Effects of Disease Prevention Bundles and Delivery on Health in the Horn of Africa– A comprehensive study

Cover picture - UN News

Development Studies: Malaria and Gastro-intestinal diseases in the Horn of Africa Evidence suggests that providing bundles at no cost maximises accessibility, particularly for vulnerable groups like children under 5 and pregnant women. By using Community Health Workers (CHWs) to deliver preventative goods directly to households, the use of fixed facilities as a centre for goods would not be required, it creates trust, increases usage, and improves health outcomes.

Background Healthcare plays an essential role in allowing basic societies to function through the treatment of diseases and the maintenance of health. However, arguably more important than treatment in some regards, is the prevention of ill-health in the first place. This can be achieved through various means, and a prominent method is the use of mosquito nets to reduce the risks of infectious and gastrointestinal illnesses which are incredibly dangerous in hot, humid, equatorial climates. Diseases spread through vectors contribute to more than 700,000 deaths each year, where 95% of malaria induced deaths in 2022 were in the WHO Africa Region. (CDC, 2023) Those most vulnerable to malaria and other vector - borne - diseases include children under the age of 5 who have not yet developed partial immunity to malaria, and pregnant women whose immunity is altered during the pregnancy, especially during the first and second pregnancies. However, there has been a significant decline in multiple vector-borne diseases (UK Health Security Agency, 2023) , (The Royal Society, 2001), including those transmitted by Anopheles funestus, following the desertification of tropical regions, especially in the Sahel and Amazon Rainforest, due to human actions such as deforestation and agriculture. A decline in soil's water retention capacity and evapotranspiration creates a less humid environment, which sterilises the breeding grounds of these vectors The decline in illness following a decline in the population of vectors highlights the importance of them in disease transmission, as evidenced by the decline in the malaria parasite index in children from 40-80% in 1967 to 10% in 1991 and 1992. The substantial decrease in malaria transmission underscores the critical role vectors play in spreading disease, illustrating how environmental changes can directly impact public health by altering vector populations.

Preventative healthcare goods, such as bed nets, are vastly under-consumed. Regions in the midst of development contend with heightened vulnerability to an array of health hazards, including vector-borne diseases, diarrheal afflictions, and exposure to pathogens in contaminated water sources, in contrast to their counterparts in well-developed nations. The under-consumption of healthcare materials can be attributed to a range of factors, with psychological barriers and monetary constraints being the main drivers. People's reluctance to travel and the effort required to collect merit goods constitute significant impediments. Despite the advantages associated with such goods, the prevalence of quasi-hyperbolic discounting significantly influences decision-making processes, leading to less investment in long-term gains and increased spending on short-term benefits.

A proposed solution to this problem of under-consumption emerged while reading Esther Duflo and Abhijit Banerjee's "Poor Economics." They identified a discrepancy where people acknowledged the benefits of using healthcare materials but did not act on that knowledge. This gap between awareness and action suggests that financial and psychological struggles hinder individuals from taking necessary steps to achieve good health. The salience of benefits alone had not led to a significant increase in consumption, highlighting the need for strategies that address both financial constraints and psychological barriers.

To tackle this issue, a practical step is to introduce a system for delivering and bundling healthcare materials. This involves combining various goods into a single package sold as a single unit, while optimising the distribution of these bundles.

1. Introduction 1.1 Context of Absolute Poverty

Absolute poverty is defined as living on less than $1.90 PPP per day. Diseases significantly contribute to the persistence of poverty traps and exacerbate poverty, particularly in nations that under-consume healthcare goods. Vector-borne diseases and the consumption of contaminated water significantly hinder individuals' ability to work, resulting in a loss of income. This issue is particularly severe in low-income countries, where the absence of adequate social safety nets, such as sick leave and unemployment benefits, worsens the financial strain on affected households. Consequently, families may be forced to deplete their savings, if available, or resort to borrowing from predatory lenders to meet their basic needs.

1.2 Impact of Diseases on Poverty The adverse effects of disease on households in the absence of adequate welfare infrastructure can be illustrated by examining the impact of COVID-19 on highly developed countries such as the UK (M. Brewer, 2021) , (M Brewer et al., 2020). To measure the effectiveness of the welfare state in protecting household incomes, it is important to understand the extent to which the UK was affected by the pandemic due to its status as a developed nation. In 2021, the UK ranked among the top five countries globally for deaths per million and was in the top twenty for the highest number of infected cases (HM Treasury, 2020). The government implemented strict lockdown measures, resulting in household net income decreasing by an estimated average of 6.9%. To calculate the effectiveness of welfare policy, post-pandemic statistics can be compared to pre-pandemic levels of household net income. A 100% effectiveness rating would correspond to a return to pre-pandemic levels of household net income.

The changes in social security specifically included a £20 per week increase in the standard allowance of Universal Credit (UC) and Working Tax Credit (WTC). Additionally, the Job Retention Scheme (JRS) and the Self-Employment Income Support Scheme (SEISS) helped prevent sharp falls in household income. However, it is unreasonable to assume that household income changes are equivalent across all deciles of pre-pandemic income levels. Those with higher household incomes were affected more severely than those with lower pre-pandemic incomes. For ease of analysis, this paper will address the average change in household income in all working households between February 2020 and May 2020. Overall, net household income fell by roughly 7%. Without a social security net, households might have lost about 30% of their net income, indicating that these policies were around 77% effective in returning household net income to pre-pandemic levels. This highlights the importance of social security nets, which are often insufficient in developing nations. The parallel drawn between the UK during COVID-19 and developing nations arises from the rampant effect of the virus even in a cold climate, akin to the rapid spread of diseases like malaria in hot equatorial climates. Losses of income are not mitigated, and households bear the full brunt of unemployment and productivity declines due to illness. These effects can be illustrated by examining the impact of malaria on rural small farmers (RSF) and agri-business/industry labourers (AB/IL) in Kenya, excluding urban middle-income workers (C. Leighton et al., 1993). The losses experienced by urban middle-income workers are proportionately less, relative to their annual income, than those experienced by RSFs and AB/ILs. Assuming annual household incomes for RSFs and AB/ILs are 20,000 KES and 35,000 KES respectively, malaria results in a loss of 1,920 KES (£12 PPP) and 660 KES (£3.97 PPP), respectively. Including healthcare costs, these households may lose around 17.6% and 5.25% of their earnings, respectively, with similar figures reported in Nigeria and other sub-Saharan nations. Rampant inflation due to the pandemic and conflicts, such as Russia’s invasion of Ukraine, exacerbates these impacts by increasing food prices, which account for 20% of global caloric intake (World Gain, 2024)

Poor families in developing countries spend approximately 50-80% of their incomes on food. Consequently, when food prices increase, these families have less money available for savings and are often compelled to use their savings to cover essential expenses such as medicines and food (H. J. Brinkman et al., 2009).

