Hostname: page-component-78c5997874-mlc7c Total loading time: 0 Render date: 2024-10-31T05:11:36.112Z Has data issue: false hasContentIssue false

Adolescent vegetable consumption: the role of socioemotional family characteristics

Published online by Cambridge University Press:  16 April 2021

Elisabeth L Melbye*
Affiliation:
Faculty of Social Sciences, Norwegian School of Hotel Management, University of Stavanger, 4036Stavanger, Norway Oral Health Center of Expertise Rogaland, 4016Stavanger, Norway
Solveig E Hausken-Sutter
Affiliation:
Faculty of Medicine, Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, 0316Oslo, Norway
Nanna Lien
Affiliation:
Faculty of Medicine, Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, 0316Oslo, Norway
Mona Bjelland
Affiliation:
Faculty of Medicine, Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, 0316Oslo, Norway
*
*Corresponding author: Email elisabeth.lind.melbye@throg.no
Rights & Permissions [Opens in a new window]

Abstract

Objective:

To describe associations between adolescents’ frequency of vegetable consumption, food parenting practices and socioemotional family characteristics, and to explore potential mediated relationships that may contribute to an understanding of the family processes involved.

Design:

Cross-sectional survey among adolescents aged 13–15 years.

Setting:

A survey questionnaire including self-report measures on adolescents’ frequency of vegetable consumption, perceived food parenting practices (i.e. family dinner frequency, maternal/paternal healthy eating guidance (HEG), maternal/paternal social support for vegetable consumption) and socioemotional family characteristics (i.e. general family functioning and level of cohesion and conflict within the family) was distributed in a convenience sample of secondary school students.

Participants:

Four hundred forty students from five secondary schools in eastern Norway completed the questionnaire.

Results:

Results from multiple linear regression analysis revealed positive and statistically significant associations between adolescents’ frequency of vegetable consumption, maternal HEG and family cohesion. A partial indirect (mediated) association between family cohesion and adolescents’ frequency of vegetable consumption, working through maternal HEG, was also found.

Conclusions:

Results from the present study suggest that perceived family cohesion may influence adolescents’ frequency of vegetable consumption both directly and indirectly. However, there is a need for continued investigation of family-related factors influencing adolescent eating. In particular, the role of socioemotional family characteristics should be further scrutinised in future studies.

Type
Research paper
Copyright
© The Author(s), 2021. Published by Cambridge University Press on behalf of The Nutrition Society

An insufficient intake of fruit and vegetables (FV) is found to be among the leading risk factors of the global burden of non-communicable diseases(Reference Lim, Vos and Flaxman1). Therefore, it is a concern that most adolescents in the Nordic countries and elsewhere have a lower consumption of FV, particularly vegetables, than recommended by the authorities(Reference Fismen, Smith and Torsheim2Reference Golley, Hendrie and McNaughton4). Adolescence is known as a critical period for the development of dietary behaviours(Reference Striegel-Moore and Bulik5), and since such behaviours are likely to track into adult life(Reference Lipsky, Haynie and Liu6), it is important to increase FV consumption among adolescents to reduce morbidity and mortality from non-communicable diseases. Thus, continued research aiming to reveal key influences on adolescent FV consumption and to develop effective interventions tailored at this group of the population seems imperative.

In Norway, which is the setting of the present study, the Norwegian School Fruit Scheme was launched in 2007 as part of a Norwegian governmental initiative to promote and increase the consumption of FV among children and adolescents. The Norwegian School Fruit Scheme provided students in all secondary schools (grades 8–10) and all combined schools (grades 1–10) with a free piece of fruit or vegetable every school day. The programme lasted for 7 years and resulted in an increased fruit consumption among adolescents regardless of gender and socio-economic status. However, the same positive effect was not found for vegetables(Reference Bere, Hilsen and Klepp7Reference Øvrum and Bere9). One obvious reason is that the programme primarily delivered fruits to the students. Thus, the potential for increasing vegetable consumption through the Norwegian School Fruit Scheme was limited. Correspondingly, a review by Evans et al. (Reference Evans, Christian and Cleghorn10) found that school-based interventions moderately improved fruit intake but had minimal impact on vegetable intake.

The influence of the traditional Norwegian meal pattern, which typically includes one hot meal (dinner) and two or three cold meals(Reference Mäkelä, Kjærnes and Ekström11), must be taken into consideration in the assessment of Norwegian adolescents’ vegetable consumption. The cold meals usually consist of bread or cereals. Fruit is more practical to eat with these cold meals and in between meals than vegetables, as they come in convenient portion sizes, ‘in their own package’ and need little treatment prior to eating compared with vegetables(Reference Bere, Veierød and Klepp12). Consequently, in Norway, vegetables are mostly eaten at dinner(Reference Myhre, Løken and Wandel13), which most children and adolescents share with their families(Reference Hausken, Lie and Lien14,Reference Melbye, Øgaard and Øverby15) . The importance of family meals for more healthful food choices has been stated in reviews by Berge(Reference Berge16) and Fulkerson et al. (Reference Fulkerson, Larson and Horning17), and the presence of at least one parent at meals has been associated with a higher FV consumption(Reference Fulkerson, Larson and Horning17,Reference Videon and Manning18) . Moreover, irregular family meals (breakfasts and dinners) have been associated with less vegetable consumption in a recent study by Totland et al. (Reference Totland, Knudsen and Paulsen19). Hence, a relevant approach for understanding predictors of adolescents’ vegetable consumption might be recognising family-related factors, besides shared family dinners, that could influence this behaviour.

Since the family environment has been acknowledged as a fundamental context for the development of eating behaviours(Reference Pearson, Atkin and Biddle20), several studies have addressed factors such as socio-economic position(Reference Fismen, Smith and Torsheim2) and various food parenting practices, including the arrangement of family meals(Reference Gebremariam, Henjum and Terragni21Reference Pearson, Biddle and Gorely23). However, less research has focused on fundamental, socioemotional family characteristics such as general family functioning and level of family cohesion or conflict as correlates of adolescent eating(Reference Larson, MacLehose and Fulkerson24). Previous research has linked socioemotional family characteristics to social and emotional outcomes in youth(Reference Balistreri and Alvira-Hammond25Reference Shek27), and Kitzmann and Beech(Reference Kitzmann and Beech28) have accentuated the importance of exploring these fundamental family characteristics in relation to (un)healthy eating among adolescents. Moreover, these features of the family environment have been suggested as contexts that may enhance or limit the effectiveness of family-based interventions(Reference Kitzmann and Beech28,Reference Sleddens, Gerards and Thijs29) . Thus, it seems relevant to scrutinise the role of socioemotional family characteristics as potential determinants of adolescent vegetable consumption.

