Introduction

In low-income countries, the persistent challenge of securing adequate and consistent food access is a multifaceted issue. It revolves around the fundamental goal of providing all individuals with access to safe, nutritious, and culturally appropriate food to support active and healthy lifestyles. Food security hinges on the alignment of food intake with nutritional requirements in a socially acceptable manner, while insufficient food access leads to inadequate nutrition. Addressing food insecurity is a complex endeavor due to its multifaceted nature, encompassing dimensions such as availability, access, use, and stability. Clearly defined terms and benchmarks are indispensable for shaping targeted policies. Sustainable food security on a global scale relies on dependable and equitable food systems that ensure all households and communities have access to sufficient and high-quality food (Coleman-Jensen, 2011). Food insecurity serves as a critical indicator of poverty and presents a formidable challenge in low-income countries (Jama & Pizarro, 2008). Contributing factors include population growth, rising living costs, climate change, civil conflicts, and societal norms. Environmental issues, droughts, and political crises have hampered governmental efforts to tackle these challenges, leading to the impoverishment of communities and ordinary households (Ibnouf, 2011). In Sudan, as in households worldwide, many communities depend heavily on natural resources for their sustenance. Unfortunately, these resources are under threat from unpredictable rainfall patterns and extended droughts, making it increasingly challenging to maintain living standards.

The Global Hunger Index, 2022 paints a bleak picture for Sudan, ranking it 106th out of 121 countries with available data. With a GHI score of 28.8, Sudan is classified among nations grappling with severe hunger (Von Grebmer et al., 2022). A combination of conflict and economic downturn has left a staggering 20.3 million people in Sudan – more than 42% of the population – in acute food insecurity. As a result, they fall into the Integrated Food Security (IPC) classification phase 3 or higher, indicating a crisis or worse, particularly between July and September 2023. Within this affected population, approximately 14 million individuals (29% of the population) are in IPC Phase 3, an emergency situation. Furthermore, nearly 6.3 million people (13% of the population) are in IPC Phase 4, denoting an even more dire emergency (ipc, 2023; wfp, 2023). These stark findings underscore the gravity of Sudan's food insecurity crisis, demanding immediate and comprehensive action.

Ongoing conflict, inadequate provision of basic services, and economic downturns are the primary drivers of humanitarian needs in the Red Sea state. This persistent conflict has exacerbated the state's already challenging food security situation. Vulnerable households in the region face reduced purchasing power due to economic hardships and rising prices, negatively impacting nutrition and overall food security. Indeed, food security is a pressing concern in the Red Sea state, exacerbated by the effects of climate change. In recent months, escalating fuel and food prices have resulted in the loss of up to 40% of the state's food production due to import, storage, and transportation challenges. Market access remains weak, further devaluing food. According to the Integrated Classification of Food Security Phase(ipc, 2023), the Red Sea Province is classified as moderately food insecure, with eight sites corresponding to pre-season falling into IPC Phase 3 and two sites classified as Emergency (IPC Phase 4), i.e., severity, foods insecure.

Given the multifaceted challenges associated with food security in the Red Sea State, it is imperative to gain a comprehensive understanding of the factors influencing household food security. While previous research has explored food security determinants in various countries worldwide such as Indonesia (Hakim et al., 2021), Zimbabwe (Mango et al., 2014), Afghanistan (Akbay & Ahmadzai, 2020), South Africa (Maziya et al., 2017), Ethiopia (Agidew & Singh, 2018), Niger (Zakari et al., 2014), Nigeria (Mustapha et al., 2018), and Pakistan (Zhou et al., 2019), there is a notable gap when it comes to understanding these factors specifically in the Red Sea State. This study aims to bridge this gap by conducting an in-depth investigation into the socioeconomic determinants that impact food security in Sudan's Red Sea state. By uncovering these critical factors, policymakers and stakeholders can tailor interventions to enhance food security, promote sustainable agricultural practices, and ensure equitable access to nutritious food.

Section "Literature Review" provides an in-depth exploration of existing literature on determinants of household food security in Sudan. Section "Materials & Methods" outlines the data collection and methodology employed in the study. The statistical analysis using linear regression is given in Section "Statistical Analysis". Dependence and independence variables are provided in Section "Dependence variable"&"Independent variables". Section "Result and discussion" includes the results and discussion related to linear regression results and classification of food security status. Finally, Section "Conclusion" concludes the study.

