Hostname: page-component-586b7cd67f-t8hqh Total loading time: 0 Render date: 2024-11-24T23:53:00.411Z Has data issue: false hasContentIssue false

Applying the socio-ecological model to understand factors associated with sugar-sweetened beverage behaviours among rural Appalachian adolescents

Published online by Cambridge University Press:  11 January 2021

Brittany A McCormick
Affiliation:
Department of Public Health Sciences, UVA Cancer Center Research and Outreach Office, University of Virginia, 16 East Main Street, Christiansburg, VA24073, USA
Kathleen J Porter
Affiliation:
Department of Public Health Sciences, UVA Cancer Center Research and Outreach Office, University of Virginia, 16 East Main Street, Christiansburg, VA24073, USA
Wen You
Affiliation:
Department of Public Health Sciences, School of Medicine, University of Virginia, Charlottesville, VA22903, USA
Maryam Yuhas
Affiliation:
Department of Nutrition and Food Studies, Syracuse University, Syracuse, NY13244, USA
Annie L Reid
Affiliation:
Department of Public Health Sciences, UVA Cancer Center Research and Outreach Office, University of Virginia, 16 East Main Street, Christiansburg, VA24073, USA
Esther J Thatcher
Affiliation:
Department of Population Health, University Hospitals, Cleveland, OH44106, USA
Jamie M Zoellner*
Affiliation:
Department of Public Health Sciences, UVA Cancer Center Research and Outreach Office, University of Virginia, 16 East Main Street, Christiansburg, VA24073, USA
*
*Corresponding author: Email [email protected]
Rights & Permissions [Opens in a new window]

Abstract

Objective:

The objective of the current study was to identify factors across the socio-ecological model (SEM) associated with adolescents’ sugar-sweetened beverage (SSB) intake.

Design:

This cross-sectional study surveyed adolescents using previously validated instruments. Analyses included descriptive statistics, ANOVA tests and stepwise nonlinear regression models (i.e., two-part models) adjusted to be cluster robust. Guided by SEM, a four-step model was used to identify factors associated with adolescent SSB intake – step 1: demographics (i.e., age, gender), step 2: intrapersonal (i.e., theory of planned behaviour (attitudes, subjective norms, perceived behavioural control, behavioural intentions), health literacy, media literacy, public health literacy), step 3: interpersonal (i.e., caregiver’s SSB behaviours, caregiver’s SSB rules) and step 4: environmental (i.e., home SSB availability) level variables.

Setting:

Eight middle schools across four rural southwest Virginia counties in Appalachia.

Participants:

Seven hundred ninety seventh grade students (55·4 % female, 44·6 % males, mean age 12 (sd 0·5) years).

Results:

Mean SSB intake was 36·3 (sd 42·5) fluid ounces or 433·4 (sd 493·6) calories per day. In the final step of the regression model, seven variables significantly explained adolescent’s SSB consumption: behavioural intention (P < 0·05), affective attitude (P < 0·05), perceived behavioural control (P < 0·05), health literacy (P < 0·001), caregiver behaviours (P < 0·05), caregiver rules (P < 0·05) and home availability (P < 0·001).

Conclusions:

SSB intake among adolescents in rural Appalachia was nearly three times above national mean. Home environment was the strongest predictor of adolescent SSB intake, followed by caregiver rules, caregiver behaviours and health literacy. Future interventions targeting these factors may provide the greatest opportunity to improve adolescent SSB intake.

Type
Research paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
© The Author(s), 2021. Published by Cambridge University Press on behalf of The Nutrition Society

Adolescence, aged 12–19 years, is a transitional period and health habits developed during this time often continue into adulthood(Reference Lake, Mathers and Rugg-Gunn1,Reference Kim, Sim and Park2) . Addressing and encouraging healthy dietary habits, including limiting sugar-sweetened beverages (SSB), are especially important during adolescence(Reference Kim, Sim and Park2). SSB include sweetened fruit flavoured drinks, regular soda or soft drinks, energy drinks, sports drinks, and sweetened coffees and teas(Reference Hedrick, Savla and Comber3). Excessive consumption of SSB has been linked to multiple health concerns such as obesity, obesity-related cancers, type 2 diabetes, CVD and dental caries(Reference Malik, Popkin and Bray4,Reference Tahmassebi, Duggal and Malik-Kotru5) .

Within the USA, SSB are substantial sources of increased calories and added sugar within the diets of adolescents(Reference Malik, Popkin and Bray4,Reference Rosinger, Herrick and Gahche6) . Among US adolescents, about 63 % consume at least one SSB per day and SSB contribute a mean of 143 calories per day(Reference Rosinger, Herrick and Gahche6). Adolescents are the highest consumers of SSB with more than 9 % of their total daily calories attributed to SSB consumption(Reference Rosinger, Herrick and Gahche6), with SSB consumption highest among adolescent males and increasing with age(Reference Rosinger, Herrick and Gahche6,Reference Park, Blanck and Sherry7) . Regional disparities in adolescent SSB intake are also evident, especially in rural Appalachia, the targeted region of this research(Reference Gutschall, Thompson and Lawrence8,Reference Southerland, Dula and Slawson9) . Adolescents within rural Appalachia have disproportionately high intakes of SSB(Reference Lane, Porter and Hecht10Reference Zoellner, Estabrooks and Davy12). More specifically, one regional study indicated mean SSB intake among middle school adolescents was about 457 calories per day, nearly three times the national mean(Reference Lane, Porter and Hecht10Reference Zoellner, Estabrooks and Davy12). Along with SSB concerns, this region suffers from high prevalence of SSB-related chronic health conditions (e.g., obesity, poor oral health) and faces substantial barriers to health, economic and social equality, such as lack of access to medical and preventative services, financial struggles, lack of health insurance, transportation issues, geographical isolation and food insecurity(Reference Gutschall, Thompson and Lawrence8,Reference Southerland, Dula and Slawson9) . Other Appalachian focused studies indicate important socio-cultural influences may account for higher SSB intake among adolescents, such as peer influence of cultural norms, being resistant to change or accepting help, matriarchal food gatekeeper and strong family ties(Reference Gutschall, Thompson and Lawrence8,Reference Southerland, Dula and Slawson9) .

