Food and nutritional security remain a major public health concern worldwide, particularly in the face of the current economic crisis and climate-related disruptions [1]-[3]. Despite the synergy among the food programs and policies discussed in the context of SDG 2, which aims to ensure food and nutritional security, a portion of the world’s population still struggles to obtain sufficient and well-diversified food [4][5].

Food insecurity is not merely a matter of insufficient food intake [6], but also due to poor-quality and undiversified diets [7][8]. Limited dietary diversity and inadequate food intake are major determinants of malnutrition, particularly among vulnerable groups such as children under 5 and women of reproductive age, with a 45% increase in the risk of mortality and morbidity in 2011 and a 75% increase in 2019, especially among pregnant and breastfeeding women in developing countries [9]-[12]. Several indicators, including the Food Consumption Score (FCS) and the Household Dietary Diversity Score (HDDS), are used by various international organizations such as the WFP, the FAO, the WHO, and UNICEF [13].

In sub-Saharan Africa, many households continue to face food insecurity and a lack of dietary diversity [1][14]-[16]. The DRC is among the hardest-hit countries, despite its resource potential. In 2023, the IPC assessment found that 40% of the population was vulnerable to chronic food and nutrition insecurity, including 25.1 million people facing moderate food insecurity and 15.7 million facing severe food insecurity [1][17]. And in 2016, more than 55% of households had a food consumption score that was borderline or poor [18].

In Tshopo Province, particularly in Yangambi, poverty and difficult socioeconomic conditions contribute to inadequate and undiversified diets, according to the 2023 CPI analysis [17]. These conditions directly affect the dietary habits of households, women, and children by leading to low food intake and limited dietary diversity. Several studies have shown that greater dietary diversity is associated with higher socioeconomic status, higher levels of household food security, and better living conditions for households [19]-[22].

Although the relationship between these factors and dietary scores has been documented internationally and in some African countries, there are almost no studies that specifically explore the factors determining food consumption and dietary diversity in local contexts within the DRC. Many studies are conducted in different contexts and do not take into account the specific local conditions of the DRC.

The same is true for Tshopo Province, particularly in the City of Yangambi, where socioeconomic conditions remain difficult, limiting households’ access to a good, diverse, and high-quality diet. This results in low food consumption and diversity scores, with 55% of households having a borderline or poor FCS, and 80% of children aged 6 - 23 months who have never achieved their minimum dietary diversity [18][23].

Few studies have thoroughly analyzed the determinants of dietary intake and diversity scores in the specific context of Yangambi. Knowledge and understanding of these factors are essential for guiding nutrition research, interventions, and public health policies appropriate to the local context. This study therefore aims to analyze the factors associated with low dietary intake and diversity scores among households in the city of Yangambi, in the Democratic Republic of the Congo.

ScopeoftheStudy: The study was conducted in the City of Yangambi, located in Tshopo Province in the Democratic Republic of the Congo [24]. The city is primarily agricultural, and the population relies mainly on subsistence farming, livestock raising, fishing, and small-scale trade [25].

Studypopulation: The study population consisted mainly of households in the Yangambi city.

StudyPeriod: The study was conducted between January and April 2026.

Typesofstudies: This was a cross-sectional analytical study designed to examine the factors associated with low scores for dietary intake and dietary diversity among households in Yangambi.

Sampling: A two-stage probability cluster sampling method was used. The various accessible villages in the City of Yangambi, selected at random, were treated as clusters.

Samplesize: The sample size was calculated using Cochran’s formula, based on a prevalence of 41% for chronic food insecurity reported by the 2023 IPC in the DRC [17], a precision of 5%, and a 95% confidence interval. After accounting for a cluster effect of 1.5 and a nonresponse rate of 10%, the final sample size was 614 households. A two-stage cluster sampling method was used, with 31 clusters surveyed. Thirty clusters contained 20 households each, whereas the final cluster contained 14 households, yielding a total of 614 households.

Inclusionandexclusioncriteria: The study included households selected from the Yangambi city whose heads had agreed to participate in the study. Households whose heads had not given their consent to participate in the study were excluded from the study.

DataCollection: Data were collected using a structured questionnaire that covered the sociodemographic, socioeconomic, environmental, and dietary characteristics of households.

