Attachments
By Helen Young, Sophie Ochola, Anastasia Marshak, Elizabeth Stites, Achiba Gargule, Elijah Odundo, Ayane Ezaki, Josue Awonon
Executive Summary
The United States Agency for International Development (USAID) Nawiri longitudinal study (LS) in Isiolo and Marsabit Counties in northern Kenya investigates the factors behind the stubborn persistence of child global acute malnutrition (GAM) rates that surpass the 15% emergency threshold. Across the two years of study, on average, wasting prevalence varied from a minimum of 15% (September 2023) to a maximum of 21% (November 2022), with an overall average of 18.6% (95% CI: 18.0–19.1%). These findings correspond to earlier trends identified in an analysis of malnutrition hotspots, indicating long-term, structurally embedded drivers of child wasting (acute malnutrition with no edema). Despite this alarming outlook, our review of seasonal variability in child wasting shows that achieving a prevalence below the emergency threshold and even below 10% is possible, as evidenced by the significant seasonal reductions in two of the four study sites.
The study investigated the drivers of child acute malnutrition—immediate, underlying, and basic—and their temporal distribution (seasonality and change over time) across four sentinel sites in Isiolo and Marsabit Counties. The sites correspond to the predominant livelihood system in each of the locations: Laisamis (pastoralism) and Loiyangalani (fisher pastoralism) in Marsabit County, and Garbatulla (agro-pastoralism) and Ngaremara (mostly mixed peri-urban with pastoralism) in Isiolo. Bimonthly surveys were conducted from September 2021 to September 2023, with extended annual surveys conducted in rounds six and twelve. Qualitative research teams made five rounds of visits to the sites to provide deeper insights into key themes and investigate quantitative findings.
This report follows the different levels of the malnutrition causal framework—from basic, to immediate, to underlying drivers of child acute malnutrition (see Figure 1 in the PDF). The findings begin with a review of the basic drivers to contextualize the sentinel sites, a critical step for interpretation of both the nutritional outcomes and immediate and underlying drivers covered in the report's sections. Understanding how the basic drivers differ helps explain the variations in what factors are driving or mitigating child acute malnutrition in each site and across time.
The basic drivers: contextualizing child acute malnutrition
This study provides unprecedented insights into the basic drivers of child acute malnutrition in the Kenyan ASALs and contextualizes and explains the consistently high rates of child malnutrition. The complex results reveal seasonal and spatial differences linked to livelihoods, climate, the environment and—most importantly—how these factors interact with institutions that are foundational to resilient livelihoods.
Pastoralism is vital but under pressure
Livestock and pastoralism remain the bedrock of most livelihoods even as the system of pastoral production continues to evolve. Pastoralist systems are adapted to manage extreme climate variability. While one or even two poor rainy seasons may not overstretch the system, the increasing frequency of droughts and consecutive failed rains, combined with multiple external pressures, has placed increasing strains on these systems. Growing pressures and constraints—combined with innovations and opportunities—have led to the evolution and transformation of these systems. Privatization of land tenure undermines sustainable resource management, restricts livestock movements, and increases resource competition, while the weakening of customary authority hampers sustainable land use norms. Encroachment of urban settlements and conservation areas further limits access to pastures. Economic development has led to sedentarization but has provided minimal benefits due to limited market access and high transport costs, forcing many into marginal economic activities. The commercialization of livestock has weakened traditional kinship systems and livestock redistribution, increasing vulnerability and undermining social support institutions, which impacts child nutrition and well-being.
Critical linkages between pastoralism and human nutrition
Numerous institutions support the pastoralist system and the dynamic relationships between people, livestock, and the environment; these dynamic relationships underpin positive human nutrition. Strategic mobility is critical to pastoral livelihoods, influencing human nutrition by allowing herders to take advantage of the nutrient distribution in the rangelands, thereby maintaining animal productivity, which translates directly into food security, nutrition, and health benefits for humans, particularly children.
The interaction between permanent settlements and mobile fora also plays a crucial role in managing nutrition, with women and children moving to the fora to access milk and other resources. The study found significantly higher odds that a child was acutely malnourished in the sedentary settlements compared to children in the fora (in Laisamis), illustrating the nutritional benefits of spending time in the fora even during severe droughts.
Social institutions, particularly reciprocity-based networks for sharing and support, are critical for managing human health and nutrition. Childcare by secondary caregivers enables women to engage in labor to cover household needs, while the sharing of milk supports vulnerable households. Extended droughts reduce the productivity of the herd while also increasing the distance between homesteads and the fora, thereby further reducing access to available milk for those in settlements.
