Hostname: page-component-848d4c4894-cjp7w Total loading time: 0 Render date: 2024-06-18T03:44:33.194Z Has data issue: false hasContentIssue false

Behavioural and environmental risk factors associated with primary schoolchildren’s overweight and obesity in urban Indonesia

Published online by Cambridge University Press:  04 May 2023

Margarita de Vries Mecheva*
Affiliation:
The International Institute of Social Studies of Erasmus University Rotterdam, 2518 AX The Hague, the Netherlands
Matthias Rieger
Affiliation:
The International Institute of Social Studies of Erasmus University Rotterdam, 2518 AX The Hague, the Netherlands
Robert Sparrow
Affiliation:
The International Institute of Social Studies of Erasmus University Rotterdam, 2518 AX The Hague, the Netherlands Development Economics Group, Wageningen University, Wageningen, the Netherlands
Erfi Prafiantini
Affiliation:
Department of Nutrition, Faculty of Medicine, Universitas Indonesia – Dr Cipto Mangunkusumo General Hospital, Jakarta, Indonesia Human Nutrition Research Center, Indonesian Medical Education and Research Institute (HNRC-IMERI), Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia
Rina Agustina
Affiliation:
Department of Nutrition, Faculty of Medicine, Universitas Indonesia – Dr Cipto Mangunkusumo General Hospital, Jakarta, Indonesia Human Nutrition Research Center, Indonesian Medical Education and Research Institute (HNRC-IMERI), Faculty of Medicine, Universitas Indonesia, Jakarta, Indonesia
*
*Corresponding author: Email mecheva@iss.nl
Rights & Permissions [Opens in a new window]

Abstract

Objectives:

To aid the design of nutrition interventions in low- and middle-income countries undergoing a nutrition transition, this study examined behavioural and environmental risk factors associated with childhood overweight and obesity in urban Indonesia.

Design:

Body height and weight of children were measured to determine BMI-for-age Z-scores and childhood overweight and obesity status. A self-administered parental survey measured socio-economic background, children’s diet, physical activity, screen time and parental practices. Logistic and quantile regression models were used to assess the association between risk factors and the BMI-for-age Z-score distribution.

Setting:

Public primary schools in Central Jakarta, sampled at random.

Participants:

Children (n 1674) aged 6–13 years from 18 public primary schools.

Results:

Among the children, 31·0 % were overweight or obese. The prevalence of obesity was higher in boys (21·0 %) than in girls (12·0 %). Male sex and height (aOR = 1·67; 95 % CI 1·30, 2·14 and aOR = 1·16; 95 % CI 1·14, 1·18, respectively) increased the odds of being overweight or obese, while the odds reduced with every year of age (aOR = 0·43; 95 % CI 0·37, 0·50). Maternal education was positively associated with children’s BMI at the median of the Z-score distribution (P = 0·026). Dietary and physical activity risk scores were not associated with children’s BMI at any quantile. The obesogenic home food environment score was significantly and positively associated with the BMI-for-age Z-score at the 75th and 90th percentiles (P = 0·022 and 0·023, respectively).

Conclusions:

This study illustrated the demographic, behavioural and environmental risk factors for overweight and obesity among primary schoolchildren in a middle-income country. To foster healthy behaviours in primary schoolchildren, parents need to ensure a positive home food environment. Future sex-responsive interventions should involve both parents and children, promote healthy diets and physical activity and improve food environments in homes and schools.

Type
Research Paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press on behalf of The Nutrition Society

The increasing prevalence of overweight and obesity in low- and middle-income countries (LMIC) has been linked to a nutritional transition towards increased consumption of energy-dense processed foods that is correlated with decreased levels of people’s physical activity and various environmental factors(Reference Popkin and Slining1). These changes in energy intake and expenditure were also observed in children and may explain the rapid increase in childhood overweight and obesity in many regions(2).

The setting of this study, urban Indonesia, reflects general trends in LMIC. The obesogenic environment and widespread dietary and lifestyle changes are driven by urbanisation and rising income. These changes also affect children and may contribute to overnutrition(Reference Angkurawaranon, Jiraporncharoen and Chenthanakij3Reference Rachmi, Li and Baur5). Indonesia currently has one of the highest and fastest growing rates of childhood overweight and obesity in Southeast Asia (for review of the available evidence see ref. Reference Lindsay, Sitthisongkram and Greaney6) with sex-specific risks(Reference Agustina, Fenny and Suparmi7). The latest estimates from the 2018 National Basic Health survey indicate that nationwide 20 % of children of 5–12 years old are overweight or obese(8). The prevalence of childhood overweight is higher in urban areas, reaching 29·2 % in Jakarta(8).

Previous research has documented a variety of factors contributing to the development of childhood overweight and obesity, such as family socio-economic background, children’s energy intake and expenditure, as well as environmental, socio-cultural, and media influences. The apparent heterogeneity of findings from high-, middle- and low-income countries indicates the complex relationships between these factors, which may vary both between and within countries and households(Reference Sigman-Grant, Hayes and VanBrackle9,Reference Williamson, Dilip and Dillard10) . Family socio-economic markers (household income and parental education) could be both positively and negatively associated with childhood overweight and obesity(Reference Tzioumis and Adair11). While the growing consumption of unhealthy foods and corresponding overweight might be attributed to rising incomes(Reference Fox, Feng and Asal12) (for Indonesia see refs. Reference Rachmi, Li and Baur5 and Reference Lipoeto, Lin and Angeles-Agdeppa13), the income–obesity link might change over time with a country’s economic development(Reference Mathieu-Bolh14) and advancement of nutrition transition(Reference Tzioumis and Adair11). Similarly, significant nutritional returns to maternal education were documented in LMIC(Reference Alderman and Headey15), while at the same time, the increase in female education and labour force participation might lead to an increasing prevalence of overweight and obesity(Reference Fox, Feng and Asal12) (for Indonesia see refs. Reference Rachmi, Li and Baur5 and Reference Rachmi, Agho and Li16).

Emerging literature explores the potential co-occurrence of various factors associated with childhood overweight. However, to date, childhood obesity research has primarily focused on the joint influence of dietary and activity patterns(Reference Bleich, Vercammen and Zatz17,Reference Katzmarzyk, Chaput and Fogelholm18) . The concurrent influence of physical and social environments on children’s healthy behaviours and weight status requires further consideration(Reference Turner, Aggarwal and Walls19). The available conceptualisations of this environment illustrate the interaction of numerous contextual factors operating at the macro (community) and micro (family) levels(Reference Davison, Jurkowski and Lawson20). Previous food environment research in LMIC was predominantly concerned with the macro-level environment and dealt mostly with the availability of unhealthy foods and beverages in neighbourhoods and schools(Reference Turner, Aggarwal and Walls19). Family or parental influence was often used as a component of school-based programmes rather than a central target of the research or policy(Reference Davison, Jurkowski and Lawson20). The home food environment is composed of a broad range of components, spanning from the availability and accessibility of healthy and unhealthy foods to parental attitudes and practices(Reference Bates, Buscemi and Nicholson21,Reference Yee, Lwin and Ho22) . The complexity of factors, their interaction and their influence on children’s behaviours and health outcomes have been conceptualised in various models(Reference Davison and Birch23,Reference Rosenkranz and Dzewaltowski24) . However, most food environment research has been conducted in high-income countries, and there is little evidence from LMIC, where the food environment and diets are under transition, and where the rising prevalence of childhood overweight requires urgent attention(Reference Turner, Aggarwal and Walls19). A summary of a scoping review of the available studies in LMIC is provided in online Supplementary Appendix A.