With growing health concerns related to high US consumption of SSB, interventions and strategies have been developed with foundations in the socio-ecological model (SEM). The SEM is centred on highlighting interdependence, or lack thereof, of internal and external factors that influence an individual’s behaviours(Reference Townsend and Foster13,Reference McLeroy, Bibeau and Steckler14) . According to SEM, an individual’s behaviour is influenced by intrapersonal (e.g., knowledge, attitudes, self-concept), interpersonal (e.g., social norms, family, peers) and environmental factors (e.g., home environment, community, public policy) (Reference McLeroy, Bibeau and Steckler14). Intervention strategies that have applied multi-level SEM approaches have shown improvements related to SSB intake, healthy eating patterns, physical activity and childhood obesity(Reference Townsend and Foster13,Reference King, Stokols and Talen15,Reference Yuhas, Porter and Hedrick16) .

Specific to SSB, studies that have incorporated single-level approaches, such as focus on intrapersonal level factors, have shown constructs such as subjective norms, perceived behavioural control and media literacy play crucial roles in an adolescent’s intentions to consume SSB(Reference Kassem, Lee and Modeste17Reference Kumar, Onufrak and Zytnick22). Constructs from the theory of planned behaviour (TPB) and health literacy concepts are intrapersonal variables associated with SSB intake. TPB constructs include behavioural intention, affective and instrumental attitudes, subjective norms and perceived behavioural control(Reference Azjen23). Health literacy concepts are often referred to as a general skill set (i.e., the ability to obtain, find, understand and use information to make health decisions) (Reference Parker, Williams and Weiss24Reference Chen, Porter and Estabrooks26). On the contrary, health literacy concepts are sometimes considered in specific domains (i.e., media literacy as the ability to access, analyse, process and produce media messages; public health literacy as the ability to obtain, interpret and act on information needed to make decisions that benefit the health of a community) (Reference Chen, Porter and Estabrooks26,Reference Freedman, Bess and Tucker27) . Collectively, these concepts refer to an adolescent’s ability to internally process and understand factors contributing to their own health outcomes and the health of their community(Reference Parker, Williams and Weiss24Reference Freedman, Bess and Tucker27).

In other single-level studies on either interpersonal or environmental factors, evidence shows caregiver behaviours, caregiver practice and home availability are important factors influencing adolescent’s SSB intake(Reference Haughton, Waring and Wang28Reference Fleary and Ettienne30). In addition, a recent observational and cross-sectional study applied an SEM approach to explore factors contributing to SSB intake among a nationally representative sample of US adolescents(Reference Yuhas, Porter and Hedrick16). Relative to intrapersonal and other social factors examined, caregiver practices and home availability showed the strongest influence on adolescent SSB behaviours, since these factors provide social support and behaviour modelling, as caregivers are usually the gatekeeper of food and beverages within the home and set an example for their adolescent about what are healthy or unhealthy choices and help develop the adolescent’s perceptions related to SSB(Reference Townsend and Foster13,Reference Yuhas, Porter and Hedrick16,Reference Riebl, MacDougal and Hill19,Reference Haughton, Waring and Wang28,Reference Fleary and Ettienne30Reference van de Gaar, van Grieken and Jansen35) . Among adolescents, home SSB environment and availability is one of the most influential predictors of food and beverage choices, and SSB behaviours(Reference Yee, Lwin and Ho36,Reference Ezendam, Evans and Stigler37) . Adolescents with home access to SSB are twice as likely to be moderate consumers of SSB and five times more likely to be high consumers of SSB(Reference Hebden, Hector and Hardy38). Furthermore, about 55–70 % of all SSB consumed by adolescents were consumed in the adolescent’s home(Reference Hafekost, Mitrou and Lawrence39,Reference Wang, Bleich and Gortmaker40) .

However, current literature is limited regarding how multiple levels of the SEM concurrently influence an adolescent’s consumption of SSB. Understanding these impacts within health disparate regions with excessive SSB intake, like rural Appalachia, is necessary to inform design and implementation of health promotion programmes. This cross-sectional study addresses gaps in literature by targeting a regional sample of Appalachian adolescents with the primary aim of identifying SEM factors associated with adolescents’ SSB intake, while controlling for relevant demographic factors. SEM levels targeted in the current study included intrapersonal factors (i.e., attitudes, subjective norms, perceived behavioural control, behavioural intentions, health literacy, media literacy, public health literacy), interpersonal factors (i.e., caregiver’s SSB behaviours, caregiver’s SSB rules) and environment (i.e., home SSB availability).

Methodology

Study design

This cross-sectional study is a secondary analysis of data from the Kids SIPsmartER trial targeting Appalachian middle school students and their caregivers. The Kids SIPsmartER intervention is a school-based, behaviour and health literacy programme aimed at improving SSB behaviours among seventh grade middle school students and engages caregivers in SSB role modelling and supporting home SSB environment changes. Kids SIPsmartER is grounded by TPB constructs and health literacy, media literacy, numeracy and public health literacy concepts. Evaluation of effectiveness and implementation of the multi-level Kids SIPsmartER intervention across twelve Appalachian middle schools through a type 1 hybrid design and cluster randomised controlled trial is on-going (Clincialtrials.gov: NCT03740113; 2018–2022) (Reference Zoellner, Porter and You41,Reference Curran, Bauer and Mittman42) . The current study utilises baseline data from eight middle schools enrolled in the first 2 years of the study.

Study setting

For this cross-sectional analysis, four Appalachian counties are represented. As indicated by scores on the United States Department of Agriculture Rural-Urban Continuum Code (1 = urban, 9 = rural), the included counties are mostly rural: Buchannan = 9, Smyth = 7, Tazewell = 5 and Wythe 6(43). According to school-level data, over 90 % of all adolescents meet on-time graduation rates, despite chronic absenteeism reaching between 11 and 22 %(44). Finally, state assessments in these counties for reading range from 72 to 86 %, math range from 79 to 91 % and science range from 77 to 90 %(44).

Eligibility and recruitment

To be eligible for inclusion, schools had to be in the geographical Appalachian region, have approximately 80–200 adolescents enrolled in seventh grade and have eighth grade within the same school building as seventh grade(Reference Zoellner, Porter and You41). Within each school, all seventh grade adolescents were eligible to participate. Specific to this secondary analysis, only students with complete data were included.

At each school, an informational letter signed by the school’s principal, a study flyer and consent form were sent home to the adolescent’s caregivers(Reference Zoellner, Porter and You41). Additional strategies at some schools included members of the research team attending ‘Back-to-School Nights’, addressing individual classes within each school about the programme, or personalised phone calls to caregivers to obtain verbal consent or remind caregivers to return the consent forms(Reference Zoellner, Porter and You41). Recruitment efforts for caregivers and adolescents were customised to the needs of each school and based on a combination of strategies to increase response rates. Adolescent participation in the current study required provision of both parental consent and adolescent assent. Teachers assisted with distributing, collecting and following up with adolescents for caregiver consent forms. Adolescents who returned the signed consent form (permission granted or denied) received a nominal prize (e.g., highlighter). The assent procedure was conducted immediately before data collection. An assent statement was read with adolescents and all questions were answered before signature was obtained(Reference Zoellner, Porter and You41).

