The role of maternal employment in explaining inequalities in child nutritional status in Cameroon: a decomposition approach Open Access

Adequate nutrition is vital for early childhood development, healthy growth, organ function and a strong immune system (UNICEF, 2020). The United Nations Sustainable Development Goal (SDG) 2.2 aims to eliminate all forms of malnutrition by 2030 (Abeba et al., 2023). The World Health Organization (WHO) defines malnutrition as imbalances in energy and nutrient intake, encompassing undernutrition, overweight, obesity and diet-related non-communicable diseases (WHO, 2021). Africa bears the brunt of this crisis, hosting more than one-third of the world's malnourished population (UNICEF and Group, 2023). Children under five are particularly vulnerable to undernutrition, which includes underweight, wasting, stunting and nutritional deficiencies, while adults are more commonly affected by overweight and obesity (WHO, 2021). Undernutrition, significantly contributes to childhood illness, disability and mortality, with long-term impacts on brain development, skills acquisition and adult income (Black et al., 2008; UNICEF, 2013).

The 2025 edition of the Joint Child Malnutrition Estimates (JME) highlights that global efforts remain off track to meet the 2025 World Health Assembly (WHA) nutrition targets and the 2030 Sustainable Development Goal (SDG) 2 objectives (World Health Orgnization, 2025). In 2024, an estimated 150.2 million children under age 5 were stunted, 42.8 million were wasted and 35.5 million were overweight globally. While stunting has shown a steady decline over the past decade, it still affected 23.2% of children under age 5 in 2024. Alarmingly, Africa accounts for nearly half (43%) of the global burden of stunting, underscoring the persistent and critical challenge the continent faces in addressing child malnutrition.

In Cameroon, malnutrition remains a significant public health challenge, particularly among vulnerable groups such as pregnant women, nursing mothers and children under five. It is a leading cause of morbidity and mortality in children, second only to malaria (Nguenda, 2018). Malnutrition arises from inadequate nutrition due to poor dietary practices, limited food availability and persistent infectious or parasitic diseases, worsened by poor hygiene conditions (Institut National de la Statistique (INS), 2020).

INS (2020) reports that in 2018, 29% of children under the age of five in Cameroon experienced growth delays or chronic malnutrition, with 14% suffering from severe developmental delays. Furthermore, 11% of children in this age group were underweight, including 3% experiencing severe weight loss. While the prevalence of stunting among children under five decreased from 31% in 1991 to 29% in 2018 and the prevalence of underweight children declined slightly from 12% to 11% over the same period, these reductions remain modest. Despite this gradual decline in undernutrition, the current rate of progress is insufficient to achieve Sustainable Development Goal 3, which aims to eliminate all forms of malnutrition by 2030.

Ongoing security crises in Cameroon have significantly impacted child malnutrition, particularly in vulnerable regions. A 2022 nutritional survey revealed deteriorating conditions among children aged 6–59 months in the Far North, North, Adamaoua and East regions, with up to one in two refugee and internally displaced children affected by stunting, surpassing emergency thresholds (OCHA, 2023). Additionally, wasting rates among Central African refugees in camps remain alarming. Projections for 2023 indicate that 48,800 children under five in the North-West and South-West regions and 291,862 children in the Far North, North-West, South-West and East regions will face acute malnutrition. These figures highlight the urgent need for coordinated interventions to combat malnutrition among vulnerable populations.

A significant body of literature explores the causes of child malnutrition and its prevention strategies. According to the United Nations Children's Fund (UNICEF), (1990)'s framework, malnutrition results from basic, underlying and immediate factors. While immediate causes like inadequate diets are well-documented, societal trends, such as maternal labor supply, also play a role (Datar et al., 2014; Sturm, 2005). Maternal employment can affect child nutrition through two mechanisms: increased household income, which reduces poverty and food insecurity, or reduced caregiving time, which can hinder activities like meal preparation, medical visits and breastfeeding (Cawley and Liu, 2012).

