Climate-Sensitive Health Outcomes in Kenya: A Scoping Review of Environmental Exposures and Health Outcomes Research, 2000–2024

Abstract

Climate change threatens health and social development gains in Kenya, necessitating health policy planning for risk reduction and mitigation. To understand the state of knowledge on climate-related health impacts in Kenya, a scoping review of 25 years of environmental health research was conducted. In compliance with a pre-registered protocol, nine bibliographic databases and grey literature sources were searched for articles published from 2000 to 2024. Of 19,234 articles screened, 816 full texts were reviewed in duplicate, and a final 348 articles underwent data extraction for topic categorisation, trend analysis, and narrative summary. Most of the studies (97%, n = 336) were journal articles, with 64% published after 2014 (n = 224). The health topics centred on vector-borne diseases (45%, n = 165), primarily vector abundance (n = 111) and malaria (n = 67), while mental health (n = 12) and heat exposure (n = 9) studies were less frequent. The research was geographically concentrated on the Lake Victoria Basin, Rift Valley, and Coastal regions, with fewer studies from the northern arid and semi-arid regions. The findings show a shift from a focus on infectious diseases towards broader non-communicable outcomes, as well as regional disparities in research coverage. This review highlights the development of baseline associations between environmental exposures and health outcomes in Kenya, providing a necessary foundation for evidence-informed climate change and health policy. However, challenges in data and study designs limit some of the evidentiary value.

1. Introduction

The recent convergence of increasing extreme weather events, along with a better understanding of attribution to anthropogenic climate change, has sharpened global attention to the impact of the environment on human health, wellbeing, and livelihoods. This is particularly evident in sub-Saharan Africa (SSA) which experiences adverse effects of climate change, despite minimal contributions to global greenhouse gas emissions [1]. Kenya is highly susceptible to climate change due to its varied topography, diverse climatic zones, and reliance on natural resources and is the largest economy in East Africa with a growing population of 2.3% that is projected to nearly double by the turn of the century [2].

Currently, Kenya stands at an inflection point toward rapid industrialisation and sustainable growth, with the potential to provide a blueprint for climate-resilient economic development in the region. Kenya was the first country in Africa to enact legislation exclusively on climate change via the 2016 Climate Change Act [3], which sets out pathways towards sustainable development through the National Climate Change Action Plan. This plan in turn advises on mechanisms of integrating sectoral climate change mitigation and adaptation actions at national and sub-national levels [4].

Various environmental exposures (EEs), including weather, hydrometeorological hazards, and air pollution, pose risks to social development gains due to their influence on human health [5]. However, many of these causal relationships have not been clearly defined in Kenya. This lack of geographically specific, long-term trend analyses and climate attribution research limits the ability to capture the burden of disease attributable to anthropogenic climate change. Accordingly, these evidence gaps have critical implications for the development of evidence-informed climate change and health policy, including international climate finance negotiations. To better understand the state of environmental health research, and to support a broader project aimed at building evidence for a planetary health approach to policy making in sub-Saharan Africa, a comprehensive synthesis of published output on the influence of EEs on health outcomes (HOs) in Kenya was undertaken. This was performed with the following objectives: (a) to undertake a scoping review of the literature on relationships between EEs and HOs; (b) to map the links between these exposures and climate-sensitive HOs and health equity through bibliometric analysis, topic mapping, and narrative synthesis; and (c) to identify knowledge gaps and future research needs to strengthen the evidence base underpinning climate change and health (CCH) attribution for policy development. This article summarises the topic and trend analysis and narrative findings from twenty-five years of environmental health research in Kenya.

2. Materials and Methods

2.1. Protocol and Registration

The reporting of this review was guided by the PRISMA extension for scoping reviews [6] and was conducted by an established team of article reviewers, a library information professional, and subject matter experts in medicine, veterinary medicine, nutrition, demography, water, sanitation and health, and child health and development. A scoping review was deemed most suitable given the wide breadth of the subject matter and lack of similar reviews, as well as broader project aims [7]. The protocol for this review was registered on the Open Science Framework on 14 April 2023 [8].

2.2. Eligibility Criteria

We aimed to identify a wide range of original literature describing the relationships between EEs and HOs. A full list of EEs is found in Appendix A,

Table A1. Environmental exposure topic categories and their corresponding subcategories.

