Postharvest Storage Practices of Maize in Rift Valley and Lower Eastern Regions of Kenya: A Cross-Sectional Study

An assessment of local farmers' knowledge, attitude, and practices on postharvest maize storage and management was carried out with a view of understanding its role in maize contamination with mycotoxins and postharvest losses in Rift Valley and Lower Eastern Regions of Kenya among 165 and 149 farmers, respectively. Differences between the two regions were analyzed using the Chi-square test, Fisher exact test, and two-sample t-test. The median quantity of maize harvested by farmers in the two regions after shelling was 585 kg. A median of 20 kg of maize was put aside as a result of rotting before shelling, and there was a significant mean difference in maize set aside as a result of rotting between the two regions (107.88 kg vs. 31.96 kg; t (306.25) = 5.707, P value <0.001). The quantity of discoloured and mouldy maize consumed ranged from 0 to 90 kg; 7 (2.2%) respondents consumed mouldy maize, 36 (11.5%) fed it to cows, and 19 (6.1%) fed it to poultry. A small percentage (3.5%) believed mouldy maize is safe for human consumption, 23.6% for animal consumption, while 15.0% considered it safe for brewing, with the differences between the two regions being statistically significant (P value <0.05). Nearly half of the respondents (48.4%) kept maize on cobs indoors, 47.1% left it in the field without covering, and 33.1% consumed and sold maize while still green, with more farmers from Lower Eastern practicing this. The results of the study suggest that there were poor postharvest practices and low awareness levels among maize farmers and that this can lead to postharvest losses due to Fusarium spp. infection and mycotoxin contamination that poses a threat to human and animal food safety. This calls for interventions on better postharvest practices.


Introduction
Agriculture is the backbone of the economy in most Sub-Saharan African (SSA) countries, contributing significantly to their Gross Domestic Product (GDP) [1]. In this sector, grains are its major product [2], of which maize is the main contributor.
Maize is a vital food crop cultivated in most parts of the world, especially in low-and middle-income countries (LMICs). Globally, it is the third most grown cereal crop that serves as the primary source of food to more than one billion people [3]. It is a staple food in many LMICs, especially in SSA, with the annual per capita consumption varying from country to country [4,5].
Most of the consumed calories in East and West Africa are from maize [6]. In Kenya, maize and its products are the staple food to the majority of the population [7], with an annual per capita consumption of approximately 77 kg [8]. Most of the maize harvested is stored in many parts of the country to ensure continuous supply between seasons. Traditional storage methods are mainly used by farmers [9] but have the disadvantage of being susceptible to insects and pest attacks [10], among other unfavourable conditions that might lead to contamination and spoilage.
More than 34 million hectares of maize were grown in SSA in the season of 2014/2015, producing approximately 70 million metric tonnes. Even though the area under maize cultivation in Africa has increased, the production and supply have declined [11]. is has been attributed partly to climate changes and heavy postharvest losses [12]. ese losses have been estimated at 20-30% of the total staple maize harvested [13], valued at approximately US$4 billion annually [14]. e high postharvest losses are associated with poor maize postharvest practices, informal marketing systems, and unfavourable physical and environmental factors. Other factors include pests and fungal attacks [15,16]. In particular, insect pests have been associated with high maize grain damage and losses. Approximately 30% of the losses in maize grains in Africa are a result of postharvest insect pests, which cause nearly 40% weight loss in maize grains [16,17]. e postharvest losses affect food security and food safety situation in Africa. ey contribute to high food prices due to scarcity and hunger [13]. e limited food supply has further escalated many conflicts, armed wars, and political instability in some parts of Africa.
With a lack of proper effective and affordable storage methods, farmers resort to crude storage methods. Maize stored using such methods experience substantial pest damage, and rodent attack [18]. Poor postharvest practices also result in mould growth, dry matter loss in the grains, and reduced grain quality [19].
Maize is also susceptible to fungal infestation, especially of the Aspergillus and Fusarium species. is is common from the period of its growth to harvest, transport, and storage, which exposes it to mycotoxin contamination associated with these fungal species, especially aflatoxins and fumonisins [20]. Maize and its products have been frequently cited to be contaminated with high aflatoxin and fumonisin levels [21,22]. Mycotoxin contamination has been reported to be a major problem in SSA. is is enhanced by poor farming practices, climatic conditions, and postharvest handling practices, which provide a conducive environment for fungal growth and subsequent production of mycotoxins, as well as insect infestation [23]. Mycotoxin contamination is associated with huge losses to the farmers and to human health [24,25].
Mycotoxins are described as "silent killers" as they are hard to detect, and some are extremely toxic to both humans and animals [26] because of the damage they cause to the immune system [27]. e mycotoxins that often occur in maize grain, which are of public health concern, include aflatoxins, zearalenone, deoxynivalenol, fumonisins, and ochratoxin [28,29]. However, in SSA, the most prevalent classes of mycotoxins are aflatoxins and fumonisins [30,31].
ere is a scarcity of information on how local farmers manage their maize after harvest, with only a few studies carried out on the same. In a study in Kaiti district of the Lower Eastern part of Kenya, most farmers were observed to apply traditional practices in determining the time for maize harvesting and subsequent storage. Maize was mainly stored in polypropylene bags in granaries and living houses. is was associated with high cases of rodent attacks, mould growth, and insect damage. Aspergillus species and aflatoxin contamination were high in the maize of farmers who relied on such traditional postharvest practices [32].
With limited data available, this study sought to assess the farmers' knowledge, attitude, and postharvest practices with a view of understanding their role in maize contamination and making policy recommendations to reduce postharvest losses and maize contamination, especially by mycotoxins. e findings of this study will also be useful in informing the design of intervention measures that would create postharvest management awareness among farmers.