Literature review

Factors influencing household food security globally

Food insecurity, according to Anderson (Anderson, 1990), is characterized as the limited or uncertain access to nutritionally adequate and safe foods, resulting from insufficient financial resources or other means of obtaining acceptable foods. In simpler terms, it occurs when households struggle to meet their basic needs due to financial constraints, leading to reduced food quality and variety. Understanding the determinants of household food security is vital for effective hunger and malnutrition mitigation (Haysom & Tawodzera, 2018). Numerous studies have explored how household characteristics impact food security. For instance, (Asare-Nuamah, 2021) found that factors like gender, household size, farming experience, access to forests, and adaptation strategies significantly affect food security in rural Ghana. Similarly, Bolarinwa et al. revealed that sociodemographic aspects such as wealth, income diversification, and location influence food insecurity in Rwanda (Bolarinwa et al., 2020).

These scholars emphasize that multiple determinants influence household food insecurity across diverse contexts. Grimaccia and Naccarato (2019) identified economic and social characteristics as key factors affecting food security worldwide (Grimaccia & Naccarato, 2019). Abdullah et al. found that age, gender, education, remittances, unemployment, wealth, and health significantly impact household resilience in rural Pakistan (Abdullah et al., 2019). Environmental factors also play a role; Shim et al. (2018) demonstrated spatial variations in nutritious food availability for rural Korean households based on economic resources (Shim et al., 2018). Importantly, food security status fluctuates over time, as shown by Bolarinwa et al. (2020), highlighting the need for interventions addressing these dynamics (Bolarinwa et al., 2020). In summary, individual and household characteristics, community environment, and temporal changes interact to shape household food security outcomes. This nuanced understanding is crucial for tailoring interventions and initiatives to improve food security in diverse socioeconomic contexts worldwide.

Determinants of household food security in Sudan

Research on the determinants of household food security in Sudan is limited. This review provides an overview of relevant studies focusing on gender, monitoring methods, grain reserves, and gender contributions to food security. Mohamed Nour and Abdalla (2021) identified key factors contributing to household food insecurity in Kassala State, Sudan, including household food production, income, land ownership, livestock, and access to markets and infrastructure. Increasing household income and improving food production were found to significantly reduce the likelihood of food insecurity (Mohamed Nour & Abdalla, 2021). Similarly, Dafalla (2022) discovered that household size, education level, employment status, livestock ownership, healthcare access, and income positively influenced food security in female-headed households in Sennar State, Sudan. Larger households were more vulnerable to food insecurity, while higher education and income levels improved food security (Fatima Mohammed Ali, 2022).

Additionally, Elzaki et al. (2021) examined food consumption patterns in rural Sudanese households using the QUAIDS model, revealing significant price inelasticity for food items. Demographic factors and place of residence also impacted consumption patterns, with potential policy implications (Elzaki et al., 2021). Ibnouf (2009) emphasized the importance of women's participation in enhancing household food security, as women contribute not only to food production but also to dietary diversity and quality. Ibnouf (2012) further underscored women's indigenous knowledge in food processing and preservation, advocating for sustainable solutions in rural Sudan (Ibnouf, 2009; Ibnouf, 2012). Similarly, Ibnouf et al. evaluated the roles of rural men and women in maintaining household supplies in western Sudan, aiding policymakers in addressing regional variations in food security (Ibnouf & Ibnouf, 2016). In post-conflict South Sudan, Lokosang et al. (2011) employed ordinal logistic regression to identify predictors of food insecurity levels, contributing to effective surveillance and intervention strategies (Lokosang et al., 2011). Although these studies highlight various aspects of food security in Sudan, including gender roles, monitoring methods, grain reserves, and men's and women's contributions, their findings can guide policymakers and practitioners in developing effective strategies to alleviate hunger and malnutrition and promote sustainable solutions in Sudan.

Logistic regression model

Logistic regression models have been widely employed to investigate household food security determinants. Ahmed et al. (2017) conducted a study in rural Pakistan, identifying factors such as household size, income, food prices, healthcare expenses, and debt as significant determinants of food security among small agricultural households (Ahmed et al., 2017). Similarly, Ibok et al. (2014) in Nigeria found that factors like formal education, farming knowledge, farmer age, farming as the main occupation, household size, income from farming, and food crop production influenced food security (Ibok et al., 2014). Bolarinwa et al. (2020) explored food security determinants in Rwanda, highlighting household sociodemographic characteristics, asset ownership, diversification strategies, and place of residence as consistent influencers of food insecurity (Bolarinwa et al., 2020).

