Abstract
Background
Child undernutrition remains a major public health challenge in Malawi, where inadequate dietary protein contributes to persistent stunting and poor child health outcomes. Soybean (Glycine max) is a locally cultivable and protein-rich crop with potential to strengthen complementary feeding in resource-constrained rural settings. However, the nutritional and practical effectiveness of soybean-based complementary foods depends on household-level processing methods and caregiver acceptability.
Methods
A community-based experimental study was conducted in Dowa District, Central Malawi, in July 2025. Four locally cultivated soybean varieties were subjected to five household-level processing methods: untreated (control), soaked, roasted, dehulled, and dehulled-and-roasted. Protein content was determined using the Kjeldahl method with triplicate laboratory analyses. Caregiver sensory evaluation was conducted among 60 caregivers attending under-five clinics using structured tasting sessions assessing aroma, flavour, texture, and overall acceptability on a 10-point Likert scale. Data were analyzed using Kruskal-Wallis H tests, Mann-Whitney U tests, repeated-measures ANOVA, and bootstrap resampling to estimate 95% confidence intervals.
Results
Measured protein content differed significantly across soybean processing methods (H = 12.68, df = 4, p = 0.013). Dehulled soybean flour demonstrated the highest measured protein content (21.26% ± 0.12; 95% CI: 21.12–21.40), while soaked soybean flour demonstrated the lowest (11.85% ± 0.11; 95% CI: 11.73–11.97). Caregiver sensory evaluation additionally differed significantly across processing methods (p < 0.05). Dehulled-and-roasted soybean porridge achieved the highest overall acceptability score (8.2 ± 0.2), followed by dehulled soybean porridge (7.8 ± 0.4). Integrated assessment demonstrated that dehulled and dehulled-and-roasted soybean processing methods produced the most favorable combined nutritional and caregiver acceptability outcomes.
Conclusion
Household soybean processing methods substantially influence both measured protein content and caregiver sensory acceptability of soybean porridge in rural complementary feeding contexts. Dehulling, with or without roasting, represents a locally feasible household processing strategy with potential to strengthen soybean-based complementary feeding using existing agricultural resources. The findings highlight the importance of integrating nutritional quality with caregiver acceptability in the development of sustainable community-based child nutrition interventions in low-resource settings.
Keywords: Soybean processing, complementary feeding, caregiver acceptability, child nutrition, community nutrition, household food environments, protein content, rural Malawi
Introduction
Child undernutrition remains a major public health challenge globally, particularly in low- and middle-income countries where inadequate complementary feeding, household food insecurity, and poverty continue to compromise child growth and development [1]. In sub-Saharan Africa, undernutrition contributes substantially to childhood morbidity, impaired cognitive development, weakened immunity, and preventable mortality [1, 2]. Malawi continues to experience a high burden of child undernutrition despite longstanding nutrition-sensitive interventions, with stunting remaining especially prevalent among children living in rural and food-insecure communities [2, 3]. Previous studies conducted in Central Malawi have similarly identified inadequate complementary feeding practices, limited dietary diversity, and constrained household food environments as important determinants of child undernutrition [4–6].
Complementary feeding represents a critical pathway through which household food systems influence child nutritional outcomes during the first two years of life [7–10]. However, complementary foods in many rural African settings remain heavily dependent on cereal-based porridges with limited protein diversity and inadequate nutrient density [11, 12]. In Malawi, maize porridge remains the dominant complementary food despite concerns regarding insufficient dietary protein quality and overall nutritional adequacy [6, 13]. Consequently, there is increasing interest in locally feasible food-based strategies capable of improving complementary feeding using accessible household resources.
Soybean (Glycine max) has emerged as an important candidate within community nutrition interventions because of its relatively high protein content, affordability, and adaptability to local agricultural systems [14–17]. Previous studies have demonstrated that soybean fortification may improve the nutritional quality of complementary foods in low-resource settings [17–19]. However, the nutritional and practical effectiveness of soybean-based complementary foods depends not only on nutrient composition, but also on household processing methods and caregiver acceptability.
Household soybean processing methods such as soaking, roasting, and dehulling may substantially alter protein concentration, digestibility, flavour, texture, and anti-nutritional compounds [12, 19, 20]. Although food science research demonstrates that processing techniques can improve nutrient quality and sensory properties [12, 19], existing studies in sub-Saharan Africa have largely prioritized laboratory nutrient analysis without adequately integrating caregiver sensory evaluation and household feasibility considerations [11, 18]. Yet caregivers remain central to complementary feeding because they determine how foods are prepared, accepted, and sustained within routine household feeding practices [8–10].
The present study addresses this gap by integrating laboratory-based protein analysis with caregiver sensory evaluation to examine how household soybean processing methods influence measured protein content and caregiver acceptability of soybean porridge in Central Malawi. Conceptually, the study situates soybean processing within broader household food environments and complementary feeding realities, recognizing caregivers as active decision-makers within resource-constrained food systems [21, 22]. Understanding how soybean processing methods influence both nutritional quality and caregiver acceptability may contribute to the development of more contextually grounded and sustainable complementary feeding interventions in rural Malawi and similar low-resource settings.