Materials and Methods
A comparative cross-sectional study was carried out in Rift Valley and Lower Eastern regions of Kenya. ree counties from each of the regions were selected and interviews conducted among maize farmers. In the Rift Valley region, the study was carried out in Bomet, Nakuru, and Trans-Nzoia counties, which are the grain baskets of the country, while in the Lower Eastern region, the research was conducted in Machakos, Makueni, and Kitui counties that are considered aflatoxin hotspots in Kenya [33].
A random sample of 314 farmers was used in the study, of which 165 respondents were from the Rift Valley region and 149 were from the Lower Eastern region. To be eligible to participate in the study, the respondent had to be a household head above 18 years of age who was a primary decision-maker on household maize storage practices.
A pretested semi-structured questionnaire, divided into three sections, was used to collect data from the respondents. e first section was on respondents' demographic information. e second section was on maize consumption practices, while the last part was on the respondents' knowledge, attitude, and practices on maize postharvest handling and storage. e questionnaires were administered with the help of trained research assistants. e collected data were entered into MS Excel, where cleaning was done. It was then imported into Statistical Package for Social Sciences (SPSS) version 24.0 for analysis. Descriptive and comparative statistics were used for the analysis. For comparative analysis between the two regions, the Chisquare test and Fisher exact test were used for categorical data, such as gender and level of education of respondents, quality of harvested maize, insect control practices, maize storage practices, and other maize postharvest practices. A two-sample t-test was used to compare the means of continuous variables, such as age of respondents, acres of land, quantity of maize taken from storage for consumption that was discoloured, quantity of maize after shelling, and the average maize selling price. A P value of less than 0.05 was considered statistically significant.

Ethical Considerations.
Ethical approval for the study was granted by the Institutional Research and Ethics Committee (IREC) of Moi University and Moi Teaching and Referral Hospital, approval no. FAN: IREC 1829 on 2 March 2017. We obtained permission for the study from the respective county governments through the assistance of the National Commission of Science, Technology, and Innovation (NACOSTI) Research Clearance Permit No. 14093 on 12 th May 2017. Kiswahili language was used to explain the study details to all respondents who spoke and understood it, being a national language. Both oral and written informed consent was obtained from the respondents.

Characteristics of the Farmers.
Of the 314 respondents in the study, 165 (52.5%) were from the Rift Valley region, while 149 (47.5%) were from the Lower Eastern region. e distribution of respondents per each of the region and counties included is shown in Table 1.
e mean age of the respondents was 42 years. Farming was the main economic activity for 233 (74.2%) of the respondents. e median acreage of land owned in the previous year and cultivated land in the last 12 months was 2.0 acres and 1.4 acres, respectively. e demographic information of the respondents is presented in Table 2.

Quantity of Maize Harvested from the Study Sites.
e median quantity of maize harvested after shelling was 6.