The lack of savings is a critical issue for many in developing communities and even in the developed world. Without savings, families have no buffer against unexpected events, forcing them to borrow from loan sharks with rates of 50% APR or higher, as seen in the UK following the pandemic (BBC, 2023). High interest on loans increases the amount borrowers must repay, leading to psychological stress and reducing disposable income. This prevents low-income families from planning for the future and making investments, keeping them trapped in poverty.

1.2 Overview of Healthcare Consumption

Besides the effects of poverty on welfare and the cycle of poverty in developing nations, this paper focuses on the under-consumption of healthcare goods. This issue is evident in various unsuccessful campaigns aimed at promoting the use of chlorine tablets and bed nets. For example, door-to-door marketing efforts in India have led to some sales, but often these purchases were driven by guilt rather than genuine intent to use the products. As a result, the items were frequently never utilised, undermining the campaigns' effectiveness in improving public health (Royal Economic Society, 2018). Despite these challenges, there have been successes through co-selling more well-known products alongside health goods. This approach leverages a positive framing effect and an understanding of consumer behaviour, resulting in approximately double the free chlorine content detected in households' drinking water. Given the success of projects that focus on behavioural elements, the remainder of this paper will explore other methods to encourage the consumption of these health-promoting goods (S. Ahmed et al.,2011).

1.3 Hypotheses Hypothesis 1: The use of bundles to combine disease preventing commodities will increase consumption of them and therefore reduce the likelihood of disease in rural communities in the Horn of Africa Hypothesis 2: The use of a community health worker (CHW) led delivery system of the bundles will contribute to increasing uptake of the disease preventing technologies and therefore reduce the likelihood of disease in rural communities in the Horn of Africa 2. Challenges in Healthcare Access 2.1 Socioeconomic Factors Affecting Healthcare Access Rural communities in developing countries generally have less access to healthcare compared to their urban counterparts. This disparity arises from factors such as cost, educational gaps, and distance from medical facilities. A study conducted in Kenya revealed that 23.3% of participants did not seek healthcare during illness. Among these, 42.8% resorted to self-medication by purchasing non-prescription drugs from shops or pharmacies (A. Ngugi et al., 2017) , (A. Mehmood et al., 2016). Another 20% avoided healthcare facilities due to high costs or lack of funds, 9.9% considered their illness not serious enough to warrant a visit, and 8.1% found the nearest facility too far away or lacked transport

2.2 Transport as a Barrier to Seeking Healthcare Travel costs, especially in terms of time, pose a significant burden for the 8.1% who need to travel for healthcare. They forgo wages during this time, further diminishing their motivation to seek medical care. This travel time represents an opportunity cost, which could otherwise be spent on income generation or educational activities vital for lifting families out of poverty. Additionally, responsibilities such as childcare and household maintenance often take precedence over healthcare, as these are viewed as non-negotiable (J. Taber et al., 2014).

In contrast, developed nations benefit from well-established transport infrastructure, including rail, buses, and other public transport options. Unfortunately, such infrastructure is often limited to urban centres in developing countries. Furthermore, inadequate road infrastructure in rural areas can make travel difficult, whether that’s via walking, bike or car, discouraging the 8.1% from seeking necessary medical services due to distance and travel challenges.

2.3 Illness as a Barrier to Seeking Healthcare in Rural Areas The inability to access healthcare can be exacerbated by illnesses that cause immobility. Many infectious diseases prevalent in impoverished equatorial regions, such as Dengue fever, Chagas disease, Onchocerciasis, Schistosomiasis, Lymphatic Filariasis, and Polio, fall into this category. These vector-borne or sanitation-related diseases are preventable. As discussed earlier, prevention is crucial. Measures such as insecticide-treated bed nets (ITNs) and chlorine tablets can significantly reduce the risk of these diseases, helping individuals avoid the struggles associated with immobility and access to healthcare.

Figure 1: Regions in the Horn of Africa with high malaria prevalence (red) and poor infrastructure (blue). Overlapping areas indicate where healthcare delivery faces the greatest challenges. Adapted from (Mapsland,n/d)

3. Health Goods Utilisation 3.1 Overview of ITNs and Chlorine Tablets 3.1.1 Vector Resistance to Insecticides Pyrethroid and deltamethrin are insecticides commonly used to treat nets to kill vectors. However, more than 100 mosquito species worldwide have developed resistance to one or more insecticides, including 56 species of Anopheline and 39 species of Culicine mosquitoes. This resistance threatens the effectiveness of ITNs in preventing the transmission of infectious diseases (L. Smith et al., 2016), (N. Liu et al., 2006). Outside of insect resistance, the combined insecticidal and irritant effects of pyrethroids, along with the physical barrier of the bed net, have been found to reduce vector density, sporozoite rates (the first form of the malaria parasite entering the human body, with the rate being the number of mosquitoes infected with sporozoites divided by the total number of mosquitoes examined, expressed as a percentage), malaria parasite prevalence, disease incidence, and all-cause child mortality.

The success of ITNs in achieving these goals has led to the transportation of 300 million ITNs to Africa at a cost of more than US$1 billion for their purchase and distribution. While there is controversy over the effectiveness of foreign aid, this paper focuses solely on the effectiveness of prevention methods, including the use of ITNs, bundles, and delivery services

3.1.2 ITNs’ Impact on Malaria Transmission and Child Mortality Although mosquitoes are gradually developing resistance to pyrethroids, numerous studies have demonstrated that ITNs remain effective in protecting against malaria transmission in areas with pyrethroid-resistant mosquito populations when properly deployed (K. Lindblade, 2015). In Malawi, household ownership of at least one ITN increased from 27% in 2004 to 55% in 2012. The use of ITNs by children under the age of 5, one of the most vulnerable groups, contributed to a decrease in national malaria parasite prevalence from 43% in 2010 to 28% in 2012. In regions where deltamethrin killed only 38% of An. funestus (the main malaria vector) and 53% of An. gambiae s.l. (the secondary vector), ITNs reduced the incidence of malaria by 30% among children aged six to 59 months, even with the vectors' resistance to deltamethrin. Beyond individual protection, high ITN usage rates can suppress malaria transmission community-wide, benefiting even those who do not use bed nets through herd immunity. These findings demonstrate that ITNs continue to be effective in preventing malaria infections, even in areas where pyrethroid-resistant An. funestus is the dominant vector (M. Kilama et al.,2014).

