Fast Food Intake: Longitudinal Trends during the Transition to Young Adulthood and Correlates of Intake
Article Outline
Abstract
Purpose
Frequent fast food intake is associated with poorer diet quality and greater weight gain. The aims of this study were to describe changes in fast food intake during the transition from middle adolescence to young adulthood, and to identify baseline correlates of this eating behavior in early young adulthood.
Methods
Data were drawn from Project EAT, a population-based, longitudinal study in Minnesota. Surveys were completed by 935 females and 751 males in high school classrooms at baseline (1998–1999; mean age = 15.9 years) and by mail at follow-up (2003–2004; mean age = 20.5 years).
Results
Frequent intake of fast food (≥3 times/week) was reported by 24% of males and 21% of females during adolescence. At follow-up, in early young adulthood the eating behavior increased among males (33%, p < .001), and there was no further increase among females (23%; p = .16). Baseline snack frequency was positively associated with frequency of fast food intake at follow-up among both genders. Baseline peer support for healthy eating among males and both concern about health and self-efficacy for healthy eating among females were inversely related to follow-up fast food intake. Among females, baseline perceptions of time and taste barriers to healthy eating, lunch frequency, television viewing, and unhealthy food availability at home were also positively associated with follow-up fast food intake.
Conclusions
Interventions are needed to address the high prevalence of frequent fast food intake among adolescents and young adults. Health professionals should help young people identify convenient and healthful food choices for meals and snacks consumed away from home.
Keywords: Fast food, Young adult, Adolescent, Longitudinal study
Frequent intake of fast food is associated with poorer diet quality and greater weight gain [1], [2], [3]. The transition from adolescence to young adulthood may be an important time for interventions to address fast food intake. This developmental period is a high-risk time for the development of being overweight [4], and increases in fast food intake during the transition to young adulthood have been found to predict greater weight gain [5], [6].
Designing effective interventions will require further exploration of longitudinal trends in fast food restaurant use during the transition to young adulthood and a strong understanding of what factors during adolescence might influence the behavior as young people progress to adulthood. Only a few studies have investigated influences on fast food intake in early young adulthood [7], [8], [9]. These studies have considered a limited number of potential influencing factors (e.g., household residence) within the scope of health behavior theories (e.g., Social Cognitive Theory [SCT]) that form the framework of effective interventions. Although research among adolescents and adults has considered a broader scope of potential influences on fast food intake than research among young adults, there are important differences between these three life stages. It cannot be assumed that factors relevant to fast food intake in adolescence and adulthood are also relevant in early young adulthood.
The present study aimed to build on prior research by examining longitudinal trends in fast food intake during the transition from adolescence to young adulthood, and by identifying factors during adolescence that are correlated with fast food intake in early young adulthood. It was hypothesized there would be a longitudinal increase in the percentages of males and females with frequent fast food intake (≥3 times/week). Prior research and the framework of the SCT [10] were used to identify potential correlates of fast food intake. Correlates were hypothesized to be similar across gender, and include factors from each of the following domains of influence: personal, behavioral, and socioenvironmental.
Methods
Sample and study design
Data for the current study were drawn from Project EAT (Eating Among Teens), a prospective, population-based study designed to examine determinants of dietary intake and weight status [11], [12], [13]. The sample consisted of 1686 young adults (45% male) who completed questions assessing fast food intake on the Project EAT survey at baseline and follow-up. At baseline the mean age of participants was 15.9 years (SD = .8 years) and, at follow-up their mean age was 20.5 years (SD = .9 years).
For Project EAT-I (1998–1999), 3074 Minnesota high school students completed the Project EAT survey and food frequency questionnaires in their classrooms. Five years later, Project EAT-II (2003–2004) aimed to resurvey by mail all original participants for whom contact information was available to examine changes in their eating patterns and weight status over time. The Project EAT survey, food frequency questionnaires, and a letter explaining that return of completed surveys implied written consent were sent by mail to the addresses provided by participants during EAT-I. Follow-up survey data was collected from 68% of those contacted (n = 1710 young adults), representing 56% of the original cohort. All study protocols were approved by the University of Minnesota's Institutional Review Board Human Subjects Committee.
Surveys and measures
The development of the baseline Project EAT survey was guided by SCT, focus group discussions with adolescents [14], an in-depth literature review, and pilot testing. For the follow-up Project EAT survey, 55% of items were retained without modification, but some revisions were made to improve the relevance of items for young adults.
