Mother’s nutrition-related knowledge and child nutrition outcomes: Empirical evidence from Nigeria

PLoS ONE, Volume 14, No. 2, Year 2019

Interpretation

Background Nutrition outcomes among young children in Nigeria are among the worse globally. Mother’s limited knowledge about food choices, feeding, and health care seeking practices contributes significantly to negative nutrition outcomes for children in most developing countries. Much less is known about the relationship between mother’s nutrition-related knowledge and child nutritional outcomes in rural Nigeria. This paper investigates therefore: (i) the association of mother’s nutrition-related knowledge with nutrition outcomes of young children living in rural Nigeria, where access to education is limited, and (ii) whether mother’s education has a complementary effect on such knowledge in producing positive child nutrition outcomes in such settings. Methods Using the Demographic and Health Survey data for Nigeria, we employ both descriptive and regression analyses approaches in analyzing the study’s objectives. In particular, we apply ordinary least square (OLS) to investigate the association of mother’s nutrition-related knowledge with child HAZ and WHZ while controlling for maternal, child, household and regional characteristics. An index was constructed for mother’s nutrition-related knowledge using information on dietary practices, disease treatment and prevention, child immunization, and family planning. Results We found that mother’s knowledge is independently and positively associated with HAZ and WHZ scores in young children. Higher levels of mother’s education, typically above primary, have a significant, positive association with child HAZ and WHZ scores. We argue that mother’s knowledge of health and nutrition may substitute for education in reducing undernutrition in young children among populations with limited access to formal education. However, the present level of mother’s education in rural Nigeria appears insufficient to reinforce knowledge in producing better nutrition outcomes for children. Conclusions This study suggests promotion of out-of-school (informal) education, such as adult literacy and numeracy classes where women without formal education can gain health and nutrition knowledge, and practices that could enhance child nutrition outcomes in Nigeria. The 2013 DHS data employed for this study is the most recent one for Nigeria. The sample was selected using a stratified three-stage cluster design consisting of 904 clusters, with a total of 38,522 households. Sampling errors are computed statistically using appropriate tools and methodologies as provided in the material and methodology document of the 2013 Nigeria Demographic and Health Survey [16]. The children module of the DHS comprises of 31,482 children under five years with their mothers between 15 to 49 years. Children living in rural households are 21,131 (67%), and 5,950 of them are between 6 to 23 years. We only included the youngest child between 6 to 23 months per family, and following deletion of observations with incomplete anthropometric measures, 4,941 mother-child pairs were finally available for analysis. The dataset captures mothers, children, and household information such as socioeconomic characteristics, anthropometric measurements for mother and child, immunization records, health care seeking and feeding practices, water and sanitation, and demographic information among others. Children are being introduced to complementary feeding between 6–23 months, as such, they are the most vulnerable group to undernutrition and consequent growth faltering [3, 37]. Hence, children in this age bracket benefit the most when their mothers have basic knowledge of health and nutritional care. This is a very important, as the consequences of undernutrition, stunting in particular are difficult to reverse in children after the age of two. Height-for-age z-scores (HAZ) and weight-for-height z-score (WHZ) are good indicators of child nutrition and health. Children with HAZ less than two standard deviations below the median measurement for the reference group are said to be stunted or chronically undernourished. While children with WHZ less than two standard deviations below the median measurement for the reference group are regarded as wasted or acutely undernourished [38]. Hence, these two indicators measure whether a child is undernourished (stunted or wasted) or not. It is of policy relevance to investigate separately the factors producing wasting and stunting gauging from these indicators. Our choice of the determinant variables was informed by the literature on the determinants of child nutrition outcomes [39]. We included child and mother characteristics and household characteristics. Child characteristics are sex, age, birthweight and child from multiple birth. Mother characteristics include age at first birth, number of children ever born, mother’s level of education and nutrition and health knowledge of mother (an index computed). Household characteristics are household size; education; age; and an asset-based wealth index, among others. An asset-based wealth index is constructed using a principal components analysis of the NDHS assets data, combining variables on ownership of a radio, bicycle, car, and other items with household dwelling characteristics [40]. This allows households to be ranked by wealth