Processingtechniquesandstatisticalanalysis: The data were entered into Excel and analyzed using STATA version 13. Categorical variables were presented as frequencies and percentages, while quantitative variables were summarized as means ± standard deviations or medians with interquartile ranges. Associations between dependent and independent variables were assessed using Pearson’s chi-square test, Student’s t-test, ANOVA with Bonferroni correction, and binary logistic regression. The results were expressed as crude and adjusted odds ratios (OR) with their 95% confidence intervals. The threshold for statistical significance was set at p < 0.05. Variables associated with the outcomes in bivariate analyses were initially considered for inclusion in the multivariable logistic regression models. Preliminary models revealed collinearity and unstable estimates for some variables and catégories. Consequently, model refinement was performed by removing variables contributing to collinearity or generating excessively large odds ratios and wide confidence intervals. Multicollinearity among variables retained in the final models was assessed using variance inflation factors (VIF). The maximum VIF values were 2.84 and 2.75 for the household dietary diversity and food consumption models, respectively, indicating no evidence of problematic multicollinearity. For logistic regression analysis, the food consumption score (FCS) was dichotomized into acceptable (>35) and unacceptable (poor and borderline ≤ 35) categories according to the World Food Programme (WFP) classification. The household dietary diversity score (HDDS) was also dichotomized into high dietary diversity and low to moderate dietary diversity categories. Households with acceptable food consumption scores and high dietary diversity scores were considered as having favorable outcomes in the regression models.

Ethicalconsiderations: The study was approved by the Ethics Committee of the Faculty of Medicine at the University of Kisangani. Administrative approvals were obtained from local authorities. Informed consent was obtained from participants before the start of data collection. The data were anonymized and treated confidentially. The authors declare that they have not conflict of interest.

Sociodemographic,socioeconomic,environmental,anddietarycharacteristicsofthehouseholdssurveyed.

Table 1 presents the sociodemographic, socioeconomic, environmental, and dietary characteristics of the households surveyed. The average age of household heads is 40.70 ± 11.71 years, with the youngest being 18 years old and the oldest 71 years old. The 30 - 44 age group is the most represented, at 43.81%, and households are generally headed by men, at 88.27%. Half of the heads of households (50.00%) have a primary school education, and agriculture is the main occupation of half of the heads of households (50.2%). Most heads of households are predominantly monogamous (70.03%).

Table 1.Sociodemographic, socioeconomic, environmental, and dietary characteristics of the households surveyed.

Women had a median age of 32, with two age ranges extending from 25 to 41, and two dominant age groups: those under 29 (41.53%) and those aged 30 - 44 (40.55%). Regarding their physiological status, 30.1% were breastfeeding, and 18.72% were pregnant. The majority of children under 5 years of age were in the 24 - 59-month age group (69.73%), and half of the households had 5 - 7 members (52.61%), with an average household size of 6.1 ± 1.6 members, ranging from 2 to 10 people.

More than half of the households were indigenous (64.73%), and many of the surveyed households were from Yalikombo (26.71%) and Yaselia (23.13%). More than half of households had a monthly income of less than 75,000 Congolese francs (52.93%), with a median monthly income of 75,000 Congolese francs and a range of 50,000 to 150,000 Congolese francs. Half of children under 18 contributed (57.49%), while many households reported that their income had decreased compared to the previous year (87.79%).

Agriculture was the primary source of household income for half of households (50.00%) and some households spent less than 20,000 Congolese francs per month on food (19.06%), with a median monthly food expenditure of 40,000 Congolese francs and interquartile ranges of 23,000 to 60,000 Congolese francs. 37.13% of households used their income primarily for food consumption.

More than half of households did not have access to safe drinking water (64.66%); the primary sources of water used by households were river water (35.83%) and pump wells (35.34%). Many households (71.5%) had an unacceptable food consumption score, with a median score of 26.5, an interquartile range of 24.5 to 37.5, and extremes ranging from 20.5 to 59.5. Most households had a moderate dietary diversity score (75.41%), with a median score of 5, an interquartile range of 5 to 6, and extremes ranging from 4 to 11 food groups, whereas 24.59% had a high dietary diversity score. The household hunger index showed that 23.45% of households experienced severe hunger and 21.17% experienced moderate hunger, with a median score of 1.

Multivariate Logistic Regression of Factors Associated with the Household Food Consumption Score

Table 2 presents the factors associated with the household food consumption score following a multivariate logistic regression. After adjustment, the age of the household head remained significantly associated with the household food consumption score (<0.001). Households headed by heads aged 30 - 44, 45 - 59, and over 60 were 27 times (ORa = 27.04; 95% CI = 9.74 - 75.08; p < 0.001), 46 times (ORa = 45.74; 95% CI = 14.13 - 148.03; p < 0.001), and 105 times (ORa = 104.63; 95% CI = 21.49 - 509.41; p < 0.001) more likely to have a good dietary intake score compared to households headed by heads of household under the age of 29.