The study aimed to assess the influence of household soybean processing methods on measured protein content and caregiver sensory acceptability of soybean porridge in Dowa District, Central Malawi. Specifically, the study examined differences in measured protein content across soybean processing methods and evaluated caregiver sensory perceptions of soybean porridges prepared using different household processing techniques.
Theoretical and conceptual framework
This study is grounded in an integrated public health nutrition framework that draws on the UNICEF conceptual framework of child malnutrition, socio-ecological theory, household food environment scholarship, and strengths-based community nutrition practice [21–23]. Together, these perspectives provide a theoretically informed basis for understanding how household soybean processing practices may influence the nutritional quality and caregiver acceptability of soybean-based complementary foods within rural Malawian settings.
The UNICEF conceptual framework positions child undernutrition as the outcome of interacting immediate, underlying, and structural determinants, including inadequate dietary intake, household food insecurity, insufficient care practices, unhealthy living environments, and broader socioeconomic inequalities [23]. Within this framework, complementary feeding practices represent a critical pathway through which household food resources are translated into child dietary intake and nutritional wellbeing. In many rural communities in Malawi, complementary feeding remains heavily dependent on cereal-based porridges with limited protein diversity and inadequate nutrient density [6, 13]. The present study therefore situates soybean processing within this broader nutritional context by examining how locally available soybeans may be processed into nutritionally improved and potentially acceptable complementary foods.
The study is further informed by socio-ecological perspectives which recognize that child feeding practices are shaped by multiple interacting levels of influence, including caregiver knowledge, household food access, cultural food norms, agricultural production systems, and structural conditions such as poverty and food insecurity [21, 24–28]. Within rural household food environments, caregivers make feeding decisions under constrained but adaptive circumstances where affordability, preparation burden, fuel availability, sensory preferences, and cultural familiarity influence whether foods are consistently prepared and consumed [8–10]. Soybean processing is therefore conceptualized not simply as a technical food preparation activity, but as a household nutrition practice embedded within broader social, economic, and environmental realities.
Food science and complementary feeding scholarship additionally provide the biochemical and nutritional foundation for the framework. Household soybean processing methods such as soaking, roasting, and dehulling may alter protein concentration, digestibility, texture, flavour, and anti-nutritional compounds [12, 19, 20]. Previous studies have demonstrated that dehulling and roasting can improve nutrient density and sensory quality of soybean-based complementary foods [18, 19]. In the present study, soybean processing methods are conceptualized as modifiable household practices capable of influencing two proximal outcomes: measured protein content and caregiver sensory acceptability. Importantly, measured protein content is interpreted as an indicator of nutritional composition rather than a direct measure of child nutritional status or dietary intake.
The framework further draws on caregiver-centered and strengths-based community nutrition perspectives which emphasize that nutritional quality alone may not determine household adoption of complementary foods [8–10, 22]. Caregivers are central actors in complementary feeding because they determine how foods are selected, prepared, and incorporated into routine child feeding practices. Sensory attributes such as aroma, flavour, texture, and overall acceptability therefore represent important adoption-related dimensions of household nutrition interventions [11, 18]. A nutritionally improved soybean preparation may have limited practical value if caregivers perceive it as culturally unfamiliar, unpalatable, or difficult to prepare within everyday household conditions.
Conceptually, the study proposes that household soybean processing methods influence both measured protein content and caregiver sensory acceptability, which together shape the potential feasibility of soybean-based complementary feeding practices within rural household food environments. The framework does not assume that improved protein concentration alone will directly translate into improved child nutritional outcomes. Rather, it positions nutritional quality and caregiver acceptability as interrelated and necessary conditions for the successful integration of soybean into household complementary feeding practices. By integrating laboratory-based nutritional assessment with caregiver sensory evaluation, the study aligns food science evidence with community nutrition realities and contributes to more contextually grounded approaches to child nutrition intervention development in low-resource settings.
Methods
Study design and setting
This study employed a community-based experimental nutrition design integrating laboratory-based protein analysis with caregiver sensory evaluation to examine how household soybean processing methods influence the nutritional quality and acceptability of soybean-based complementary foods in rural Malawi. The study was informed by a translational public health nutrition perspective recognizing that complementary feeding practices are shaped not only by nutritional composition, but also by household food environments, caregiver decision-making, and the practical feasibility of food preparation within resource-constrained settings. Fig. 1.
Fig. 1.
Conceptual Framework: Household Soybean Processing, Nutritional Quality, Caregiver Acceptability and Complementary Feeding Practices
The study was specifically designed to move beyond isolated laboratory nutrient assessment by situating soybean processing within broader complementary feeding realities in rural communities. The integration of biochemical nutrient analysis with caregiver sensory evaluation therefore enabled simultaneous assessment of nutritional quality and household acceptability-related dimensions of soybean-based complementary foods. The study was conducted between June and July 2025 in rural communities within Traditional Authority (TA) Chiwere in Dowa District, Central Malawi (Fig. 2). Dowa is predominantly rural and agriculturally dependent district characterized by subsistence farming, recurrent household food insecurity, and persistent child undernutrition [5, 6]. Complementary feeding practices within the district remain heavily dependent on maize-based porridges with limited dietary protein diversity, particularly among low-income rural households.