Household Maize Consumption.
More than half (62.4%) of the respondents' households consumed more than 30 kg of maize grains in 30 days, while 118 (37.6%) consumed less than 30 kg in the same period. Of the consumed maize, 178 (56.7%) of respondents reported that all of it was from their production, 75 (23.9%) said that half of the consumed maize was from their production, while 61 (19.4%) reported having bought all of the consumed maize. ere was a statistically significant difference in the quantity of maize consumed from own production between the two regions, with many farmers from the Rift Valley consuming maize from their production as compared to those from the Lower Eastern region (X 2 (2) = 55.709, P value <0.001) ( Table 3). e median number of days the maize stayed in storage was 90 days. e mean difference in the number of days the maize lasted in storage for Rift Valley (M = 140.04, SD = 76.73) and Lower Eastern (M = 71.24, SD = 51.69) regions was statistically significant; t (219.82) = 7.981, P value <0.001. e quantity of maize grains taken from storage for human consumption in the month of the study ranged from 0 to 1,350 kg, with a median of 30 kg. Of this maize, only four (1.3%) of the respondents reported that it was of bad quality. e quantity of discoloured maize consumed ranged from 0 to 90 kg, with a mean of 3.6 kg.
Mouldy maize was reported to be disposed by 98 (31.2%) respondents, while 36 (11.5%) use it to feed the cows, 19 (6.1%) feed it to poultry, 7 (2.2%) consumed it, 2 (0.6%) sold it, and 2 (0.6%) used it for brewing. e respondents' beliefs and perceptions regarding mouldy maize are presented in Table 4. ere was a significant difference between the two regions with regard to perceptions on mouldy maize (P value <0.05), except on the perceptions on the consumption of wet/smelling but goodlooking maize (Table 4).

Storage Insect Control Methods.
Chemical insecticides were the most widely used method of insect control by 222 (70.7%) of the farmers, while 102 (32.5%) used sun-drying and 25 (8.0%) used ash. On chemical insecticide use, 108 (48.6%) respondents used Actellic Super, 99 (44.6%) used Actellic Dust, 6 (2.7%) used Skana Super, while 2 (0.6%) used Malathion Dust. Many farmers from the Lower Eastern region used Actellic Dust compared to those from the Rift Valley (X 2 (3) = 29.622, P value <0.001), while Actellic Super was used by a high proportion of farmers from the Rift Valley (X 2 (3) = 53.645, P value <0.001). e differences were statistically significant (Table 5). Less than half, 148 (47.1%), left the maize in the field without covering after harvesting, with more farmers from the Lower Eastern region leaving their maize on cobs in the Only eleven respondents (3.5%) took maize to commercial storage facilities, while the rest stored it at home. e practice of consuming and selling the maize while still green was practiced by 104 (33.1%) of the respondents, with a high proportion of respondents from the Lower Eastern region practicing it (71 (47.7%) vs 33 (20.0%)) (X 2 (1) = 27.025, P value <0.001).

Other Postharvest
All farmers interviewed used sun-drying and airing to dry their maize. One hundred eight (34.4%) respondents practiced storage of maize on dehusked cobs, 52 (26.6%) practiced the storage of maize in husks, and 307 (97.8%) stored maize after shelling. Most of the respondents, 306 (97.5%), reported that their grains were dry last season. For those who reported that their grains were not dry, they said that it was due to poor weather, poor seed quality, and short drying season. e majority, 309 (98.4%), reported that their   International Journal of Microbiology storage facilities were cleaned before storing the new maize grains (Table 6). Among the respondents, 120 (38.2%) knew their maize grains were dry when they were hard to chew, 95 (30.3%) by listening to the grains' sound and measuring its weight, 49 (15.6%) by touching the grains, 17 (5.4%) by observing changes in colour, and 16 (5.1%) when the maize cobs are hit during shelling. Other methods used to know when the maize grains were dry included observation during shelling as reported by 8 (2.5%) respondents, and storage in store for one week as reported by 4 (1.3%), after sun-drying as reported by 2 (0.6%) and during milling as was reported by 2 (0.6%) of the respondents.