The continued effectiveness of ITNs in the presence of insecticide resistance may be attributed to the sublethal effects of pyrethroids. For instance, increased irritancy upon contact with treated netting can reduce a mosquito's ability to find and feed on a host or incubate malaria parasites. Additionally, in the absence of insecticides, vector competence may decline, particularly affecting the survival of insecticide-resistant mosquitoes (SW Lindsay et al., 2021). While ITNs are still effective, their ability to kill mosquitoes and prevent disease transmission is declining due to rising resistance among mosquito populations. Understanding how mosquitoes are developing resistance, particularly to pyrethroids, is crucial. This knowledge could lead to new approaches for deploying ITNs to maintain their effectiveness despite these challenges.

3.1.3 Effects of Faecal Water Contamination and Diarrhoea While ITNs can mitigate the risk of infectious diseases spreading through vector transmission, humans have a need for water. In developing communities without access to a sanitary freshwater well or a pipeline, water may be taken and used for consumption and bathing in unsanitary static water sources where pathogens are able to grow and diversify. Contamination of water is not only at the source, however, can occur during transportation or when being stored at home or in the village.

In Africa and Latin America, studies indicate that a village child may have as many as 6 to 10 bouts of diarrhoea a year, each lasting an average of 3 days. Diarrhoea can cause death, and is also a significant contributor to malnourishment in the children that survive (J. Gasana, 2002). The primary source of microbial contamination is human faeces, and children may come into contact with these harmful agents indirectly, often through contaminated drinking water, bottle formula, or weaning foods. In Rwanda, diarrheal diseases rank as the second leading cause of morbidity after malaria with more than 89% of children affected by diarrhoea being between the ages of 0 and 2, and the greatest effects among those aged 6 to 11 months, likely due to this being the critical period when weaning begins (Palintest, 2020) , (Food Standards Scotland, n.d) , (K. Sankaranarayanan, 2014).

NGO guidelines suggest that 10-100 Colony Forming Units (CFU - a measure of viable colonogenic cell numbers in CFU/mL) per 100 mL is tolerable for short periods of time, but above 100 CFU per 100 mL the risk to life is too great (New Brunswick, n.d). With faecal coliforms, the WHO suggests that the CFU/100mL in irrigation water for crops to be eaten raw should be less than 1000 and for drinking water to be safe CFU/100mL should be at 0. Despite this, an average was taken from three different counties across Kenya, total coliforms per 100mL in boreholes were 982 and 1203 in river water which are both dangerous levels of total coliforms to human health when consumed (A. Onyango et al., 2018) , (R. Abila et al., 2012) , (M. Osiemo et al., 2019)

Elevated coliform levels in water are associated with a range of severe health implications, including typhoid fever, cholera, bacillary and amoebic dysenteries, various gastrointestinal disorders, respiratory illnesses, peripheral vascular diseases, and gangrene (A. Bhargava et al.,2003)

3.1.4 Methods of Water Treatment and Practicality The effects of fecal coliforms on human health are great, therefore it is important for those in developing communities to use chlorine tablets to disinfect their water and to ensure that it is safe for use and consumption. However, to determine the correct method of disinfecting it is important to find the main factors influencing the presence of faecal bacteria pollution in groundwater systems in developing countries. The design of the wells and their maintenance, the distance to latrines and the geological structure of the water points are the most significant variables affecting the presence of E. coli and total coliforms. The presence of fecal bacteria in these areas correlates negatively with the concentration of ion Na+ (being an indirect indicator of fast recharge in the study site). The negative correlation between the presence of Na+ ions and fecal coliform levels suggests that tubewells with high concentrations of fecal coliforms are typically located in areas with stagnant waters and slow recharge rates. In such cases, chlorine dosing becomes a viable alternative to continuous chlorination systems, with chlorine being capable of reducing total coliforms to <1/100mL (A. Rahman et al., 2019). Unlike continuous chlorination systems, which are costly and unsuitable for bundling with ITNs, chlorine tablets present a feasible solution. This limitation reduces the appeal of bundles to communities dependent on wells with fast recharge rates, thereby restricting the effectiveness and reach of these health-promoting bundles (A. Habib et al., 2023).

3.1.5 Optimal Dosage of Chlorine to Prevent Negative Side-Effects

Chlorine tablets are typically designed for one tablet to treat a litre of water. Thus, it is crucial for communities to accurately measure the volume of water in their wells to avoid over- or underdosing. Incorrect dosing can lead to significant health risks, either by leaving the water insufficiently treated and still contaminated, or by introducing excessive chlorine, which can be harmful to human health as chlorine can combine with naturally occurring organic matter in the water to form disinfection byproducts (DBPs) which are unhealthy after regular, long-term exposure (N. Ferrer et al., 2020) , (Minnesota Department of Health, 2022). In rural developing communities, wells used for water are often hand-dug, employing PVC or bamboo strainers to create a clear path to the surface. These shallow wells, which are typically either strainer wells or cavity wells, can have their volume be approximated to that of a cylinder, πr2 h, where r is the radius of the well and h is the height of the water column (B. C. Mal et al.,1986). This method is simple, allowing communities to estimate the well's volume for appropriate tablet dosing in order to avoid negative health-effects. To ensure higher likelihoods of success in sanitising well-water, communities should use chlorine tablets as the primary disinfectant while pre-settling water and ensuring safe storage post sanitation. For protozoa which is somewhat resistant to chlorine, households may boil after disinfection to ensure the bacteria dies. If boiling is not feasible, they should allow treated water to sit for a longer time, which can improve chlorine’s efficacy against some protozoa. 3.2 Use of Healthcare Goods 3.2.1 Problems of Free-ITN Distribution While chlorine tablets and ITNs are effective in reducing the transmission of disease, they still come at a cost, each will respectively be discussed. There is a history of ITNs being delivered without charge due to foreign aid, however this hasn’t led to universal usage by those who the ITNs are available to which is partially due to the communities’ psychology. The free distribution of nets has not necessarily led to the actual use of them. While free, or subsidised distribution has caused greater equity of ownership across social-classes, ITN ownership does not directly correlate with proper utilisation within a household (P. Moscibrodzki et al., 2018). Without adequate education, simply providing free nets may not suffice to achieve the 80% household coverage recommended by the WHO for effective community-wide malaria protection.

3.2.2 The Misuse of ITN’s

There is also the problem of the misuse of nets (while not a particularly large problem), which can be solved through information campaigns to help communities understand how to properly use ITNs for the potential health impact to be fully realised. It has been speculated that the free distribution of mosquito bed nets leads to greater net attrition than in areas where nets were attained via other means (H. Koenker et al., 2014) - with nets from a free distribution campaign being six-times more likely to be given away. In more marginalised communities also, nets may be used for alternative purposes with documented cases in coastal Kenya, for drying fish near Lake Victoria, fishing in the Tamatave region of Madagascar and for fishing in Lake Tanganyika.

