PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Am J Health Behav. Author manuscript; available in PMC 2016 July 1.
Published in final edited form as:
PMCID: PMC4448118
NIHMSID: NIHMS666136

College Women’s Weight-related Behavior Profiles Differ by Sexual Identity

Nicole A. VanKim, PhD, MPH, Darin J. Erickson, PhD, Marla E. Eisenberg, ScD, MPH, Katherine Lust, PhD, MPH, RD, B. R. Simon Rosser, PhD, and Melissa N. Laska, PhD, RD

Abstract

Objectives

To identify and describe homogenous profiles of female college students based on weight-related behaviors and examine differences across 5 sexual orientation groups.

Methods

Data from the 2009–2013 College Student Health Survey (Minnesota-based survey of 2- and 4-year college students) were used to fit latent class models.

Results

Four profiles were identified across all sexual orientation groups: “healthier eating habits,” “moderate eating habits,” “unhealthy weight control,” and “healthier eating habits, more physically active.” Differences in patterns and prevalence of profiles across sexual orientation suggest need for interventions addressing insufficient physical activity and unhealthy weight control behaviors.

Conclusions

Future interventions should consider the diversity of behavioral patterns across sexual orientation to more effectively address weight-related behavioral disparities.

Keywords: sexual orientation, college students, weight-related behaviors, disparities

Existing research suggests that lesbian, gay, and bisexual (LGB) adult women are more likely to be obese than heterosexual women.18 Findings for disparities in diet and physical activity across sexual orientation among women mostly have been mixed, and measurement of these behaviors has been inconsistent.3,5,6,915

These disparities exist even during emerging adulthood (typically defined as 18–25 years), when independence is generally established and new responsibilities, life skills, and identities are negotiated and formed.16 In studying weight-related health, particularly across sexual orientation, it is important to consider age.7,9 Research has suggested that weight-related health generally declines during emerging adulthood, with noted weight gain, deterioration of diet quality and physical activity, and increasing sedentary behaviors.1725 Some studies have indicated that during emerging adulthood, health disparities widen.2628 Given that adolescence and emerging adulthood is also a period for sexual development and exploration,2931 this period in the life-course may be a critical time where individual trajectories can have an adverse impact on weight-related health disparities. Nearly half of emerging adults attend college, representing a large proportion of this age group and an accessible population in which to study emerging adult disparities and to design and deliver interventions to address these disparities.32

In our previous work on weight-related disparities among emerging adults, findings suggested that among women, differences exist across sexual orientation for breakfast, fast food, and restaurant food consumption, physical activity, unhealthy weight control behaviors (such as vomiting, taking diet pills or laxatives), and binge eating.33 Other studies using longitudinal cohort data also found disparities across sexual orientation in unhealthy weight control behaviors and physical activity from adolescence into emerging adulthood.15,34 However, none of these studies has explored the patterning of weight-related behaviors.

Patterning of weight-related behaviors is complex, and traditional methods of analysis, such as regression, may not be sufficient in modeling of relationships among multiple behaviors. This is particularly relevant for weight-related factors, where eating habits, sedentary behaviors, and physical activity, might not consistently track together. It is likely that in these instances, a more holistic analytic approach might better capture the heterogeneity in a population. It is this heterogeneity that might improve understanding of the differences observed in weight-related health disparities across sexual orientation that can then be used to target and tailor interventions. For example, in a study of female college students by Laska et al,35 the authors were able to identify 4 subgroups of women based on a wide-range of health behaviors, including physical activity and nutrition. “Poor lifestyle, low risk” had the highest membership (40% of students), highlighting a large proportion of the female student population for which to address health-related programming targeting wellness behaviors (eg, physical activity and nutrition) with a concurrent focus on other risky behaviors (eg, alcohol and tobacco use).35 Further, the findings from the Laska et al study,35 as well as other studies, highlight that important differences in weight-related behaviors across sex exist, likely due to factors such as varying social norms.4,5,34,35 Therefore, we are including only women in this paper.

The purpose of this study was to identify and describe homogenous profiles of female college as students based on patterning of healthful weight-related behaviors (eg, eating habits, physical activity, weight control behaviors), and to examine differences across 5 sexual orientation groups. We hypothesized that there are distinct classes that share common patterns of weight-related behaviors, and for whom interventions can be developed and tailored. In addition, we hypothesized that differences in proportions of classes and patterning of weight-related behaviors exist across sexual orientation groups, with greater proportions of LGB women in unhealthy classes and exhibiting unhealthier patterns than heterosexual women.