Fig. 2.
Study setting showing Dowa District, Traditional Authority Chiwere, and participating outreach clinics in Central Malawi
The study was implemented within the catchment areas of Mvera Mission Hospital and surrounding outreach clinics, including Gogo, Ching’amba, Kalinyengo, Lavu, and Mphande. These outreach clinics serve predominantly rural populations with limited access to diversified complementary foods and nutrition-sensitive household resources. Dowa District therefore provided an analytically relevant setting for examining the interaction between locally available food resources, household processing practices, and complementary feeding feasibility within food-insecure rural environments.
The district has additionally been the focus of previous community nutrition and child health research examining household food insecurity, undernutrition, and rural health inequities [4–6]. The established community engagement within the study setting strengthened contextual relevance and supported implementation of the caregiver sensory evaluation component.
Soybean procurement and household processing procedures
Soybean samples were obtained from locally cultivated soybean varieties commonly grown and consumed within rural communities in Dowa District. Soybeans were sourced from farming households and local agricultural markets to ensure relevance to locally accessible household food resources. Prior to processing, soybean grains were cleaned to remove damaged seeds and foreign materials.
Five household-level soybean processing methods were evaluated based on their relevance to commonly practiced food preparation techniques within rural Malawian settings:
Untreated soybean flour (control).
Soaked soybean flour.
Roasted soybean flour.
Dehulled soybean flour.
Dehulled-and-roasted soybean flour.
For the soaking treatment, soybean grains were immersed in clean water for approximately 12 h at ambient temperature prior to drying. Roasting procedures were conducted using traditional household roasting techniques over moderate heat until light browning was achieved. Dehulling involved manual removal of soybean seed coats following soaking and drying procedures. Processed soybean samples were subsequently milled into flour using locally accessible milling facilities commonly utilized within rural communities.
The selected processing methods were conceptualized as modifiable household food preparation practices capable of influencing nutrient composition, sensory properties, and the practical feasibility of soybean incorporation into complementary feeding practices.
Laboratory determination of protein content
Measured protein content of soybean flour samples was determined using the Kjeldahl method following Association of Official Analytical Chemists (AOAC) procedures [29–32]. The Kjeldahl method was selected because of its established reliability and widespread application in food protein determination within nutrition and food science research.
Laboratory analyses were conducted in triplicate analytical replicates for each soybean processing treatment to improve analytical precision and assess measurement reliability. These triplicate measurements represented repeated laboratory analyses of the same processed samples rather than independent biological replicates.
Analytical quality assurance procedures were maintained throughout laboratory analysis to ensure consistency and reliability of measurements across all treatment groups.
Preparation of soybean porridge samples
Soybean and maize flour were combined using a standardized 1:3 ratio to reflect commonly practiced household complementary feeding preparations within rural Malawian communities while maintaining consistency across treatment groups. The selected ratio was informed by previous complementary feeding studies utilizing soybean-maize blends and was considered appropriate for producing porridges with acceptable consistency for sensory evaluation [31, 32].
Caregiver sensory evaluation
Sensory evaluation was conducted among 60 caregivers attending under-five outreach clinics within the study area. Participants were purposively recruited because of their direct involvement in complementary feeding and household food preparation practices for young children.
Sensory evaluation sessions were conducted in community clinic settings following preparation of fresh soybean porridges from each processing treatment. All participating caregivers evaluated each of the five soybean porridge samples. To minimize potential assessment bias, porridge samples were blind-coded using randomly assigned three-digit identifiers and presented in randomized order across participants. Caregivers independently evaluated each soybean porridge sample for aroma, flavour, texture (mouthfeel), and overall acceptability using a structured 10-point hedonic Likert scale ranging from 1 (“dislike extremely”) to 10 (“like extremely”).
Given varying educational backgrounds among participants, trained research assistants administered the sensory evaluation tool in Chichewa, the local language spoken within the study area. The Likert scale was explained verbally using culturally familiar descriptors and visual response aids prior to evaluation. Participants were provided with practice examples to ensure understanding of the scoring procedure before formal sensory assessment commenced.
The sensory evaluation component was incorporated to assess the practical household acceptability and potential adoption feasibility of soybean porridges within complementary feeding contexts. This approach reflects growing recognition within community nutrition scholarship that nutritionally improved foods are unlikely to achieve sustained household adoption unless they are simultaneously acceptable, culturally feasible, and practically adaptable within routine household feeding environments [8–10, 18]. Caregivers were provided with clean water between tasting sessions to minimize carryover sensory effects across soybean porridge samples.
Statistical analysis
Descriptive statistics were computed for measured protein content and caregiver sensory evaluation outcomes across soybean processing methods. Mean protein content, standard deviations, coefficients of variation, and bootstrapped 95% confidence intervals were calculated for laboratory outcomes.