3.3 Bundling as a Strategy to Increase Consumption

3.3.1 Tailoring of Bundling Services This leads onto the discussion of bundling services, where households will receive information within each bundle about the effective uses of chlorine tablets and bed nets in order to prevent misuse. The leaflets especially must be localised to each region where these practices would be implemented, this would ensure that they are culturally relevant and available in local languages. This adaptation process should also consider the specific cultural practices and health behaviours of the target communities, meaning that surveys and door-to-door questionnaires should be conducted beforehand in order to tailor these leaflets to each community.

3.3.2 Effectiveness of Bundles in Increasing Consumption

Bundling is a strategy that has the potential to be effective as it simplifies decision-making for individuals and reduces the mental burden of selecting and purchasing multiple items separately. It also encourages their combined use, making it easier for individuals to adopt these health practices without the need and use of flawed heuristics. Although there is a limited amount of research on the impacts of bundling for social purposes, its effectiveness can be measured by drawing parallels with disadvantaged populations in developed countries.

3.3.3 Composition and Pricing of Healthcare Bundles

The proposed service to communities would utilise a pure bundling strategy, offering goods exclusively as a bundled package rather than individually (although they could still be bought separately if necessary) (E. Knutsson, 2011), where this bundle would include multiple units of both chlorine tablets and ITNs. By requiring households to purchase the entire bundle, the risk of misuse of the additional nets or tablets is supposedly reduced. Additionally, the bundle would be priced lower than the net cost of the individual ITNs and tablets, providing a psychological incentive for purchase by presenting them as discounted items within an affordable range, or even with a 100% subsidy, as, ‘child mortality is reduced more under free distribution than any cost sharing strategy.’ (Cohen et. al, 2008), where cost-sharing is placing a cost on ITNs (which would be analogous to placing a cost on the bundles). The case for bundles being provided at no cost to the user is the same study by Cohen which focuses on the most vulnerable demographics. ‘... a large share of ITN users is necessary for a substantial externality to take place, we find that cost sharing is again slightly cheaper than free distribution … even free distribution to pregnant women is not enough to generate significant community-wide effects since not all households experience a pregnancy… cost-sharing is at best marginally more cost-effective than free distribution, but free distribution leads to many more lives saved’. This signals towards the bundles being provided for free, alongside the fact that ‘pregnant women who receive free ITNs are no less likely to put them to intended use than pregnant women who pay for their nets’, further reinforcing the fact that bundles should be provided for free. Despite the clear advantages of providing ITNs at no cost, there are also significant benefits to cost-sharing particularly in terms of sustainability within the retail sector for them, Therefore, to account for all possibilities, we will consider the optimal bundle price to promote retail sustainability. The pricing of these bundles would vary based on where they’re being provided, with lower prices set for areas of lower affluence and higher prices in more affluent regions. Additionally, other factors such as access to these goods, transportation costs, and existing healthcare infrastructure must be considered, as they can affect the affordability and efficiency of distributing the bundles. For simplicity, the Horn of Africa will be divided into general groups— capital cities and surrounding regions—where prices will be adjusted according to the Gross National Income (GNI), especially as large percentages of their income come from agriculture. To calculate the optimal price of bundles, the Gross National Income (GNI) of select countries in the Horn of Africa will be used alongside the contribution of the agriculture, forestry and fishing sector to GDP based on statistics from Statista (2023), (2022) and International Trade Administration (2024) - including: Kenya, Ethiopia, Somalia in US$ PPP/annum. This is because of the relatively inverse relationship between GNI and Agriculture as a % of GDP in these countries. Where G(ɑ) = ß/ɑ* presents the inverse relationship between GNI (G as a function of ɑ, where ɑ denotes agriculture as a percentage of GDP and ß is a constant > 0).

Table 1: FOA (2024), Statista (2022 , 2023)

The pricing of bundles in each country will be determined as a percentage of GNI, considering the local cost of living and consumption patterns. ITNs are most effectively priced between 0-10 KES (approximately 0-0.075 USD), as demand within this price range is inelastic, meaning that price fluctuations do not significantly impact demand, as demonstrated by Cohen et al. (2008). The prices should be set at a level that discourages those with less need from purchasing ITNs, while remaining affordable enough for individuals facing credit constraints to access them when necessary. Given that the optimal ITN price is in the range of 0-10 KES, pricing for ITNs within the bundle will decrease in proportion to the distance from the capital city. This approach reflects the increasing dependence on agriculture in more remote areas. For instance, In Ethiopia and Kenya, income from farming constitutes over 60% of total income for both small and larger farms. In contrast, households with more diversified income sources experience reduced food insecurity, as wealthier farmers and urban households are better positioned to invest in alternative income streams. Off-farm self-employment offers the highest return among all income-generating activities (Etea et al., 2019; Woldenhanna, 2019). Furthermore, rural farming incomes are highly vulnerable to climate change and drought, contributing to greater relative poverty in these regions compared to urban areas, where incomes are less dependent on weather conditions. These lower wages in rural areas, partly due to limited income diversification, support the argument for reducing the price of ITN bundles in more remote regions to sustain a viable retail sector, especially in areas with high malaria prevalence. The composition of the bundles will be based on the average size of households in the horn of Africa, and according to Statista (2022), between Ethiopia and Kenya, the average size was 4.81. To ensure full coverage, each bundle will include 3 ITNs under the recommended practice that one ITN should be offered for every 2 household members. ITNs also have a survival time of 22 months.

As for chlorine tablets, it will be based on the amount needed every 22 months, due to the ITN degradation period being on average 22 months. Wells need to be dosed once every 6-12 months however due to sunlight breaking down chlorine and to prepare for any unforeseen events, it will be assumed that they will need to be used once every 6 months. To do this, we will include several chlorine tablets which can last for the 22-month period (approximately 2 years), therefore 2 doses of chlorine tablets will be provided per household, as wells are generally communal.

In order to optimise the price of the bundles we’ll set the variables and budget constraint alongside a price function.

Pi as the optimal price of bundles in the region i di is the distance from the capital city to region i GNIi is the Gross National Income per capita in region i Ai is the contribution of agriculture to GDP in region i ɑ, β, γ are weight coefficients to balance the influence of each variable We will set a constraint

𝑄+ is the quantity of bundles demanded in region i B is the total budget allocated for the distribution and bundle operation We’ll now set the Lagrange function to help us find the optimal price while satisfying the budget constraint with the equation below representing the sum of the prices in all regions, adjusted by the constraint

To find the relationship between the Lagrange multiplier (𝜆) and the demand (𝑄+) in region i we will now differentiate with respect to 𝑃i

Now we differentiate with respect to 𝜆 to attain the original budget constraint

Now we substitute the price function into the constraint

To complete the equation, we need to set values for 𝛼, 𝛽 𝑎𝑛𝑑 𝛾. This can be achieved through estimation.

