Potential contribution of cereal and milk based fermented foods to dietary nutrient intake of 1-5 years old children in Central province in Zambia

Abstract

Zambia is still facing undernutrition and micronutrient deficiencies despite fortification and supplementation programmes stressing the need for additional solutions. Fermented foods have the potential to improve nutrient intake and, therefore, could have an important role in food based recommendations (FBRs) to ensure adequate intake of nutrients for optimal health of populations. Secondary dietary intake data was used in Optifood, a linear programming software to develop FBRs, for children aged 1–3 and 4–5 years in Mkushi district of Zambia. Three scenarios per age group were modeled to determine FBRs based on: (1) FBRs based on local available foods (2) FBR and Mabisi, a fermented milk beverage, and (3) FBR with Munkoyo, a cereal fermented beverage. The scenarios were compared to assess whether addition of Mabisi or Munkoyo achieved a better nutrient intake. FBRs based on only locally available non-fermented foods did not meet ≥70% of recommended nutrient intake (RNI) for calcium, fat, iron and zinc, so-called problem nutrients. The addition of Munkoyo to the FBRs did not reduce the number of problem nutrients, but after adding Mabisi to the FBR’s only iron (67% of RNI) in the 1–3 year age group and only zinc (67% of RNI) in the 4–5 year age group remained problem nutrients. Mabisi, a fermented milk product in combination with the local food pattern is a good additional source of nutrients for these age groups. However, additional nutrition sensitive and cost-effective measures would still be needed to improve nutrient intake, especially that of iron and zinc.

About the Authors:

Justin Chileshe

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

Affiliations Laboratory of Genetics, Plant Sciences Group, Wageningen University and Research, The Netherlands, Division of Human Nutrition and Health, Wageningen University and Research, The Netherlands, Tropical Diseases Research Centre, Ndola, Zambia

ORCID logo http://orcid.org/0000-0003-4134-2851

Elise F. Talsma

Roles Methodology, Supervision, Writing – review & editing

Affiliation: Division of Human Nutrition and Health, Wageningen University and Research, The Netherlands

Sijmen E. Schoustra

Roles Conceptualization, Funding acquisition, Investigation, Supervision, Writing – review & editing

Affiliations Laboratory of Genetics, Plant Sciences Group, Wageningen University and Research, The Netherlands, Department of Food Science and Nutrition, School of Agricultural Sciences, University of Zambia, Zambia

Karin J. Borgonjen-van den Berg

Roles Data curation, Formal analysis, Methodology, Validation

Affiliation: Division of Human Nutrition and Health, Wageningen University and Research, The Netherlands

Ray Handema

Roles Investigation, Project administration, Resources, Supervision, Writing – review & editing

Affiliation: Tropical Diseases Research Centre, Ndola, Zambia

Bas J. Zwaan

Roles Funding acquisition, Project administration, Supervision, Validation, Visualization, Writing – review & editing

Affiliation: Laboratory of Genetics, Plant Sciences Group, Wageningen University and Research, The Netherlands

Inge D. Brouwer

Roles Conceptualization, Data curation, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing

Affiliation: Division of Human Nutrition and Health, Wageningen University and Research, The Netherlands

Introduction

Undernutrition remains a severe public health problem in Zambia where 40% of the children under the age of five years are stunted, 15% have underweight and 6% are wasted [1]. Undernutrition in children leads to reduced growth, cognitive development impairment, greater susceptibility to infections and higher risk of mortality [2, 3]. Micronutrient deficiencies are also high in Zambia with 60% of the children between 6–59 months old suffering from anaemia which may result from iron deficiency among others and 54% of them affected by vitamin A deficiency [4, 5]. Different strategies and interventions exist to prevent or reduce undernutrition and micronutrient deficiencies. These include food-based strategies such as dietary diversification and food fortification, as well as nutrition education, public health and food safety measures, and supplementation [6]. Fortification and supplementation of micronutrients are seen as short and medium term solutions but are more expensive, and adherence might be low if intensive stimulation programs are not put in place [7]. Despite fortification of sugar with vitamin A and salt with iodine and supplementation implemented to reduce micronutrient deficiencies for more than two decades, Zambia is still facing undernutrition and micronutrient deficiencies stressing the need for additional solutions [8].

In Zambia there are a number of nutritionally dense foods locally available that can potentially be utilised to alleviate these nutrition problems [9]. The formulation, testing and implementation of food based recommendations (FBRs) can contribute to improve the nutrient intake, especially when these local nutrient dense foods are incorporated [10]. Such FBRs facilitate the formulation of Infant and Young Child Feeding (IYCF) practices being identified as one of the most effective public health interventions to improve young child survival in developing countries [11]. Especially in young children, the composition of diets and quality of foods consumed by populations have a direct impact on their health and wellbeing and thus FBRs are important in providing population-level and context specific guidance on consumption of appropriate foods to meet their nutritional needs [11].

Optifood is a linear programming software that allows formulation of FBRs based on the best combinations of local foods to optimize nutrient intake and model the extent to which these can supply nutritionally adequate diets [12–16]. In previous studies conducted with intake data from children of 6 months to 23 months old, gaps in intake of specific nutrients have been found especially that of calcium, iron and zinc [14–17]. Adding recommendations on increase of intake of animal source foods can potentially improve intake of calcium, iron and zinc but consumption of such foods in Zambia is low where diets are predominantly plant-based. In plant-based diets, preparation methods such as fermentation can be a method to make nutrients more bioavailable [18]. Phytates are present as storage compounds of phosphorous in cereals as complexes with metal cations such as iron, zinc and calcium. Phytates can be degraded by fermentation, in which microbial activity lowers the pH providing an optimum environment for enzymatic degradation by phytase leading to an increase in the cations making them bioavailable [19]. Fermentation is also beneficial in dairy products, as this process helps to convert lactose into more digestible components hence making milk more tolerable [20]. More generally, fermentation also makes micronutrients more bioavailable especially in plant based foods, gives food longer shelf life and contributes to a healthy ecology of intestinal bacteria which promotes general health, through fermenting bacteria (usually lactic acid bacteria) that serve as probiotics [18, 21–24]. Also fermentation of dairy products increases the amounts of micronutrients such as folate [25], vitamin B2 [26] and vitamin B12 [27] among others and are made more bioavailable.

