Maternal and child undernutrition causes more than 10% of the total global burden of disease and more than one-third of child deaths (1, 2). The poor nutritional status of children and mothers has long-term health consequences, and the ramifications extend to inter-generational low productivity and perpetuation of poverty (3–6). Despite growing emphasis on promotion of maternal and child nutrition through various policies, plans, and programs (5, 7), Nepal has not made significant progress in the reduction of maternal and child undernutrition as expected (8, 9). For example, stunting in under-five children declined from 57% in 2001 to 36% in 2016, and underweight in under-five children declined from 43% in 2001 to 27% in 2016 (9). Global literature has revealed that 45% of under-five deaths are contributed by undernutrition as malnourished children are highly susceptible to childhood diseases (10). Therefore, reduction of underweight can be an important step toward reducing childhood morbidity and mortality in Nepal.

Multiple factors affect childhood nutritional and development status, including household socioeconomic position, child feeding practices, environmental factors, access to and use of health services, and household level hygiene practices (2). Childhood illnesses such as diarrhea and acute respiratory infections, which are associated with poor hygiene and access to sanitation, are common causes of under-nutrition in developing countries including Nepal (11–14). It has been suggested that nutritional interventions and environmental sanitation strategies such as access to clean water, waste disposal, and use of clean household fuels can reduce childhood underweight and reduce child mortality by 14–31% in developing countries in South Asia and Africa (15).

While there has been some research on the effect of feeding practices and socioeconomic conditions (14, 16–18) on childhood undernutrition in Nepal, there is little empirical research on how socioeconomic and environmental conditions may influence childhood underweight. Furthermore, understanding these relationships in rural settings may offer evidence for designing contextual promotion activities. Thus, this paper investigates the association between social, environmental, and maternal and child health (MCH) factors and low weight for age (weight for age Z score <−2 SD) among children under 5 years old from a rural district of eastern Nepal.

The prevalence of underweight in the study population was 37% [95% confidence interval (CI): 33–43%] (table is not shown).

The prevalence of underweight (weight for age < −2 SD) was found to be 37% (95% CI: 32–43) in the current study. This is higher than the national average (27%) and that of the eastern region of Nepal (24%) (9). More than two-thirds of our sample were from the disadvantaged ethnic communities. Those communities usually have low socioeconomic status, lower land ownership, high rates of childbirth at home, and childhood illnesses, which might be accounted for the prevalence of underweight of children aged under 5 years (26, 27). Pooled figures of three consecutive NDHS data showed higher odds of early initiation of breastfeeding among mothers with primary education (aOR: 1.24) and secondary or higher education (aOR: 1.63) (27). In our study, about one-third of mothers were either illiterate or only had primary level education, and nearly half of their children were underweight. Lower maternal education has been shown to be associated with late initiation of breastfeeding after birth (27) and poor complementary feeding practices of children aged under 2 years, consequently affecting childhood nutritional status (28, 29). Furthermore, mothers who belong to the disadvantaged ethnic groups face food insufficiency around the year and are more likely to engage in labor works for subsistence in Nepal (30). Because of various household chores, mothers from such ethnic groups have less time for child feeding, child caring, and health-seeking during an illness in comparison to the advantaged, educated and wealthy mothers. This consequently affects dietary frequency and diversity in infants and young children (26, 29). Context-specific health information education and communication (IEC)/behavior change communication (BCC) materials such as radio jingle, posters, pamphlets, and other pictorials can be designed in the local languages and delivered through multiple channels so that message can be reached out to disadvantaged communities.

In the current study, children aged 24–59 months were more likely to be underweight than children aged 0–23 months. A survey carried out in Lao PDR found that children in higher age groups had higher odds of underweight, which is consistent with the current study finding (31, 32). Poor socioeconomic status and housing conditions in our study may have contributed to the high prevalence of underweight among children aged 24 months or above. In the rural Nepali contexts, mothers usually give less attention to their children when they reach 2 years of age. At that point, elder siblings or other members of the family take over as their caregivers, which might lead to poor feeding and hygiene practices. As a result, such children are more susceptible to childhood illness, ultimately become underweight (33). Furthermore, nearly one-third of children between the age of 6 and 9 months (30%) do not receive complementary foods in addition to breastfeeding (12). Such practices are more prevalent among poor and marginalized families, which might also be contributing to children being underweight. This finding suggests that later age within under 5 years’ age carry a higher risk of underweight. They need more attention, especially in the frequent feeding of diversified foodstuff in adequate quantity during pre-school period of the children, and the practice should be promoted.

In this study, children from families who consumed water without any treatment, such as boiling, or filtration were found to be more underweight. Poor hygiene and sanitation practices are primarily responsible for childhood illnesses worldwide (34). In this study, about 25% of children had suffered from diarrhea 1 month prior to the data collection. The higher number of diarrheal cases is reported during the summer season in the district where our study was carried out. Moreover, mothers must do their household chores, in addition to being involved in agricultural activities and labor work for subsistence. This also contributes to poor child caring and such children are more prone to undernutrition (35). Moreover, poor child caring practices increases susceptibility to childhood illnesses such as diarrhea and worm infestations which also leads to underweight (36).

