UBC Faculty Research and Publications

The home food environment and associations with dietary intake among adolescents presenting for a lifestyle… Watts, Allison W; Barr, Susan I; Hanning, Rhona M; Lovato, Chris Y; Mâsse, Louise C Feb 6, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-40795_2018_Article_210.pdf [ 553.42kB ]
JSON: 52383-1.0363455.json
JSON-LD: 52383-1.0363455-ld.json
RDF/XML (Pretty): 52383-1.0363455-rdf.xml
RDF/JSON: 52383-1.0363455-rdf.json
Turtle: 52383-1.0363455-turtle.txt
N-Triples: 52383-1.0363455-rdf-ntriples.txt
Original Record: 52383-1.0363455-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessThe home food environment andassociations with dietary intake amongadolescents presenting for a lifestylemodification interventionAllison W. Watts1,2, Susan I. Barr3, Rhona M. Hanning4, Chris Y. Lovato5 and Louise C. Mâsse6,7*AbstractBackground: The home food environment may be an important target for addressing adolescent obesity. The aimof this study was to investigate associations between aspects of the home food environment and the diets of adolescentswho present for obesity treatment.Methods: Cross-sectional baseline data were collected from 167 overweight/obese adolescent-parent pairs participatingin an e-health lifestyle modification intervention. Adolescent intake of specific foods (fruit and vegetables, totalfat, sugar-sweetened beverages, desserts/treats, and snacking occasions) was assessed by three 24-h dietary recalls, whilehousehold factors were collected from adolescent and parent questionnaires. Structural Equation Modeling, controllingfor relevant covariates, was used to examine the relationship between adolescent diet and the following householdfactors: parent modeling, parenting style, family meal practices, and home food/beverage availability.Results: Findings reveal that few characteristics of the home food environment were associated with adolescentdietary intake. Greater home availability of high-fat foods was moderately associated with adolescent snack intake(β = 0.27, p < .001). Associations with fruit/vegetables and fat intake were small and some were in unexpecteddirections. Parent modeling of healthful food choices and healthier family meal practices were associated withlower availability of high-fat foods and treats in the home, but were not directly associated with adolescent diets.Conclusions: Parent modeling of healthy foods and positive mealtime routines might contribute to thehealthfulness of foods offered in the homes of adolescents who are overweight/obese. Additional research isneeded to better characterize the complex aspects of the household environment that influence adolescent diet.Keywords: Home food environment, Adolescent overweight, Obesity treatment, Dietary intakeBackgroundA healthy diet during adolescence is important for optimalgrowth and for preventing the development of conditionssuch as diabetes, dental carries, and obesity [1]. Currently,adolescents consume too few fruits and vegetables and toomany energy-dense nutrient-poor foods and beverages (e.g.sugary drinks, fast foods, and snacks) [2–5], and severalstudies report that these markers of poor diet quality are as-sociated with obesity [6, 7]. Furthermore, one third ofCanadian and American adolescents are overweight orobese [8, 9]; therefore, promoting a healthier diet is likelyan important strategy for addressing childhood obesity.However, interventions have had limited success in chan-ging adolescent dietary behavior, particularly in the con-text of obesity treatment programs [10].In obesity treatment programs, parents are increasinglyseen as important agents of behavior change because theyare in control of broader aspects of the home, including theavailability of foods and the rules that may support or hin-der their children’s dietary choices [11]. Several models of* Correspondence: lmasse@cfri.ubc.ca6School of Population and Public Health, University of British Columbia, 2206East Mall, Vancouver, British Columbia V6T 1Z9, Canada7Child and Family Research Institute, University of British Columbia, 4480 OakStreet, room F512a, Vancouver, BC V6H 3V4, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Watts et al. BMC Nutrition  (2018) 4:3 https://doi.org/10.1186/s40795-018-0210-6the home food environment informed by social-cognitiveand socio-ecological theories suggest that familial influ-ences including parenting practices and other aspects ofthe home will shape the uptake of healthy dietary behav-iors [12, 13]. Social aspects (e.g., parent role modeling,parenting style, mealtime routines, socio-demographicand economic characteristics), physical aspects (e.g, whatfoods and beverages are available and easily accessible)and the interplay between them have been associated withadolescent diet in previous studies [14–16]. This type ofmodel has been tested in a population-based study examin-ing adolescent fruit and vegetable (FV) intake. In particular,FV intake was influenced by availability in the home, and inturn, availability in the home was influenced by social sup-port for healthful eating, family meal patterns, food securityand socio-economic status (SES) [16].There is limited evidence testing these home environ-ment models for youth who are overweight or obese andseeking treatment. Some evidence comes from results of