Maharjan and Joshi (2011) conducted a Nepal-focused study showing that programs targeting professional groups and small landowners or landless individuals significantly reduced food insecurity, as indicated by their logistic regression model (Maharjan & Joshi, 2011). Additionally, Abdullah et al. (2019) explored factors affecting household vulnerability to insufficient access to nutritious food in rural Pakistan, revealing variables such as age, gender, education, remittances, unemployment, wealth, and morbidity as crucial determinants of overall household nutrition levels (Zhou et al., 2019).Collectively, these results demonstrate the effectiveness and applicability of the logistic regression model in diverse contexts across different countries, providing valuable insights into understanding the primary drivers of food security dynamics at the household level. By synthesizing the literature on these factors, this study aims to offer evidence-based insights into the specific determinants of household food security in the Red Sea state. These insights will inform strategies and interventions aimed at improving food security, enhancing nutritional outcomes, and promoting sustainable agricultural practices in the region.

Materials & methods

Study location

The Red Sea state, one of Sudan's 18 states, occupies the northeastern region of the country. Bordered by the Red Sea to the east, covering an area of approximately 216,000 km2. It shares its borders with Kassala State and Eritrea to the south, Nile State to the west, and Egypt to the north. Port Sudan, the state's capital and primary seaport, facilitates approximately 90% of Sudan's international trade. The Red Sea state comprises 10 localities: Hala’ib, Al ganab, Port Sudan, Swaken, Sinkat, Haya, Tawkar, Agig, Jubayt elma’adin, and Dordib, with two villages serving as refuge for internally displaced persons according to the International Organization for Migration (IOM). International Organization for Migration (IOM), Dec 29 2022. DTM Sudan — Baseline Assessment — Returnees from Abroad — Round 5. IOM, Sudan, 2022). The state is home to diverse tribal groups, enriching its cultural tapestry. Port Sudan, the bustling port city, holds significant economic importance as a gateway to international trade for the entire state and Sudan as a whole.

The study was conducted in the Red Sea state in Sudan, encompassing both urban and rural communities within the district. The inclusion of rural sites aimed to capture a wide spectrum of agricultural practices, resource availability, and socio-economic conditions typical of rural areas. Factors such as proximity to markets, infrastructure availability, and varying environmental conditions were carefully considered when selecting the locations. The primary goal was to conduct a comprehensive examination of the characteristics and challenges faced by the population residing in the Red Sea District. This approach facilitated an in-depth analysis of the diverse factors influencing household food security. According to the 2023 Comprehensive Food Security and Vulnerability Analysis (CFSVA) report (wfp, 2023), the Red Sea state has witnessed a significant rise in food insecurity, with food insecurity levels surging from 18% in 2022 to 49% in 2023, marking a 31% increase. Severe food insecurity has also seen a substantial increase during this period, rising from 2 to 16%.

Data source

The data for this study in the Red Sea State was obtained from the Consumption Patterns and Nutrition Study (CPNS), conducted in Sudan by the Ministry of Agriculture and Forestry and the Food Security Technical Secretariat (FSTS). The CPNS was a cross-sectional survey carried out in 18 districts (states) of Sudan between August and November 2019. This survey marked the first nationwide sample survey in Sudan since the national household survey in 2009. The CPNS employed a three-tier approach and a stratified sampling design. The primary sampling units were the listing areas identified during the 2008 Sudan population and housing census. The sampling method involved the random selection of three administrative units per sub-province (locality), three blocks per urban area, and three villages per rural area, all chosen through systematic sampling. This methodology was applied to a total of 174 localities, with each locality housing 96 households. The allocation of households was proportional to the administrative units. After excluding households with data errors, the valid sample size in the Red Sea state was determined to be 740 households. The stratified sampling technique ensured that the resulting sample accurately represented the various subgroups within the community, allowing for the classification and analysis of population subgroups. The data collected as part of the CPNS survey in the Red Sea State serves as a valuable foundation for examining household consumption patterns and diet-related factors in the specific context of the state (Fig. 1).

Fig. 1
figure 1

IPC Acute food insecurity projections of Sudan for July 2023 – September 2023. 2023 https://www.ipcinfo.org/ipcinfo-website/alerts-archive/issue-23/en/.