Fermented foods traditionally are an important part of the diet in many countries and are now being advocated for inclusion in food based recommendations for regular consumption in some countries [28]. Zambia has a wide range of local fermented foods similar to western yoghurt, wine, and beer [29]. A number of traditional non-alcoholic fermented beverages are available such as Mabisi and Munkoyo which are consumed by all age groups [30]. Munkoyo is a fermented beverage made from maize porridge with Rhynchosia venulosa (Munkoyo) roots added whereas Mabisi is a fermented milk product made by allowing raw milk to ferment at ambient temperature in containers such as buckets [30, 31].

The prevailing levels of undernutrition among children in Zambia are severe especially stunting which is an indication of long term deficits of the quantity and quality of food. To combat undernutrition, there is urgent need to find possible solutions of improving nutrient intake. This study used linear programming to explore secondary dietary intake data of children, 1–5 years of age, in Mkushi, Zambia to assess the potential effect on nutrient adequacy of adding traditional fermented foods (Mabisi and Munkoyo) into FBRs. This is with a view to identify nutrient gaps and suggest food combinations the local diets can come as close to filling as possible with addition of fermented foods. This secondary dietary intake data was initially collected to determine maize intakes and vitamin A intakes with a view to introducing bio-fortified orange maize.

Methods

Study design

This study was based on cross-sectional dietary intake data collected previously with children in a rural, maize consuming population in Mkushi (Central Province, Zambia) [32]. Data was collected during harvest/early post-harvest season (May–June 2009) using the 24-hour recall method and was used to model weekly food based recommendations for children aged 1–5 years, with and without inclusion of Mabisi or Munkoyo. The harvest/early post-harvest season was chosen because the initial study was designed to capture data for the period of plenty and going into the lean period to determine the maize intakes and vitamin A intakes with a view to introducing bio-fortified orange maize.

The initial study was approved by the Tropical Diseases Research Centre (TDRC) Ethics review committee (Ndola, Zambia) reference number TDRC/ERC/0705/0409 and the International Food Policy Research Institute (Washington, DC, USA) Institutional review board [32]. All the data were fully anonymized before accessing them and the ethics committee waived the requirement for informed consent for the present study.

Subjects

The original study assessed dietary intake of 320 children in Mkushi [32] and included children aged 6–59 months, for whom parents gave written informed consent and were residing within the project catchment area were included into the study. Children were excluded based on the following criteria: being outside the age group 6–59 months; if parents did not give consent; children not residing in the catchment area as defined by the project; severely malnourished children (WAZ or WHZ <-3 SD, based on anthropometric measurements and WHO growth reference data); and children with severe anemia (Hb <7.0 g/dl).

From this data set we selected two groups of non-breastfed children aged 1–3 years (n = 156) and 4–5 years (n = 65) for Optifood programming, based on the different recommended nutrient intakes of these two age groups. Breastfed children were excluded because of a too small sample size (n = 15).

Dietary intake assessment

The dietary intake data was collected using the multi pass 24 hour recall as described elsewhere [33]. To collect the dietary intake data [32], the mothers of the children were asked to recall all foods and beverages consumed by their children, including amounts, during 24-hours of the previous day. Food portion sizes were estimated using examples of real foods, scaled photographs, or in volumes using standardized measuring spoons and cylinders carried by interviewers, and calibrated modelling clay. Information on feeding habits including plate sharing during meal times by children or family members, and on foods and quantities consumed outside of the home was collected. To convert portion sizes recorded in volumes to gram weight equivalents, a local conversion table was developed. Grams of ingredients consumed from the composite dishes was derived from the recipe data collected during the recall study or from standard recipes compiled prior to the initial study. Nutrient intake calculations were based on the food composition tables (FCT) developed by the initial study[32] [33].

Data preparation

Data from the 24 hour recalls was used to generate model parameters for Optifood using Excel and Access 2010 (Microsoft Corporation) [34]. Model parameters per age group were defined as follows: (I) A list of non-condiment foods and drinks consumed by ≥5% of the children during the recall period to ensure the commonly consumed foods are included; (II) The serving size of each food was defined as the medium serving size in grams per day for all children who consumed that particular food; (III) The minimum and maximum number of servings per week for each food group and food subgroup defined as the 10th and 90th percentiles of the serving counts respectively. The maximum and minimum number of servings per food within a food subgroup was estimated based on percentage of children who consumed the food [15, 16, 35, 36]. In the initial study only 2 children in the 4–5 age group were reported to have consumed Mabisi with the volume similar to that of tea consumed and thus the serving size/day for Mabisi in the two age groups were estimated based on the tea serving volume/day; (IV) Staple foods were identified as foods belonging to the food groups’ grain and grain products or starchy roots. Snacks were defined as foods consumed only in between meals. The type of meal (snack or staple) was determined based on the nature of the food and time of the food consumption.

Thirteen nutrients were selected for analysis of nutrient adequacy including total fat, total protein, calcium, vitamin C, thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, vitamin A as RAE, iron and zinc using the FAO/WHO RNIs [37].

Energy constraints were used to model FBR’s that ensured average energy requirement for the two age groups, by using mean body weight obtained from the initial study for each age group [38]. Dietary patterns in Zambia like in other developing countries are often rich in plant based foods and are high in Phytates, thus low bioavailability of iron (RNI: 11.6 mg/day assuming 5% bioavailability) and zinc (RNI: 8.3 mg/day assuming 15% bioavailability) were taken into account [39, 40].