aparent-centered program focusing on promoting an au-thoritative parenting style, role modeling, and a healthierhome food environment (e.g. availability, accessibility, mealroutines), which found greater reductions in body massindex (BMI) than when children alone are targeted [17].These findings have sparked increased interest in the rolethat parenting and home food environments may play foryouth in obesity treatment programs. Further explorationof these influences on the diets of overweight/obese adoles-cents will inform intervention targets and help to improvethe effectiveness of obesity treatment programs.To build on the existing literature and gain insightsthat are directly relevant to improving adolescent obes-ity treatment programs, the aim of the present studywas to test a structural equation model of associationsbetween the home food environment and dietary intakeamong obesity treatment seeking adolescents. It was hy-pothesized that an authoritative parenting style, parentmodeling of healthful food choices (FV and low-fatsnacks), more healthful family meal practices (fewermeals in front of the TV and at fast food restaurants),reduced availability of less healthful foods and drinks(availability of select high-fat foods or non-diet softdrinks in their home) and high SES (higher education orincome level) would be associated with more healthfuldietary habits among overweight/obese adolescents. Itwas also hypothesized that social influences may indir-ectly influence adolescent dietary intake through associ-ations with availability of less healthful foods in thehome (Fig. 1).MethodsParticipants and proceduresThis study utilizes baseline data collected from adoles-cent participants of an eight-month e-health obesityintervention, which included anthropometric measure-ments, questionnaires (Additional file 1), and three 24-hdietary recall assessments. In addition, one of their par-ents completed a baseline questionnaire on the homefood environment. Participants were recruited fromnewspaper advertisements (62%), invitations sent to pre-vious patients of a Children’s Hospital Endocrinology &Diabetes Unit (13%) and healthy weights clinic (15%),and other sources (e.g., word of mouth) (10%). Eligibleadolescents were 11–16 years old and had BMI z-scoresgreater than one standard deviation from the mean, ac-cording to WHO age-and-gender matched growth charts[18]. Participants had to be residents of the greater Van-couver area with no plans to move within the studyperiod, read at the grade 6 level and speak English. Ex-clusion criteria included comorbidities that required im-mediate medical attention, medical reasons that madephysical activity too difficult, use of medication affectingbody weight, diagnosis of Type 1 diabetes, or participa-tion in another weight-loss program. Of the 183 parent-child pairs who completed the baseline assessment,seven did not meet eligibility requirements (e.g. BMI,reading level), three did not complete any 24-h dietaryrecalls, and six parents did not complete the home en-vironment questionnaire yielding a sample of 167parent-adolescent pairs for the present analyses. Writtenconsent was obtained from all participants and thisstudy was approved by the University of BritishColumbia and the University of Waterloo ethics boards.MeasuresOutcome variableDietary Intake was assessed using a previously validated[19], computer-based 24-h dietary recall program employ-ing a three-pass technique where participants were askedto report all foods/beverages that they consumed theFig. 1 Proposed model of factors within the home foodenvironment and their association with adolescent dietary intake.This conceptual model details the primary processes of interest,however, modeling will also take into account important covariatessuch as child age and sex, maternal education andhousehold incomeWatts et al. BMC Nutrition  (2018) 4:3 Page 2 of 9previous day at breakfast, morning snack, lunch, afternoonsnack, dinner, and evening snack. Over 900 brand or gen-eric food items were available and participants wereinstructed to substitute foods not found (20% of recallshad at least one food item substituted). Photographsdepicting measured portion sizes helped to estimate por-tion sizes and prompts allowed for the selection of top-pings commonly eaten with certain foods (e.g. spreads ontoast). A summary screen allowed participants to confirmor delete their selections. Dietary data were downloadedfrom the web survey and processed with The Food Proces-sor software package (version 8.0, ESHA Research, Salem,OR, 2002) that uses the 2007 Canadian Nutrient File data(http://www.hc-sc.gc.ca/fn-an/nutrition/fiche-nutri-data/index-eng.php) to calculate nutrient and Canadian foodgroup estimates.Of the 167 adolescents examined in the present study,76 provided all three days of dietary recalls, while 46provided two days and 45 provided only one day. No dif-ferences by number of dietary recalls completed werefound except for consumption of desserts/treats, whichwas significantly greater among those who completedmore days of dietary recall (data not shown). Becausefew differences were found, dietary intakes were aver-aged across all available recalls to obtain daily estimatesof: 1) servings of FV, 2) percentage of energy from totalfat (Fat), 3) servings of sugar-sweetened beverages (SSB),4) servings of desserts or treats (Desserts/treats), and 5)percentage of energy from snacking occasions (Snacks).Desserts/treats included food items commonly con-sumed for dessert or as a treat (e.g. cookies, cake, candy,chocolate, ice cream and chips), which are typically en-ergy dense yet nutrient poor. Servings of SSB and des-serts/treats were dichotomized (any vs. none) becausethey had a highly left-skewed distribution.Independent variablesParent Modeling was assessed with five items fromthe adolescent questionnaire: 1) My parents eat vege-tables when I am with them; 2) My parents eat fruitswhen I am with them; 3) My parents eat salad at arestaurant when I am with them; 4) My parents eatlow-fat snacks when I am with them; 5) My parentseat low-fat dressings with salads when I am withthem. Responses to each item were coded on a 4-point scale (Never, Sometimes, Frequently, Always).These items were adapted from Cullen’s 15-item par-ent modeling scale [20], which also included additionalitems specific to particular meal times. Similar items havealso been used to predict diet outcomes in adolescentsamples [21].Parenting Style was assessed with eleven items fromthe parent questionnaire such as wanting to hear aboutmy child’s problems, knowing where my child is afterschool, and telling my child that I like him/her just theway he/she is. Responses to each item were coded on a4-point scale (Never, Sometimes, Often, Always). Theseitems were derived from Cullen’s 11-item authoritativeparenting scale [22].Family Meal Practices was assessed with seven itemsdrawn from the Family Nutrition and Physical ActivityScreening Tool [23], which was completed by parents: 1)eating breakfast together, 2) eating at fast food restau-rants, 3) eating while watching television, 4) eating fruitsand vegetables with meals or as snacks, 5) using pre-packaged foods for meals, 6) eating dessert regularlyafter dinner, and 7) eating dessert regularly in the even-ing. Responses were coded on a 4-point scale so that ahigher score indicated more healthful meal practices.Home Food Availability was assessed with eight itemsfrom the parent questionnaire. Participants were asked ifthe following seven food types were available in the pastweek (yes/no) and if they were low-fat (yes/no): 1) cook-ies, pies, cakes or snack cakes; 2) chips (e.g. potato, corn,tortilla or Doritos chips); 3) ice cream or frozen yogurt;4) granola bars; 5) bacon/sausage; 6) hot dogs; and 7)frozen dinners. Similar to previous studies that summedfood items into the total number of core foods versusnon-core foods available in the home or the number ofenergy-dense snack foods [24, 25], availability items weresplit into two indices and summed to generate: 1) Avail-ability of high-fat foods (bacon/sausage, hot dogs, frozendinners; range = 0–3), and 2) Availability of high-fattreats (cookies/pies/cakes/snack cakes, chips, ice cream/frozen yogurt, and granola bars; range = 0–4). Itemsidentified as low-fat versions were omitted. Availabilityof soft drinks was assessed with the following item: “Didyou have regular sodas or soft drinks in your home inthe past week?” These items were derived from a list of15 items used in the Girls Health Enrichment MultisiteStudy [26, 27]. Similar items have been used to predictdietary intake in adolescent samples [21].CovariatesAdolescent Age and Gender, Parent Ethnicity, MaternalEducation and Household Income were based on parentself-report. Highest degree, certificate, or diploma ofmother was obtained and responses were grouped intothree categories: 1) Less than or equal to high schooleducation; 2) Trade certificate, diploma, non-universitycertificate, or university certificate below a bachelorlevel; and 3) University degree or greater. Total income,before taxes and deductions, of all household membersfrom all sources in the past 12 months was obtained andresponses were collapsed into four categories:1) ≤ $40,000; 2) $40,001–$80,000; 3) $80,001–$120,000;and 4) ≥ $120,000. Body Mass Index z-scores, based onsex and age, were computed from measured height andWatts et al. BMC Nutrition  (2018) 4:3 Page 3 of 9weight using the WHO method for children and adoles-cents (5–19 years old) [18].AnalysisConfirmatory Factor Analysis (CFA) was performed todetermine if scale factor structures were supported inthis sample. Availability of high-fat foods was conceptu-alized as an index and availability of soft drinks wasassessed by only one item; therefore, they were not ex-amined using CFA. Model fit was assessed using com-monly accepted fit indices: Chi-square goodness of fittest (p-value ≥.15), Comparative Fit Index (CFI > .95),Root Mean Square Error of Approximation (RMSEA<.06with an upper CI ≤ .08 and a p-value > .05), and theStandardized Root Mean Square Residual (SRMR<.08)[28]. Since the chi-square test is highly influenced bymodel complexity and sample size, and CFI and SRMRare highly influenced by the inclusion of non-significantpaths, the RMSEA was the main index used to deter-mine model fit [28]. A single model was built with allthree latent constructs and the Maximum Likelihood Es-timator was used. Internal consistency of items in eachscale was determined by computing Cronbach’s alpha.After the measurement models were refined, twostructural equation models tested the conceptual modellinking