Impact of socio-economic, demographic and cultural factors on the development of children’s mental health: a cross-sectional study in Bangladesh

Abstract

Mental health in childhood brings up the state of a child’s emotional and psychological well-being, encompassing their ability to navigate challenges, form relationships, and get by stressors, laying the foundation for their future mental and emotional resilience. This cross-sectional study, conducted between December 1, 2023, and February 1, 2024, intended to identify the substantial risk factors of children’s mental health (CMH) development in Bangladesh. The data were collected using two-phase sampling through a questionnaire filled out by trained interviewers. To test the normality of the dependent variable, Kolmogorov–Smirnov and Shapiro–Wilk tests were applied, and the tests confirmed the non-normality assumption (P < 0.05) of the study variables. Mann–Whitney U and Kruskal–Wallis H tests were applied as bivariate analyses, and generalized gamma regression was performed to determine the significant risk factors of mental development. The odds ratio (OR) and 95% confidence interval (CI) were the measuring parameters of the risk factors of CMH. The results revealed that early childhood disease (OR 0.9537, CI 0.93–0.97, p = 0.000769), monthly family income (OR 0.9247, CI 0.87–0.98, p = 0.015046), and providing supplementary food sometimes (OR 1.0583, CI 1.02–1.09, p = 0.001819) were significant socio-economic factors influencing CMH. Among demographic factors, gender (OR 0.9693, CI 0.94–0.99, p = 0.025684) and division (OR 1.0496, CI 1.007–1.09, p = 0.021429) showed significant associations. Additionally, child opportunities to play outside (OR 0.9451, CI 0.91–0.98, p = 0.007235), which may be shaped by cultural attitudes toward child supervision, gender roles, and norms around outdoor activity, was also found to be significantly associated with children’s mental health. The results of this study would assist policymakers to take initiatives in mental health development including the funding of interventions so that children in Bangladesh can achieve their developmental potential.

Introduction

Children’s mental health (CMH) development refers to the gradual acquisition and enhancement of emotional, psychological, and social skills throughout life. It encompasses various aspects, including managing emotions, forming positive relationships, coping with stress, making decisions, and adapting to life’s challenges. It is vital to recognize and address mental health issues early on, as they can significantly impact later life. Several factors contribute to children’s mental health, including genetic predisposition, environmental influences, family dynamics, and life experiences. Recently, there has been a growing acknowledgment in the UK, Europe, and worldwide of the significance of mental health and well-being to overall health [1]. Approximately 200 million children in developing countries are incapable of growing to their full potential during their first five years of life due to factors including poverty, malnutrition, and inadequate health care [2, 3].

An analysis of data using the Early Childhood Development Index (ECDI) across 35 low and middle-income countries estimate that 80.8 million children ages 3 and 4 y in LMICs experienced low cognitive and socio-emotional development in 2010, with the largest number of affected children in sub-Saharan Africa, followed by South Asia and the East Asia and Pacific region [4]. Maternal demographics, birth-weight, gestational age, and emotional development in children were found to be related to mental health and emotional development in British Columbia and Canada [5]. Increasing maternal age (15 years to about 30 years) was associated with a decreased risk of developmental vulnerability in contrast, maternal age beyond 35 years was mostly related to increased susceptibility in Australia [6]. According to estimates, 17.2% of children in Vietnam, did not reach their full developmental potential within the first five years of life [7].

According to the World Health Organization (WHO), Bangladesh is one of the ten countries that host the most disadvantaged children at extreme risk of impaired cognitive and social-emotional development [8]. In Bangladesh, various obstacles hinder the fulfilments of the UN Sustainable Development Goals and the readiness of children for school such as family poverty, early marriage, child malnutrition, and inadequate maternal healthcare [2]. The National Children’s Policy 2011 in Bangladesh aimed to raise a scientifically inquisitive generation, which could keep pace in the modern world [9]. The main goal of this policy is to raise awareness of the beneficial effects that fostering learning and kid-friendly environments for young children’s development, particularly concerning their positive learning styles and cognitive growth.

Different studies have revealed that socio-demographic factors, such as gender, economic status, place of residence, and maternal education contribute to early development trajectories [10]. Children who were breastfed had mothers with higher education levels, resided in urban areas, and were engaged in four or more stimulating activities and reading books with family members, demonstrated a higher likelihood of consistent and positive developmental trajectories [11]. Literature is also evident that children’s psychological and socio-emotional development is adversely impacted by their mother’s mental health conditions; however, these effects fade between the ages of 11 and 13 [12]. Socioeconomic status affects children’s well-being at multiple levels, including both neighbourhood and family. Its effects are moderated by children’s characteristics, family characteristics, and external support systems [13].

Stress during pregnancy may be a risk factor for future developmental problems and appears to be one of the determinants of delay in motor and mental development in infants of 8 months of age [14]. Higher levels of prenatal anxiety, nonspecific stress, and depressive symptoms is also associated with more advanced motor development in children after postnatal control for each psychological measure [15]. A child born to an older mother is more likely to experience physical problems such as anomalies [16], short birth-weight, or early gestational age, which may affect brain functioning and impair the best possible cognitive and socio-emotional development. To facilitate effective support strategies, it is imperative to identify the underlying factors influencing children’s mental health.

However, research on the determinants of children’s mental well-being in Bangladesh remains scarce, with a notable lack of comprehensive insights into the role of socioeconomic, demographic, and cultural factors. To address this gap, we propose a cross-sectional study spanning the central and south eastern regions of Bangladesh. This study aims to investigate the impact of socioeconomic, demographic, and cultural factors on the development of children’s mental health. More specifically, we seek to uncover potential risk factors, including neonatal conditions, parental characteristics, and socioeconomic status, contributing to mental health outcomes among children in Bangladesh.