index. We ranked households as poorest tercile wealth index, middle tercile wealth index and wealthiest tercile wealth index. The region dummies where the household belongs, which comprises of the six geopolitical regions of Nigeria. Under this sub-heading, we discuss the computation of mother’s nutrition-related knowledge index, the measurement of mother’s education variable and its interaction with mother’s knowledge. The DHS data contains some important information on dietary practices, disease treatment and prevention, child immunization and family planning. We follow the guidelines for assessing nutrition-related knowledge, attitudes, and practices (KAP) as contained in [41]. Mother’s nutrition-related knowledge was assessed based on five key nutrition and health information as follows: (i) mother’s knowledge of the important of colostrum; (ii) knowledge of continued breastfeeding; (iii) knowledge of diarrhea prevention and treatment using Oral Rehydration Solution (ORS); (iv) knowledge of child immunization; and (v) knowledge of family planning. A detailed description of these variables and their measurement is presented in the result section. Past studies have assessed mother’s nutrition knowledge by either assigning scores to observed knowledge (practice) [42] or scoring mother’s ability to answer “yes” or “no” to a set of questions relating to child health and nutrition [43, 44] or a combination of these [33]. A strong linear relationship between knowledge of young child nutrition and practices has been established in the literature, especially where there are no sociocultural barriers to such practices [45, 46]. We then apply principal component analysis (PCA) as adapted from Filmer and Pritchett [47] to construct the mother’s Nutrition Knowledge Index using the five components highlighted above. Variables for each component are assigned indicator weights that are first standardized; that is, z-scores are calculated and then factor coefficient (factor loading) scores are calculated. More details are shown in S1 File. Since this study is interested in knowing at what level of education is the association of mother’s education most significant with nutrition outcomes, mother’s levels of educational attainment are categorized as follows in the empirical model used: no education, primary education, secondary education, and tertiary education. We further used the interaction between the four educational level dummy variables and mother’s knowledge index to produce four interaction terms. Adding interaction terms to the model helps to better understand the relationship between knowledge and education and the association with outcomes. In other words, to test whether the association of mother’s knowledge with HAZ and WHZ is different at the different levels of mother’s education. For the purpose of nonparametric analysis between mother’s knowledge versus mother’s education on the distribution of HAZ and WHZ scores, we defined mothers with high knowledge as those with above the mean of nutrition-related knowledge in our sample, while mothers with low knowledge are below the mean cut off for nutrition-related knowledge index. For the purpose of interaction model, we categorize mother’s education as no education and with some education. We first adopt a bivariate (nonparametric) analytical approach to understand the relationship between mother’s knowledge versus mother’s education on the distribution of HAZ and WHZ scores using kernel density plots. We also report descriptive statistics that show the means comparisons of variables by maternal education and maternal nutrition-related knowledge. The analytical model employed for this study is a production function similar to the one applied in Rosenzweig and Schultz [48]. The child nutrition outcomes N of child i in household j depend on mother’s nutrition knowledge K, a set of maternal inputs Y, observable individual child’s characteristics I, household characteristics H, and regional characteristics G. This mathematical relationship is specified as: where Nij is the child nutrition outcomes measured by HAZ and WHZ as indicators for stunting and wasting, respectively. Kij is mother’s nutrition knowledge index vector. Yij includes mother’s education, mother’s age at first birth, and the number of children she has borne. Vector Iij includes child’s sex, child’s age and whether child is from multiple birth. The household characteristics vector Hij includes household size and wealth status; while Gij captures the geographic location (zone) where a child grows up (six dummy variables for zones). Ɛij is a vector representing the net effect of all other relevant unobserved factors. This relationship is expressed by the linear function: Where N is child nutrition outcomes and mother’s nutrition knowledge K. Y, I, H and G represent other determinants as specified in Eq (1)). To test whether the association of mother’s knowledge with HAZ and WHZ is different at the different levels of mother’s education, we add interaction terms of knowledge and different levels of mother’s education E. In particular, we estimate the following specification: Mother’s nutrition-related knowledge K is an index constructed based on key nutrition and health information described above. The levels of mother’s education used in the empirical model are no education, primary education, secondary education, and tertiary education.