Non-parametric analytical procedures were selected because of the distributional characteristics and relatively small analytical sample size associated with laboratory protein measurements. Kruskal-Wallis H tests were therefore used to assess differences in measured protein content across soybean processing methods. Pairwise Mann-Whitney U tests were subsequently conducted for post-hoc comparisons where appropriate [33].
Repeated-measures analysis of variance was used to assess differences in caregiver sensory evaluation scores across soybean porridge treatments [34]. Statistical significance was established at p < 0.05. All statistical analyses were conducted using SPSS version 29 (IBM Corp., Armonk, NY, USA).
Results
Measured protein content across household soybean processing methods
Measured protein content differed across the five household soybean processing methods evaluated (Table 1; Fig. 3). Dehulled soybean flour demonstrated the highest mean protein content (21.26% ± 0.12), whereas soaked soybean flour demonstrated the lowest mean protein content (11.85% ± 0.11). Intermediate protein values were observed for untreated control flour (17.47% ± 0.13), roasted flour (17.15% ± 0.14), and dehulled-and-roasted flour (16.92% ± 0.15).
Table 1.
Measured protein content of soybean flour across household processing methods
Values represent mean ± standard deviation from triplicate analytical measurements. Confidence intervals were estimated using bootsrap resampling (10,000 iterations). Different superscript letters indicate satisfically differences between treatments at p<0.05
Fig. 3.
Measured protein content of soybean flour across household processing methods
Analytical variability remained low across all treatments, with coefficients of variation ranging from 0.56% to 0.93%. Bootstrapped 95% confidence intervals demonstrated separation between the dehulled and soaked treatments (Table 1).
The Kruskal-Wallis H test demonstrated statistically significant differences in measured protein content across processing methods (H = 12.68, df = 4, p = 0.013). Pairwise Mann–Whitney U comparisons demonstrated significantly higher protein content in dehulled soybean flour compared with soaked soybean flour (p < 0.01). Figure 3 visually demonstrates the distribution of measured protein content across the five soybean processing methods, highlighting the comparatively higher protein concentration observed in dehulled soybean flour relative to the remaining treatments.
Caregiver sensory evaluation of soybean porridge
A total of 60 caregivers participated in the sensory evaluation component of the study. Sensory evaluation scores differed across soybean processing methods for aroma, flavour, texture, and overall acceptability (Table 2; Fig. 4).
Table 2.
Caregiver sensory evaluation scores for soybean porridge prepared using different household processing methods
Fig. 4.
Caregiver sensory evaluation profile of soybean porridge across processing methods
Dehulled-and-roasted soybean porridge demonstrated the highest overall acceptability score (8.2 ± 0.2), followed by dehulled soybean porridge (7.8 ± 0.4). Dehulled-and-roasted porridge additionally demonstrated the highest flavour score (8.1 ± 0.2), while dehulled soybean porridge demonstrated the highest texture score (7.8 ± 0.3).
Untreated soybean porridge demonstrated the lowest sensory scores across all evaluated attributes, including aroma (4.8 ± 0.9), flavour (4.2 ± 1.1), and overall acceptability (4.6 ± 1.0).
Repeated-measures analysis demonstrated statistically significant differences across soybean processing methods for all evaluated sensory attributes (p < 0.05).
Figure 4 demonstrates the multidimensional sensory evaluation profile across soybean processing methods, illustrating comparatively stronger sensory performance among dehulled and dehulled-and-roasted soybean porridges across aroma, flavour, texture, and overall acceptability domains.
Integrated nutritional and caregiver acceptability outcomes
Integrated assessment of measured protein content and caregiver overall acceptability demonstrated variation in nutritional and sensory outcomes across household soybean processing methods (Fig. 5).
Fig. 5.
Integrated assessment of measured protein content and caregiver overall acceptability across household soybean processing methods. Caregiver willingness to adopt soybean porridge across processing methods
Dehulled soybean flour demonstrated the highest measured protein content (21.26%), whereas dehulled-and-roasted soybean porridge demonstrated the highest overall caregiver acceptability score (8.2 ± 0.2). Untreated soybean porridge demonstrated comparatively lower overall acceptability despite intermediate protein content values.
Figure 5 integrates nutritional and sensory outcomes by jointly plotting measured protein content and caregiver acceptability scores across processing methods. Dehulled and dehulled-and-roasted soybean treatments demonstrated the most favorable combined distribution relative to the remaining processing methods.
The integrated distribution further demonstrated that soaked soybean flour was characterized by comparatively lower protein content despite moderate sensory acceptability, while untreated soybean flour demonstrated relatively lower caregiver acceptability despite intermediate protein content values.
Discussion
The present study examined the influence of household soybean processing methods on measured protein content and caregiver sensory acceptability of soybean porridge within rural complementary feeding contexts in Central Malawi. The findings demonstrated that household soybean processing substantially influenced both nutritional composition and caregiver sensory evaluation outcomes. Dehulled soybean flour demonstrated the highest measured protein content, while dehulled-and-roasted soybean porridge achieved the highest overall caregiver acceptability scores. Collectively, the findings suggest that dehulling, with or without roasting, may provide a favorable balance between nutritional quality and household acceptability for soybean-based complementary feeding interventions in low-resource rural settings.