the home food environment to adolescent dietaryoutcomes: FV, Fat, SSB, Desserts/treats, and Snacks. Forthe analyses, servings of FV were expressed per 1000 kcal(to account for energy intake and to maintain a scalecomparable with the other dietary variables). First, all ofthe independent variables were regressed on each dietaryoutcome to determine direct effects. Second, the inde-pendent variables were regressed on dietary outcomes aswell as on home availability variables. Covariates in-cluded adolescent age, sex, maternal education andhousehold income. The Means- and Variance- adjustedWeighted Least Squares (WLSMV) method of estima-tion was used to handle a combination of continuousand dichotomous outcome variables. WLSMV has beenproposed as the best estimator when categorical data arepresent [29], was designed specifically for use with smalland moderate sample sizes, and is fairly robust to non-normality [30, 31]. Model fit was assessed using the indi-ces described earlier as well as the Weighted Root MeanSquare Residual (WRMR). When using the WLSMV es-timator, the RMSEA and WRMR are the best indices ofmodel fit, with a WRMR of less than 1.0 and a RMSEAof less than 0.6 suggesting a good fit [28].Missing data were handled using pairwise deletion(< 5% missing). All conceptual paths were included inthe model and were considered significant at p-value< 0.05.All statistical analyses were conducted using MPlus®(version 7, Los Angeles, CA).ResultsSample characteristicsThe average age of adolescents was 13 and slightlymore females participated than boys. Families werefairly evenly distributed across household income cat-egories, while twice as many mothers had a universitydegree as compared to a high school degree or less.Families reported having more high-fat treats in thehouse than other high-fat foods and just over onethird reported having non-diet soft drinks in thehouse (Table 1).Measurement modelInitial results did not support the original factor struc-ture of the data; however, after examination of modifica-tion indices, several post-hoc modifications withconceptual relevance were made to produce a measure-ment model that demonstrated good model fit. Retaineditems and fit indices are presented in Table 2.Structural equation modelFirst, a model of direct effects was fit: χ2 (df = 252) = 352,p < .001; RMSEA = .05 [.04–.06], p = .56; CFI = .73; andWRMR= 0.98. No direct associations were seen with au-thoritative parenting, parent modeling, or family mealpractices and dietary outcomes (data not shown). Second,a model with the addition of variables regressed on homefood and beverage availability was fit: χ2 (df = 252) = 339,p < .001; RMSEA = .05 [.03–.06], p = .72; CFI = .76; andWRMR= 0.90 (see significant standardized coefficients inFig. 2 and full solution in Table 3). The findings revealedthat social variables (authoritative parenting, parent mod-eling, and family meal practices) had no direct effect ondietary outcomes; however, several social variables had adirect association with the availability of food and bever-ages in the home, which in turn, had a direct effect ondietary outcomes (Fig. 2). In both models, the CFI and χ2p-value were not within suggested ranges, but the RMSEAand WRMR were. Examination of the modification indicesdid not uncover ways to improve the model and deletionof non-significant paths was not considered, given theconfirmatory nature of our analyses.A small number of variables were associated with ado-lescent dietary outcomes (Fig. 2). As hypothesized, avail-ability of high-fat foods was associated with a greaterpercentage of energy from fat and from snacks. Greateravailability of high-fat treats was associated with lowerFV and unexpectedly with lower fat intake. Despite hy-pothesized associations, no relationships were found forDesserts/Treats and SSB intake or with the availabilityof soft drinks. Among demographic and socio-economicfactors, adolescents from families with higher maternaleducation and with a lower income consumed lowerpercentage of energy from fat. In addition, males hadWatts et al. BMC Nutrition  (2018) 4:3 Page 4 of 9lower odds of reporting SSB consumption (Table 3).Note that analyses with the percentage of energy fromsaturated fat versus total fat yielded similar results.Some hypothesized relationships between factors inthe social environment and the physical environmentof the home were observed (Fig. 2). Healthful parentmodeling and more healthful family meal practiceswere indirectly associated with dietary outcomesthrough home food availability. Adolescents who re-ported that their parents modeled healthful food con-sumption had fewer high-fat treats in their homes.Similarly, families reporting healthier family mealpractices also reported reduced availability of high-fatfoods and high-fat treats. Families with a higher ma-ternal education and higher household income hadlower high-fat food availability.DiscussionFew studies have examined the home food environmentamong adolescents with overweight/obesity. Families ofadolescents who present for obesity treatment may pro-vide valuable insights about what home environmentcharacteristics need to be addressed to improve the ef-fectiveness of these interventions. Results suggest thatlimited aspects of the home food environment are asso-ciated with the diets of