Topic Category	Topic Subcategories
Climate Change and Seasonality	Climate change
	Seasonality
	Solar radiation
Ambient Temperature	Excess heat
	Excess cold
	Temperature variation
Rainfall and Humidity	Rainfall
	Humidity
Wind Characteristics	Wind speed
Flooding and Water Characteristics	Flooding
	Water level change
	Water quality parameters
Drought	Drought
Air Pollution	Particulate matter
	NOx, SO2, Ozone
	Cooking fuel emissions
	Lighting emissions
	Second-hand tobacco smoke
	Other air pollutants
Terrestrial and Water Pollution	Organic matter contamination
	Inorganic compound contamination
	Microbial contamination
	Plastic pollution
Topography	Soil type and moisture
Climate-Associated Land Use Change	Deforestation
	Urban development
	Habitat and vegetation change

. Given the various effects of land use change and pollution on health, these environmental drivers were also incorporated. The eligible health outcomes encompassed direct and indirectly impacted outcomes, such as heat stroke and vector-borne diseases (VBD), respectively, and are inclusive of wellbeing outcomes (

Table 1. Health outcome topic categories and their corresponding subcategories.

Topic Category	Topic Subcategories
Injury or Death	Trauma Death	Burden estimate
Heat Exposure and Skin Conditions	Heat stress
	Skin conditions
Cardiovascular, Circulatory, and Respiratory Disorders	Heart disease and circulatory disorders Lung and airway conditions	Respiratory infections
Waterborne Diseases and Water Access Disorders	Cholera Diarrheal diseases Leptospirosis Cryptosporidiosis Schistosomiasis	Giardia Dehydration and kidney disorders Water insecurity Harmful algal blooms
Vector-borne Diseases	Vector or parasite abundance or prevalence Malaria Dengue Trypanosomiasis Lymphatic filariasis Leishmaniasis	Tick-borne diseases Soil-transmitted helminths Yellow Fever Chikungunya West Nile Virus
Zoonoses	Animal reservoir abundance or zoonotic disease prevalence Anthrax Coxiella burnetti	Bartonellosis Rift Valley fever Brucellosis
Malnutrition and Foodborne Diseases	Stunting Wasting Malnutrition	Food insecurity Escherichia coli and Salmonella
Mental Health Conditions	Eco-anxiety and depression Stress and resilience	Cognitive capacity
Adverse Birth or Pregnancy Outcomes	Neonatal or infant outcomes
	Maternal health outcomes
Health Equity Research	Neoplasia Pre-existing conditions Displacement and migration Occupational hazards	Health vulnerability Awareness and Perceptions Conflict Gender-based violence Access to health facilities Health inequities

).

We included original research published in English between 1 January 2000 and 20 February 2023 and updated the search on 13 August 2024. This timeframe reflects the growing interest and discourse on the health implications of climate change as well as improvements in environmental impact assessments in public health sciences [9]. Eligible studies were required to include some measure of an HO produced by either qualitative or quantitative analysis. As our review focused on Kenya, we included studies on any demographic populations, including transborder pastoralist communities, as well as global studies, if the data from Kenya was disaggregated and extractable.

2.3. Information Source

The search strategy was informed by the population–exposure–comparator–outcome (PECO) model [10]:

• P: Population of Kenya;
• E: Environmental exposures, including weather, hydrometeorological hazards, and air quality variables;
• C: No effect of environmental exposures on health conditions (as available, studies will not be excluded for lack of comparison groups);
• O: Disease burdens or measures of association or effect of environmental exposures on health outcomes.

We included synonyms for HOs guided by categories listed in the World Health Organization report, “Quality Criteria for Health National Adaptation Planning” [11]. The full methods, search terms, and database search results were conducted by a library information professional and are hosted in an open access digital repository maintained by the London School of Hygiene & Tropical Medicine [12]. Nine bibliographic databases were searched: Medline, Embase, Global Health, Food Science and Technology Abstract and Econlit via OvidSP, GreenFile and Africa-Wide Information via EBSCOhost, Clarivate Analytics Web of Science core content, and Scopus. Grey literature sources included Google Scholar and websites of 10 organisations known to be working on environmental or CCH research in Kenya. The results from the database search were compared for Kenya versus five East African countries (Tanzania, Uganda, Somalia, Sudan, and Ethiopia); given that the single largest proportion of studies originated from the former (40% of the search results), the study was narrowed to Kenyan research results only.