Empirical findings on the relationship between maternal employment and child malnutrition are mixed. Some studies identify maternal employment as a risk factor for child malnutrition (Andrade and Gil, 2023; Kyanjo et al., 2025; Mim et al., 2025; Rashad and Sharaf, 2019), while others suggest it serves as a protective factor (Nankinga et al., 2019; Ngenzebuke and Akachi, 2017; Tchakounté, 2023). However, no study has yet explored the factors contributing to disparities in child nutrition between children of employed and unemployed mothers. This study addresses this gap by examining these disparities. The article is structured as follows: a review of relevant literature, a description of methodology and data, an analysis of econometric results and a conclusion with key findings and implications.

The literature highlights the determinants of malnutrition among children under five, emphasizing factors such as gender, birth order, parental education and maternal employment. Bras and Mandemakers (2022) found that boys generally have better height-for-age and weight-for-age outcomes than girls and earlier-born children fare better nutritionally than later-born siblings. Parental education plays a critical role, with Semba et al. (2008) showing that higher parental education levels reduce stunting risk. Bras and Mandemakers (2022) also noted that educated mothers are more likely to secure better-paying jobs, improving nutrition and healthcare for their children. However, evidence on maternal employment's impact on child nutrition is mixed, with studies reporting both positive and negative effects.

Maternal employment, as discussed by Morrill (2011) and Rashad and Sharaf (2019), can increase household income, benefiting child nutrition, but also reduces time for caregiving, potentially compromising tasks like preparing meals or maintaining hygiene. Meyer (2016) showed that full-time maternal employment increases the risk of childhood overweight, attributing this to unhealthy behaviors in children, particularly poor dietary habits and reduced physical activity, arising from a decreased amount of maternal time dedicated to childcare. In developed countries, maternal employment has been linked to higher risks of overweight and obesity in children, with Meyer (2016) in Germany and Datar et al. (2014) in the United States, showing a positive association between maternal work hours and children's BMI. In contrast, studies in China (Nie and Sousa-Poza, 2014) and Denmark (Greve, 2011) reported no significant associations.

In developing countries, research focuses on undernutrition. Waghode et al. (2025) reviewed studies in South Asia and found mixed results, with maternal employment associated with both positive and negative effects. Rashad and Sharaf (2019) in Egypt and Andrade and Gil (2023) in Ecuador reported significant negative impacts of maternal employment on child nutrition, including increased stunting likelihood. The type of employment is crucial, as Nankinga et al. (2019) found that children of mothers in agriculture or manual labor are more likely to be stunted than those of mothers in formal jobs. Similarly, Ketema et al. (2022) reported higher rates of stunting among children of employed mothers in Uganda, but Ngenzebuke and Akachi (2017) found a positive association between female labor force participation and child weight-for-age in Nigeria. Tchakounté (2023) in Cameroon observed mixed effects, with maternal employment linked to higher odds of stunting but lower odds of wasting.

The literature also addresses disparities across rural-urban divides and social groups. Nie and Lin (2025), employed an inequality decomposition approach and demonstrated that rural Chinese children are at greater risk of stunting due to limited access to resources, healthcare and education. Similarly, Ali et al. (2025), used Fairlie's decomposition method and found that rural Indian women exhibit increased vulnerability to undernutrition, with economic status, education and dietary diversity identified as key determinants. Kalinda et al. (2023) reported similar findings in Rwanda, attributing disparities in stunting to socio-economic and structural factors. In a study focusing on Dalit and non-Dalit children in India, Sharma and Smieliauskas (2022) applied the Oaxaca-Blinder decomposition and documented significant reductions in nutritional inequalities over time. Despite these contributions, the literature reveals a notable gap: children of employed mothers appear to face higher malnutrition risks, yet few studies investigate the underlying causes or offer potential solutions. Addressing this gap is essential for the development of effective policies aimed at combating child malnutrition.

The methodology section is organized into three main parts: data collection, a description of the study variables and the specification of the econometric approach.

This study utilized data from the 2018 Cameroon Demographic and Health Survey (DHS), conducted by the National Institute of Statistics in collaboration with the Ministry of Public Health. The survey covered 13,160 households across 12 regions, encompassing both urban and rural areas. Information was collected from women aged 15–49 and children under five, drawn from a subsample of households not included in the men's survey.

