1 Introduction

The availability of sufficient quantities of essential micro and macronutrients is key to supporting people’s health. While global food production has the potential to supply sufficient quantities of food to meet nutrient requirements, access has not been equally met, causing malnourishment and famine, despite over 40 years of international efforts to address these (Chappell & LaValle, 2011; FAO, 1974). Inadequate access to nutritious food is most persistent across Sub-Saharan Africa (Muringai et al., 2022; Slater et al., 2022; White et al., 2021), resulting in a high prevalence of stunting, and other forms of malnutrition (Béné et al., 2010; Byrd et al., 2021a, b; FAO, 2021; Ruel & Alderman, 2013). During 2020–2022, 72% of the total population in Kenya experienced moderate or severe food insecurity with 28% experiencing severe food insecurity (FAO et al., 2023). The vulnerability to food insecurity in this region has been attributed to various factors including domestic challenges such as low socio-economic status, low investment in irrigated agriculture and research, and climate change effects (FAO et al., 2023; Wudil et al., 2022) that have exacerbated the situation, with up to 74% (39 million people) of Kenyans unable to afford a healthy diet (FAO et al., 2023; Herforth et al., 2020).

Food and nutrition security requires the maintenance of sufficient yields and a stable supply of nutritionally diverse foods (Bernhardt & O’Connor, 2021). Therefore, policy reforms for food and nutrition security improvement must have clearly defined nutrition attainment commitments. The government of Kenya has committed to ensuring that all citizens are free from hunger and have adequate food of acceptable quality as stipulated in The Constitution (2010) and the Food and Nutrition Security Policy (FNSP) (Ministry of Agriculture, 2011). The country has committed as a signatory to several nutrition-related global agreements and mechanisms including the United Nations (UN) Decade of Action on Nutrition (2016–2025) (United Nations System Standing Committee on Nutrition, 2016), the Scaling Up Nutrition (SUN) movement, the World Health Assembly (WHA) 2025 nutrition targets (WHO, 2024), the Sustainable Development Goals (SDGs), and the Second International Conference on Nutrition (ICN2) Declaration and Plan of Action. These agreements have defined the foundation for addressing the immediate, underlying, and basic causes of malnutrition including expanding the political, economic, social, and technological space for nutrition actions (Nutrition International, 2021). All these commitments require empirical research and greater sharing of information among all sectors that are key to ensuring the availability and accessibility of nutritious food in the nation.

While fish has been recognized as a nutrient-dense food, and evidence of the positive impact of fish on human health at a global scale is prevalent in the scientific literature (Crona et al., 2023; Golden et al., 2021; Thilsted et al., 2016), its contribution to national food and nutrition security remains uncertain in most countries (Chan et al., 2019). As a result, there is a paucity of fisheries-related scientific information available to those developing management strategies and food security policy directions, particularly in tropical developing countries, thus marginalizing the role that fisheries and aquaculture can and should play in national food security and nutrition policies (FAO, 2020; FAO et al., 2022). This is in part due to limited data on the nutrient composition of fish species (Byrd et al., 2021a, b; Nordhagen et al., 2020). To help address this limitation, the ability to estimate nutrient concentrations of different species of fish has recently been advanced (Hicks et al., 2019; MacNeil, 2021), and this information has been made publicly available through the FishNutrients tool in FishBase (Froese & Pauly, 2021). Using this information, many developing regions and emerging economies can gain empirical evidence to guide plans to use their fisheries resources to help address nutrient deficiencies.

Here we quantify nutrient content and nutrient supply for marine and freshwater capture, aquaculture, imported and exported fish in Kenya, to determine the contribution fish can make to food and nutritional security. The objectives of the study were to: (1) Establish trends in the supply of fish over 16 years (2005–2020) in Kenya; (2) Assess the nutrient content of current (2020 estimates) fish supply; and (3) Evaluate over time how the fish supply could contribute to nutrition security. We used Kenya as a case study due to the high prevalence of food and nutrition insecurity (FAO et al., 2023), and its binding commitments to address this. Further, the importance of fish resources in Kenya has not been comprehensively incorporated into food security policy and strategies. We focus on children under five years of age, for whom micronutrient deficiencies remain a public health concern, with inadequate intakes of calcium, iron, zinc, vitamin A, and animal-sourced protein (Byrd et al., 2018; Codjia et al., 2024; Kamau-Mbuthia et al., 2023; Ryckman et al., 2022; Stewart et al., 2019) resulting in nearly one-third of children aged 6–59 months being anemic, 9.2% with Vitamin A Deficiencies (VAD), and the prevalence of zinc deficiency at 83.3% (Kenya Ministry of Health, 2011).

2 Materials and methods

2.1 Study sites

Kenya, in East Africa, lying between latitudes 4.5ºN and 4.5ºS and longitudes 34ºE and 41ºE, is currently classified as a lower-middle-income country based on its per capita gross national income (GNI) of US$2,080 for 2021 (United Nations, 2021). Kenya’s population of over 50 million is growing rapidly at 2.3% (United Nations Department of Economic and Social Affairs, Population Division, 2019) with 13% of the population being children below five years of age.

Kenya’s supply of fish comes from marine and freshwater wild capture fisheries, aquaculture, and imports. Kenya’s freshwater wild captured fish comes from Lake Turkana, Lake Victoria, several small inland freshwater lakes, dams, and rivers (Schubert et al., 2021). The marine capture fisheries comes from the 9,700 km2 of territorial seas and 142,400 km2 of exclusive economic zone (EEZ) (Kimani et al., 2018). Kenya’s domestic aquaculture supplies come from an emerging aquaculture sector involving extensive, semi-intensive, and intensive production (Munguti et al., 2021). Kenya is also focused on meeting national demand for fish by increasing imports of mackerel and tilapia, largely from China, but also from Japan, Oman, Korea, Somalia, and Djibouti (Fish Inspection and Quality Assurance Report, 2020). Some imports also come from the neighboring East African countries of Tanzania and Uganda (Ibengwe et al., 2022).