2.4. Selection Process

All citations were deduplicated and transferred into the reference manager software EPPI-Reviewer Web 4.14.2.0 [13] for two-stage screening. Title and abstract screening was conducted by a single trained reviewer against a priori inclusion criteria. All reviewers were trained on an initial sample of 800 abstracts in duplicate to ensure consistency. Following primary screening, full-text articles were reviewed in duplicate for eligibility in the initial search, with the updated search results screened by a single experienced reviewer. The reviewers recorded the reason(s) for the exclusion of any articles.

2.5. Data Extraction

Data extraction was undertaken by two independent reviewers, and confirmation of the final extracted data was arbitrated by a third reviewer; data extraction from the literature included in the updated search was performed by a single experienced reviewer. Articles were categorised by main HO (Table 1) and EE (Appendix A, Table A1). The main categories were further divided into subcategories that were refined using an iterative approach during data extraction. Data on publication year; author institutional affiliation; study type; article type; funder(s); location(s); and analysis method(s) were recorded. To investigate the extent of research funding available for climate and health studies, data on six funding models was extracted, categorisation of which can be found in Appendix A,

Table A2. Multi-selectable bibliometric characteristics of included literature. Table A2 summarises health outcome categories, methods of analysis, and funder type. Articles could be allocated more than once within each category.

Characteristic	Included literature (n)
Health Outcome Categories	431
Vector-borne Diseases	165
Health Equity Research	80
Waterborne Diseases and Water Access Disorders	39
Cardiovascular, Circulatory, and Respiratory Disorders	29
Malnutrition and Foodborne Diseases	36
Zoonoses	26
Adverse Birth or Pregnancy Outcomes	19
Injury or Death	16
Mental Health Conditions	12
Heat Exposure	9
Methods of Analysis	663
Descriptive statistics	214
Parametric and non-parametric statistical tests	138
Regression analysis	145
Advanced modelling methods	63
Qualitative methods	55
Time-series analysis	38
Health risk assessment	7
Meta-analysis	3
Funder(s)	402
International funder(s)	246
No funding cited	61
International university funding	48
Kenya Government	28
University funding	11
Private/local funder	8

. Analytical methods for each article were assessed through study design, and brief summaries of the EE and HO results from each study were tabulated.

2.6. Evidence Synthesis

We used topic categorisation, trend analysis, and narrative summaries to synthesise the evidence. To explore characteristics of environment and health research in Kenya, we mapped locations of empirical research on HOs to Kenya’s main climatological zones [14], used temporal frequency in topic categorisation for key trend mapping, and developed a Sankey diagram to illustrate environmental drivers of HOs using R software version 4.4.2. [15,16].

2.7. Protocol Amendments

We applied one protocol amendment to our review. The amendment was made to the HO categories informed by the Quality Criteria for Health National Adaptation Plans [11]. We added the category “Adverse Pregnancy or Birth Conditions” to reflect gender disparities in HOs and expanded equity-related subthemes under the category “Health Equity”.

3. Results

The bibliographic database search identified 33,494 records, of which 14,051 were duplicates removed prior to screening (Figure 1). The grey literature search identified 77 texts and six records from cited references. The resulting 19,234 unique references underwent title and abstract screening, with 18,419 removed. Of these, 816 underwent full-text screening and 348 met inclusion criteria for the final article set (Table S1). A total of 38 full-text reports, including 19 conference abstracts, could not be retrieved.

3.1. Article Characteristics

The characteristics of the final set of articles are shown in Figure 2 and Figure 3A,B, and Appendix A, Table A2 and

Table A3. Single selectable bibliometric characteristics of included literature. Table A3 summarises publication year, institutional collaboration, study type, and geographic scale. Articles were allocated once within each category.

Characteristic	Included Literature (n)
Year Published	348
2000–2004	21
2005–2009	39
2010–2014	64
2015–2019	100
2020–2024	124
Institutional Collaborations	348
Kenyan	59
African Region	19
International	198
International without African collaborations	72
Study Type	348
Observational	183
Modelling Study	76
Mixed methods	26
Trials *	24
Qualitative	36
Meta-analysis	3
Geographic Scale	348
National	63
Regional	195
City/community	78
Not specified/relevant	12

* Includes both randomised and non-randomised trials.

and include both single and multi-selectable characteristics, as specified. The study set is primarily comprised of journal articles (96.7%, n = 336). Research output on the environmental determinants of health in Kenya steadily increased by an average 26% per 5-year period from 2000 to 2024. Most study designs were observational in nature (52.6%, n = 183), while modelling, qualitative, and randomised and non-randomised trials constituted 21.8% (n = 76), 10.3% (n = 36), and 6.9% (n = 24), respectively (as shown in Appendix A, Table A3). The methods of analysis reflected the main study designs, with most studies using multiple modes of analysis (n = 663) (Appendix A, Table A2). These studies included 21.9% regression analysis, 9.5% advanced modelling methods, 8.3% qualitative methods, such as interviews and focus groups, and 1.1% applied health risk exposure assessments.