treatment seeking adolescents;however, many expected associations were not found. Inparticular, home availability of non-diet soft drinks wasnot associated with decreased consumption of lesshealthful foods or beverages. In addition, more positiveparent modeling and family meal practices were not dir-ectly associated with any dietary outcomes, but were as-sociated with reduced availability of certain less healthfulTable 1 Adolescent and household characteristicsN Mean SD Range n (%)Demographic CharacteristicsAge 167 13.2 1.8 11.0–16.0Sex (Female) 167 89 (53.3)Body Mass Index (BMI zscore)a 167 2.7 0.9 1.1–6.7Weight (kg) 167 83.5 22.9 48.0–175.8Height (m) 167 1.63 0.1 1.4–2.0Maternal Education 167≤ High SchoolTrade Certificate/Diploma≥ University Degree32 (19.2)64 (38.3)71 (42.5)Household Income 167≤ $40,000$40,001–$80,000$80,001–120,000≥ $120,00033 (19.8)54 (32.2)45 (27.0)35 (21.0)Parent Ethnicity (White) 165 77 (46.7)Home Food EnvironmentAvailability of High-Fat Foods (0–3) 167 0.6 0.7 0–3Availability of High-Fat Treats (0–4) 167 1.9 1.2 0–4Availability of Soft Drinks (yes) 167 61 (36.5)Authoritative Parenting (1–4) 159 3.5 0.5 2.1–4.0Parent Modeling (1–4) 162 2.5 0.6 1.0–4.0Family Meal Practices (1–4) 154 2.8 0.7 1.0–4.0Dietary IntakeFruit & Vegetables, servings/d 167 3.4 2.0 0.0–8.8Fat, % kcal/d 167 32.8 8.1 3.4–56.7SSB, consumed (yes) 167 88 (52.7)Desserts/Treats, consumed (yes) 167 104 (62.3)Snacks, % kcal/d 167 17.3 11.5 0.0–67.7SD standard deviation, SSB sugar-sweetened beverages, BMI Body Mass IndexaBased on WHO growth chartsTable 2 Measurement model of parenting constructs usingconfirmatory factor analysisStandardizedFactor LoadingaStandardErrorCronbach’salphaAuthoritative Parenting 0.81Listens to child’s problems 0.45 0.07Aware of where child isgoing0.56 0.06Tells child when doing agood job0.65 0.06Checks child’s homework 0.63 0.06Knows what child doeswith friends0.72 0.05Likes child the way they are 0.60 0.06Tells child when to comehome0.75 0.05Parent Modeling 0.76Parents eat fruits aroundchild0.50 0.07Parents eat salad atrestaurants around child0.60 0.06Parents eat low fat snacksaround child0.75 0.05Parents eat low fatdressings around child0.82 0.05Family Meal Practices 0.60Family eats fast food 0.71 0.09Family eats while watchingtelevision0.42 0.09Family uses pre-packagedmeals0.66 0.09Family eats dessert afterdinner0.41 0.09Initial model fit: χ2(df = 249) = 494, p < .001; RMSEA = .08 [.07–.09], p < .001; CFI= .78; and SRMR = .09Final model fit: χ2(df = 87) = 123, p < .01; RMSEA = .05 [.03–.07], p = .50; CFI= .94; and SRMR = .06aStandardized factor loadings of final model, all significant at p < .001Correlations between factors were as follows: 0.15 between authoritativeparenting and parent modeling; 0.16 between authoritative parenting andfamily meal practices; and 0.25 between parent modeling and familymeal practicesWatts et al. BMC Nutrition  (2018) 4:3 Page 5 of 9foods in the home. Mixed findings suggest that interven-tions that target both aspects of the social and physicalenvironment of the home may help to support dietaryintake among adolescents who are overweight/obese, butthat they may be limited. Individual preferences and in-fluences outside the home including peers, and commu-nity and school environments are likely shaping thediets of adolescents who are overweight/obese.The strongest associations found in this study werebetween social aspects of the home food environment(modeling and meal routines) and having less health-ful foods in the home. This finding is not surprisingas parent preferences likely impact food purchasesand have been found to predict the foods served toyounger children [32]. In addition, families with mealroutines such as consuming fast food meals moreFig. 2 Structural equation model of factors within the home environment associated with the dietary intake of 167 overweight/obeseadolescents. This figure presents only the significant standardized regression coefficients (which can be interpreted as correlations) and the fullsolution is presented in Table 3. These effects are corrected for the following covariates: child age and sex, maternal education, and householdincome. Non-significant paths are not shown for clarityTable 3 All estimated paths of the structural equation model examining direct and indirect effects (n = 167)Dietary OutcomesFruit & Vegetables Fat SSB Desserts/Treats SnacksHome Food Environment Standardized regression coefficient, p-valueAvailability of High-Fat Foods −.145, p = .09 .157, p = .04 .190, p = .06 .159, p = .10 .243, p < .001Availability of High-Fat Treats −.178, p = .03 −.186, p = .01 .174, p = .10 .119, p = .28 −.018, p = .82Availability of Soft Drinks −.127, p = .12 −.060, p = .42 .154, p = .12 .142, p = .16 .003, p = .97Authoritative Parenting −.112, p = .15 .060, p = .36 .045, p = .63 −.019, p = .86 .061, p = .44Parent Modeling .068, p = .50 −.030, p = .75 −.039, p = .76 −.176, p = .20 −.046, p = .67Family Meal Practices −.190, p = .08 −.063, p = .53 .071, p = .59 .033, p = .82 −.159, p = .13CovariatesMaternal Education .100, p = .24 −.281, p < .001 .007, p = .94 .036, p = .74 −.006, p = .94Household Income −.147, p = .08 .176, p = .03 −.004, p = .97 .075, p = .48 .001, p = .99Age −.116, p = .16 .087, p = .29 .016, p = .87 .012, p = .91 .026, p = .75Sex (male) −.065, p = .44 .031, p = .68 −.205, p = .04 −.110, p = .27 −.146, p = .09Home Availability OutcomesHigh-Fat Food High-Fat Treats Soft DrinksHome Food Environment Standardized regression coefficient, p-valueAuthoritative Parenting .062, p = .41 .014, p = .85 .011, p = .61Parent Modeling −.143, p = .10 −.198, p = .03 −.321, p = .31Family Meal Practices −.283, p = .01 −.286, p = .01 −.054, p = .61CovariatesMaternal Education −.204, p = .003 −.028, p = .72 .010, p = .90Household Income −.165, p = .048 −.134, p = .10 −.124, p = .40SSB sugar-sweetened beveragesBolded values are significant at p < 0.05Watts et al. BMC Nutrition  (2018) 4:3 Page 6 of 9frequently have been found to report having chipsand soft drinks available in the home and a higher in-take of fast food and salty snacks by adolescents [33].In contrast, we did not find a direct relationship withadolescent dietary intake, but social influences mayindirectly shape what foods are made available orbroader aspects of the home environment. For ex-ample, parents of overweight/obese adolescents whomodel healthful eating and create healthier meal rou-tines may be more actively engaged in promotinghealthful eating as a whole and thus, also makingchanges to other aspects of the home eating environ-ment that influence diet. Interventions aimed at im-proving the quality of foods made available in homesmay benefit from also targeting parenting behaviors,such as modeling and family meal practices. In con-trast to previous studies [34], we did not identify anyassociations with an authoritative parenting style.These null findings may be explained by our measure,which did not identify the typical typologies of par-enting (authoritarian, authoritative, permissive anddisengaged) resulting in overlap with authoritarianparenting styles.Previous studies have found positive associations be-tween availability and adolescent consumption of a var-iety of foods/beverages including FV [35], non-corefoods [24], less healthful foods [36], energy-dense snacks[35], and soft drinks [37], and many similar associationswere found in the present study. However, there weremany null findings and some associations were in an un-expected direction. In light of the small number of fooditems that were assessed for availability, those measuredmay represent less healthful food items that are in thehome along with healthier options or that may be inmost households for special occasions only (e.g., parties,the weekend). Since these families had presented for anobesity treatment program, they may have made changesto the home environment (e.g., eliminating particularfoods) after enrolling in the intervention, but prior tothe baseline data collection that did not yet translateinto dietary change (and obscuring longer term pat-terns). Therefore, results may not reflect families thathave not yet contemplated making environmental or be-havioral changes in response to their child’s weight [38].It should also be noted that several associations weresignificant, but had small effect sizes (< 0.23 or < 5% ofthe variance explained) and may explain some of the in-consistent associations observed. Associations with smalleffect sizes tend to be less stable and these findingsshould be interpreted with caution. While it remains dif-ficult to determine how many subjects should be in-cluded in a SEM analysis to yield enough power, ourstudy was likely powered to detect moderate effectsbased on the findings from simulation studies [39]. Thusit would be useful to replicate these analyses in a largersample to test the stability of these associations. Otherlimitations of this study include that families of adoles-cents who are overweight/obese and who present fortreatment may be influenced by a more complex set ofindividual and psycho-social factors influencing foodchoices or may make changes to their environment inresponse to their own or their children’s weight. Thus,findings are most applicable to the families of adoles-cents who present for obesity treatment in urban or sub-urban settings. This study also utilized cross-sectionaldata; therefore, precludes causal inferences. Further-more, measures were not validated in a sample of ado-lescents with overweight/obesity and their parents andself-reported parenting practices and diets are suscep-tible to social desirability bias, which may have influ-enced the results towards a null finding [40]. Themeasure for family meal practices had particularly lowreliability and may highlight the difficulty in measuringthe home food environment, particularly in unique sam-ples. Improved measures for assessing the home foodenvironments of adolescents are needed. Finally, only aselect number of dietary outcomes were examined as in-dicators of diet quality. Our dietary database precludedthe examination of added sugars, for example, whichmay be an important indicator of a suboptimal dietamong adolescents.ConclusionsDespite confirmation of some hypothesized relationshipsin the present study, many dietary factors were not asso-ciated with aspects of the home food environment.However, parent modeling of healthy foods and positivemealtime routines were associated with the healthfulnessof foods offered in the homes of adolescents who areoverweight/obese. It remains a challenge to characterizeboth dietary intake and the complex aspects of thehousehold