Proper feeding practices for infants and young children are essential for reducing infant morbidity and mortality . World Health Organization recommends early breastfeeding, exclusive breastfeeding until 6 months of age, dietary diversification from 6 to 24 months with continued breastfeeding until 24 months or longer . This period of dietary diversification should take into account dietary diversity, which means that children should receive at least four distinct food groups in their daily diet in order to maintain adequate growth and development . Poor dietary diversity is one of the major determinants of malnutrition in children aged 6 to 23 months. In addition to clinically apparent forms of malnutrition, dietary diversity plays a key role in the occurrence of micronutrient deficiencies, also known as “hidden hunger.” According to estimates, hidden hunger affects up to 2 billion people worldwide . Vitamin and mineral deficiencies account for 7.3% of the global burden of disease, with iron and vitamin A deficiencies among the 15 leading causes . It is an essential indicator for achieving the Sustainable Development Goals . However, a study conducted by Ortenzy et al. in 50 countries showed that the percentage of children consuming a poorly diversified diet remained stable, from 21% in 2010 to 24% in 2020 . In West and Central Africa, only one in four children (25%) has a sufficiently diverse diet. The proportion of children aged 6-23 months with a minimum dietary diversity is 34% in Nigeria, 28% in Mauritania, 26% in Benin, 24% in Ghana, 20% in Senegal, and 11% in Mali . In Côte d'Ivoire, despite an increase from 26% in 2016 to 28% in 2020 , minimum dietary diversity remains low among children aged 6 to 23 months. This is despite the activities to promote good infant and young child feeding practices implemented by the National Nutrition Program as part of the fight against malnutrition, particularly chronic malnutrition. In light of this situation, a study based on secondary data was conducted by the executive secretariat of the National Council for Nutrition, Food, and Early Childhood Development to identify the factors explaining this low dietary diversity in Côte d'Ivoire . The study found that low levels of maternal education and poor access to drinking water were the causes of this weakness. However, the study was unable to highlight the specific characteristics of each area, especially those areas severely affected by malnutrition. The objective of this study, based on primary data, was to examine the factors associated with dietary diversity among children aged 6 to 23 months in the localities with the highest prevalence of chronic malnutrition in Côte d'Ivoire.

This study was conducted in Woroba and Zanzan administrative districts. Woroba district, located in the northwest of Côte d'Ivoire, is the fifth largest in the country. It covers an area of 31,088 km² and has an estimated population of 1,184,813 in 2021, with an average of 3,000 assisted births per year in the district and 14 health facilities. Zanzan district, located in the east of Côte d'Ivoire, has an area of 38,251 km² and an estimated population of 1,235,886 inhabitants in 2021. It has 10 health facilities and records an average of 2,700 assisted deliveries per year. These districts are among the 14 districts of Côte d'Ivoire and have the highest prevalence of chronic malnutrition (30.4% and 30.5%) .

This was a cross-sectional analytical study conducted in 12-day period from March 19 to 30, 2023.

The study population consisted of mother-child pairs.

The study included, mother-child pairs who had lived in the study areas for six months or more and who were present during the survey period were included.

Those who were ill at the time of the survey and who did not consent to participate in the study were excluded.

The sample size calculation, which took into account the prevalence of dietary diversity in the Zanzan district (18.2%) resulted in a sample size of 229 mother-child pairs.

Mother-child pairs were selected from households as follows: within each administrative district, one locality was selected at random. The locality of Bondoukou was selected for the Zanzan district and that of Touba for the Zanzan district. Within each locality, seven clusters of 20 households were selected. The distribution of clusters according to environment was based on population density. Thus, three clusters were chosen in urban areas and four clusters in rural areas. Within each cluster, the investigators sought the center and randomly selected the direction in which to begin the survey. The first household in that direction was chosen, and a step of two households was applied for the following households. In each household, a mother-child pair was selected, and if there were several targets, the target was selected at random. This distribution resulted in a total of 280 mother-child pairs.

Data collection was carried out during a face-to-face interview with the mothers using a questionnaire that gathered information on the sociodemographic characteristics of the children and the household. To obtain data on children's feeding practices, the 24-hour recall method was used.