The comparatively higher measured protein content observed in dehulled soybean flour is consistent with food science evidence demonstrating that removal of the fibrous seed coat concentrates cotyledon-associated nutrients and reduces fibre-related dilution effects in legumes [12, 19]. Soybean hulls contain substantially lower protein concentrations relative to cotyledon tissue, and dehulling therefore increases the relative proportion of protein-rich components within the processed flour [16, 19]. Similar findings have been reported in studies examining soybean and pulse processing technologies, where dehulling improved protein concentration, nutrient density, and digestibility in complementary food formulations intended for young children [16, 19, 35]. These findings reinforce the importance of simple household-level mechanical processing techniques in improving the nutritional potential of locally available plant-based complementary foods within resource-constrained settings.
In contrast, soaked soybean flour demonstrated the lowest measured protein content across all evaluated processing methods. This finding aligns with previous evidence suggesting that soaking may contribute to soluble nitrogen loss, nutrient leaching, and compositional dilution during water absorption and draining processes [16, 35]. Agume et al. observed reductions in soluble proteins following wet soybean processing and highlighted the influence of soaking duration on nutrient composition [16]. Similar reductions in measurable nutrient concentration following soaking have also been reported in legume-processing studies conducted in complementary feeding research [12, 35]. Although soaking is frequently promoted for reducing anti-nutritional compounds and improving digestibility, the present findings suggest that uncontrolled or prolonged soaking may simultaneously reduce measurable protein concentration within soybean flour preparations intended for household complementary feeding applications.
Importantly, the study demonstrates that nutritional composition alone may not adequately predict the feasibility of household adoption of soybean-based complementary foods. While dehulled soybean flour demonstrated the highest measured protein content, dehulled-and-roasted soybean porridge achieved the highest caregiver acceptability scores across flavour and overall acceptability domains. These findings are consistent with sensory evaluation and legume-processing literature demonstrating that roasting improves aroma development, flavour complexity, and palatability while reducing undesirable “beany” flavour characteristics associated with soybean products [1, 3]. Thermal processing has been shown to reduce lipoxygenase activity and volatile compounds associated with undesirable sensory properties in soybean-based foods, thereby improving consumer acceptability [1, 36]. The present findings therefore suggest that roasting may enhance the sensory feasibility of soybean incorporation within household complementary feeding practices.
The comparatively lower acceptability scores observed for untreated soybean porridge further reinforce the importance of sensory quality within complementary feeding interventions. Untreated soybean preparations are commonly associated with undesirable aroma, poor mouthfeel, and persistent beany flavour profiles that may limit household acceptance and sustained utilization [1, 3]. Previous studies conducted in sub-Saharan Africa similarly report that caregiver acceptance of legume-based complementary foods is strongly influenced by organoleptic characteristics, cultural familiarity, and preparation quality [11, 17, 24]. The present findings therefore support broader community nutrition scholarship emphasizing that nutritional interventions must integrate biochemical quality with sensory acceptability and household feasibility if sustained adoption is to occur [8–10].
From a public health nutrition perspective, these findings are particularly important within low-resource rural food environments where commercially fortified complementary foods remain inaccessible to many households due to financial and geographic barriers [13, 15, 29]. Soybean represents a relatively affordable and locally cultivable source of plant protein capable of strengthening household complementary feeding practices using existing agricultural systems [14, 16, 17]. However, the effectiveness of soybean-based interventions depends not only on nutrient composition, but also on whether caregivers are willing and able to prepare and repeatedly use soybean products within routine household feeding practices. The integrated findings presented in Fig. 5 therefore provide an important translational contribution by demonstrating the intersection between nutritional quality and caregiver acceptability within real-world household settings.
The study additionally contributes to community nutrition scholarship by demonstrating the importance of caregiver-centered approaches in complementary feeding research. Feeding practices are socially embedded behaviours shaped by household labour demands, food norms, caregiving responsibilities, cultural familiarity, and caregiver perceptions regarding food suitability for young children [8–10, 21]. Within socio-ecological and household food environment frameworks, caregivers function as active decision-makers navigating constrained but adaptive food systems rather than passive recipients of nutritional recommendations [21, 27, 28]. The comparatively high acceptability of dehulled-and-roasted soybean porridge suggests that processing methods capable of improving sensory characteristics may enhance the practical feasibility and sustainability of soybean-based nutrition interventions in rural communities.
The findings further align with strengths-based community nutrition approaches emphasizing locally available foods, household adaptive capacities, and culturally feasible preparation methods [22, 23]. Rather than relying exclusively on externally supplied fortified foods, community-based nutrition strategies that strengthen utilization of locally cultivable legumes may support greater sustainability, food sovereignty, and household-level resilience in low-resource settings [13, 29]. In Malawi and similar contexts where maize-based porridges remain dominant complementary foods, soybean incorporation through acceptable household processing techniques may represent a feasible strategy for improving dietary protein quality among young children.