METHODS

Study Population and Data Source

Data were from the 2009–2013 College Student Health Survey (CSHS), an ongoing statewide surveillance system of 2- and 4-year colleges and universities across Minnesota. For most schools participating in the CSHS, students were selected randomly through a registrar’s enrollment list furnished by participating educational institutions. For smaller schools, all students were invited to participate to have sufficient sample sizes for reports generated for each school; at larger schools, only a proportion of students were invited. Eligible participants were sent multiple invitations, including postcards and emails, to complete an anonymous online survey. Participants who completed the survey were entered into a raffle to win prizes such as iPods and iPads. The overall response rate was 33.2%. Additional details on the CSHS are available online (http://www.bhs.umn.edu/surveys/index.htm).

Between 2009 and 2013, 46 institutions participated in CSHS (26 2-year and 20 4-year). Thirty colleges participated in the CSHS in more than one year between 2009 and 2013. To ensure that participants were not included in the dataset more than once and to maximize sample size, a college’s second year of data was included when the possibility of overlap in participants was expected to be negligible (ie, less than 2%), as others and us have done previously.33,36,37 Six schools with a sampling percentage of less than 50% (ie, less than 50% of the student body were invited to participate in the survey) had a negligible percentage of overlap (estimated sample overlap range: 0.45%–1.57%). Thus, an additional year of data was included for these schools (Nstudents = 6912). This yielded a final merged 2009–2013 CSHS dataset consisting of 29,118 students.

Measures

Sexual orientation was assessed on the CSHS as both identity and behavior. Given the importance of both identity and behavior, as well as evidence that discordant behavior among heterosexual-identified adults is salient in addressing health disparities,38,39 we created the following categories for sexual orientation: “heterosexual” (identified heterosexual and did not report engaging in any same-sex sexual behavior in the past year), “discordant heterosexual” (identified and reported engaging in any same-sex sexual behavior in the past year), “gay/lesbian” (identified as gay or lesbian, regardless of sexual behavior), “bisexual” (identified as bisexual, regardless of sexual behavior), and “unsure” (identified as unsure about their sexual orientation, regardless of sexual behavior). This categorization is consistent with previous research using the Youth Risk Behavior Survey (YRBS)37 and CSHS data.33

A variety of weight-related behaviors were included in these analyses: dietary intake and eating habits, physical activity, sedentary behavior, and unhealthy weight control behaviors. All variables were dichotomized based on existing public health recommendations, which have practical significance in that they serve as a meaningful threshold for health. Furthermore, dichotomization facilitates interpretation of results and was the most appropriate approach given the non-normality of the majority of the data.

Three aspects of dietary intake were assessed: fruit and vegetable, soda, and diet soda consumption. These items used standard questions adapted from the YRBS,40 “During the past 7 days, how many times did you eat/drink the following?” Six items assessed specific foods/drinks for fruit and vegetable consumption. Frequency response options ranged from, “I did not eat or drink this,” to “4 or more times per day.” Participants met recommendations if they reported consuming fruits and vegetables ≥5 times/day. For soda and diet soda, participants met recommendations for each item if they reported consuming <1/day.41,42

To assess eating habits, participants reported the number of days that they ate breakfast.43 Breakfast consumption was dichotomized as ≥5 days/week or <5 days/week. The frequency of eating (1) fast food meals and (2) at other restaurants (not including fast food establishments) also was assessed. Response options ranged from “never” to “several times per day.” Frequent consumption of fast food or restaurant food is associated with increased portion sizes and excess weight.44,45 Therefore, both fast food and restaurant food consumption were dichotomized as≥several times/week vs <several times/week.36

Three types of physical activity were assessed: strenuous, moderate, and strengthening. The question asked: “In the past 7 days, how many hours did you spend doing the following activities?” Examples were provided for each type of activity. Response options ranged from “None,” to “6½ + hours.” Given conceptual similarities between moderate and strenuous physical activity, they were combined into a single “moderate-to-vigorous physical activity” indicator. Meeting recommendations was ≥5 hours/week of moderate and vigorous physical activity combined or ≥4.5 hours/week of either moderate or vigorous physical activity (guided by recommendations for weight maintenance, which include ≥1 hour on most days of the week).46 Consistent with previous research using CSHS data, strengthening physical activity was categorized as ≥2.5 hours/week or ≤2 hours/week.47

Time spent watching television and using a computer (for things besides school or work) on an average day were used to assess sedentary behaviors. Response options ranged from “None” to “5+ hours.” Categories of ≥14 hours/week vs <14 hours/week were created for screen time in line with recommendations for young people of <2 hours/day.48

