duration and weight gain: a systematic review. Obesity. 2008;16:643–653. 35. McKnight-Eily L, Eaton D, Lowry R, et al. Relationships between hours of sleep and health-risk behaviors in US adolescent students. Prev Med. 2011;53:271–273. 36. Wheaton A, Perry G, Chapman D, et al. Relationship between body mass index and perceived insufficient sleep among US adults: an analysis of 2008 BRFSS data. BMC Public Health. 2011;11:295. 37. Redline S, Clark K, Graham G. Risk factors for sleep disordered breathing in children: associations with obesity, race, and respiratory problems. Am J Respir Crit Care Med. 1999;159:1527–1532. 38. Vorona R, Winn M, Babineau T, et al. Overweight and obese patients in a primary care population report less sleep than patients with a normal body mass index. Arch Intern Med. 2005;165:25–30. 39. Chen X, Beydoun M, Wang Y. Is sleep duration associated with childhood obesity: a systematic review and metaanalysis. Obesity. 2008;16:265–274. 40. Gupta N, Mueller W, Chan W, Meininger J. Is obesity associated with poor sleep quality in adolescents? Am J Human Biol. 2002;14:762–768. 41. Shlisky J, Hartman T, Kris-Etherton P, et al. Partial sleep deprivation and energy balance in adults: an emerging issue for consideration by dietetics practioners. J Acad Nutr and Diet. 2012;112:1785–1797. 42. Psota T, Lohse B, West S. Association between eating competence and cardiovascular disease biomarkers. J Nutr Educ Behav. 2007;39:S171–S178. 43. Lohse B, Psota T, Estruch R, et al. Eating competence of elderly Spanish adults is associated with a healthy diet and a favorable cardiovascular disease risk profile. J Nutr. 2010;140:1322–1327. 44. Stotts J, Lohse B. Reliability of the ecSatter Inventory as a tool to measure eating competence. J Nutr Educ Behav. 2007;39: S167–S170. 45. Brown LB, Larsen KJ, Nyland NK, Eggett DL. Eating competence of college student in an introductory nutrition course. J Nutr Educ Behav. 2013;45:269–273. 46. Dour CA, Horacek TM, Schembre SM, et al. Process evaluation of Project WebHealth: a non-dieting web-based intervention for obesity prevention in college students. J Nutr Educ Behav. 2013; 45:288–295. 47. Lohse B, Bailey R, Krall J, et al. Diet quality is related to eating competence in crosssectional sample of low-income females surveyed in pennsylvania. Appetite. 2012;58: 645–650. 48. Lohse B, Arnold K, Wamboldt P. Evaluation of About Being Active, an online lesson about physical activity shows that perception of being physically active is higher in eating competent low-income women. BMC Women’s Health. 2013;13:12.

November/December 2014, Vol. 29, No. 2

e71

For individual use only. Duplication or distribution prohibited by law. 49. Greene G, White AA, Hoerr S, et al. Impact of an online healthful eating and physical actiity program for college students. Am J Health Promot. 2012;27:47– 58. 50. Hoek H. Incidence, prevalence and mortality of anorexia and other eating disorders. Curr Opin Pyschiatr. 2007;19: 389–394. 51. Neumark-Sztainer D, Wall M, Larson N, et al. Dieting and disordered eating behaviors from adolescence to young adulthood: findings from a 10-year longitudinal study. J Am Diet Assoc. 2011; 111:1004–1011. 52. Ganley R. Emotion and eating in obesity: a review of the literature. Int J Eat Disord. 2006;8:343–361. 53. Bennett J, Greene G, Schwartz-Barcott D. Perceptions of emotional eating behavior: a qualitative study of college students. Appetite. 2013;60:187–192. 54. Prochaska J, Redding C, Evers K. Health Behavior and Health Education: Theory, Research and Practice: The Transtheoretical Model and Stages of Change. 4th ed. San Francisco, Calif: Jossey-Bass Inc; 2008.