Adolescents were administered the validated survey instrument at baseline during one class period (approximately 45 min). One researcher read each question and answer option aloud to the entire class, while 1–3 additional research staff answered individual student questions (staffing dependent on size and structure of classroom).

Measures

Demographic characteristics included variables of gender and age. Scaled response options are illustrated in Tables 1 and 2. All anchors for the Likert scales used can be found in Table 3.

Table 1 Bivariate associations between intrapersonal level variables and adolescent sugar-sweetened beverage (SSB) intake (n 793)

a,bOne-way ANOVA tests were used to assess if there were any significant differences between the means of the categories. Post-hoc analyses were done using the Tukey method. Values without the same superscript letter are significantly different (P < 0·05).

cFor each of the five SSB questions (i.e., regular soft drinks, sweetened juice beverage/drink, sweetened tea, coffee with sugar, energy drinks), adolescents report consumption frequency across seven response categories (ranging from never or < 1 time per week to three or more times per day) and reported portion sizes across six response categories (ranging from 6 ounces or less to greater than 20 ounces). Total SSB was derived using standardised and validated scoring procedures.

Table 2 Bivariate associations between interpersonal and environmental level variables and adolescent sugar-sweetened beverage (SSB) intake (n 793)

a,bOne-way ANOVA tests were used to assess if there were any significant differences between the means of the categories. Post-hoc analyses were done using the Tukey method. Values without the same superscript letter are significantly different (P < 0·05).

cFor each of the five SSB questions (i.e., regular soft drinks, sweetened juice beverage/drink, sweetened tea, coffee with sugar, energy drinks), adolescents report consumption frequency across seven response categories (ranging from never or < 1 time per week to three or more times per day) and reported portion sizes across six response categories (ranging from 6 ounces or less to greater than 20 ounces). Total SSB was derived using standardised and validated scoring procedures.

Table 3 Stepwise regression model to explain adolescent sugar-sweetened beverage (SSB) intake using factors across the socio-ecological model (n 793)

Avg ME: average marginal effects; robust SE: cluster robust standard errors.

*Significant at P < 0·05, **P < 0·01, ***P < 0·001.

Sugar-sweetened behaviours

An adapted version of the validated Beverage Intake Questionnaire was used to assess the dependent variable, adolescent SSB behaviours(Reference Hedrick, Savla and Comber3,Reference Hedrick, Comber and Ferguson45,Reference Hill, MacDougall and Riebl46) . The Beverage Intake Questionnaire-15 focuses on frequency and portion sizes of fifteen beverage categories. To meet the needs of the current study, alcohol items were removed and the three types of milk were consolidated into a single milk category, resulting in a ten-item assessment. Importantly, the five items necessary to compute amounts of SSB (i.e., regular soft drinks, sweetened juice beverage/drink, sweetened tea, coffee with sugar, energy drinks) were not altered. For each SSB question, adolescents report consumption frequency with seven response categories ranging from never or < 1 time per week to 3 or more times per day. Portion sizes were also reported across six response categories ranging from 6 ounces or less to > 20 ounces. Using standardised and validated scoring procedures, daily totals for each type of SSB were totalled by multiplying intake frequency of the SSB category by the portion reported for that SSB category(Reference Hedrick, Savla and Comber3,43,Reference Hedrick, Comber and Ferguson45) . Likewise, the five categories of SSB were summed to obtain total intake of all SSB per day.

Intrapersonal variables

Intrapersonal level variables included TPB constructs, health literacy, media literacy and public health literacy. TPB constructs were assessed on a seven-point Likert scale and included behavioural intention (two items), affective attitude (one item), instrumental attitude (one item), subjective norms (one item) and perceived behavioural control (one item) (Reference Lane, Porter and Hecht10,Reference Zoellner, Estabrooks and Davy12,Reference Riebl, MacDougal and Hill19) .

Health literacy was assessed with the six-item Newest Vital Sign. Using validated procedures, scores were categorised on total score values: 0–1 indicates high likelihood of limited health literacy, 2–3 indicates the possibility of limited health literacy and 4–6 designates adequate health literacy(Reference Weiss, Mays and Martz47Reference Linnebur and Linnebur49). Media literacy was assessed as the mean of six items measured on a seven-point Likert scale (Cronbach’s α = 0·59)(Reference Lane, Porter and Hecht10,Reference Chen, Porter and Estabrooks26) . Public health literacy was measured as the mean of four items assessed on a five-point Likert scale (Cronbach’s α = 0·68)(Reference Lane, Porter and Hecht10,Reference Chen, Porter and Estabrooks26,Reference Rogers, Fine and Handley50) .

Interpersonal variables

Interpersonal level variables were measured on a five-point Likert scale and included one item on caregiver’s SSB behaviours and ten items on caregiver’s rules around adolescent SSB consumption (Cronbach’s α = 0·66) (Reference Bogart, Elliott and Ober34,Reference Nebeling, Hennessy and Oh51) .

Environmental variable

Reported on a five-point Likert scale, environmental level variables assessed home availability of SSB with the same five SSB item reported for total SSB adolescent consumption (i.e., regular soft drinks, sweetened juice beverage/drink, sweetened tea, coffee with sugar, energy drinks) (Cronbach’s α = 0·51)(Reference Bogart, Elliott and Ober34,Reference Nebeling, Hennessy and Oh51) .

Analyses

SPSS version 26.0 was used for summary statistics and ANOVA analyses, while Stata 16.0 was used for regression analysis. For independent variable measures including more than two questions, Cronbach’s α values were conducted to determine internal consistency of scales. Although several constructs in our model were measured with multiple questions, there are no consistent methods in the literature in terms of how to aggregate variables into one single index or multiple ones. To serve our purpose of identifying the impacts of categories of SEM, we used the mean scores across multiple questions as the level of those constructs with more than one variable information collected. The procedure of averaging those Likert scale variables resulted in continuous-like values for those constructs and was treated as continuous variables in the interpretation of regression model results.