To assess unhealthy weight control behaviors, participants indicated the frequency of 4 behaviors in the past 12 months: using laxatives to control weight, taking diet pills, binge eating, and inducing vomiting to control weight.35,42,43 These are similar to items that have been used extensively in other research, most notably the YRBS.34,49,50 Due, in part, to low prevalence of each, using laxatives, taking diet pills, and inducing vomiting were combined into a single unhealthy weight control behaviors variable (any vs none) whereas binge eating was examined separately.33

Analysis

Latent class analysis (LCA) is a technique designed to identify a small number of homogenous subgroups within a larger heterogeneous group,51,52 based on responses to select indicators. Details on this technique have been described in existing work.51,53

After assessing initial LCA models, fruit and vegetable consumption and sedentary behavior were dropped as indicators due to no separation between classes; that is, across classes, the probabilities for these 2 indicators were similar and did not help in characterizing different classes. Thus, 9 indicators were included in final LCA models to identify healthy weight-related behavioral patterning: soda, diet soda, fast food, restaurant food, and breakfast consumption, moderate-to-vigorous and strengthening physical activity, unhealthy weight control behaviors, and binge eating.

For these analyses, we included only women as participants (64.0% of original sample). Furthermore, we excluded participants with missing data for sexual orientation (N = 44), participants who reported being currently pregnant, due to different recommendations for weight and related behaviors while pregnant (N = 255). This yielded final analytic sample of 18,297 female college students. All data management and analyses were performed using SAS (SAS version 9.1, Cary, NC: SAS Institute Inc).

RESULTS

Overall, most female students were heterosexual (92.3%), 0.8% were discordant heterosexual, 1.2% were gay/lesbian, 3.7% were bisexual, and 2.0% were unsure of their sexuality. Two-thirds (62.3%) attended a 4-year school, most were white (82.5%), and the median age was 22 years.

The prevalence of the healthy weight-related behavioral indicators used in the final LCA models, by sexual orientation, is presented in Table 1. Overall, large majorities of women, across sexual orientation, met recommendations for soda, diet soda, fast food, and restaurant food consumption. However, across all sexual orientation groups, less than half of the women met recommendations for breakfast consumption, only about one-fourth to one-third met recommendations for moderate-to-vigorous physical activity, and few met recommendations for strengthening physical activity. Most female students did not engage in unhealthy weight control or binge eating (81.0–89.8% for unhealthy weight control, 70.4%–83.4% for binge eating).

Table 1
Prevalence of Meeting Weight-Related Behavioral Recommendations among Female College Students by Sexual Orientation, CSHS 2009–2013

Table 2 shows the fit statistics for the LCA models for each of the 5 sexual orientation groups. After considering fit statistics and interpretability of the latent class models, a 4-class solution was selected for all sexual orientation groups.

Table 2
Fit Statistics for Unconditional Independent LCA Models across Sexual Orientation Groups

Figure 1 shows the item-response probabilities and healthy weight-related behavioral patterning of classes. In final models, across all sexual orientation groups, 4 distinct classes were identified. Class 1 (“Healthier eating habits”) was characterized by high probabilities of meeting recommendations for regular soda, diet soda, fast food, and restaurant consumption (heterosexual: 0.91–0.98; discordant heterosexual: 0.90–0.97; gay/lesbian: 0.86–0.96; bisexual: 0.91–1.00; unsure: 0.93–1.00), a moderate probability of eating breakfast ≥5 days/week (range across sexual orientation groups: 0.41–0.65), low probability of meeting physical activity recommendations (moderate-to-vigorous physical activity: 0.01–0.25; strengthening physical activity: 0.00–0.06), and low probability of engaging in unhealthy weight control or binge eating (no unhealthy weight control: 0.94–0.98; no binge eating: 0.81–0.97).

Figure 1
Item-response Probabilitiesa across Sexual Orientation Groups

Class 2 (“Moderate eating habits”) had similar patterning to Class 1 including low physical activity, no unhealthy weight control, and no binge eating. However, this class was characterized by lower probabilities of meeting recommendations for regular soda (0.51–0.70), diet soda (0.64–0.86), fast food (0.39–0.69), restaurant food (0.57–0.89), and eating breakfast (0.08–0.23).

Class 3 (“Unhealthy weight control”) also had similar patterning to Class 1 including high probabilities for meeting dietary intake, fast food and restaurant food recommendations, low-to-moderate probabilities of breakfast consumption and physical activity. Class 3 was characterized by a lower probability of reporting no unhealthy weight control (0.06–0.36) and no binge eating (0.03–0.20).