The FCT used for this study included nutrient values for Mabisi (FCT as Sour milk). Nutrient values for Munkoyo were based on a composited recipe consisting of maize meal and water. The World Health Organization (WHO) estimated that the maximum iron and zinc bioavailability in maize based diets ranges between 10 and 15% [41]. Fermentation of maize based foods or beverages improves bioavailability of iron and zinc mainly by the breakdown of Phytates and is estimated to result in a 5% increase in bioavailability for iron, but not clear for zinc [42]. The values of iron and zinc for Munkoyo in the FCT were therefore increased by 5% to 10% and 20% respectively to take into account the increase in bioavailability through fermentation.

Analysis in Optifood

All analyses were carried out with Optifood program version 4.0.9.0 using a three module approach, based on linear programming to design population specific weekly FBRs. Per target group, three scenarios were modelled to develop FBR, namely: a) local foods without Munkoyo or Mabisi, b) local foods with Munkoyo included and c) local foods with Mabisi included. The data analysis was done by the lead author and checked by other authors as quality control at every stage from data preparation in Microsoft excel and in Microsoft access to importing the data into Optifood. Independent analysis was performed by a second person and the results were compared. The linear programming analyses used in this study have been described in detail by others [10, 15, 35, 36]. In summary, for each scenario in each target group data was checked by running module 1 (to set up model parameters) to ensure that model parameters were producing realistic diets with energy contents within a sufficient range to allow for modelling. An expert who was familiar with local dietary patterns then examined the foods selected in these 16 optimized 7-day diets to decide whether at least some individuals from the population could consume them.

Module 2 (to identify food based recommendations) was run for each of the three scenarios and for each age group to develop two optimized diets called the food pattern diet and no-food pattern diet. The food pattern diet is the best diet with minimized deviations from RNIs within the target population’s average food pattern. The no-food pattern diet is the best diet with minimized deviations from RNIs deviating from the average food pattern while remaining within lower and upper food group constraints. Module 2 was used to identify the nutrient dense food (sub) groups that were likely to improve nutrient adequacy and needed to be tested in module 3. Foods that contributed at least 5% to any of the nutrients were identified as nutrient dense foods. [16, 35, 36].

In Module 3 (to test alternative sets of FBRs to select the best dietary recommendations for the target population) diets were modelled for all three scenarios in both target groups, two 7-day diets per nutrient (i.e. in total 26 diets) were modelled of which 13 maximised (best-case scenario) and 13 minimized the nutrient content of the diet, for one nutrient (worst-case scenario), by preferentially selecting respectively the lowest and highest nutrient dense foods for that specific nutrient.

In step (i) of module 3, a no recommendation diet was run for each scenario to identify problem nutrients. Problem nutrients were defined as nutrients that were less than 100% of RNI in the best-case scenario (maximized diet) of module 3 diet modelled without FBR constraints [16, 35, 36].

In step (ii) of module 3, food groups with weekly servings above zero obtained in module 2 best food pattern diet, and nutrient dense foods and their accompanying food (sub) groups identified in module 2 were tested individually and were combined in step (iii) of module 3, the final FBRs per scenario in each target group was selected based on the combination of foods and food (sub) groups that covered 70% of the RNI in the worst case scenario for most nutrients, minimizing the deviation from the local food pattern as much as possible. The final FBRs were compared between the three scenarios. The outline of our approach of the modules for each scenario are presented in Fig 1.

[Figure omitted. See PDF.]

Fig 1. Graphical representation of Optifood analysis.

Each age group was done separately and for each age group three scenarios were done. For each scenario, the three modules were applied to determine the FBRs.

https://doi.org/10.1371/journal.pone.0232824.g001

Results

Data of 156 children in the 1–3 year age group with an average age of 2.8 years (SD = 8.1) and 65 children in the 4–5 year age group with an average age of 4.3 years (SD = 3.7) were used in this study. Girls represented 47% (n = 74) in 1–3 age group and 61% (n = 45) in 4–5 age group.

A total of 164 foods (with 28 foods consumed by ≥5% of children) were consumed by children in 1–3 year group and 115 foods (with 31 foods consumed by ≥5% of children) were consumed by children in the 4–5 year group. The most commonly consumed foods in both age groups included vegetable oil, onions, tomatoes, maize flour, and rape leaves as a pro-vitamin A rich source green leafy vegetable. Maize meal (medium serving per day was 242g for 1–3 year olds and 255g for 4–5 year olds) was consumed in high amounts compared to other foods. The smallest serving sizes were for white sugar (14g/day) for the 1–3 year olds and for brown sugar (20g/day) for the 4–5 year olds. The consumption frequencies of the foods varied from 0 to 14 times per week. An overview of all the foods, their corresponding serving sizes and consumption frequencies used for analysis is shown in Table 1 for each of the two age groups.

[Figure omitted. See PDF.]

Table 1. Foods and food (sub) groups consumed by children in the two age groups (1–3 years old and 4–5 years old) also describing the percentage of children who consumed a particular food, the serving size with the minimum and maximum number of servings per week.

https://doi.org/10.1371/journal.pone.0232824.t001

Linear programming

Diets produced in Module 1 (to set up model parameters) were feasible and no changes in model parameters were needed. The number of nutrients reaching ≥100% of RNI in the two best diets of the three scenarios for each age group as analysed in module 2 (to identify food based recommendations), are shown in Table 2. For 1–3 years old children, 8 nutrients for the average food pattern and 10 nutrients for best food pattern reached ≥100%, with no differences whether in- or excluding Mabisi or Munkoyo. For 4–5 years old children, the average food pattern diet covered >100% RNI for 5 nutrients whether or not including Mabisi, however, including Munkoyo increased number of nutrients covering ≥100% RNI to 6. For this age group, 11 nutrients reached ≥100% RNI for the best food pattern without differences with or without Mabisi or Munkoyo. Fat, calcium and zinc were identified as problem nutrients for this age group. In module 2, food groups that included grains & grain products and vegetables; food sub groups that included nuts/seeds & unsweetened products, cooked beans/lentils/peas and small whole fish with bones; and rape leaves as a single food were identified as nutrient-dense foods to be included to develop food based recommendations in module 3 (to test alternative sets of FBRs to select the best dietary recommendations for the target population).