2.2 Fish production data

We gathered fish production data (landings, aquaculture, imports, and exports) from the national fisheries’ statistical bulletins, Kenya National Economic Survey Reports, and the Fish Inspection and Quality Assurance Report, Coast edition (2020) (Table 1). These sources of data are produced and compiled by the Kenya Fisheries Service (KeFS). Data from 2016 was used to standardize fish families/categories to ensure that taxa groupings were consistent through the whole time series. The data from the sources were disaggregated into mixed categories of fish families and fish groups which were at species, genera, family, or broader grouping; thus, our calculations were limited to analysis of these mixed fisheries categories (henceforth referred to as “fish taxa” in the rest of this article). We also use the term fish to include all aquatic animals, including finfish, shellfish (molluscs and crustaceans) and other invertebrates (such as sea cucumbers) (FAO, 2022). Due to COVID-19 restrictions during 2020, we indicate that there may be some uncertainty associated with the data from 2020. The COVID-19 curfew affected many fishers who undertake night fishing as the timing of their fishing expedition was interrupted. The same was noted for fishers who leave early morning in search of baits when the curfew hours extended to 5 AM, yet this is the ideal hour for their bait searching (Kenya Fisheries Statistical Bulletin, 2020).

Table 1 Fish production data sources that were used in this study
Full size table

2.3 Fish supply and fish consumption

To calculate the supply of fish available for human consumption, we first subtracted exported quantities from the reported supply quantities for the respective years and fish taxa to obtain net supply of marine and freshwater capture quantities in Kenya. Among the imported species/taxa reported in Kenya include Sphyraenidae (frozen barracudas), frozen mixed fish and fish products, Scombridae (frozen mackerels), Salmonidae (frozen salmon and sea trout) and Cichlidae (tilapias) while exports include the taxa Sciaenidae (drums or croakers), Latidae (Nile perch), Lutjanidae (snappers), Lethrinidae (emperors), Serranidae (groupers), Haemulidae (grunts), Scombridae (tunas and mackerels), Carangidae (kingfishes), Engraulidae (anchovies), Istiophoridae (billfishes), Sharks andRays, Palinuridae (lobsters), Portunidae (crabs), Penaeidae (shrimps/prawns), Octopodidae (octopus) and Loliginidae (squids). We then calculated the total supply of fish by summing net supply from inland freshwater systems, coastal and marine systems, aquaculture, and imports (Lofstedt et al., 2021). We multiplied the net weights of fish supply data (capture, aquaculture, imports, and exports) by established edible portion conversion factors for finfish, molluscs, and crustaceans, which represent the portion remaining after gutting and deboning (Table 2) (Edwards et al., 2019; Thurstan & Roberts, 2014). Fish that are consumed whole (e.g., small pelagic species) or that were reported as processed products (e.g., fillets) were assigned a conversion factor of 1 (Table 2).

Table 2 Edible fish conversion percentages and multiplier factor for the fish and fish products extracted from the Kenya fisheries data sources
Full size table

Next, to calculate apparent consumption of fish (per capita fish supply), the total supply of fish was divided by Kenya’s total human population for the applicable year to establish the per capita fish supply. The annual Kenya human population estimates for 2005–2020 were obtained from the UN (United Nations Department of Economic and Social Affairs, Population Division, 2019) who conduct demographic research, support intergovernmental processes in the area of population and development, and assists countries in developing their capacity to produce and analyze population data and information.

We evaluated the change in the supply of fish across all landed taxa through time. The time-series did not meet the main assumptions for fitting a trend line by linear regression. Therefore, we used the nonparametric Mann-Kendall (MK) trend test, followed by the Thiel-Sen estimator or Sen’s Slope test to determine and fit the trend’s slope to examine whether the changes in trends in total supply, and per capita supply over 2005–2020 were significant (Aditya et al., 2021; Chen et al., 2022; Helsel et al., 2020).

2.4 Nutrient content of fish and nutrient supply

Our analysis focused on the concentration of five key micronutrients (calcium, iron, selenium, vitamin A, and zinc) and two macronutrients (total omega-3 polyunsaturated fatty acids (PUFA) and protein) across the fish taxa reported in the statistical bulletin reports. These nutrients were chosen because they are known to be lacking in diets in East Africa, yet rich in fish (Hicks et al., 2019; Ngare & Muttunga, 1999; White et al., 2021). Estimates were sourced for these nutrients from FishBase (Froese & Pauly, 2021; Hicks et al., 2019; MacNeil, 2021). Where data were reported to species level, species nutrient values were taken directly from FishBase. Where data was reported at broader fish taxa level, we used the average nutrient values across species in that fish taxa that were found in Kenya’s geographical region on FishBase. For non-finfish species (molluscs, crustaceans, sea cucumbers), where there are fewer data and species-specific estimates, nutrient values were obtained from the Food and Agriculture Organization’s International Network of Food Data Systems (INFOODS) (INFOODS: Software Tools, 2021; SELF Nutrition Data | Food Facts, Information and Calorie Calculator, 2021) based on estimates averaged to the lowest taxonomic unit. We used the preceding nutrient information to develop a national fish nutrient content database for the 52 consumed fish taxa reported in the most recent (2020) statistical bulletin in Kenya (Table 3). We used these 52 taxa since they were the most recently reported in the Fisheries Statistical Bulletin and a representation of what is currently available for human consumption at the time of our analysis. We further used this nutrient information data to calculate the nutrient supply (amount of each nutrient available in Kenya from the fish supplies) by multiplying nutrient content per 100 g of each fish taxon by the total weight of each fish taxon. We then calculated nutrient supply from the four sources (marine fisheries, freshwater fisheries, aquaculture, and import) by summing up the nutrient supply of fish within each source.

