Journal of Public Health Advance Access originally published online on May 16, 2007
Journal of Public Health 2007 29(2):123-131; doi:10.1093/pubmed/fdm023
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Community beliefs about childhood obesity: its causes, consequences and potential solutions
Tanya Covic, Coordinator of Research Higher Degrees and Lecturer in Psychology1,
Louise Roufeil, Program Director, NSW Central West Division of GP and Adjunct Lecturer in Psychology2,3
Suzanne Dziurawiec, Coordinator of Research Higher Degrees and Senior Lecturer in Psychology4
1 School of Psychology, University of Western Sydney, Locked Bag 1797, Penrith South DC, NSW 1797, Australia
2 NSW Central West Division of General Practice, Bathurst, NSW, Australia
3 School of Social Sciences and Liberal Studies, Charles Sturt University, Bathurst, NSW, Australia
4 School of Psychology, Murdoch University, Perth, WA, Australia
Address correspondence to Tanya Covic, E-mail: t.covic{at}uws.edu.au
Background The objective of this study was to explore community beliefs about the causes, consequences and potential solutions of childhood obesity.
Methods A convenience sample of 434 adults (41.2 ± 13.3 years; 61% parents) in New South Wales, Australia, was surveyed using a newly developed childhood obesity scale.
Results Five causal (emotional eating; eating habits and food knowledge; environmental dysfunction; abundance of contemporary lifestyle; cost of contemporary lifestyle), four consequences (known consequences of obesity; behavioural consequences; social consequences; less-known physical consequences) and three potential solutions factors (parental actions; professional assistance; limiting behaviours) were identified. Parents did not differ from non-parents across the 12 factors nor were there any differences based on the level of education. There were, however, gender differences across two causal factors (emotional eating and abundance of contemporary lifestyle) and two consequences factors (behavioural consequences and social consequences), with females endorsing all four factors more strongly than males.
Conclusions The results of this study suggest that this sample was aware of the complex nature of childhood obesity in terms of its causes, consequences and a range of potential solutions, but they endorsed more family rather than community-based interventions.
Keywords: beliefs, childhood obesity, general community
Increasing trends in the prevalence of childhood obesity have been reported throughout the developed and developing world.1 In addition, the distribution of body mass index has shifted so that the heaviest children have become heavier, thus further compounding the potential physical and psychosocial consequences of obesity.2 A comprehensive review of the evidence regarding the physical and psychosocial consequences of childhood obesity was described by Lobstein et al.3
While childhood obesity has become a focus of considerable attention by researchers, media, health professionals and government authorities, it is of considerable relevance, especially to community-based intervention programmes, to understand how the general community perceives childhood obesity. The understanding of those lay beliefs is essential for engaging individuals in community education programmes and prevention/intervention programmes for childhood obesity.4
Only a handful of studies have considered community beliefs about childhood obesity. These were limited to exploring views of specific groups such as health professionals and care-givers,5,6 and focusing on individual causal factors such as parental control5,7 or on specific outcomes (i.e. obesity-related health risk).5,6
Myers and Vargas5 found that 35% of parents of obese children did not believe that their child was overweight, but 78% were concerned about heart disease as a possible consequence of childhood obesity. Furthermore, 19% of the health professional interviewers also failed to recognize obese children. Perceptions of obesity are also culturally influenced. Myers and Vargas' study was based on an Hispanic population, whereas in Young-Hyman et al.'s6 study based on African-American population, substantially fewer parents (44%) perceived their child's weight as a possible health problem.
To date, only one study4 has explored a broader community's views in terms of the causes and preventative measures regarding childhood obesity, in a sample of 315 adults (53% parents) from Melbourne, Australia, using a questionnaire based on a literature review on the causes and prevention of obesity and with reference to primary school children. The public's perception of the most important causes of childhood obesity related to over-consumption of unhealthy foods, parental responsibility, modern technology and the mass media. In terms of preventative measures, the public-endorsed actions aimed at children (i.e. school-based health foods and daily physical education) more so than changes at the government level (i.e. banning high fat foods advertisement, more safe cycling and walking tracks). While the findings of this study suggest that community views regarding childhood obesity are closely related to the scientific views, perceptions of the consequences and potential solutions were not explored.