e72

American Journal of Health Promotion

55. Lallukka T, Lahelma E, Rahkonen O, et al. Associations of job strain and working overtime with adverse health behaviors and obesity: evidence from the Whitehall II Study, Helsinki Health Study, and the Japenese Civil Servants Study. Soc Sci Med. 2008;66:1681–1698. 56. Adams S, Matthews C, Ebbeling C, et al. The effect of social desirability and social approval on self-reports of physical activity. Am J Epidemiol. 2005;161:389–398. 57. Herbert J, Ma Y, Clemow L, et al. Gender differences in social desirability and social approval bias in dietary self-report. Am J Epidemiol. 1997;146:1046–1055. 58. Larson N, Neumark-Sztainer D, Story M, et al. Identifying correlates of young adults’ weight behavior: survey development. Am J Health Behav. 2011;35:712–725. 59. Krall J, Lohse B. Interviews with lowincome Pennsylvanians verify a need to enhance eating competence. J Am Diet Assoc. 2009;109:468–473. 60. Ball K, Brown W, Crawford D. Who does not gain weight: prevalence and predictors of weight maintenance in young women.

61.

62.

63.

64.

65.

Int J Obes Relat Metab Disord. 2002;26:1570– 1580. Hoffman D, Policastro P, Quick V, Lee S. Changes in body weight and fat mass of men and women in the first year of college: a study of the ‘‘Freshman 15’’. J Am Coll Health. 2006;55:41–50. Matvienko O, Lewis D, Schafer E. A college nutrition science course as an intervention to prevent weight gain in female college freshman. J Nutr Educ. 2001;33:95–101. Leslie E, Sparling P, Owen N. University campus setting and the promotion of physical activity in young adults: lessons from research in Australia and the USA. Health Educ. 2001;101:116–125. Hivert M-F, Langlois M-F, Berard P, et al. Prevention of weight gain in young adults through a seminar-based intervention program. Int J Obes. 2007;31:1262–1269. Ferrara C. The college experience: physical activity, nutrition and implications for intervention and future research. J Exerc Physiol. 2009;12:23–35.

November/December 2014, Vol. 29, No. 2

www.HealthPromotionConference.com Editor in Chief Michael P. O’Donnell, PhD, MBA, MPH Associate Editors in Chief Jennifer DiNoia, PhD Jennie Jacobs Kronenfeld, PhD Kwame Owusu-Edusei Jr., PhD Kerry J. Redican, MPH, PhD, CHES

The Wisdom of Practice and the Rigor of Research

“The American Journal of Health Promotion provides a forum for that rare commodity — practical and intellectual exchange between researchers and practitioners.” Kenneth E. Warner, PhD Dean and Avedis Donabedian Distinguished University Professor of Public Health School of Public Health, University of Michigan

“The contents of the American Journal of Health Promotion are timely, relevant, and most important, written and reviewed by the most respected researchers in our field.” David R. Anderson, PhD, LP Senior Vice President & Chief Health Officer, StayWell Health Management

Be the first to know.

onlineFirst

Available exclusively to Online Subscribers The American Journal of Health Promotion is now publishing all articles online, ahead of print. Articles are available as a PDF document for download as soon as they have completed the review process. This means you can access the very latest papers in the field of health promotion – in some cases up to a year before they appear in print.

Subscribe Online at www.HealthPromotionJournal.com Call 800-783-9913 (US only) or 731-645-4496

Subscribe Today.

ANNUAL SUBSCRIPTION RATES:

(Effective 1-1-14 through 12-31-14)

Individual

Print + Online

Institution

Tier 1 Print + Online Online Only

Tier 2 Print + Online Online Only

Tier 3 Print + Online Online Only

Print Only*

U.S.

$139

$359

$459

$559

$184

Canada and Mexico

$148

$368

$468

$568

$193

Other Countries

$157

$377

$477

$577

$202

*Print-only subscriptions are based on location. For multi-site institutions, each site must have its own subscription. Tier 1 — Most Employers and Corporations except Health Organizations, Libraries and Schools Tier 2 — Health Organizations including Hospitals, Clinics, Health Promotion Providers, Insurance Companies and Voluntary Health Agencies Tier 3 — Libraries, Colleges and Universities

2014_ad_sub.indd 1

10/4/13 12:03 PM