AI Digest

Study Justification:

– The study aims to investigate the association between mother’s nutrition-related knowledge and child nutrition outcomes in rural Nigeria, where access to education is limited.
– The study also examines whether mother’s education has a complementary effect on nutrition knowledge in producing positive child nutrition outcomes.
– The findings of this study will contribute to understanding the role of mother’s knowledge in reducing undernutrition in young children and inform strategies to enhance child nutrition outcomes in Nigeria.

Highlights:

– The study found that mother’s knowledge is independently and positively associated with child height-for-age and weight-for-height scores.
– Higher levels of mother’s education, particularly above primary education, have a significant positive association with child nutrition outcomes.
– The study suggests that mother’s knowledge of health and nutrition may substitute for formal education in reducing undernutrition in populations with limited access to education.
– However, the current level of mother’s education in rural Nigeria appears insufficient to reinforce knowledge and produce better nutrition outcomes for children.

Recommendations for Lay Reader and Policy Maker:

– Promote out-of-school (informal) education, such as adult literacy and numeracy classes, to provide women without formal education with health and nutrition knowledge.
– Enhance efforts to improve mother’s education level, particularly in rural areas, to further improve child nutrition outcomes.
– Develop and implement targeted interventions to improve dietary practices, disease treatment and prevention, child immunization, and family planning knowledge among mothers.
– Strengthen health and nutrition education programs to empower mothers with the necessary knowledge and skills to improve child nutrition outcomes.

Key Role Players:

– Government agencies responsible for education, health, and nutrition policies and programs.
– Non-governmental organizations (NGOs) working in the field of education, health, and nutrition.
– Community leaders and local authorities.
– Health workers, including nurses and community health workers.
– Educators and trainers for adult literacy and numeracy classes.

Cost Items for Planning Recommendations:

– Development and implementation of adult literacy and numeracy classes.
– Training programs for health workers and educators.
– Production and dissemination of educational materials on health and nutrition.
– Monitoring and evaluation of interventions.
– Coordination and collaboration between government agencies, NGOs, and community stakeholders.
– Research and data collection to inform evidence-based interventions.
– Advocacy and awareness campaigns to promote the importance of mother’s knowledge in child nutrition outcomes.

Strength of Evidence

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong because it is based on empirical data from the 2013 Nigeria Demographic and Health Survey. The study uses a large sample size and employs both descriptive and regression analyses to investigate the association between mother’s nutrition-related knowledge and child nutrition outcomes. The study also controls for various maternal, child, household, and regional characteristics. However, to improve the evidence, the abstract could provide more information on the statistical significance of the findings and the effect sizes observed. Additionally, it would be helpful to include information on any limitations of the study and suggestions for future research.

