24‐hour movement guidelines and weight status among preschool‐aged children in Bangladesh: A community‐level cross‐sectional study

Abstract Background Increasing levels of urbanization and digitization in Bangladesh may be adversely associated with children's weight status and related movement behaviors. This study examined the prevalence of obesity, physical activity (PA), sedentary behavior, and sleep among young children from a district town in northern Bangladesh and identified factors associated with unhealthy weight status. Methods Population‐based cross‐sectional study involving all kindergarten schools in Jamalpur District town. Schools and children aged 4–7 years were randomly selected and had their weight and height measured. Mothers completed a questionnaire on their child's PA, recreational screen time, and sleep and sociodemographic factors. Children's weight status was based on World Health Organization (WHO) categories. Meeting the PA recommendation was based on the WHO guidelines for children and adolescents, and meeting the sedentary behavior and sleep recommendations was based on the Canadian/Australian 24‐hour movement guidelines for children and young people. Results A total of 585 children and their parents were included in the study. Overall, 15% of children were overweight or obese. Three‐quarters of children met the sleep guideline, and 50% met the PA guideline. Less than one third of children (31%) met the recreational screen time guideline, whereas 15% met all three guidelines. However, when adjusted for all predictors in the model, maternal education, family income, and child's age were significantly associated with overweight/obesity. Children with higher maternal education level were 2.3 times (AOR = 2.33, 95% CI: 1.19–4.78) more likely to be overweight/obese. Children in families with a higher monthly income had 1.9 times (AOR = 1.95, 95% CI: 1.14, 3.35) higher risk of being overweight/obese. Conclusions Prioritizing maternal education (mother‐centric interventions) can help address the high levels of childhood obesity in Bangladesh.


INTRODUCTION
The World Health Organization (WHO) considers childhood obesity as "one of the most serious public health challenges in the 21st century" (WHO, 2020a(WHO, , 2020b. The prevalence of overweight and obesity in adolescents is defined according to the WHO growth reference for school-aged children and adolescents. More than 41 million children under the age of 5 are estimated to be overweight worldwide. Over two thirds of these children live in low-and middle-income countries (LMICs) (WHO, 2020a(WHO, , 2020b, where the prevalence of childhood obesity has accelerated in recent years (Abarca-Gómez et al., 2017).
Many LMICs, particularly in Southern Asia where there is a history of child malnutrition, now see childhood obesity as a major public health issue (Abarca-Gómez et al., 2017;Hoque et al., 2014;Popkin et al., 2012). Childhood obesity is not well understood by parents in this region and is often perceived as a sign of good health . Against the backdrop of managing undernutrition and life-threatening communicable diseases, childhood obesity is often less prioritized in Southern Asia than undernutrition (Khan & Talukder, 2013).
Childhood obesity is influenced by many lifestyle, socioeconomic, genetic, and environmental factors (Gupta et al., 2012;Sahoo et al., 2015). Systematic reviews have shown that physical activity (PA) is negatively associated with adiposity in children Jiménez-Pavón et al., 2010;Poitras et al., 2016). More recently, time spent in 24-hour movement behaviors (sleep, screen time, and PA) is associated with childhood obesity (Roman-Viñas et al., 2016).
WHO's Commission on Ending Childhood Obesity has also recognized the importance of movement behaviors in preventing obesity among children (WHO, 2016).
Bangladesh is a densely populated middle-income country with a population of over 165 million. It is undergoing rapid unplanned urbanization whereby the proportion of people who live in urban areas will increase from 32% in 2020 to more than 50% by 2040 (Hayes & Jones, 2015). Concomitantly, the prevalence of childhood obesity (which includes overweight) has increased from 3.6% in 2000 to 9% in 2016 (Biswas et al., 2017;UNICEF, 2019). Data on the behavioral epidemiology of childhood obesity are scarce in Bangladesh (Biswas et al., 2017). Most studies among school children residing in the capital city, Dhaka, (population ≈ 18 M) have focused on the prevalence but not the risk factors or correlates of childhood obesity (Biswas et al., 2017).
Overweight/obesity (OWOB) should be prevented as early as possible as children with excess weight are likely to become obese in adulthood (Geserick et al., 2018). Targeting preschool-aged children could be highly effective as diet and physical habits are usually developed during preschool ages (Hodges, 2003). Identification of modifiable risk factors or correlates is critically important to develop intervention strategies for preventing childhood obesity in LMICs. This study aimed to: (a) determine the prevalence of OWOB among preschool children (4-7 years) in a district town and (b) to identify how OWOB is associated with sociodemographic characteristics and adherence to PA guidelines.