[Figure omitted. See PDF.]

Table 2. Nutrient composition of the two optimized diets (Average food pattern and the best food pattern) for each of the three scenarios for the two age categories in module 2.

https://doi.org/10.1371/journal.pone.0232824.t002

In module 3 (i), fat, calcium, iron and zinc were identified as problem nutrients (reaching <100% RNI) in the best case scenario diet for the 1–3 year old group and fat, calcium and zinc for the 4–5 year old group. Testing food (sub) groups identified as nutrient dense in module 2, individually and combined (in module 3ii and iii), resulted in an increased coverage of the problem nutrients but none reached ≥70% RNI, and, hence, remained as problem nutrients (Table 3). Testing with inclusion of Munkoyo did not change these results (Table 4), however, inclusion of Mabisi resulted in fat, calcium and zinc disappearing as problem nutrient as their intake reached ≥70% RNI. Only iron remained as problem nutrient (Table 5).

[Figure omitted. See PDF.]

Table 3. Shows Food based recommendations for scenario 1 with only local foods for the 1–3 year old group.

https://doi.org/10.1371/journal.pone.0232824.t003

[Figure omitted. See PDF.]

Table 4. Shows food based recommendations for scenario 2 based on local foods with Munkoyo added for the 1–3 year old group.

https://doi.org/10.1371/journal.pone.0232824.t004

[Figure omitted. See PDF.]

Table 5. Shows Food based recommendations for scenario 3 based on local foods with Mabisi added for the 1–3 year old group.

https://doi.org/10.1371/journal.pone.0232824.t005

Fat, calcium and zinc were identified as problem nutrients (reaching <100% RNI) in the best-case scenario diet for the 4–5 year olds and, when testing food (sub) groups identified as nutrient dense in module 2, individually and combined (in module 3b and c), these nutrients remained as problem nutrients irrespective whether or not Munkoyo was included (Tables 6 and 7). However, including Mabisi showed improvement in calcium (≥70% RNI) with fat and zinc remaining below 70% RNI (Table 8).

[Figure omitted. See PDF.]

Table 6. Shows Food based recommendations for scenario 1 with only local foods for the 4–5 year old group.

https://doi.org/10.1371/journal.pone.0232824.t006

[Figure omitted. See PDF.]

Table 7. Show Food based recommendations for scenario 2 based on local foods with Munkoyo added for the 4–5 year old group.

https://doi.org/10.1371/journal.pone.0232824.t007

[Figure omitted. See PDF.]

Table 8. Shows Food based recommendations for scenario 3 based on local foods with Mabisi added for the 4–5 year old group.

https://doi.org/10.1371/journal.pone.0232824.t008

The final recommendations (with most nutrients reaching ≥70% RNI in the worst case scenario of the FBRs) for the 1-3year age group included the following serves per week: Vegetables 21 servings; Mabisi 7 servings; Grains and grain products 7 servings; Cooked beans, lentils and peas 3 servings; Nuts, seeds and unsweetened products 4 servings; and Rape leaves 7 servings; Small whole fish with bones 2 servings (Table 5).

The final recommendations (with most nutrients reaching ≥70% RNI in the worst case scenario of the FBRs) for 4–5 year age group included the following servings per week: Vegetables 21 servings; Mabisi 6 servings; grains and grain products 6 servings; Cooked beans, lentils and peas 3 servings; Nuts, seeds and unsweetened products 5 servings; and Rape leaves 7 servings; Small whole fish with bones 2 servings (Table 8).

Discussion

The current analysis revealed that nutrient intake among children aged 1–5 years in Zambia can be profoundly improved through carefully selected combinations of locally available foods but will not be able to cover the nutrient requirements for fat, calcium, iron and zinc. Only the inclusion of Mabisi (fermented milk) and not Munkoyo can sufficiently improve fat, calcium, iron (for 1–3 years olds only) and zinc (for 4–5 years olds only) intake for 1–5 year olds as indicated In Fig 2.

[Figure omitted. See PDF.]

Fig 2. Comparison of problem nutrients for the three scenarios (FBRs with local foods only, FBRs with Munkoyo added and FBRs with Mabisi added) against the worst case scenarios for the two age groups, 1–3 years old in graph (a) and 4–5 years old in graph (b).

Notes: The y-axis shows the % of RNI and the x-axis shows the problem nutrients. RNI is Recommended Nutrient Intake; FBRs is Food Based Recommendations. These are the nutrients that did not reach 70% RNI when diets were modelled for the 3 scenarios compared with the worst-case-scenario. The rest of the nutrients selected and modelled in Optifood are not depicted in this figure because they were all ≥70% RNI for all the scenarios. In the figure, the diet with Mabisi recorded the highest increase compared to the worst-case-scenario for the problem nutrients in both age groups).