The current study, therefore, aims to broaden the examination of community beliefs regarding the causes, consequences and potential solutions of childhood obesity and to develop a questionnaire that is primarily driven by the current community views about childhood obesity.
Specifically, this study aims to
- explore community beliefs about the causes, consequences and potential solutions of childhood obesity in a broad sample of the Australian population
- explore the relationship between demographic variables and beliefs about the causes, consequences and potential solutions of childhood obesity.
 |
Methods |
|---|
 TOP Methods ResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
A total of 434 adults (78% response rate) volunteered to participate in this study. Participants were recruited by a small group of seven research trainees who made personal approaches to prospective volunteers, primarily through their local schools and neighbourhoods. The communities involved comprised small rural towns (population <10 000), medium sized regional centres and suburbs surrounding a major urban centre, mostly in New South Wales, Australia. The relevant university's ethics committee approval was obtained for this study. Participants' consent was implied in the completion and return of the questionnaire. The questionnaire consisted of the following two parts:
Demographic details: age, gender, education, marital and parental status, as well as items about awareness of media reports on childhood obesity.
Childhood obesity questionnaire (COQ): this scale was developed as part of this study and consists of three subscales of causes, consequences and potential solutions statements which are rated on a five-point Likert scale ranging from strongly disagree to strongly agree (1–5). The items were generated based on a pilot study with a community sample (n = 35) and university students (n = 7). The pilot study participants were asked to respond to three open-ended questions on what they may think are the causes, consequences and potential solutions to childhood obesity. The generated statements were pooled together, duplicated and similar responses were removed and expressions uniformed. These items were then tested with a different sample (n = 10) for clarity prior to the proper study. A total of 46 causes, 48 consequences and 21 potential solutions items were included in the final questionnaire. The three subscales were preceded with an instruction to read each item and indicate a level of agreement/disagreement in terms of the causes, consequences and potential solutions of childhood obesity. The items are listed in the Tables 2–4.
Analyses were conducted using Statistical Package for the Social Sciences (SPSS) 12.0 for Windows. Basic frequency counts were performed to describe the distribution of responses for each item across all subscales. Factor analysis, a data reduction technique that allows for identification of clusters of items that represent specific dimensions of a measured construct, was performed separately for the causes, consequences and potential solutions items using principal components analysis with varimax rotation. Items that loaded <0.40 were removed, as recommended by Tabachnick and Fidell.8 The choice of how many factors to extract was made on the basis of Scree plots.9 Items that loaded significantly on more than one factor were retained on the factor where they loaded higher. For each participant, items loading on each factor were summed and divided by the number of items on the factor to create subscale scores. Individual items were ranked based on their mean level of endorsement scores. Cronbach's alphas were calculated to determine the internal reliability of the resulting factors or subscales. Relationships between demographic variables and beliefs were assessed using t-tests and, where required, two-way ANOVA. Two-way ANOVA (sex by parental status and sex by education level) were conducted on each of the emergent subscales and examined using an alpha level of 0.01. A more stringent alpha (0.01) was used due to multiple tests and hence need to reduce Type I error rate.8
|
Results |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
Descriptive data
The demographic profile of the 434 participants is presented in Table 1. The majority of participants were female (70%), parents (65%) and living with a partner (71%). The mean age was 41.2 years (range, 18–85 years; SD = 13.3), and 47% of participants were tertiary educated. The majority (95%) of participants were aware of media reports on childhood obesity. The majority of participants had three or fewer children (31% has three children, 39% two children and 16% one child). The majority of children were aged 18 and younger (54% of first child, 57% of second and third child).
|
Beliefs about causes of childhood obesity
Of the 46 causal items, the most strongly endorsed item, as indicated by the highest mean scores, was ‘family eating habits’ followed by ‘eating lots of fatty foods’, ‘overeating’ and ‘easy access to fast food’ (Table 2).
|
The five factors derived from the principal components analysis of beliefs about the causal factors of childhood obesity are presented in Table 2, along with their respective item loadings, internal reliabilities and percentages of explained variance. The first factor comprised eight items involving emotional aspects of eating and was labelled as emotional eating. The second factor, labelled as eating habits and food knowledge, included 10 items about unbalanced eating and deficits in parental and child knowledge or behaviour. The third factor, environmental dysfunction, included 6 items broadly concerned with past or current problems in the family. The fourth factor included 6 items involving the excesses of modern life and was labelled as abundance of contemporary lifestyle. The fifth and final factor, cost of contemporary lifestyle, consisted of 7 items concerned with the monetary, temporal and social costs of modern life. These five factors together accounted for 42% of the variance.