The present study is part of the Family and Dietary Habits project and is based on a framework constructed to describe various family environmental levels and constructs included in questionnaires developed for this project(Reference Bjelland, Hausken and Sleddens30). The Family and Dietary Habits framework constitutes an ecological model emphasising factors within the family environment that may contribute to explain dietary behaviours in adolescents. The factors included in this framework are organised in three levels: an individual level (adolescent eating and personal characteristics), a parental level (parenting style and food parenting practices, including parents’ arrangement of shared family meals) and a family level (fundamental socioemotional family characteristics), with factors interacting within and across levels. Adolescent vegetable consumption was the individual level factor of interest in the present study and served as the dependent variable in model analyses. Parental level factors hypothesised to be related to adolescent vegetable consumption were food parenting practices such as family dinner frequency, healthy eating guidance (HEG) and positive encouragement for vegetable consumption. Family level factors hypothesised to be related to adolescent vegetable consumption were fundamental socioemotional family characteristics such as general family functioning, family cohesion and family conflict.

Based on this introductory section, the objectives of the present study were to: (1) describe associations between adolescent vegetable consumption and the parental and family level factors presented above and (2) explore potential mediated relationships that may contribute to an increased understanding of the family processes involved.

Methods

Participants and procedures

Secondary school students aged 13–15 years were recruited through a convenience sample of five public schools in eastern Norway. Since the students were underaged, a parental consent form including questions assessing household educational level was distributed to the students’ parents. After receiving written consent from parents, students who agreed to participate were asked to complete a web-based questionnaire during school hours. Of the 1136 students invited to take part in this cross-sectional study, 440 (39 %) completed the questionnaire.

Questionnaire

The students spent between 25 and 45 min on completing the questionnaire, which consisted of 141 questions assessing dietary intake (vegetables and sugar-sweetened beverages), accessibility and availability (of vegetables and sugar-sweetened beverages), personal characteristics, family meals, parenting styles, food parenting practices, socioemotional family characteristics and sociodemographic factors(Reference Bjelland, Hausken and Sleddens30). All measures in the questionnaire, including parental and family level factors, were assessed from the perspective of the students. The subset of measures used in the current study is presented in the following paragraphs.

Frequency of vegetable consumption

Vegetable consumption was assessed using frequency measures reproduced from Lien et al. (Reference Lien, Bjelland and Bergh31). Students were asked two questions to report their usual intake of cold (raw) and heated (boiled, fried, roasted, etc.) vegetables, respectively, on an eight-point frequency scale (1 = never/seldom, 2 = less than once a week, 3 = 1–2 times a week, 4 = 3–4 times a week, 5 = 5–6 times a week, 6 = once a day, 7 = 2 times a day, 8 = 3 times a day or more). Vegetable juices were not included in this measure. As suggested by Andersen et al. (Reference Andersen, Bere and Kolbjornsen32), the response categories were recoded to reflect vegetable consumption in times/week prior to data analyses (0 = never/seldom; 0·5 = less than once a week; 1·5 = 1–2 times a week; 3·5 = 3–4 times a week; 5·5 = 5–6 times a week; 7 = once a day; 14 = twice a day; 21 = 3 times a day). Consequently, all response categories had a common denominator (times a week), which improved the readability of the results and increased comparability with studies using similar measures(Reference Lien, Bjelland and Bergh31Reference Melbye, Øverby and Øgaard34). Total frequency of vegetable consumption was calculated by adding up the consumption of cold and heated vegetables.

Family dinner frequency

Family dinner frequency was measured with one item: ‘How often does your mother and/or father usually sit down and eat dinner with you?’. Response alternatives were given on an eight-point frequency scale (1 = never/seldom; 2 = once a week; 3 = twice a week; 4 = 3 times a week; 5 = 4 times a week; 6 = 5 times a week; 7 = 6 times a week; 8 = 7 times a week). This variable was not normally distributed as most of the adolescents ate dinner together with their parent(s) 6 or 7 times/week (80·5 %). Therefore, responses were dichotomised into ‘0–5 times a week’ and ‘6–7 times a week’.

Healthy eating guidance

The concept of HEG was developed by Haszard et al. (Reference Haszard, Williams and Dawson35) and includes food parenting practices like teaching about nutrition, modelling healthy eating, encouraging a balanced and varied diet and making healthy foods and beverages accessible in the home. In the present study, perceived maternal and paternal HEG was measured separately for mothers and fathers by the nine-item HEG subscale adapted from Haszard et al.’s(Reference Haszard, Williams and Dawson35) five-factor version of Musher-Eizenman and Holub’s(Reference Musher-Eizenman and Holub36) Comprehensive Feeding Practices Questionnaire. To the authors’ knowledge, the Comprehensive Feeding Practices Questionnaire has previously been used to assess food parenting practices from a parental perspective only. Thus, the items had to be slightly modified to represent the perspective of adolescents in the current research. For example, ‘My mother/father discusses with me why it is important to eat healthy foods’ (see Appendix 1 for a complete list of HEG items). The HEG items were scored on a five-point Likert scale ranging from 1 (disagree) to 5 (agree), where the sum of scores was divided by 9 to give a total average score ranging from 1·0 to 5·0. Higher scores indicate higher levels of HEG. Haszard et al. (Reference Haszard, Williams and Dawson35) reported good internal consistency reliability for the HEG subscale with an α coefficient of 0·82. Also, construct validity was supported by Haszard et al. (Reference Haszard, Williams and Dawson35), as parents with concern for child overweight, and parents who rated a healthy diet as very important for their child, were found to report higher levels of HEG. Previous testing of the Comprehensive Feeding Practices Questionnaire with parents in a Norwegian setting indicated that this instrument is also a valid tool for measuring multiple parental feeding practices with parents of 10–12-year-olds(Reference Melbye, Øgaard and Øverby37).