The present study contributes to childhood obesity research in LMIC by providing novel evidence of a comprehensive set of demographic, behavioural and environmental risk factors and their associations with children’s weight across its entire distribution using recent data on a focused high-risk population in the urban setting of an upper middle-income country.

Conceptual framework

The ecological systems theory states that children’s behaviour occurs in and is influenced by ‘its ecological context; that is, in the actual environments in which human beings live their lives’(Reference Bronfenbrenner, Morri, Lerner and Damon25) (as put forward by Bronfenbrenner, p.794). This theory has been applied in public health interventions to assess the intersecting influences of cultural, social, economic, psychological and environmental factors on children’s health behaviours(Reference Davison, Jurkowski and Lawson20).

The present study adopted the family ecological model that extends the ecological systems theory and focuses on the family as a domain for intervention(Reference Davison, Jurkowski and Lawson20). The study retained some community (neighbourhood walkability and crime levels), policy (food labelling, school physical education and food policies), media (advertising to children) and organisational characteristics (school environment, job characteristics and work demand). The home environment is conceptualised as a broad range of social and physical factors within the family, such as food availability and routines, parental knowledge, and practices that promote or impede children’s health behaviours related to eating and physical activity and may contribute to an increased risk of overweight and obesity (for review of previous research see ref. Reference Bates, Buscemi and Nicholson21).

The conceptual framework (Fig. 1) includes three domains of behavioural (dietary habits and physical activity) and environmental (obesogenic home food environment) risk factors, each of which has a set of measurable dimensions. The domain of dietary habits includes measurements of children’s dietary intake, such as consumption of breakfast, fruits, vegetables, snacks, and sugar-sweetened and soft drinks. The physical activity domain comprises measurements of a child’s physical activity level and sedentary behaviour (screen time). The obesogenic home food environment is related to parental nutrition knowledge and family eating practices. All domains are associated with family social (parental influences) and physical (home environment) factors, as well as with parent–child socio-economic correlations. This conceptual framework informs the empirical specifications of this study.

Fig. 1 Conceptual framework. FV, fruit and vegetables; Veg., vegetables; TW, times a week; SSB/CSD, sugar-sweetened beverages/carbonated soft drinks

Methods

Data collection and participants

The sample consisted of 1674 children aged 6–13 years in grades 2–5 from eighteen public primary schools in Central Jakarta, Indonesia. Schools were randomly sampled from a list of 279 schools in Central Jakarta in the Ministry of Education and Culture’s database (rand function in Excel), as described elsewhere(Reference de Vries Mecheva, Rieger and Sparrow26). School principals invited class teachers for a briefing session. Among the teachers present during the session, at least one class in each grade was selected by the principal and teachers to be included in the study. Children who were reported to have no history of food allergies and were available at school on a certain day participated in a separate snack choice experiment(Reference de Vries Mecheva, Rieger and Sparrow26). All but four parents/guardians of the participating children completed a self-administered parental survey (the English version of the survey questionnaire is provided in online Supplementary Appendix B) that was sent home along with an informed consent letter to obtain parental permission for the children to participate in the experiment. During the experiment, the school food environment was not formally assessed. However, as observed during the data collection, informal food vendors often operate near or even inside the school premises, offering an assortment of prepared foods and snacks to children.

The self-administered parental survey collected information on family socio-economic background, parental recall of children’s last 7 d consumption of various foods, physical activity, sedentary behaviour, home food environment, and parental practices. The survey instruments were obtained from existing validated tools such as the FFQ(Reference Livingstone, Robson and Wallace27,Reference Fatihah, Ng and Hazwanie28) and Child Feeding Questionnaire(Reference Birch, Fisher and Grimm-Thomas29,Reference Purwaningrum, Sibagariang and Arcot30) . A bilingual researcher translated the survey instruments into Bahasa Indonesia. Local experts then reviewed the translated instruments to ensure that the scales and selected food items were appropriate for the study settings. The survey instruments were pretested among thirteen randomly selected mothers of primary schoolchildren. Anthropometrics (height and weight) of students whose parents provided consent were measured by a trained nutritionist.

Measures

Child’s height and weight

Two measurements were obtained for each child’s height and weight. Height and weight variables used in the analysis were constructed using average of these two measurements. Anthropometric data were missing for three children. BMI was calculated as the weight in kilograms divided by the squared height in metres. The respective BMI Z-score, adjusted for age and sex, was calculated using the WHO Child Growth Standards and WHO Reference 2007 data and the command zanthro (Reference Vidmar, Cole and Pan31) in STATA version 16.1 (StataCorp. LLC, College Station, TX, USA). The WHO Child Growth Standards (age- and sex-specific) cut-offs were applied to estimate the proportion of normal, overweight (BMI-for-age Z-score above +1 sd), and obese (BMI-for-age Z-score above +2 sd) children(32,Reference De Onis and Lobstein33) . Children with extreme BMI values (equal to or greater than 5 sd from the mean for their age) were not excluded from the sample as they were accommodated by quantile regression analysis.

Socio-economic and demographic variables

The parental survey included information on the children’s age, sex, maternal and paternal income, and educational attainment. Parental income was estimated as the sum of the monthly salaries of the mothers and fathers. The variable was then dichotomised according to whether parental income was equal or lower than the minimum wage. The highest level of education attained by each parent was converted into years: 6 years for elementary school, 9 years for junior high school, 12 years for senior high school, 16 years for undergraduate studies and 18 years for master’s and PhD degrees.