Table 3 Nutrient content of the 52 fish taxa reported in the most recent Fisheries Statistical Bulletin (sorted by their percentage abundance in supplied edible biomass)
Full size table

We examined the correlative associations among the seven nutrients for all the currently supplied 52 fish taxa using principal component analysis (PCA) (Kassambara, 2017). PCA was used to identify relative associations among the fish taxa and nutrient types and examine the significance of the contribution of the nutrients on the principal axes of the associations. This approach helped identify the principal directions in which the nutrient content data varied, and where similarities and differences between fish taxa occurred. PCA factors with eigen values greater than 1, and which the scree plot indicated a break or elbow cut-off were used to identify patterns in nutrient content among fish taxa (Kassambara, 2017).

2.5 Nutrient density of fish supply

We calculated the nutrient density of each fish taxa recorded in Kenya’s fish supply data based on the nutrient content of each fish taxa relative to established recommended nutrient intake values. Nutrient density was defined and calculated as the contribution a 100 g portion of fish (Drewnowski, 2009) would make towards the recommended nutrient intakes (RNI) of a target population group (in our case, children under five years) (Hicks et al., 2021; Maire et al., 2021). We used per 100 g as a standard unit used for nutrient composition on FishBase, FAO INFOODS database and the Kenya Food composition tables. Targeting children under five years was considered an appropriate strategy since interventions that improve early nutrition have the potential to improve outcomes throughout life (Black et al., 2020; Heidkamp et al., 2021). Nutrient density is therefore the sum of the percentage contribution that a 100 g of raw fish makes to the recommended nutrient intake across all seven nutrients (calcium, iron, selenium, vitamin A, zinc, omega-3 PUFA, and protein) (Maire et al., 2021). The percentage contribution to RNI for each nutrient was capped at 100% to prevent extreme values dominating patterns of variation in density scores such as fish sources or fish taxa with especially high values for some nutrients such as protein and selenium (Maire et al., 2021). Nutrient density scores can therefore scale up to a maximum potential value of 700%, where all seven nutrients are fulfilling recommended nutrient intakes. We calculated the weighted average of nutrient density across all reported fish taxa in six categories representing each source (marine and freshwater capture, aquaculture, and imports) as well as exports (marine and freshwater), and across the twenty top taxa (seven freshwater and thirteen marine taxa).

The Kenya country-level information on nutritional intakes has been reported qualitatively based on the significance of micronutrients and macronutrients to human development. Therefore, there are no nutrient-specific quantitative recommendations except for overall animal protein intakes (Ministry of Health, 2017). To overcome this limitation, estimations of the RNI for children aged under five years were based on the recommended values by the FAO and WHO (FAO/WHO, 2004) for micronutrients, recommended intakes by the Institute of Medicine (Institute of Medicine, 2006) for protein adequate intakes (AI) and RNI for total omega-3 PUFA adequate intakes based on the FAO (FAO, 2010) (Table 4).

Table 4 Average recommended nutrient intakes (RNI) for micronutrients, adequate intakes (AI) for protein and omega-3 PUFA in children under five years, Source: (FAO, 2010; FAO/WHO, 2004; Institute of Medicine, 2006)
Full size table

For each of the 84 taxa reported in the entire 2005–2020 dataset, we calculated the amount or portion size of fish in grams that would be required to meet an average of 33.3% of the recommended nutrient intake across the six nutrients (calcium, iron, selenium, vitamin A, zinc, and omega-3 PUFA) for a child under five years old (Beal, 2020; Hicks et al., 2021; White et al., 2021). We focused on children under five years of age and made the assumption that one-third of their recommended nutrient intake would come from fish (Ortenzi & Beal, 2021). The contribution of each nutrient is capped at 100%, such that an average of 33% of requirements can be met by two nutrients providing 100% of requirement or six nutrients providing 33.3% of the requirements (Beal, 2020; White et al., 2021). We then present the number of taxa, and the biomass of supply across the portion sizes, and what proportions of these are drawn from taxa originating from inland freshwater systems versus coastal and marine systems. Here, we used all (84) taxa that have ever been reported in the Fisheries Statistical Bulletins during 2005–2020 to document their nutritional contribution by portion sizes, even though some of the taxa were not reported in the most recent (2020) Statistical Bulletin.

2.6 Changes in fish and nutrient supply and potential contributions to nutritional security

To examine changes in fish supply, nutrient supply, and apparent consumption, we used 2005 as a reference year, and 2020 as our most recent year for which there was supply data and calculated percent change in fish supply, nutrient supply, and apparent fish consumption using the equation:

$$Percent\,change=\frac{Fish\,supply\,in\,2020-Fish\,supply\,in\,2005}{Fish\,supply\,in\,2005}*100$$

We substituted the fish supply in the above equation with per capita fish consumption and nutrient supply, to get the change in apparent consumption and nutrient supply respectively. Nutrient supply was calculated by multiplying nutrient content of fish taxa by weight of production. To determine the potential contributions of the available fish supply to the Kenyan population, we divided the available fish supply by the recommended per capita consumption of 10.4 kg/yr based on WHO recommended intake of regularly eating 1–2 servings of fish per week. We used 100 g as the standard serving and 52 weeks per year to get 10.4 kg/yr. To estimate the potential contributions of the available fish supply to the Kenyan population residing within 20 km of the inland and marine fisheries water bodies, we assumed that 20.4% of Kenya’s population was residing within 20 km of inland and marine fishery water bodies (Robinson et al., 2022) and therefore divided the available fish supply by this population. To assess the potential contribution to nutrition security of the available nutrient supply, we divided the available supply of each nutrient by the nutrient needs of our target population. All the data analyses were completed in the freely available R and RStudio software (R Core Team, 2022; RStudio Team, 2020).