The anthropometric data collected during the 2018 EDSC-V survey enabled the assessment of the nutritional status of children under five using various anthropometric indices. This analysis allowed for the identification of subpopulations affected by stunting, acute malnutrition and underweight related to these conditions. Initially, 5,296 children under five were eligible for anthropometric measurements and weighing. However, after excluding those with missing, incomplete, or out-of-range data, the final sample comprised 4,618 children with valid measurements.

In this study, child nutritional status is assessed using underweight status, a widely recognized anthropometric indicator frequently employed in previous research analyzing child nutrition (Andrade and Gil, 2023; Hossain et al., 2023; Islam et al., 2020; Ketema et al., 2022; Rashad and Sharaf, 2019; Sharma and Smieliauskas, 2022). Underweight in children under five is assessed using the weight-for-age Z-score (WAZ), which compares a child's weight to a reference population of healthy children of the same age and sex based on WHO growth standards. Specifically, the weight-for-age Z-score measures how many standard deviations a child's weight is from the median weight of the reference group. It is given by the following formula:

Children with a WAZ below −2 standard deviations are classified as underweight, indicating they weigh significantly less than their peers, reflecting both acute and chronic malnutrition. This standardized method is widely used to monitor child nutritional status globally.

The primary explanatory variable, maternal employment, is defined as a binary indicator reflecting whether the mother is currently employed, either as a wage earner or self-employed. Maternal employment plays a significant role in child nutrition by affecting both the time available for caregiving and the household's economic resources. In Cameroon, where women often balance work and family responsibilities, this relationship is complex but essential for understanding socioeconomic inequalities in child nutritional status. This study therefore focuses on maternal employment to examine its contribution to these disparities. The additional variables relevant to child nutrition-related health outcomes, drawn from previous studies in the literature, are outlined in Table 1.

This study employs a Logit model to analyse the determinants of child nutritional status, as measured by underweight, given that the dependent variable $Y$ is dichotomous, indicating whether a child is underweight or not (Zamo Akono and Medjo Obia, 2025). In the particular case of our study, if we consider a sample of n individuals with an index of $i=1,...,n$ , then:

The probability of a child being underweight is assumed to be a function of the mother's employment status, as well as various socio-economic, demographic and environmental factors (Zamo Akono and Medjo Obia, 2025). The following functional form can be used:

$yi*$ being a latent variable defined as follows:

It is observed that the probability of a child being underweight corresponds to the mathematical expectation of the variable.

The aim being to explain the probability of a child being underweight $(Yi=1)$ on the basis of a set of characteristics $xi$⁠, then:

Since the logistic function is symmetrical, we can replace $F−ε(xiβ)$ by $Fε(xiβ)$ . We therefore have a distribution function of the logistic function which represents the probability of a child being underweight as follows:

By dividing the probability of a child being underweight by the probability of not being underweight, we obtain the ratio:

$Pr(Yi=1)Pr(Yi=0)=Pr(Yi=1)1−Pr(Yi=1)=exiβ$ This represents the risk of a child being underweight, known as the “odds ratio”.

Applying log to linearize gives:

Where $β$ represents the vector of coefficients for the explanatory variables $xi$ described in Table 1.

Decomposition of disparities in underweight prevalence between children of employed and unemployed mothers.

This analysis applies the multivariate decomposition method for nonlinear response models (MVDCMP), as developed by Powers et al. (2011), to explore the key determinants of the disparity in underweight prevalence between children of employed and unemployed mothers. Denoting $y̅e$ and $y̅u$ as the average predicted probabilities of underweight status for children of employed and unemployed mothers, respectively, the overall difference in these probabilities is decomposed as follows:

In this context $Xe$ and $Xu$ denote the covariates associated with children of employed and unemployed mothers, respectively, while $βe$ and $βu$ represent their corresponding coefficients. The model applies a logistic link function for the analysis. The E component, often termed the explained component, accounts for the portion of the disparity attributable to differences in characteristics or endowments. In contrast, the C component, commonly referred to as the unexplained component or coefficient effects, captures the disparity arising from differences in the coefficients (Zamo Akono and Medjo Obia, 2025). A comprehensive decomposition enables a detailed examination of the specific contributions of individual variables to these components of the overall observed disparity. This process entails breaking down E and C into distinct segments, $Ej$ and $Cj$ (for each j = 1,2,…k), which represent the specific contribution of the jth covariate to the explained and unexplained components, respectively. Yun (2004) proposed a simplified method for detailed decomposition by utilizing weights derived from a first-order Taylor expansion of Equation (7) around $Xeβe$ and $Xuβu$⁠. Following the linearization, the weight component for E is expressed as:

and the $kth$ weight component for C is:

Where,

The distinct weights $WΔXk$ represent the contribution of the kth covariate to the linearization of E, while $WΔβk$ captures the contribution of the kth covariate to the linearization of C. These weights are proportional to the individual contributions in the decomposition of the linear predictor. The relative magnitude of these contributions to the explained or unexplained portions of the outcome disparity aligns with their relative influence in the decomposition of the linear predictor. Consequently, the overall difference can be expressed as the sum of weighted contributions from each specific component.

Table 2 summarizes the demographic and socioeconomic characteristics of women and their children, highlighting key differences between employed and unemployed mothers. Among the sample, 11.49% of children are underweight, with a higher prevalence observed in children of employed mothers (13.21%) compared to unemployed mothers (9.90%). The age distribution is balanced across groups (19–22%) and male and female children are equally represented (50.7 and 49.3%, respectively). Illness is slightly more common in children of employed mothers (31.7%) than unemployed mothers (28.6%), potentially reflecting disparities in caregiving or healthcare access.

Educational levels show unemployed mothers are more likely to have no education (24.7%) compared to employed mothers (21.3%), while primary education is more prevalent among employed mothers (38.5 vs. 28.5%). Higher education levels are similar across both groups, suggesting limited influence on employment status. Employment is also linked to age: younger mothers (15–24 years) are more likely to be unemployed (38.0%) compared to employed mothers (25.1%), while middle-aged (25–34 years) and older mothers (35–49 years) are more likely to be employed.

Economic and geographic disparities are evident, with employed mothers concentrated in poorer households (24.12 and 25.62% in the poorest and poorer wealth quintiles) and rural areas (61.2%), while unemployed mothers are more represented in wealthier households (23.06 and 15.54% in the richer and richest quintiles) and urban areas (55.9%). These patterns underscore the complex relationship between employment, socioeconomic status and residence.

The prevalence of underweight among children under five reported in Table 3 is influenced by age, maternal, economic and geographic factors. Younger children (0–11 months) and females have lower rates of underweight compared to older children (48–59 months) and males. Maternal education and nutritional status significantly affect children nutritional status, with children of uneducated (23.50%) and underweight mothers (33.74%) having the highest prevalence. Economic disparities are stark, as children in the poorest households (22.02%) and rural areas (14.86%) are more affected than those in wealthier households (3.23%) and urban areas (7.38%). These findings underscore the need for targeted strategies addressing education, economic inequality and healthcare access to reduce underweight prevalence.

Table 4 presents the estimates of the determinants of underweight among children in Cameroon. The results of the multivariate logit regression are reported as marginal effects with 95% confidence intervals. The analysis distinguishes between women's employment status in the third and fourth columns. All three estimated models are statistically significant at the 1% level, as indicated by Prob > χ² = 0.000 across the models.

The results of the primary regression analysis reveal that maternal employment status is associated with a 2.22% increase in the likelihood of children being underweight. This finding suggests that employed mothers may face challenges in balancing work responsibilities with caregiving duties. Specifically, the income from maternal employment often fails to offset the reduced time for childcare, creating challenges for mothers as primary caregivers in ensuring their children's proper nutrition, which can lead to poorer nutritional outcomes and increased malnutrition risk.

The observed relationship between maternal employment and child underweight aligns with broader evidence on its association with child stunting in low- and lower-middle-income countries. For example, Andrade and Gil (2023) report that employed mothers in Ecuador are more likely to have children who experience stunted growth or chronic malnutrition compared to their unemployed or inactive counterparts. Similarly, Rashad and Sharaf (2019) demonstrate that the probability of stunting in children with employed mothers in Egypt is 18.6% higher than in those with employed mothers.