https://doi.org/10.1371/journal.pone.0232824.g002

Results from our study indicate that among 1–5 year old children in Mkushi, the intake of fat, iron, zinc and calcium is below the requirements. The identified problem nutrients in our study are consistent with other findings in Zambia. Calcium shows the lowest achievable % of RNI in the scenario’s with and without Munkoyo added. In Mkushi, another study also found nutrient adequacy for children between 4 and 8 years old for most nutrients except for calcium [43]. Another study in Zambia found that children between 6–18 months had inadequate nutrient intakes for iron, zinc and calcium [44]. Low consumption of animal source foods could partly explain low intake of fat, calcium, iron and zinc nutrients in our population due to a substantial increase in production and consumption of cassava, maize and vegetable oils and a corresponding decrease in that of animal source foods observed in the past decades [8]. Inclusion of Mabisi in the food based dietary recommendations increased the potential of adequate intake of fat, calcium, zinc and iron in our study. A review on the impact of lipid-based nutrient supplements (LNS) plus complementary foods on health, nutrition and developmental outcomes among infants and young children [45] suggested that LNS plus complementary feeding compared to no intervention is effective at improving growth outcomes and anaemia among children aged 6 to 23 months in low- and middle-income countries (LMIC) in Asia and Africa, and more effective if provided over a longer duration of time (over 12 months). A trial in Malawi comparing the developmental outcomes of 18-month-old infants who received complementary feeding for 1 year either with lipid-based nutrient supplements or micronutrient fortified corn-soy porridge [46] found that the two types of interventions have comparable developmental outcomes by 18 months of age. A number of other trials conducted on yogurt (a fermented milk product with Lactobacillus bulgaricus and Streptococcus thermophilus) consumption have linked this yogurt intake to better health outcomes including weight management, type 2 diabetes, cardiovascular, disease risk, bone health, gastrointestinal (GI) health, malnutrition, immunological parameters and overall mortality [46–54]. In many of these studies microbial diversity appeared to increase in subjects consuming yogurt and the association between a greater microbial diversity and better health conditions has been attributed to yogurt consumption. Since Mabisi is a fermented milk product with the potential to contribute towards meeting reference intake of fat, calcium and zinc for 1–3 year aged children and fat, calcium and iron for 4–5 year aged children, regular consumption could contribute to the improvement in nutritional status, development functions, and gut function through fermenting bacteria.

Munkoyo (a cereal based fermented product) did not confer significant benefits in improving nutrient intake but the children may benefit from the fermenting microbes—that include high abundance of lactic acid bacteria—in this product. Chilton et al, in a review promoting inclusion of fermented foods in dietary guidelines established that the extensive use of and nutritional and health benefits derived from the fermented foods are evident enough for recommendation of regular consumption [28].

The feasibility and generalizability of identified food based dietary recommendations need further attention and it should be evaluated before implementation; for instance whether the required behaviour change is feasible [35]. Further, the foods included in the food based dietary guidelines should be available in sufficient amounts, although availability of foods does not always result in increased consumption as other aspects influence consumption, for example accessibility and affordability [55]. Cost of the locally available foods was not taken into account when modelling the different scenario diets. While Mabisi is generally considered a low-cost food, it is not known whether the final FBRs (including Mabisi) are affordable for the parents or caretakers of the children. Generally socio-economic factors influence food choices and nutrient dense foods are in general more expensive than low nutrient dense foods [56, 57]. Furthermore, Mabisi should be acceptable for consumption. From our dietary intake it appeared that Mabisi was consumed by only three children (one from the 1–3 year age group and 2 from the 4–5 year age group). However other studies about traditional fermented foods in Zambia [30] have shown that these foods are consumed on a regular basis in all age groups and are culturally accepted. Lastly, data from only one district in Zambia (Mkushi) was used. Although generally in Zambia similar dietary patterns are observed throughout the country with main foods being cereals, roots & tubers and vegetables [58], some foods in our FBRs may not be available in all areas of Zambia. Therefore use of data from a national representative sample would probably result in FBRs that are generalizable for the whole country.

There are some limitations in this study that are acknowledged. First, we acknowledge that recall bias in dietary intake data collection using 24 hour recall is inevitable. Precautions were taken through well trained interviewers, the use of a food frequency questionnaire (FFQ) to determine the frequency of specific food items, proper calibrations and the use of pictures and plastic bowls to better visualize and estimate the serving sizes. Often in recalls, nutrition dense foods that are irregularly consumed and snacks and fruits are often forgotten. Nevertheless, we could not avoid introducing errors in the data which could have affected our results, but we tried to keep it to a minimum. We also made sure to quality control the data analysis where the lead author carried out the analysis and the second person did the same analysis and the results were comparable.

Second, FBRs were developed for children between 1 and 5 years who are not breastfed due to very few children being breastfed after 1 year of age. The World Health Organization and UNICEF recommends initiation of breastfeeding within the first hour after the birth, exclusive breastfeeding for the first six months and continued breastfeeding for two years or more, together with safe, nutritionally adequate, age appropriate, responsive complementary feeding starting at the age of sixth months [41]. Part of the target group in this study falls in this particular age range. If breastfeeding were included in the modelling for the 1–2 years old, probably the effect of Mabisi would have been less as the breast milk would have provided for some of the problem nutrients. However, breastfeeding after 1 year was very low, accounting for less than 5% children, and was far lower than that of the national average of 92% and 42% of all children still breastfed at age 1 year and at 2 years respectively [1]. It may be that in this region of the country breastfeeding after age 1 year is low and that changing behaviour towards extending the breastfeeding period to 2 years or beyond may be challenging due to socio-cultural issues. We think that inclusion of Mabisi is more acceptable than extension of breastfeeding and that also breastmilk lacks the beneficial effects of fermentation, however it is acknowledged that Mabisi may have a higher safety risk.

Third, we used a combination of food composition tables (FCTs) to develop the one used in this study and this may have introduced errors due to nutrient variations introduced. We did not have the nutrient composition of Mabisi and used the composition of sour milk that is considered similar to Mabisi from the FCT. We do not expect any significant differences in nutrients between Mabisi and sour milk as they are similar products made in the same way but with different names in different regions of the country. It is clear that fermentation reduces phytates and increases bioavailability of minerals such iron and zinc in foods and that information on iron is available in literature but it is unclear for zinc [42]. We increased zinc values modestly by 5%, similar to that of iron in the FCT taking into account the reduction of phytate that has been shown before of 15–46% after maize porridge fermentation [59] and that the increase in iron bioavailability positively correlated with zinc amounts determined from in vitro studies [60]. We also took into account results from in vivo studies in rats that have also shown a higher zinc bioavailability when fed on diets containing fermented cassava than the unfermented cassava [61]. It is remarkable that dried small whole fish species in our FCT contained very high vitamin B12 values reaching ≥300% of the RNI for vitamin B12. These values were based on data published in the African Journal of Food, Agriculture, Nutrition and Development in 2010 and a report published by Nyirenda et al in 2007 [9, 62]. In these studies vitamin B12 values were similar to the FCT values used in this study for dried small whole fish species. It is known that dietary patterns in low and middle income countries are mostly plant based and as animal source products are rich sources of vitamin B12, it would be expected that deficiencies in vitamin B12 exist. In a study that assessed vitamin B12 status among Zambian children under the age of five years, high levels of vitamin B12 deficiencies were found (87%) [63]. The high values for vitamin B12 in fish in our FCT might have led to an overestimation of the dietary intake of vitamin B12 for the target group. To improve the data, a direct chemical analysis of the fish consumed would have given better estimates of vitamin B12 intake.