Child’s dietary intake and physical activity

A score for the behavioural risk associated with an unhealthy diet was created for each child(Reference Katzmarzyk, Chaput and Fogelholm18). Four questions of the self-administered parental survey recorded the parental recall of the number of times a week the child consumed fruits (five items: orange, watermelon, papaya, mango and banana), vegetables (five items: carrot, spinach, kale, tomato and beans), snacks (five items: chips, sweet and savoury snacks, confectionery, biscuits, and ice cream) and sugar-sweetened beverages (four items: carbonated soft drinks, flavoured drinks, juice and sweetened tea). Additional fruits (apples, grapes, dragon fruit, snake fruit, durian, guava, etc.), vegetables (mustard green, bean sprout, beans, corn, broccoli, cabbage, etc.) and snacks (bread, fried food, pizza, pudding, cake, instant noodles, etc.) were reported by parents in the open question (Other) and used in the respective variables. No other items of sugar-sweetened beverages reported in the open question (Other) were different from the provided answer options. The place of consumption (home, school or elsewhere) was not specifically asked in the questions on food and beverages intake. One question captured the frequency of the child’s breakfast consumption over the last 7 d. The average weekly intake of the four food groups and breakfast consumption were used to compute a dietary risk score. The cut-off points(Reference Battakova, Mukasheva and Abdrakhmanova34) were applied to create dichotomised variables indicating whether the child consumed less than one fruit (< 7 times a week) and one vegetable (< 7 times a week) per d, frequently consumed snacks (> 12 times a week) and sugar-sweetened beverages (> 3 times a week) and did not eat breakfast every day (< 7 times a week). The scores of the five components were summed to obtain the child’s overall dietary risk score.

The physical activity risk score was estimated using the number of times the child was running, playing a ball or cycling during the last 7 d, and the average number of hours per d that the child spent playing computer games, using the Internet, or watching TV. Additional physical activities (such as walking, playing badminton, volleyball, basketball, dancing, swimming and performing martial arts) were reported by parents in the open question (Other) and were also considered. Similar to the dietary risk score, the cut-off points for limited physical activity (< 3 times a week) and prolonged screen time (≥ 2 h per d) were applied(Reference Battakova, Mukasheva and Abdrakhmanova34). The two scores were then combined to obtain the child’s overall physical activity risk score.

Home food environment

The obesogenic home food environment was measured using six components of parental self-reports on the encouragement of children’s fruit and vegetable consumption (one item; five-point Likert scale), frequency of children helping to prepare dinner (one item; number of times a week), eating dinner with the family (one item; number of times a week), having vegetables at dinner (one item; five-point Likert scale), parental knowledge of the guidelines for children’s fruit and vegetable intake (two items, number of servings per d), and restrictions on snack consumption (one item; five-point Likert scale). The choice of variables was informed by the conceptual framework and previous studies(Reference Dallacker, Hertwig and Mata35Reference Pearson and Biddle37). Six dichotomised and reverse-coded variables for each component of the home food environment were constructed to indicate whether the parents failed to encourage the child’s fruit and vegetable consumption, the child did not help prepare dinner (< 1 time a week), the child did not eat dinner with the family every day (< 7 times a week), vegetables were not usually served at dinner, parental knowledge of the required child’s fruit and vegetable intake did not meet the guidelines of at least five servings of fruits and vegetables a day, as recommended by the WHO(38) and Indonesian Ministry of Health(39), and parents did not restrict the child’s snack consumption. The scores of the six components were summed to obtain the overall obesogenic home food environment score.

Statistical analysis

Means and standard deviations were used to describe continuous variables, and frequencies and percentages were used for categorical variables. Fisher’s exact test was used to determine if there was a significant association between categorical variables, and t test was performed for continuous variables. Missing observations of children’s age (n 131) and maternal education (n 63) were interpolated using the class and pooled sample means, respectively. The indicator of parental income was not used in the analysis for two reasons. First, 14·0 % of observations were missing due to the potential sensitivity of reporting, as was found in the pretest. Second, the variable was strongly and negatively correlated with years of maternal education. Parental education was found to be associated with the higher socio-economic status of households in many LMIC(Reference Peltzer, Pengpid and Samuels40). As in Indonesia mothers are the main caregivers(Reference Rachmi, Hunter and Li41), the indicator of maternal years of education was used in the empirical specification. Missing observations for behavioural and environmental risk factors were excluded from the analysis (listwise deletion). Spearman’s rank correlation was used to assess the strength of relationship between the risk scores and their components.

The first set of analyses tested the associations between overweight/obesity status and children’s demographics and exposure to behavioural and environmental risks described in the conceptual framework. The relationships between child–parent characteristics and three risk scores (diet, physical activity and home food environment) that captured various risk factors were modelled using logistic regression. As the influence of the risk factors may vary depending on the child’s weight status(Reference Lindsay, Sitthisongkram and Greaney6), the second set of analyses examined the associations between behavioural and environmental risk scores and children’s weight across five percentiles (10th, 25th, median, 75th and 95th percentiles) of the standardised BMI-for-age Z-score distribution using a quantile regression model(Reference Koenker42) (sqreg command in STATA). Standard errors stemmed from 2000 bootstrap replications. The model included controls for children’s age, height and maternal education and was fitted for the pooled sample as well as separately for boys and girls. The additional analysis explored the associations between the components that might contribute to the obesogenic home food environment and children’s consumption of fruits, vegetables, and snacks. In all analyses, the statistical significance of the findings was highlighted using a 5 % significance cut-off.

Results

Sample characteristics

The socio-economic characteristics of the children and their parents are shown in Table 1. The mean age was 9·5 years. The average height was 134·2 cm, and the average weight was 32·3 kg. Almost half (48·0 %) of the 1674 children were male. Most parents (77·0 %) had a combined income equal to or lower than the minimum wage of Rp3,941 000 (US$284). The average education of parents was about 11·2 years and did not differ across parents’ or children’s sexes.

Table 1 Descriptive statistics of children in selected primary schools in Jakarta

d, dummy variable; L7D, last 7 d; FV, fruit and vegetables.

Differences of the means assessed by using the Fisher’s exact test for categorical variables and t test for continuous variables.

Childhood overweight and obesity

Approximately one-third (31·0 %) of the children in the sample were overweight or obese, whereas 17·0 % were obese (Table 1). Obesity was higher in boys (21·0 %) than in girls (12·0 %) (two-tailed P < 0·0001). The average BMI-for-age Z-score was 0·19 sd with a minimum of −5·85 and a maximum of 4·16.

Children’s dietary intake and activity

On average, the children consumed at least one serving of fruit and one serving of vegetables daily (7·0 and 5·5 times a week, respectively, Table 1). Boys ate fruits and vegetables significantly less frequently than girls did (both P < 0·05). Children consumed snacks and sugar-sweetened beverages almost every day (9·8 and 5·5 times a week, respectively), and the consumption frequency was higher among girls (both P < 0·05). A majority of parents (84·0 %) agreed that they limit snack consumption (‘I have to be sure that my child does not eat too many sweet/savoury snacks–chips, chocolate, candy, ice cream). Additionally, most parents (93 %, not reported in the table) were aware of children’s food intake at school as children indicated in their responses to the question ‘Do your parents often ask what food you eat at school?’ (the English version of the children’s questionnaire is provided in online Supplementary Appendix B). On average, children ate breakfast and participated in family dinners approximately four times a week. Children rarely helped prepare dinner (on average 1·6 times a week, girls more frequently; P < 0·0001).