Methodology

Study design

This cross-sectional study aims to assess the socio-economic, demographic, and cultural factors that significantly affect the development of mental health among children aged 6 to 10 years in Bangladesh. Data were collected from 33 schools in four division-Dhaka, Chittagong, Mymensingh and Rajshahi. The survey followed a two-phase sampling approach. In the first phase, four out of the eight administrative divisions in Bangladesh Dhaka, Chittagong, Mymensingh, and Rajshahi were randomly selected to ensure geographic diversity. In the second phase, primary schools within these divisions were chosen based on several criteria: willingness to participate, accessibility, administrative approval, and variation in location (urban vs. rural), school size, and student demographic profiles. A total of 40 schools were initially approached through written requests, selected for convenience and their active operational status during the study period. Permission for data collection was granted by 33 of these schools, from which the final sample was drawn. Dhaka Division, located in the central part of Bangladesh, is the country’s most populous and economically significant region. It is bounded by Mymensingh division to the north, Chittagong division to the east, and Rajshahi division to the west. Chittagong, Rajshahi, and Mymensingh are three distinct divisions of Bangladesh, each contributing uniquely to the nation’s cultural tapestry, economic development, and historical heritage. Rajshahi is a major administrative, educational, cultural, and commercial hub in Bangladesh. It is a historically significant silk production center. Sometimes the city is known as the "City of Education." Chittagong division is geographically the largest of the eight administrative divisions of Bangladesh and the country’s main port city. Among the selected schools, 17 were categorized as Government Primary Schools and 16 Private Primary Schools. All the schools were mixed-sex, and their student populations ranged from 30 to 300 pupils. The inclusion criterion for this study was children aged between 6 and 10, and whose parents were present at the school on visiting day and gave verbal consent for the child to participate. Most of the questions were related to parents’ socioeconomic, demographics, and children’s mental development. Therefore, we have taken consent and interviews from parents only. Children absent in school or not given consent by parents were considered exclusion criteria of this study. Parents willing to participate in the study but unable to respond to the questionnaire due to memory lapse and limited knowledge were also excluded from the survey.

Dependent variable

The dependent variable for this study is Children’s Mental Health, which is measured by children’s learning ability, Problem-solving ability, Concentration quality, Academic performance, Understanding ability, Memory ability, Imagination ability, Level of consciousness, Moral concepts, Attitude and insights, Affect and mood, Thoughts and perceptions. All of these variable were measured in a 5-point Likert scale and coded as (Very Poor = 0, Poor = 1, Fair = 2, Good = 3, Very Good = 4). This scale was selected due to its simplicity, ease of administration, and suitability for capturing subjective assessments in a culturally diverse and resource-limited context like Bangladesh. This approach was guided by existing literature and large-scale surveys (e.g., the National Survey of Children’s Health, U.S.; UNICEF MICS), where similar ordinal response formats have been successfully employed.

Explanatory variable

The explanatory variable included a set of socioeconomic and demographic characteristics namely Child age (6–10) and Gender (boy, girl); Parents’ age, Place of family residence (Urban, Rural); Number of household members, Types of house was included in three categories (Single-family homes, Apartments, Multi-family homes); Family types (Joint family, Nuclear family). Demographic Birth factors included birth weight (< 2.5 kg, 2.5–3.5 kg, > 3.5 kg); Child’s delivery by caesarean section (yes, no); premature birth (yes, no); breastfeeding status in first 6 months (Always, Breastfeed and formula milk, Always formula milk); Mother poor nutrition and psychological stress during pregnancy ( yes, no); and Have taken drug related to Calcium during pregnancy (yes, no); Mother’s age at the time of birth of that children (less than 18, 18–25, 25–30, 30 + years); Parents level of education in four categories (Illiterate, primary, secondary, higher); Father occupation in six categories (government job, private job, daily work, unemployed, remittance worker, others); Monthly family income in four categories (less than 10 k, 10 k–20 k, 20 k–40 k, more than 40 k) and religion in 3 categories (Muslim, Hindu and Others).

Data collection tool

After carefully revising the previous literature, the authors prepared the questionnaire for this study. The questionnaire included information on parents’ health status during pregnancy and complications of the child during and after birth, like early childhood disease, premature birth, etc. Trained interviewers collected the responses by direct interview method from the parents. The responses related to mental health development were based on parents’ observations of their children. The questionnaire was designed in three major sections. The first section included neonatal information like the age of the children (number), gender (Male, Female), religion (Muslim, Hindu), types of school (government or private), Birth weight (< 2.5 kg, 2.5–3.5 kg, > 3.5 kg), early childhood disease (yes, no), premature birth (yes, no), child’s delivery by caesarean section (yes, no), breastfeeding status in the first 6 months (always, breastfeed and formula milk, always formula milk), have given/giving child supplementary food/milk (e.g. Horlicks, Maltova. Etc) are added in four categories (always, sometimes, seldom, never), give child opportunity to play /going outside/meeting with friends (always, sometimes, seldom, never). The second section contains information on parental socio-demographic characteristics and health status. The variables included in the second section were the Father’s age, Mother’s age, Number of household members, Father’s and Mother’s education (Less than secondary level, Secondary level, Higher secondary level), Father’s occupation (Not employed, Private service, Government service, Daily work, Remittance worker), Monthly family income (less than 10 k, 10 k-20 k, 20 k-40 k, more than 40 k), Family types (Joint family, nuclear family), Types of place of residence (Urban, Rural), Mother’s age at the time of birth of children (less than 18, 18–25, 25–30, 30 +), Maternal history of specific drug use during pregnancy (No, Yes), Poor nutrition during pregnancy (No, Yes), Psychological stress during pregnancy (yes, no). Socioeconomic status was measured by key variables parents’ education, parents’ occupation, family types (joint/nuclear), present residents, and monthly family income. Finally, socioeconomic status was categorized into three classes- low, middle, and high. The information on the psychological stress of mothers during pregnancy was collected based on their self-assessment and the use of antidepressant drugs at that time. In the third section, 12 Likert scale variables were considered to assess the dependent variable of children’s mental health condition (CMHC) and discussed in detail in the next section.