Perceptions on Factors Attributable to Maize Grain
Spoilage. Of the respondents, 98 (31.2%) thought that poor soil condition contributes to the spoilage of the maize grains.
Higher proportions of the respondents (88.5%, 89.8%, 91.4%, and 82.2%) thought that bad weather, wetness of the piles of the harvested maize, dampness of the storage place, and harvesting maize earlier than usual, respectively, were the main reasons for the spoilage. A small proportion (16.9%) thought that drying maize longer than usual can lead to maize spoilage.
ere were significant differences in perceptions on poor soils, wet weather, and wetness in maize pile, dampness in the storage places, and the presence of insects and pests in the storage places between the farmers in the Rift Valley and Lower Eastern regions. e proportion of farmers who thought that these caused spoilage of maize grains was higher among Rift Valley respondents than that of the Lower Eastern region (P value <0.05) ( Table 7).

Perceptions on Minimizing Mould
Infestation. Up to 186 (59.2%) of the participants thought that spreading chemical  International Journal of Microbiology insecticides over the grains prior to storage would reduce mould growth, with the proportion being significantly higher in the Rift Valley than in the Lower Eastern region (110 (66.7%) vs. 76 (51.0%)), (X 2 (1) = 7.952, P value = 0.005). Two hundred ninety-three (93.3%) thought that the storage of completely dry maize only would reduce mould growth. ere were smaller proportions of participants who believed that storage of maize in plastic bags (70 (22.3%)), plastic containers (39 (12.4%)), metallic silos (96 (30.6%)), and clay pots (107 (34.1%)) would help minimize mould growth.

Discussion
e low education level among the majority of the respondents contributed negatively to their postharvest practices and, by extension, their lack of awareness of the consequences of maize contamination [34]. Most of the respondents had low income, a likely indicator of the low economic status of most respondents who depended on farming.
Most farmers harvested 90 kg of maize per season. With the per capita maize consumption in Kenya having been reported to be 77 kg [8], the majority of the farmers' production in our study could not meet the requirements for the household consumption per season, using the national per capita consumption as the reference. e majority of the farmers in this study exhaust their maize from the stores by the ninth month after harvest, with only 39.3% having maize in their stores by the next harvest season. is was also the case in a study by amaga-Chitja et al. [35] where the average length the maize lasted ranged from 5.6 months to 8.6 months, showing that the maize is exhausted before the next harvest. In case the maize was exhausted before the next harvest, the farmers bought maize to cater for deficits [35], as was shown in our study. is could be as a result of low maize production. It could also be due to the farmers selling off most of the harvested maize, thus leading to food insecurity.
Some respondents believed that mouldy maize is safe for human and animal consumption. e belief, as mentioned above, was also reported in the AfloSTOP survey carried out   [36]. In Ghana, Akowuah et al.
found that the majority of maize farmers and traders believed that there were no health effects in consuming mouldy maize due to the rigorous cooking process of maize-based foods [37].

Storage Insect Control Methods.
Insect attack has been reported to contribute significantly to the loss of maize grains. In a study by Midega and colleagues, insect pests were reported to result in approximately 40% grain loss, with the maize weevil and grain borer being perceived to be the most dangerous [38]. e majority of the farmers (70.7%) in our study areas practiced some form of insect control with the use of chemical insecticides, while 32.5% used sun-drying and 8.0% used ash. However, the effectiveness of traditional insect control methods such as the use of ash is not known and needs to be evaluated further.
Our study found out that the use of chemical insecticides was the primary insect control method followed by sundrying. is was different from the study by Midega et al., who found out that in Kenya, aeration/sun-drying were the leading insect mitigation practice in their study areas, as was practiced by 88.8% of the study participants [38].
While chemical insecticides were the most widely used storage insect control method in our study areas, their effectiveness has been shown to reduce with the period of storage [38]. In the study by Midega and colleagues, the addition of chemical insecticides kept the insect damage at 25% during the first four months. However, by the sixth month, the damage increased up to 80% in the chemical insecticides added maize. e damage was more in insecticides added maize stored as grains than those stored in the unshelled form [38].