Finally, Class 4 (“Healthier eating habits, more physically active”) had high probabilities for meeting recommendations for regular soda, diet soda, fast food, and restaurant food consumption, similar to Class 1. Furthermore, although probabilities were slightly lower for no unhealthy weight control and binge eating compared to Classes 1 and 2, probabilities were still high (no unhealthy weight control: 0.60–1.00; no binge eating: 0.65–0.96). Class 4 is distinguishable by having the highest probabilities on physical activity compared to other classes (moderate-to-vigorous physical activity: 0.68–0.99; strengthening physical activity: 0.30–0.95). It should be noted that among gay/lesbian and unsure women, probabilities for meeting physical activity recommendations were lower than other sexual orientation groups (ie, heterosexual, discordant heterosexual, and bisexual). For example, the probability of exhibiting high moderate-to-vigorous physical activity was 0.68 for gay/lesbian and 0.70 for unsure women, compared to 0.99, 0.92, 0.84 for heterosexual, discordant heterosexual, and bisexual women, respectively. Similar differences were noted for strengthening PA.

In addition to examining patterning, the prevalence of each class was also assessed (Table 3). Across all sexual orientation groups except gay/lesbian women, Class 1 (“Healthier eating habits”) had the highest prevalence. For gay/lesbian women, Class 2 (“Moderate eating habits) had the highest prevalence. Class 3 (“Unhealthy weight control”) had the lowest prevalence for heterosexual, gay/lesbian, and unsure women, and the prevalence for heterosexual women was nearly half of the prevalence for all other sexual orientation groups (except gay/lesbian women). For discordant heterosexual and bisexual women the class with the lowest prevalence was Class 4 (“Healthier eating habits, more physically active”). Prevalence estimates for Class 4 ranged from 8.7% for bisexual women to 24.0% for gay/lesbian women.

Table 3
Probability of Latent Class Membership

DISCUSSION

Overall, our results indicated that 4 distinct classes of healthy weight-related behaviors exist: Class 1 (“Healthier eating habits”), Class 2 (“Moderate eating habits”), Class 3 (“Unhealthy weight control”), and Class 4 (“Healthier eating habits, more physically active”). Despite conceptual similarities between the classes, differences in the prevalence of classes across sexual orientation highlight an area of concern. For example, compared to heterosexual women, a relatively small proportion of bisexual women were characterized by one of the 2 healthier eating classes (Class 1 and 4). Additionally, nearly twice the proportion of discordant heterosexual, bisexual, and unsure females exhibited the “unhealthy weight control” profile compared to heterosexual women, highlighting an additional area of concern. Consistent with previous work, these findings suggest that discordant heterosexual, LGB, and unsure women experience poorer health with regard to weight-related behaviors, particularly in the areas of unhealthy weight control and physical activity.5,9,14,15

These findings help advance our understanding of the disparities in weight-related behaviors across sexual orientation and may help with intervention design, targeting, and outreach. More specifically, the patterns allow us to examine how healthy weight-related behaviors correlate in differing ways within diverse heterogeneous populations, and subsequently, inform our ability to intervene on multiple related behaviors simultaneously. For example, whereas our previous work using traditional regression methods found that discordant heterosexual, LGB, and unsure female students were more likely to engage in unhealthy weight control behaviors and binge eating,33 a finding consistent in the present study, the LCA results indicated that unhealthy weight control was coupled with lower probability of meeting recommendations for physical activity. Existing longitudinal research has found that dieting behaviors (including unhealthy weight control behaviors assessed here) are associated with weight gain over time, possibly because dieting yields less in the way of healthy, sustainable food choices and physical activity.54 Our LCA results partially support this hypothesis, with unhealthy weight control co-occurring with physical inactivity and skipping breakfast, though not with other aspects of nutrition. Overall, this finding suggests that to support healthier patterns of weight-related behaviors, it may be important to develop interventions that concurrently address unhealthy weight control and physical activity behaviors. These interventions are particularly needed for discordant heterosexual, bisexual, and unsure women as they shoulder a greater burden of this behavioral pattern with nearly twice the proportion of people in this class compared to heterosexual women (heterosexual: 7.8%, discordant heterosexual: 15.0%, bisexual: 18.9%, unsure: 12.1%).