Fast food intake was assessed at baseline and follow-up using a single item on the Project EAT survey: “In the past week, how often did you eat something from a fast-food restaurant (like McDonald's, Burger King, Hardee's, etc)?” Six response categories ranged from “never” to “more than seven times.” As only 7% of young adults reported having fast food five or more times in the past week, response categories were collapsed to 0 times (low), one to two times (moderate), and three or more times (frequent). Although this measure of fast food intake has not been formally validated, other studies [6], [15], [16] have used similar measures. Analyses examining the relationship of usual past-year dietary intake, assessed by a food frequency questionnaire, with report of fast food intake at follow-up were conducted for the present study to internally investigate the construct validity of this measure (data available from the first author). Young adult intakes of total energy, fat energy, saturated fat energy, sodium, french fries, and soft drinks were higher among those who reported frequent fast food intake in comparison to those who reported moderate and low fast food intake.
Sociodemographic characteristics of participants were assessed using the Project EAT surveys administered at baseline or at follow-up. Race/ethnicity was assessed at baseline with the question: “Do you think of yourself as…White, Black or African American, Hispanic or Latino, Asian American, Hawaiian or Pacific Islander, or American Indian or Native American.” Subjects could choose more than one category; responses indicating multiple categories were coded as “mixed or other.” Classification tree methodology [17] was used to generate five categories of socioeconomic status (SES). The prime determinant of SES was the higher educational level of either parent at baseline [11]. At follow-up, participants also reported their hours of weekly employment, postsecondary student status, and current living situation as defined by where they had lived for the majority of the past year.
Personal, behavioral, and socioenvironmental variables were assessed using the Project EAT survey at baseline. The majority of variables briefly described below, have been described in more detail in other publications [2], [18]. An indicator of internal consistency, Cronbach's alpha, is provided when variables were based on the sum of three or more items [19].
Personal variablesConcern about health was based on five items that assessed how much adolescents cared about being healthy and eating healthy foods (range: 5–20, Cronbach's α = .69). Perceived taste barriers to healthy eating was based on four items assessing taste preferences for fruit, vegetables, healthy foods, and unhealthy foods (range: 4–16, Cronbach's α = .65). Similarly, perceived time barriers to healthy eating was based on four items assessing attitudes about the amount of time it takes to eat healthy foods (range: 4–16, Cronbach's α = .68). Perceived benefits of healthy eating was based on five items assessing attitudes about the extent to which dietary intake can affect health, appearance, and performance (range: 5–20, Cronbach's α = .80) [20]. Self-efficacy for healthy eating was based on nine statements assessing how sure adolescents were they could eat healthy food in various situations (range: 9–54, Cronbach's α = .87) [21]. Body satisfaction was assessed using a modified version of the Body Shape Satisfaction Scale [22] (range: 10–50, Cronbach's α = .92). Weight concern was based on response to four statements such as “I think a lot about being thinner” (range: 4–16, Cronbach's α = .80).
Behavioral variablesFrequency of eating breakfast, lunch, and dinner (range: 0–7) were self-reported for the past week [21]. Frequency of eating snacks (range: 0–6) was assessed with the question: “How many times did you snack (eat in-between meals) yesterday?” [21]. Food preparation (range: 0–7) and shopping involvement were assessed using questions that asked about how often adolescents helped with these tasks over the past week. Responses to the shopping item were dichotomized (yes/no). Sport involvement was assessed with a question about the number of sport teams played on during the past year [23]; responses were dichotomized (1 or more/none). Television viewing was assessed using two items that separately asked about hours watched on an average weekday and weekend day [24], [25]; weekly hours of television viewing was computed (range: 0–35). Use of healthy and unhealthy weight control behaviors were self-reported for the past year. Adolescents indicated whether they had used (yes/no) any of four healthy methods (e.g., ate more fruits and vegetables) or any of nine unhealthy methods (e.g., took diet pills).
Socioenvironmental variablesParental support for healthy eating was based on four items assessing perceptions of how strongly one's mother and father care about and encourage eating healthy food (range: 4–16, Cronbach's α = .64). Peer support for healthy eating (range: 1–4) was based on agreement with the statement “Many of my friends care about eating healthy food.” Family meal frequency (range: 0–9) was assessed with the question: “During the past 7 days, how many times did all, or most, of your family living in your house eat a meal together?” [26]. Availability of healthy and unhealthy foods at home was based on eight statements. Participants reported the frequency of having four healthy foods (e.g., fruit and vegetables, range: 4–16, Cronbach's α = .60) and four unhealthy foods (e.g., potato chips or other salty snacks, range: 4–16, Cronbach's α = .79) available at home. Food insecurity was assessed with the question: “How often during the last 12 months have you been hungry because your family could not afford food?” and responses were dichotomized to represent the presence or absence of food insecurity.