Children in the sample spent 5·1 h a day watching TV, playing games on a computer/mobile, or using the Internet. Boys had a significantly longer screen time than girls (P < 0·0001). Children went outside to run, bike or play ball on average 4·1 times a week, and boys were significantly more physically active than girls (P < 0·0001).

Behavioural and environmental risk factors

Table 2 presents the indicators for behavioural (i.e. diet and physical activity) and environmental risk scores as well as the prevalence of their components.

Table 2 Prevalence and scores of behavioural and environmental risk factors

d, dummy variable; FV, fruit and vegetables.

Differences of the means assessed by using the Fisher’s exact test for categorical variables.

The average dietary risk score in the pooled sample was 2·6 on a scale of 0 to 5. There was no sex difference in the overall score (two-tailed P = 0·450), but there were in all its components excluding breakfast consumption. Most children (71·7 %) did not consume vegetables daily, and slightly more than half did not eat breakfast and fruits every day (54·7 % and 57·5 %, respectively). Excessive consumption of snacks was reported in one-third of children (29·7 %). Two-thirds of the children frequently (more than three times a week) consumed sugar-sweetened beverages (61·8 %). Only 2·9 % of the children scored 0 points, while 18·8 % had four or five dietary risk factors.

The average risk score of physical activity was 1·4 on a scale from 0 to 2 but was more prominent among girls (two-tailed P < 0·0001). Of the children, 94·6 % had two or more hours of screen time daily, and 44·8 % did not participate in active sports or games at least three times a week. The combination of a low level of physical activity and high level of sedentary behaviour (a maximum score of 2) was achieved by 36·0 % of the children and almost half of the girls (45·4 %).

The score for obesogenic home food environment was 2·5 out of 6, without significant differences across sex (two-tailed P = 0·201). Only 3·0 % of the sample met the WHO(38) and Indonesian Ministry of Health(39) guidelines for the five servings of fruits and vegetables daily. An average parent in the sample believed that the child should consume 3·6 servings of fruits and vegetables a day. Most parents (84·0 %) reported that they encouraged their children to consume fruits and vegetables, and 33·0 % agreed with the statement that vegetables are usually served at dinner. At the same time, every sixth parent did not encourage their children to eat fruits and vegetables (15·6 %), and most parents were unaware of the required child’s daily consumption of five servings of fruits and vegetables (78·9 %). Moreover, 67·0 % of parents did not usually serve vegetables at dinner, and 16·1 % reported that they did not limit their children’s snack consumption. A large proportion of the children did not regularly help prepare or eat dinner with their families (53·9 % and 48·5 %, respectively). Only 3·9 % of the children were not exposed to any environmental risk factors, while 5·5 % of the children scored a maximum of 5–6 points. Approximately two-thirds of the children (55·0 %) faced two or three negative environmental components.

Each risk score was highly correlated with the factors underlying other risk scores (Table 3). The dietary risk score (five items, α = 0·54) was strongly associated with the obesogenic home food environment and its components, such as parental encouragement of fruit and vegetable consumption, parental limits of snack consumption, serving vegetables at dinner, and having regular family dinners. The physical activity risk score (two items, α = 0·15) was correlated with children’s consumption of fruits, vegetables, drinks and snacks. The obesogenic home food environment score (six items, α = 0·31) was strongly associated with irregular consumption of fruits, vegetables, sugar-sweetened beverages and breakfast as well as with the level of children’s physical activity. Low values of Cronbach’s alpha (not reported in the table) are suggestive of the heterogeneity in latent sources of risk within the same domain.

Table 3 Correlation of risk scores and their components

d, dummy variable; FV, fruit and vegetables.

Number of paired ranks n 1121. Spearman’s rank correlation.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Risk factor relationships and their association with childhood overweight and obesity

Table 4 presents the estimates of the logit models that assess associations between the odds of childhood overweight or obesity and risk factors, including child’s (sex, age and height) and maternal (years of education) characteristics, and behavioural (child’s diet and physical activity) and environmental risk factors at home (the estimates of a linear probability model, showing qualitatively similar results, are provided in online Supplementary Table SC1 in Supplementary Appendix C). There were strong positive associations between the child’s male sex (aOR = 1·67; 95 % CI 1·30, 2·14) and height (aOR = 1·16; 95 % CI 1·14, 1·18 with weight status (Column 1). Every year of age reduced the odds of being overweight or obese by a factor of 0·43 (95 % CI 0·37, 0·50) (Column 1). Maternal years of education displayed a positive but insignificant association with childhood overweight in all models. The influence of behavioural risk factors (dietary and physical activity risk scores) was statistically insignificant (Columns 1, 2 and 3). The score for the obesogenic home food environment significantly increased the odds of childhood overweight or obesity (OR = 1·11; 95 % CI 1·01, 1·22) (Column 4). Similar results were obtained when estimating a model that included seventeen school dummy variables to control for the school environment (online Supplementary Table SC2 in Supplementary Appendix C).

Table 4 Risk factors of overweight and obesity among children in selected primary schools in Jakarta

d, dummy variable; environm., environment.

Logistic regression model. Robust standard errors. Dependent variable: a binary overweight indicator based on the BMI-for-age Z-score above +1 sd according to the WHO Child Growth Standards(32,Reference De Onis and Lobstein33) . All specifications included a dummy for missing observations of age and maternal years of education, which were interpolated with the sample mean.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Behavioural (dietary and physical activity) risk scores were not associated with children’s weight across five percentiles of the standardised BMI-for-age Z-score distribution in the pooled sample (Table 5). The obesogenic home food environment risk core was significantly associated with the children’s BMI-for-age Z-score at the 75th and 90th percentiles (obesity) (P = 0·022 and 0·023, respectively). Maternal education was positively and significantly associated with children’s BMI-for age Z-score at the median (P = 0·026). Male sex was predictive of higher BMI-for-age Z-scores (75th and 90th percentiles, both P < 0·0001). This finding underlines the rationale for estimating the model separately for girls and boys. Table 6 provides the detailed estimates of the quantile regression model by sex. The dietary and physical activity risk scores were not associated with the child’s BMI across its Z-score distribution (Panels A and B). The coefficients of the obesogenic home food environment at the 75th and 90th percentiles were somewhat larger for boys than for girls but were not precisely estimated (P = 0·105 and 0·069, respectively).