Abstract

The 2013 DHS data employed for this study is the most recent one for Nigeria. The sample was selected using a stratified three-stage cluster design consisting of 904 clusters, with a total of 38,522 households. Sampling errors are computed statistically using appropriate tools and methodologies as provided in the material and methodology document of the 2013 Nigeria Demographic and Health Survey [16]. The children module of the DHS comprises of 31,482 children under five years with their mothers between 15 to 49 years. Children living in rural households are 21,131 (67%), and 5,950 of them are between 6 to 23 years. We only included the youngest child between 6 to 23 months per family, and following deletion of observations with incomplete anthropometric measures, 4,941 mother-child pairs were finally available for analysis. The dataset captures mothers, children, and household information such as socioeconomic characteristics, anthropometric measurements for mother and child, immunization records, health care seeking and feeding practices, water and sanitation, and demographic information among others. Children are being introduced to complementary feeding between 6–23 months, as such, they are the most vulnerable group to undernutrition and consequent growth faltering [3, 37]. Hence, children in this age bracket benefit the most when their mothers have basic knowledge of health and nutritional care. This is a very important, as the consequences of undernutrition, stunting in particular are difficult to reverse in children after the age of two. Height-for-age z-scores (HAZ) and weight-for-height z-score (WHZ) are good indicators of child nutrition and health. Children with HAZ less than two standard deviations below the median measurement for the reference group are said to be stunted or chronically undernourished. While children with WHZ less than two standard deviations below the median measurement for the reference group are regarded as wasted or acutely undernourished [38]. Hence, these two indicators measure whether a child is undernourished (stunted or wasted) or not. It is of policy relevance to investigate separately the factors producing wasting and stunting gauging from these indicators. Our choice of the determinant variables was informed by the literature on the determinants of child nutrition outcomes [39]. We included child and mother characteristics and household characteristics. Child characteristics are sex, age, birthweight and child from multiple birth. Mother characteristics include age at first birth, number of children ever born, mother’s level of education and nutrition and health knowledge of mother (an index computed). Household characteristics are household size; education; age; and an asset-based wealth index, among others. An asset-based wealth index is constructed using a principal components analysis of the NDHS assets data, combining variables on ownership of a radio, bicycle, car, and other items with household dwelling characteristics [40]. This allows households to be ranked by wealth index. We ranked households as poorest tercile wealth index, middle tercile wealth index and wealthiest tercile wealth index. The region dummies where the household belongs, which comprises of the six geopolitical regions of Nigeria. Under this sub-heading, we discuss the computation of mother’s nutrition-related knowledge index, the measurement of mother’s education variable and its interaction with mother’s knowledge. The DHS data contains some important information on dietary practices, disease treatment and prevention, child immunization and family planning. We follow the guidelines for assessing nutrition-related knowledge, attitudes, and practices (KAP) as contained in [41]. Mother’s nutrition-related knowledge was assessed based on five key nutrition and health information as follows: (i) mother’s knowledge of the important of colostrum; (ii) knowledge of continued breastfeeding; (iii) knowledge of diarrhea prevention and treatment using Oral Rehydration Solution (ORS); (iv) knowledge of child immunization; and (v) knowledge of family planning. A detailed description of these variables and their measurement is presented in the result section. Past studies have assessed mother’s nutrition knowledge by either assigning scores to observed knowledge (practice) [42] or scoring mother’s ability to answer “yes” or “no” to a set of questions relating to child health and nutrition [43, 44] or a combination of these [33]. A strong linear relationship between knowledge of young child nutrition and practices has been established in the literature, especially where there are no sociocultural barriers to such practices [45, 46]. We then apply principal component analysis (PCA) as adapted from Filmer and Pritchett [47] to construct the mother’s Nutrition Knowledge Index using the five components highlighted above. Variables for each component are assigned indicator weights that are first standardized; that is, z-scores are calculated and then factor coefficient (factor loading) scores are calculated. More details are shown in S1 File. Since this study is interested in knowing at what level of education is the association of mother’s education most significant with nutrition outcomes, mother’s levels of educational attainment are categorized as follows in the empirical model used: no education, primary education, secondary education, and tertiary education. We further used the interaction between the four educational level dummy variables and mother’s knowledge index to produce four interaction terms. Adding interaction terms to the model helps to better understand the relationship between knowledge and education and the association with outcomes. In other words, to test whether the association of mother’s knowledge with HAZ and WHZ is different at the different levels of mother’s education. For the purpose of nonparametric analysis between mother’s knowledge versus mother’s education on the distribution of HAZ and WHZ scores, we defined mothers with high knowledge as those with above the mean of nutrition-related knowledge in our sample, while mothers with low knowledge are below the mean cut off for nutrition-related knowledge index. For the purpose of interaction model, we categorize mother’s education as no education and with some education. We first adopt a bivariate (nonparametric) analytical approach to understand the relationship between mother’s knowledge versus mother’s education on the distribution of HAZ and WHZ scores using kernel density plots. We also report descriptive statistics that show the means comparisons of variables by maternal education and maternal nutrition-related knowledge. The analytical model employed for this study is a production function similar to the one applied in Rosenzweig and Schultz [48]. The child nutrition outcomes N of child i in household j depend on mother’s nutrition knowledge K, a set of maternal inputs Y, observable individual child’s characteristics I, household characteristics H, and regional characteristics G. This mathematical relationship is specified as: where Nij is the child nutrition outcomes measured by HAZ and WHZ as indicators for stunting and wasting, respectively. Kij is mother’s nutrition knowledge index vector. Yij includes mother’s education, mother’s age at first birth, and the number of children she has borne. Vector Iij includes child’s sex, child’s age and whether child is from multiple birth. The household characteristics vector Hij includes household size and wealth status; while Gij captures the geographic location (zone) where a child grows up (six dummy variables for zones). Ɛij is a vector representing the net effect of all other relevant unobserved factors. This relationship is expressed by the linear function: Where N is child nutrition outcomes and mother’s nutrition knowledge K. Y, I, H and G represent other determinants as specified in Eq (1)). To test whether the association of mother’s knowledge with HAZ and WHZ is different at the different levels of mother’s education, we add interaction terms of knowledge and different levels of mother’s education E. In particular, we estimate the following specification: Mother’s nutrition-related knowledge K is an index constructed based on key nutrition and health information described above. The levels of mother’s education used in the empirical model are no education, primary education, secondary education, and tertiary education.