Full size image

Statistical analysis

Model specification

In this paper, the dependent variable is the composite Food Security Status (FSS). The FSS is calculated using the Food Consumption Score (FCS) as a measure. To construct a binary dependent variable, households were classified into three categories based on established guidelines by Coates J, Swindale A (Coates et al., 2007; Webb et al., 2006)): poor, borderline, and acceptable. To determine the overall food security status of each household, indicators were examined to assess whether they met certain thresholds. This analysis resulted in a dummy variable. As a result, the FCS was converted into two scores: 1 for food secure and 0 for food insecure, based on predetermined cut-offs for these indicators. In order to the dependent variable of food security is qualitative and dichotomous in nature, it can only take on two values: either present or absent. Therefore, using a traditional binary response method assigns a value of 1 to indicate food security and 0 to indicate its absence. This binary measure of food security has significant policy implications as it allows for clearer interpretation and practical application (Coleman-Jensen, 2011).

The study used a logistic regression model to analyze the association between the dependent variable (FSS) and the independent variables. The logistic regression model is suitable for binary response variables, such as food security status, which can take on two values: 1 for food secure and 0 for food insecure.

The logistic regression model can be represented as follows:

$$ln(Pi / (1 - Pi)) = {b}_{0} + {b}_{1}{X}_{1} + {b}_{2}{X}_{2} + ... + {b}_{ni}{X}_{ni}$$
(1)

Pi represents the likelihood of a household being food secure, and (1—Pi) represents the likelihood of a household being food insecure. The coefficients \({b}_{0},{b}_{1},{b}_{2}\) and \({b}_{ni}\) represent the estimated impacts of the independent variables \({X}_{1},{X}_{2},{X}_{3}\) and \({X}_{ni}\) on the log odds of food security.

In logistic regression analysis, the odds ratio is applied to assess the likelihood of food security based on the values of independent variables. It represents the ratio between the probability of a household is food secure and the probability of it being food insecure. The odds ratio provides valuable insights into the relationship between different categorical variables [3]. Several observations can be made when interpreting odds ratios in logistic regression:1.) If the odds ratio is greater than 1, it means that subjects within a given category have higher chances of achieving food security than those in the reference category.2.) Conversely, if the odds ratio is less than 1, this indicates that subjects within a given category have lower odds of achieving food security compared to those in the reference category.3) An odds ratio of 1 implies that subjects within a given category have a lower probability of achieving food security Category have similar or equivalent chances of achieving food security as those in the reference category. These interpretations provide powerful insights into understanding how different categorical variables affect and contribute to the likelihood of food security or insecurity in households (Table 1).

Table 1 Description of the variables used in the model
Full size table

Dependence variable

Food consumption score

In line with Wiesmann D's argument, the Food Consumption Score (FCS) serves as a quantitative measure to assess a household's food consumption habits over one week. It gauges diet diversity and nutritional quality by monitoring the frequency of food group consumption. The FCS offers insights into food accessibility, reflecting the range and regularity of food intake based on a week-long record of consumption (Wiesmann et al., 2009). This scoring system serves as a practical proxy for evaluating household food security, revealing levels of food sufficiency and potential insecurity (Carletto et al., 2013). To implement the FCS, household-level data is collected concerning the consumption patterns of 23 specific foods, as outlined in the 2006 Sudan Comprehensive Food Security and Vulnerability Analysis (CFSVA) report (Programme, 2006). These food items are categorized into distinct food groups using standardized data gathered over a seven-day period. Respondents are then queried about the frequency of their household's consumption of each food group or product during the specified time frame (Table 2).

Table 2 Food groups and standard weights used to calculate FCS
Full size table

Once the data is gathered, it is categorized into eight distinct food groups, and the consumption rates for each group are calculated by summing up the consumption frequencies of the individual foods within them. Any frequencies exceeding seven are truncated for consistency. The resulting consumption rates are then multiplied by pre-assigned weights for each food group, generating weighted scores for each group. These weighted scores across all food groups are summed to create the Food Consumption Score (FCS). Cutoff standards for the Food Consumption Groups (FCGs) are determined after calculating the FCS, considering the frequency distribution of the ratings and understanding the typical consumer behavior in the studied country or region. FCS scores falling between 0 and 21 typically indicate inadequate food consumption, while scores ranging from 21.5 to 35 suggest borderline consumption, and scores above 35 signify satisfactory consumption.