The value of 𝛼 can be estimated through trial and error. The value of ITNs in each bundle should be a value less than 30 KES (0.23 USD) in order to satisfy the optimal price between 0-10 KES per net. The value of chlorine tablets should be 4.3 KES (0.033), with the reasoning being that in Somalia (the country with the lowest GNI in our sample), the GNI per capita is $620. The poorest quintile of earners might earn $1-2 a day and pricing should be around 0.5-2% of income to ensure affordability of necessities such as food and housing. In total, the value of each chlorine tablets and ITNs combined would be around 34.3 KES (0.27 USD). This is useful as we should have a target price range, which in our case would be a value less than or equal to $0.27. 𝛼 is a multiplier of the sum of all the variables, therefore a change in it affects the overall price. To set the price at approximately 30 KES, we will set 𝛼 to have a value of around 25.

The value of 𝛽 reflects the value of the countries’ purchasing power as higher GNI generally means higher purchasing power and vice versa. Having a low value of 𝛽 of around 0.005 would ensure that for each unit increase in GNI, the price increases by 0.005 KES, relative to Kenya and the surrounding regions’ weak currencies.

The value of 𝛾 measures the contribution of agriculture to GDP, with a higher contribution indicating a stronger reliance on agriculture, potentially affecting the demand for the bundles. Therefore, the value of 𝛾 should be low to ensure that those in highly agricultural economies (which in most cases have lower incomes) are able to afford the bundles. The value of 𝛾 we will set is -0.02, therefore for every percentage point increase in agricultural contribution to GDP, the price decreases by 0.02 KES.

With the new completed budget function:

To find the optimal price

We get 𝑃+ instead of ∑+ 𝑃+ as we’re finding a weighted average price by dividing the total weighted sum B by the total quantity ∑+ 𝑄+ . Using this principle, we can re-arrange the budget function to factor in the price.

This is our final price function, and each variable ca n be inputted to find the optimal price for region i.

To ensure greater accuracy, readers can alter the values of ɑ, β and γ based on different data available to them. 3.4 Delivery as a Strategy to Increase Consumption

3.4.1 Proposed Method of Delivery

Now that we have established a price function tailored to different regions within the Horn of Africa and determined the optimal bundle composition, we can develop a delivery strategy to maximize bundle usage.

Delivering healthcare goods to fixed facilities, while beneficial in some cases, often falls short of maximising community health outcomes. For instance, a study from BMJ Open found that although new primary health centres in Côte d’Ivoire improved antenatal attendance and reduced child mortality, these results were limited by restricted access for rural residents who faced travel barriers (Koné et al., 2022). This underscores that direct community deliveries can more effectively increase healthcare usage, especially in remote areas with difficult access to fixed facilities. We propose implementing a Community Health Worker (CHW) program. CHWs have been effective in increasing treatment uptake, decreasing mortality rates, and preventing unhealthy behaviours while being “inextricably linked” to positive outcomes in patient and health systemsoriented themes (Maes et al., 2014), (Mwai et al., 2013). Local health-workers can deliver healthcare goods to remote, underserved regions, understanding the local culture and building long-term trust in healthcare systems. For non-local workers, training on cultural norms and encouraging positive connections within the community will be crucial to increase healthcare bundle usage. CHWs not only bridge the logistical gap in healthcare delivery but also play a significant role in fostering long-term trust. Involving CHWs familiar with local customs allows for culturally sensitive communication, which builds comfort and helps counter initial hesitancy, ultimately supporting sustained healthcare engagement within the community.

Delivery volumes will be based on the number of households per community. Given poor road infrastructure, alternative transport methods will be used. CHWs, familiar with the terrain, can deliver on bicycles, motorbikes, or on foot, collecting bundles from a central hub to maximise efficiency. Other than delivery, CHWs could play a central role in an information campaign describing the most effective use of ITNs and water purifying tablets to communities. This idea could be used in conjunction with leaflets.

A similar delivery model involves countries like Rwanda successfully using drone deliveries to overcome road infrastructure limitations. Drones were employed to deliver critical medical supplies directly to rural clinics, ensuring consistent stock and timely access, while cutting down on transportation times by 79 minutes and reducing blood expiry by 67% 12 months after the implementation of the service (Nisingizwe et al., 2022). Such models demonstrate how alternative delivery strategies can overcome terrain challenges and build trust within the community by showing that their healthcare needs are prioritised. This approach could be adapted to large-volume deliveries of healthcare bundles in the Horn of Africa.

In summary, the role of CHWs is essential in supporting both patient outcomes and the broader health system. Their work not only improves individual health measures but also reinforces the healthcare system’s capacity to deliver effective services in underserved regions. This dual impact shows that CHWs play a crucial part in not only addressing immediate healthcare needs but also in strengthening the infrastructure that makes long-term improvements possible

Discussion

Essential Interpretation Based on Key Findings

This study demonstrates that replacing travel to fixed facilities to purchase health commodities with community deliveries and bundling of the goods, has the potential to significantly reduce the transmission of vector-borne and water induced gastrointestinal diseases. The most impactful factor within the programme would be the use of CHWs to deliver the bundles and communicate the correct method of using the goods, such as how to correctly dose water purifying tablets without ill side-effects. CHWs would potentially lead to increased health commodity uptake and bridge the cultural gap between the program directors and the program. While it is unclear the direct effect of the program on disease reduction, the reader may attempt to find it using regression analysis.

Main Supporting Evidence

A key finding was that the use of CHWs have been essential to the fight against HIV in sub-Saharan Africa (Mwai et al., 2013). While this research focuses on vector and waterborne diseases such as malaria and E. Coli, the success of CHWs in managing HIV interventions provides a strong foundation for their potential effectiveness in reducing the consequences of other endemic health issues.

CHWs are likely to be successful in aiding the prevention of diseases in the Horn of Africa due to the topography and conditions being similar to those in countries where their success has been demonstrated. The conditions that promote HIV transmission such as socio-economic inequalities and poor access to health facilities often overlap with high prevalence of malaria and waterborne disease. CHWs provide interventions tailored to local needs and are highly adaptable to the terrain through the use of varying transportation methods to reach remote areas. With their strong ties to local communities, the success of CHWs in reducing HIV is transferable to their potential in addressing vector-borne and waterborne diseases. Comparisons to Previous Studies