Analysis of research funders found that most funding came from international funders (n = 246, 61.2%), with the remaining funders (n = 47) cited as international universities (n = 48) or Kenyan governmental (n = 28), university (n = 11), or private funders (n = 8) (Appendix A, Table A2). Sixty-one studies did not cite a funding source, the largest proportion of which were studies from research teams based solely in Kenyan institutions.

3.2. Key Trends in Health Impacts Assessments

To track research interests and methods, temporal stacked bar charts were produced to map trends in health topics and study designs in the study set (Figure 3). An almost six-fold increase in overall research output occurred between 2000 and 2024. The focus on infectious diseases, which dominated output between 2000 and 2019, has declined, giving way to a greater share of research on non-communicable diseases since 2020 (Figure 3A). Despite this recent decline, VBD research was nonetheless the most common health outcome researched overall, peaking between 2015 and 2019 (n = 44). Research concerning the intersection between zoonoses and climate has had a relatively low representation in Kenyan literature but similarly peaked in 2015–2019 (n = 12), declining recently (2020–2024, n = 8). Interest in the role of climate drivers of waterborne diseases and malnutrition has also increased since 2020 (Figure 3A). Additionally, research intensity on topics of health equity, such as occupational health, awareness of the climate threat to health, and gendered vulnerabilities, has risen sharply, as seen by a 650% growth rate from 2005–2009 to 2010–2014. Figure 3B illustrates, amongst other things, the growth in qualitative research methods and modelling studies over the study period.

3.3. Environmental Drivers of Health Outcomes in Kenya

Occurrences of environmental covariates of HOs were extracted from 348 articles (Figure 4). Rainfall represented the most frequent exposure studied in relation to a health outcome (n = 168) and was specifically investigated as a driver of VBD in 123 occurrences within the article set. Other exposures linked to VBD included temperature (n = 123), habitat change (n = 94), and seasonality (n = 63)—the latter of which was frequently defined as a categorical variable that identified Kenya’s rainy and dry seasons and was the third-most studied exposure (n = 103) after temperature (n = 134). Climate change was most often defined as a non-specific exposure term and frequently used in surveys that evaluated awareness of participants of climate change impacts on health (n = 26). Less studied EEs by frequency of relationship included plastic pollution (n = 1), water level change (n = 5), and deforestation (n = 6).

The most frequently studied HO was VBD with 407 occurrences linking different VBDs to an environmental exposure, followed by research relevant to health equity (n = 165). Health outcomes, including heat stress (n = 14), maternal health outcomes (n = 11), access to health facilities (n = 6), eco-anxiety, and depression (n = 7) were less frequently studied in relation to an EE. A full listing of all the studies with summarised environmental health relationships is available in Table S1.

3.4. Research Locations in Kenya

The geographical distribution of empirical research with sub-location data and based on climatological zones in Kenya is illustrated in Figure 5 [14]. This figure shows health outcome results extracted from studies that cited a location for data collection or findings. As some studies report findings from and for multiple climatological zones and health outcomes, the geographic distribution refers to number of research locations and not to the number of articles. The highest density of research occurs in Kenya’s humid southwest region, an area which includes the Lake Victoria Basin, Rift Valley region and Nairobi, and in which 73% of VBD (159/218) and 60% of health equity research (59/99) was conducted. In contrast, the HOs studied in the northeastern region of the northern arid and semi-arid lands (ASALs) of Kenya more frequently focused on malnutrition and zoonoses, areas where pastoralism is a common livelihood (Table S1). Empirical research was least frequently conducted in the southeastern lowlands and northwestern region, the latter being the least research-diverse of all the regions, encompassing just half of all possible outcome categories.