However, in countries like Sudan, where the customary daily intake of oil, fat, butter, and sugar is prevalent, there can be artificial inflation in FCGs values. To account for this, the minimum limits are adjusted accordingly, particularly for households with regular consumption of these items. In such cases, the revised thresholds are set at 28.5 for marginal consumption and 42 for satisfactory consumption. It's crucial to acknowledge that these limits may require testing and adaptation based on the specific context and dietary habits. This article proposes an alternative set of FCGs limits designed for home use, which consider the routine consumption of oil and sugar. These adjusted thresholds were utilized in the 2006 Sudan Comprehensive Food Security and Vulnerability Analysis (CFSVA) report (Programme, 2006).

Independent variables

Location factors

This analysis delves into demographic and socio-economic factors to gain insights into varying levels of food security among different households. In a Mali study conducted by Coulibaly et al. (2023), a significant disparity in food security was uncovered between urban and rural households. The study measured food security using per capita household food expenditure, highlighting the profound influence of urbanization on food security dynamics. Findings suggested that urban households experienced lower food security levels compared to their rural counterparts (Coulibaly et al., 2023). Conversely, a different perspective emerged from a study in Rwanda by Bolarinwa et al. (2020), which identified geographic factors as key determinants of food security. Specifically, households in urban areas were significantly less prone to food insecurity than their rural counterparts. This contrast underscores the vital role of spatial location in shaping household food security status (Bolarinwa et al., 2020).

Household size

Household size serves as an indicator of consumption needs and highlights the challenges in providing for household members. Several studies have explored the impact of household size on food security. For instance, research by Oduniyi and Tekana (2020) on food security in South African agricultural households found that even a single-member increase in household size correlated with a reduced likelihood of achieving food security (Oduniyi & Tekana, 2020). Similarly, Asare-Nuamah's study on food insecurity in rural Ghana emphasized the significant influence of household size on food security. These studies underscore the pivotal role of household size in shaping food security outcomes (Asare-Nuamah, 2021).

Age of household head

The age of the household head is an independent variable that can correlate with labor force participation and its subsequent impact on household food availability. Younger, more agile household heads often engage in extensive farming activities, while older heads may participate less in labor-intensive farming. Younger household members may also seek non-farm employment to supplement their income.

However, previous studies yield conflicting results regarding the association between age and food security. Some suggest that as the household head ages, farming knowledge increases, risk aversion grows, and production diversifies, potentially leading to improved food security outcomes, as observed in a study in Mudzi District, Zimbabwe (Mango et al., 2014). Conversely, a study by Muhammad NA and Sidique indicated a slightly negative impact of age on food security (Muhammad & Sidique, 2019), suggesting that age may have limited or negative effects on household food security.

Education

Education plays a pivotal role in enhancing food security by influencing individuals' understanding of nutrition, food preparation techniques, and empowering them to make informed dietary choices. Research indicates that higher education levels are associated with improved food security, as those with more education tend to secure higher-paying jobs, resulting in greater spending power and better access to nutritious food. For example, a study in rural northern Pakistan (Abdullah et al., 2019) demonstrated a positive relationship between the educational level of the household head and food security. These findings emphasize the importance of education in addressing food security challenges and promoting better nutrition within households.

Gender

Gender dynamics of the household head are crucial in the context of food security. Research on this subject presents mixed findings. Ibnouf (2011) highlights the positive role of women in household food production, as they are often responsible for food preparation, processing, and preservation. In cases where male household members migrate, women may assume greater responsibility for ensuring food security (Ibnouf, 2011). Conversely, authors found that female-headed households may experience higher levels of food insecurity than male-headed households (Kharisma & Abe, 2020). Further investigation is needed to comprehend the specifics of this relationship in the study area's context.

Employment status

Employment status significantly affects household food security. Employment provides households with income, enabling them to access an adequate and nutritious food supply. Conversely, unemployment or underemployment raises the risk of food insecurity due to limited income and restricted access to food. Employment type, stability, and wage levels further influence the relationship between employment status and food security. In-depth research is necessary to fully understand the dynamics between employment and food security in specific contexts and among various populations. For example, Ahmed et al. (2017) found that creating non-agricultural employment opportunities can enhance local food security (Ahmed et al., 2017). This suggests that diversifying income sources beyond agriculture can positively impact access to sufficient and nutritious food within communities. Additionally, Ramsey et al. (2012) emphasized that low-income households are particularly vulnerable to food insecurity, highlighting the increased risk faced by these households due to limited access to affordable and nutritious food (Ramsey et al., 2012). These findings contribute to our understanding of the relationship between employment opportunities, income levels, and food security among different socioeconomic groups.