It is worth noting that two items that were ranked in the top 10 (‘promotional gimmicks used by fast food companies’ and ‘eating too many processed foods’) did not load >0.40 on any factor and were therefore excluded.
Factors scores were calculated by adding the responses to items loading at 0.40 or higher on each factor, and then dividing by the number of items comprising each factor. The mean factor scores are also shown in Table 2. Of the five factors, the most endorsed was Factor 2, eating habits and food knowledge (mean = 4.00; SD = 0.65), followed by Factor 4, abundance of contemporary lifestyle (mean = 3.97; SD = 0.73), and then Factor 1, emotional eating (mean = 3.59; SD = 0.63), Factor 3, environmental dysfunction (mean = 3.29; SD = 0.80) and Factor 5, cost of contemporary lifestyle (mean = 3.09; SD = 1.18).
Beliefs about consequences of childhood obesity
In terms of the 48 consequences of childhood obesity, the most strongly endorsed item by the participants was ‘have poor fitness’ followed by ‘heart disease’, ‘adulthood obesity’ and ‘high cholesterol’, respectively. Item rankings are given in Table 3.
|
The consequences items were also subjected to a principal component analysis and varimax rotation. On the basis of the Scree plot, four factors were extracted and together accounted for 43.6% of the variance in the scores. Three items did not reach the cut-off value of 0.40, but it is worth noting that one of those items (‘more prone to diabetes in later life’) was ranked as the seventh most endorsed item in the consequences subscale. The four factors, along with their respective item loadings, internal reliabilities and percentages of explained variance, are shown in Table 3.
The first factor, labelled as known consequences of obesity, included 14 items, which primarily concerned with the well-known physical consequences of obesity. The second factor, which included 10 items, was labelled as behavioural consequences. The third factor, social consequences, included 12 items. The fourth and final factor, labelled as less-known physical consequences, comprised 9 items.
Of the four factors, the most endorsed was Factor 1, known consequences of obesity (mean = 3.87; SD = 0.57), followed by the Factor 3, social consequences (mean = 3.70; SD = 0.70), and then Factor 2, behavioural consequences (mean = 3.70; SD = 0.70), and Factor 4, less-known physical consequences (mean = 3.35; SD = 0.59).
Beliefs about potential solutions to childhood obesity
The most strongly endorsed of the 21 potential solutions items was ‘more physical exercise’, followed by ‘parents provide healthy food choices’, ‘educate child about healthy eating’ and ‘parents provide more positive role models for healthy life styles’ (Table 4).
|
The 21 items yielded 3 factors which accounted for 44.86% of the variance in the scores. These factors, along with their respective item loadings, their internal reliabilities and the percentages of explained variance, are shown in Table 4. The first factor, labelled as parental actions, contained 10 items that emphasized parental responsibility for action. The second factor, labelled as professional assistance, contained 8 items concerned with seeking professional assistance. The third and last factor consisted of only 3 items concerned with setting limits and was labelled as limiting behaviour.
In terms of the mean factor scores, Factor 1, parental actions, was most endorsed (mean = 4.19; SD = 0.37), followed by Factor 3, limiting behaviour (mean = 4.14; SD = 0.53), and lastly Factor 2, professional assistance (mean = 3.20; SD = 0.53).
Relationship between demographic variables and beliefs
Independent t-tests were conducted to assess if parents differed across the three beliefs scales from non-parents. No significant differences were found. Likewise, no significant difference was found when beliefs were compared in terms of education (tertiary degree versus no tertiary degree). There were, however, differences in terms of the age of the children, with participants who had older children (>19 years old) endorsing the abundance of contemporary lifestyle (t = –3.45, P < 0.01) significantly more than the parents who had younger children (t = –3.45, P < 0.01). This was further evident in the age comparison across the sample (both parents and non-parents) with the only significant age difference being in the higher endorsement of the abundance of contemporary lifestyle (t = –3.13, P < 0.01) by the participants over the age of 41 compared with those aged 41 and younger.