In the current study, the likelihood of being underweight was higher among those children whose mother perceived smaller size at birth than average—a finding consistent with a study conducted in 2014 in Nepal (37). Mothers’ nutritional status determines the nutrition of their children at birth (38). In Nepal, about half of the pregnant women are anemic or have low body mass index (12). A randomized controlled trial conducted in Nepal reported that the supplementation of multiple micronutrients during pregnancy increased the birth weight and reduced low birth by 14% (39). However, there is a widespread cultural belief that if a pregnant woman consumes a lot of nutritious food during her pregnancy, the baby growing in her womb will become very big in size which will then make childbirth more painful and difficult (40). Such belief subjects mothers to nutritional deficiency during pregnancy which might result in intrauterine growth retardation, preterm births, and/or children being born with small size at births. If parents perceive that their newborn is smaller than average or is born with low birth weight, such babies need additional postnatal care and frequent feedings because such children face higher chance of becoming ill (39, 41). So, maternal nutrition education and counseling session during antenatal care visits should be promoted focusing on undernourished pregnant women. Parents of low birth weight, preterm, and small babies also need to be made aware of the future implications. The need for adequate attention and nutritional support, caring, and health seeking of such children should also be highlighted.

The children who were not growth-monitored in the last 6 months preceding the survey were over two times more likely to be underweight than those who were monitored. Studies have shown that regular growth monitoring is associated with better feeding practices of children (18, 42). A review of growth monitoring and promotion impact conducted in 2008 reported that growth monitoring could provide an entry point for preventive and curative health care and were an integral part of programs that were associated with significant reductions in malnutrition (43). Further, it provides an opportunity to promote and encourage breastfeeding and appropriate complementary feeding which results in good nutrition status of children (43). In Nepal, apart from health facilities, growth monitoring sessions are conducted in at least three to five places within the community at the primary health care/primary outreach clinics (PHC/ORC) on a monthly basis. The national nutrition program has recommended that all children need to be growth-monitored at least six times in the first year, four times in the second year, and three times each in remaining three years (44). Such intervention increases the access to child health services to mothers who live in hard-to-reach areas from the health facilities. Growth monitoring is the entry point which determines children’s nutritional status and conducts counseling to the parents/caretakers for improved nutritional knowledge, attitude, and household level practices.

The findings from the current study have several implications for policy, research, and practice. Comprehensive interventions which focus on educating future mothers, improving MCH service utilization, and improving the economic status of households are required to improve the nutritional situation of children under 5 years of age. Improved feeding practices of pre-school children should be promoted. An earlier study suggested enhancing complementary feeding practice in Nepal by focusing on women from poorer households and those less exposed to media for the better nutritional status of children (29). Strategies to improve the nutritional status of children include improving women’s educational status (25) and promoting regular growth monitoring practices. These provide an opportunity for counseling mothers about their children’s nutritional needs and feeding practices. Furthermore, long-term strategies for improvement of economic status such as homestead food production, kitchen garden, and income-generating activities may also improve the child caring and feeding practices, which can impact nutritional status of children. Similarly, adequate health seeking practices during pregnancy may reduce the chance of low birth weight or preterm babies and if babies are born smaller than average size, more postnatal care, and extra care for such babies may minimize the possibility of being underweight in later months. Because of the high density of indigenous people living in the district (20), there might be many cultural and linguistic barriers to implementation of targeted interventions. More importantly, cultural practices such as eating less during pregnancy, poor food diversity need to be addressed by targeting pregnant women, families, and communities in behavior change communication activities (45). Mass media mobilization, distribution of IEC/BCC materials as per local context, counseling through peer mobilization (46), or involvement of mother-in-law in prenatal care (47) could be potential strategies to reduce such contextual barriers.

Some limitations of the present study should be acknowledged. A study conducted in eastern Nepal may not be generalizable to communities from other parts of the country, and the cross-sectional nature of the survey does not allow for inferences regarding causality. Self-reported response from the mothers is another limitation of this study, though probing questions were asked during interview and mothers had been invited to provide documents (for example birth certificate, antenatal care card, and others) to reduce the recall bias. Nonetheless, the strength of this study includes the objective assessment of the weight of the children and objective observations of some environmental factors during data collection.

Underweight among under-five children was significantly associated with the age of a child, purified water consumption practices, growth monitoring practices, and the mother’s perception of small size at birth. These findings suggest the need for broader and longer-term interventions focusing on maternal and child nutrition promotion to reduce underweight among under-five children.

The data and findings related to this research are included in the manuscript.

DA holds an MPH and public health officer at the Ministry of Health Nepal. RK is a PhD candidate at The University Queensland School of Public Health, Australia. YP holds an MPH and is a Research Manager Nepal Health Sector Support Programme, Nepal. AP holds a PhD and is Associate Professor of Biostatistics at Tribhuvan University, Institute of Medicine, Department of Community Medicine and Public Health, Nepal.

DA designed the study concept and conducted statistical analysis of the manuscript. RK and YP contributed to data analysis, literature review, drafting and finalizing the manuscript. AP supervised the project. All the authors have read and approved the final version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.