Table 5 Child’s BMI Z-score and behavioural and environmental risk scores

d, dummy variable; environm., environment.

n 1538. Dependent variable: BMI-for-age Z-score that was calculated using the WHO Child Growth Standards and WHO Reference 2007 data(Reference Vidmar, Cole and Pan31). Column 1: pooled linear regression model. Robust standard errors. Columns 2–6: quantile regression model. Bootstrapped standard errors (2000 replication). Pseudo-R-squared values are reported for quantile regression estimates. Both models included a dummy for missing observations of age and maternal years of education, which were interpolated with the sample mean.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Table 6 Child’s BMI Z-score and behavioural and environmental risk scores, by sex

d, dummy variable; environm., environment.

n 732 (boys) and 806 (girls). Dependent variable: BMI-for-age Z-score that was calculated using the WHO Child Growth Standards and WHO Reference 2007 data(Reference Vidmar, Cole and Pan31). Column 1: linear regression model. Robust standard errors. Columns 2–6: quantile regression model. Bootstrapped standard errors (2000 replication). Pseudo-R-squared values are reported for quantile regression estimates. Both models include a dummy for missing observations of age and maternal years of education, which were interpolated with the sample mean.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Additional analysis of the associations between the components of the obesogenic home food environment and children’s fruit, vegetable, and snack consumption is provided in Table 7. Children whose parents did not encourage fruit and vegetable consumption had lower odds of eating at least one serving of them daily (aOR = 0·40; 95 % CI 0·27, 0·58). Similarly, child’s involvement in dinner preparation (aOR = 0·51; 95 % CI 0·38, 0·68) and inadequate parental knowledge of the required fruit and vegetable intake (aOR = 0·67; 95 % CI 0·46, 0·96) reduced the odds of everyday consumption of fruit and/or vegetable. Higher maternal education (aOR = 0·86; 95 % CI 0·74, 0·98), lack of parental encouragement (aOR = 0·11; 95 % CI 0·01, 0·89), irregular family dinners (aOR = 0·31; 95 % CI 0·14, 0·68) and unavailability of vegetables at dinner (aOR = 0·35; 95 % CI 0·18, 0·69) decreased the odds of meeting the recommended fruit and vegetable intake of five or more servings per d(38,39) . At the same time, children whose parents did not limit their children’s snack consumption were 3·5 times more likely to consume five or more fruits and vegetables daily (aOR = 3·47; 95 % CI 1·53, 7·87). Interestingly, children who did not help prepare dinner were less likely to snack every day (aOR = 0·64; 95 % CI 0·49, 0·82) and 1·4 times more likely to meet the guidelines of restricted (less than 12 snacks per week) consumption of snacks (aOR = 1·37; 95 % CI 1·06, 1·76). Moreover, as Table 8 illustrates, parental and children’s responses to similar questions were strongly correlated (the English version of the children’s questionnaire is provided in online Supplementary Appendix B).

Table 7 Child’s fruit, vegetables, and snacks consumption and the components of obesogenic home food environment

d, dummy variable; FV, fruit and vegetables.

Logistic regression models. Robust standard errors. n 1188. Dependent variables: Column 1 – child consumes at least one serving of fruit or vegetable a day; Column 2 – a binary indicator of child’s daily fruit and vegetable intake that meets the recommended five or more servings; Column 3 – child consumes at least one snack a day; Column 4 – a binary indicator of a child’s weekly consumption of snacks that meets the restriction of less than 12. All specifications included a dummy for missing observations of age and maternal years of education, which were interpolated with the sample mean.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Table 8 Correlation of parental and children’s responses

d, dummy variable; FV, fruit and vegetables, rs, Spearman’s rank correlation coefficient.

Number of paired ranks n 1398. Spearman’s rank correlation.

*P < 0·05,

**P < 0·01,

***P < 0·001.

Discussion

The present study examined the associations of behavioural and environmental risk factors with childhood overweight and obesity among 6–13 years old children from public primary schools in Central Jakarta, Indonesia. The findings emphasise childhood overweight and obesity in Indonesia as an emerging and important public health issue that requires policy attention. Of the children, 31·0 % were overweight and more boys (21·0 %) than girls (12·0 %) were obese. Overall, childhood overweight and obesity in the present study were higher than the 15·6 % prevalence among 6–12-year-old children reported in a recent study by Oddo and colleagues(Reference Oddo, Maehara and Rah4) and a 16·0 % prevalence in a nationally representative sample of young adolescents in Agustina et al.(Reference Agustina, Fenny and Suparmi7). The higher prevalence of obesity among boys than that among girls conforms with the findings of previous research in Indonesia(Reference Rachmi, Li and Baur5,Reference Agustina, Fenny and Suparmi7) .

Sex heterogeneity was also found in the children’s dietary habits and physical activity levels. Overall, only a few children met the recommendations for daily fruit and vegetable intake, and the average fruit and vegetable consumption by boys was significantly lower than that of girls. Girls consumed more snacks and sugar-sweetened drinks than boys. More than half of the children did not eat breakfast daily. Although boys spent more time running, cycling or playing a ball, they also spent significantly longer hours watching TV, using the Internet or playing computer games. In general, almost all children had two or more hours of daily screen time. In line with previous studies in Indonesia, girls were less physically active than boys(Reference Agustina, Fenny and Suparmi7,Reference Sultana, Rahman and Sigel43,Reference Roshita, Riddell-Carre and Sjahrial44) and had a comparatively similar level of screen time(Reference Nurwanti, Hadi and Chang45). Previous study in Indonesia explained the lower level of girls’ physical activity by the cultural restrictions of girls’ participation in sports (‘girls should not be sporty’)(Reference Roshita, Riddell-Carre and Sjahrial44) as well as by the domestic activities they have to undertake(Reference Agustina, Fenny and Suparmi7). Children’s low levels of physical activity and prolonged screen time were correlated with low intake of fruits and vegetables. This finding corresponds with previous research in high-income settings that reported a link between children’s sedentary behaviour and unhealthy eating(Reference Pearson and Biddle37).

The use of behavioural and environmental risk scores that account for the co-occurrence of various risk factors of childhood overweight is one of the innovations of the present study. To the best of the authors’ knowledge, to date, only one study in LMIC has explored the prevalence of dietary and inactivity risk scores from the sets of related factors(Reference Sultana, Rahman and Sigel43). Although it was not possible to compare the findings directly owing to the use of different factors and cut-off points, the results of the present study resemble those of previous reports. The average dietary risk score was 2·6 (scale 0–5) and sex homogeneous. Previous research reported a 79·7 % prevalence of unhealthy food intake among boys and 78·2 % among girls(Reference Sultana, Rahman and Sigel43). Similarly, the average risk score for physical activity was 1·4 (scale 0–2) and somewhat higher for girls as compared with 84·5 %/64·5 % and 86·1 %/64·7 % prevalence of physical activity/sedentary behaviour among girls and boys reported by Sultana and colleagues(Reference Sultana, Rahman and Sigel43).

Research in high-income countries has emphasised the importance of the home food environment for children’s healthy eating(Reference Davison, Jurkowski and Lawson20), especially among younger children(Reference Ong, Ullah and Magarey36). The present study found that the overall obesogenic home food environment score was correlated with a child’s irregular consumption of fruits, vegetables and breakfast, as well as with a low level of physical activity. Although most parents in the present study reported encouragement of their children’s fruit and vegetable intake and restriction of snack consumption, many were not familiar with dietary guidelines and did not provide vegetables at dinner. About half of the children did not help prepare and had dinner with their family on a regular basis.