Land tenure

Land tenure status significantly influences household food security by impacting access to land and associated natural resources. Secure land ownership is crucial for ensuring a reliable food supply. It involves more than formal land rights registration; it includes farmers' legally recognized access and control over land, fostering trust and stability needed for long-term investments in production and land management (Khalid, 2016). Secure land ownership motivates farmers to make substantial capital and labor investments to boost agricultural productivity and sustainable land resource management. Farmers with secured land rights are inclined to adopt advanced technologies, implement soil conservation practices, and engage in sustainable land use, leading to improved agricultural production and enhanced household food security.

A study by Ajefu and Abiona (2020) underscores the connection between land tenure security and improved agricultural productivity, resulting in enhanced household food security. The researchers emphasize that households owning land are more likely to be food-secure and have greater dietary diversity than landless households, especially in drought-prone regions (Ajefu & Abiona, 2020). Conversely, Pritchard et al. (2019) conducted research suggesting that while land plays a crucial role in food security, this relationship is significantly influenced by livelihood strategies and seasonal circumstances. This implies that non-agricultural activities also play a crucial role in ensuring household food security (Pritchard et al., 2019).

Livelihood sources

Rural households that primarily rely on crop cultivation for income are among the most vulnerable groups regarding food security. Poverty is closely linked to food insecurity, reducing a household's purchasing power for essential goods and services like housing, energy, and water. Economic activities or sources of food for households, including agriculture, animal husbandry, wages/salaries, own businesses, fishing, pensions/transfers from family members, have been identified as factors influencing household food security (Mango et al., 2014). Factors such as access to markets, availability of credit, household support programs/remittances, and food aid also affect household food security status (Mango et al., 2014). Asghar and Muhammad (2013) emphasized that low household income plays a vital role in determining household nutritional status. Households with limited financial resources often struggle to purchase sufficient food during scarcity or high prices, increasing vulnerability to food insecurity (Asghar & Muhammad, 2013). Additionally, land ownership and access to credit facilities and support are necessary. Kassie et al. (2015) demonstrated that when combined with the adoption of advanced agricultural technologies, such as improved crop varieties and effective agronomic practices, agricultural production and productivity in rain-dependent households relying on agriculture for their livelihood can be enhanced. The study also revealed that improvements in agricultural technologies, coupled with the implementation of effective agronomic practices, can significantly reduce household food insecurity (Kassie et al., 2015).

Livestock ownership is particularly crucial for household food security, reflecting the socioeconomic status of farmers. Owning livestock provides various benefits, including the ability to sell livestock for income and access to food when local availability is limited. Moreover, animal husbandry yields products like milk, dairy, and meat for consumption or sale, diversifying nutritional offerings and generating income. Livestock farming also contributes to income generation while providing traction for agricultural activities, enabling households to achieve self-sufficiency, generate income, and meet their nutritional needs. Consequently, households with livestock tend to have higher food security compared to those without such assets (Bogale & Shimelis, 2009).

Result and discussion

The primary aim of this paper is to investigate the factors leading to insufficient and uncertain food access in Sudan's Red Sea state. We gathered cross-sectional data from 720 households during September to November 2020. These data were part of a larger study focusing on consumption patterns, nutrition, and various aspects of household food insecurity across Sudan. We utilized descriptive statistical techniques to assess the food security status in the Red Sea State. Additionally, we conducted logistic regression analysis to identify factors impacting household food security.

Table 3 summarizes key household characteristics. About 61% of the surveyed households resided in urban areas, reflecting significant urban representation. Nearly all participants (98%) were interviewed as heads of their households, recognizing their central role in household decision-making, particularly in financial matters. Demographically, there was noticeable gender inequality among respondents, with 95.4% being men and only 4.6% women. This skew towards male participation aligns with prevailing societal norms where men traditionally assume primary responsibilities for household well-being. However, exceptions arise in cases of unique circumstances like death or divorce, where women take on leadership roles. The participants in this study were generally older and responsible for providing food to their households. Age distribution showed 2% were under 20, 35.7% were between 21–40, 52.3% were between 41–60, 9.2% were between 61–80, and only 0.8% were over 81. On average, households had a relatively large size of 6–7 members (72.2%).