3 Results

3.1 Trends in the supply of fish

Most of the supply of fish in 2020 came from inland freshwater systems (90,783 tonnes (t) representing 67% of total supplies), followed by coastal and marine systems (19,179 t representing 14% of supplies), aquaculture (17,352 t representing 13% of fish supply), and finally imports (8,430 t representing 6% of the supply quantities) (Fig. 1a). Between 2005 and 2020, the supply of fish from inland waters declined by 21% (Fig. 1a). This decline was compensated by increases in coastal and marine systems, aquaculture, and imports, resulting in an overall 11% marginal increase (p = 0.048) of fish supply (Fig. 1a, b). However, this compensation for the decline in inland freshwater supply did not translate into adequate fish supplies to meet the national per capita supply, which declined significantly (p = 0.015) by 24% from an estimated 3.3 kg/capita/year in 2005 to 2.5 kg/capita/year in 2020 (Fig. 1c).

Fig. 1
figure 1

Supply of fish in Kenya a disaggregated by sector (inland freshwater capture fisheries, coastal and marine capture fisheries, aquaculture, and imports) b as total supply of edible fish, and c as apparent per capita fish consumption. Data sources: Kenya Fisheries Service Statistical Bulletins, Fish inspection and quality assurance reports and Kenya National Bureau of Statistics (2021) Economic Survey Report

Full size image

3.2 Nutrient content of the available fish

The nutrient content of fish varied widely across taxa with an indication of overlaps of inland freshwater-derived and marine water-derived taxa except for the freshwater silver cyprinid (family Danionidae, locally called “omena”) and trout (family Salmonidae) (Fig. 2a). Principal component analysis (PCA) of all 52 fish taxa revealed three nutrient content patterns (Fig. 2b, c, d). The three nutrient content patterns together explained up to 72.4% of the variation in the nutrient content. Table 5 shows the statistical significance of the nutrients with high factor loadings on each of the three PCA axes representing the nutrient content patterns. The first nutrient content pattern identified fish that were high in calcium, iron, selenium, zinc, and total omega-3 PUFA, and accounted for 32.3% of the variance. This nutrient content pattern was associated with freshwater silver cyprinids (Danionidae), bivalve molluscs, trout (Salmonidae), anchovies (Engraulidae), mullets (Mugilidae), squids (Loliginidae), octopuses (Octopodidae), cuttlefishes (Sepiidae), sea cucumbers (Holothuridae) and swordfish (Xiphidae) as the first 10 fish taxa with greatest PCA factor loadings. The second nutrient content pattern revealed fish that were rich in selenium, total omega-3 PUFA, and protein, accounting for 24.5% of the variation. This nutrient pattern was associated with trout (Salmonidae), bivalve molluscs, Nile perch (Latidae), whiptails (Trichiuridae), freshwater silver cyprinids (Danionidae), needlefishes (Belonidae), cuttlefishes (Sepiidae), sea cucumbers (Holothuridae), groupers (Serranidae) and goatfishes (Mullidae). The third nutrient content pattern was associated with calcium, zinc, and protein, and explained 15.6% of the variation. The top 10 fish taxa with the greatest PCA factor loadings this pattern included lungfish (Protopteridae), bivalve molluscs, freshwater silver cyprinids (Danionidae), trout (Salmonidae), needlefishes (Belonidae), anchovies (Engraulidae), edible crabs (Portunidae), spiny lobsters (Palinuridae), slipper lobsters (Scyllaridae) and cuttlefishes (Sepiidae). The larger overlap of fish taxa at the centre of the PCA plot reflects similarity in nutrient contents of these fish per 100 g portion size. Fish taxa at extreme directions of the nutrient arrows imply high content of specific nutrients per 100 g portion size.

Fig. 2
figure 2

a Principal Component Analysis (PCA) of the nutrient content across seven nutrients among the 52 fish taxa reported in the Kenya fisheries statistics. Fish taxa in green refer to those of freshwater origin while those in blue refer to those from coastal and marine waters, Salmonidae is sourced from aquaculture, b Fish nutrient content pattern 1, c Fish nutrient content pattern 2, d Fish nutrient content pattern 3

Full size image
Table 5 Significant correlation associations of the fish nutrient content on the three PCA axes representing the 3 nutrient content patterns
Full size table

3.3 Nutrient density of targeted fish by source and taxa

The nutrient density of an average fish varied by source (i.e., marine and freshwater capture, aquaculture, imports, and marine and freshwater export) (Fig. 3a). An average 100 g serving of fish from any of the sources has the potential to meet all the selenium and protein dietary requirements for a child under five years (Fig. 3a). Fish from the inland freshwater systems have the highest nutrient densities. An average 100 g serving has the potential to meet over 45% of the nutrient needs for a child under five years across all nutrients except vitamin A, with potential calcium contribution of 90%, and total omega-3 PUFA contribution of 64% (Fig. 3a). The coastal and marine systems, aquaculture, imports, and exports have the potential to support over 25% fulfillment of recommended intakes for iron, selenium, zinc, total omega-3 PUFA, and protein.