Childhood overweight in the present study was strongly associated with children’s (male) sex, taller stature and younger age. A study of Indonesian adolescents from the 2013 wave of the Basic Health Survey (n 155 645) reported a higher overweight prevalence among 10–14-year-olds as compared with 15–18-year-old children(Reference Nurwanti, Hadi and Chang45). That trend persisted in 2018 as reported in a recently published study by Agustina and colleagues(Reference Agustina, Fenny and Suparmi7). Rachmi et al. analysed the data from four cross-sectional waves of the Indonesian Family Life Survey (n 4101) and found that younger children (2·0–2·9-year-old) had a higher prevalence of overweight(Reference Rachmi, Agho and Li16). While cultural norms that favour boys over girls(Reference Rachmi, Li and Baur5) may translate into parental feeding practices and consequent overnutrition of boys, sex- and age-specific concerns about body image(Reference Agustina, Fenny and Suparmi7) may also play a role in higher prevalence of overweight and obesity among boys and younger children in Indonesia.

The risk of being overweight increased with higher maternal education after adjusting for other predictors of obesity such as diet and home food environment. To date, there has been mixed evidence on the influence of maternal education on children’s weight status. Previous research in developed countries has often used maternal education as a proxy for overall knowledge about healthy diet and other health-promoting behaviours(Reference Alderman and Headey15). This expectation was not supported by evidence from developing countries, where previous studies found a positive association between parental education and childhood overweight and obesity(Reference Katzmarzyk, Chaput and Fogelholm18,Reference Kim, Roesler and von dem Knesebeck46) . Educated mothers may have a higher socio-economic status and work-time pressure and may tend to replace nutritious home-cooked meals with ready-to-eat processed foods and snacks(Reference Rachmi, Li and Baur5,Reference Lipoeto, Lin and Angeles-Agdeppa13,Reference Peltzer, Pengpid and Samuels40) . Conversely, a recent study of malnutrition risk factors in two Indonesian districts (n 2160) reported a negative association between maternal education and adolescent overweight(Reference Maehara, Rah and Roshita47). Further research is required to explore the associations between parental characteristics and childhood overweight and obesity in Indonesia.

In line with previous research in ASEAN countries(Reference Pengpid and Peltzer48) and Indonesia(Reference Maehara, Rah and Roshita47), the dietary and physical activity risk scores in this study were not associated with the children’s weight status. Exposure to an obesogenic home food environment resulted in higher odds of being overweight. However, the studied risk factors differed across the five quantiles of the standardised BMI-for-age Z-score distribution. Maternal education was positively and significantly associated with children’s BMI-for-age Z-scores at the median. The child’s male sex was predictive of a higher BMI-for-age Z-score (75th and 90th percentiles). Exposure to the obesogenic food environment at home was strongly and significantly associated with a high BMI (75th and 90th percentiles of the Z-score distribution) in the pooled sample.

An adequate level of consumption of fruits and vegetables, together with a limited intake of nutrient-poor processed foods, should be the main component of a healthy diet(49). Parental knowledge, encouragement and family routines play an important role in promoting children’s healthy behaviours(Reference Sirasa, Mitchell and Rigby50). The findings of this study were consistent with this notion. Home food environment was found to correlate with the child’s diet, particularly the intake of fruits, vegetables and breakfast. Children whose parents did not encourage fruit and vegetable intake consumed fewer fruits and vegetables. Children who lacked parental encouragement were not served vegetables at dinner and did not regularly eat dinner with their family were less likely to meet the recommended five servings of fruit and vegetables per d. At the same time, children whose parents did not limit snack consumption had a much greater chance of consuming five or more fruits and vegetables a day than children with restrictive parents. Previous research in Southeast Asia reported positive associations between parental control, restriction of food consumption and childhood overweight(Reference Lindsay, Sitthisongkram and Greaney6). Parental practices are shaped by socio-cultural norms and beliefs especially in contexts where awareness of the negative influence of environmental factors and detrimental consequences of childhood overweight is low(Reference Rachmi, Hunter and Li41).

This study had some limitations. First, it used self-reported recall data. Parental responses might suffer from social desirability bias by over-reporting healthy behaviours and under-reporting unhealthy behaviours. Despite that the survey instruments were obtained from existing validated tools(Reference Livingstone, Robson and Wallace27Reference Purwaningrum, Sibagariang and Arcot30), the number of dietary and activity components was limited due to time and logistical constraints. The respondents’ self-reports to frequency questions might have been subject to systematic measurement error. Future research could use 24-h recall and food records to measure total intakes of different food components as well as clinical tools for physical activity measurement (accelerometers, pedometers and activity diaries) for more precise assessment of energy expenditure. Second, the fitted models did not control for parental income. Therefore, the reported associations between children’s behaviours and weight status might be confounded by income effects for unhealthy food and sedentary lifestyles. Third, the studied indicators were highly correlated. Dissection of specific influences requires experimental evidence. Fourth, the sample might lack the statistical power to detect the association of children’s diet and physical activity and weight status. Future research might address the issue by leveraging panel data and/or by employing more precise measures of risk factors than self-reported data. Fifth, the sample was relatively homogeneous in terms of demographic characteristics compared with the rest of the country. Six and related, although the previous study based on the same sample(Reference de Vries Mecheva, Rieger and Sparrow26) reported that the sample characteristics were well aligned with the national census, the classes in schools were not selected at random and that might have affected the representativity of the data. Notwithstanding these limitations, the findings provide useful evidence that could inform future research and family-based interventions to reduce the risk of childhood overweight and obesity in Indonesia.

Conclusion

This study illustrates the potential association between demographic, behavioural, and environmental risk factors and overweight or obesity among primary schoolchildren in a middle-income country setting. This adds to the growing literature on the prevalence of health risk factors, their coexistence and associations with childhood overweight and obesity in LMIC. To the best of the authors’ knowledge, this is the first attempt to assess the complex relationship between various risk factors and their association with children’s weight across the entire BMI-for-age Z-score distribution in urban Indonesia. This knowledge is required to inform future research and to develop effective interventions and policies to counter the rising prevalence of childhood overweight and obesity in LMIC. Parents of primary schoolchildren can encourage healthy behaviours in their children by raising their nutrition knowledge and providing a supportive home food environment. Future interventions should be sex-responsive given the documented sex differences, target both parents and children, and aim to improve the food environment at homes and schools and continuously promote healthy diet and adequate physical activity.