Setting and participants
This cross-sectional study followed the STROBE (strengthening the reporting of observational studies in epidemiology) statement for the transparent reporting of observational studies (von Elm et al., 2014).
A list of kindergarten schools (n = 54) in the Jamalpur district town

Questionnaire
In this study, we conducted secondary analyses of a dataset from a previous study. The study used a questionnaire used was part of a larger survey that aimed to study maternal perception   and 11 h of uninterrupted sleep per night. PA was operationalized as time spent in energetic play outside. For young children, the term energetic or active play has been used in recent PA guidelines developed for children in Australia and in the Asia-Pacific (Loo et al., 2023;Loo et al., 2022;Okely et al., 2017). We have found that it is an easier concept for parents to understand than MVPA. Using energetic or active play to operationalize PA has been validated in a parent-reported questionnaire against accelerometry and found to be adequate (Okely et al., 2009).
All data collectors underwent 2 days of training and orientation.
This was provided by the same trainer using a set protocol. The trainer checked that data collectors administering the protocol in the same way and data collectors rehearsed the data collection procedures on one another and children as part of their training.

Outcome variables
Weight status: The height of the participating children was measured using a stadiometer (Seca, Hamburg, Germany) with a unit of 0.1 cm.
Digital scales with a unit of 0.01 kg (Mega, Dhaka, Bangladesh) were used for assessing weight. Body mass index was calculated using height and weight measurements. Children removed their shoes and any heavy clothing before the height and weight measurements. All measures were taken once and then recorded.

Predictor variables
Predictor variables were PA, sedentary recreational screen time, sleep, and selected sociodemographic factors (child sex, child age, maternal and paternal education level, and monthly family income). Parents who wished to participate in the study attended the school and gathered in the school hall. All parents who attended were mothers and will be referred to as mothers from this point forward. A self-or interviewer-

Statistical analyses
Children were classified as underweight, normal weight, overweight,

Characteristics of study participants
A total of 649 children and their mothers from 12 preschools participated in the study, of which data from 585 participants (mother and child) were eligible for analysis after removing randomly missing data (9.9%). The mean age (SD) of included children and mothers was 5.5 years (SD 0.9) and 28.4 years (SD 4.6), respectively. Of the participants, 57.1% were boys. Nearly all mothers (98%) lived with their husbands and over 87% were housewives. Compared with children from Noorani preschools, children from secular preschools had a significantly higher socioeconomic status based on maternal education and family income (Table 1).