In the ANOVA analysis, we rounded the mean construct level values into the nearest appropriate Likert category to create Likert scale at the construct level (e.g., on a 5-point Likert scale 3–3·99 = neither agree nor disagree, 4–4·99 = somewhat agree, etc.). ANOVA were completed to identify bivariate associations between SSB consumption within our adolescent sample and demographic characteristics, intrapersonal, interpersonal, environmental and exploratory variables. Tukey’s test of significance level at P ≤ 0·05 determined statistically significant post hoc relationships. We conducted ANOVA using the full five- or seven-point Likert scale range for each variable, and by consistently collapsing into three category response options. Overall statistical and post hoc interpretations were remarkably similar, so the reduced option is presented.

Gender, age, intrapersonal, interpersonal and environmental variables were entered into a modified two-part model in a stepwise fashion(Reference Mullahy52,Reference Buntin and Zaslavsky53) . Our sample contains modest numbers of zero SSB consumption reported by adolescents. These ‘zeros’ are true zero instead of missing or censoring, and they cause sizeable skewness to the SSB outcome distribution. Therefore, we choose the modified two-part model that generalises the Tobit model to analyse those data with true zeros and is more robust to distribution assumptions(Reference Buntin and Zaslavsky53,Reference Stewart54) . The two-part models we estimated contain the first part that handles the nonlinear process of generating consume v. not-consume decisions as probit model and the second part that estimates the nonzero continuous SSB consumption via a log-link generalised model and the se are adjusted to be school-year cluster robust. To ensure comparability across models, we restricted the sample size to be the same across all models (i.e., the smallest set of observations that have non-missing values across all the variables in the largest model specification in step 4). Following the conceptual framework of Kids SIPsmartER(Reference Zoellner, Porter and You41), independent variables were included based on their hypothesised proximal influence on SSB consumption (i.e., behavioural intention most proximal, home environment most distal): step 1: demographic factors, step 2: intrapersonal variables, step 3: interpersonal variables and step 4: environmental variables.

Results

Participants

Within the eight schools, 1360 seventh grade adolescents were eligible to participate. After caregiver consent and adolescent assent processes, 874 adolescents were consented, 862 (63 %) adolescents completed the baseline survey and 793 (58 %) were included in the current study. The seventy-two adolescents not included in the analysis had missing data for the variables evaluated. Gender distribution was relatively equal with 55·4 % females and 44·6 % males. Age distribution of adolescents included 3·9 % aged 11 years, 80·5 % aged 12 years and 15·7 % aged 13 or older.

Bivariate association between adolescent sugar-sweetened behaviour intake and socio-ecological model factors

Across all students, SSB intake was a mean of 36·3 (sd = 42·5) ounces and 433·4 (sd = 493·6) calories per day. Compared with reported SSB ounces per day among females (32·2, sd = 34·6), males consumed significantly more SSB ounces per day (41·5, sd = 50·3) (P = 0·002). Reported SSB ounces per day increased with each year older from 11 years of age (32·9, sd = 29·6), 12 years of age (34·3, sd = 40·1) to 13 years of age or older (47·6, sd = 54·2) (P = 0·006).

As illustrated in Tables 1 and 2, all intrapersonal, interpersonal and environmental factors were associated with SSB intake in the direction hypothesised (all P < 0·05). For example, higher affective attitudes, instrumental attitudes, subjective norms, perceived behavioural control, behavioural intentions, health literacy, media literacy and public health literacy were associated with less reported SSB ounces per day (all P < 0·05). Similarly, as adolescents reported stronger agreement with caregiver SSB rules and more positive caregiver SSB behaviour, the reported SSB ounces per day was lower (both P < 0·01). Finally, as adolescents reported more prominent availability of SSB within their home environment, the reported SSB ounces per day was higher (P < 0·001).

Stepwise regression results

Table 3 presents the associated average marginal effects and cluster robust se of those marginal effects for variables in each step of the stepwise modelling process. In the final step, seven variables show statistically significant contribution in explaining adolescent’s SSB consumption: behavioural intention, affective attitude, perceived behavioural control, health literacy, caregiver behaviours, caregiver rules and home availability. For each 1-unit increase in behavioural intentions and affective attitudes, adolescents had a mean reduction in SSB consumption at the rate of about 1·5 ounces and 2·1 ounces per day, respectively (P < 0·05). Perceived behavioural control was also statistically significant at 5 % significance level and is similarly interpreted: adolescents had a mean SSB intake reduction by about 1·2 ounces per day for each 1-unit increase in perceived behavioural control. For every 1-unit increase in health literacy, adolescents showed a mean of 4·0-ounce decrease in SSB consumption per day (P < 0·001). Related to interpersonal and environmental factors, all three variables showed statistically significant influences in adolescents’ SSB daily consumption. For each 1-unit increase in caregiver behaviours and caregiver rules, there was a mean of 3·4-ounce and 4·6-ounce decrease in SSB intake per day (P < 0·05), respectively. Related to the environmental level variable, for each 1-unit increase in home availability there is a mean of 15·1-ounce increase in adolescents’ SSB consumption per day (P < 0·001). This large magnitude of the net influence, as compared with other constructs, signals the clinical significance of home availability.

Steps 1 through 3 of the regression models also reveal notable findings. For demographic factors, gender and age were both significant in step 1. More specifically, male adolescents on average consumed 6·9 more ounces of SSB per day relative to female adolescents and age accounted for an additional 9·3 ounces of SSB for each 1-unit increase of age in years. However, gender was no longer significant after including intrapersonal factors. Age remained significant when including both intrapersonal and interpersonal factors, but was no longer significant when environmental factors were included. For TPB constructs, behavioural intentions and affective attitudes were significant in each step, yet the marginal effects decreased with the addition of interpersonal and environmental factors which is to be expected since inclusion of those additional factors resulting in the net effect of TPB constructs (i.e., netting the influence of interpersonal and environmental factors that overlapped with TPB). Instrumental attitudes and subjective norms were not significant in any steps, while perceived behavioural control was only significant in the final step 4 of the model. Health literacy was significant across each step, yet neither media literacy nor public health literacy significantly contributed to the model. Caregiver behaviours and caregiver rules related to SSB were significant in step 3 and remained significant, although the marginal effects somewhat decreased, with the addition of the environmental variable in step 4.