Positive encouragement for vegetable consumption

Parents’ encouragement of healthy eating behaviours has been associated with positive outcomes(Reference Dave, Evans and Condrasky38). For example, Melbye, Øgaard and Øverby(Reference Melbye, Øgaard and Øverby39) found a positive association between parental encouragement of a balanced and varied diet and vegetable intake in 10–12-year-olds. Furthermore, Young, Fors and Hayes(Reference Young, Fors and Hayes40) found that perceived parental support for FV consumption was a significant predictor of FV consumption in adolescents. In the current research, perceived maternal and paternal positive encouragement was measured by the five-item Positive Encouragement Subscale (PES) adapted from Dave et al.’s(Reference Dave, Evans and Condrasky38) Emotional Social Support Scale for FV intake. Since vegetable consumption was the dependent variable of interest in the present study, the PES was modified to cover intake of vegetables only: ‘How often, during the past month, did your mother/father (1) compliment you for your vegetable consumption; (2) encourage you to eat vegetables when you were tempted not to; (3) discuss your vegetable consumption with you; (4) remind you to eat vegetables; and (5) asked you on ideas on how you could eat more vegetables’. The PES items were scored on a five-point frequency scale ranging from 1 (never) to 5 (very often), where the sum of scores was divided by 5 to give a total average score ranging from 1·0 to 5·0. Higher scores indicate higher levels of positive encouragement. Dave et al. (Reference Dave, Evans and Condrasky38) observed good internal consistency for the PES with an α coefficient of 0·82. They also found evidence of construct validity, as the PES correlated with related measures such as reinforcement, availability and accessibility.

General family functioning

General family functioning includes structural, organisational and interactional patterns of the family as described by Bowen’s(Reference Bowen41) Family Systems Theory. According to this theory, the interactions that occur within a family are reciprocal in that each member of the family is being shaped by other family members’ behaviours. These mutual influences may provide insight into behaviours that ultimately determine health outcomes in individual family members(Reference Larson, MacLehose and Fulkerson24). In the current research, perceived family functioning was measured with the General Functioning Scale, which is a twelve-item subscale extracted from the McMaster Family Assessment Device(Reference Epstein, Baldwin and Bishop42). For example, ‘Planning family activities is difficult because we misunderstand each other’ (see Appendix 1 for a complete list of General Functioning Scale items). The General Functioning Scale response categories ranged from 1 (strongly agree) to 4 (strongly disagree), where the sum of scores was divided by 12 to give a total average score ranging from 1·0 to 4·0. A higher score (i.e. ≥2·0) indicates poorer family functioning(Reference Byles, Byrne and Boyle43,Reference Ridenour, Daley and Reich44) . The General Functioning Scale has demonstrated good psychometric properties with adolescents in various cultural contexts(Reference Kazarian45Reference Shek47).

Family cohesion and family conflict

Family cohesion has been defined as the degree of perceived commitment, support and help family members provide for each other, or as the emotional connection between family members(Reference Moos, Moos, Fredman and Sherman48). Cohesion is recognised as an important influence on children’s development and functioning(Reference Ackard, Neumark-Sztainer and Story49,Reference Franko, Thompson and Affenito50) and has been shown to affect adolescents’ feeling of control over their own health(Reference Zdanowicz, Janne and Reynaert51). Interestingly, previous studies have suggested a link between family cohesion and healthy dietary behaviours among adolescents(Reference Franko, Thompson and Affenito50,Reference Franko, Thompson and Bauserman52Reference Kalavana, Maes and De Gucht54) . Family conflict has been defined as the degree of perceived aggression and conflict among family members(Reference Moos55), and in contrast to cohesion, it has been associated with negative outcomes in children and adolescents(Reference Moos55Reference Wadsworth and Compas58). For example, in a study by Schuetzmann et al. (Reference Schuetzmann, Richter-Appelt and Schulte-Markwort59), conflict and rejection were linked to deviant eating behaviour in preadolescents. Furthermore, family conflict has been associated with unhealthy eating in high-school students(Reference Kalavana, Maes and De Gucht54). In the present study, family cohesion and family conflict were measured by items derived from the Cohesion (9 items) and Conflict (8 items) subscales included in the Family Environment Scale (FES)(Reference Moos, Moos, Fredman and Sherman48). For example, cohesion: ‘Family members really help and support one another’. For example, conflict: ‘We fight a lot in our family’ (see Appendix 1 for a complete list of items measuring cohesion and conflict). The cohesion and conflict items were scored on a four-point scale ranging from 1 (true) to 4 (false). As for other rating scale variables in this study, averaged sum scores for the cohesion and conflict subscales were calculated. Higher scores indicate higher levels of cohesion and conflict, respectively(Reference Charalampous, Kokkinos and Panayiotou60). Previous research assessing the psychometric properties of the FES, from which the cohesion and conflict scales are derived, has shown inconsistent results. In terms of internal consistency reliability for the FES subscales, the originally reported α coefficients varied between 0·64 and 0·79, with an acceptable benchmark to be above 0·60 (this value was justified by the emphasis placed on the breadth of the measured constructs)(Reference Moos, Moos, Fredman and Sherman48). In a study by Charalampous, Kokkinos and Panayiutou(Reference Charalampous, Kokkinos and Panayiotou60), where the validity and reliability of the FES were tested with individuals aged 16–60 years, the cohesion and conflict scales emerged as unidimensional, supporting the convergent validity of the scales. The α coefficients were found to be similar to the α coefficients originally reported by Moos and Moos(Reference Moos, Moos, Fredman and Sherman48), 0·74 and 0·64 for cohesion and conflict, respectively. Charalampous, Kokkinos and Panayiutou(Reference Charalampous, Kokkinos and Panayiotou60) endorsed the strong theoretical basis and predictive utility of the scales which make them fruitful for examining the family environment. Furthermore, in a study by Kalavana, Maes and de Gucht(Reference Kalavana, Maes and De Gucht54), where the FES was administered to senior high-school students (mean age 16·6 years, sd = 4·8), the construct validity of the cohesion and conflict subscales was supported and both factors had an acceptable internal consistency with α of 0·76 (cohesion) and 0·74 (conflict). The internal consistency for the cohesion and conflict subscales was also found to be acceptable in a more recent study on adolescents aged 11–18 years, with reported α of 0·80 (cohesion) and 0·75 (conflict)(Reference Xu, Boyd and Butler61).

Sociodemographic factors

Sociodemographic factors are well-known, inflexible correlates of dietary behaviours and were included as covariates in the current study. Data from the parent with the longest education were used as a measure of highest household educational level and were classified as ‘less than or equal to 12 years’, ‘between 13 and 16 years’ and ‘more than 16 years’. Gender was classified as ‘boy’ or ‘girl’. Family structure was classified as ‘living with both parents’ v. ‘other living arrangements’.