Findings from Maslove et al. (2009) present the significant barriers to malaria prevention and treatment in rural areas, including a lack of understanding of transmission, logistical challenges, and mistrust in conventional medicines. CHWs, as proposed in this study, directly address these issues by delivering healthcare bundles to households while simultaneously providing information and education to promote the most effective application of prevention measures. For example, the belief that malaria cannot be prevented, or communities’ reliance on traditional medicines, highlights the importance of CHWS in delivering the healthcare bundles while providing community-specific education. The role of fixed healthcare facilities in improving local health is often constrained in developing areas such as our study area, due to inadequate infrastructure, workforce shortages, and resource limitations (Al-Worafi, 2021). These issues, combined with the physical inaccessibility of many facilities, particularly for rural communities, make it challenging for them to rely solely on fixed facilities for healthcare services. In such conditions, the delivery of prevention bundles, directly to households offers a more practical solution. However, as Al-Worafi (2021) suggests, fixed healthcare facilities are essential in the maintenance of a successful healthcare system. In the long term, given improved infrastructure and better resource allocation, these facilities could become more accessible and provide better quality services that extend beyond the preventative measures offered through deliveries. Therefore, while the current limitations make the delivery of healthcare bundles a more practical solution to improve accessibility and increase usage, potential future improvements in fixed healthcare services could reduce the need for bundle provision. Strength and Limitations of the Study

This study demonstrates strength in addressing healthcare disparities in remote regions by focusing on direct delivery of healthcare bundles to households. The proposed system overcomes significant logistical challenges posed by poor infrastructure, ensuring that essential healthcare goods reach those most in need. The pricing analysis provides an optimal solution for ensuring accessibility, while prioritising vulnerable groups such as children under 5 and pregnant women. However, the study assumes that fixed healthcare facilities will remain underutilised, potentially overlooking their future role as infrastructure improves and institutions develop. Additionally, while the findings promote free bundle distribution, the financial sustainability of such an approach should be studied further. Finally, the study primarily addresses logistical and economic barriers, with less focus on cultural factors that could influence the uptake of healthcare goods.

Unexpected Findings

No unexpected findings were observed; rather, the results reaffirm the challenges documented in previous studies regarding the relationship between healthcare commodity distribution and usage.

Summary of Hypothesis and Purpose

This study hypothesised that the bundling and direct delivery of healthcare goods would significantly increase their usage, compared to their collection from fixed facilities. The purpose of this study is to explore the impact of the direct-delivery of healthcare bundles on improving access and usage in less-developed rural communities.

Significance of the Study

This study provides a cost-effective healthcare delivery model for rural regions, improving access to essential goods for vulnerable populations. The findings contribute to the literature on healthcare distribution by identifying the partnership between bundle composition and community-based delivery strategies and provides recommendations for NGOs and governments to optimise healthcare access in these under-developed regions. While the delivery and bundle model shows promise in the short term, its long-term adoption remain uncertain. Further research is needed to evaluate its sustainability, particularly the role of governments and institutions in scaling such programmes. Additionally, the impact of free deliveries and bundles on retail markets should be studied to ensure sustainability in the long run.

Unanswered Questions

While we tried to offer insights into healthcare delivery in rural regions, there are several questions that remain unanswered and need further exploration. Firstly, the long-term adoption and effectiveness of the proposed delivery model remains uncertain. Will households continue to utilise healthcare bundles consistently over time, or will ongoing interventions and incentives be required to ensure the ownership and particularly usage of these goods. Additionally, the role of governments and institutions in scaling and maintain the programme is not fully understood as of now. To what extent can government intervention or public-private partnerships enhance the reach and sustainability of healthcare delivery? Another critical area involves the effects on local economies. How do free deliveries and healthcare bundles impact local retail markets, and what measures can be implemented to ensure that the intervention does not disrupt existing supply chains?

While this study focuses on distribution and usage, the potential behavioural barriers to healthcare consumption such as cultural factors are unidentified. What strategies, aside from the use of CHWs, can further improve trust, cooperation and encourage consistent usage of the bundles in communities where traditional health practices may be seen as superior compared to modern healthcare?

Finally, the use of bundles has potential to increase healthcare usage, however it is uncertain whether the studies in Chile and developed nations pertaining to bundles can have their conclusions be transferrable to those in the Horn of Africa, given the limited literature available. What are the direct effects of bundling goods, on ownership and usage in different regions in the Horn of Africa, and how do these translate to reduced disease incidence? Addressing these questions is essential in understanding how free-bundling and deliveries can improve healthcare access and usage in developing regions, contributing to the reduction of preventable diseases. Future research and collaboration between policymakers, researchers, and local communities will be crucial to improve these approaches and ensure their long-term success. Bibliography

Abila, R., Muthangya, M., Mutuku, E., & Mutati, K. (2012). Physico-chemical and bacteriological quality assessment of shallow wells in Kitui town, Kenya. Journal of Environmental Science and Water Resources, 1(2), pp. 27-33. [Online]. Available at: http://irlibrary.mmarau.ac.ke:8080/bitstream/handle/123456789/2564/Prof%20Abila%201.pdf?sequence=1&

Ahmed, S. M., Hossain, S., Kabir, M. M., & Roy, S. (2011). Free distribution of insecticidal bed nets improves possession and preferential use by households and is equitable: findings from a study. Malaria Journal, 10(357), p... [Online]. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3266224/ Al-Worafi, Y.M. (2024). Healthcare Facilities in Developing Countries: Infrastructure. In: AlWorafi, Y.M. (eds) Handbook of Medical and Health Sciences in Developing Countries. Springer, Cham. https://doi.org/10.1007/978-3-030-74786-2_209-1 Bhargava A., Bouis, H. E., Hallman, K., & Hoque, B. A. (2003). Coliforms in the water and hemoglobin concentration as predictors of gastrointestinal morbidity of Bangladeshi children. American Journal of Human Biology, 15(.), pp. 209-219. [Online]. Available at: https://d1wqtxts1xzle7.cloudfront.net/75724061/ajhb.1014120211204-13986-o58cn3- libre.pdf?1638681499=

Brewer, M. (2021). Did the UK policy response to Covid-19 protect household incomes? The Journal of Economic Inequality, 19(.), pp. 433-458. [Online]. Available at: https://link.springer.com/content/pdf/10.1007/s10888-021-09491-w.pdf Brewer, M., & Gardiner, L. (2020). The initial impact of COVID-19 and policy responses on household incomes. Oxford Review of Economic Policy, 36(S1), pp. 187-199. [Online]. Available at: https://watermark.silverchair.com/graa024.pdf?token=AQECAHi208BE49Ooan9kkhW_Ercy7Dm3ZL_9Cf3qfKAc485y

CDC. (2023). Malaria's Impact Worldwide. [Online]. cdc.gov. Available at: https://www.cdc.gov/malaria/php/impact/index.html#:~:text=According%20to%20the%202023%20World,clinic

Chong, A., Hentschel, J., & Saavedra, J. (2007). Bundling of basic public services and household welfare in developing countries: The case of Peru. Oxford Development Studies, 35(3), p.1. [Online]. Available at: https://www.tandfonline.com/doi/pdf/10.1080/13600810701514894