Data were analyzed using SPSS 26 software. The dependent variable was dietary diversity. It was recoded as 0 for low dietary diversity and 1 for good dietary diversity. Good dietary diversity was defined as the child having consumed at least 5 of the 8 recommended food groups (i.e., a score ≥ 5 out of 8) . The eight food groups are: breast milk; cereals, roots, and tubers; legumes and nuts; dairy products; meat products (meat/fish/poultry); eggs; vitamin A-rich fruits and vegetables; and other fruits and vegetables. The independent variables were derived from the mother's sociodemographic factors and the characteristics of the household and the child. Analysis consisted of performing the usual descriptive statistics with, depending on the nature of the variables, frequencies, proportions, and means with their standard deviations. Bivariate analysis used, Chi2 test and identified associations between the variables studied. Independent variables for which the associations had a p-value of less than 0.20 were used in the multivariate analysis through logistic regression. Significance was confirmed for a p-value ˂0.05.

Approval was obtained from the National Committee for Ethics in Life Sciences and Health. The informed consent of each mother was sought after she had been informed of the voluntary, free, and unpaid nature of the study. The survey only began once written consent had been obtained.

A total of 293 mother-child pairs were surveyed. More than half (53.2%) of mothers were aged between 20 and 29, with extremes of 15 and 50 and an average age of 26.3 ± 6.6 years. Most mothers (63.5%) had no schooling, were engaged in income-generating activities (74.1%), and lived with a partner (81.9%). The average age of the children was 13.54 ± 5.1 months, and more than half (65.2%) were between 12 and 24 months old. More than half of the children were female (53.2%) and lived in rural areas (54.6%). The majority of children lived in households with no other children aged 0 to 24 months (88.7%) (Table 1).

With regard to advice on child nutrition, 22.2% of mothers received it from health workers and 24.9% from their social circle. Three out of four children (75.1%) had poor dietary diversity (Table 2). The most commonly consumed foods were cereals, roots, and tubers (87.4%), fruits and vegetables (74.7%), and animal proteins (59.4%). Less than 3 out of 10 children consumed dairy products (23.9%), legumes and nuts (13.7%), eggs (10.6%), and other fruits and vegetables (2.4%).

In bivariate analysis (Table 3), minimal dietary diversity was significantly associated with place of residence (p = 0.015), mother's educational level (p = 0.042), marital status (p = 0.001), and child's age (p = 0.000). In multivariate analyses (Table 4), the factors significantly associated with dietary diversity were single marital status among mothers [p = 0.002; OR = 2.93 (1.5–5.73)] and minimum primary education level [p = 0.042; OR = 2.15 (1.03–4.48)]. Among children, age 12–24 months was significantly associated with dietary diversity [p=0.000; OR= 3.72 (1.86–7.43)].