Conclusions

This study shows that FBRs for Zambian non-breastfed, 1–5 year old children based on only locally available foods into FBRs do not meet the required intakes for fat, calcium, iron and zinc and that the inclusion of Mabisi (and not Munkoyo) can have a major impact on the nutrient adequacy. Results in this study indicates the importance of the inclusion of Mabisi, a fermented milk product in the local food diet as a good additional source of nutrients for these age groups. However, to improve iron (for 1–3 years old) and zinc (for 4–5 years old) intake, alternative strategies should be found. The results from this analysis can serve as a guide for designing evidence based dietary recommendations under local conditions. Furthermore, additional data on for example cost of foods would still be needed to improve feasibility of developed FBRs.

Supporting information

[Figure omitted. See PDF.]

S1 Data.

https://doi.org/10.1371/journal.pone.0232824.s001

(CSV)

S2 Data.

https://doi.org/10.1371/journal.pone.0232824.s002

(XLSX)

Acknowledgments

We acknowledge the contributions of Dr Modest Mulenga, Executive Director of the Tropical Diseases Research Centre, Ward Siamusantu at the Zambian National Food and Nutrition Commission and Dr. Christine Hotz of the HarvestPlus, Washington DC, USA during the initial study. We also sincerely thank Marith Blom for her contributions to Optifood analysis.

Citation: Chileshe J, Talsma EF, Schoustra SE, Borgonjen-van den Berg KJ, Handema R, Zwaan BJ, et al. (2020) Potential contribution of cereal and milk based fermented foods to dietary nutrient intake of 1-5 years old children in Central province in Zambia. PLoS ONE 15(5): e0232824. https://doi.org/10.1371/journal.pone.0232824

References

1. Central Statistical Office/Zambia, Ministry of Health/Zambia, University of Zambia Teaching Hospital Virology Laboratory, University of Zambia Department of Population Studies, Tropical Diseases Research Centre/Zambia, ICF International: Zambia Demographic and Health Survey 2013–14. In. Rockville, Maryland, USA: Central Statistical Office/Zambia, Ministry of Health/Zambia, and ICF International; 2015.

2. Reinhardt K, Fanzo J: Addressing Chronic Malnutrition through Multi-Sectoral, Sustainable Approaches: A Review of the Causes and Consequences. Frontiers in Nutrition 2014, 1:13. pmid:25988116

3. Walker SP, Chang SM, Powell CA, Simonoff E, Grantham-McGregor SM: Early childhood stunting is associated with poor psychological functioning in late adolescence and effects are reduced by psychosocial stimulation. The Journal of nutrition 2007, 137(11):2464–2469. pmid:17951486

4. Zambia MoH: Zambia national malaria indicator survey 2015. In. Lusaka; 2015.

5. Food N, Nutrition Commission of Zambia UoZ, MOST: the USAID Micronutrient Program, United States Center for Disease Control: Report of the national survey to evaluate the impact of vitamin A interventions in Zambia in July and November 2003. In.: NFNC Lusaka; 2004.

6. Burchi F, Fanzo J, Frison E: The Role of Food and Nutrition System Approaches in Tackling Hidden Hunger. International Journal of Environmental Research and Public Health 2011, 8(2):358. pmid:21556191

7. Mildon A, Klaas N, O’Leary M, Yiannakis M: Can fortification be implemented in rural African communities where micronutrient deficiencies are greatest? Lessons from projects in Malawi, Tanzania, and Senegal. Food and Nutrition Bulletin 2015, 36(1):3–13. pmid:25898711

8. Zhang Z, Goldsmith PD, Winter-Nelson A: The Importance of Animal Source Foods for Nutrient Sufficiency in the Developing World: The Zambia Scenario. Food and Nutrition Bulletin 2016, 37(3):303–316. pmid:27150300

9. Nyirenda DB, Musukwa M, Mugode RH: The common Zambian foodstuff, ethnicity, preparation and nutrient composition of selected foods. Zambia: Child Health Project of the Ministry of Health, and Boston, USA: Boston University 2007.

10. Ferguson EL, Darmon N, Briend A, Premachandra IM: Food-based dietary guidelines can be developed and tested using linear programming analysis. The Journal of nutrition 2004, 134(4):951–957. pmid:15051853

11. Bhutta ZA, Das JK, Rizvi A, Gaffey MF, Walker N, Horton S, et al: Evidence-based interventions for improvement of maternal and child nutrition: what can be done and at what cost? The Lancet 2013, 382(9890):452–477.

12. Daelmans B, Ferguson E, Lutter CK, Singh N, Pachón H, Creed-Kanashiro H, et al: Designing appropriate complementary feeding recommendations: tools for programmatic action. Maternal & Child Nutrition 2013, 9(S2):116–130.

13. Levesque S, Delisle H, Agueh V: Contribution to the development of a food guide in Benin: linear programming for the optimization of local diets. Public health nutrition 2015, 18(04):622–631.

14. Santika O, Fahmida U, Ferguson EL: Development of Food-Based Complementary Feeding Recommendations for 9- to 11-Month-Old Peri-Urban Indonesian Infants Using Linear Programming. The Journal of nutrition 2009, 139(1):135–141. pmid:19056658

15. Skau JK, Bunthang T, Chamnan C, Wieringa FT, Dijkhuizen MA, Roos N, et al: The use of linear programming to determine whether a formulated complementary food product can ensure adequate nutrients for 6- to 11-month-old Cambodian infants. The American Journal of Clinical Nutrition 2014.