Assessment of children’s mental health condition (CMHC)

The current study considered 12 items (Fig. 1) to distinguish children’s mental health. Each item appeared with 5 levels coded 0, 1, 2, 3, and 4. Where 0 indicates that the child’s mental health is very poor, 1 for poor mental health, 2 indicates the mental health condition is fair, 3 indicates the mental health condition is good, and 4 for very good mental health. The items considered for the CMHC are the child’s learning ability, Problem-solving ability, Concentration quality, Academic performance, Understanding ability, Memory ability, Imagination ability, Level of consciousness, Moral concepts, Attitude and insights, Affect and mood and Thought and perception. By adding all 12 items, each child could receive a score from 0 to 4 on each of the 12 items, resulting in a Total Mental Health Score (TMHS) ranging from 0 (lowest possible mental health status) to 48 (highest possible mental health status).

Fig. 1.

Indicators of mental health

To standardize the score across all children, we calculated the Child Mental Health Score (CMHSc) [17], by dividing each child’s TMHS by the maximum possible score of 48:

1

The range of the CMHSc lies between 0 and 1, i.e., .

The score tending from 0 to 1 indicates that the severity of MH increased where 0 indicates very poor mental health, 1 indicates excellent mental health.

Sampling and data collection

This study used two-phase sampling to select Government and Private Primary Schools and participants for collecting data on children’s mental development. Firstly, four administrative divisions named: Dhaka, Chittagong, Rajshahi and Mymensingh in Bangladesh, were chosen randomly. As per the Annual Primary School Census 2021, there were a total of 1,18,891 primary-level educational institutions in Bangladesh, including 65,566 government primary schools, and 4,799 private primary schools. Written applications were sent for data collection to 40 schools convenient to the authors and actively running during the study period. Permission has been obtained from the authority of 33 institutions for data collection. This research used Cochran’s formula for the finite population to determine the required sample size. The Cochran’s formula for sample size determination for the finite population is

2

where N is the size of the finite population, is the sample size for an infinite population, and can be defined as

3

Here, z is the critical value of the excepted confidence level, p is the proportion of a certain attribute presented in the population, q = 1-p, and e is the level of precision. Now if we consider a 5% level of significance or 95% confidence interval, we have z = 1.96, considering the expected proportion of the study’s attribute is 50%, i.e., p = 0.5, q = 1 − p = 0.5, and e = 0.05. Putting all these values in Eq. (2), we can get = 384.16 384. According to the Ministry of Primary and Mass Education, the estimated number of primary school children in Bangladesh is 16,230,000. Hence, the population size for this study is N = 16,230,000. Now, using the value of N and , we have from Eq. (1)

4

The adequate sample size for this study population is 384, and this investigation targeted 384 children for data collection. However, the study reached 450 parents, considering the possibility of incomplete response, participant withdrawal from the study, etc. Finally, we got complete responses from 401, with a response rate of 89.11%. A group of skilled and trained interviewers collected data through face-to-face interviews with the parents. Furthermore, the parents were informed that completing the questionnaire would take about 5–10 min. All the data have been recorded in an Excel sheet and used for further analysis. Data collection took place between December 1, 2023, and February 1, 2024. A detailed illustration of the children’s mental health considered in this study has been presented in Fig. 1. As the children cannot respond to their mental, socioeconomic and demographic conditions, the data has been collected from their parents.

Statistical analysis

The dependent variable constructed in this study is a numeric variable (CMH), which ranges from 0 to 1. The increasing value of CMH from 0 to 1 indicates good mental health conditions. To test the normality of the outcome variable Kolmogorov–Smirnov and Shapiro–Wilk tests were applied by using SPSS v25, and the significance of the test result confirmed the non-normality assumption (P < 0.05) of the study variables. First, the frequency distribution of all the explanatory variables (Children, parental socio-economic, and demographic features) was constructed to understand the general information about the study. Secondly, Bivariate analyses were conducted using the Mann–Whitney U test and the Kruskal–Wallis H test in SPSS. The non-parametric Mann–Whitney test was applied as the numeric outcome variable and the independent variables are qualitative (2 categories). Similarly, the Kruskal–Wallis H test was used in the same situation but when explanatory qualitative variables have 3 or more categories. The variables found significant in the bivariate analysis were further considered in the multivariate analysis. Two generalized linear regression models (GLRM): generalized linear gamma regression (GLGR), and generalized linear beta regression (GLBR) models, were implemented using R software (version 4.3.2) to assess the significant impact of different socioeconomic and demographic variables on Mental Health. The Akaike information criterion (AIC) and Bayesian information criteria (BIC) have been calculated for both models to identify the best one. The AIC and BIC value for the GLGR model (AIC: -488.9162, BIC: -365.1034) was found least compared to the GLBR model (AIC: -444.6004, BIC: -320.7876) indicating that the generalized gamma regression model is better than the beta regression model. Hence, the results of the GLGR model have been extracted for this study and prepared the results accordingly.

Result

Figure 1 illustrates the frequency of the indicators of mental health considered in this study. Specifically, 45.7% of children demonstrated good learning ability (n = 183), 48.4% problem-solving ability (n = 194), 43.9% exhibited good concentration quality (n = 147), and 48.7% favorable academic performance (n = 195).