An average 100 g serving of fish from all the 20 most abundant fish taxa by edible weight, has the potential to fulfill the selenium and protein requirements for an under five years old child, but varied in contributions to other nutrients (Fig. 3b). Seven of these fish taxa were from inland freshwater systems while 13 were coastal and marine fishes. Family Danionidae, which consists of one species, the freshwater silver cyprinid (Rastrineobola argentea, locally called “omena”) had the highest nutrient density, with a 100 g serving providing over 60% to the recommended intake of all nutrients except for vitamin A where it contributed 19%. Furthermore, a 100 g serving of this freshwater silver cyprinid (Danionidae) could meet 100% of the recommended intake of calcium for children under five years. Among the coastal and marine species, octopus (Octopodidae) had the highest nutrient density followed by sardines (Clupeidae). Two marine fish families, namely Belonidae (needlefishes) and Lutjanidae (snappers) were found to contribute over 25% to vitamin A requirements, higher than the other taxa. The different fishes contributed to some degree to different nutrients and therefore hold some potential to contribute to recommended nutrient intakes.

Fig. 3
figure 3

Nutrient density of fish presented as the percent contribution an average 100 g portion of fish could make towards the daily recommended nutrient intake (RNI) for a child under five years across five micronutrients (calcium, iron, selenium, vitamin A and zinc) and two macronutrients (omega-3 PUFA and protein) (maximum 700%) grouped by a production source (marine capture, freshwater capture, aquaculture, imports, marine and freshwater exports (numbers in brackets refer to the number of fish taxa as reported in the data sources); b top twenty fish taxa (making up to 95% of reported edible catch). Bars represent the contribution a 100 g portion of fish provides of the seven key nutrients relative to the recommended nutrient intake (RNI) for children under five years. For each nutrient, the value within the bar represents the percentage contribution to RNI. The top seven species are from inland freshwater systems and the bottom thirteen from coastal and marine systems (sorted by their quantities from most to least within the two systems)

Full size image

3.4 Portion sizes required to meet nutrient needs

Based on the fish available in Kenya, consuming portions that range in size between 46 and 125 g, would be sufficient to meet an average of 33% of the daily nutrient needs across six nutrients (calcium, iron, selenium, vitamin A, zinc, and omega-3 PUFA) for a child under five years (Fig. 4a, b). For the most abundant fish taxa by biomass, consuming a portion of 80–90 g would be sufficient to meet these needs; the next most abundant taxa (family Danionidae) would only require a 40–50 g portion size (Fig. 4b). Despite the range of taxa and sources that comprise Kenya’s aquatic food system, which is dominated by coastal and marine taxa, freshwater fishes constitute the largest biomass in the 40–50 g, 70–80 g, 80–90 g and 100–110 g portion size categories. The freshwater silver cyprinid (Danionidae) was the only species in the 40–50 g portion size, mixed other freshwater species constituted 46% of the 70–80 g portion size, and families Latidae – Nile perch (one species - Lates niloticus and Cichlidae - Tilapias (mainly Oreochromis niloticus) made up 50% and 36% respectively of the 80–90 g portion size, and Clariidae - Catfish (main species - Clarias gariepinus) was the most abundant in the 100–110 g portion size categories where it made up 57% of the biomass (Fig. 4b). On average, a child under five years old would need to consume approximately 66 g, 80 g, 89–93 g of a fish sourced from inland freshwater systems, import, aquaculture systems, or coastal and marine systems respectively to meet the 33% RNI across the six nutrients (calcium, iron, selenium, vitamin A, zinc, and total omega-3 PUFA) (Table 6). The portion sizes required to reach an average of 33.3% RNI across the 84 fish taxa are presented in the Supplementary material (Fig. S1).

Fig. 4
figure 4

a The fish portion size (g) required to reach an average of 33.3% of daily recommended nutrient intakes across six nutrients (calcium, iron, selenium, vitamin A, zinc, and total omega-3 PUFA) for all 84 fish taxa (2005–2020); the equivalent values for terrestrial animal source foods (beef, pork and chicken) shown above were from Tacon & Metian, 2013. Recommended intakes are for children under five years old. All fish taxa were assigned to either freshwater or coastal and marine ecosystem (including aquaculture and imports). Chicken is highlighted* since it falls beyond the range of 40–130 g but forms part of dietary animal protein for most Kenyans. b Edible fish biomass supply across the portion sizes, indicating that most of the available biomass falls between 80–90 g and 40–50 g portion size respectively

Full size image
Table 6 The average fish portion size by main fish taxa within each fish supply system that would fulfill 33.3% of recommended nutrient intakes across six nutrients (calcium, iron, selenium, vitamin A, zinc, and omega-3 PUFA) for children under five years old. Fish taxa are ordered from most abundant within each system for the top 10 except aquaculture which has four species
Full size table

3.5 Potential and changes in fish contributions to nutritional security

There was a small (11%) increase in fish supply between 2005 and 2020 (Fig. 5a). However, over the same period, the potential of this fish supply to meet per capita fish consumption decreased by 24% and nutrient supply declined by between 25 and 40% (Fig. 5a). The largest decline in nutrient supply was for calcium while the smallest was for vitamin A (Fig. 5a). Despite these declines, in 2020, calcium, selenium, and protein supplies had the potential to meet the nutrient requirements (based on the assumption that 33% of each nutrient requirement comes from fish) for all children under five years (Fig. 5b). Supplies of omega-3 PUFA, iron, zinc, and vitamin A had the potential to meet the nutrient requirements for 93%, 72%, 70% and 17% of the children under five years, respectively (Fig. 5b). The current fish supply of 135,745 t has the potential to provide fish for 13,052,432 Kenyans (24% of the national population) based on the World Health Organization (WHO) recommendations of two servings (200 g) of fish per week. If targeted for local consumption by households living within 20 km of the inland and marine fishery waters, this supply would meet the recommended per capita consumption of 10.4 kg/yr for all the population (10,979,600) residing here.