environment that influence diet. The homeenvironment and its influence on diet may be unique foroverweight/obese adolescents; thus, future research isneeded to identify important influences of diet amongthis understudied group. Future research should alsoconsider the role of environments outside the home onadolescent dietary behaviors, such as the school food en-vironment, and eating out with peers.Additional fileAdditional file 1: Parent questionnaire items assessing demographiccharacteristics and the home food environment. (DOCX 375 kb)AbbreviationsBMI: Body Mass Index; CFA: Confirmatory Factor Analysis; CFI: Comparatie FitIndex; FV: Fruits and Vegetables; RMSEA: Root Mean Square Error ofApproximation; SES: Socioeconomic Status; SRMR: Standardized Root MeanWatts et al. BMC Nutrition  (2018) 4:3 Page 7 of 9Square Residual; SSB: Sugar-sweetened Beverages; WHO: World HealthOrganization; WLSMV: Means- and Variance- adjusted Weighted Least Squares;WRMR: Weighted Root Mean Square ResidualAcknowledgementsThe data presented in this paper were previously published as part of the PhDthesis of the first author titled, “Influences in home, school and communityenvironments on the dietary behaviours of overweight/obese adolescents”.FundingThe data collection was supported by the Canadian Institutes of HealthResearch (CIHR) Institute of Nutrition, Metabolism and Diabetes and theHealth Research Foundation [Agreement number 00789–000] to LCM. AWWreceived doctoral support from the Canadian Institutes of Health Research,the Danone Institute of Canada, and the Heart and Stroke Foundation ofCanada. LCM received salary support from the Child and Family ResearchInstitute at the BC Children’s Hospital.Availability of data and materialsThe datasets used and/or analysed during the current study are availablefrom the corresponding author on reasonable request.Authors’ contributionsAW was involved in conceptualizing the study, collecting the data, data analysis,data interpretation and writing the manuscript. LM provided guidance on thestatistical analysis and along with CY, RH and SB was involved in study design,data interpretation, and critically reviewing the manuscript. All authors haveread and approved the final version of the manuscript.Ethics approval and consent to participateWritten consent was obtained from adolescent participants and one of theirparents. All study protocols were approved by the University of British Columbiaand the University of Waterloo ethics boards.Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.Author details1School of Population and Public Health, University of British Columbia, 2206East Mall, Vancouver, British Columbia V6T 1Z9, Canada. 2Present address:Division of Epidemiology and Community Health, School of Public Health,University of Minnesota, Minneapolis, Minnesota 55401, USA. 3Department ofFood, Nutrition & Health, University of British Columbia, 2357 Main Mall,Vancouver, British Columbia V6T 1Z4, Canada. 4School of Public Health andHealth Systems, University of Waterloo, 200 University Avenue West,Waterloo, Ontario N2L 3G5, Canada. 5School of Population and Public Health,University of British Columbia, 2206 East Mall, Vancouver, British ColumbiaV6T 1Z9, Canada. 6School of Population and Public Health, University ofBritish Columbia, 2206 East Mall, Vancouver, British Columbia V6T 1Z9,Canada. 7Child and Family Research Institute, University of British Columbia,4480 Oak Street, room F512a, Vancouver, BC V6H 3V4, Canada.Received: 3 January 2017 Accepted: 29 January 2018References1. Story M. Nutritional requirements during adolescence. In: McAnarney ER,Kreipe RE, Orr DE, Comerci GD, editors. Textbook of adolescent medicine.Philadelphia: W.B. Saunders; 1992. p. 75–84.2. Garriguet D. Canadians’ eating habits. Health Rep. 2007;18:17–32.3. Garriguet D. Beverage consumption of children and teens. Health Rep. 2008;19:1–7.4. Keast DR, Fulgoni VL, Nicklas TA, O’Neil CE. Food sources of energy andnutrients among children in the United States: National Health andnutrition examination survey 2003–2006. Nutrients. 2013;5:283–301.5. Lillico HG, Hammond D, Manske S, Murnaghan D. The prevalence of eatingbehaviors among Canadian youth using cross-sectional school-basedsurveys. BMC Public Health. 2014;14:323.6. Black JL, Billette J-M. Fast food intake in Canada: differences amongCanadians with diverse demographic, socio-economic and lifestylecharacteristics. Can J Public Health. 2015;106:e52–8.7. Ebbeling CB, Feldman HA, Chomitz VR, Antonelli TA, Gortmaker SL,Osganian SK, et al. A randomized trial of sugar-sweetened beverages andadolescent body weight. N Engl J Med. 2012;367:1407–16.8. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity and trends inbody mass index among us children and adolescents, 1999-2010. JAMA.2012;307:483–90.9. Roberts KC, Shields M, de Groh M, Aziz A, Gilbert J-A. Overweight andobesity in children and adolescents: results from the 2009 to 2011 Canadianhealth measures survey. Health Rep. 2012;23:37–41.10. Oude Luttikhuis H, Baur L, Jansen H, Shrewsbury VA, O’Malley C, Stolk RP,et al. Interventions for treating obesity in children. Cochrane Database SystRev. 