Statistical analyses
Mixed linear regression models were used to generate adjusted prevalences of frequent fast food intake (≥3 times/week) in 1999 and 2004 and to test whether changes in prevalence were statistically significantly different from the null. Models included a main effect term for year and a random effect term for individuals to account for the tracking of behavior within individuals (longitudinal or repeat correlation). Chi-squared tests were used to separately examine associations of the three-level fast food intake variable (0 = low, 1 = moderate, 2 = frequent) with each sociodemographic characteristic considered.
To examine associations between each of the potential correlates (personal, behavioral, and socioenvironmental variables) assessed at baseline with follow-up fast food intake, ordinal logistic regression analyses were conducted on the three-level dependent fast food intake variable using two models. Model 1 controlled for sociodemographic characteristics and Model 2 controlled for both sociodemographic characteristics and baseline fast food intake. Model 1 was used to examine the total association of potential correlates with fast food intake at follow-up. Model 2 was used to examine associations between potential correlates and change in fast food intake over the 5-year study period. Separate models 1 and 2 were fit for each variable examined.
For Models 1 and 2, nondichotomous personal, behavioral, and socioenvironmental variables were standardized to allow for relative comparisons of strength between the observed associations. Models for every potential correlate satisfied the proportionality of odds assumption for ordinal logistic regression, indicating that odds ratios (ORs) for being in either of the two higher intake categories (frequent or moderate) versus the lowest intake category (low) were not statistically different from the ORs for being in the highest intake category (frequent) versus either of the two lower intake categories (moderate or low) [27]. The only exception was for snack frequency in males. To examine the association of snack frequency with fast food intake in males, separate binary logistic regression models were fit.
Of the 1710 young adults who provided follow-up survey data, exclusions were made for those missing data on fast food intake (n = 24). Data were weighted to adjust for demographic differences in response to Projects EAT-I and EAT-II using the response propensity method [28]. The weighted ethnic/racial and SES proportions of the study sample are as follows: 55.5% white, 17.4% Asian, 16.2% African American, and 10.9% mixed or other race, whereas SES percentages are low (17.9%), low-middle (18.8%), middle (24.9%), upper-middle (25.7%), and high (12.7%). Concern regarding the missing-at-random assumption was lessened by the finding that after weighting and adjustment for demographic characteristics, baseline fast food intake among responders was not significantly different from baseline fast food intake among nonresponders to Project EAT-II.
All analyses were stratified by gender. A 95% confidence level was used to interpret the statistical significance of probability tests. Analyses were conducted using the Statistical Analysis System (SAS version 8.2, SAS Institute, Cary, NC, 2001).
Results
Longitudinal trends in fast food intake
During the transition from middle adolescence (baseline) to young adulthood (follow-up) the proportion of males that reported frequent fast food intake rose from 23.6% to 33.0% (change = 9.4%, t = 4.56, p < .001). In contrast, the proportion of females that reported frequent fast food intake did not significantly increase (change = 2.3%, t = 1.41, p = .16). Among females, the proportion that reported frequent fast food intake was 20.5% during adolescence and 22.8% during young adulthood.
Unadjusted associations of sociodemographic characteristics with follow-up fast food intake
Low, moderate, and frequent levels of fast food intake were reported by 21.8%, 50.9%, and 27.3% of young adults, respectively. A greater proportion of males (33.0%) than females (22.8%) reported frequent fast food intake (χ2 = 25.11, p < .001). Frequent fast food intake was most common among males and females of low-middle SES (males: 43.8%; females: 29.5%) and in the race category mixed or other (males: 36.2%; females: 32.7%; Table 1). Univariate analyses also indicated that among females, fast food intake was related to hours of weekly employment, postsecondary student status, and living situation. Frequent fast food intake was most common among females who were working full-time or ≥40 hours per week (26.3%). Frequent fast food intake was least common among females attending a 4-year college full-time (14.5%) and those who were living on a college/university campus (9.8%).