Furthermore, the study found that 48.7% (n = 195) of children displayed good understanding ability, 42.4% (n = 170) showed good memory ability and 40.9% (n = 164) showed good imagination ability. Similarly, the level of consciousness was within normal limits of 41.2% (n = 165), indicating a generally stable state. Additionally, a substantial proportion of children displayed strengths in moral concepts 40.6% (n = 163), attitude and insights 39.7% (n = 159), and thoughts and perceptions 40.3% (n = 162). These findings underscore the multifaceted nature of children’s mental health, encompassing cognitive, emotional, and moral dimensions.

While the majority of indicators showed favourable results, some areas exhibited slightly lower percentages. For instance, affect and mood 35.8% (144) indicated a need for additional attention to emotional well-being among specific children.

Characteristics of socioeconomic and demographic features

General frequency distribution tables have been prepared to understand the characteristics of this study’s explanatory variables and are presented in.

Tables 1 and 2. Age distribution revealed that the largest proportion of children fell within the 7-year-old category (25.4%), followed by 8 year-olds (19.5%). Gender distribution was nearly equal, with 50.6% girls and 49.4% boys participating. In terms of school type, a majority attended government schools (68.1%) compared to private schools (31.9%). The religious composition leaned towards Muslim participants (82.8%) compared to Hindu participants (17.2%). Birth weight varied, with the majority falling within the 2.5–3.5 kg range (52.9%). A notable portion of children (55.6%) had experienced early childhood diseases, while an equal proportion was delivered via caesarean Sect. (48.6%). A significant number of children were not born prematurely (63.8%) and were breastfed along with formula milk during the first 6 months of life (56.4%). Additionally, a majority of parents provided supplementary food/milk to their children sometimes (70.6%) and granted them opportunities to play/go outside/meet friends sometimes (46%).

Table 1.

The frequency distribution of the demographic characteristics of children (n = 401)

Variables	Categories	Frequency	Percent
Child’s age	5	61	15.2
	6	54	13.5
	7	102	25.4
	8	78	19.5
	9	61	15.2
	10	45	11.2
School Type	Private	128	31.9
	Government	273	68.1
Gender	Girl	203	50.6
	Boy	198	49.4
Religion	Hindu	69	17.2
	Muslim	332	82.8
Birthweight	< 2.5 kg	143	35.7
	2.5–3.5 kg	213	52.9
	> 3.5 kg	45	11.2
Early childhood disease	No	178	44.4
	Yes	223	55.6
Child’s Delivery by Caesarean Section	No	206	51.4
	Yes	195	48.6
Premature Birth	No	256	63.8
	Yes	145	36.2
Breastfeeding Status in the first 6 months of the child	Always	157	39.2
	Breastfeed and formula milk	226	56.4
	Always Formula milk	18	4.5
Have given/giving child supplementary food/milk (e.g. Horlicks, Maltova. etc.)	Always	70	17.5
	Sometimes	283	70.6
	Seldom	40	10.0
	Never	8	2.0
Give the child an opportunity to play/go outside/meet with friends	Always	136	33.9
	Sometimes	188	46.9
	Seldom	65	16.2
	Never	12	3.0

Table 2.

The frequency distribution of the parent’s socioeconomic and demographic characteristics

Variables	Categories	Frequency	Percent
Are you currently using any family planning method	No	104	25.9
	Yes	297	74.1
Father’s education level	Illiterate	17	4.24
	Primary	65	16.21
	Secondary	197	49.13
	Higher	122	30.42
Mother’s education level	Illiterate	33	8.2
	Primary	123	30.7
	Secondary	192	47.9
	Higher	53	13.2
Father’s occupation	Government jobs	93	23.2
	Private jobs	129	32.2
	Daily work	59	14.7
	Remittance Worker	85	21.2
	Others	35	8.73
Monthly family income	Less than 10 k	22	5.5
	10 k-20 k	103	25.7
	20 k-40 k	207	51.6
	More than 40 k	69	17.2
Types of house	Single-family homes	176	43.9
	Apartments	105	26.2
	Multi-family homes	120	29.9
Types of Place of Residence	Rural	163	40.6
	Urban	238	59.4
Mother’s age at the time of birth of that child (in Years)	Less than 18	31	7.7
	18–25	239	59.6
	25–30	102	25.4
	30 +	29	7.2
Family Type	Nuclear Family	215	53.5
	Joint Family	186	46.4
Mother’s poor nutrition during pregnancy	No	192	47.8
	Yes	209	52.0
Psychological Stress during pregnancy	No	133	33.1
	Yes	268	66.7
Have taken drugs related to Calcium during pregnancy	No	146	36.3
	Yes	255	63.4

Regarding family planning, Table 2 shows that the majority of parents (74.1%) currently use a family planning method. In the case of Fathers’ education level, it is observed that only 4.24% (n = 17) possess less than a primary level, 49.13% (n = 197) possess secondary, and 30.42% (n = 122) of them possess higher education. Table 2 revealed that 13.2% (n = 53) of the mothers possess higher education. Father’s occupations varied, with a notable representation in private jobs (32.2%) and government jobs (23.2%). Monthly family income distribution ranged from less than 10 k (5.5%) to 20 k–40 k (51.6%). In terms of housing, single-family homes were the most common (43.9%), followed by apartments (26.2%) and multi-family homes (29.9%). Residence types showed a slight predominance of urban areas (59.4%) compared to rural areas (40.6%). Maternal age at childbirth varied, with a significant proportion falling within the 18–25 age range (59.6%). A considerable number of children (n = 215, 53.5%) belong to the nuclear family, while 46.4% (n = 186) child belongs to the joint family. The result showed that 47.8% (n = 192) of the mothers had poor nutrition, and 52.0% (n = 209) had Psychological stress during pregnancy. In addition, Table 2 revealed that 63.4% (n = 255) have taken drugs related to calcium during pregnancy.