Pastoralist institutions have shown remarkable adaptability in addressing challenges through customary environmental governance and herder drought management strategies, which have extended livestock mobility and maintained intergroup relations. Despite the strain on community social support systems, sharing and reciprocity remain prevalent, providing essential food, assistance, and childcare across all sites. These institutions are not only foundational to resilient livelihoods but also serve as the frontline of disaster response.
Livelihood diversification—the pros and cons for child wasting
The shift towards sedentarization and diversification in pastoral systems has created a mix of adaptive, coping, and maladaptive strategies affecting nutrition and resilience. The study highlights site-specific diversification, such as fishing in Loyangalani, farming in Garbatulla, and casual labor in Ngaremara. Some activities—such as fishing—may be both a survival strategy and longer-term adaptation. Diversification differs by gender, with many women taking on labor-intensive, low-return economic activities that increase their workload and time away from young children, negatively impacting caregiving and child nutrition.
The study distinguishes between diversification for survival, such as some casual labor and collection and sale of firewood during drought, and strategic adaptations, such as trade and small business, which spread risk and enhance resilience. Hence, in Loyangalani we see that fishing is protective for acute malnutrition while casual labor is protective in Ngaremara but has the opposite effect in Laisamis. Thus, diversification has varying impacts on child nutrition and so requires careful evaluation to ensure positive contributions.
Pathways to recovery and resilience
Some stakeholders may perceive the crisis to be over with the return of the rains in 2023. For pastoralists, however, rain can bring additional problems and more livestock losses when animals are severely weakened due to protracted drought. The resumption of reproduction and re-establishment of herds is a slow process. This lack of or slow recovery is reflected in the acute malnutrition data: both showing no significant improvement in all the sites except Laisamis after the return of the rains, as well as the continued presence of critical and emergency level rates of GAM. Effective and sustainable recovery requires effective environmental governance, adaptive drought management, and a shift in women’s livelihoods from high-risk, marginal endeavors to sustainable, adaptive opportunities.
The basic drivers of child wasting—climate and seasonality, systems and institutions, and livelihoods—affect entire communities, not just specific households. Both drought and conflict impact food security, water access, and disease across entire communities, not just individual households. Given that these crises increase vulnerability for all, effective humanitarian and preventative action requires a community-wide response strategy. Ensuring the resilience and recovery of pastoralist systems involves a comprehensive approach that addresses immediate needs, supports long-term positive adaptation strategies, and fosters sustainable livelihoods for drought-affected communities.
Nutritional outcomes
Seasonality of wasting
The study reveals that seasonal variability in child wasting is significant, with an average difference of six percentage points between peak and nonpeak seasons, which is greater than the difference between the two years of study. For Laisamis, Ngaremara, and Garbatulla, two general peaks in wasting occur during the rainy seasons (November/December and April/May), with the lowest prevalence in the long dry season (August/September). In contrast, Loiyangalani shows different patterns, with peaks in February and June corresponding to temperature peaks, and the lowest prevalence in September. Each site has distinct primary peaks, with Laisamis showing a significant peak in April/May, Ngaremara having two significant peaks, with November/December being greater than the April/May peak, and Garbatulla showing the least seasonal variability, with one significant peak in November. The presence of two peaks in three sentinel sites and the differences between these sites and Loiyangalani indicates different drivers and protective factors across sites. Understanding these patterns is crucial for developing targeted interventions to address the drivers of malnutrition.
Consistent characteristics of a wasted child—age, gender, and female caretaker nutritional status
The study consistently found that children aged 3–5 years have much higher odds of being wasted than younger children. This was unexpected because global trends show that the older group tends to have lower rates of wasting compared to younger children; however, the most recent Kenya Demographic Health Survey (2022) shows a similar reversal of the expected relationship between age and wasting. Qualitative data from our study suggests that younger children are less likely to experience wasting due to more time with their mothers as compared to older children. Increased women’s workload means older children are often with caregivers who lack sufficient food, thereby raising the malnutrition risk for this group. The accumulation of multiple wasting episodes over time may also explain the higher risk of wasting in older children. Additionally, boys are generally at greater risk of wasting, stunting, and being underweight. The study also underscores the link between the nutritional status of female caretakers and child wasting, highlighting the vulnerability of households with lower caretaker mid-upper arm circumference (MUAC).
A synthesis of the immediate and underlying drivers in each site
Overall, the study found that the immediate and underlying drivers of wasting varied by site and time of year, with the lowest prevalence of wasting consistently occurring during the long dry season. Diarrhea was the only driver consistently correlated with wasting across all four sites, but it was present in only 16% of wasting cases, indicating that most episodes of wasting did not coincide with diarrhea.