Table 1. Sociodemographic characteristics of male and female participants in Project EAT (Eating Among Teens) by follow-up fast food intake
| Fast food intake (past week) | ||||||
|---|---|---|---|---|---|---|
| Males (n = 751) | Females (n = 935) | |||||
| Low (0 times) | Moderate (1–2 times) | Frequent (≥3 times) | Low (0 times) | Moderate (1–2 times) | Frequent (≥3 times) | |
| % (n) | 18.2 (137) | 48.8(366) | 33.0(248) | 24.7(231) | 52.5(491) | 22.8(213) |
| Race/ethnicity | ||||||
| White | 19.3(84) | 46.1(200) | 34.6(150) | 29.7(146) | 50.0(246) | 20.3(100) |
| Asian | 24.0(28) | 53.7(62) | 22.3(26) | 22.1(39) | 58.9(103) | 19.0(33) |
| African American | 8.5(10) | 57.8(65) | 33.7(38) | 17.2(27) | 52.8(82) | 30.0(47) |
| Mixed/other | 18.1(15) | 45.7(38) | 36.2(30) | 17.3(17) | 50.0(50) | 32.7(33) |
| χ2 = 16.06; p = .01 | χ2 = 23.22; p < .001 | |||||
| Socioeconomic status | ||||||
| Low | 20.5(25) | 53.6(64) | 25.9(31) | 24.2(43) | 48.5(86) | 27.3(48) |
| Low-middle | 15.9(21) | 40.3(54) | 43.8(59) | 15.5(27) | 55.0(97) | 29.5(52) |
| Middle | 17.2(29) | 52.9(91) | 29.9(51) | 23.5(57) | 54.4(131) | 22.1(53) |
| Upper-middle | 16.4(35) | 51.6(110) | 32.0(68) | 26.0(55) | 55.9(119) | 18.1(39) |
| High | 26.5(26) | 40.5(40) | 33.0(33) | 42.0(47) | 41.6(46) | 16.4(18) |
| χ2 = 16.95; p = .03 | χ2 = 32.79; p < .001 | |||||
| Hours of weekly employment | ||||||
| 0 hours/week | 21.0(21) | 57.3(58) | 21.7(22) | 28.6(46) | 50.3(82) | 21.1(34) |
| 1–39 hours/week | 17.1(63) | 49.0(179) | 33.9(124) | 27.4(142) | 51.3(266) | 21.3(110) |
| ≥40 hours/week | 19.2(53) | 44.0(121) | 36.8(101) | 17.2(43) | 56.5(140) | 26.3(65) |
| χ2 = 8.67; p = .07 | χ2 = 11.21; p = .02 | |||||
| Postsecondary student status | ||||||
| Not in college | 17.1(57) | 49.8(167) | 33.1(111) | 21.8(69) | 49.2(155) | 29.0(91) |
| Part-time | 17.6(13) | 53.1(40) | 29.3(22) | 19.9(27) | 56.1(77) | 24.0(33) |
| Full-time, community college | 10.6(10) | 47.1(46) | 42.3(42) | 20.2(29) | 55.0(80) | 24.8(36) |
| Full-time, 4-year college | 23.6(56) | 46.5(110) | 29.9(71) | 31.9(104) | 53.6(175) | 14.5(47) |
| χ2 = 11.42; p = .08 | χ2 = 27.24; p < .001 | |||||
| Living situation | ||||||
| Own/rent | 18.1(45) | 45.8(113) | 36.1(89) | 25.8(95) | 51.3(189) | 22.9(84) |
| With parents | 16.0(62) | 51.8(201) | 32.2(125) | 20.0(82) | 52.7(216) | 27.3(111) |
| On campus | 25.0(23) | 45.8(42) | 29.2(27) | 35.7(46) | 54.5(70) | 9.8(13) |
| Other | 31.7(7) | 42.2(9) | 26.1(6) | 26.7(7) | 60.5(16) | 21.8(3) |
| χ2 = 8.55; p = .20 | χ2 = 24.97; p < .001 | |||||
Associations of personal, behavioral, and socioenvironmental factors with follow-up fast food intake
Among males, peer support for healthy eating at baseline was associated with lower fast food intake at follow-up (OR = 0.85, confidence interval [CI] = 0.74, 0.98) and decreases in intake (OR = 0.86, CI = 0.75, 0.99) between baseline and follow-up (Table 2, Models 1 and 2). For every standard deviation increase in perceived peer support, the odds of males being in a higher fast food intake category (vs. the next lower category) at follow-up were reduced by 14% (Model 2). Snack frequency was not associated with increases in fast food intake at follow-up (Model 2). However, snack frequency was positively associated with increased odds of having frequent versus low fast food intake at follow-up (Model 1: OR = 1.45, CI = 1.01, 2.08).