Acknowledgements

Acknowledgements: The author are grateful to Khalida Fauzia and Apriliya Tri Hidayati for their outstanding work during the data collection. We thank the Editor and two anonymous referees for their constructive comments. The study implementation by an international visiting scholar was approved by the Dean Faculty of Medicine, the Rector of Universitas Indonesia and the Ministry of Research, Technology and Higher Education of Indonesia. We are grateful to the Head of the Doctorate Study Program in Nutrition (2017–2020), Human Nutrition Research Centre of Indonesian Medical Education and Research Institute (HNRC-IMERI), International Relation Office (IRO) and Dean Office, Faculty of Medicine, Universitas Indonesia for the arrangement of the permission for an international vising scholar by the Ministry of Research, Technology and Higher Education of Indonesia. The authors thank all students, parents, teachers and principals of eighteen participating primary schools in Central Jakarta and Government of DKI Jakarta Province for their cooperation during the study. Authorship: MdVM: Conceptualisation, Methodology, Formal analysis, Investigation, Data curation, Writing – Original Draft, Visualisation and Project administration; M.R.: Methodology, Investigation, Writing – Review & Editing, Supervision, Project administration and Funding acquisition; R.S.: Methodology, Writing – Review & Editing and Supervision, and Funding acquisition; E.P.: Writing – Review & Editing and Resources; R.A.: Writing – Review & Editing, Resources, Project administration and Supervision. 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 study participants were approved by the Ethics Committees of the Faculty of Medicine, Universitas Indonesia, and the International Institute of Social Studies of Erasmus University Rotterdam, the Netherlands. Written informed consent was obtained from all subjects’ parents.

Financial support:

The International Institute of Social Studies Erasmus University and the Development Economics Group at Wageningen University provided the funding for the study.

Conflict of interest:

There are no conflicts of interest.

Supplementary material

For supplementary material/s referred to in this article, please visit https://doi.org/10.1017/S1368980023000897