16. Vossenaar M, Knight FA, Tumilowicz A, Hotz C, Chege P, Ferguson EL: Context-specific complementary feeding recommendations developed using Optifood could improve the diets of breast-fed infants and young children from diverse livelihood groups in northern Kenya. Public health nutrition 2017, 20(6):971–983. pmid:27917743

17. Ferguson E, Chege P, Kimiywe J, Wiesmann D, Hotz CCMCNOAR: Zinc, iron and calcium are major limiting nutrients in the complementary diets of rural Kenyan children. Maternal & Child Nutrition 2015, 11:6–20.

18. Hotz C, Gibson RS: Traditional food-processing and preparation practices to enhance the bioavailability of micronutrients in plant-based diets. The Journal of nutrition 2007, 137(4):1097–1100. pmid:17374686

19. Gupta RK, Gangoliya SS, Singh NK: Reduction of phytic acid and enhancement of bioavailable micronutrients in food grains. Journal of food science and technology 2015, 52(2):676–684. pmid:25694676

20. de Vrese M, Stegelmann A, Richter B, Fenselau S, Laue C, Schrezenmeir J: Probiotics—compensation for lactase insufficiency. Am J Clin Nutr 2001, 73(2 Suppl):421s–429s. pmid:11157352

21. Egounlety M, Aworh OC, Akingbala JO, Houben JH, Nago MC: Nutritional and sensory evaluation of tempe-fortified maize-based weaning foods. Int J Food Sci Nutr 2002, 53(1):15–27. pmid:11820093

22. Kort R, Sybesma W: Probiotics for every body. Trends in biotechnology 2012, 30(12):613–615. pmid:23031355

23. Singh RK, Chang H-W, Yan D, Lee KM, Ucmak D, Wong K, et al: Influence of diet on the gut microbiome and implications for human health. Journal of Translational Medicine 2017, 15(1):73. pmid:28388917

24. Zimmermann MB, Chassard C, Rohner F, N’Goran E K, Nindjin C, Dostal A, et al: The effects of iron fortification on the gut microbiota in African children: a randomized controlled trial in Cote d’Ivoire. Am J Clin Nutr 2010, 92(6):1406–1415. pmid:20962160

25. Wouters JT, Ayad EH, Hugenholtz J, Smit G: Microbes from raw milk for fermented dairy products. International Dairy Journal 2002, 12(2–3):91–109.

26. LeBlanc J, Laiño JE, del Valle MJ, Vannini Vv, van Sinderen D, Taranto MP, et al: B-Group vitamin production by lactic acid bacteria–current knowledge and potential applications. Journal of Applied Microbiology 2011, 111(6):1297–1309. pmid:21933312

27. Zironi E, Gazzotti T, Barbarossa A, Farabegoli F, Serraino A, Pagliuca G: Determination of vitamin B12 in dairy products by ultra performance liquid chromatography-tandem mass spectrometry. Italian journal of food safety 2014, 3(4).

28. Chilton SN, Burton JP, Reid G: Inclusion of Fermented Foods in Food Guides around the World. Nutrients 2015, 7(1):390–404. pmid:25580813

29. Nguz K, Kasase C, Phiri C, Shindano J, Hikeezi D: Isolation and inhibitory effects of lactic acid bacteria from selected traditional fermented beverages from rural Zambia on some food pathogens. Journal of Science and Technology (Lusaka, Zambia) 2004(special ed.):9–16.

30. Schoustra SE, Kasase C, Toarta C, Kassen R, Poulain AJ: Microbial Community Structure of Three Traditional Zambian Fermented Products: Mabisi, Chibwantu and Munkoyo. PloS one 2013, 8(5):e63948. pmid:23691123

31. Moonga HB, Schoustra SE, Linnemann AR, Kuntashula E, Shindano J, Smid EJ: The art of mabisi production: A traditional fermented milk. PloS one 2019, 14(3):e0213541. pmid:30870441

32. Hotz C, Chileshe J, Siamusantu W, Palaniappan U, Kafwembe E: Vitamin A intake and infection are associated with plasma retinol among pre-school children in rural Zambia. Public Health Nutr 2012, 15(9):1688–1696. pmid:22443986

33. Hotz C, Chileshe, J., Siamusantu, W. and Palaniappan, U: Nutrition Survey in Central and Eastern Provinces, Zambia. In.; 2009.

34. Ferguson EL, Darmon N, Fahmida U, Fitriyanti S, Harper TB, Premachandra IM: Design of optimal food-based complementary feeding recommendations and identification of key “problem nutrients” using goal programming. The Journal of nutrition 2006, 136(9):2399–2404. pmid:16920861

35. Talsma EF, Borgonjen-van den Berg KJ, Melse-Boonstra A, Mayer EV, Verhoef H, Demir AY, et al: The potential contribution of yellow cassava to dietary nutrient adequacy of primary-school children in Eastern Kenya; the use of linear programming. Public health nutrition 2018, 21(2):365–376. pmid:28965533

36. Kujinga P, Borgonjen-van den Berg KJ, Superchi C, ten Hove HJ, Onyango EO, Andang’o P, et al: Combining food-based dietary recommendations using Optifood with zinc-fortified water potentially improves nutrient adequacy among 4-to 6-year-old children in Kisumu West district, Kenya. Maternal & Child Nutrition 2018, 14(2):e12515.

37. Joint FWU: Human energy requirements. Report of a Joint FAO/WHO/UNU Expert Consultation, Rome, 17–24 October 2001. 2004.

38. Group WHOMGRS: WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 2006, 450:76–85. pmid:16817681

39. Gibson RS, Perlas L, Hotz C: Improving the bioavailability of nutrients in plant foods at the household level. Proceedings of the Nutrition Society 2006, 65(02):160–168.