Discussion

Our study fills an important gap in demonstrating how multiple levels of SEM influence rural Appalachia adolescents who are disproportionately burdened with numerous health disparities impacted by excessive SSB intake(Reference Hales, Carroll and Fryar55,Reference Fleming and Afful56) . Our study showed rural Appalachia adolescents consumed a mean of 439 calories per day, which is similar to other research in this region(Reference Rosinger, Herrick and Gahche6,Reference Lane, Porter and Hecht10,Reference Lane, Porter and Hecht11) . This is excessively higher than recommended daily added sugar intake of < 10 % of daily caloric intake and over 300 % higher than national mean adolescent SSB intake(Reference Rosinger, Herrick and Gahche6,Reference Lane, Porter and Hecht10,Reference Lane, Porter and Hecht11) . When including all SEM levels within our stepwise regression to explain adolescent’s SSB intake, the marginal effects and level of significance were strongest from home availability, followed by caregiver rules, health literacy, caregiver behaviours, affective attitudes, behavioural intentions and behavioural control.

Within other literature, home availability of SSB has shown to be one of the strongest factors influencing adolescent SSB intake(Reference Lane, Porter and Estabrooks31,Reference Yee, Lwin and Ho36,Reference Hebden, Hector and Hardy38Reference Wang, Bleich and Gortmaker40,Reference Watts, Miller and Larson57,Reference Ortega-Avila, Papadaki and Jago58) . Similar to our findings, studies have shown significant associations with increased adolescent SSB intake when SSB are more readily available in the home; however, few have evaluated home availability of SSB in conjunction with other variables (e.g., interpersonal, intrapersonal and macro-level factors) (Reference Haughton, Waring and Wang28,Reference van de Gaar, van Grieken and Jansen35,Reference Hebden, Hector and Hardy38,Reference Wang, Bleich and Gortmaker40) . Another qualitative study among adolescents highlights the important influence of home SSB availability on adolescent recognition of SSB intake at home and norms around availability in the home(Reference Ortega-Avila, Papadaki and Jago58). However, a large portion of the intervention literature does not address the influence of home SSB availability. For example, one systematic review evaluated fifty-five interventions targeting child and adolescent SSB intake and found only four studies addressed the adolescent’s home environment(Reference van de Gaar, van Grieken and Jansen35). Collectively, these findings suggest the importance, yet limited intervention approaches, of targeting home environment with efforts focused on improving adolescent SSB behaviours.

Similar to home environment, interpersonal level factors that included caregiver rules and caregiver behaviours also explained adolescent SSB intake. Our finding supports prior literature that demonstrates the important association among caregiver behaviours and adolescent SSB intake(Reference van der Horst, Kremers and Ferreira29,Reference Lane, Porter and Estabrooks31Reference Bogart, Elliott and Ober34,Reference Yee, Lwin and Ho36) . Although fewer studies have evaluated caregiver rules, some showed caregivers who take a role in active guidance could help reduce adolescent SSB intake(Reference Lane, Porter and Estabrooks31,Reference Yee, Lwin and Ho36) . Other studies further illustrate how additional caregiver traits, such as education level and attitudes towards SSB, can influence adolescent SSB behaviours(Reference Lane, Porter and Estabrooks31,Reference Pettigrew, Jongenelis and Chapman33,Reference Yee, Lwin and Ho36) . While published findings related to the important role of caregiver rules and behaviours on adolescent SSB intake are consistent, most research in this area is cross-sectional. There is limited evidence of interventions that have successfully targeted caregivers’ rules and behaviours to help improve their adolescents SSB intake. Given limited evidence of interventions having successfully targeted parents’ rules and behaviours to help improve their adolescent’s SSB intake, this is an important focus for future research.

While our findings suggest environmental and interpersonal factors strongly predict adolescent SSB intake within rural Appalachia, four intrapersonal factors also significantly contributed to the final model. Three of these intrapersonal factors were related to TPB: behavioural intention, affective attitudes and perceived behavioural control. The important influence of behavioural intention and perceived behavioural control on adolescent SSB behaviour in our study is consistent with previous TPB literature(Reference Lane, Porter and Hecht10,Reference Zoellner, Estabrooks and Davy12,Reference Riebl, MacDougal and Hill19,Reference Riebl, Estabrooks and Dunsmore20,Reference Hackman and Knowlden59,Reference Branscum and Sharma60) . Interestingly, the TPB construct with the greatest influence on adolescent SSB intake in our study was affective attitude (i.e., emotional reaction to the outcome of a behaviour) (Reference Azjen23). Although subjective norms were significant in the ANOVA analysis, it was not significant in the regression model when accounting for all other constructs. This suggests that the perceived influence of the friends on adolescents SSB behaviours is not as important as other included factors. The fourth significant intrapersonal factor in the final model was health literacy. This finding is reflective of existing literature that suggests health literacy is one of the strongest intrapersonal predictors of health status and outcomes in adults(Reference Parker, Williams and Weiss24). While less frequently studied in adolescents, our study affirms findings from a previous study suggestion that lower health literacy scores were associated with higher SSB intake(Reference Park, Eckert and Zaso61). While neither media literacy nor public health literacy contributed to the final model, we believe they warrant further investigation in future studies to understand their impact on SSB intake. These factors were individually associated with SSB intake in the ANOVA tests and reflect factors known to influence SSB intake and other health behaviours (i.e., marketing exposure, perceptions of supporting the health of their community) (Reference Yuhas, Porter and Hedrick16,Reference Rogers, Fine and Handley50,Reference Chen62Reference Bergsma and Carney65) .

In step 1 of our regression model, when only considering demographics, males consumed significantly higher amounts of SSB relative to females and older adolescents consumed significantly more SSB than their younger counterparts. These gender and age findings are similar to existing literature(Reference Rosinger, Herrick and Gahche6). Yet these demographic factors no longer remained significant when adding higher-level SEM variables of influences. Collectively, our findings underscore the importance of each SEM level to understand adolescent SSB intake and in the design and evaluation of interventions targeting adolescent SSB intake.

Limitations and strengths

Several study limitations should be considered. First, all variables within our study were self-reported by the adolescents and could be subject to self-report bias. Similarly, responses related to caregiver SSB rules and SSB behaviours were the adolescent’s perceptions of their caregivers. There may be discrepancies with the perceived caregiver actions as reported by adolescents v. actual caregiver actions. Second, student demographic data related to race and ethnicity were collected but not included in our analysis due to inconsistencies between self-reported and census data for southwest Virginia region of Appalachia. It was concluded that questions related to race and ethnicity were unfamiliar knowledge to most adolescents, leading to inaccurate report. Yet, given limited diversity in this region (95·1 % White and 98·8 % non-Hispanic), the implications of not including race and ethnicity variables are postulated to have little impact on model interpretations or study conclusions(44). Third, since our study was cross-sectional in design, cause and effect cannot be determined. Fourth, Cronbach’s α values for media literacy and public health literacy were on the low end of satisfactory for internal consistency and should be interpreted somewhat cautiously.(Reference Taber66) Finally, due to targeted rural status and unique cultural norms within southwest Virginia related to SSB intake, the current study may lack generalisability to other regions(Reference Lane, Porter and Hecht10,Reference Lane, Porter and Hecht11) . These limitations should be considered within the study strengths, including strong theoretical approach, use of previously validated questionnaires, standardisation in survey administration and an adequate sample across the eight schools and four counties in the rural, health disparate Appalachia region.