The present study should nevertheless be interpreted alongside several limitations. First, the study assessed measured protein content rather than protein bioavailability, amino acid utilization, or child nutritional outcomes. Although protein concentration provides an important nutritional indicator, it does not directly measure digestibility, dietary intake, or growth effects among children. Second, Finally, anti-nutritional compounds such as phytates, tannins, and trypsin inhibitors were not directly quantified despite their potential influence on protein digestibility, nutrient bioavailability, and overall nutritional quality of soybean-based complementary foods [12, 19]. Third, sensory evaluation was conducted among caregivers rather than children, who are the ultimate beneficiaries of complementary feeding interventions. Although caregivers are primary decision-makers regarding food preparation and feeding practices, child sensory responses, food intake behaviours, and consumption patterns may differ from caregiver perceptions. Future studies should therefore incorporate direct child acceptability assessments and feeding observations to better understand the relationship between caregiver preferences and child consumption outcomes. Fourth, the study was conducted within a single rural district in Central Malawi, which may limit transferability to urban or culturally distinct settings.
Despite these limitations, the study possesses several important strengths. The integration of laboratory-based protein analysis with community-based caregiver sensory evaluation allowed simultaneous examination of nutritional composition and household adoption-related outcomes. The study additionally benefited from sustained community engagement within the study setting, enhancing contextual relevance and participant trust. Importantly, the study moves beyond narrow laboratory assessment by situating soybean processing within broader household food environments and community nutrition realities. This integrated approach strengthens the public health relevance of the findings and contributes to emerging scholarship emphasizing translational and contextually grounded nutrition interventions [13, 24, 29].
Future research should extend this work by examining protein digestibility, anti-nutritional compounds, dietary intake outcomes, and child growth indicators associated with soybean-based complementary feeding interventions. Longitudinal intervention studies evaluating sustained household adoption and child nutritional outcomes would further strengthen evidence regarding the effectiveness of soybean processing strategies within rural complementary feeding programs. Additional implementation-focused research examining caregiver preparation burden, fuel requirements, cost implications, and household feasibility may also strengthen understanding of practical scalability within low-resource settings.
Overall, the findings demonstrate that household soybean processing methods significantly influence both measured protein content and caregiver sensory acceptability. Dehulled and dehulled-and-roasted soybean processing methods demonstrated the most favorable combined nutritional and acceptability profiles within the study setting. These findings support the potential value of contextually grounded, household-level soybean processing approaches as part of community-based complementary feeding strategies aimed at strengthening child nutrition in resource-constrained rural environments.
Conclusion
This study demonstrated that household soybean processing methods significantly influence both measured protein content and caregiver sensory acceptability of soybean porridge within rural complementary feeding contexts in Central Malawi. Dehulled soybean flour demonstrated the highest measured protein content, while dehulled-and-roasted soybean porridge achieved the highest caregiver acceptability scores, indicating that household processing practices shape both nutritional quality and the practical feasibility of soybean-based complementary foods.
The findings further demonstrate that nutritional composition alone may not be sufficient to support sustained household adoption of complementary foods. Processing methods that improve sensory acceptability alongside nutritional quality may enhance the integration of soybean into routine child feeding practices in resource-constrained settings.
From a public health nutrition perspective, dehulling, with or without roasting, represents a locally feasible household processing strategy with potential to strengthen complementary feeding using existing agricultural resources. The study therefore contributes to community nutrition scholarship by integrating food science evidence with caregiver-centered acceptability assessment within real-world household food environments.
Future research should examine protein digestibility, anti-nutritional compounds, child dietary intake, and nutritional outcomes associated with soybean-based complementary feeding interventions in low-resource rural settings.
Acknowledgements
The author gratefully acknowledges the invaluable contributions of the Health Surveillance Assistants who served as research assistants and played a critical role in data collection for this study. We also extend our sincere appreciation to the social work and biology students whose technical and logistical support was essential for the successful implementation of the research.We are further grateful to the community elders and community promoters for facilitating engagement, coordination, and local participation, and to all caregivers and participants who generously contributed their time and experiences. Their involvement was fundamental to the success of this study.
Abbreviations
- AOAC
Association of Official Analytical Chemists
- ANOVA
Analysis of Variance
- CI
Confidence Interval
- CV
Coefficient of Variation
- HCl
Hydrochloric Acid
- H₂SO₄
Sulfuric Acid
- ISO
International Organization for Standardization
- NaOH
Sodium Hydroxide
- SD
Standard Deviation
- TA
Traditional Authority
- UNICEF
United Nations Children’s Fund
- w/v
Weight per Volume
Authors’ contributions
Emmanuel Chilanga (EC) designed the study, collected the data, and conducted the analyses. EC drafted the initial manuscript and critically revised the manuscript for intellectual content.
Funding
The author received no specific funding for this study.
Data availability
Soybean flour sample data is available upon request from the author. The sensory evaluation data can be shared in deidentified form.