DISCUSSION
In this study, we found that there was a higher prevalence of obesity among boys and among children who attended secular preschools and higher maternal education and higher family income were associated with an increased likelihood of childhood overweight and obesity.
In our study, the combined prevalence of OWOB was 14%. A recent meta-analysis has reported a similar prevalence (13%) for children aged 0-19 years in this country (Biswas et al., 2017). Nationally representative data on OWOB among school children are lacking in South Asian countries. Another meta-analysis estimated that the combined prevalence of OWOB was 19.3% in India (Ranjani et al., 2016) and 14% in Pakistan (Warraich et al., 2009). A third metaanalysis at the subcontinent level (India, Bangladesh, and Pakistan) found that the prevalence of OWOB was 2%-6% among rural children and adolescents, 16%-18% in average urban settings (nonaffluent), and 23%-36% among affluent urban children (Hoque et al., 2014).
The rates of OWOB in wealthy countries are generally higher in rural areas and among people of lower socioeconomic status but the reverse is observed in LMICs (PHE,; Popkin et al., 2012). Higher rates of OWOB among urban affluent children residing in cities in South Asia are comparable to the situation in high-income countries. These findings reinforce that major lifestyle changes are happening in rapidly urbanized South Asian countries, and OWOB is emerging as a major public health threat in these countries.
We found that the prevalence of obesity was nearly two times higher in boys than girls, whereas the prevalence of overweight was similar. These findings for obesity were similar to another study in Bangladesh (Das et al., 2013). Potential reasons for a higher proportion of obesity among boys include the patriarchal culture in Bangladesh where boys are generally more favored than girls. We also found that the prevalence of obesity was higher among children of secular schools compared to their Noorani counterparts. This disparity could be due to more socioeconomic disadvantaged conditions among families of Noorani school children whose parents have relatively less education and lower family income. This finding is consistent with another study in Bangladesh .
In our study setting, the proportion of preschool-aged children who met all PA, screen time, and sleep guidelines (15%) was similar to several other countries, including Australia (14.9%), China (15%), Sweden (19.4%), and Canada (12.7%) (Chaput et al., 2017;Cliff et al., 2017;Delisle Nyström et al., 2020;Guan et al., 2020). On the contrary, in our recent study conducted in Dhaka city (the most densely populated megacity on earth), a lower proportion (4.7%) of preschool-aged children met all guidelines . The proportion of children in the present study who met the screen time recommendation was found to be 31%, which was similar to Sweden (37.8%) and South Africa (48%) (Delisle Nyström et al., 2020;Draper et al., 2020). The proportion of children who met the PA guideline was approximately 50%, which was lower than several other countries, such as Sweden (∼90%)  our survey reported a lack of adequate neighborhood facilities such as playgrounds for their children. Our field-level inspections revealed that there were no playgrounds in two thirds of the participating schools and none of the schools included sports activities in their curriculum.
In this context, intervention strategies involving families and schools must be prioritized in developing countries like Bangladesh, which is undergoing rapid urbanization and digitization.
We found that a higher proportion of children from Noorani backgrounds met all guidelines as compared to that of secular schools.
Cultural and socioeconomic factors might explain this finding. Noorani families are more economically disadvantaged and practice more conservative lifestyles. This means that their children would have less access to electronic media devices and spend more time playing outdoors In the present study, higher maternal education and higher family income were associated with higher odds of being overweight/obese.
These findings are consistent with our prior study in children under 5 years of age , and studies reported from some However, interestingly, in our study setting, most participating mothers were stay-home mothers (87%) who did not perceive TA B L E 5 Unadjusted (UOR) and adjusted odds ratios (AOR) and 95% confidence intervals (95% CI) for sociodemographic and movement behaviors and overweight and obesity among children.  . Therefore, prioritizing mother-centric interventions is expected to be effective in preventing childhood obesity in LMICs, including Bangladesh (Hossain et al., 2020).

Limitations
Because of the stratified nature of sampling that was employed, the results are representative of the target population of 4-7-yearold school children in Jamalpur. The findings provide a baseline for similarly sized district towns in Bangladesh, a country with a large population with homogeneous characteristics with respect to language and sociocultural perspectives (MdB & Hossain, 2019). Due to the cross-sectional nature of the study, it was not possible to make firm conclusions about correlates as the outcome (overweight and obesity) and exposure were assessed at the same time point and data on temporal sequence were not available. Adjusted odd ratios were derived using binary regression to incorporate the effects of confounders.
However, this would not include the effect of any other factors not included in the study and in the analyses. For the assessment of PA and sleep, we relied on parent-report, rather than device-based measurements using accelerometry. Parent-reports from high-income countries have been shown to overreport time spent in PA and in sleep, although it is not clear if this is also true in LMICs. We operationalized MVPA as time spent in energetic play outside, which has been shown to be adequately valid (Okely et al., 2009), although it may have been difficult for mothers to report accurately about their child's daily play when their child was at school. We chose to use the guidelines for school-aged children (rather preschool-aged children) because only a very small number of children in the study were aged less than 5 years (n = 67). Moreover, our participating schools were mainly part of primary-level schools where there were sections for preschool children (KG-I, KG-II).

CONCLUSIONS
This is the first study to examine associations between 24-hour movement behaviors and childhood overweight/obesity among communitylevel preschoolers in Bangladesh. In our study, children from affluent families and highly educated were more likely to be overweight/obese.
Our findings provide a baseline for future studies and public health policies and interventions to reduce the increasing level of obesity in