Table 2. Associations of baseline personal, behavioral, and socioenvironmental factors with fast food intake at follow-up among males (n = 751)
| Social cognitive theory factors | Model 1a,b | Model 2a,c | ||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Personal factors | ||||
| Concern about health | 0.87 | 0.75, 1.00 | 0.90 | 0.78, 1.04 |
| Perceived taste barriers to healthy eating | 1.05 | 0.91, 1.21 | 1.03 | 0.89, 1.19 |
| Perceived time barriers to healthy eating | 1.03 | 0.90, 1.19 | 1.00 | 0.86, 1.15 |
| Perceived benefits of healthy eating | 1.01 | 0.87, 1.17 | 1.02 | 0.88, 1.18 |
| Self-efficacy for healthy eating | 1.06 | 0.92, 1.22 | 1.07 | 0.93, 1.24 |
| Body satisfaction | 0.97 | 0.85, 1.12 | 0.97 | 0.84, 1.12 |
| Weight concerns | 0.99 | 0.86, 1.14 | 1.01 | 0.87, 1.16 |
| Behavioral factors | ||||
| Breakfast frequency | 0.99 | 0.86, 1.15 | 1.03 | 0.89, 1.19 |
| Lunch frequency | 1.14 | 0.99, 1.32 | 1.12 | 0.96, 1.29 |
| Dinner frequency | 1.01 | 0.87, 1.17 | 0.99 | 0.86, 1.15 |
| Snack frequencyd | ||||
| high vs. low fast food intake | 1.45 | 1.01, 2.08 | 1.18 | 0.80, 1.73 |
| moderate vs. low fast food intake | 0.86 | 0.68, 1.08 | 0.84 | 0.67, 1.06 |
| Food preparation involvement | 1.09 | 0.95, 1.26 | 1.09 | 0.94, 1.26 |
| Shopping involvement | 0.84 | 0.63, 1.12 | 0.81 | 0.61, 1.09 |
| Sport involvement | 1.31 | 0.96, 1.79 | 1.33 | 0.94, 1.80 |
| Television viewing | 1.09 | 0.95, 1.26 | 1.06 | 0.92, 1.22 |
| Healthy weight control behaviors | 0.97 | 0.72, 1.32 | 1.01 | 0.74, 1.37 |
| Unhealthy weight control behaviors | 1.00 | 0.73, 1.37 | 1.00 | 0.72, 1.38 |
| Socioenvironmental factors | ||||
| Parental support for healthy eating | 0.94 | 0.81, 1.08 | 0.94 | 0.81, 1.09 |
| Peer support for healthy eating | 0.85 | 0.74, 0.98 | 0.86 | 0.75, 0.99 |
| Family meal frequency | 0.98 | 0.85, 1.13 | 0.99 | 0.85, 1.14 |
| Healthy food availability at home | 1.11 | 0.96, 1.30 | 1.10 | 0.95, 1.28 |
| Unhealthy food availability at home | 1.12 | 0.96, 1.30 | 1.04 | 0.89, 1.21 |
| Food insecurity | 1.41 | 0.80, 2.50 | 1.46 | 0.82, 2.59 |
aOdds ratios are standardized and are interpreted as the difference in odds of being in each category of fast food intake compared to the next lower category (i.e., frequent versus moderate, moderate vs. low) associated with a 1 standard deviation change in the personal, behavioral, or socioenvironmental variable. Statistically significant odds ratios are shown in bold. |
bModel 1: Each association was tested separately, weighted for nonresponse, and adjusted for race/ethnicity, postsecondary student status, employment status, socioeconomic status, and living situation. |
cModel 2: Each association was tested separately, weighted for nonresponse, and adjusted for the covariates in Model 1 plus baseline fast food intake. |
dSeparate binary logistic models were fit because the parallel regression assumption was rejected. |
Among females, increases in fast food intake at follow-up (Table 3, Model 2) were associated with perceived taste barriers to healthy eating (OR = 1.18, CI = 1.03, 1.35), lunch frequency (OR = 1.18, CI = 1.04, 1.35), snack frequency (OR = 1.17, CI = 1.03, 1.34), time spent watching television (OR = 1.30, CI = 1.14, 1.49), and unhealthy food availability at home (OR = 1.19, CI = 1.04, 1.37). For example, with every standard deviation increase in time spent watching television at baseline, the odds of females being in a higher fast food intake category (vs. the next lower category) at follow-up were increased by 30% (Model 2). Before adjustment for baseline fast food intake, follow-up fast food intake was also found to be inversely associated with concern about health and self-efficacy for healthy eating, and positively associated with time barriers to healthy eating.