Fig. 5
figure 5

a Change in fish supply, apparent per capita fish consumption, and nutrient supply between 2005 and 2020, and b the proportion of Kenya’s children under five years old whose recommended nutrient intakes (33% of requirements per nutrient) could be met with current (2020) supply from fish, bars with positive values shown indicate those nutrients that are in surplus by the indicated value

Full size image

4 Discussion

Our findings demonstrate that while fish supply in Kenya has increased by 11%, per capita fish consumption has declined by 24% and the total nutrient supply has decreased by 25–40% over the 2005–2020 period, due to both changes in the quantity and types of species produced and an increase in the national population. Here, we discuss the implication of this for food and nutrition security in Kenya and highlight the potential that the current supply has if it were to be targeted to meet the nutrient needs of vulnerable children under five years. We further evaluate two reference populations (percent of the national population and populations near fishery water bodies) to show the potential of current fish supplies to meet the recommended per capita fish consumption of 10.4 kg/yr.

4.1 Declines in fish supply for consumption threaten food and nutrition security in Kenya

Kenya experienced a 25–40% decline in nutrient supply from fish, an even greater decline than in per capita fish supply, despite the 11% increase in overall fish supply. The decline in supplied nutrients is associated with changes in the composition and quantity of taxa, resulting from declines in freshwater taxa and increases in aquaculture, imported, and coastal and marine taxa. Similar declines in nutrient supply from fish have been shown in Bangladesh and Brazil, when aquaculture or imported taxa replace traditional wild species (Bogard et al., 2017; Heilpern et al., 2023). The changes in the quantity and types of species produced could be attributed to various factors. These include declining fish stocks in both marine and freshwater systems (Aura et al., 2020; Hicks & McClanahan, 2012; Kimani et al., 2018; Schubert et al., 2021), and alien species invasion, rising eutrophication, and climate change impacts in Lake Victoria (Nyamweya et al., 2022). There have also been large negative impacts on local fish supply due to an interplay of direct and indirect impacts of fish trade arising from higher demand for fish of Lake Victoria, mainly Nile perch (Latidae) and the freshwater silver cyprinid (Danionidae), in the export market and for fishmeal respectively (Abila, 2005). The decline of nutrients could be attributed to declines of the inland freshwater catches from Lake Victoria (Aura et al., 2020). However, given the fast life history of the freshwater silver cyprinid (family Danionidae) and stock management spanning the three countries (Kenya, Uganda, and Tanzania), there is some uncertainty over potential landings for this species with some analyses suggesting catches could be sustainably increased (Wessels et al., 2023). Therefore, policy interventions to increase supplies of the freshwater silver cyprinid could contribute to more nutrition security. The biomass dominance of the fish supply by freshwater species is a key feature of the Kenya fishery (Schubert et al., 2021). Therefore the decline in this biomass highlights the need for improved fisheries management and conservation to rebuild fish stocks, focus on nutrient-rich aquaculture and imports and harnessing nutritious small and medium pelagic coastal and marine fish that are locally important in order to increase fish supply (Njiru et al., 2021). Promoting sustainable on-farm aquaculture aligns with the global trend in fish supplies that is transitioning from dependence on wild fisheries to aquaculture (Belton & Thilsted, 2014). However, such aquaculture interventions need to move from maximizing production and embrace a nutrition-sensitive approach that considers nutritional quality (Bogard et al., 2017; Heilpern et al., 2023). Although there have been expectations that fish imports and aquaculture growth would narrow the fish supply gaps in East Africa (Obiero et al., 2019), our findings highlight the fact that there will need to be large increases in supply from these other sources, to match the increase in population (and thus, increase in demand). The supply of nutrients from these sources also must be addressed as the change in the supply will lead to a drop in the supply of essential micronutrients (Bogard et al., 2017).

Between 2005 and 2020, Kenya’s population grew by 46%, from an estimated 37 million to 54 million, exceeding the 11% rate of increase in fish supply, and resulting in a 24% decline in the per capita fish available for consumption. This decline in fish consumption in Kenya is set in the context of a longer-term decline in consumption across Sub-Saharan Africa of 14% between 1990 and 2002, reaching a low of 6.7 kg/yr, with Sub-Saharan Africa being the only part of the developing world where fish supply per person was declining while production was seemingly increasing (Béné et al., 2010). Human populations are expected to grow further across the wider East Africa and western Indian Ocean region, suggesting an increased demand for seafood at this larger regional scale (Nyamweya et al., 2022; Taylor et al., 2019).

4.2 Kenya’s aquatic food systems hold potential to support nutrition security

Improving the contribution of small-scale fisheries to overall national food and nutrition demand is Kenya’s Vision 2030 long-term development blueprint aspiration. However, with the characteristic fish deficits reported here and elsewhere (Obiero et al., 2019; Odoli et al., 2019; Schubert et al., 2021; Taylor et al., 2019), targeting vulnerable groups like children under five years is important. We established that fish nutrient supplies in Kenya are sufficient to meet one-third of the calcium, selenium, and protein requirements for all children under five years and one-third of the omega-3 PUFA, iron, and zinc requirements for over 70% of these children highlighting the potential for targeted nutrition policies to increase the consumption of fish among children. However, food systems are complex and multi-sectoral, interventions will thus need to involve multiple actors, including educationists, health and nutrition specialists, fisheries scientists, governance and policy actors (Codjia et al., 2024; Kamau-Mbuthia et al., 2023; Yazdanpanah et al., 2023).