Cocoon. (n.d). Mosquito nets are an effective way to protect you from bites when traveling and camping. [Online]. cocoonusa.com. Available at: https://www.cocoonusa.com/blogs/news/mosquito-nets-are-an-effective-way-to-protect-youfrom-mosquito

Cohen, J. et al. (2008). Free distribution or cost-sharing? Evidence from a malaria prevention experiment. NBER Working Paper Series. Available at: https://www.nber.org/system/files/working_papers/w14406/w14406.pdf

Etea, B. et al. (2019). Household income diversification and food security: Evidence from rural and semi-urban areas in Ethiopia. MDPI, 11(12), p.3232. [Online]. Available at: https://www.mdpi.com/2071-1050/11/12/3232

FAO. (2012, 2005). Family Farming Knowledge Platform. [Online]. Food and Agriculture Organization UN. Available at: https://www.fao.org/family-farming/datasources/dataportrait/income/en/

Ferrer, N., Folch, A., Maso, G., Sanchez, S., & Sanchez-Vila, X. (2020). What are the main factors influencing the presence of faecal bacteria pollution in groundwater systems indeveloping countries? Journal of Contaminant Hydrology, 228(.), p... [Online]. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0169772219301202#:~:text=The%20study%20reveal Food Standards Scotland. (n.d). Water Testing. [Online]. foodstandards.gov.scot. Available at: https://www.foodstandards.gov.scot/self-assessment-resources/water-testing

Gasana, J. (2002). Impact of Water Supply and Sanitation on Diarrheal Morbidity among Young Children in the Socioeconomic and Cultural Cont. Environmental & Occupational Health, 90(.), pp. 76-88. [Online]. Available at: https://digitalcommons.fiu.edu/cgi/viewcontent.cgiarticle=1003&context=eoh&httpsredir=1&referer=

Habib, A., Cocciolo, S., Haque, A., Raihan, M., & Bhattacharya, P. (2023). How to clean a tubewell: the effectiveness of three approaches in reducing coliform bacteria. Science of The Total Environment, 872(.), p... [Online]. Available at: https://www.sciencedirect.com/science/article/pii/S0048969723005478

HM Treasury. (2020). Impact of COVID-19 on working household incomes: Distributionalanalysis as of May 2020. HM Treasury, 1(1), pp. 2-7. [Online]. Available at: https://assets.publishing.service.gov.uk/media/5f0570673a6f40041295c1a5/Impact_of_COVID-19_on_workin

Kilama, M., Smith, D. L., Hutchinson, R., Kigali, R., & Yeka, A. (2014). Estimating the annual entomological inoculation rate for Plasmodium falciparum transmitted by Anopheles gambiae s.l. Malaria Journal, 13(111), p... [Online]. Available at: https://malariajournal.biomedcentral.com/articles/10.1186/1475-2875-13-111#:~:text=Sporozoite%20rate

Knutsson, E. (2011). Bundling for consumers? Understanding complementarity and its effect on consumers' preferences and satisfaction. Dissertation, 1(.), pp. 1-193. [Online]. Available at:

https://www.diva portal.org/smash/get/diva2:453408/FULLTEXT01.pdf

Koenker, H., Kilian, A., Bell, C. Z., Onyefunafoa, E. O., Selby, R., & Abeku, T. (2014). What happens to lost nets: A multi-country analysis of reasons for LLIN attrition using 14 household surveys in four countries. Malaria Journal, 13(464), pp. 1-10. [Online]. Available at: https://link.springer.com/content/pdf/10.1186/1475-2875-13-464.pdf

Koné S, Bonfoh B, Probst-Hensch N, et al. Impact of newly constructed primary healthcare centres on antenatal care attendance, facility delivery and all-cause mortality: quasi-experimental evidence from Taabo health and demographic surveillance system, Côte d’Ivoire. BMJ Open 2022 https://bmjopen.bmj.com/content/12/1/e054355 Leighton, C., & Foster, R. (1993). Economic impacts of malaria in Kenya and Nigeria. Major Applied Research Paper, 6(.), pp. 45-53. [Online]. Available at: https://pdf.usaid.gov/pdf_docs/pnabs294.pdf

Lindblade, K. A. (2015). A cohort study of the effectiveness of insecticide-treated bed nets to prevent malaria in an area of moderate pyrethroid resistance. Malaria Journal, 13(31), p... [Online]. Available at: https://link.springer.com/content/pdf/10.1186/s12936-015-0554-1.pdf

Lindsay, S. W., Kleinschmidt, I., & Thomas, M. B. (2021). Threats to the effectiveness of insecticide-treated bednets for malaria control: Thinking beyond insecticide resistance. Lancet Global Health, 9(.), pp. 1325-1331. [Online]. Available at: https://www.thelancet.com/pdfs/journals/langlo/PIIS2214-109X(21)00216-3.pdf

Liu, N., Xu, Q., Zhu, F., & Zhang, L. (2006). Pyrethroid resistance in mosquitoes. Insect Science, 13(.), pp. 159-166. [Online]. Available at: https://d1wqtxts1xzle7.cloudfront.net/47493639/Pyrethroid_resistance_in_mosquitoes20160725-32671-5fr

Maes, K., et al. (2014). Listening to community health workers: How ethnographic research can inform positive relationships among community health workers. American Journal of Public Health, 104(5), p... [Online]. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC3987580/

Mal, B. C., & Mishra, A. P. (1986). Low-cost tubewells for developing countries. International Journal for Development Technology, 4(.), pp. 197-203. [Online]. Available at: https://es.ircwash.org/sites/default/files/212.6-3350.pdf

Mapsland. (1972). Detailed map of the Horn of Africa with relief [Map]. Available at: https://www.mapsland.com/africa/horn-of-africa/detailed-map-of-horn-of-africa-with-relief1972

Maslove, D.M., Mnyusiwalla, A., Mills, E.J., McGowan, J., Attaran, A. and Wilson. (2009). Barriers to the effective treatment and prevention of malaria in Africa: a systematic review of qualitative studies. BMC International Health and Human Rights. 9(1), p.26. [Online]. Available at: https://bmcinthealthhumrights.biomedcentral.com/articles/10.1186/1472-698X-9-26

Mehmood, A., Rehman, A. U., & Zaman, M. (2016). Self-medication; An emerging trend. British Journal of Pharmaceutical Research, 14(1), pp. 1-8. [Online]. Available at: https://www.researchgate.net/profile/AsimMehmood/publication/311360501_Selfmedication_An_Emerging_

Mn Department of Health. (2022). Drinking Water Chlorination: Frequently Asked Questions. [Online]. health.state.mn.us. Available at: https://www.health.state.mn.us/communities/environment/water/factsheet/chlorination.html#:~:text=Is% Moscibrodzki, P., Dobelle, M., Stone, J., Kalumuna, C., Chiu, Y. H. M., & Hennig, N. (2018). Free versus purchased mosquito net ownership and use in Budondo sub-county, Uganda. National Library of Medicine, 17(.), p. 363. [Online]. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6191916/