Bivariate analysis of the significant mean difference

Table 3 represents the results of bivariate analysis exploring the association or significant mean rank or median difference of mental health among different categories of the explanatory variables.

Table 3.

Mann–Whitney U test of significant mean difference of Mental Health between the categories of binary explanatory variables

		Mann–Whitney U Test
Variables	Categories	Mean	Std. Deviation	Minimum	Maximum	Test statistics	P value
School type	Private	0.5885	0.15109	0.17	0.88	16,899	0.594
	Government	0.5948	0.13159	0.08	1
Gender	Girl	0.5766	0.13865	0.08	1	17,337	0.017
	Boy	0.6098	0.13551	0.17	1
Religion	Hindu	0.5628	0.15697	0.08	0.88	9796	0.106
	Muslim	0.5988	0.13327	0.17	1
Early childhood disease	No	0.6254	0.12995	0.29	1	15,134	0.00
	Yes	0.5668	0.13891	0.08	1
Child’s delivery by caesarean section	No	0.6023	0.12813	0.25	1	18,638	0.210
	Yes	0.5828	0.14730	0.08	1
Premature birth	No	0.6080	0.13097	0.08	1	15,390	0.004
	Yes	0.5660	0.14613	0.17	1
Are you currently using any family planning method	No	0.5700	0.15863	0.08	0.88	14,237.5	0.271
	Yes	0.6007	0.12941	0.25	1
Type of place of resident	Rural	0.5882	0.12374	0.17	0.88	18,366.5	0.363
	Urban	0.5960	0.14707	0.08	1
Family type	Nuclear family	0.5874	0.1446	0.08	1	19,433	0.625
	Joint family	0.5991	0.13014	0.17	1
Mother’s poor nutrition during pregnancy	No	0.5761	0.13521	0.08	1	17,192	0.013
	Yes	0.6111	0.13896	0.17	1
Psychological stress during pregnancy	No	0.5795	0.13373	0.08	1	15,066.5	0.011
	Yes	0.6196	0.14290	0.25	1
Have taken drug related to calcium during pregnancy	No	0.5819	0.13216	0.17	.92	17,343	0.252
	Yes	0.5991	0.14104	0.08	1

Considering a 5% level of significance the result of Mann–Whitney U tests demonstrate no significant association of CMH between school types (Mann–Whitney U = 16,899, p > 0.05) and religion (Mann–Whitney U = 9796, p > 0.05). However, genders (Mann–Whitney U = 17,337, p < 0.05) and early childhood disease (Mann–Whitney U = 15,134, p < 0.05) emerge as significant factors associated with mental health. Specifically, premature birth (Mann–Whitney U = 15,390, p < 0.05), mother’s poor nutrition (Mann–Whitney U = 17,192, p < 0.05) and psychological stress during pregnancy (Mann–Whitney U = 15,066.5, p < 0.05) also demonstrates significance in influencing mental health conditions. Conversely, variables such as child’s delivery by caesarean section, family planning method, type of place of residence, and family type show no significant associations with mental health.

In addition to the Mann–Whitney U test, the Kruskal–Wallis H test for testing the significant mean rank difference of CPHSc has been calculated and presented in Table 4. A significant difference is found between the children who have the opportunity to play or going outside or meeting with friends (Kruskal–Wallis H = 10.824, P < 0.05) and the highest mean mental health score was observed in the children whose always given opportunity to play/going outside/meeting with friends (0.6239 . Given child supplementary food/milk (Kruskal–Wallis H = 6.309, P < 0.05) and the division (Kruskal–Wallis H = 12.326, P < 0.05) were found to be significant with mental health outcome. Furthermore, significant disparities are observed based on the fathers education levels (Kruskal–Wallis H = 14.326, P < 0.05) and fathers occupation (Kruskal–Wallis H = 11.177, P < 0.05) with higher-educated fathers and fathers who do private jobs correlating with higher mental health scores. Additionally, monthly family income (Kruskal–Wallis H = 28.517, P < 0.05) and the type of house (Kruskal–Wallis H = 6.012, P < 0.05) also exhibit significant associations with mental health outcomes, with lower income households and individuals living in multi-family homes reporting lower mental health scores. Moreover, the table showed that mother’s age at the time of birth of that child’s were also significant (Kruskal–Wallis H = 8.844, P < 0.05) with the score of mental health, with individuals born to mothers aged less than 18 showing lower mental health scores (0.5524 compared to other age groups.

Table 4.

Kruskal–Wallis H test of significant mean difference of CMHSc among the multiple categories (more than two) explanatory variables