Children under five years are in an important life stage, during which adequate nutrition can support longer-term health benefits. Therefore, by reducing the likelihood of undernutrition, stunting, micronutrient deficiencies, and children not reaching their developmental potential (Black et al., 2020; Byrd et al., 2021a, b; Shrestha et al., 2022), aquatic food systems can offer an eminent ecosystem service to be anchored in food nutrition policy and guidelines (Bennett et al., 2021). Although it is unrealistic to expect all fish to be consumed by children under five, a range of strategies are available to include or increase fish in the diets of children. Strategies could include: incorporating fish into complementary feeding (Codjia et al., 2024; World Health Organization, 2023); promoting the inclusion of fish into school feeding programs (Ahern et al., 2021; Khan et al., 2021); raising awareness of locally available and affordable nutritious fish (Kimiywe et al., 2024; Ryckman et al., 2022); sharing knowledge on fish processing strategies (e.g. converting fish into powders or pastes) to enhance uptake of fish in preferred forms by children (Kimiywe et al., 2024); and revision of dietary guidelines or nutrition action plans to target this age group through recommendations that they are prioritized in the household for fish consumption (Khan et al., 2021). However, achievement of target interventions would require further stakeholder engagement and strategically working together for successful implementation to improve nutrition status (Codjia et al., 2024; Kamau-Mbuthia et al., 2023).

There are a large number and diversity of fish taxa available for consumption in Kenya, and all are an important source of nutrition. The portion sizes for fish (46–125 g) needed to meet one-third of the nutrient needs of a child under five years are in the same range, or smaller, as other animal-sourced foods such as beef (74 g), pork (93 g) and chicken (135 g) (Tacon & Metian, 2013). Fish has the additional benefit of having many options (many fish taxa) to choose from, and locally being often cheaper (Robinson et al., 2022; Ryckman et al., 2021). This diversity increases the likelihood of species being more affordable and accessible within local contexts (Ryckman et al., 2021). Furthermore, fish have unique richness in nutrients that are mostly lacking in diets (Bernstein et al., 2019; White et al., 2021) that makes them standout as significant nutritious food for human health improvement in tropical developing nations. Certain species, such as the small freshwater silver cyprinid (Danionidae) have particularly high nutrient density (Byrd et al., 2021a, b; Isaacs, 2016; Wessels et al., 2023). Freshwater silver cyprinids are both the most nutritious and most abundant species in Kenya; a 40–50 g portion is sufficient to meet one third of the dietary needs of a child under five years across six key nutrients. Freshwater silver cyprinids are also one of the most affordable fish and remains a good source of nutrition when dried (Robinson et al., 2022; Ryckman et al., 2021; White et al., 2021). Shellfishes (bivalve molluscs, octopus, squids, crustaceans), small marine species such as families Clupeidae (mainly sardines) and Engraulidae (anchovies) were also found to be nutritious (Bernstein et al., 2019; Isaacs, 2016; Robinson et al., 2022). Small freshwater silver cyprinid, sardines and anchovies reported here are rich in calcium, zinc and omega-3 PUFA (Robinson et al., 2022; Wessels et al., 2023). It is important to note that small portion sizes needed for the most nutritious fish are particularly valuable for targeting infants who have small stomachs, so as not to displace breast milk intake (Byrd et al., 2021a, b; Dewey & Brown, 2003). Fish consumption in general in Kenya could be highlighted for its importance to human health according to nutrient content profiles, and particularly nutrient needs for vulnerable stages such as when pregnant, lactating or under the age of five years (Bonham et al., 2009; Byrd et al., 2021a, b; Cartmill et al., 2022; Gibson et al., 2020).

The portion size ranges we found here call for a reconsideration of the dietary guidelines in Kenya. The Kenya National Guidelines for Healthy Diets and Physical Activity 2017 recommend at least two servings/week of lean meat including fish, with each serving for fish approximated at 30 g (Ministry of Health, 2017). This translates to a minimum of 60 g/week. Our findings suggest that consuming the most nutritious and abundant fish (Danionidae) would require consuming 46 g/ day while consuming the less nutritious family Gempylidae (snake mackerels) would require consuming 125 g/day (Supplementary Fig. S1) to meet recommended nutrient intakes across the six valuable micronutrients (calcium, iron, selenium, vitamin A, zinc, and omega-3 PUFA) for a child under five years. Hence to help tackle malnutrition among target population groups such as in children under five years, whose nutrient needs we used in this study, there would be need to have taxa-specific guidelines such as 46 g/day for the nutritious freshwater silver cyprinid (Danionidae) and a minimum average of 80 g/day for the more abundant fish taxa that includes Nile perch (Latidae) and tilapias (Cichlidae). These proposed ranges of daily portions match recommendations by many nations and global bodies on how much fish a population must eat to have beneficial health outcomes that have been found to vary widely from 97 to 550 g/week (Thurstan & Roberts, 2014). Our findings highlight a range from 92 to 160 g/week if the twice per week consumption recommendation is maintained or maximized at 322 to 560 g/week assuming daily fish consumption. However, in Kenya, achieving daily consumption of fish would be challenging due to the declining per capita fish supply and declining nutrient density. Therefore, the potential of fish supply and nutrients to support nutritional security would require a targeting strategy, such as identified populations with specific nutrient needs or those near fishery water bodies while considering the interplay of factors that may influence fish consumption.