Our findings also suggest the need to refine interventions to address unique, identified barriers or experiences specifically related to unhealthy weight control behaviors and physical activity for sexual minority female students. For example, online intervention designs may provide a cost-efficient means of adapting and tailoring intervention messaging to meet the needs of many different sub-groups within the college population – including heterosexual women, as well as sexual minority women. Previous work evaluating an online nutrition, physical activity, and sleep course in an applied, university setting found several improvements in self-reported student behaviors as well as high student satisfaction with the course content.55 Additional advantages of online interventions include more personalization toward individual needs based on specific behavioral profiles (eg, individuals identified as exhibiting an “unhealthy weight control” profile, for example through a brief series of 5–10 screening questions, could receive an intervention designed to address key features of this profile: unhealthy weight control behaviors, physical inactivity and skipping breakfast) as well as privacy for students who may be less comfortable disclosing their sexual orientation to college wellness program staff.

Our results also indicate that the majority of women, regardless of sexual orientation, were in classes with low physical activity, a finding consistent with previous work demonstrating low physical activity among college students.47,56,57 Although classes with low physical activity also included varying levels of diet and unhealthy weight control, high physical activity co-occurred with healthier eating habits and low unhealthy weight control (ie, in Class 4). This class exhibited a critical pattern in that it was the healthiest pattern identified. An area of concern, however, is that both gay/lesbian and unsure females showed lower probabilities of physical activity (especially strengthening physical activity) than did other sexual orientation groups, suggesting that there is still a need for targeted physical activity interventions for these classes of gay/lesbian and unsure females. Further, it was interesting that the presence of a more uniquely “unhealthy” pattern (such as one that exhibited both poor diet and low physical activity) among any of the sexual orientation groups was not present. This finding suggests that there may not be a notable subgroup of college females, regardless of sexual orientation, that needs interventions for all aspects of weight-related behaviors, an encouraging possibility that allows us to refine and target our intervention efforts.

In addition to targeted strategies that could be employed to address these issues, broader environmental and structural changes on college campuses also may need to be considered. For example, with regard to physical activity, facilities such as college recreation centers should be safe, supportive spaces for discordant heterosexual, gay/lesbian, bisexual, and unsure female students. Creating these safer spaces could include factors such as availability and promotion of single stall changing rooms and/or restrooms for students, better signage to promote the need for safe spaces and respectful behavior in the gym, hours at the gym designated as LGB hours to promote safety, and inclusivity for LGB students, or LGB-specific fitness groups for students to engage in a variety of activities such as hiking, biking, yoga, or organized sports. Coupled with improvements in the physical and social environment to promote physical activity, resources that address social norms related to unhealthy weight control behaviors among LGB students may also be valuable. Moreover, to address overall disparities in weight-related outcomes by sexual orientation that have been documented by numerous studies to date,29,14,15,33 campus wellness programs may need to employ more targeted outreach strategies directed at LGB students and use materials that are inclusive of diverse students, including LGB students.

To our knowledge, this is the first study to use LCA to characterize healthy weight-related behavioral patterns and examine disparities across sexual orientation among emerging adults. A strength of this study includes the large sample of gay/lesbian, bisexual, and unsure participants (which allowed separate sexual orientation groups rather than treating non-heterosexual as a homogenous group) as well as the inclusion of the discordant heterosexual group, thereby allowing for a more robust and fine grain examination of sexual orientation disparities. More research is needed to understand potential mechanisms for LGB college students, such as the role of social norms or stigma, as well as issues related to poor body image, which lend to poorer weight-related behavioral patterns. Better understanding of mechanisms also may help to improve development of intervention strategies to address known weight-related disparities more effectively. Finally, because this was a population-based sample of college students in Minnesota only, the results may not be generalizable to college students in other geographic areas or to emerging adults not attending a post-secondary institution.

Conclusions

Overall, these findings highlight unique patterning of healthy weight-related behaviors across sexual orientation among college women. Future research should examine how these behavioral patterns are related to relevant health outcomes, such as overweight and obesity, high cholesterol, high blood pressure, or diabetes. Regarding interventions, future work should tailor intervention components and target recruitment to specific patterns of weight-related behaviors and to specific sexual orientation groups. Areas of greatest importance include addressing unhealthy weight control and low physical activity among discordant heterosexual, gay/lesbian, bisexual, and unsure college women.

Acknowledgments

At the time of this study, N. A. VanKim was a graduate school trainee with the Division of Epidemiology and Community at the University of Minnesota School of Public Health. The study was supported primarily by the Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD) of the National Institutes of Health under Award Number R21HD073120 (PI: M. Laska). Further support on this project was provided by NIDDK Award Number T32 DK083250. The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institutes of Health.