Table 3 Characteristics of the households
Full size table

Regarding education, respondents had varying levels of achievement, with 32.3% being illiterate, 13.8% receiving education in traditional Islamic schools (khalwa), 20.8% completing kindergarten, 21.1% completing primary school, and only a small fraction (10.8%) completing secondary education. Furthermore, just 1.2% held university degrees. Among the respondents, a significant proportion (36.9%) held steady, paid employment, providing them with economic stability. Another group included employers (20.4%) and the self-employed (30.3%) with successful businesses. It's crucial to note that even with economic stability, some individuals in these categories, such as farmers, fishermen, or pastoralists, could still be vulnerable due to the nature of their occupations. Other employment categories represented only a small percentage in the study area.

Source of livelihood

In the study area, it was observed that the majority of households resided in urban areas. As shown in Table 4, approximately 39.5% of respondents relied on wages and salaries as their main source of income. In contrast, a smaller percentage of rural households (27.5%) cultivate crops and livestock for food production, while others derive their income from owning businesses (6.4%) or pensions (3.5%). Some individuals relied on grants (1.1%), transfers from family members (2.6%), and other sources (19.6%) for financial support.

Table 4 Household food sources
Full size table

Analysis of land use in the study area revealed that a clear majority (70.4%) of households did not own or use any land, forest or pasture. In contrast, a smaller proportion (29.6%) reported owning such land. Among those who owned land, ownership status varied: 23.6% had full ownership rights. In addition, a small percentage (2.4%) rented their land, while an even smaller proportion (1.4%) owned fractional ownership of their land. In addition, a minority of respondents (2.3%) indicated that land was communal (Table 5).

Table 5 Land tenure status
Full size table

Food security status

The analysis of the food security status measured by the FCS shows that the conditions in the study area are extremely inadequate, as shown in Table 6. Only 23.8% of households demonstrated safe food intake, while 76.2% demonstrated unsafe consumption patterns. Furthermore, the breakdown of food consumption groups in Table 6 shows that 76.2% of households fall into the marginal category. Meanwhile, 19.1% are classified as acceptable consumption, only 4.2% have poor consumption. Taken together, these results highlight that the vast majority of households across the study area have inadequate and unstable access to food, indicating prevalent food insecurity.

Table 6 Food security status in the study area
Full size table

Logistic regression model result

The variability of the dependent variable, as justified by the model, can be measured using the pseudo-R-square or Nagelkerke's R-square (Field, 2013). In the model summary Table 7 of this study for the dependent variable FSS, it was found that the model explained between 0.124 (Cox & Snell R-squared) and 0.185 (Nagelkerke R-squared) of the variances in the dependent variable (Pallant, 2020). A higher Nagelkerke's R-square indicates a better fit of the model. In addition, Hosmer and Lemshow's model fit test indicates significance when it is not statistically significant. Overall, the classification table shows that about 78% of the cases were correctly classified by the model and the accuracy rate was very good. The model has good sensitivity as 98.6% of HHs who will be food insecure were correctly predicted to be food insecure based on the model (Table 8).

Table 7 FCGs generated by FCS
Full size table
Table 8 Classification table of the model*
Full size table

The omnibus test of the model coefficient for the dependent variable (FSS) indicates that the full model exhibits significant predictive power, with a chi-square value of 97.5, degrees of freedom (df) equal to 46, and a p-value below 0.05. This result leads to the rejection of the null hypothesis (p < 0.05). However, the addition of independent variables to the model did not significantly enhance its predictive capacity regarding respondents' decisions. Most variables did not make statistically significant contributions to the model, as their p-values exceeded 0.05. The exceptions were geographical location, age, gender, employment status, and source of livelihood. Geographical location was determined based on the study area, where the majority of households were situated in urban areas. Notably, geographical location had a statistically significant negative effect on FSS (odds ratio = 0.118, p = 0.017). This suggests that stable households in urban areas were more likely to experience food security compared to those residing in rural areas. Similar findings were reported by (Sultana & Kiani, 2011) in their study on household-level determinants of food security in Pakistan.