Declarations
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and received ethical approval from the University of Livingstonia Research Ethics Committee, Malawi. Written informed consent was obtained from all participating caregivers prior to participation in the sensory evaluation sessions. Participants were informed about the study objectives, voluntary nature of participation, confidentiality procedures, and their right to withdraw from the study at any stage without consequence.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Soliman N, Soliman A, Alyafei F, et al. Persistent global burden of stunting among children. Eur J Med Health Sci. 2024;6(2):15–20. 10.24018/ejmed.2024.6.2.2080. [Google Scholar]
- 2.Kodish SR, Farhikhtah A, Mlambo T, Hambayi MN, Jones V, Aburto NJ. Leveraging the Scaling Up Nutrition Movement to operationalize stunting prevention activities: implementation lessons from rural Malawi. Food Nutr Bull. 2022;43(1):104–20. 10.1177/03795721211046140. [DOI] [PubMed] [Google Scholar]
- 3.Chikhungu LC. Trends and patterns of stunted-only and stunted-underweight children in Malawi: a confirmation for child nutrition practitioners to continue focusing on stunting. Malawi Med J. 2022;34(2):111–7. 10.4314/mmj.v34i2.6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chilanga E, Collin-Vézina D, Khan MN, Riley L. Prevalence and determinants of intimate partner violence against mothers of children under-five years in Central Malawi. BMC Public Health. 2020;20(1):1848. 10.1186/s12889-020-09910-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chilanga E, Collin-Vézina D, MacIntosh H, Mitchell C, Cherney K. Prevalence and determinants of malaria infection among children of local farmers in Central Malawi. Malar J. 2020;19(1):308. 10.1186/s12936-020-03382-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chilanga E, Chilanga M. Predisposing and reinforcing factors of undernutrition among 0- to 59-months-old children in rural communities of Central Malawi. Soc Sci Humanit Open. 2023;7(1):100454. 10.1016/j.ssaho.2023.100454. [Google Scholar]
- 7.Martin SL, Matare CR, Kayanda RA, et al. Engaging fathers to improve complementary feeding is acceptable and feasible in the Lake Zone, Tanzania. Matern Child Nutr. 2021;17(S1):e13144. 10.1111/mcn.13144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Schratz LM, Larkin O, Dos Santos N, Martin C. Caregiver influences on eating behaviors in children: an opportunity for preventing obesity. Curr Atheroscler Rep. 2023;25(12):1035–45. 10.1007/s11883-023-01171-6. [DOI] [PubMed] [Google Scholar]
- 9.Stelle I, Kinshella MLW, Moore SE. Caregiver perceptions of nutrition interventions in infants and children under 24 months of age: a systematic review. Public Health Nutr. 2023;26(9):1907–16. 10.1017/S1368980023001246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang J, Chang YS, Wei X, Cao Y, Winkley K. The effectiveness of interventions on changing caregivers’ feeding practices with preschool children: a systematic review and meta-analysis. Obes Rev. 2024;25(4):e13688. 10.1111/obr.13688. [DOI] [PubMed] [Google Scholar]
- 11.Løvdal T, Skaret J, Drobac G, et al. Canned complementary porridges for infants and young children (6–23 months) based on African indigenous crops: nutritional content, consistency, sensory, and affordability compared to traditional porridges based on maize and finger millet. Matern Child Nutr. 2025;21(1):e13752. 10.1111/mcn.13752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Oladiran DA, Emmambux NM. Locally available African complementary foods: nutritional limitations and processing technologies to improve nutritional quality-a review. Food Rev Int. 2022;38(5):1033–63. 10.1080/87559129.2020.1762640. [Google Scholar]
- 13.Kerr RB, Chilanga E, Nyantakyi-Frimpong H, Luginaah I, Lupafya E. Integrated agriculture programs to address malnutrition in northern Malawi. BMC Public Health. 2016;16(1):1197. 10.1186/s12889-016-3840-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ndovie P, Nkhata SG, Geresomo N, et al. Nutritional, functional, and microbial qualities of legume-based flour blends processed by SMEs in Zambia and Malawi compared to standard Corn-Soy Blend Plus (CSB+): a cross-sectional study. BMC Nutr. 2025;11(1):49. 10.1186/s40795-025-01034-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ndovie P, Geresomo NC, Nkhata SG, Fungo R, Nyau V, Munthali J. Exploring the impact of legume consumption on undernutrition in rural Malawian children aged 6–59 months old: a community-based cross-sectional study. J Nutr Sci. 2026;15:e10. 10.1017/jns.2025.10073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gulkirpik E, Donnelly A, Nowakunda K, Liu K, Andrade Laborde JE. Evaluation of a low-resource soy protein production method and its products. Front Nutr. 2023;10:1067621. 10.3389/fnut.2023.1067621. [DOI] [PMC free article] [PubMed]
- 17.Ngaha Damndja W, Ngangoum ES, Saidou C, Mohamadou S. Formulation of three infant foods from plantain flour fortified with sesame (Sesamum indicum), soybean (Glycine max) and cashew nut (Anacardium occidentale L). Food Chem Adv. 2023;3:100313. 10.1016/j.focha.2023.100313. [Google Scholar]
- 18.Daagema AA, Ajayi K, Talabi JY, Dada IO, Chipili G, Adewuyi OO. Quality evaluation and acceptability of improved ‘Luam-Nahan’ porridge flour developed from maize, cassava, soybean, and Jatropha tanjorensis for preschool children. J Nutr Food Secur. 2025;10(3). 10.18502/jnfs.v10i3.19234.