Materials

https://efashare.b-cdn.net/materials/701bce71ca6014d0a2bd0fda177bf5175c292e893ade07ccbe1b7dc2bd70e59b/pone.0212775.s001.docx

Innovations

Innovation 1: Mobile-based Health Education Platform
Develop a mobile-based health education platform that delivers nutrition-related knowledge and practices to women without formal education. This platform can provide interactive lessons, videos, and quizzes to educate women on topics such as healthy food choices, feeding practices, disease prevention, child immunization, and family planning. The platform can be accessed through basic mobile phones, making it accessible to women in rural areas with limited access to formal education.

Innovation 2: Community Health Workers Training Program
Implement a training program for community health workers (CHWs) to educate and empower women without formal education on maternal and child health. CHWs can be trained to deliver health education sessions in local communities, focusing on nutrition-related knowledge and practices. These sessions can include demonstrations, discussions, and hands-on activities to engage women and promote behavior change. CHWs can also provide ongoing support and guidance to women, reinforcing the knowledge gained through the training program.

Innovation 3: Peer-to-Peer Support Groups
Establish peer-to-peer support groups for women without formal education, where they can share experiences, learn from each other, and receive guidance from trained facilitators. These support groups can provide a safe and supportive environment for women to discuss maternal and child health topics, including nutrition. Facilitators can lead discussions, provide information, and encourage women to adopt healthy practices. The support groups can also serve as a platform for women to access additional resources and services related to maternal and child health.

Innovation 4: Radio and Television Programs
Develop radio and television programs that specifically target women without formal education and provide them with nutrition-related knowledge and practices. These programs can be broadcasted in local languages and cover a range of topics, including healthy food choices, breastfeeding, hygiene practices, and child immunization. The programs can feature expert interviews, success stories, and interactive segments to engage and educate women. Additionally, phone-in or text message features can be included to allow women to ask questions and seek further information.

Innovation 5: Partnerships with Non-Governmental Organizations (NGOs)
Establish partnerships with NGOs that specialize in adult literacy and numeracy programs to integrate health and nutrition education into their curriculum. This collaboration can ensure that women without formal education have access to both basic literacy skills and essential health knowledge. NGOs can provide trained instructors, learning materials, and resources to support the integration of health education into their existing programs. This approach can reach a larger number of women and create sustainable solutions for improving maternal and child health outcomes.