Poor quality diets are one of the major barriers to survival, growth, development and learning for children . Throughout the world, young children do not benefit from appropriate diets, the diet they receive is neither age-appropriate nor satisfactory in terms of frequency and diversity . It appears from our study that the foods most consumed by children were cereals, roots, and tubers. This observation is the same in all regions of Côte d'Ivoire, regardless of age and in those of several studies carried out . Moreover, our study noted a low consumption of dairy products, legumes, and eggs. The low consumption of legumes and eggs was noted in Annane study and in 2019 report of UNICEF on children situation in the world . This low consumption of legumes is justified by the lack of knowledge about the use of legumes in children’s diets. Emphasis must be placed on these foods in nutritional education provided as part of children under 2 years old nutrition. The low consumption of eggs would find its explanation in taboos which stipulate that its consumption by children would promote the practice of theft among them . Regarding food diversity, its prevalence in the two localities with a high prevalence of malnutrition in Côte d'Ivoire (Touba and Bondoukou) was 24.9%. This result is slightly below those found by the national multisectoral platform for nutrition in Côte d'Ivoire in 2016, which noted a food diversity of 26% et and by Agboaye et al (25.5%) in 2021 . Khamis et al in Tanzania (26%) and Opoku et al in Ghana (35.3%) noted higher prevalence of dietary diversity in their studies. This low diversity could be explained by mothers' lack of knowledge about the use of certain foods in children’s meals, by the existence of taboos, and by difficulties in accessing certain foods. This observation raises the need to strengthen all the strategies that have been implemented by the executive secretariat of the national council for nutrition, food, and early childhood development and by the multisectoral national nutrition plan, to improve the diet of the population and especially children living in areas with low dietary diversity . Regarding the factors associated with dietary diversity, our study found a significant association with an children age of 12 months and older, the level of education, and the marital status of the mother. No significant association was found between place of residence and dietary diversity, even if place of residence explained urban and rural differences in educational attainment. Indeed, mothers in urban areas were significantly more educated than those in rural areas. This association between place of residence and level of education could make place of residence as a confounding variable. However, logistic regression considering place of residence along with other variables made it possible to objectively assess the influence of each variable on children's dietary diversity. Concerning age, our study noted that being older than 12 months was positively related to dietary diversity. In fact, children aged 12 to 23 months were 4.5 times more likely to have good diversity compared to those aged 6 to 11 months. Our results are consistent with those of Opoku et al. in Ghana . These findings may be due to mothers perceiving young children as having a low intestinal capacity to digest certain foods. Consequently, the complementary foods offered to young children lack variety and are often based solely on cereals and legumes. Furthermore, the older the child becomes, the more this perception of mothers diminishes, and the child is integrated into the family meal, which is more diversified. This explains the greater diversity among children aged 9 to even 12 months. Moreover, the good diversity among older children could also be attributed to their preparation to accept foods in various forms and their greater familiarity with certain foods compared to younger infants . Regarding education level, our study showed that mothers with at least a primary education were 2.8 times more likely to have good dietary diversity. This finding was also made by Frempong in Ghana . Several studies have shown a strong association between maternal education and positive nutritional outcomes in children . Indeed, education enables women to provide appropriate care for their children . Furthermore, it is linked to higher incomes and better nutritional knowledge, both of which are associated with child diets and nutrition . This suggests the importance of intensifying strategies to increase women's education and strengthen nutrition education at the school level to fight poor dietary diversity. Our study noted that, children of single mothers were nearly three times more likely to have good dietary diversity. This could be explained by the fact that these mothers receive more support from their extended families and have more time to care for their children, unlike mothers living with a partner, who are more concerned with running their households and face different economic realities than single mothers. Our result are in contradiction with studies conducted by Dhami and Ng who noted that children with single mothers were more likely to have poor dietary diversity due to the fact that single mothers lack support from their families or communities, which leads to poor infant feeding practices. In order, to improve dietary diversity among children, it would be appropriate to strengthen the training and supervision of health workers on feeding indicators for infants and young child especially among uneducated mothers and mothers living with a partner, as well as children under 1 year of age. The message on dietary diversity for children should be tailored to uneducated mothers. Families and friends of mothers should be encouraged to support them, and community cooking demonstrations on preparation of complementary foods should be organized.

This work has limits. On the one hand, its transversal nature prevents establishing the causal link, the temporal sequence between explanatory variables and good dietary diversity. On the other hand, dietary diversity as measured in this study refers to whether or not food is consumed on the eve of the survey. However, not having consumed a food or group of foods on the eve of the interview with the investigators does not imply that this food or group is not part of the infant’s eating habits. However, it should be noted that these limitations do not alter the informative nature of this study. It has the merit of highlighting the factors that explain dietary diversity among children aged 6 to 23 months in our study area.

Good dietary diversity among children aged 6 to 23 months in the study areas was relatively low. It was associated with the child's age, mother’s marital status and education. Interventions aimed at improving maternal education and diversified food consumption among children aged 6 to 59 months should be strengthened to break the intergenerational cycle of malnutrition.

Sable Parfait Stephane: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Ilupeju Victoire: Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

Loukou Gilbert Konan: Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

Ekou Franck Kokora: Visualization, Writing – original draft, Writing – review & editing

Bomouan Jean Fabien: Visualization, Writing – original draft, Writing – review & editing

Kouassi Kaudjhis Regine: Visualization, Writing – original draft, Writing – review & editing

Abinan Audrey: Visualization, Writing – original draft, Writing – review & editing

Coulibaly Amed: Visualization, Writing – original draft, Writing – review & editing

Konan Yao Eugène: Visualization, Writing – original draft, Writing – review & editing

Ake-Tano Odile: Visualization, Writing – original draft, Writing – review & editing

The authors declare no conflicts of interest.