Table 3. Associations of baseline personal, behavioral, and socioenvironmental factors with fast food intake at follow-up among females (n=935)
| Social cognitive theory factors | Model 1a,b | Model 2a,c | ||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Personal Factors | ||||
| Concern about health | 0.83 | 0.73, 0.95 | 0.88 | 0.77, 1.00 |
| Perceived taste barriers to healthy eating | 1.26 | 1.11, 1.44 | 1.18 | 1.03, 1.35 |
| Perceived time barriers to healthy eating | 1.18 | 1.04, 1.34 | 1.09 | 0.96, 1.24 |
| Perceived benefits of healthy eating | 0.92 | 0.83, 1.05 | 0.97 | 0.85, 1.11 |
| Self-efficacy for healthy eating | 0.85 | 0.75, 0.98 | 0.89 | 0.78, 1.02 |
| Body satisfaction | 1.01 | 0.89, 1.16 | 0.98 | 0.86, 1.12 |
| Weight concerns | 0.92 | 0.81, 1.05 | 0.95 | 0.83, 1.08 |
| Behavioral Factors | ||||
| Breakfast frequency | 0.97 | 0.85, 1.11 | 0.94 | 0.82, 1.08 |
| Lunch frequency | 1.24 | 1.09, 1.41 | 1.18 | 1.04, 1.35 |
| Dinner frequency | 0.93 | 0.82, 1.06 | 0.92 | 0.81, 1.05 |
| Snack frequency | 1.25 | 1.10, 1.43 | 1.17 | 1.03, 1.34 |
| Food preparation involvement | 0.92 | 0.81, 1.06 | 0.99 | 0.86, 1.14 |
| Shopping involvement | 0.98 | 0.76, 1.27 | 0.96 | 0.74, 1.25 |
| Sport involvement | 0.97 | 0.74, 1.28 | 0.85 | 0.64, 1.12 |
| Television viewing | 1.34 | 1.17, 1.53 | 1.30 | 1.14, 1.49 |
| Healthy weight control behaviors | 0.93 | 0.64, 1.36 | 0.93 | 0.64, 1.37 |
| Unhealthy weight control behaviors | 0.85 | 0.65, 1.11 | 0.87 | 0.67, 1.14 |
| Socioenvironmental Factors | ||||
| Parental support for healthy eating | 0.92 | 0.81, 1.05 | 0.97 | 0.85, 1.12 |
| Peer support for healthy eating | 0.97 | 0.85, 1.11 | 1.02 | 0.89, 1.16 |
| Family meal frequency | 0.95 | 0.83, 1.08 | 0.95 | 0.83, 1.08 |
| Healthy food availability at home | 0.90 | 0.78, 1.03 | 0.88 | 0.76, 1.01 |
| Unhealthy food availability at home | 1.32 | 1.15, 1.51 | 1.19 | 1.04, 1.37 |
| Food insecurity | 1.11 | 0.69, 1.77 | 1.05 | 0.65, 1.69 |
aOdds ratios are standardized and are interpreted as the difference in odds of being in each category of fast food intake compared to the next lower category (i.e., frequent versus moderate, moderate versus low) associated with a 1 standard deviation change in the personal, behavioral, or socioenvironmental variable. Statistically significant odds ratios are shown in bold. |
bModel 1: Each association was tested separately, weighted for nonresponse and adjusted for race/ethnicity, postsecondary student status, employment status, socioeconomic status, and living situation. |
cModel 2: Each association was tested separately, weighted for nonresponse and adjusted for the covariates in Model 1 plus baseline fast food intake. |
To estimate the total predictive ability of the covariates examined here to determine fast food intake at follow-up, final ordinal logistic regression models were run for males and females including sociodemographic characteristics, fast food intake at baseline, and all of the baseline personal, behavioral, and socioenvironmental factors simultaneously as independent variables. The total predictive ability of covariates was similar for males (generalized R2 = .12) and females (generalized R2 = .17). For comparison, the total predictive ability of a model including only fast food intake at baseline and sociodemographic characteristics as independent variables was even lower for males (generalized R2= .07) and females (generalized R2 = .12).
Discussion
This study investigated longitudinal changes in fast food intake during the transition from adolescence to young adulthood. Frequent fast food intake (≥3 times/week) was reported by 20% to 33% of young people during the transition to young adulthood. Although no longitudinal change during the transition period was found in the prevalence of frequent fast food intake among females, there was an increase in the prevalence of frequent fast food intake among males. Other survey data has similarly shown there is an overall increase in the number of days per week that young people eat at a fast food restaurant during the transition from adolescence to young adulthood and smaller increases among females than males [5].