Footnotes

Human Subjects Statement

These analyses were considered secondary analysis of anonymous data and therefore deemed exempt from IRB review. The University of Minnesota IRB approved all CSHS data collection.

Conflict of Interest Statement

The authors declare no conflict of interest.

Contributor Information

Nicole A. VanKim, Postdoctoral Research Fellow, Institute for Behavioral and Community Health, San Diego State University, San Diego, CA.

Darin J. Erickson, Assistant Professor, University of Minnesota School of Public Health, Division of Epidemiology and Community Health, Minneapolis, MN.

Marla E. Eisenberg, Associate Professor, University of Minnesota, Division of General Pediatrics and Adolescent Health, Minneapolis, MN.

Katherine Lust, Director of Research, University of Minnesota, Boynton Health Service, Minneapolis, MN.

B. R. Simon Rosser, Professor, University of Minnesota School of Public Health, Division of Epidemiology and Community Health, Minneapolis, MN.

Melissa N. Laska, Associate Professor, University of Minnesota School of Public Health, Division of Epidemiology and Community Health, Minneapolis, MN.

References

1. Bowen DJ, Balsam KF, Ender SR. A review of obesity issues in sexual minority women. Obesity. 2008;16(2):221–228. [PubMed]
2. Boehmer U, Bowen DJ, Bauer GR. Overweight and obesity in sexual-minority women: evidence from population-based data. Am J Public Health. 2007;97(6):1134–1140. [PubMed]
3. Boehmer U, Bowen DJ. Examining factors linked to overweight and obesity in women of different sexual orientations. Prev Med. 2009;48(4):357–361. [PubMed]
4. Conron KJ, Mimiaga MJ, Landers SJ. A population-based study of sexual orientation identity and gender differences in adult health. Am J Public Health. 2010;100(10):1953–60. [PubMed]
5. Dilley JA, Simmons KW, Boysun MJ, et al. Demonstrating the importance of feasibility of including sexual orientation in public health surveys: health disparities in the Pacific Northwest. Am J Public Health. 2009;99(10):1–8. [PubMed]
6. Fredriksen-Goldsen KI, Kim HJ, Barkan SE, et al. Health disparities among lesbian, gay, and bisexual older adults: results from a population-based study. Am J Public Health. 2013;103(10):1802–1809. [PMC free article] [PubMed]
7. Deputy NP, Boehmer U. Weight status and sexual orientation: differences by age and within racial and ethnic subgroups. Am J Public Health. 2014;104(1):103–109. [PubMed]
8. Jun H, Corliss HL, Nichols LP, et al. Adult body mass index trajectories and sexual orientation: the Nurses’ Health Study II. Am J Prev Med. 2012;42(4):348–354. [PMC free article] [PubMed]
9. Boehmer U, Miao X, Linkletter C, Clark MA. Adult health behaviors over the life course by sexual orientation. Am J Public Health. 2012;102:292–300. [PubMed]
10. Case P, Austin SB, Hunter DJ, et al. Sexual orientation, health risk factors, and physical functioning in the Nurses’ Health Study II. J Womens Health. 2004;13(9):1033–1047. [PubMed]
11. Aaron DJ, Markovic N, Danielson ME, et al. Behavioral risk factors for disease and preventive health practices among lesbians. Am J Public Health. 2001;91(6):972–975. [PubMed]
12. Roberts SA, Dibble SL, Nussey B, Casey K. Cardiovascular disease risk in lesbian women. Womens Health Issues. 2003;13(4):167–174. [PubMed]
13. Valanis BG, Bowen DJ, Bassford T, et al. Sexual orientation and health: comparisons in the women’s health initiative sample. Arch Fam Med. 2000;9(9):843–853. [PubMed]
14. Rosario M, Corliss HL, Everett BG, et al. Sexual orientation disparities in cancer-related risk behaviors of tobacco, alcohol, sexual behaviors, and diet and phyiscal activity: pooled Youth Risk Behavior Surveys. Am J Public Health. 2014;104(2):245–254. [PubMed]
15. Calzo JP, Roberts AL, Corliss HL, et al. Physical activity disparities in heterosexual and sexual minority youth ages 12–22 years old: roles of childhood gender nonconformity and athletic self-esteem. Ann Behav Med. 2014;47(1):17–27. [PMC free article] [PubMed]
16. Arnett JJ. Emerging adulthood: a theory of development from the late teens through the twenties. Am Psychol. 2000;55(5):469–480. [PubMed]