Implications for future research

Based on findings from our cross-sectional study, future research should include more robust intervention approaches focused on a multi-level SEM approach to improve SSB intake for adolescents with an emphasis on home availability, caregiver influences and personal influences (i.e., behavioural intentions, affective attitudes, perceived behavioural control and health literacy). Importantly, the on-going Kids SIPsmartER intervention trial is filling this gap by focusing on these levels of influence with a school-based curriculum targeting SSB behaviours among seventh grade middle school students and an integrated text messaging programme targeting caregivers(Reference Zoellner, Porter and You41). The 6-month, 12-session intervention curriculum is guided by TPB constructs and health literacy concepts and applies numerous evidence-based behavioural change techniques to target adolescents SSB behaviour change. For caregivers who consent to participate in the text message component, the content aligns with the adolescents’ curriculum, with assessments every 5–6 weeks, in which caregivers select personal, interpersonal or environmental barriers and receive targeted strategy messages focused on these factors. These strategies offer tips and techniques guiding caregivers to decrease their own SSB intake, as well as improve SSB-related caregiver practices, rules and home environment. Future findings from the multi-level Kids SIPsmartER intervention trial will help to identify causal factors influencing high SSB intake among rural Appalachian adolescents, as well as verify the findings from this cross-sectional analysis.

Conclusion

Our study identifies home environment as the strongest predictor for adolescent SSB intake within rural Appalachia across all levels of the SEM. There is also strong evidence of the influence caregiver rules and behaviours have on adolescent SSB intake, and various other intrapersonal factors including adolescent health literacy and behavioural intentions. Thus, our study adds to the literature by identifying home environment and caregiver rules and behaviours as the most influential factors to adolescent SSB intake. The use of these findings can help develop further research in the area of multi-level health interventions that help reduce SSB behaviours for adolescents, especially those targeting home environment and caregivers. Given the negative influence of SSB on adolescent health, especially in rural Appalachia, but also nationally, there is need for further understanding about various levels of influence so effective interventions can be developed and implemented.

Acknowledgements

Acknowledgements: We would like to acknowledge the superintendents, middle school principals and teachers for their time and dedication in this research. Likewise, the authors thank all the students who have participated. The authors also appreciate the insights from Dr. Phillip Chow, Dr. Lee Ritterband and Dr. Deborah Tate for their role in conceptualising the larger Kids SIPsmartER framework and trial. Finally, we acknowledge the data collection and management support of Donna Brock, Jacob Nottingham and Katherine Eisner. Financial support: The current study was funded by the National Institute of Health (NIH), National Institute on Minority Health and Health Disparities (R01MD012603). NIH was not involved in the design of the current study or writing of this manuscript. Conflict of interest: None. Authorship: B.A.M. drafted the initial manuscript. Principal Investigator J.M.Z. and co-investigators K.J.P. and W.Y. obtained funding for the study and also helped drafted sections of the manuscripts. M.Y., A.L.R. and E.J.T. provided manuscript revisions and all authors approve and take responsibility for the final manuscript. B.A.M. led data entry and management and B.A.M. and W.Y. conducted all analyses. J.M.Z., K.J.P., M.Y. and A.L.R. led survey development. K.J.P., M.Y., A.L.R. and E.J.T. assisted in sample recruitment and data collection. Ethics of human subject participation: The current study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving research study participants were approved by the University of Virginia’s Institutional Review Board for the Social and Behavioral Sciences (IRB-SBS). Written or verbal informed consent was obtained from all caregivers and assent was obtained from all student participants. Verbal consent was witnessed and formally recorded.