Statistical analyses

Initial analyses

The SPSS statistical software package version 25 (SPSS Inc.) was used for statistical analyses. Initial analyses included frequencies for categorical variables, and mean scores, standard deviations, skewness, kurtosis, Cronbach’s α and intra-class correlation coefficients for rating scale variables. As suggested by Kline(Reference Kline62), we applied cut-off values of 3·0 and 8·0 for skewness and kurtosis, respectively. Cronbach’s α was used to assess internal consistency reliability for all rating scale variables and was classified as >0·70 = ‘acceptable’ and >0·80 = ‘preferable’(Reference Field63). Intra-class correlation coefficients were used to assess test–retest reliability in a subsample of adolescents (n 54) and were classified as ≥0·81 = ‘excellent’, 0·61–0·80 = ‘moderate’ and ≤0·40 = ‘poor’(Reference Singh, Vik and Chinapaw64). Prior to regression analyses, bivariate correlations were run to test for multicollinearity between independent variables. We applied a cut-off value of 0·80 or greater for multicollinearity, as suggested by Haerens et al. (Reference Haerens, Craeynest and Deforche53).

Model analyses

Two different regression strategies were applied to address the research objectives. First, a multiple linear regression analysis was run to describe associations between adolescents’ frequency of vegetable consumption (individual level), perceived food parenting practices (parental level) and socioemotional family characteristics (family level) derived from the Family and Dietary Habits framework. Sociodemographic factors were also included and treated as covariates. Next, based on results from the multiple linear regression, potential mediated relationships were explored. The analytical strategy applied to test for mediation was based on Hayes’(Reference Hayes65) modelling tool PROCESS, version 3. This tool includes bootstrapping resampling techniques resulting in more robust results than standard methods relying on parametric assumptions(Reference Cerin and MacKinnon66). Since approximately 5000 bootstrap samples are considered sufficient for most applications(Reference Fritz, Taylor and Mackinnon67), and since it is the current PROCESS default, we generated 5000 bootstrap samples for the mediation analysis by resampling with replacement from the original sample. Associations between predictor and mediator, and between mediator and outcome variables, were reported in traditional manner by unstandardised coefficients and associated P-values. This was also the case for total and direct associations between predictor and outcome variables. Since P-values for indirect (mediated) effects are not displayed in the PROCESS output, the indirect effect was reported by unstandardised regression coefficients with 95 % CI. Conforming to the bootstrapping approach, an indirect effect which CI did not include zero was considered statistically significant(Reference Cerin and MacKinnon66).

Results

Initial analyses

Frequencies for categorical variables are presented in Table 1. As can be seen from this table, the sample consisted of 52 % girls and 48 % boys. Most adolescents (74 %) came from highly educated households (34 % with 13–16 years of education, 27 % with more than 16 years of education), and a large proportion (69 %) lived together with both parents. Mean scores, standard deviations, skewness, kurtosis, Cronbach’s α and intra-class correlation coefficients for rating scale variables are presented in Table 2. As depicted in this table, all variables had values within the range of chosen cut-offs for skewness and kurtosis, Cronbach’s α were satisfactory to preferable (range: 0·72–0·89) and intra-class correlation coefficients were good to excellent (range: 0·68–0·83). Finally, no multicollinearities were found for the independent variables to be included in subsequent model analyses (range: 0·01–0·33).

Table 1 Frequencies for categorical variables (n 440)

Table 2 Means, standard deviations, skewness, kurtosis, Cronbach’s α and intra-class correlations (ICC) for rating scale variables

HEG, healthy eating guidance; PEV, positive encouragement for vegetable consumption.

* n 440.

n 54.

Times/week.

Model analyses

The multiple linear regression analysis resulted in positive and statistically significant associations between adolescents’ frequency of vegetable consumption and perceived maternal HEG (β = 0·22, P = 0·04), family cohesion (β = 0·21, P = 0·02) and household educational level (β = 0·11, P = 0·04), explaining 9 % of the variance in vegetable consumption (Table 3). Based on these results, a single-mediator model including these variables was tested to explore the potential processes involved.

Table 3 Regression coefficients (β) and variance explained (R 2) for multiple linear regression on vegetable consumption frequency

HEG, healthy eating guidance; PEV, positive encouragement for vegetable consumption.

* P < 0·05.

Since a causal model is the theoretical basis for the examination of potential mediating mechanisms, the temporal order assumption of a causal model was taken into account when specifying this model. To be more specific, in a three-variable mediation model, the independent variable X is hypothesised to precede (and cause) mediator M, which, in turn, precedes (and causes) dependent variable Y, such that accounting for the effect of X on M and of M on Y explains, in part or in whole, the influence of X on Y(Reference Gelfand, Mensinger and Tenhave68). Following from this, it seems reasonable that family cohesion (which is a fundamental family characteristic) temporally precedes and thus may have the potential to influence, adolescents’ perceptions or acknowledgement, of maternal HEG (which is a context-specific, food-related behaviour). The opposite (perceived maternal HEG influencing family cohesion) seems less likely. Likewise, it seems reasonable that maternal HEG temporally precedes and may have the potential to influence adolescents’ frequency of vegetable consumption. Consequently, family cohesion was included as the predictor (X), while maternal HEG was included as the potential mediator (M) of the association between family cohesion and adolescents’ frequency of vegetable consumption (Y). Household educational level was included as a covariate. Results from mediation analysis showed that family cohesion was significantly and positively associated with maternal HEG (β = 0·65, P < 0·001) and that maternal HEG was significantly and positively associated with adolescents’ frequency of vegetable consumption (β = 0·90, P < 0·05). Positive, statistically significant total (β = 2·63, P < 0·001) and direct (β = 2·04, P < 0·01) associations were also found between family cohesion and adolescents’ frequency of vegetable consumption. Finally, a statistically significant indirect (mediating) effect of perceived maternal HEG on the association between family cohesion and adolescents’ frequency of vegetable consumption (β = 0·58, 95 % CI 0·11, 1·15) was found. This effect accounted for about 22 % of the total effect of family cohesion on adolescent vegetable consumption (i.e. ratio of indirect to total effect, PM = 0·22) and thus represents a partial mediation (Fig. 1).

Fig. 1. Path diagram for modelling family cohesion as a predictor of adolescents’ frequency of vegetable consumption, partly mediated by maternal healthy eating guidance (HEG). Coefficient estimates (b) and statistical significance tests (P-values and CI) were obtained using the PROCESS script for SPSS

Discussion

The present study aimed to: (1) describe associations between adolescents’ frequency of vegetable consumption, selected food parenting practices and socioemotional family characteristics; and (2) explore potential mediated relationships that may contribute to an increased understanding of the family processes involved. Family cohesion and maternal HEG was found to be the most important correlates of vegetable consumption frequency in the multiple linear regression model, while household educational level appeared as a weaker correlate. When testing for mediated relationships (adjusting for household educational level), maternal HEG was found to act as a partial mediator of the positive association between family cohesion and adolescents’ frequency of vegetable consumption.