Next, gender differences were explored across the 12 factors. On the basis of an alpha level of 0.01, females endorsed four factors significantly more than males. These refer to two causal factors, emotional eating (t = –5.79, P < 0.01) and abundance of contemporary lifestyle (t = –2.63, P < 0.01), and two consequences factors, behavioural consequences (t = –3.17, P < 0.01) and social consequences (t = –4.27, P < 0.01).
In terms of the interaction between gender (male versus female participants) and education (non-tertiary educated versus tertiary educated), there were no significant differences across the 12 factors.
|
Discussion |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
Main finding of this study
This study examined community beliefs about childhood obesity in terms of causes, consequences and possible solutions. The main findings of this study are that this sample was well aware of the complexity of obesity in children, but appeared to perceive it as an issue to be addressed at the family level, rather than a broader social level.
What is already known on this topic?
Only a few studies have explored the community's perceptions about childhood obesity, and these have been limited to health professional groups and parents5,6 and to causal factors5,7 or specific health outcomes.5,6 The only other general community-based study4 was limited to causes and preventative measures regarding the childhood obesity.
What this study adds
Specifically, the most endorsed causes of childhood obesity in our study related to poor eating habits, a lack of parental diet supervision and sedentary lifestyle dominated by computers, TV and easy access to fast food and snacks. These views correspond to research findings on the causes of childhood obesity10 and these reported by the general community.4 It is interesting, however, to note that ‘poor parental supervision of diet’ was ranked the seventh most endorsed item among causes of childhood obesity, given Robinson et al.7 finding that parental supervision was not related to children's weight status. Although this view may not be empirically supported, it seems that it is strongly held by the community.
In terms of the consequences of childhood obesity, the participants were well aware of the common physical consequences, including heart disease, adulthood obesity and back-problems but also low self-esteem. More surprisingly and contrary to some other studies,5 the adults in our sample were also knowledgeable about social and behavioural consequences, such as obese children being bullied and teased at school, poor fitness levels and disturbed body image, as well as behavioural issues, such as sleep problems and underachieving. While in Myers and Vargas's5 study only 3% of parents identified low self-esteem as an issue for obese children and only 11% considered that obese children would have little energy to play, in our study 56% endorsed (agree-strongly agree) low self-esteem as a consequence of childhood obesity and 91% ranked poor fitness (which influences ability to play) as the single most significant consequence of childhood obesity. These differences may be due to the samples of the studies. Myers and Vargas5 utilized a sample of 200 parents of obese pre-school children, whereas our study used a considerably larger (n = 434) community-based (parents and non-parents) sample. Furthermore, the differences may be temporal with the 5-year gap between the two studies reflecting a dramatic increase in media and community awareness about childhood obesity, as is evident in 95% of our participants reporting awareness of media coverage of childhood obesity.
As for the potential solutions of childhood obesity, the participants strongly believed that the solution to the problem rests with parents, through encouraging physical exercise (98%), healthy food choice (97%) and education of children about healthy eating (96%). The focus on parents reflects, as Young-Hyman et al.6 (p. 241) state, that ‘Unlike adults, children are only partially responsible for their eating and exercise choices’. It is, however, of concern to note that the blame and solution to childhood obesity is perceived to rest within the family to the exclusion of the broader society (i.e. media, fast food manufacturers, government regulations, etc.). Therein lies potential for parental guilt and anxiety regarding their children's weight status. Moreover, it is unlikely that a solution to the complex problem of childhood obesity is solely to be found at the individual and familial level. For example, in a recent large Australian survey of children, it has been found that walking or cycling to school has dropped by 90%, largely due to safety fears and fewer secondary students attending their local school.11 Broader social and political action is clearly necessary.