Maternal educational achievement significantly reduces the odds of children underweight in Cameroon. The negative marginal effects of maternal education are large and consistent across all levels, with secondary education (6.3% reduction overall) and higher education (7.2% reduction overall) having the most substantial protective effects. This effect is consistent across both employed and unemployed mothers, emphasizing the critical role of maternal education in enhancing child nutrition. This underscores the transformative role of maternal education in improving child nutritional outcomes. Yabancı et al. (2014) highlight that higher maternal education improves nutritional knowledge, influencing healthier dietary habits, feeding practices and health-seeking behaviors. These findings are consistent with previous studies that found that mothers education significantly reduce the odds of malnutrition among Children (Semba et al., 2008; Datar et al., 2014; Brauner-Otto et al., 2019; Nankinga et al., 2019; Islam et al., 2020; Mori et al., 2021; Bras and Mandemakers, 2022; Andrade and Gil, 2023).

Older children face a higher risk of being underweight, with the likelihood peaking at 48–59 months (4.8% overall and 6.1% for children of employed mothers). Marginal effects are smaller for children of unemployed mothers, suggesting that caregiving practices may mitigate some age-related risks. However, reduced caregiving time in employed households contributes to significant nutritional disadvantages for older children.

The findings align with Hossain et al. (2023), showing that underweight prevalence among children under five increases with age, likely due to common breastfeeding practices in Cameroon, where most women breastfeed until 12 months. Breastfeeding reduces underweight risks, but discontinuation may expose children to improperly sterilized feeding equipment. Unlike Shajan and Sumalatha (2022), who found that infants of employed mothers face nutritional challenges, this study suggests that older children benefit from maternal employment.

As shown in the results presented in Table 4, male children have a higher probability of being underweight compared to their female counterparts. This finding is consistent with the work of Nankinga et al. (2019) and Quamme and Iversen (2022), who observed that female children are less likely to experience wasting and stunting than male children. According to these studies, this pattern may be attributed to the genetic composition of female children, which provides them with greater resilience to conditions of limited food supply. Additionally, household size is associated with an increased likelihood of children being underweight in Cameroon.

Consistent with findings from previous studies (Abuya et al., 2012; Elmighrabi et al., 2023; Quamme and Iversen, 2022; Sandra et al., 2022; Victora et al., 2008), a child's birth weight is significantly associated with the likelihood of being underweight. Children with low birth weight have a higher propensity to be underweight compared to those with higher birth weights. Underweight births are more common among malnourished mothers, and the factors contributing to these outcomes often persist postpartum.

Mothers in the lowest income quintile face a heightened risk of raising malnourished children. In contrast, children of mothers from wealthier households are less likely to be underweight compared to their counterparts in less affluent households. These findings align with Hossain et al. (2023), who reported a lower prevalence of underweight children in households with higher per capita income in Bangladesh. Wealthier families typically have access to better healthcare, nutrient-rich diets and healthier living conditions. Conversely, children from low-income families are more prone to being underweight due to insufficient food intake, limited access to basic healthcare and a greater vulnerability to infections (Hossain et al., 2023).

Table 5 presents the main results of the multivariate decomposition of disparities in the predicted probability of underweight children between employed and unemployed mothers in Cameroon. The total gap is 0.0325, statistically significant (p < 0.001), highlighting a notable disparity in child health outcomes based on maternal employment status. This disparity is attributed 31.93% to measurable differences in resources or characteristics (endowments), suggesting a need to equalize these factors between employed and unemployed mothers. The remaining 68.07% stems from differences in how these resources impact child health outcomes (coefficients), indicating that employment may provide benefits like income while imposing challenges such as reduced childcare time. These findings align with previous studies on child nutrition gaps, which reveal that only a small portion of disparities can be explained by observable factors. For instance, Kalinda et al. (2023) found that 23% of the rural-urban stunting gap in Rwanda was attributable to measurable characteristics. Similarly, Nie et al. (2019) reported that 38% of the change in wasting and only 9% of the change in stunting in India between 2005 and 2012 were explained by such factors.

The results from the detailed decomposition presented in Table 6 provide important insights into the complex factors driving disparities in child undernutrition between employed and unemployed mothers in Cameroon. Notably, the analysis reveals that while differences in characteristics (endowments) between these groups do contribute to the gap, the majority of the disparity arises from differences in how these characteristics influence child nutrition (coefficients). This distinction is critical as it suggests that equalizing access to resources alone may not be sufficient to close the nutritional gap; instead, the context in which these resources are utilized and their differential impacts must also be addressed.