The finding of a positive association between family cohesion and adolescents’ frequency of vegetable consumption is supported by previous research by Franko et al. (Reference Franko, Thompson and Bauserman52), which indicates that family cohesion may be linked to healthy eating in numerous ways: First, a cohesive family may be a family that explicitly promotes healthy behaviours: for example, parents encourage healthy eating, and adolescents who feel a high level of connectedness with their parents may be more inclined to follow their suggestions. Second, cohesion has been linked to psychological health, which may have a direct effect on the development of healthy attitudes and behaviours (including healthy eating) in children and adolescents(Reference Franko, Thompson and Bauserman52,Reference Haerens, Craeynest and Deforche53,Reference Moore and Harré69) .

The positive association between maternal HEG and adolescents’ frequency of vegetable consumption is in line with former studies on social influences postulating that the influence of important others is an essential element in explaining child and adolescent eating behaviours. For example, in a cohort study among children aged 6–11 years and their parents, Couch et al. (Reference Couch, Glanz and Zhou70) found that food parenting practices such as encouragement, modelling and family rules showed strong positive relationships with child FV intake. Positively framed practices such as these were also associated with increased consumption of vegetables and decreased consumption of sugar-sweetened beverages in a Norwegian study on 10–12-year-olds(Reference Melbye and Hansen71). Mothers are of special interest because their food intake has been shown to be related to that of their children, presumably due to their role as ‘gatekeepers’ of food in the household(Reference Campbell and Crawford72). Conforming to this, results from a study by Pinard et al. (Reference Pinard, Yaroch and Hart73), where home environmental contributors to obesity among children and adolescents aged 5–17 years were explored, indicated that mothers provide much of both the physical (availability/accessibility) and social (role modelling/policies/feeding styles) context in which child and adolescent food choices are made. Furthermore, the finding of perceived maternal HEG as a mediator of the relationship between family cohesion and adolescent vegetable consumption frequency is in line with the above-mentioned suggestions by Franko et al. (Reference Franko, Thompson and Bauserman52), as a high extent of family cohesion (as perceived by adolescents) may prepare the ground for effective maternal HEG. This could possibly reflect a mechanism where adolescents’ perception and appreciation of a cohesive family environment make them more open and responsive to maternal advice and guidance which, in turn, has a favourable effect on adolescent vegetable consumption. Hence, the findings from the present study shed light on potential mechanisms involved in the dynamic relationships between different family environmental levels and adolescent eating behaviours.

No associations were found between adolescents’ frequency of vegetable consumption and the parental level factors family dinner frequency and positive encouragement for vegetable consumption (PEV). The lack of association between vegetable consumption frequency and family dinner frequency may be explained by: (1) the fact that in Norway, vegetables are mostly eaten at dinner(Reference Myhre, Løken and Wandel13); and (2) the limited variation in family dinner frequency in the population of interest(Reference Hausken, Lie and Lien14,Reference Melbye, Øgaard and Øverby15) (the latter was confirmed in the present study). The lack of association seen for PEV could possibly be explained by a lack of parental encouragement specifically targeting vegetable consumption. Another possible explanation may be the adolescents’ lack of recognition of such encouragement. The relatively low mean values for perceived maternal and paternal PEV in the present sample (see Table 2) support this line of reasoning. It is worth noting that the way children perceive getting support from their parents and the way parents perceive offering their support may be very different from each other(Reference Dave, Evans and Condrasky38).

The lack of associations seen for the family level factors general family functioning and family conflict may be due to these factors’ ‘distance’ to the behaviour of interest, as more proximal environmental (e.g. home availability/accessibility of vegetables), individual (e.g. taste preferences) and social (e.g. peer influence) factors may play a greater role in influencing adolescent dietary behaviours(Reference Berge, Wall and Larson74). Nevertheless, both general family functioning and conflict may have an impact on the relationships between the more proximal factors and adolescent dietary behaviours, even if we were not able to detect it with the measures and analyses applied in the present study. For example, previous research has indicated that family conflict can significantly predict unhealthy dietary behaviours in adolescents(Reference Kalavana, Maes and De Gucht54,Reference Byely, Archibald and Graber75) .

Strengths and limitations

There has been a call for research relating fundamental socioemotional family characteristics to adolescent eating(Reference Larson, MacLehose and Fulkerson24). Furthermore, research applying ecological models to increase the understanding of how processes within the family may influence adolescent dietary behaviours have been requested(Reference Patrick, Hennessy and McSpadden76). Thus, one strength of the present study is that it adds to the current literature by its ecological approach in assessing influences of the family environment on adolescent vegetable consumption, thereby acknowledging the dynamic interplay of various factors and levels of the home food environment. More specifically, this work combines well-researched food parenting practices with less explored fundamental socioemotional family characteristics to uncover family environmental influences on adolescent vegetable consumption.

Among the limitations of the current work is the study’s cross-sectional nature, which hampers causal inferences. The self-report on all study variables is another limitation, increasing the risk of social desirability responses and common methods bias. The application of a frequency measure for vegetable consumption may also be considered a limitation because of its limited accuracy regarding the amount of vegetables ingested. However, such accuracy was not a key issue in the present work where the intention was to rank individuals according to their usual consumption of vegetables in terms of frequency (i.e. times/week). The lower respondent burden of frequency measures compared with more accurate methods such as repeated 24-h recalls or food diaries, and their ability to capture long-term dietary intake(Reference Cade, Thompson and Burley77), was also reasons for choosing this approach. Frequency measures appear to be feasible instruments in survey research aiming at exploring associations between dietary habits and a wide range of potential determinants without wearing out the respondents. Also, the use of a convenience sample with a large proportion of adolescents from highly educated households may limit the generalisability of our findings. Moreover, the relatively low explanatory power of the multiple linear regression model may be considered a limitation. However, the objective of this study was not to adapt models with the greatest possible explanatory power, but to describe associations between adolescents’ frequency of vegetable consumption, food parenting practices and socioemotional family characteristics – and to explore the potential processes involved.