Another important outcome of the current study was the relative lack of emphasis placed on the health professionals as a resource in treating childhood obesity. Health professionals, such as family general practitioners (GP), may start with simple initiatives such as regularly weighing children, which, it has been reported, less than 50% of GPs actually do.12
In terms of potential demographic influences, no education differences were detected across the beliefs factors. Gender differences were found across two causal factors, emotional eating and abundance of contemporary lifestyle, and two consequences factors, behavioural consequences and social consequences, all of which were endorsed more strongly by females than males. Similar gender differences were reported by Hardus et al.4 These findings suggest that women are still traditionally spending more time involved in child-care,13 and therefore may be more aware of the context (i.e. technology use, non-physical extracurricular activities) and consequences of childhood obesity. Furthermore, age differences were found only in reference to abundance of contemporary lifestyle, with older participants (parents and non-parents) endorsing it more so than the younger participants.
Limitations of this study
The findings of this study should be interpreted with caution due to the following limitations. First, the sampling method was not truly random and resulted in an over-representation of tertiary-educated women.14 However, the education level of the sample as a whole was slightly lower, and therefore somewhat more representative, than that reported in the only other community-based study of the Australian public's perceptions of childhood obesity4 and, in contrast to that study, younger persons were not over represented. Second, the scale items are not a comprehensive set in terms of a scientific summary of causes, consequences and potential solutions of childhood obesity. However, their limited range is an informative reflection of the actual levels of awareness in the general community and these findings should be taken into account when developing community-based educational interventions.
|
Conclusions |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
In summary, the current study of a large, general community-based sample showed that this sample was aware of the complex nature of childhood obesity in terms of causes, consequences and a range of potential solutions. However, this study also highlighted that this sample saw the childhood obesity problem predominantly as a family rather than a social issue. These preliminary findings may be considered in the development and implementation of community-based educational programmes.
|
Conflict of interest statement |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
None declared.
|
Acknowledgements |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
The authors would like to thank Susan Doughty, Amy Goodwin, Sue Hawkins, Wanda Little, Lisa Terrill, Jenny Tudberry and Stacey Turner for data collection and input into this study, and David Wyett for data preparation.
|
References |
|---|
|
TOPMethodsResultsDiscussionConclusionsConflict of interest statementAcknowledgementsReferences |
|---|
- Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity: public-health crisis, common sense cure. Lancet (2002) 360:473–82.[CrossRef][Web of Science][Medline]
- Flegal KM, Troiano RP. Changes in the distribution of body mass index of adults and children in the US population. Int J Obes Relat Metab Disord (2000) 24:807–18.[CrossRef][Web of Science][Medline]
- Lobstein T, Baur L, Uauy R. Obesity in children and young people: a crisis in public health. Obes Rev (2004) 5((Suppl 1)):4–85.[CrossRef][Medline]
- Hardus PM, van Vuuren CL, Crawford D, et al. Public perceptions of the causes and prevention of obesity among primary school children. Int J Obes Relat Metab Disord (2003) 27:1465–71.[CrossRef][Web of Science][Medline]
- Myers S, Vargas Z. Parental perceptions of the preschool obese child. Pediatr Nurs (2000) 26:23–30.[Medline]
- Young-Hyman D, Herman LJ, Scott DL, et al. Care giver perception of children's obesity-related health risk: a study of African American families. Obes Res (2000) 8:241–8.[Web of Science][Medline]
- Robinson TN. Television viewing and childhood obesity. Paediatr Clin N Am (2001) 19:169–81.
- Tabachnick BG, Fidell LS. Using Multivariate Statistics (2001) Boston: Allyn and Bacon.
- Cattell RR. The scree test for number of factors. Mulitvariate Behav Res (1966) 1:245–76.[CrossRef]
- Jeffery RW. Public health strategies for obesity treatment and prevention. Am J Health Behav (2001) 25:252–9.[Web of Science][Medline]
- Booth M, Okely AD, Denney-Wilson E, et al. NSW schools physical activity and nutrition survey (SPANS) 2004: Summary Report. (2006) Sydney: NSW Department of Health.
- Gerner B, McCallum Z, Sheehan J, et al. Are general practitioners equipped to detect child overweight/obesity? Survey and audit. J Paediatr Child Health (2006) 42:206–11.[CrossRef][Web of Science][Medline]
- Bittman M. Juggling Time: How Australian Families Use Time (1991) Canberra: Department of Prime Minister and Cabinet, Office of the Status of Women.
- Australian Bureau of Statistics. Population by age and sex and education and work, Australia, 2005. (2005) Australia: Australian Bureau of Statistics. Statistics 3201.0/ 6227.0.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||