- 19.Muoki PN, de Kock HL, Emmambux MN. Effect of soy flour addition and heat-processing method on nutritional quality and consumer acceptability of cassava complementary porridges. J Sci Food Agric. 2012;92(8):1771–9. 10.1002/jsfa.5545. [DOI] [PubMed] [Google Scholar]
- 20.Makinde FM, Abolarin OO. Effect of post-dehulling treatments on anti-nutritional and functional properties of cowpea (Vigna unguiculata) flour. J Appl Sci Environ Manag. 2020;24(9):1641–7. 10.4314/jasem.v24i9.23. [Google Scholar]
- 21.Dulmus CN, Sowers KM. The profession of social work: guided by history, led by evidence. Hoboken: Wiley; 2012. [Google Scholar]
- 22.Saleebey D. The strengths perspective in social work practice. 6th ed. Boston: Pearson; 2012. [PubMed] [Google Scholar]
- 23.UNICEF. Strategy for improved nutrition of children and women in developing countries. New York: UNICEF; 1990. [DOI] [PubMed] [Google Scholar]
- 24.Boseley S. Michael Marmot: leader in the social determinants of health. Lancet. 2008;372(9650):1625. 10.1016/S0140-6736(08)61675-X. [DOI] [PubMed] [Google Scholar]
- 25.Bronfenbrenner M. A conversation with Martin Bronfenbrenner. East Econ J. 1987;13(1):1–6. [Google Scholar]
- 26.Food insecurity in Malawi. regional disparities, sociocultural determinants, and the role of mobile advisory services. Food Humanity. 2025;5:100695. 10.1016/j.foohum.2025.100695. [Google Scholar]
- 27.Tembo D, Manja LP, Joshua MK. Faith and food in rural Malawi: an investigation into the sociocultural determinants of household food security. Food Cult Soc. 2025;28(4):1055–83. 10.1080/15528014.2024.2407195. [Google Scholar]
- 28.Chiutsi-Phiri G, Kalimbira AA, Banda L, et al. Preparing for a community-based agriculture-to-nutrition trial in rural Malawi: formative research to assess feasibility and inform design and implementation decisions. Pilot Feasibility Stud. 2021;7(1):141. 10.1186/s40814-021-00877-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Beljkaš B, Matić J, Milovanović I, Jovanov P, Mišan A, Šarić L. Rapid method for determination of protein content in cereals and oilseeds: validation, measurement uncertainty and comparison with the Kjeldahl method. Accred Qual Assur. 2010;15(10):555–61. 10.1007/s00769-010-0677-6. [Google Scholar]
- 30.Wang H, Pampati N, McCormick WM, Bhattacharyya L. Protein nitrogen determination by Kjeldahl digestion and ion chromatography. J Pharm Sci. 2016;105(6):1851–7. 10.1016/j.xphs.2016.03.039. [DOI] [PubMed] [Google Scholar]
- 31.Alamu EO, Maziya-Dixon BB, Popoola I, Gondwe TNP, Chikoye D. Nutritional evaluation and consumer preference of legume fortified maize-meal porridge. Ibadan: International Institute of Tropical Agriculture; 2016. [Google Scholar]
- 32.Nnam NM, Baiyeri GT. Evaluation of the nutrient and sensory properties of multimixes and porridges made from maize, soybean, and plantain for use as complementary food. Ecol Food Nutr. 2008;47(1):64–76. 10.1080/03670240701615374. [Google Scholar]
- 33.Chicco D, Sichenze A, Jurman G. A simple guide to the use of Student’s t-test, Mann-Whitney U test, Chi-squared test, and Kruskal-Wallis test in biostatistics. BioData Min. 2025;18(1):56. 10.1186/s13040-025-00465-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Van Den Bergh D, Wagenmakers EJ, Aust F. Bayesian repeated-measures analysis of variance: An updated methodology implemented in JASP. Adv Methods Practices Psychol Sci. 2023;6(2):25152459231168024. [Google Scholar]
- 35.Gelli A, Margolies A, Santacroce M, et al. Improving child nutrition and development through community-based childcare centres in Malawi – the NEEP-IE study: study protocol for a randomised controlled trial. Trials. 2017;18(1):284. 10.1186/s13063-017-2003-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chikafa M, Nejadhashemi AP, Moller K, Razavi H, Bizimana JC. Multidimensional evaluation of the impacts of agricultural interventions to achieve food security in Malawi. Food Energy Secur. 2023;12(4):e486. 10.1002/fes3.486. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Soybean flour sample data is available upon request from the author. The sensory evaluation data can be shared in deidentified form.