17. Park MJ, Mulye TP, Adams SH, et al. The health status of young adults in the United States. J Adolesc Health. 2006;39(3):305–317. [PubMed]
18. Gordon-Larsen P, The NS, Adair LS. Longitudinal trends in obesity in the United States from adolescence to the third decade of life. Obesity. 2010;18(9):1801–1804. [PMC free article] [PubMed]
19. Larson NI, Neumark-Sztainer D, Hannan PJ, Story M. Trends in adolescent fruit and vegetable consumption, 1999–2004 - Project EAT. Am J Prev Med. 2007;32(2):147–150. [PubMed]
20. Popkin BM. Patterns of beverage use across the lifecycle. Physiol Behav. 2010;100(1):4–9. [PMC free article] [PubMed]
21. Niemeier HM, Raynor HA, Lloyd-Richardson EE, et al. Fast food consumption and breakfast skipping: predictors of weight gain from adolescence to adulthood in a nationally representative sample. J Adolesc Health. 2006;39(6):842–849. [PubMed]
22. Gordon-Larsen P, Nelson MC, Popkin BM. Longitudinal physical activity and sedentary behavior trends: adolescence to adulthood. Am J Prev Med. 2004;27(4):277–283. [PubMed]
23. Nelson MC, Neumark-Sztainer D, Hannan PJ, et al. Longitudinal and secular trends in physical activity and sedentary behavior during adolescence. Pediatrics. 2006;118(6):E1627–E1634. [PubMed]
24. Matthews CE, Chen KY, Freedson PS, et al. Amount of time spent in sedentary behaviors in the United States, 2003–2004. Am J Epidemiol. 2008;167(7):875–881. [PMC free article] [PubMed]
25. Nelson MC, Story M, Larson NI, et al. Emerging adulthood and college-aged youth: an overlooked age for weight-related behavior change. Obesity. 2008;16(10):2205–2211. [PubMed]
26. Harris KM, Gordon-Larsen P, Chantala K, Udry JR. Longitudinal trends in race/ethnic disparities in leading health indicators from adolescence to young adulthood. Arch Pediatr Adolesc Med. 2006;160(1):74–81. [PubMed]
27. Scharoun-Lee M, Kaufman JS, Popkin BM, Gordon-Larsen P. Obesity, race/ethnicity and life course socioeconomic status across the transition from adolescence to adulthood. J Epidemiol Community Health. 2009;63(2):133–164. [PMC free article] [PubMed]
28. Scharoun-Lee M, Adair LS, Kaufman JS, Gordon-Larsen P. Obesity, race/ethnicity and the multiple dimensions of socioeconomic status during the transition to adulthood: a factor analysis approach. Soc Sci Med. 2009;68(4):708–716. [PMC free article] [PubMed]
29. Savin-Williams RC, Ream GL. Prevalence and stability of sexual orientation components during adolescence and young adulthood. Arch Sex Behav. 2007;36(3):385–394. [PubMed]
30. Igartua K, Thombs BD, Burgos G, Montoro R. Concordance and discrepancy in sexual identity, attraction, and behavior among adolescents. J Adolesc Health. 2009;45(6):602–608. [PubMed]
31. Rosario M, Schrimshaw EW, Hunter J, Braun L. Sexual identity development among gay, lesbian, and bisexual youths: consistency and change over time. J Sex Res. 2006;43(1):46–58. [PMC free article] [PubMed]
32. National Center for Education Statistics. [Accessed June 24, 2014];Enrollment rates of 18- to 24-year-olds in degree-granting institutions, by level of institution and sex and race/ethnicity of student: 1967 through 2012 (on-line) Available at: http://nces.ed.gov/programs/digest/d13/tables/dt13_302.60.asp.
33. Laska MN, VanKim NA, Erickson DJ, et al. Disparities in weight and weight behaviors by sexual orientation in college students. Am J Public Health. 2015;105(1):111–121. [PubMed]
34. Austin SB, Ziyadeh NJ, Corliss HL, et al. Sexual orientation disparities in purging and binge eating from early to late adolescence. J Adolesc Health. 2009;45(3):238–245. [PMC free article] [PubMed]
35. Laska MN, Pasch KE, Lust K, et al. Latent class analysis of lifestyle characteristics and health risk behaviors among college youth. Prev Sci. 2009;10(4):376–386. [PMC free article] [PubMed]
36. VanKim NA, Erickson DJ, Eisenberg ME, et al. Weight disparities for transgender college students. Health Behav Policy Rev. 2014;1(2):161–171. [PMC free article] [PubMed]