Table 2. Factors associated with the household food consumption score (Multivariate logistic regression).

Even after adjustment, educational level remained associated with the dietary intake score. Households headed by heads of household with a secondary education level were 4 times (ORa = 3.61; 95% CI = 1.64 - 7.94; p = 0.001) as likely to have a good dietary intake score compared to those without a secondary education. However, primary education remained non-significant after adjustment (p = 0.774), and the woman’s physiological status also remained non-significantly associated with the dietary intake score (p = 0.585).

Household size and average monthly household income remained significantly associated with the household food consumption score (p < 0.005). Households with 5 to 7 members (ORa = 0.07; 95% CI = 0.03 - 0.14; p < 0.001) and those with more than 8 members (ORa = 0.04; 95% CI = 0.01 - 0.09; p < 0.001) were respectively less likely to have a good dietary intake score compared to households with fewer than 5 members.

Households with a monthly income of between 76,000 and 200,000 Congolese francs were three times more likely (ORa = 2.17; 95% CI = 1.59 - 4.63; p < 0.001) to have a good food consumption score, unlike households with an income of more than 201,000 Congolese francs.

Multivariate Logistic Regression of Factors Associated with the Household Food Diversity Score

Table 3 presents the factors associated with the household dietary diversity score following a multivariate logistic regression. The age of the head of household remains significantly associated with the household dietary diversity score after adjustment (<0.001).

Table 3. Factors associated with household dietary diversity scores (Multivariate logistic regression).

Thus, households headed by individuals aged 30 - 44, 45 - 59, and 60 or older are 25 times (ORa = 24.72; 95% CI = 7.57 - 80.70; p < 0.001), 435 times (ORa = 434.65; 95% CI = 102.10 - 1850.28; p < 0.001), and 612 times (ORa = 611.65; 95% CI = 114.12 - 3278.26; p < 0.001), respectively, compared to households headed by individuals under 29 years of age. The household head’s educational level remained significantly associated with the household’s diversity score after adjustment (<0.001); households whose heads were educated (primary, secondary, higher) were 10 times (ORa = 9.84; 95% CI = 2.99 - 32.35; p < 0.001), 11 times (ORa = 10.51; 95% CI = 2.89 - 38.19; p < 0.001), and 567 times (ORa = 567.22; 95% CI = 80.92 - 3976.01; p < 0.001) more likely to have a high dietary diversity score.

Household size remains significantly associated with the household dietary diversity score after adjustment (<0.001); specifically, households with 5 to 7 members (ORa = 0.01; 95% CI = 0.01 - 0.04; p < 0.001) and those with 8 or more members (ORa = 0.03; 95% CI = 0.01 - 0.08; p < 0.001) are less likely to have a high dietary diversity score compared to smaller households with fewer than 4 members.

The village (Weko, Yaekama, Yalisombo) was also significantly associated with the household dietary diversity score (<0.005); households in the villages of Weko (ORa = 0.03; 95% CI = 0.01 - 0.15; p < 0.001), Yaekama (ORa = 0.11; 95% CI = 0.03 - 0.38; p < 0.005), and Yalisombo (ORa = 0.03; 95% CI = 0.01 - 0.19; p < 0.001) had, respectively, a zero chance of achieving a high dietary diversity score compared to the village of Lilanda; however, the villages of Yalikombo (p = 0.362) and Yaselia (p = 0.69) did not, respectively, show a significant association after adjustment for household dietary diversity score.

Average monthly household income also remained a significant factor after adjustment (<0.05); households with incomes between 76,000 and 200,000 Congolese francs were less likely to achieve a high dietary diversity score (ORa = 0.38; 95% CI = 0.19 - 0.77) compared with households earning less than 75,000 congolese francs. However, households with a monthly income exceeding 201,000 Congolese francs showed no significant association with the dietary diversity score after adjustment (p = 0.1).

Households with incomes between 76,000 and 200,000 Congolese francs were less likely to achieve a high dietary diversity score compared with households earning less than 75,000 Congolese francs.

The objective of this study was to analyze the factors associated with low household food consumption and dietary diversity scores in the Yangambi city of the Democratic Republic of the Congo. The results show that many households had a borderline or poor food consumption score and also a low or moderate dietary diversity score. After adjustment using binary logistic regression in the multivariate analysis, several variables remained significantly associated with household food consumption and diversity scores. These include the age of the head of household, the head of household’s educational level, household size, monthly household income, and the household’s village of origin.