G. Mwai, G. Mburu, K. Torpey, P. Frost, N. Ford, J. Seeley. (2013). Role and outcomes of community health workers in HIV care in sub-Saharan Africa. Journal of the International AIDS Society. 16(1), p.n/d. [Online]. Available at: https://onlinelibrary.wiley.com/doi/pdfdirect/10.7448/IAS.16.1.18586

M. Nisingizwe, P. Ndishimye, K. Swaibu, L. Nshimiyimana, P. Karame, V. Dushimiyimana, J. Musabyimana, C. Musanabaganwa, S. Nsanzimana, M. Law (2022). Effect of unmanned aerial vehicle (drone) delivery on blood product delivery time and wastage in Rwanda. Research Gate.10(4), pp.564-569. [Online]. Available at: https://www.researchgate.net/publication/359248373_Effect_of_unmanned_aerial_vehicle_drone_delivery_

Ngugi, A. K., Agoi, F., Mahoney, M. R., Lakhani, A., & Mang'ong'o, D. (2017). Utilization of health services in a resource-limited rural area in Kenya: Prevalence and household-level factors. PLoS One, 12(2), p... [Online]. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5328402/

Numbeo.com. (2024). Cost of Living in Somalia. [Online]. Numbeo. Available at: https://www.numbeo.com/costofliving/country_result.jspcountry=Somalia#:~:text=A%20family%20of%20four%20estimated,lowe

Palintest - Water Analysis Technologies. (2020). Total Coliform Testing. [Online]. Palintest.com. Available at: https://www.palintest.com/parameters/totalcoliforms/#:~:text=Total%20Coliforms%20In%20Emergency%20W

P. Freeman, H.B. Perry, S.K. Gupta, B. Rassekh. (2009). Accelerating progress in achieving the millennium development goal for children through community-based approaches. Global Public Health. 7(4), pp.400-419. [Online]. Available at: https://www.tandfonline.com/doi/pdf/10.1080/17441690903330305

Royal Economic Society. (2018). Getting poor people to take up preventive health products. [Online]. res.org.uk. Last Updated: 28 March 2018. Available at: https://res.org.uk/mediabriefing/getting-poor-people-to-take-up-preventive-health-products/

Smith, L. B., Kasai, S., & Scott, J. G. (2016). Pyrethroid resistance in Aedes aegypti and Aedes albopictus: Important mosquito vectors of human diseases. Pesticide Biochemistry and Physiology, 133(1), p. 12. [Online]. Available at: https://pubmed.ncbi.nlm.nih.gov/27742355/ Statista.com. (2022). Average household size in Africa as of 2022, by country. [Online]. Statista. Available at: https://www.statista.com/statistics/1228286/average-household-size-in-africa-bycountry/

Statista.com. (2022). Contribution of agriculture, forestry, and fishing sector to the Gross Domestic. [Online]. Statista. Available at: https://www.statistacom.bris.idm.oclc.org/statistics/1265139/agriculture-as-a-share-of-gdp-in-africa-by-country/

Statista.com. (2023). Gross National Income (GNI) per capita in Africa in 2023, by country. [Online]. Statista. Available at: https://www.statistacom.bris.idm.oclc.org/statistics/1290903/gross-national-income-per-capita-in-africa-by-country/

Taber, J. M., Leyva, B., & Persoskie, A. (2014). Why do people avoid medical care? A qualitative study using national data. General Internal Medicine, 30(3), pp. 290-297. [Online] Available at: https://link.springer.com/content/pdf/10.1007/s11606-014-3089-1.pdf

The Royal Society. (2001). Early effects of climate change: do they include changes in vectorborne diseases? [Online]. ncbi.nlm.nih.gov. Available at: https://www.c/pmc/articles/PMC1088500/pdf/TB011057.pdf

Trade.gov. (2024). Somalia - Country Commercial Guide. [Online]. ITA. Available at: https://www.trade.gov/country-commercial-guides/somalia-agribusiness-andfood#:~:text=Somalia's%20agricultural%20sector%

UK Health Security Agency. (2023). Health Effects of Climate Change: The health threat from vector-borne diseases. [Online]. GOV.UK. Last Updated: 11 December 2023. Available at: https://ukhsa.blog.gov.uk/2023/12/11/health-effects-of-climate-change-the-health-threatfrom-vector

Woldenhanna, T., et al. (2001). Income diversification and entry barriers: Evidence from the Tigray region of northern Ethiopia. Food Policy, 25(4), pp. 351-365. [Online]. Available at: https://www.sciencedirect.com/science/article/pii/S0306919201000094?casa_token=ArXg8KvtuxQAAAAA:EFOi

World Gain. (2024). Study details Ukraine war's wheat market impact. [Online]. Worldgrain.com. Last Updated: 14 February 2024. Available at: https://www.worldgrain.com/articles/19623-study-details-ukraine-wars-wheat-market-impact

World Health Organization. (2023). World Malaria Day 2023 – zeroing in on malaria. [Online]. emro.who.int. Last Updated: 2023. Available at: https://www.emro.who.int/somalia/news/worldmalaria-day-2023-zeroing-in-on-malaria.html

pmi.gov. (2024). Ethiopia Malaria Operational Plans. [Online]. U.S. PMI. Available at: https://www.pmi.gov/wp-content/uploads/2024/04/FY-2024-Ethiopia-MOP.pdf Statista.com. (2022). Average household size in Africa as of 2022, by country. [Online]. Statista. Available at: https://www.statista.com/statistics/1228286/average-household-size-in-africa-bycountry/

UK Health Security Agency. (2023). Health Effects of Climate Change: the health threat from vector-borne diseases. [Online]. GOV.UK. Last Updated: 11 December 2023. Available at: https://ukhsa.blog.gov.uk/2023/12/11/health-effects-of-climate-change-the-health-threatfrom-vector

Woldenhanna, T., et al. (2001). Income diversification and entry barriers: Evidence from the Tigray region of northern Ethiopia. Food Policy, 25(4), pp. 351-365. [Online]. Available at: https://www.sciencedirect.com/science/article/pii/S0306919201000094?casa_token=ArXg8KvtuxQAAAAA:EFOi

World Gain. (2024). Study details Ukraine war's wheat market impact. [Online]. Worldgrain.com. Last Updated: 14 February 2024. Available at: https://www.worldgrain.com/articles/19623-study-details-ukraine-wars-wheat-market-impact

World Health Organization. (2023). World Malaria Day 2023 – zeroing in on malaria. [Online].emro.who.int. Last Updated: 2023. Available at: https://www.emro.who.int/somalia/news/worldmalaria-day-2023-zeroing-in-on-malaria.html