This study also examined correlates of fast food intake during the transition to young adulthood. Consistent with geographic analyses of fast food restaurant density [29] and prior observational research both in samples of adolescents [2], [16], [30] and adults [1], [15], the current study found frequent fast food intake was more common among young adults in nonwhite race categories (African American and mixed or other) and the low-middle category of SES. The potential value of using SCT in the design of interventions to reduce fast food consumption was suggested by associations of baseline personal (i.e., concern about health), behavioral (i.e., lunch and snack frequency, time spent watching television), and socioenvironmental (i.e., peer support for healthy eating, unhealthy food availability at home) factors with increases or decreases in fast food intake.
Although a range of factors were associated with fast food intake among females, a limited number of factors were associated with changes in fast food intake or intake at follow-up among males. Factors not included in this manuscript such as characteristics of neighborhood and school environments (e.g., fast food sales and advertising on high school or college campuses, hours of operation for neighborhood restaurants) and economic factors (e.g., fast food prices, personal income) may have greater influence on fast food intake among males than the personal, behavioral, and socioenvironmental factors that were considered here and related to fast food intake among females. Further, the sum predictive ability of factors examined in this study was low for both genders. Future research will need to examine a broader range of potential influences on fast food intake, and might consider whether the strength of influences vary according to the type of fast food restaurant considered. In particular, previous studies have suggested that there is a need to examine the influence of young adults' social and physical environments (e.g., social norms among peers and coworkers, neighborhood availability of fast food) on their intake of fast food [7].
Of all the personal, behavioral, and socioenvironmental factors examined, frequency of snacking was the only factor found to be associated with higher fast food intake in both males and females. The observed associations between baseline snack frequency and follow-up fast food intake were consistent with cross-sectional research in the Project EAT sample at baseline [2], and might be because of frequent purchases of snacks from fast food restaurants. Young people who frequently purchase snacks from fast food restaurants during adolescence may maintain or increase the frequency of this behavior as they transition to adulthood. These findings suggest a need for helping adolescents to choose healthy food options when eating in between meals.
This study had several strengths, but was not without limitations. Strengths include the prospective design, the large and diverse sample, and the use of a theoretical model to guide our selection of factors for examination from the Project EAT survey. Although the sample was diverse in terms of young adults' race/ethnicity, family SES, employment status, and postsecondary student status, participants were drawn from one Midwestern state. Sampling weights correcting for nonresponse were used in all analyses; however, attrition from the baseline population may have still reduced the representativeness of the sample.
Although our measure of fast food intake was similar to measures used by other studies [6], [15], [16] and was related to usual past year dietary intake in the expected way at baseline [2] and follow-up, the item inquired only about fast food intake in the past week. This study assumed that fast food intake reported for the past 7-day period, on average, reflects usual fast food intake. Multiple univariate tests were conducted to examine associations of sociodemographic characteristics with frequent fast food intake, and therefore, the findings presented do not necessarily represent independent associations. Many variables were also examined, and were statistically tested for their association with fast food intake in multivariate models. Given the scope of the survey, some brief and unvalidated measures were used. With multiple comparisons it is important to remember that some of the associations observed could be spurious; however, factors associated with fast food intake in the current study were largely consistent with other studies examining influences on fast food intake among adolescents and adults [2], [15], [31].
Findings of this study indicate there is a need for nutrition interventions to address fast food intake during the transition to young adulthood. Interventions targeted to male adolescents should emphasize supporting peers to engage in healthy eating behaviors. Interventions for female adolescents should build concern for healthy eating, reduce exposure to television advertisements for fast food restaurants, and could aim to reduce perceived barriers to healthy eating by providing opportunities to taste and prepare nutritious, convenient foods. Taking available food dollars into consideration, health professionals should help young people to identify food and beverage options that could be consumed away from home and that would best allow for meeting nutritional requirements within their energy needs [32], [33]. Healthy and portable options that could be brought from home (e.g., bottled water, fresh fruit) should be discussed as well as guidelines for managing portion sizes (e.g., ordering smaller portions, sharing orders) and selecting nutrient-dense options from fast food menus (e.g., low-fat milk) [34].
Acknowledgements
Data collection was supported by grant number R40 MC 00319 (D. Neumark-Sztainer, principal investigator) from the Maternal and Child Health Bureau (Title V, Social Security Act), Health Resources and Service Administration, Department of Health and Human Services. Analyses were supported by the Adolescent Health Protection Program, grant T01-DP000112 (L. Bearinger, principal investigator) from the Centers for Disease Control and Prevention (CDC), Department of Health and Human Services. The content of the manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the CDC.