		Kruskal Wallis H test
Variables	Categories	Mean	Std. Deviation	Minimum	Maximum	Test statistics	P value
Division	Dhaka	0.5730	0.13780	0.17	0.88	14.408	0.002
	Rajshahi	0.6229	0.11360	0.38	1
	Chattogram	0.5680	0.13089	0.08	1
	Maymansing	0.6138	0.15576	0.21	0.88
Birth weight	< 2.5 kg	0.5797	0.14947	0.17	1	4.796	0.091
	2.5–3.5 kg	0.5933	0.13032	0.08	1
	> 3.5 kg	0.6313	0.13028	0.33	.92
Breastfeeding status in the first 6 month of child	Always	0.5994	0.13988	0.17	1	8.173	0.17
	Breastfeed and formula milk	0.5962	0.13321	0.17	1
	Always Formula milk	0.4931	0.14938	0.08	0.71
Given child supplementary food/milk	Always	0.5514	0.17920	0.17	1	8.629	0.035
	Sometimes	0.5981	0.12217	0..08	1
	Seldom	0.6167	0.13874	0.29	0.88
	Never	0.6563	0.18332	0.38	0.88
Give child opportunity to play/going outside/meeting with friends	Always	0.6239	0.14700	0.17	1	10.824	0.013
	Sometimes	0.5831	0.12333	0.21	1
	Seldom	0.5577	0.15498	0.08	0.79
	Never	0.5833	0.08704	0.42	0.71
Father’s education level	Primary	0.657	0.0292	0.47	0.95	12.326	0.0012
	Secondary	0.5797	0.13528	0.08	0.88
	Higher	0.6292	0.14405	0.25	1
Mother’s education level	Illiterate	0.5682	0.08704	0.42	0.75	5.650	0.130
	Primary	0.5766	0.13699	0.08	0.88
	Secondary	0.6051	0.13969	0.17	1
	Higher	0.6012	0.15617	0.25	1
Father’s occupation	Government jobs	0.6013	0.13359	0.29	1	11.177	0.025
	Private jobs	0.6079	0.13667	0.21	1
	Daily work	0.5465	0.10409	0.29	0.71
	Remittance Worker	0.5887	0.15796	0.08	0.88
	Others	0.6049	0.14331	0.25	0.88
Monthly family income	Less than 10 k	0.5591	0.11351	0.29	0.92	28.517	0.00
	10 k-20 k	0.4886	0.14155	0.08	.67
	20 k-40 k	0.61687	0.13645	0.21	1
	More than 40 k	0.6045	0.15135	0.17	0.88
Types of house	Single-family homes	0.5856	0.14536	0.08	1	6.012	0.049
	Apartments	0.6166	0.12980	0.21	1
	Multi-family homes	0.5778	0.13190	0.17	0.83
Mother age at the time of birth of that children	Less than 18	0.5524	0.14149	0.29	0.79	8.844	0.031
	18–25	0.5860	0.14263	0.08	1
	25–30	0.6111	0.12373	0.29	0.88
	30 +	0.6279	0.13267	0.25	0.83

Table 5 represents a positive correlation coefficient between mental health conditions and several demographic variables including the child’s age, mother’s age, and number of household members. Positive coefficients signify that as the age of parents increases, there is a corresponding improvement in the mental health condition of their children. Conversely, negative coefficients suggest that with the increase in the number of household members, mental health conditions are deteriorating.

Table 5.

Kendall’s tau_b and Spearman’s rho correlation between mental health and different explanatory variable

			Child’s age	Father’s age	mother’s age	Number of household member
kendall’s tau_b	Mental health	Correlation coefficient	0.199	0.064	0.130	0.098
		Sig. (2-tailed)	0.002	0.076	0.000	0.010
Spearman’s rho	Mental health	Correlation Coefficient	0.151	0.086	0.182	0.129
		Sig. (2-tailed)	0.002	0.084	0.000	0.010

Impact of neonatal, parental socioeconomic, and demographic factors on mental health

The variables found significant in the bivariate analysis were further considered in the multivariate analysis. Table 6 represents the impact of socioeconomic and demographic factors on children’s mental health. Early childhood disease shows a significant negative impact on mental health (OR 0.9537, p = 0.000769), suggesting that children with early childhood diseases are about 4.63% less likely to have good mental health compared to those without such diseases. Infrequent opportunities for outdoor play also negatively impact mental health. Specifically, children who only sometimes (OR 0.9644, p = 0.025740) or seldom (OR 0.9451, p = 0.007235) have the chance to play outside are respectively 3.56% and 5.49% less likely to have good mental health, compared to children who always play outside. The influence of gender seems to be minor but statistically significant compared to boys, to be in better mental condition (OR 0.9693, p = 0.025684). Additionally, children from the Rajshahi division have better mental health outcomes, being 1.05 times more likely (OR 1.0496, p = 0.021429) to be in good health compared to children from other divisions. Moreover income has a significant impact on mental health. Children from lower income families are roughly 7.53% less likely than those from families with higher incomes to be in good mental health (OR 0.9247, p = 0.015046). Furthermore, providing supplementary food sometimes (OR 1.0583, p = 0.001819) or seldom (OR 1.0605, p = 0.026123) significantly improves children’s mental health, with these children being about 5.83% and 6.05% more likely, respectively, to have good mental health compared to those who are never provided supplementary food.

Table 6.

Impact of parental socioeconomic and demographic factors on children’s mental health (Generalized linear gamma regression model)