The coefficient effect dominates across all age categories, with older children (24–35 months: 58.15%; 36–47 months: 32.40%) contributing significantly to disparities. The prominent role of child age through the coefficients effect indicates that the nutritional vulnerability of children varies considerably depending on maternal employment status, especially in the critical early years. This could reflect differences in caregiving practices or time availability, with employed mothers potentially having less time for direct child care, impacting children's nutritional outcomes disproportionately at certain ages. Birth order's strong contribution to the disparity highlights the cumulative challenges faced by higher-order births, which might be more pronounced among unemployed mothers due to limited resources or among employed mothers due to time constraints. The nuanced effects observed with birth size further suggest that biological factors interact with socioeconomic and maternal employment conditions, shaping child nutritional status in complex ways.

Maternal education demonstrates mixed effects. Primary education contributes negatively via endowments (−14.41%) but positively via coefficients (6.84%). Secondary education shows the reverse, with a positive endowment effect (11.23%) but a substantial negative coefficient effect (−65.13%), suggesting that the benefits of secondary education on child nutrition differ markedly between employed and unemployed mothers. The mixed effects of maternal education underscore its multifaceted role. While education generally enhances maternal knowledge and practices related to child nutrition, its benefits appear to manifest differently for employed and unemployed mothers. This may reflect variations in how education influences employment opportunities, income and childcare arrangements, emphasizing the need for context-sensitive interventions that consider maternal employment status when leveraging education to improve child health.

The results indicate that maternal age significantly influences disparities in child nutrition between employed and unemployed mothers in Cameroon. Endowment effects show that older mothers (35–49 years) have characteristics that reduce the likelihood of underweight children, contributing −18.32% to the disparity, while middle-aged mothers (25–34 years) contribute −4.81%. However, coefficient effects highlight the differential impact of employment: middle-aged mothers (25–34 years) disproportionately experience adverse outcomes (−149.62%), followed by older mothers (−54.92%). These findings emphasize the need for targeted interventions, such as childcare and nutrition support for employed mothers, with a particular focus on younger and middle-aged groups to address both structural inequalities and employment-related challenges.

The contrasting effects of household wealth – where greater wealth widens disparities through differences in endowments but narrows them through coefficients – suggest that wealthier families possess more resources, yet the impact of these resources on child nutrition tends to be more evenly distributed between employed and unemployed mothers. This implies that while wealth can mitigate some of the adverse effects of maternal employment on child nutrition, it is not sufficient on its own to eliminate the disparities. Additionally, the relatively limited influence of place of residence and maternal underweight status highlights that other factors play a more significant role in driving nutritional inequalities. Nonetheless, including these variables remains important to present a comprehensive understanding of the determinants.

This article investigates how maternal employment status influences child undernutrition and identifies underlying factors contributing to nutritional disparities between children of employed and unemployed mothers. The study finds that children of employed mothers are more likely to be underweight compared to those of unemployed mothers, with structural barriers playing a larger role than differences in observable characteristics. Employed mothers face challenges such as limited maternity leave (often just three months or less, especially in the informal sector), time constraints and lack of affordable, quality childcare, all of which negatively impact child care and nutrition.

To address these issues, the study recommends that the government support private childcare initiatives and enforce regulations to protect child health. It also suggests creating focus groups where experienced mothers can share advice with younger mothers on balancing work and child nutrition. Policies should prioritize young children from disadvantaged households, as well as young and single mothers.

The study highlights the mixed effects of maternal education depending on employment status, emphasizing the need for tailored nutrition and parenting education programs that teach practical skills like meal preparation, breastfeeding and managing childhood illnesses. Birth order is another important factor, with children of higher birth orders more at risk of poor nutrition. Strengthening family planning and reproductive health services can help improve outcomes by promoting adequate birth spacing and targeted nutrition support for larger families.

Finally, while these findings provide important guidance for child health policy in Cameroon, the study notes limitations such as using a simple employed/unemployed classification and focusing on short-term nutritional outcomes. It calls for further research on the long-term effects of maternal employment and the influence of job characteristics.

LOMO contributed in the writing of the manuscript. SAA contributed in the writing and supervision of the manuscript. All authors read and approved the final version of the manuscript.

The data are available upon request.