References

Popkin, BM & Slining, MM (2013) New dynamics in global obesity facing low-and middle-income countries. Obes Rev 14, 1120.CrossRefGoogle ScholarPubMed
LBD Double Burden of Malnutrition Collaborators (2020) Mapping local patterns of childhood overweight and wasting in low-and middle-income countries between 2000 and 2017. Nat Med 26, 750759.CrossRefGoogle Scholar
Angkurawaranon, C, Jiraporncharoen, W, Chenthanakij, B et al. (2014) Urban environments and obesity in Southeast Asia: a systematic review, meta-analysis and meta-regression. PLoS One 9, e11354.CrossRefGoogle ScholarPubMed
Oddo, VM, Maehara, M & Rah, JH (2019) Overweight in Indonesia: an observational study of trends and risk factors among adults and children. BMJ Open 9, e031198.CrossRefGoogle ScholarPubMed
Rachmi, CN, Li, M & Baur, LA (2017) Overweight and obesity in Indonesia: prevalence and risk factors—a literature review. Public Health 147, 2029.CrossRefGoogle ScholarPubMed
Lindsay, AC, Sitthisongkram, S, Greaney, ML et al. (2017) Non-responsive feeding practices, unhealthy eating behaviors, and risk of child overweight and obesity in Southeast Asia: a systematic review. Int J Environ Res Public Health 14, 436.CrossRefGoogle ScholarPubMed
Agustina, R, Fenny, MD, Suparmi, MK et al. (2021) Psychosocial, eating behavior, and lifestyle factors influencing overweight and obesity in adolescents. Food Nutr Bull 42, S72S91.CrossRefGoogle ScholarPubMed
Riskesdas (2022) Laporan Hasil Riset Kesehatan Dasar (Report on Indonesian National Health Research). https://www.litbang.kemkes.go.id/laporan-riset-kesehatan-dasar-riskesdas (accessed November 2022).Google Scholar
Sigman-Grant, M, Hayes, J, VanBrackle, A et al. (2015) Family resiliency: a neglected perspective in addressing obesity in young children. Child Obes 11, 664673.CrossRefGoogle ScholarPubMed
Williamson, VG, Dilip, A, Dillard, JR et al. (2020) The influence of socioeconomic status on snacking and weight among adolescents: a scoping review. Nutrients 12, 167.CrossRefGoogle ScholarPubMed
Tzioumis, E & Adair, LS (2014) Childhood dual burden of under-and overnutrition in low-and middle-income countries: a critical review. Food Nutr Bull 35, 230243.CrossRefGoogle ScholarPubMed
Fox, A, Feng, W & Asal, V (2019) What is driving global obesity trends? Globalization or ‘modernization’? Global Health 15, 116.Google ScholarPubMed
Lipoeto, NI, Lin, KG & Angeles-Agdeppa, I (2013) Food consumption patterns and nutrition transition in South-East Asia. Public Health Nutr 16, 16371643.CrossRefGoogle ScholarPubMed
Mathieu-Bolh, N (2021) The elusive link between income and obesity. J Econ Survey 36, 935968.CrossRefGoogle Scholar
Alderman, H & Headey, DD (2017) How important is parental education for child nutrition? World Dev 94, 448464.CrossRefGoogle ScholarPubMed
Rachmi, CN, Agho, KE, Li, M et al. (2016) Stunting coexisting with overweight in 2.0–4.9-year-old Indonesian children: prevalence, trends and associated risk factors from repeated cross-sectional surveys. Public Health Nutr 19, 26982707.CrossRefGoogle ScholarPubMed
Bleich, SN, Vercammen, KA, Zatz, LY et al. (2018) Interventions to prevent global childhood overweight and obesity: a systematic review. Lancet Diabetes Endocrinol 6, 332346.CrossRefGoogle ScholarPubMed
Katzmarzyk, PT, Chaput, JP, Fogelholm, M et al. (2019) International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE): contributions to understanding the global obesity epidemic. Nutrients 11, 848.CrossRefGoogle Scholar
Turner, C, Aggarwal, A, Walls, H et al. (2018) Concepts and critical perspectives for food environment research: a global framework with implications for action in low-and middle-income countries. Glob Food Sec 18, 93101.CrossRefGoogle Scholar
Davison, KK, Jurkowski, JM & Lawson, HA (2013) Reframing family-centred obesity prevention using the Family Ecological Model. Public Health Nutr 16, 18611869.CrossRefGoogle ScholarPubMed
Bates, CR, Buscemi, J, Nicholson, LM et al. (2018) Links between the organization of the family home environment and child obesity: a systematic review. Obes Rev 19, 16727.CrossRefGoogle ScholarPubMed
Yee, AZ, Lwin, MO & Ho, SS (2017) The influence of parental practices on child promotive and preventive food consumption behaviors: a systematic review and meta-analysis. Int J Behav Nutr Phys Act 14, 114.CrossRefGoogle ScholarPubMed
Davison, KK & Birch, LL (2001) Childhood overweight: a contextual model and recommendations for future research. Obes Rev 2, 159171.CrossRefGoogle ScholarPubMed
Rosenkranz, RR & Dzewaltowski, DA (2008) Model of the home food environment pertaining to childhood obesity. Nutr Rev 66, 123140.CrossRefGoogle ScholarPubMed
Bronfenbrenner, U & Morri, PA (2006) The bioecological model of human development. In Handbook of Child Psychology, pp. 793828 [Lerner, RM and Damon, W, editors]. New York: Wiley.Google Scholar
de Vries Mecheva, M, Rieger, M, Sparrow, R et al. (2021) Snacks, nudges and asymmetric peer influence: evidence from food choice experiments with children in Indonesia. J Health Econ 79, 102508.CrossRefGoogle Scholar
Livingstone, MBE, Robson, PJ & Wallace, JMW (2004) Issues in dietary intake assessment of children and adolescents. Br J Nutr 92, S213S222.CrossRefGoogle ScholarPubMed
Fatihah, F, Ng, BK, Hazwanie, H et al. (2015) Development and validation of a food frequency questionnaire for dietary intake assessment among multi-ethnic primary school-aged children. Singap Med J 56, 687694.CrossRefGoogle ScholarPubMed
Birch, LL, Fisher, JO, Grimm-Thomas, K et al. (2001) Confirmatory factor analysis of the Child Feeding Questionnaire: a measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite 36, 201210.CrossRefGoogle ScholarPubMed
Purwaningrum, DN, Sibagariang, HYM, Arcot, J et al. (2018) Validation of a measurement instrument for parental child feeding in a low and middle-income country. Int J Behav Nutr Phys Act 15, 112.CrossRefGoogle Scholar
Vidmar, SI, Cole, TJ & Pan, H (2013) Standardizing anthropometric measures in children and adolescents with functions for egen: update. Stata J 13, 366378.CrossRefGoogle Scholar
WHO Multicentre Growth Reference Study Group (2006) WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 450, 7685.Google Scholar
De Onis, M & Lobstein, T (2010) Defining obesity risk status in the general childhood population: which cut-offs should we use? Int J Pediatr Obes 5, 458460.CrossRefGoogle ScholarPubMed
Battakova, Z, Mukasheva, S, Abdrakhmanova, S et al. (2017) Childhood obesity in Kazakhstan: behavioural health risks associated with diet and physical activity. Public Health Panorama 3, 695702.Google Scholar
Dallacker, M, Hertwig, R & Mata, J (2018) The frequency of family meals and nutritional health in children: a meta-analysis. Obes Rev 19, 638653.CrossRefGoogle ScholarPubMed
Ong, JX, Ullah, S, Magarey, A et al. (2016) Positive influences of home food environment on primary-school children’s diet and weight status: a structural equation model approach. Public Health Nutr 19, 25252534.CrossRefGoogle ScholarPubMed
Pearson, N & Biddle, SJ (2011) Sedentary behavior and dietary intake in children, adolescents, and adults: a systematic review. Am J Prev Med 41, 178188.CrossRefGoogle ScholarPubMed
WHO (2003) Diet, Nutrition, and the Prevention of Chronic Diseases: Report of a Joint WHO/FAO Expert Consultation. https://www.who.int/publications/i/item/924120916X (accessed November 2022).Google Scholar
Ministry of Health (2014) Peraturan Menteri Kesehatan Republik Indonesia Nomor 41 Tahun 2014 Tentang Pedoman Gizi Seimbang (Regulations Concerning Guidelines for Balanced Nutrition by the Minister of Health of the Republic of Indonesia Number 41 of 2014). https://peraturan.bpk.go.id/Home/Details/119080/permenkes-no-41-tahun-201 (accessed November 2022).Google Scholar
Peltzer, K, Pengpid, S, Samuels, T et al. (2014) Prevalence of overweight/obesity and its associated factors among university students from 22 countries. Int J Environ Res Public Health 11, 74257441.CrossRefGoogle ScholarPubMed
Rachmi, CN, Hunter, CL, Li, M et al. (2018) Food choices made by primary carers (mothers/grandmothers) in West Java, Indonesia. Appetite 130, 8492.CrossRefGoogle ScholarPubMed
Koenker, R (2005) Quantile Regression. New York: Cambridge University Press.CrossRefGoogle Scholar
Sultana, S, Rahman, MM, Sigel, B et al. (2021) Associations of lifestyle risk factors with overweight or obesity among adolescents: a multicountry analysis. Am J Clin Nutr 113, 742750.CrossRefGoogle ScholarPubMed
Roshita, A, Riddell-Carre, P, Sjahrial, R et al. (2021) A qualitative inquiry into the eating behavior and physical activity of adolescent girls and boys in Indonesia. Food Nutr Bull 42, S122S131.CrossRefGoogle ScholarPubMed
Nurwanti, E, Hadi, H, Chang, JS et al. (2019) Rural–urban differences in dietary behavior and obesity: results of the Riskesdas study in 10–18-year-old Indonesian children and adolescents. Nutrients 11, 2813.CrossRefGoogle ScholarPubMed
Kim, TJ, Roesler, NM & von dem Knesebeck, O (2017) Causation or selection–examining the relation between education and overweight/obesity in prospective observational studies: a meta-analysis. Obes Rev 18, 660672.CrossRefGoogle ScholarPubMed
Maehara, M, Rah, JH & Roshita, A (2019) Patterns and risk factors of double burden of malnutrition among adolescent girls and boys in Indonesia. PLoS One 14, e0221273.CrossRefGoogle ScholarPubMed
Pengpid, S & Peltzer, K (2019) Underweight and overweight or obesity and associated factors among school-going adolescents in five ASEAN countries, 2015. Diabetes Metab Syndr 13, 30753080.CrossRefGoogle ScholarPubMed
WHO (2004) Global Strategy on Diet, Physical Activity and Health. https://www.who.int/publications/i/item/9241592222 (accessed November 2022).Google Scholar
Sirasa, F, Mitchell, LJ, Rigby, R et al. (2019) Family and community factors shaping the eating behaviour of preschool-aged children in low and middle-income countries: a systematic review of interventions. Prev Med 129, 105827.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1 Conceptual framework. FV, fruit and vegetables; Veg., vegetables; TW, times a week; SSB/CSD, sugar-sweetened beverages/carbonated soft drinks

Figure 1

Table 1 Descriptive statistics of children in selected primary schools in Jakarta

Figure 2

Table 2 Prevalence and scores of behavioural and environmental risk factors

Figure 3

Table 3 Correlation of risk scores and their components

Figure 4

Table 4 Risk factors of overweight and obesity among children in selected primary schools in Jakarta

Figure 5

Table 5 Child’s BMI Z-score and behavioural and environmental risk scores

Figure 6

Table 6 Child’s BMI Z-score and behavioural and environmental risk scores, by sex

Figure 7

Table 7 Child’s fruit, vegetables, and snacks consumption and the components of obesogenic home food environment

Figure 8

Table 8 Correlation of parental and children’s responses

Supplementary material: File

De Vries Mecheva et al. supplementary material

Appendices A-C

Download De Vries Mecheva et al. supplementary material(File)
File 4.5 MB