Variables	Estimates	St. Error	t value	Odds ratio	95% CI of Odds		P value
					Lower	Upper
(Intercept)	0.5139253	0.0488404	10.523	1.6718408	1.5192243	1.8397888	< 2e-16
Early Childhood disease (Yes)	-0.0473975	00.0139744	− 3.392	0.9537082	0.9279412	0.9650271	0.000769
Give child opportunity to play outside (Sometimes)	-0.0362865	0.0162057	− 2.239	0.9643639	0.9342146	0.9954863	0.025740
Give child opportunity to play outside (Seldom)	-0.0564322	0.0208949	− 2.701	0.9451306	0.9072063	0.9846403	0.007235
Give child opportunity to play outside (Never)	-0.0498860	0.0407141	− 1.225	0.9513379	0.8783728	1.0303641	0.221249
Father education (Primary)	0.0305681	0.015324	0.037	1.0405283	0.9535426	1.0724433	0.78271
Father education (Secondary)	0.0238619	0.0329876	0.723	1.0241489	0.9600282	1.0925523	0.469915
Father education (Higher)	0.0421906	0.0371691	1.135	1.0430933	0.9698059	1.1219191	0.257066
Gender (Girl)	-0.0312269	0.0139407	− 2.240	0.9692557	0.9431310	0.9961040	0.025684
Psychological stress during pregnancy(No)	0.0249493	0.0141492	1.763	1.0252631	0.9972212	1.0540935	0.078673
Division(Chattogram)	0.0293067	0.0211742	1.384	1.0297404	0.9878801	1.0733745	0.167167
Division(Maymensing)	0.0248376	0.0201369	1.233	1.0251486	0.9854766	1.0664177	0.218193
Division (Rajshahi)	0.0484528	0.0209741	2.310	1.0496458	1.0073714	1.0936944	0.021429
Father Occupation(Remittance Worker)	-0.0347716	0.0282444	− 1.231	0.9658259	0.9138127	1.0207997	0.219066
Father Occupation(Private jobs)	-0.0014880	0.0247795	− 0.060	0.9985132	0.9511772	1.0482049	0.952150
Father Occupation (Government jobs)	-0.0179068	0.0273674	− 0.654	0.9822526	0.9822526	1.0363784	0.513316
Father Occupation (Others)	0.9992384	0.0323618	− 0.024	0.9992384	0.9378269	1.0646713	0.981230
Monthly family income (10 k-20 k)	-0.0783060	0.0320580	− 2.443	0.9246815	0.8683691	0.9846456	0.015046
Monthly family income (20 k-40 k)	-0.0783060	0.0205201	1.612	1.0336267	0.9928806	1.0760450	0.107862
Monthly family income (More than 40 k)	0.0351794	0.0278532	1.263	1.0358055	0.9807753	1.0939233	0.207372
Types of house (Apartments)	0.0198826	0.0167475	1.187	1.0200816	0.9871415	1.0541208	0.235908
Types of house (Multi-family homes)	-0.0103121	0.0174225	-0.592	0.9897409	0.9565143	1.0241216	0.554287
Mother age at the time of birth of that children(18–25)	0.0236955	0.0256580	0.924	1.0239785	0.9737572	1.0767900	0.356340
Mother age at the time of birth of that children (25–30)	0.0358890	0.0280586	1.279	1.0365408	0.9810765	1.0951408	0.201671
Mother age at the time of birth of that child (30 +)	0.0397072	0.0354369	1.121	1.0405061	0.9706903	1.1153432	0.263223
Have given child supplementary food(Sometimes)	0.0566233	0.0180273	3.141	1.0582571	1.0215188	1.0963167	0.001819
Have given child supplementary food(Seldom)	1.0604611	0.0262854	2.233	1.0604611	1.0072112	1.1165263	0.026123
Have given child supplementary food(Never)	0.0900250	0.0494611	1.820	1.0942017	0.9931070	1.2055875	0.069546
Mothers poor nutrition during pregnancy(No)	0.0162536	0.0139471	1.165	1.0163864	0.9889790	1.0445533	0.244616
Child premature birth(Yes)	-0.0062899	0.0148477	− 0.424	0.9937299	0.9652282	1.0230732	0.672084

Discussion

The study investigated how socioeconomic, demographic, nutritional, parental, and birth-related factors influenced children’s mental health development in Bangladesh. The bivariate analysis found that a parent’s socioeconomic status has a significant impact on their children’s mental health. Prior research in many settings has indicated that children from the richest households are more likely to obtain therapy than those from the poorest households [18]. A previous study indicated that children of young mothers (≤ 19 years) in low- and middle-income countries are disadvantaged at birth and during childhood, with increased risks of low birth weight, preterm birth, stunting, and failure to complete secondary schooling [19]. A population-based longitudinal study in the United States discovered that low levels of household income are connected to various lifetime physical disorders and suicide attempts, and a reduction in household income is associated with an increased risk for incident mental disorders [20].

This study found that maternal poor nutrition during pregnancy had a significant impact on children’s mental health. Previous research revealed that hunger caused developmental delays, and children who were hungry or regularly ill were more likely to experience developmental difficulties [21]. In addition, the study discovered no link between birth weight and children’s mental health. On the contrary, a study in the US found a significant association in children with ‘‘normal’’ birth weights of 2,500–2,999 g and those with birth weights of 3,500–3,999 g to have mental retardation, cerebral palsy, learning disability without mental retardation, and other developmental delay [22]. Research on the population of England has also revealed an importance for early birth characteristics such as birth weight and socioeconomic class on children’s psychological well-being [23]. This study discovered that housing style has a significant impact on children’s mental health, with lower-income households and those living in multi-family homes reporting lower mental health scores. Similarly, a study of Austrian children found that children who live in multiple family dwellings have significantly stronger associations between residential density and a standardized self-report index of psychological health as well as teacher ratings of behavioral conduct in the classroom than their counterparts who live in single-family detached homes or row houses [24].

The study discovered that father education had a significant impact on children’s mental health. A longitudinal study conducted in Indonesia found that, whereas father’s education has a considerable impact on child mental health, mother’s education had no significant association [25]. This study found that children with early childhood disorders are less likely to have good mental health than those without such ailments. A study found that poor physical fitness in childhood is linked to a number of negative health outcomes in children, including high blood pressure, cholesterol, obesity, and chronic disease [26].

Gender also plays a role, with girls having slightly lower rates of excellent mental health than boys. Gender is extensively proven to be a correlate of objectively measured physical activity in youth samples, with boys on average participating in more activity than girls, and the impact magnitude is reasonably stable throughout ages [27]. A recent study demonstrated gender disparities in health, mental growth, and development in early childhood in three groups of medically at-risk neonates, discovering that prematurely born girls showed stronger cognitive and motor development than prematurely born boys [25].