References

Lake, AA, Mathers, JC, Rugg-Gunn, AJ et al. (2006) Longitudinal change in food habits between adolescence (11–12 years) and adulthood (32–33 years): the ASH30 Study. J Public Health 28, 1016.CrossRefGoogle ScholarPubMed
Kim, SY, Sim, S, Park, B et al. (2016) Dietary habits are associated with school performance in adolescents. Medicine 95, e3096.CrossRefGoogle ScholarPubMed
Hedrick, VE, Savla, J, Comber, DL et al. (2012) Development of a brief questionnaire to assess habitual beverage intake (BEVQ-15): sugar-sweetened beverages and total beverage energy intake. J Acad Nutr Diet 112, 840849.CrossRefGoogle ScholarPubMed
Malik, VS, Popkin, BM, Bray, GA et al. (2010) Sugar-sweetened beverages, obesity, type 2 diabetes mellitus, and cardiovascular disease risk. Circulation 121, 13561364.CrossRefGoogle ScholarPubMed
Tahmassebi, JF, Duggal, MS, Malik-Kotru, G et al. (2006) Soft drinks and dental health: a review of the current literature. J Dent 34, 211.CrossRefGoogle ScholarPubMed
Rosinger, A, Herrick, K, Gahche, J et al. (2017) Sugar-sweetened beverage consumption among U.S. youth, 2011–2014. NCHS Data Brief 271, 18.Google Scholar
Park, S, Blanck, HM, Sherry, B et al. (2012) Factors associated with sugar-sweetened beverage intake among United States high school students. J Nutr 142, 306312.CrossRefGoogle ScholarPubMed
Gutschall, M, Thompson, K & Lawrence, E (2017) Addressing health disparities in rural nutrition practice: a qualitative model from rural Appalachia. J Hunger Environ Nutr 13, 8499.CrossRefGoogle Scholar
Southerland, J, Dula, T & Slawson, D (2019) Barriers to healthy eating among high school youth in rural Southern Appalachia. J Appalach Health 1, 3143.Google Scholar
Lane, H, Porter, KJ, Hecht, E et al. (2018) Kids SIP smartER: a feasibility study to reduce sugar-sweetened beverage consumption among middle school youth in central Appalachia. Am J Health Promot 32, 13861401.CrossRefGoogle ScholarPubMed
Lane, H, Porter, KJ, Hecht, E et al. (2016) Healthy Hurley: a randomized controlled feasibility study to reduce sugar-sweetened beverage consumption among middle school youth in Central Appalachia. In APHA 2016 Annual Meeting & Expo: APHA. https://apha.confex.com/apha/144am/meetingapp.cgi/Paper/356416 (accessed May 2020).Google Scholar
Zoellner, J, Estabrooks, PA, Davy, BM et al. (2012) Exploring the theory of planned behavior to explain sugar-sweetened beverage consumption. J Nutr Educ Behav 44, 172177.CrossRefGoogle ScholarPubMed
Townsend, N & Foster, C (2013) Developing and applying a socio-ecological model to the promotion of healthy eating in the school. Public Health Nutr 16, 11011108.CrossRefGoogle ScholarPubMed
McLeroy, KR, Bibeau, D, Steckler, A et al. (1988) An ecological perspective on health promotion programs. Health Educ Q 15, 351377.CrossRefGoogle ScholarPubMed
King, AC, Stokols, D, Talen, E et al. (2002) Theoretical approaches to the promotion of physical activity: forging a transdisciplinary paradigm. Am J Prev Med 23, 1525.CrossRefGoogle ScholarPubMed
Yuhas, M, Porter, KJ, Hedrick, V et al. (2020) Using a socioecological approach to identify factors associated with adolescent sugar-sweetened beverage intake. J Acad Nutr Diet 120, 15571567.CrossRefGoogle ScholarPubMed
Kassem, NO, Lee, JW, Modeste, NN et al. (2003) Understanding soft drink consumption among female adolescents using the Theory of Planned Behavior. Health Educ Res 18, 278291.CrossRefGoogle ScholarPubMed
Kassem, NO & Lee, JW (2004) Understanding soft drink consumption among male adolescents using the theory of planned behavior. J Behav Med 27, 273296.CrossRefGoogle ScholarPubMed
Riebl, SK, MacDougal, C, Hill, C et al. (2016) Beverage choices of adolescents and their parents using the theory of planned behavior: a mixed methods analysis. J Acad Nutr Diet 116, 226239.e221.CrossRefGoogle ScholarPubMed
Riebl, SK, Estabrooks, PA, Dunsmore, JC et al. (2015) A systematic literature review and meta-analysis: the Theory of Planned Behavior’s application to understand and predict nutrition-related behaviors in youth. Eat Behav 18, 160178.CrossRefGoogle ScholarPubMed
Cervi, MM, Agurs-Collins, T, Dwyer, LA et al. (2017) Susceptibility to food advertisements and sugar-sweetened beverage intake in non-Hispanic black and non-Hispanic white adolescents. J Commun Health 42, 748756.CrossRefGoogle ScholarPubMed
Kumar, G, Onufrak, S, Zytnick, D et al. (2015) Self-reported advertising exposure to sugar-sweetened beverages among US youth. Public Health Nutr 18, 11731179.CrossRefGoogle ScholarPubMed
Azjen, I (1991) The theory of planned behavior. Organ Behav Hum Decis Process 50, 179211.Google Scholar
Parker, RM, Williams, MV & Weiss, BD (1999) Health literacy: report of the Council on Scientific Affairs. Ad Hoc Committee on Health Literacy for the Council on Scientific Affairs, American Medical Association. JAMA 281, 552557.Google Scholar
Centers for Disease Control and Prevention (2019) What is Health Literacy? https://www.cdc.gov/healthliteracy/learn/index.html (accessed April 2020).Google Scholar
Chen, Y, Porter, KJ, Estabrooks, PA et al. (2017) Development and evaluation of the Sugar-Sweetened Beverages Media Literacy (SSB-ML) scale and its relationship with SSB consumption. Health Commun 32, 13101317.CrossRefGoogle ScholarPubMed
Freedman, DA, Bess, KD, Tucker, HA et al. (2009) Public health literacy defined. Am J Prev Med 36, 446451.CrossRefGoogle ScholarPubMed
Haughton, CF, Waring, ME, Wang, ML et al. (2018) Home matters: adolescents drink more sugar-sweetened beverages when available at home. J Pediatr 202, 121128.CrossRefGoogle ScholarPubMed
van der Horst, K, Kremers, S, Ferreira, I et al. (2007) Perceived parenting style and practices and the consumption of sugar-sweetened beverages by adolescents. Health Educ Res 22, 295304.CrossRefGoogle ScholarPubMed
Fleary, SA & Ettienne, R (2019) The relationship between food parenting practices, parental diet and their adolescents’ diet. Appetite 135, 7985.CrossRefGoogle ScholarPubMed
Lane, H, Porter, K, Estabrooks, P et al. (2016) A systematic review to assess sugar-sweetened beverage interventions for children and adolescents across the socioecological model. J Acad Nutr Diet 116, 12951307.e1296.CrossRefGoogle ScholarPubMed
Yee, AZH, Lwin, MO & Lau, J (2019) Parental guidance and children’s healthy food consumption: integrating the theory of planned behavior with interpersonal communication antecedents. J Health Commun 24, 183194.CrossRefGoogle ScholarPubMed
Pettigrew, S, Jongenelis, M, Chapman, K et al. (2015) Factors influencing the frequency of children’s consumption of soft drinks. Appetite 91, 393398.CrossRefGoogle ScholarPubMed
Bogart, LM, Elliott, MN, Ober, AJ et al. (2017) Home sweet home: parent and home environmental factors in adolescent consumption of sugar-sweetened beverages. Acad Pediatr 17, 529536.CrossRefGoogle ScholarPubMed