Conclusions

The lower than recommended vegetable consumption in adolescents calls not only for studies and actions tailored directly towards this group of the population. Results from the present study suggest that perceived family cohesion may influence adolescent vegetable consumption both directly and indirectly (through maternal HEG), indicating that research and development of interventions directed towards the socioemotional aspects of the family environment may also be relevant. The large number of studies stating the importance of family meals suggests that developing interventions aimed at increasing the frequency of family meals could be a first step. Based on the findings from the current study, we suggest that a possible second step could be to provide parents with knowledge about how to create a socioemotional family environment that prepares the ground for positively framed food parenting practices and favourable eating behaviours. However, since knowledge is a prerequisite, but by itself not sufficient to induce behaviour change, parent and adolescent empowerment and other underexplored factors and processes that may help explain adolescent eating behaviours in general, and vegetable consumption in particular, should also be included in future research. Ultimately, understanding the factors and mechanisms at play is essential for the development, implementation and success of any intervention.

Acknowledgements

Acknowledgements: The authors would like to thank participating schools, students and staff participating in data collection. Financial support: The present work is part of The Family & Dietary habits project, which was funded by the Norwegian Research Council (grant number 213857/H10). Conflict of interest: There are no conflict of interest. Authorship: All authors participated in the development of the study framework and measurement instruments. M.B. and N.L. designed the study. M.B. and S.H.S. collected the data. E.L.M. performed statistical analyses and drafted the manuscript. All authors provided critical revision of the paper and approved the final manuscript. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving research participants were approved by the Norwegian Social Sciences Data Services. The Regional Committee for Medical and Health Research Ethics was also informed, but no approval was needed. Since the subjects were underaged, written informed consent was obtained from all subjects and their parents.

Supplementary material

For supplementary material accompanying this paper visit https://doi.org/10.1017/S1368980021001658