2009;3:1–57.11. Golan M. Parents as agents of change in childhood obesity - from researchto practice. Int J Pediatr Obes. 2006;1:66–76.12. Rosenkranz RR, Dzewaltowski DA. Model of the home food environmentpertaining to childhood obesity. Nutr Rev. 2008;66:123–40.13. Story M, Neumark-Sztainer D, French S. Individual and environmentalinfluences on adolescent eating behaviors. J Am Diet Assoc. 2002;102:S40–51.14. Couch SC, Glanz K, Zhou C, Sallis JF, Saelens BE. Home food environment inrelation to children’s diet quality and weight status. J Acad Nutr Diet. 2014;114:1569–79.15. Pearson N, Biddle SJH, Gorely T. Family correlates of fruit and vegetableconsumption in children and adolescents: a systematic review. PublicHealth Nutr. 2009;12:267–83.16. Neumark-Sztainer D, Wall M, Perry C, Story M. Correlates of fruit andvegetable intake among adolescents. Findings from project EAT. Prev Med.2003;37:198–208.17. Golan M, Kaufman V, Shahar DR. Childhood obesity treatment: targetingparents exclusively v. Parents and children. Br J Nutr. 2006;95:1008–15.18. de Onis M, Onyango AW, Borghi E, Siyam A, Nishida C, Siekmann J.Development of a WHO growth reference for school-aged children andadolescents. Bull World Health Organ. 2007;85:660–7.19. Hanning R, Royall D, Toews J, Blashill L, Wegener J, Driezen P. Web-basedfood behaviour questionnaire: validation with grades six to eight students.Can J Diet Pract Res. 2009;70:172–8.20. Cullen KW, Baranowski T, Rittenberry L, Cosart C, Hebert D, de Moor C.Child-reported family and peer influences on fruit, juice and vegetableconsumption: reliability and validity of measures. Health Educ Res. 2001;16:187–200.21. Loth KA, MacLehose RF, Larson N, Berge JM, Neumark-Sztainer D. Foodavailability, modeling and restriction: how are these different aspects of thefamily eating environment related to adolescent dietary intake? Appetite.2016;96:80–6.22. Cullen KW, Baranowski T, Rittenberry L, Cosart C, Owens E, Hebert D, et al.Socioenvironmental influences on children’s fruit, juice and vegetableconsumption as reported by parents: reliability and validity of measures.Public Health Nutr. 2000;3:345–56.23. Ihmels MA, Welk GJ, Eisenmann JC, Nusser SM. Development andpreliminary validation of a family nutrition and physical activity (FNPA)screening tool. Int J Behav Nutr Phys Act. 2009;6:1–10.24. Johnson L, van Jaarsveld CHM, Wardle J. Individual and family environmentcorrelates differ for consumption of core and non-core foods in children. BrJ Nutr. 2011;105:950–9.25. MacFarlane A, Cleland V, Crawford D, Campbell K, Timperio A. Longitudinalexamination of the family food environment and weight status amongchildren. Int J Pediatr Obes. 2009;4:343–52.26. Baranowski T. GEMS: Home Food Availability Questionnaire. 2000. http://www.midss.org/content/girls-health-enrichment-multi-site-studies-gems.Accessed 15 Nov 2016.27. Cullen KW, Klesges LM, Sherwood NE, Baranowski T, Beech B, Pratt C, et al.Measurement characteristics of diet-related psychosocial questionnairesamong African-American parents and their 8- to 10-year-old daughters:Watts et al. BMC Nutrition  (2018) 4:3 Page 8 of 9results from the girls’ health enrichment multi-site studies. Prev Med. 2004;38(Suppl):34–42.28. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structureanalysis: conventional criteria versus new alternatives. Struct Equ ModelMultidiscip J. 1999;6:1–55.29. Brown TA. Confirmatory factor analysis for applied research. New York:Guilford Press; 2006.30. Beauducel A, Herzberg PY. On the performance of maximum likelihoodversus means and variance adjusted weighted least squares estimation inCFA. Struct Equ Model Multidiscip J. 2006;13:186–203.31. Flora DB, Curran PJ. An empirical evaluation of alternative methods ofestimation for confirmatory factor analysis with ordinal data. PsycholMethods. 2004;9:466–91.32. Skinner JD, Carruth BR, Wendy B, Ziegler PJ. Children’s food preferences: alongitudinal analysis. J Am Diet Assoc. 2002;102:1638–47.33. Boutelle KN, Fulkerson JA, Neumark-Sztainer D, Story M, French SA. Fastfood for family meals: relationships with parent and adolescent food intake,home food availability and weight status. Public Health Nutr. 2007;10:16–23.34. Berge JM. A review of familial correlates of child and adolescent obesity:what has the 21st century taught us so far? Int J Adolesc Med Health. 2009;21:457–83.35. Pearson N, Ball K, Crawford D. Predictors of changes in adolescents’consumption of fruits, vegetables and energy-dense snacks. Br J Nutr. 2011;105:795–803.36. Campbell KJ, Crawford DA, Salmon J, Carver A, Garnett SP, Baur LA.Associations between the home food environment and obesity-promotingeating behaviors in adolescence. Obesity. 2007;15:719–30.37. Hebden L, Hector D, Hardy LL, King L. A fizzy environment: availability andconsumption of sugar-sweetened beverages among school students. PrevMed. 2013;56:416–8.38. Campbell M, Benton JM, Werk LN. Parent perceptions to promote ahealthier lifestyle for their obese child. Soc Work Health Care. 2011;50:787–800.39. Wolf EJ, Harrington KM, Clark SL, Miller MW. Sample size requirements forstructural equation models: an evaluation of power, bias, and solutionpropriety. Educ Psychol Meas. 2013;76:913–34.40. Mâsse LC, Watts AW. Stimulating innovations in the measurement ofparenting constructs. Child Obes. 2013;9(Suppl):S5–13.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Watts et al. BMC Nutrition  (2018) 4:3 Page 9 of 9


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items