Recent observations indicate that food insecurity is increasingly affecting urban areas. However, development agencies have primarily focused on rural-oriented food security policies, as highlighted in the 2017 report published by UNCIF. This emphasis on rural food security has led to insufficient measurement and addressing of food security challenges faced by urban populations, contributing to the vulnerability of urban poor (Tuholske et al., 2020). These findings align with previous research in this area, emphasizing the association between place of residence and food security outcomes. Age is a significant factor in assessing household food security. Economic model analysis revealed that age had a statistically significant positive impact, with an odds ratio of 2.654 and a p-value of 0.001. This study also found significant differences in household food insecurity across different age groups, with households headed by individuals aged between 40 and 60 experiencing higher levels of food insecurity. These results align with previous research conducted by Abdullah et al. (2019), Mango et al. (2014) and Muhammad and Sidique (2019) highlighting the significant influence of age on household food insecurity. While some studies, such as the one by Bushara and Ibrahim (2017), suggest a positive impact of age on productivity in certain contexts, more research is needed to fully understand the complex relationship between age and productivity, particularly in different geographical and cultural settings in Sudan (Bushara & Ibrahim, 2017).

Gender played a significant role in determining food security, with a statistically significant negative association (p = 0.002) and an odds ratio of 1.877, indicating that male-headed households were more likely to be classified as food insecure than female-headed households. These results underscore the need for further investigation into how gender affects access to and availability of nutritious food, as well as differences in resource allocation and decision-making within households regarding food-related matters. The findings also align with the existing literature, highlighting the gender disparity in food security outcomes and emphasizing the importance of recognizing and supporting the often-overlooked contributions of women to household food security.

Employment status, referring to the occupation or activity of the household head, was statistically significant (p = 0.000) with a positive association (odds ratio = 3.145) with food security. This result supports previous research, such as the study by Loopstra and Tarasuk (2013), (Loopstra & Tarasuk, 2013), which found that higher income and better employment were linked to reduced household food insecurity. However, the relationship between female employment and food security depends on income levels, with poorer households benefiting more, as noted by Haddad (1992), (Haddad, 1992). Therefore, employment-based interventions have the potential to enhance food security in low-income households, but they may not eliminate food insecurity entirely for all households. Non-farm employment has also been shown to positively impact food security in rural areas, suggesting that diversifying employment opportunities could be a strategy to improve food security.

The study highlights the need to examine income segments and employment sectors that can most effectively improve access to food in the Red Sea state, considering local cultural and labor market dynamics, as well as the high prevalence of temporary and seasonal labor migration. A differentiated, local understanding of employment dynamics and their interaction with food security is crucial for developing effective policies. Finally, the study found a significant negative association between livelihood sources and food security (odds ratio = 0.101, p = 0.030). This underscores the importance of considering various livelihood strategies when addressing food insecurity in this context. The "source of livelihood" variable measures the contribution of different income sources (e.g., crop production, animal husbandry, wages, own businesses, pensions, and aid transfers from family members) to a household's total income (Fiedler & Mwangi, 2016). While this study highlights the value of diversified livelihood strategies in addressing food insecurity, the relationship is complex, as other studies suggest (Liu et al., 2018), (Sassi, 2021). Food security policies should consider local livelihood portfolios, vulnerabilities, and opportunities. Promoting livelihood diversification and supporting existing strategies are crucial considerations for addressing food insecurity comprehensively (Table 9).

Table 9 Logistic regression model results
Full size table

Conclusion

The aim of this study was to investigate the factors affecting the food security status of Red Sea state households in Sudan. A logistic regression model was used to assess the impact of different determinants on food security. The results showed that food security in the study region is generally good, with only 23% of households experiencing food insecurity. The most important factors positively associated with food security status were the head of household's geographic location, age, and employment status. In contrast, the gender of the head of household and dependence on limited sources of livelihood were negatively associated with food security. This study provides valuable insights into the socioeconomic and demographic variables affecting the food security of Sudanese households in the Red Sea state. The results could serve as a basis for policies and programs aimed at improving access and availability of food in this region. Further studies are needed to deepen the understanding of the complex relations between livelihood, gender, employment and the experience of food insecurity in Sudan. These results point to priority areas for policies and programs to improve access and availability of food. Promoting livelihood diversification, increasing agricultural production and expanding job opportunities could help vulnerable households improve their food security. Female-headed households could potentially benefit from targeted interventions. Although food insecurity does not appear to be pervasive at present, future risk reduction through livelihood support and social protection will be crucial. Based on this study, future research may deepen our understanding of the household-level dynamics affecting food security outcomes in Sudan. A deeper understanding could be gained by examining the interactions between socioeconomic status, access to food and livelihood strategies. Nonetheless, this study makes a significant contribution to our understanding of the underlying variables affecting household food security in the Red Sea state. Findings provide a robust evidence framework to support regional initiatives to increase food security and resilience.