Other Postharvest Maize Storage Practices.
Postharvest maize management practices, which include drying, cleaning, and storage, among others, are essential and critical along the maize value chain. Farmers dry their maize either during storage or when the maize is still in the field. Kaaya and Kyamuhangire [39] reported that any delays in maize drying while in the field could result in losses during the period of storage. e current study findings are related to what was found in the AfloSTOP survey, where the farmers in both Rift Valley and Eastern parts of Kenya dried maize on cobs for a median of seven days per season [36].
According to Addo and colleagues [40], the majority of farmers in SSA make use of different storage practices, including the use of wooden baskets, jute bags, polyethylene bags, thatched structures, and raised platforms. ese methods of maize grain storage are associated with Fusarium spp. [41] and Aspergillus spp. infestation leading to fumonisin and aflatoxin contamination.
According to Potter and Hotchkiss [42], the storage of maize and other farm produce on the farm is widely practiced in Africa, accounting for nearly 85% of the national storage. However, such storage methodologies are ineffective and are associated with maize contamination with fumonisins, aflatoxins, insect damage, and moulding.
According to the findings of a study by amaga-Chitja and colleagues, most farmers left their maize in the field to dry before harvesting [35]. is was also the case in Ghana, where most farmers heaped and left the maize on the field after harvesting [37]. e process of leaving harvested maize in the field after harvest before storage provides a suitable environment for insect and fungal infestation [15]. Udoh and colleagues [43] reported this to be a widely used practice in SSA, and this has been attributed to labour challenges and the requirement to allow the maize to dry well before it is stored.
Only eleven respondents (3.5%) took maize to commercial storage facilities, with the rest storing it at home. Poor postharvest practices and storage have been reported to play a critical role in the maize contamination with aflatoxin and fumonisins. In the study by Kamala et al., storing maize without the addition of chemical insecticides and drying the maize on bare grounds were significantly associated with aflatoxin and fumonisin contamination [44].
With large quantities of maize consumed while still green, the harvests are significantly reduced. With the small farm sizes of the majority of the farmers, the harvested maize is exhausted early. Consuming or selling the maize while still green leaves the household with low yields which cannot last till the next season. is practice is rampant in Kenya, with media reports in the country highlighting its effects of reducing the final dry maize output and hence impacting negatively on food security, with stakeholders requiring the government to come up with a policy to regulate the practice [45].
As was the case in this study, a study in Tanzania found that most farmers cleaned their storage facilities and cleared it of old maize grain stock before loading them with a new stock [44]. is was also the practice by the majority of the farmers in Guatemala, where 98% of the farmers were reported to clean their maize storage facility before storing freshly harvested stock [46]. According to a study by Mwangi and colleagues, cleaning the stores and drying the maize grains before storage were associated with minimal losses [47].
Similar to the findings of our study, Kamala and colleagues reported that farmers in Tanzania tested for grain dryness by biting or listening to the sound produced by the maize grains [44]. In Ghana, the farmers were reported to check for maize dryness using their teeth by biting [37]. In Guatemala, farmers also used these and other traditional practices, where they made use of fingernail tests (32%), mouth test (16.9%), and a combination of visualization and sound (45.4%) tests to check for maize dryness before storage [46]. Such traditional practices are not accurate and might lead to maize being stored while still having high moisture content, hence making it susceptible to fumonisin and aflatoxin contamination [48].

Conclusion
Inadequate knowledge on better postharvest practices among farmers has been reported to be among the challenges that farmers have to overcome to be able to reduce losses associated with postharvest practices [49,50]. However, training of farmers on the same practices was not associated with reduced losses since trained farmers were noted to experience a similar level of losses as those who did not have any training on the same. ese results point to the fact that farmers might not necessarily apply what they learnt, and this might be attributed to the lack of the required equipment and technologies and the necessary capital [51].
A large percentage of food products are lost during postharvest practices in SSA [8]. Hence, there is a need to invest in better postharvest practices to reduce losses. Furthermore, the demonstration of cost-benefit by putting in place mitigation measures is necessary, and hence the need to provide evidence of the effects of postharvest losses and the quantities involved [52]. Reducing postharvest losses can go a long way in the achievement of food security, hence motivating the farmers' interest in postharvest losses mitigation.
Farmers interviewed had inadequate knowledge of proper postharvest practices, with some of their practices being associated with postharvest losses.
is affects the quality of maize as poor postharvest practices expose the maize grains to contamination that might cause health effects on consumers. is calls for urgent interventions on better maize postharvest management practices.
Data Availability e raw SPSS data used in this study are included within the supplementary information file.

Additional Points
Limitations. e study depended on the respondents' recall of some aspects; a method found to be ineffective as a result of distortion and recall bias.
e study was descriptive, making it practically hard to generalize the findings even though it provides an insight into postharvest practices of farmers that might apply to many parts of the country and beyond.
e survey was carried out during one harvest season. Hence, it might not represent the general practices as they might vary from one season to the next.

Conflicts of Interest
e authors declare that they have no conflicts of interest.