In Laisamis, significant seasonal drivers included fever, malaria (not confirmed), diarrhea, and the absence of household camels in the village. Fever and diarrhea consistently follow the same seasonal pattern as wasting. Seasonal patterns suggest malaria (not confirmed) is an important contributor towards the end of the rainy season, and the presence of camels contribute to lower wasting during the dry season.
In Loiyangalani, significant drivers included fever, malaria (not confirmed), diarrhea, open defecation, and lack of cereal consumption. The lowest wasting rates in August/September correspond to reduced food insecurity, and the lowest prevalence of malaria (not confirmed), fever, and diarrhea.
Ngaremara’s significant drivers included diarrhea, lack of cereal consumption, and absence of cattle, with individual caretaker and household characteristics, such as caretaker disability and monogamous as opposed to polygamous households, also being important predictors of wasting.
Garbatulla’s significant drivers included diarrhea, respiratory illness, use of informal water sources, lack of fruit, vegetable, and meat consumption, and household food insecurity, with the consumption of roots and tubers indicating reliance on less preferred, lower-nutrient foods.
Despite improvements since the end of the drought, as of September 2023 most sites still faced critical or emergency levels of wasting, highlighting the need to address not only immediate and underlying drivers but also basic systemic factors contributing to persistently high malnutrition rates.
Conclusions and implications of the findings
The evidence and analysis have significant implications for all stakeholders and potentially reflect deeper failures and missed opportunities in both development and humanitarian systems. The sustainable prevention of child acute malnutrition is a collective responsibility, from the village level to the highest levels of government, and cannot be ignored. The report concludes with implications for a range of stakeholders, summarized below.
1. Implications for assessments, surveys, and surveillance
The study emphasizes the need for disaggregating GAM rates by sex, age, and geography, incorporating additional anthropometric variables, understanding the implications of data aggregation, avoiding simplistic seasonal assumptions, timing data collection appropriately, improving survey representation of households who have migrated to distant pasture (fora), and enhancing variables for assessing food security, livelihoods, nutrition, and health.
2. Implications for effective responses to shocks and seasonal stresses (treatment and prevention of malnutrition)
Recommendations include reviewing age-specific project targeting criteria, extending malnutrition policies to include children up to 5 years, expanding targeting in known malnutrition hotspots, enhancing screening and treatment coverage, using simplified protocols for treating malnutrition, considering the needs of mothers, monitoring relapse rates, and tailoring response strategies to specific community drivers of acute malnutrition.
3. Implications for sustainable livelihoods and adaptive diversification
Global nutritional programs have increasingly focused on food systems. In Isiolo and Marsabit, a food systems approach should build on existing pastoral livelihoods and support strategic, sustainable diversification to enhance nutrition and resilience, while considering community-specific livelihood profiles and ensuring interventions are well-coordinated and contextually adapted.
4. Implications for strengthening systems and institutions
Resilience in pastoral communities is largely due to institutions that support mobility, manage environmental resources, and facilitate social safety nets, all of which are associated with nutritional benefits. However, these institutions are vulnerable to protracted and compounded shocks (e.g., extended drought overlaid with intergroup conflict), necessitating interventions and actions at every level to ensure that these institutions can function effectively.
5. Implications for learning, uptake, and systems strengthening
To drive effective policy and institutional change, evidence must be integrated into an uptake strategy that promotes evidence-based learning and systems strengthening. Dissemination and uptake must be driven by stakeholder interests and concerns. Local dissemination with strong participation is crucial to facilitate debate and priority setting among local development actors and communities to achieve sustainable solutions to malnutrition.
6. Methodological insights and research priorities
Longitudinal data are essential in highly variable climatic contexts, where climate directly influences household activities and exposures. Analyzing climate data is crucial rather than relying on predefined seasons. Additionally, a mixed methods sentinel site approach provides in-depth data on local livelihood systems, institutions, and their wider influences and interconnections. These findings also have methodological implications for how and when we measure wasting.
Further research is recommended on the relationship between fora migration and missing data, the connection between anthropometric indicators and functional outcomes (especially for children aged 3–5 years), and the construction of livelihood variables using quantitative data. Follow-up research should include representative samples of migrating households, as current program evaluations often exclude them, potentially biasing data and misrepresenting ASAL populations.
Conclusions
Child acute malnutrition persists in these sentinel sites, known as malnutrition hotspots, and highlights systemic failures and missed opportunities in development and humanitarian systems, underscoring the collective responsibility for sustainable prevention of child malnutrition. The analysis of basic drivers contextualizes the consistently high rates of child malnutrition, demonstrating seasonal and spatial differences linked to livelihoods, climate, and environment, and shaped by institutions and social support mechanisms. The study’s implications call for a comprehensive response to address these underlying issues.