According to this study, psychological stress during pregnancy has a considerable impact on mental health outcomes. Recent research suggests that a mother’s psychological health during pregnancy can dramatically alter learning, motor development, and behavior in offspring [28]. In China, pregnant women with a second child report low to moderate levels of stress [29]. Prenatal anxiety is well known as a potential risk factor for poor birth outcomes such as preterm birth, low birth weight, and preeclampsia [30]. An Australian study discovered that high scores for stressful life experiences during pregnancy are related to BMI, overweight, and obesity in offspring [30].

According to the study, the mother’s age at the time of birth has a considerable impact on the child’s mental health. In many regions of Bangladesh, cultural norms and traditions often encourage early marriage and early childbearing, which can lead to a younger maternal age at first birth. These cultural expectations may result in limited emotional, psychological, or financial preparedness for motherhood, potentially affecting the child’s mental development. A review found that older maternal age appeared to have a protective influence on offspring’s behavioral and cognitive outcome [31]. Similarly, a study in Thailand discovered that rising maternal age was associated with a lower risk of developmental vulnerability among children born to women aged 15–30 years [6]. Therefore, cultural practices influencing maternal age at childbirth can have both direct and indirect effects on a child’s mental health development.

Children who were constantly given the opportunity to play, go outside, or meet with friends had the highest mean mental health score. The outdoors is an open setting where booming voices, large, joyful, and daring motions are tolerated, providing youngsters with a sense of exhilaration and freedom of being and doing [32]. However, cultural norms—especially in more conservative households—can limit children’s freedom to engage in outdoor activities, particularly for girls. These restrictions, rooted in tradition, may reduce opportunities for social interaction and active play, potentially affecting mental well-being. Children’s health and development gain from being exposed to sunlight, fresh air, and natural and live creatures outside [32]. Furthermore, the outdoors is a free, accessible, and constantly changing environment where children can interact with nature, experience natural phenomena, meet new people, and become familiar with their surroundings, allowing them to feel comfortable and independent while moving throughout the community [32].

On the other hand, characteristics including family planning technique, location of residence type, caesarean section birth of the child and family type do not substantially correlate with mental health in this study.

This study highlights the need for targeted mental health interventions addressing socio-economic, demographic, and cultural disparities among children in Bangladesh. Policies should prioritize vulnerable groups to promote equitable child development outcomes.

Limitation and strength

The study’s limitation was the lack of randomization in the selection of schools and children, which could result in selection bias, as schools that declined to participate or were inaccessible might systematically differ from those included, potentially affecting the representativeness of the sample. Additionally, the exclusion of madrasa students, who represent a substantial portion of the school-going population in Bangladesh, limits the generalizability of the findings across the broader educational landscape. The study relied solely on parent-reported data to evaluate children’s mental health using a Likert-type scale, which may have introduced reporting bias due to subjective perceptions and recall limitations. Parents’ understanding, attitudes, and personal stress levels may have influenced their responses. Furthermore, the study asked about the mother’s health status during pregnancy and early life which may have resulted in recollection bias. Although the study looked at a wide range of characteristics, some variables, such as environmental pollutants, such as air pollution, noise exposure, or residential density, which are well-documented determinants of child mental health were not included. This exclusion was primarily due to limitations in data availability. Future studies should consider integrating these dimensions to provide a more comprehensive understanding. The study’s findings have significant practical implications for child health interventions and policy in Bangladesh. The stratified sampling method ensured adequate representation of both urban and rural populations, highlighting disparities in child health and underscoring the need for region-specific health interventions and policies. Future research can extend these findings by investigating longitudinal correlations between the highlighted determinants and long-term child health outcomes.

Conclusion

Mental health is defined as a condition of well-being in which all internal and external body parts, organs, tissues, and cells work normally. This cross-sectional study gives useful information about the elements that influence children’s mental health in Bangladesh. According to the findings, gender, division, and early childhood disorders are all important predictors of mental health. Mother’s poor nutrition, psychological stress during pregnancy, early birth, and additional food all have a significant impact on mental health outcomes. Furthermore, access to play and social interaction were discovered to have a considerable positive impact on mental health. These findings also reflect broader cultural and behavioural contexts, such as the cultural expectations surrounding early motherhood and the social restrictions on children’s outdoor play, which can influence their mental health outcomes. Parental education levels, monthly family income, mother age, and home style are all strongly associated with higher mental health ratings. In contrast, variables such as school type, Caesarean delivery, family planning method, residential location, and family type show no significant associations. Understanding these crucial determinants allows policymakers and public health practitioners to create focused interventions to improve children’s mental health in Bangladesh.

Abbreviations

CMHSc

Child mental health score

GLBR

Generalized linear beta regression

GLGR

Generalized linear gamma regression

MICS

Multiple Indicator Cluster Survey

Author contributions

Conceived and designed the experiments—SA, MOF, RH. Supervision and validation—MOF. Performed the experiments and Analyze—SA, RH, MOF. Interpretation and explanation of the results—SA, RH, MOF. Contributed reagents, materials, analysis tools or data—SA, MOF, RH, SB. Wrote the paper—SA, RH, MOF, SB

Funding

The author(s) did not receive any specific grant for this work.

Data availability

The corresponding author will provide data upon request.

Declarations

Ethics approval and consent to participate

This study received ethical approval from the Ethical Committee of Noakhali Science and Technology University (NSTUEC), with approval number NSTU/SCI/EC/2022/90(B). Before data collection, informed written permission was obtained from the relevant primary school authorities. As the study involved no medical or surgical procedures, informed verbal consent was obtained from the parents (either mother or father) of the participating children, in line with the committee’s guidelines. The research followed the principles of the Declaration of Helsinki. Participants were assured of confidentiality and anonymity, and participation was entirely voluntary, with no financial or material incentives provided.

Consent for publication

Not applicable for this article.

Clinical trial

Not applicable for this article.

Competing interest

The authors declare no competing interests.