Kenya’s per capita consumption estimated here at 2.5 kg/capita/yr would need to increase four-fold to meet common global (10.2 kg/capita/yr) (Willett et al., 2019) or regional (10.6 kg/capita/yr) (Canty & Deichmann, 2022) recommendations for fish consumption, and is one eighth of the global average fish consumption of 20.3 kg/capita/yr (FAO, 2020). The present fish supply can meet national per capita fish consumption of 10.4 kg/yr for 24% (13,052,432 people) of Kenya’s population. Alternatively, this supply was found adequate to meet the per capita fish consumption of all Kenya’s population (11 million people) residing 20 km from inland and marine fishery water bodies. However, there is a need to seek a variety of interventions including improvements in fisheries management, if the fishery is to be enhanced to contribute to increased fulfillment of recommended nutrient intakes and food and nutrition security commitments of the nation. Thus, ensuring Kenya’s fish food system can support food and nutrition security will require policies and implementation frameworks that identify where interventions are needed for the vulnerable, lactating mothers, and children, considering inadequate nutrient intakes in children can result in poor health, educational, and economic outcomes (Byrd et al., 2021a, b; Ryckman et al., 2021; Shrestha et al., 2022). Further, mainstreaming of the information on nutrient contribution from fish in food policies would be a positive human developmental milestone especially in areas with reported cases of undernutrition including those close to small-scale fisheries resources. Therefore, maintaining the supply and accessibility of fish from the different sources and fish taxa would be critical towards supporting dietary quality. However, due to supply inadequacy, further assessment to identify ‘problem nutrients’ and identify those in need of nutrition is advised. This could be guided by the understanding that not all communities in Kenya consume fish due to various socio-cultural, economic, and geographic reasons (Ayuya et al., 2021). This calls for concerted efforts on the inclusion of these small-scale fisheries (the usually forgotten food) in food and nutrition security policies and dialogues for the country to achieve milestones in the pledges to the sustainable development goals and other food-related agenda.

5 Conclusions

This study is the first to analyze and present combined fish nutrient content data, nutrient density, portion size requirement to meet 33% needs for six nutrients and examine change in fish-derived nutrients from tropical marine, freshwater fisheries, aquaculture, and imported fish in Kenya. We framed this in the context of targeting malnutrition among children under five years of age who are vulnerable to nutritional inadequacies at such a critical developmental stage. By combining fish nutrient content information with supply, we have unveiled the nutrient supply of Kenya’s fish, teasing out how this often-forgotten food can be harnessed to alleviate nutrient deficiencies in vulnerable target population groups such as children under five years, or populations close to fishery water bodies. In doing so, we firstly established that the fish taxa supplied in Kenya can play a substantial role in ameliorating micronutrient deficiencies among target vulnerable populations. Marine and inland freshwater capture fisheries, aquaculture, and imported supplies contain key nutrients that would be beneficial to improved health. We highlight empirically that fish consumption could be beneficial to target vulnerable populations such as children under five years, based on the nutrient profiles of the available fishes from the four main sources. With this information, national food policy and dietary recommendations could be updated. For example, diets incorporating fish within the portion sizes of 40–50 g per day could be used in public health nutrition policies to advise on the prevention of micronutrient deficiencies in children under five years, based on consumption of the abundant freshwater silver cyprinids (Danionidae), commonly referred to as omena or dagaa in local contexts. Similar guidelines can be followed for other fish taxa highlighted in Table 6 of this paper, but being cognizant of available fish quantities, diverse fish preferences, quantities of fish that a child can consume, acceptability, cultural contexts, and affordability of the fish.

Secondly, we identified three pathways by which the fish supplies can help improve food and nutrition security by being mainstreamed into national food and nutrition policy:

  1. 1.

    Promoting the intake of nutrients from across diverse fish taxa. This would be dependent on which nutrient types are locally deficient. Advising on the options available based on the results of this study would be limited to the seven nutrients that we assessed.

  2. 2.

    Adopting proposed portion sizes of fish taxa that an individual would need to consume to meet 33% of nutritional requirements across the six key nutrients.

  3. 3.

    Supporting scalability of our assessment of potential contribution of fish to tackling malnutrition in children under five years to other vulnerable population groups such as pregnant women, lactating women or any other. However, this pathway would require increased fish supply to reverse the declining per capita fish supply for nutrition security.

Thirdly, we highlight that declining inland freshwater fish supply in terms of quantity may be compensated by increases in marine and coastal fisheries production, aquaculture production, and imports. However, because on average, these sources are less nutritious than freshwater sources, and human population is expected to continue to grow, greater increases than losses will be necessary. The nutritional quality of fish lost by the decline of nutritious freshwater species would not be compensated for by these three sources, highlighting the need to ensure conservation and management efforts are put in place to sustain inland freshwater fisheries, that contribute most of the fish catches by weight and coastal and marine fisheries, that contribute most of the fish catches by numbers of taxa supplied.

Finally, through this study, we make an important contribution to the body of empirical evidence showing that the inclusion of tropical small-scale fishery food systems in food policy could support efforts to tackle nutrient deficiencies by supplying key bioavailable nutrients including calcium, iron, selenium, vitamin A, zinc, total omega-3 PUFA, and protein. Although our grouping of fish at broad taxa reported in fisheries statistical bulletins could obscure some nutritious species, these findings offer an improvement from what is reported in the current Kenya food tables and is aligned with how fisheries resources are assessed and sold. For future improvements, we propose that further research be focused on disaggregating fish supplies at the species level as data collectors get trained in fish taxonomy. The need to mainstream small-scale fisheries in food system policies, and dialogues would therefore be key to meeting national and international commitments to tackling all forms of hunger. We suggest some caution in the interpretation of the nutrient contribution from aquaculture in this study since it was based on nutrient content estimated for fish taxa in the wild, but aquaculture species may have varied nutrient concentrations, hence a gap for future investigations.