References

Lim, SS, Vos, T, Flaxman, AD et al. (2012) A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 22242260.CrossRefGoogle ScholarPubMed
Fismen, AS, Smith, ORF, Torsheim, T et al. (2016) Trends in food habits and their relation to socioeconomic status among Nordic adolescents 2001/2002–2009/2010. PLoS One 11, e0148541.CrossRefGoogle ScholarPubMed
Vereecken, C, Pedersen, TP, Ojala, K et al. (2015) Fruit and vegetable consumption trends among adolescents from 2002 to 2010 in 33 countries. Eur J Public Health 25, 1619.CrossRefGoogle ScholarPubMed
Golley, RK, Hendrie, GA & McNaughton, SA (2011) Scores on the dietary guideline index for children and adolescents are associated with nutrient intake and socio-economic position but not adiposity. J Nutr 141, 13401347.CrossRefGoogle Scholar
Striegel-Moore, RH & Bulik, CM (2007) Risk factors for eating disorders. Am Psychol 62, 181198.CrossRefGoogle ScholarPubMed
Lipsky, LM, Haynie, DL, Liu, D et al. (2015) Trajectories of eating behaviors in a nationally representative cohort of U.S. adolescents during the transition to young adulthood. Int J Behav Nutr Phys Act. Published online: 4 November 2015. doi: 10.1186/s12966–015–0298-x.CrossRefGoogle Scholar
Bere, E, Hilsen, M & Klepp, KI (2010) Effect of the nationwide free school fruit scheme in Norway. Br J Nutr 104, 589594.CrossRefGoogle ScholarPubMed
Hovdenak, IM, Bere, E & Stea, TH (2019) Time trends (1995–2008) in dietary habits among adolescents in relation to the Norwegian school fruit scheme: the HUNT study. Nutr J. Published online: 20 November 2019. doi: 10.1186/s12937–019–0501-z.CrossRefGoogle Scholar
Øvrum, A & Bere, E (2014) Evaluating free school fruit: results from a natural experiment in Norway with representative data. Public Health Nutr 17, 12241231.CrossRefGoogle ScholarPubMed
Evans, C, Christian, M, Cleghorn, C et al. (2012) Systematic review and meta-analysis of school-based interventions to improve daily fruit and vegetable intake in children aged 5 to 12 years. Am J Clin Nutr 96, 889901.CrossRefGoogle Scholar
Mäkelä, J, Kjærnes, U, Ekström, MP et al. (1999) Nordic meals: methodological notes on a comparative survey. Appetite 32, 7379.CrossRefGoogle ScholarPubMed
Bere, E, Veierød, MB & Klepp, KI (2005) The Norwegian school fruit programme: evaluating paid v. no-cost subscriptions. Prev Med 41, 463470.CrossRefGoogle Scholar
Myhre, JB, Løken, EB, Wandel, M et al. (2015) Meal types as sources for intakes of fruits, vegetables, fish and whole grains among Norwegian adults. Public Health Nutr 18, 20112021.CrossRefGoogle ScholarPubMed
Hausken, SE, Lie, HC, Lien, N et al. (2019) The reliability of the general functioning scale in Norwegian 13–15-year-old adolescents and association with family dinner frequency. Nutr J. Published online: 27 March 2019. doi: 10.1186/s12937–019–0447–1.CrossRefGoogle Scholar
Melbye, EL, Øgaard, T, Øverby, NC et al. (2013) Parental food-related behaviors and family meal frequencies: associations in Norwegian dyads of parents and preadolescent children. BMC Public Health. Published online: 10 September 2013. doi: 10.1186/1471–2458–13–820.CrossRefGoogle Scholar
Berge, JM (2009) A review of familial correlates of child and adolescent obesity: what has the 21st century taught us so far? Int J Adolesc Med Health 21, 457483.CrossRefGoogle ScholarPubMed
Fulkerson, JA, Larson, N, Horning, M et al. (2014) A review of associations between family or shared meal frequency and dietary and weight status outcomes across the lifespan. J Nutr Educ Behav 46, 219.CrossRefGoogle ScholarPubMed
Videon, TM & Manning, CK (2003) Influences on adolescent eating patterns: the importance of family meals. J Adolesc Health 32, 365373.CrossRefGoogle ScholarPubMed
Totland, TH, Knudsen, MD, Paulsen, MM et al. (2017) Correlates of irregular family meal patterns among 11-year-old children from the Pro Children study. Food Nutr Res. Published online: 22 June 2017. doi: 10.1080/16546628.2017.1339554.CrossRefGoogle Scholar
Pearson, N, Atkin, AJ, Biddle, SJ et al. (2010) Parenting styles, family structure and adolescent dietary behaviour. Public Health Nutr 13, 12451253.CrossRefGoogle ScholarPubMed
Gebremariam, MK, Henjum, S, Terragni, L et al. (2016) Correlates of fruit, vegetable, soft drink, and snack intake among adolescents: the ESSENS study. Food Nutr Res. Published online: 20 September 2016. doi: 10.3402/fnr.v60.32512.CrossRefGoogle Scholar
Gebremariam, MK, Lien, N, Torheim, LE et al. (2016) Perceived rules and accessibility: measurement and mediating role in the association between parental education and vegetable and soft drink intake. Nutr J. Published online: 17 August 2016. doi: 10.1186/s12937–016–0196–3.CrossRefGoogle Scholar
Pearson, N, Biddle, SJ & Gorely, T (2009) Family correlates of breakfast consumption among children and adolescents. A systematic review. Appetite 52, 17.CrossRefGoogle ScholarPubMed
Larson, N, MacLehose, R, Fulkerson, JA et al. (2013) Eating breakfast and dinner together as a family: associations with sociodemographic characteristics and implications for diet quality and weight status. J Acad Nutr Diet 113, 16011609.CrossRefGoogle ScholarPubMed
Balistreri, KS & Alvira-Hammond, M (2016) Adverse childhood experiences, family functioning and adolescent health and emotional well-being. Public Health 132, 7278.CrossRefGoogle ScholarPubMed
Brown, SL (2006) Family structure transitions and adolescent well-being. Demography 43, 447461.CrossRefGoogle ScholarPubMed
Shek, DT (2002) Family functioning and psychological well-being, school adjustment, and problem behavior in Chinese adolescents with and without economic disadvantage. J Genet Psychol 163, 497502.CrossRefGoogle ScholarPubMed
Kitzmann, KM & Beech, BM (2006) Family-based interventions for pediatric obesity: methodological and conceptual challenges from family psychology. J Fam Psychol 20, 175189.CrossRefGoogle ScholarPubMed
Sleddens, EF, Gerards, SM, Thijs, C et al. (2011) General parenting, childhood overweight and obesity-inducing behaviors: a review. Int J Pediatr Obes 6, e12e27.CrossRefGoogle ScholarPubMed
Bjelland, M, Hausken, SE, Sleddens, EF et al. (2014) Development of family and dietary habits questionnaires: the assessment of family processes, dietary habits and adolescents’ impulsiveness in Norwegian adolescents and their parents. Int J Behav Nutr Phys Act. Published online: 15 October 2014. doi: 10.1186/s12966–014–0130-z.CrossRefGoogle Scholar
Lien, N, Bjelland, M, Bergh, IH et al. (2010) Design of a 20-month comprehensive, multicomponent school-based randomised trial to promote healthy weight development among 11–13 year olds: the health in adolescents study. Scand J Public Health 38, 3851.CrossRefGoogle ScholarPubMed
Andersen, L, Bere, E, Kolbjornsen, N et al. (2004) Validity and reproducibility of self-reported intake of fruit and vegetable among 6th graders. Eur J Clin Nutr 58, 771777.CrossRefGoogle ScholarPubMed
Bere, E & Klepp, K (2005) Changes in accessibility and preferences predict children’s future fruit and vegetable intake. Int J Behav Nutr Phys Act. Published online: 10 October 2005. doi: 10.1186/1479–5868–2–15.CrossRefGoogle Scholar
Melbye, EL, Øverby, NC & Øgaard, T (2012) Child consumption of fruit and vegetables: the roles of child cognitions and parental feeding practices. Public Health Nutr 15, 10471055.CrossRefGoogle ScholarPubMed
Haszard, JJ, Williams, SM, Dawson, AM et al. (2013) Factor analysis of the Comprehensive Feeding Practices Questionnaire in a large sample of children. Appetite 62, 110118.CrossRefGoogle Scholar
Musher-Eizenman, D & Holub, S (2007) Comprehensive feeding practices questionnaire: validation of a new measure of parental feeding practices. J Pediatr Psychol 32, 960972.CrossRefGoogle ScholarPubMed
Melbye, EL, Øgaard, T & Øverby, NC (2011) Validation of the comprehensive feeding practices questionnaire with parents of 10-to-12-year-olds. BMC Med Res Methodol. Published online: 09 August 2011. doi: 10.1186/1471–2288–11–113.CrossRefGoogle Scholar
Dave, JM, Evans, AE, Condrasky, MD et al. (2012) Parent-reported social support for child’s fruit and vegetable intake: validity of Measures. J Nutr Edu Behav 44, 132139.CrossRefGoogle ScholarPubMed