37. Corliss HL, Goodenow CS, Nichols L, Austin SB. High burden of homelessness among sexual-minority adolescents: findings from a representative Massachusetts high school sample. Am J Public Health. 2011;101(9):1683–1689. [PMC free article] [PubMed]
38. Cochran SD, Mays VM. Physical health complaints among lesbians, gay men, and bisexual and homosexually experienced heterosexual individuals: results from the California Quality of Life Survey. Am J Public Health. 2007;97(11):2048–2055. [PubMed]
39. Cochran SD, Mays VM. Burden of psychiatric morbidity among lesbian, gay, and bisexual individuals in the California Quality of Life Survey. J Abnorm Psychol. 2009;118(3):647–658. [PMC free article] [PubMed]
40. Centers for Disease Control and Prevention. [Accessed January 23, 2006];Youth Risk Behavior Surveillance System (on-line) Available at: http://www.cdc.gov/HealthyYouth/yrbs/index.htm.
41. Nelson MC, Lytle LA. Development and evaluation of a brief screener to estimate fast food and beverage consumption among adolescents. J Am Diet Assoc. 2009;109(4):730–734. [PMC free article] [PubMed]
42. Laska MN, Pasch KE, Lust K, et al. The differential prevalence of obesity and related behaviors in two- vs. four-year colleges. Obesity. 2011;19(2):453–456. [PMC free article] [PubMed]
43. Nelson MC, Lust K, Story M, Ehlinger E. Credit card debt, stress and key health risk behaviors among college students. Am J Health Promot. 2008;22(6):400–407. [PubMed]
44. Young LR, Nestle M. The contribution of expanding portion sizes to the US obesity epidemic. Am J Public Health. 2002;92:246–249. [PubMed]
45. Diliberti N, Bordi PL, Conklin MT, et al. Increased portion size leads to increased energy intake in a restaurant meal. Obes Res. 2004;12(3):562–568. [PubMed]
46. Dietary Guidelines Advisory Committee. The Report of the Dietary Guidelines Advisory Committee on Dietary Guidelines for Americans. Washington, DC: US Department of Health and Human Services; 2005. [Accessed February 16, 2015]. Available at: http://www.health.gov/dietaryguidelines/dga2005/document/pdf/dga2005.pdf.
47. VanKim NA, Laska MN, Ehlinger E, et al. Understanding young adult physical activity, alcohol and tobacco use in community colleges and 4-year post-secondary institutions: a cross-sectional analysis of epidemiological surveillance data. BMC Public Health. 2010;10:208. [PMC free article] [PubMed]
48. American Academy of Pediatrics. Children, adolescents, and television. Pediatrics. 2001;107(2):423–426. [PubMed]
49. Centers for Disease Control and Prevention. Sexual identity, sex of sexual contacts, and health-risk behaviors among students in grades 9–12 - Youth Risk Behavior Surveillance, Selected Sites, United States, 2001–2009. MMWR Morb Mortal Wkly Rep. 2011;60(SS-7):1–133. [PubMed]
50. Austin SB, Ziyadeh NJ, Kahn JA, et al. Sexual orientation, weight concerns, and eating-disordered behaviors in adolescent girls and boys. J Am Acad Child Adolesc Psychiatry. 2004;43(9):1115–1123. [PubMed]
51. Collins LM, Lanza ST. Latent Class and Latent Transition Analysis: With Applications in the Social, Behavioral, and Health Sciences. Hoboken, NJ: John Wiley & Sons, Inc; 2010.
52. McCutcheon AL. Latent Class Analysis. Thousand Oaks, CA: Sage Publications; 1987.
53. Berlin KS, Williams NA, Parra GR. An introduction to latent varibale mixture modeling (part 1): overview and cross-sectional latent class and latent profile analyses. J Pediatr Psychol. 2013:1–14. [PubMed]
54. Haines J, Neumark-Sztainer D. Prevention of obesity and eating disorders: a consideration of shared risk factors. Health Educ Res. 2006;21(6):770–782. [PubMed]
55. Rothenberger Institute. [Accessed January 20, 2015];College Sleep, Nutrition, and Exercise Courses (on-line) Available at: http://www.ri.umn.edu/Courses/SEE-course-impact.php.
56. Nelson TF, Gortmaker SL, Subramanian SV, Wechsler H. Vigorous physical activity among college students in the United States. J Phys Act Health. 2007;4:495–508. [PubMed]
57. VanKim NA, Nelson TF. Vigorous physical activity, mental health, perceived stress, and socializing among college students. Am J Health Promot. 2013;28(1):7–15. [PMC free article] [PubMed]