AI Innovations Description

The recommendation that can be developed into an innovation to improve access to maternal health based on the study “Mother’s nutrition-related knowledge and child nutrition outcomes: Empirical evidence from Nigeria” is to promote out-of-school (informal) education programs for women without formal education. These programs can provide adult literacy and numeracy classes where women can gain health and nutrition knowledge, as well as practices that could enhance child nutrition outcomes in Nigeria. This recommendation is based on the finding that mother’s knowledge of health and nutrition is independently and positively associated with child nutrition outcomes. However, the study also highlights that the present level of mother’s education in rural Nigeria appears insufficient to reinforce knowledge in producing better nutrition outcomes for children. Therefore, promoting out-of-school education programs can help bridge this gap and empower women with the necessary knowledge and skills to improve maternal and child health.

AI Innovations Methodology

The study aims to investigate the association between mother’s nutrition-related knowledge and child nutrition outcomes in rural Nigeria, where access to education is limited. The methodology used in the study involves analyzing the Demographic and Health Survey (DHS) data for Nigeria, employing both descriptive and regression analysis approaches.

The study uses ordinary least square (OLS) regression analysis to examine the association of mother’s nutrition-related knowledge with child height-for-age z-scores (HAZ) and weight-for-height z-scores (WHZ), which are indicators of child nutrition and health. The analysis controls for maternal, child, household, and regional characteristics. An index is constructed for mother’s nutrition-related knowledge using information on dietary practices, disease treatment and prevention, child immunization, and family planning.

The study also examines the complementary effect of mother’s education on nutrition-related knowledge and child nutrition outcomes. Mother’s education is categorized into four levels: no education, primary education, secondary education, and tertiary education. Interaction terms between mother’s education and the knowledge index are included in the analysis to understand the relationship between education, knowledge, and child nutrition outcomes.

The dataset used in the study is the 2013 Nigeria DHS data, which includes information on socioeconomic characteristics, anthropometric measurements, immunization records, health care seeking and feeding practices, water and sanitation, and demographic information. The sample consists of 4,941 mother-child pairs from rural households in Nigeria.

To simulate the impact of recommendations on improving access to maternal health, a potential methodology could involve conducting a randomized controlled trial (RCT) or a quasi-experimental study. The study could compare the outcomes of a group of mothers who receive interventions aimed at improving their nutrition-related knowledge and access to maternal health services with a control group that does not receive these interventions.

The interventions could include providing nutrition education and counseling to mothers, improving access to prenatal and postnatal care services, and implementing community-based programs to promote healthy behaviors and practices. The impact of these interventions on maternal health outcomes, such as maternal mortality rates, maternal nutrition, and utilization of maternal health services, can be measured and compared between the intervention and control groups.

Additionally, qualitative methods such as interviews and focus group discussions can be used to gather feedback and insights from the participants regarding the effectiveness and acceptability of the interventions. This information can further inform the development and refinement of strategies to improve access to maternal health.

Overall, the study provides empirical evidence on the association between mother’s nutrition-related knowledge and child nutrition outcomes in rural Nigeria and suggests the promotion of out-of-school education and practices to enhance child nutrition outcomes. The methodology used in the study allows for the identification of factors influencing child nutrition outcomes and the potential impact of interventions on improving access to maternal health.

Author

Dima Health

Statistics:

Citations: 51

Identifiers:

DOI: 10.1371/journal.pone.0212775

Research Areas:

Community Interventions, Environmental, Food Security, Health System and Policy, Maternal Access, Maternal and Child Health, Quality of Care, Sexual and Reproductive Health, Social Determinants

Study Countries:

Nigeria

Study Design:

Cross Sectional Study

Study Approach:

Quantitative

Participants Gender:

Female