Ageoftheheadofhousehold: The age of the household head remained significantly associated with household dietary intake and diversity scores after multivariate logistic regression. Households headed by older heads (over 30 years old) were more likely to have high dietary intake and diversity scores compared to households headed by younger heads (under 30 years old).

These results are consistent with those of Gebre T and Nkonde C where the older age of the household head also increases the likelihood of higher household food consumption and diversity scores [15][26]. This may be explained by the fact that older heads of households have greater economic stability and more experience in managing food needs and allocating household income to food compared to younger heads.

Educationalleveloftheheadofhousehold: After multivariate logistic regression, the head of household’s level of education remained significantly associated with household food consumption and diversity scores; households headed by more educated heads (secondary, higher education) were more likely to have higher food consumption and diversity scores compared to households headed by heads with little or no education. Ouedraogo DS et al., Gibson E et al. and Tanoh EF et al. confirmed the same results in their studies, where the risk of having an inadequate and undiversified diet was higher in households headed by uneducated heads compared to those headed by educated heads [12][19][27]. This would explain the role of education: a higher level of education enables households to consume a greater variety of foods. Furthermore, education improves nutritional knowledge, promotes the proper use of household income, and increases access to other economic opportunities.

Householdsize: After logistic regression, household size remained significantly associated with dietary intake and diversity scores; large households (more than 5 members) were less likely to have high dietary intake and diversity scores compared to smaller households. These results are consistent with those presented by Boulom S et al., Nikièma V et al., and Awoke W et al., who emphasize that a one-unit increase in family size reduces household food security; the presence of a large number of members in a household tends to lead to food insecurity compared to households with smaller families [3][11][28].

This observation may be linked to the fact that a household with multiple members has greater food needs, which are inevitably dependent on the household’s economic status in terms of affordability.

Household'svillageoforigin: The household’s village of origin remained significantly associated with household dietary intake and diversity after adjustments based on multivariate logistic regression. Some villages had a lower probability of achieving high scores for dietary intake and diversity compared to others that had a high probability of achieving high dietary scores.

This could be influenced by local geographic, agricultural, and socioeconomic conditions that may affect household dietary intake and diversity in certain villages. Several studies conducted in the Yangambi region, including that by Mangaza L et al., demonstrate the existence of these geographic, agricultural, and socioeconomic differences that impact households’ nutritional and dietary conditions [17][24][25].

Averagemonthlyhouseholdincome: Average monthly household income remained significantly associated with household food consumption and dietary diversity scores after multivariate logistic regression. Households with a monthly income between 76,000 and 200,000 Congolese francs were more likely to achieve an acceptable food consumption score compared with households earning less than 75,000 Congolese francs. However, regarding dietary diversity, households in this income category showed lower odds of achieving a high dietary diversity score after adjustment. Similar associations between household income and dietary outcomes have been reported in other studies [20][21][29]. These findings may reflect differences in household expenditure priorities, food purchasing patterns, and access to diverse foods.

Strengthsandlimitationsofthestudy: This study provides recent data on the factors associated with low household food consumption and dietary diversity scores in a rural city in the Democratic Republic of the Congo that has received little attention in the literature. The internationally recognized food indicators used in this study, combined with multivariate analysis, identified the main factors linked to low food consumption and dietary diversity scores in the City of Yangambi.

Nevertheless, the cross-sectional nature of the study does not allow for the establishment of a causal relationship between the factors studied and dietary scores. As a limitation, some of the information collected was based on self-reports from household heads and may have been subject to recall bias.

Additionally, certain categories were underrepresented, which may have contributed to the high odds ratios mentioned in this study. Despite this, this study provides important information that can help guide nutritional interventions and public health policies regarding food security in the Yangambi City, as well as in other similar contexts in the Democratic Republic of the Congo.

Residual confounding cannot be excluded in this study. Seasonal variations in food availability, self-reported household income and food expenditure, as well as differences in market accessibility between villages, were not fully captured and may have influenced the associations observed in this study. In addition, some adjusted odds ratios were very large and associated with wide confidence intervals, possibly reflecting sparse data in certain categories and limited sample sizes for some subgroups. These factors should be considered when interpreting the findings with caution.

This study shows that several sociodemographic and socioeconomic factors, including the age and educational level of the head of household, household size, average monthly household income, and village of origin, were associated with household dietary intake and dietary diversity scores in the City of Yangambi.

These results support the need to focus and strengthen multisectoral food security interventions on improving the socioeconomic, sociodemographic, and dietary conditions of households in rural areas of the Democratic Republic of the Congo.