40. Hurrell RF, Reddy MB, Juillerat M-A, Cook JD: Degradation of phytic acid in cereal porridges improves iron absorption by human subjects. The American Journal of Clinical Nutrition 2003, 77(5):1213–1219. pmid:12716674

41. Staff WHO, Organization WH, UNICEF.: Global strategy for infant and young child feeding: World Health Organization; 2003.

42. Suri DJ, Tanumihardjo SA: Effects of Different Processing Methods on the Micronutrient and Phytochemical Contents of Maize: From A to Z. Comprehensive Reviews in Food Science and Food Safety 2016, 15(5):912–926.

43. Caswell BL, Talegawkar SA, Dyer B, Siamusantu W, Klemm RD, Palmer AC: Assessing child nutrient intakes using a tablet-based 24-hour recall tool in rural Zambia. Food and nutrition bulletin 2015, 36(4):467–480. pmid:26487637

44. Owino V, Amadi B, Sinkala M, Filteau S, Tomkins A: Complementary feeding practices and nutrient intake from habitual complementary foods of infants and children aged 6–18 months old in Lusaka, Zambia. African Journal of Food, Agriculture, Nutrition and Development 2008, 8(1):28–47.

45. Das JK, Salam RA, Hadi YB, Sheikh SS, Bhutta AZ, Prinzo ZW, et al: Preventive lipid-based nutrient supplements given with complementary foods to infants and young children 6 to 23 months of age for health, nutrition, and developmental outcomes. Cochrane Database of Systematic Reviews 2019(5).

46. Phuka JC, Gladstone M, Maleta K, Thakwalakwa C, Cheung YB, Briend A, et al: Developmental outcomes among 18-month-old Malawians after a year of complementary feeding with lipid-based nutrient supplements or corn-soy flour. Maternal & Child Nutrition 2012, 8(2):239–248.

47. O’Connor LM, Lentjes MA, Luben RN, Khaw K-T, Wareham NJ, Forouhi NG: Dietary dairy product intake and incident type 2 diabetes: a prospective study using dietary data from a 7-day food diary. Diabetologia 2014, 57(5):909–917. pmid:24510203

48. Sazawal S, Habib AA, Dhingra U, Dutta A, Dhingra P, Sarkar A, et al: Impact of micronutrient fortification of yoghurt on micronutrient status markers and growth–a randomized double blind controlled trial among school children in Bangladesh. BMC Public Health 2013, 13(1):514.

49. Buyuktuncer Z, Fisunoğlu M, Guven GS, Unal S, Besler HT: The cholesterol lowering efficacy of plant stanol ester yoghurt in a Turkish population: a double-blind, placebo-controlled trial. Lipids in health and disease 2013, 12(1):91.

50. Goldbohm RA, Chorus AM, Galindo Garre F, Schouten LJ, van den Brandt PA: Dairy consumption and 10-y total and cardiovascular mortality: a prospective cohort study in the Netherlands–. The American journal of clinical nutrition 2011, 93(3):615–627. pmid:21270377

51. Ballesta S, Velasco C, Borobio M, Argueelles F, Perea EJ: Fresh versus pasteurized yogurt: comparative study of the effects on microbiological and immunological parameters, and gastrointestinal comfort. Enfermedades infecciosas y microbiologia clinica 2008, 26(9):552–557. pmid:19100174

52. Pashapour N, Lou S: Evaluation of yogurt effect on acute diarrhea on 6–24 months old hospitalized infants. Turkish Journal of Pediatrics 2006, 48(2):115. pmid:16848109

53. Olivares M, Díaz-Ropero MP, Gómez N, Sierra S, Lara-Villoslada F, Martín R, et al: Dietary deprivation of fermented foods causes a fall in innate immune response. Lactic acid bacteria can counteract the immunological effect of this deprivation. Journal of dairy research 2006, 73(4):492–498. pmid:16987435

54. Heaney RP, Rafferty K, Dowell MS: Effect of yogurt on a urinary marker of bone resorption in postmenopausal women. Journal of the American Dietetic Association 2002, 102(11):1672–1674.

55. Darmon N, Drewnowski A: Contribution of food prices and diet cost to socioeconomic disparities in diet quality and health: a systematic review and analysis. Nutrition Reviews 2015, 73(10):643–660. pmid:26307238

56. Darmon N, Drewnowski A: Does social class predict diet quality? The American journal of clinical nutrition 2008, 87(5):1107–1117. pmid:18469226

57. Maillot M, Darmon N, Vieux F, Drewnowski A: Low energy density and high nutritional quality are each associated with higher diet costs in French adults. The American journal of clinical nutrition 2007, 86(3):690–696. pmid:17823434

58. Food, Organization A: Nutrition Country Profile: The Republic of Zambia. In.: Food and Agriculture Organization of the United Nations Rome, Italy; 2009.

59. Hotz C, Gibson RS: Assessment of Home-Based Processing Methods To Reduce the Phytate Content and Phytate/Zinc Molar Ratio of White Maize (Zea mays). Journal of Agricultural and Food Chemistry 2001, 49(2):692–698. pmid:11262014

60. Oikeh SO, Menkir A, Maziya-Dixon B, Welch R, Glahn RP: Assessment of Concentrations of Iron and Zinc and Bioavailable Iron in Grains of Early-Maturing Tropical Maize Varieties. Journal of Agricultural and Food Chemistry 2003, 51(12):3688–3694. pmid:12769546

61. Lazarte CE, Vargas M, Granfeldt Y: Zinc bioavailability in rats fed a plant-based diet: a study of fermentation and zinc supplementation. Food Nutr Res 2015, 59:27796–27796. pmid:26626410

62. Haug A, Christophersen O, Kinabo J, Kaunda W, Eik L: Use of dried Kapenta (Limnothrissa miodon and Stolothrissa tanganicae) and other products based on whole fish for complementing maize-based diets. African Journal of Food, Agriculture, Nutrition and Development 2010, 10(5).

63. Harris J, Haddad L, Seco Grütz S: Turning rapid growth into meaningful growth: Sustaining the commitment to nutrition in Zambia; 2014.