van de Gaar, VM, van Grieken, A, Jansen, W et al. (2017) Children’s sugar-sweetened beverages consumption: associations with family and home-related factors, differences within ethnic groups explored. BMC Public Health 17, 195.CrossRefGoogle ScholarPubMed
Yee, AZ, Lwin, MO & Ho, SS (2017) The influence of parental practices on child promotive and preventive food consumption behaviors: a systematic review and meta-analysis. Int J Behav Nutr Phys Act 14, 47.CrossRefGoogle ScholarPubMed
Ezendam, NP, Evans, AE, Stigler, MH et al. (2010) Cognitive and home environmental predictors of change in sugar-sweetened beverage consumption among adolescents. Br J Nutr 103, 768774.CrossRefGoogle ScholarPubMed
Hebden, L, Hector, D, Hardy, LL et al. (2013) A fizzy environment: availability and consumption of sugar-sweetened beverages among school students. Prev Med 56, 416418.CrossRefGoogle ScholarPubMed
Hafekost, K, Mitrou, F, Lawrence, D et al. (2011) Sugar sweetened beverage consumption by Australian children: implications for public health strategy. BMC Public Health 11, 950.CrossRefGoogle ScholarPubMed
Wang, YC, Bleich, SN & Gortmaker, SL (2008) Increasing caloric contribution from sugar-sweetened beverages and 100 % fruit juices among US children and adolescents, 1988–2004. Pediatr 121, e16041614.CrossRefGoogle Scholar
Zoellner, JM, Porter, KJ, You, W et al. (2019) Kids SIPsmartER, a cluster randomized controlled trial and multi-level intervention to improve sugar-sweetened beverages behaviors among Appalachian middle-school students: rationale, design & methods. Contemp Clin Trials 83, 6480.CrossRefGoogle ScholarPubMed
Curran, GM, Bauer, M, Mittman, B et al. (2012) Effectiveness-implementation hybrid designs: combining elements of clinical effectiveness and implementation research to enhance public health impact. Med Care 50, 217226.CrossRefGoogle ScholarPubMed
United States Department of Agriculture (2013) Rural-Urban Continuum Codes. https://www.ers.usda.gov/data-products/rural-urban-continuum-codes.aspx (accessed May 2020).Google Scholar
United State Census Bureau (2018) 2014–2018 ACS 5-Year Data Profile. https://www.census.gov/acs/www/data/data-tables-and-tools/data-profiles/ (accessed June 2020).Google Scholar
Hedrick, VE, Comber, DL, Ferguson, KE et al. (2013) A rapid beverage intake questionnaire can detect changes in beverage intake. Eat Behav 14, 9094.CrossRefGoogle ScholarPubMed
Hill, CE, MacDougall, CR, Riebl, SK et al. (2017) Evaluation of the relative validity and test-retest reliability of a 15-item beverage intake questionnaire in children and adolescents. J Acad Nutr Diet 117, 17571766.e1755.CrossRefGoogle ScholarPubMed
Weiss, BD, Mays, MZ, Martz, W et al. (2005) Quick assessment of literacy in primary care: the newest vital sign. Ann Fam Med 3, 514522.CrossRefGoogle ScholarPubMed
Warsh, J, Chari, R, Badaczewski, A et al. (2014) Can the newest vital sign be used to assess health literacy in children and adolescents? Clin Pediatr 53, 141144.CrossRefGoogle ScholarPubMed
Linnebur, LA & Linnebur, SA (2018) Self-administered assessment of health literacy in adolescents using the newest vital sign. Health Promot Pract 19, 119124.CrossRefGoogle ScholarPubMed
Rogers, EA, Fine, S, Handley, MA et al. (2014) Development and early implementation of the bigger picture, a youth-targeted public health literacy campaign to prevent type 2 diabetes. J Health Commun 19, 144160.CrossRefGoogle ScholarPubMed
Nebeling, LC, Hennessy, E, Oh, AY et al. (2017) The FLASHE study: survey development, dyadic perspectives, and participant characteristics. Am J Prev Med 52, 839848.CrossRefGoogle ScholarPubMed
Mullahy, J (1998) Much ado about two: reconsidering retransformation and the two-part model in health econometrics. J Health Econ 17, 247281.CrossRefGoogle ScholarPubMed
Buntin, MB & Zaslavsky, AM (2004) Too much ado about two-part models and transformation? Comparing methods of modeling Medicare expenditures. J Health Econ 23, 525542.CrossRefGoogle ScholarPubMed
Stewart, J (2009) Tobit or not tobit? https://www.bls.gov/osmr/research-papers/2009/pdf/ec090100.pdf (accessed May 2020).Google Scholar
Hales, CM, Carroll, MD, Fryar, CD et al. (2017) Prevalence of obesity among adults and youth: United States, 2015–2016. NCHS Data Brief. https://www.cdc.gov/nchs/data/databriefs/db288.pdf (accessed June 2020).Google Scholar
Fleming, E & Afful, J (2018) Prevalence of total and untreated dental caries among youth: United States, 2015–2016. NCHS Data Brief. https://www.cdc.gov/nchs/data/databriefs/db307.pdf (accessed June 2020).Google Scholar
Watts, AW, Miller, J, Larson, NI et al. (2018) Multicontextual correlates of adolescent sugar-sweetened beverage intake. Eat Behav 30, 4248.CrossRefGoogle ScholarPubMed
Ortega-Avila, AG, Papadaki, A & Jago, R (2018) The role of the home environment in sugar-sweetened beverage intake among northern Mexican adolescents: a qualitative study. J Public Health 27, 791801.CrossRefGoogle Scholar
Hackman, CL & Knowlden, AP (2014) Theory of reasoned action and theory of planned behavior-based dietary interventions in adolescents and young adults: a systematic review. Adolesc Health Med Ther 5, 101114.CrossRefGoogle Scholar
Branscum, P & Sharma, M (2014) Comparing the utility of the theory of planned behavior between boys and girls for predicting snack food consumption: implications for practice. Health Promotion Pract 15, 134140.CrossRefGoogle ScholarPubMed
Park, A, Eckert, TL, Zaso, MJ et al. (2017) Associations between health literacy and health behaviors among urban high school students. J Sch Health 87, 885893.CrossRefGoogle ScholarPubMed
Chen, YC (2013) The effectiveness of different approaches to media literacy in modifying adolescents’ responses to alcohol. J Health Commun 18, 723739.CrossRefGoogle Scholar
Pinkleton, BE, Austin, EW, Chen, Y-CY et al. (2013) Assessing effects of a media literacy-based intervention on us adolescents’ responses to and interpretations of sexual media messages. J Child Media 7, 463479.CrossRefGoogle Scholar
Primack, BA, Douglas, EL, Land, SR et al. (2014) Comparison of media literacy and usual education to prevent tobacco use: a cluster-randomized trial. J Sch Health 84, 106115.CrossRefGoogle ScholarPubMed
Bergsma, LJ & Carney, ME (2008) Effectiveness of health-promoting media literacy education: a systematic review. Health Educ Res 23, 522542.CrossRefGoogle ScholarPubMed
Taber, KS (2017) The use of Cronbach’s alpha when developing and reporting research instruments in science education. Res Sci Educ 48, 12731296.CrossRefGoogle Scholar
Figure 0

Table 1 Bivariate associations between intrapersonal level variables and adolescent sugar-sweetened beverage (SSB) intake (n 793)

Figure 1

Table 2 Bivariate associations between interpersonal and environmental level variables and adolescent sugar-sweetened beverage (SSB) intake (n 793)

Figure 2

Table 3 Stepwise regression model to explain adolescent sugar-sweetened beverage (SSB) intake using factors across the socio-ecological model (n 793)