References
- Fast-food habits, weight gain, and insulin resistance (the CARDIA study): 15-year prospective analysis. Lancet. 2005;365:36–42
- Fast food restaurant use among adolescents: associations with nutrient intake, food choices, and behavioral and psychosocial variables. Int J Obesity. 2001;25:1823–1833
- Differential associations of fast food and restaurant food consumption with 3-y change in body mass index: the Coronary Artery Risk Development in Young Adults Study. Am J Clin Nutr. 2007;85:201–208
- Five-year obesity incidence in the transition period between adolescence and adulthood: the National Longitudinal Study of Adolescent Health. Am J Clin Nutr. 2004;80:569–575
- Longitudinal trends in race/ethnic disparities in leading health indicators from adolescence to young adulthood. Arch Pediatr Adolesc Med. 2006;160(1):74–81
- Fast food consumption and breakfast skipping: predictors of weight gain from adolescence to adulthood in a nationally representative sample. J Adolesc Health. 2006;39:842–849
- Body mass index gain, fast food, and physical activity: effects of shared environments over time. Obesity. 2006;14(4):701–709
- . Dietary status and eating out practices of college students. J Am Diet Assoc. 1992;92:867–869
- . Differences exist in the eating habits of university men and women at fast-food restaurants. Nutr Res. 2006;26:524–530
- . How individuals, environments, and health behavior interact: Social Cognitive Theory. In: Glanz K, Rimer B, Lewis F editor. Health Behavior and Health Education: Theory, Research and Practice. 3rd ed.. San Francisco, CA: Jossey-Bass; 2002;p. 165–184
- Overweight status and eating patterns among adolescents: where do youth stand in comparison to the Healthy People 2010 Objectives?. Am J Public Health. 2002;92(5):844–851
- Obesity, disordered eating, and eating disorders in a longitudinal study of adolescents: how do dieters fare 5 years later?. J Am Diet Assoc. 2006;106(4):559–568
- Overweight status and weight control behaviors in adolescents: longitudinal and secular trends from 1999 to 2004. Prev Med. 2006;43:52–59
- Factors influencing food choices of adolescents: findings from focus-group discussions with adolescents. J Am Diet Assoc. 1999;99:929–934937
- . Fast food restaurant use among women in the Pound of Prevention study: dietary, behavioral and demographic correlates. Int J Obes. 2000;24:1353–1359
- Fast-food intake and diet quality in black and white girls. Arch Pediatr Adolesc Med. 2005;159:626–631
- Classification and Regression Trees. Belmont, CA: Wadsworth International Group; 1984;
- Correlates of fruit and vegetable intake among adolescents: findings from Project EAT. Prev Med. 2003;37:198–208
- . Statistics notes: Cronbach's alpha. BMJ. 1997;314:572
- Survey development for assessing correlates of young adolescents' eating. Am J Health Behav. 2002;26(4):284–295
- . Influences on adolescent eating behavior. J Adolesc Health. 1996;18(1):27–34
- . Gender differences in body satisfaction. Obesity Res. 1997;5(5):402–409
- Reliability of the Youth Risk Behavior Survey Questionnaire. Am J Epidemiol. 1995;141(6):575–580
- Reducing obesity via a school-based interdisciplinary intervention among youth: Planet Health. Arch Pediatr Adolesc Med. 1999;153(4):409–418
- Reliability and validity of a brief questionnaire to assess television viewing and computer use by middle school children. J Sch Health. 2004;74(9):370–377
- Family meal patterns: associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc. 2003;103:317–322
- . Logistic Regression: A Self-Learning Text. 2nd ed.. New York: Springer; 2002;
- . Survey nonresponse adjustments for estimates of means. Int Stat Rev. 1986;54:139–157
- . Fast food, race/ethnicity, and income: a geographic analysis. Am J Prev Med. 2004;27(3):211–217
- Effects of fast-food consumption on energy intake and diet quality among children in a national household survey. Pediatrics. 2004;113(1 Pt 1):112–118
- . Eating at fast-food restaurants is associated with dietary intake, demographic, psychosocial and behavioural factors among African Americans in North Carolina. Public Health Nutr. 2004;7(8):1089–1096
- . Fast food: friendly?. Int J Obes. 2007;31(6):884–886
- . Dietary Guidelines for Americans 2005 [Online]. http://www.health.gov/dietaryguidelines/(accessed December 1, 2007)
- . Fast food and children and adolescents: implications for practitioners. Clin Pediatr. 2005;44:279–288
PII: S1054-139X(07)00664-7
doi:10.1016/j.jadohealth.2007.12.005
© 2008 Society for Adolescent Medicine. Published by Elsevier Inc. All rights reserved.