Melbye, EL, Øgaard, T & Øverby, NC (2013) Associations between parental feeding practices and child vegetable consumption: mediation by child cognitions? Appetite 69, 2330.CrossRefGoogle ScholarPubMed
Young, EM, Fors, SW & Hayes, DM (2004) Associations between perceived parent behaviors and middle school student fruit and vegetable consumption. J Nutr Edu Behav 36, 212.CrossRefGoogle ScholarPubMed
Bowen, M (1978) Family Therapy in Clinical Practice. New York: Jason Aronson, Inc.Google Scholar
Epstein, NB, Baldwin, LM & Bishop, DS (1983) The McMaster family assessment device. J Marital Fam Ther 9, 171180.CrossRefGoogle Scholar
Byles, J, Byrne, C, Boyle, MH et al. (1988) Ontario child health study: reliability and validity of the general functioning subscale of the Mcmaster family assessment device. Fam Process 27, 97104.CrossRefGoogle ScholarPubMed
Ridenour, TA, Daley, J & Reich, W (1999) Factor analyses of the family assessment device. Fam Process 38, 497510.CrossRefGoogle ScholarPubMed
Kazarian, SS (2010) Cultural appropriateness of the Family Assessment Device (FAD) in the case of ethnic Armenian adolescents in Lebanon. Int J Soc Psychiatry 56, 230238.CrossRefGoogle ScholarPubMed
Pedersen, MA, Kristensen, LJ, Sildorf, SM et al. (2019) Assessment of family functioning in families with a child diagnosed with type 1 diabetes: validation and clinical relevance of the general functioning subscale of the McMaster family assessment device. Pediatr Diabetes 20, 785793.Google ScholarPubMed
Shek, DT (2001) The general functioning scale of the Family Assessment Device: does it work with Chinese adolescents? J Clin Psychol 57, 15031516.CrossRefGoogle ScholarPubMed
Moos, RH & Moos, BS (1987) Family environment scale. In Handbook of Measurements for Marriage and Family Therapy, pp. 8286 [Fredman, N & Sherman, R, editors]. New York: Brunner/Mazel Inc.Google Scholar
Ackard, DM, Neumark-Sztainer, D, Story, M et al. (2006) Parent - child connectedness and behavioral and emotional health among adolescents. Am J Prev Med 30, 5966.CrossRefGoogle ScholarPubMed
Franko, DL, Thompson, D, Affenito, SG et al. (2008) What mediates the relationship between family meals and adolescent health issues. Health Psychol 27, 109117.CrossRefGoogle ScholarPubMed
Zdanowicz, N, Janne, P & Reynaert, C (2004) Family, health, and adolescence. Psychosomatics 45, 500507.CrossRefGoogle ScholarPubMed
Franko, DL, Thompson, D, Bauserman, R et al. (2008) What’s love got to do with it? Family cohesion and healthy eating behaviors in adolescent girls. Int J Eat Disord 41, 360367.CrossRefGoogle Scholar
Haerens, L, Craeynest, M, Deforche, B et al. (2008) The contribution of psychosocial and home environmental factors in explaining eating behaviours in adolescents. Eur J Clin Nutr 62, 5159.CrossRefGoogle ScholarPubMed
Kalavana, TV, Maes, S & De Gucht, V (2010) Interpersonal and self-regulation determinants of healthy and unhealthy eating behavior in adolescents. J Health Psychol 15, 4452.CrossRefGoogle ScholarPubMed
Moos, RH (1990) Conceptual and empirical approaches to developing family-based assessment procedures: resolving the case of the Family Environment Scale. Fam Process 29, 199208.CrossRefGoogle ScholarPubMed
Repetti, RL, Taylor, SE & Seeman, TE (2002) Risky families: family social environments and the mental and physical health of offspring. Psychol Bull 128, 330366.CrossRefGoogle ScholarPubMed
Ross, RD, Marrinan, S, Schattner, S et al. (1999) The relationship between perceived family environment and psychological wellbeing: mother, father, and adolescent reports. Aust Psychol 34, 5863.CrossRefGoogle Scholar
Wadsworth, ME & Compas, BE (2002) Coping with family conflict and economic strain: the adolescent perspective. J Res Adolesc 12, 243274.CrossRefGoogle Scholar
Schuetzmann, M, Richter-Appelt, H, Schulte-Markwort, M et al. (2008) Associations among the perceived parent–child relationship, eating behavior, and body weight in preadolescents: results from a community-based sample. J Pediatr Psychol 33, 772782.CrossRefGoogle ScholarPubMed
Charalampous, K, Kokkinos, CM & Panayiotou, G (2013) The Family Environment Scale: resolving psychometric problems through an examination of a Greek translation. Int J Edu Psychol Assess 13, 8199.Google Scholar
Xu, Y, Boyd, RC, Butler, L et al. (2017) Associations of parent-adolescent discrepancies in family cohesion and conflict with adolescent impairment. J Child Fam Stud 26, 33603369.CrossRefGoogle Scholar
Kline, RB (2005) Methodology in the Social Sciences. Principles and Practice of Structural Equation Modeling, 2nd ed. New York: Guilford Press.Google Scholar
Field, A (2013) Discovering Statistics using IBM SPSS Statistics, 4th ed. London: Sage publications Ltd.Google Scholar
Singh, AS, Vik, FN, Chinapaw, MJ et al. (2011) Test-retest reliability and construct validity of the ENERGY-child questionnaire on energy balance-related behaviours and their potential determinants: the ENERGY-project. Int J Behav Nutr Phys Act. Published online: 09 December 2011. doi: 10.1186/1479–5868–8–136.CrossRefGoogle Scholar
Hayes, AF (2017) Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach, 2nd ed. New York: Guilford Press.Google Scholar
Cerin, E & MacKinnon, DP (2009) A commentary on current practice in mediating variable analyses in behavioural nutrition and physical activity. Public Health Nutr 12, 11821188.CrossRefGoogle ScholarPubMed
Fritz, MS, Taylor, AB & Mackinnon, DP (2012) Explanation of two anomalous results in statistical mediation analysis. Multivar Behav Res 47, 6187.CrossRefGoogle ScholarPubMed
Gelfand, LA, Mensinger, JL & Tenhave, T (2009) Mediation analysis: a retrospective snapshot of practice and more recent directions. J Gen Psychol 136, 153178.CrossRefGoogle ScholarPubMed
Moore, J & Harré, N (2007) Eating and activity: the importance of family and environment. Health Promot J Aust 18, 143148.CrossRefGoogle ScholarPubMed
Couch, SC, Glanz, K, Zhou, C et al. (2014) Home food environment in relation to children’s diet quality and weight status. J Acad Nutr Diet 114, 15691579.CrossRefGoogle ScholarPubMed
Melbye, EL & Hansen, H (2015) Promotion and prevention focused feeding strategies: exploring the effects on healthy and unhealthy child eating. BioMed Res Int. Published online: 25 August 2015. doi: 10.1155/2015/306306.CrossRefGoogle Scholar
Campbell, K & Crawford, D (2007) Associations between the home food environment and obesity-promoting eating behaviors in adolescence. Obesity 15, 719730.CrossRefGoogle ScholarPubMed
Pinard, CA, Yaroch, AL, Hart, MH et al. (2014) The validity and reliability of the Comprehensive Home Environment Survey (CHES). Health Promot Pract 15, 109117.CrossRefGoogle Scholar
Berge, JM, Wall, M, Larson, N et al. (2013) Family functioning: associations with weight status, eating behaviors, and physical activity in adolescents. J Adolesc Health 52, 351357.CrossRefGoogle ScholarPubMed
Byely, L, Archibald, AB, Graber, J et al. (2000) A prospective study of familial and social influences on girls’ body image and dieting. Int J Eat Disord 28, 155164.3.0.CO;2-K>CrossRefGoogle ScholarPubMed
Patrick, H, Hennessy, E, McSpadden, K et al. (2013) Parenting styles and practices in children’s obesogenic behaviors: scientific gaps and future research directions. Child Obes 9, 7386.CrossRefGoogle ScholarPubMed
Cade, J, Thompson, R, Burley, V et al. (2002) Development, validation and utilisation of food-frequency questionnaires – a review. Public Health Nutr 5, 567587.CrossRefGoogle ScholarPubMed
Figure 0

Table 1 Frequencies for categorical variables (n 440)

Figure 1

Table 2 Means, standard deviations, skewness, kurtosis, Cronbach’s α and intra-class correlations (ICC) for rating scale variables

Figure 2

Table 3 Regression coefficients (β) and variance explained (R2) for multiple linear regression on vegetable consumption frequency

Figure 3

Fig. 1. Path diagram for modelling family cohesion as a predictor of adolescents’ frequency of vegetable consumption, partly mediated by maternal healthy eating guidance (HEG). Coefficient estimates (b) and statistical significance tests (P-values and CI) were obtained using the PROCESS script for SPSS

Supplementary material: File

Melbye et al. supplementary material

Melbye et al. supplementary material

Download Melbye et al. supplementary material(File)
File 16.6 KB