3.5. Health Outcome Summaries

3.5.1. Vector-Borne Diseases

VBDs were the most studied HOs in this review, evaluated in 165 studies, 67% of which focused on vector abundance (n = 111). A range of VBDs were assessed in the literature, including malaria, dengue, yellow fever, West Nile virus and others. These studies explored relationships between EEs and a variety of burden indicators, such as disease risk, incidence, and vector population dynamics. Malaria was the most studied VBD, based on both clinical case reports (n = 67) and vector abundance studies (n = 71), with most of the research conducted in southwestern Kenya (Table S1). There was comparatively less literature available on other VBDs such as dengue (n = 8), soil-transmitted helminths (n = 4), and West Nile virus (n = 3) and only a single result each for trypanosomiasis, leishmaniasis, and tick-borne diseases and two for lymphatic filariasis and yellow fever.

3.5.2. Health Equity Research

The largest subtopic of health equity was the respondents’ awareness and perceptions of the risks of climate change to health using qualitative methods (n = 57). Demographic and social determinants were frequent elements of these studies, especially air pollution exposure studies in urban informal settlements [18,19,20,21,22] and studies on gender vulnerabilities to indoor air pollution [23,24]. Gendered access or lack thereof was identified in educational attainment and accessibility to health facilities [25,26]. Two studies explored gender roles and violence in view of climate change, identifying a potential exacerbation of inequalities and harmful practices like female genital mutilation and intimate partner violence [26,27]. Water scarcity-driven conflicts in pastoralist communities were also explored in three studies [28,29,30].

3.5.3. Waterborne Diseases and Water Access Disorders

Thirty-nine studies investigated waterborne diseases and water access in Kenya. Warming temperatures both increased and decreased Schistosomiasis transmission depending on the region, while precipitation had a delayed influence on snail vector density [31,32,33,34]. Increased rainfall during the rainy season impacted the cholera risk [35], and seasonality significantly influenced snail abundance and cryptosporidium prevalence [36,37,38]. Studies on diarrheal risk assessed in unvaccinated children under five (n = 3) reported that dry season conditions may drive rotavirus infections [39,40,41].

3.5.4. Cardiovascular, Circulatory, and Respiratory Disorders

A number of studies on cardiovascular, circulatory, and respiratory disorders assessed their relationship with household air pollutants (HAPs) (n = 15), finding that exposure, especially of women and children, over extended periods of time is linked to adverse outcomes [42,43,44,45]. Exposure to HAPs such as particulate matter (PM2.5) and carbon monoxide (CO) was linked to reduced cardiac function [46], and volatile organic compounds from wood smoke are associated with increased self-reported respiratory, eye irritation, and headache symptoms [47]. In households that used firewood or unprocessed biomass, children and infants under the age of five had a greater relative risk of acute respiratory infections (ARI) compared to those using kerosene fuels; long-term exposure to PM2.5 also increased these conditions and symptoms [43,48,49].

3.5.5. Malnutrition and Foodborne Diseases

Approximately 9% of studies included in this review evaluated the impact of climate variables on nutritional deficiencies (n = 30), frequently measured in relation to drought or changes in precipitation or through proxy measures such as the normalised difference vegetation index. In northern Kenyan, temperature was positively associated with malnutrition [50], and the impact of drought was inversely correlated with validated growth measures, such as middle-upper-arm-circumference, height-for-age, and weight-for-age Z-scores in children [51,52].

3.5.6. Zoonoses

Two studies assessed how EEs impact wildlife host density and habitat suitability in relation to zoonotic disease [53,54], as well as interactions between precipitation and land use change on rodent host density [54]. Climate risk factors for Rift Valley fever (RVF), a zoonotic virus that causes disease in livestock and haemorrhagic fever in humans, were evaluated in sixteen studies through empirical and modelling methods applied to livestock, humans, and mosquito vectors.

Rainfall abnormalities, vegetation change, and measures of humidity were associated with abundance and suitability of potential habitats for RVF vectors [55,56,57] and also impact vectors and host susceptibility [58].

3.5.7. Adverse Birth or Pregnancy Outcomes

The impacts of EEs on maternal and infant health were investigated in nineteen studies. Eight articles evaluated precipitation, seasonality, and drought and the associated effect on anthropometric measurements of nutritional status [52,59]. Disease prevalence in children under five years of age, assessed in six studies, was characterised by seasonal patterns of occurrence; the diseases investigated included Rotavirus, Escherichia coli, Shigellosis, cryptosporidiosis, and other Enteropathogens [37,39,40,41]. Two additional studies investigated the incidence of infant ARIs, including human metapneumovirus [60] and respiratory syncytial virus [61].

3.5.8. Injury or Death

Sixteen studies evaluated death or injury due to trauma linked to EEs, along with studies that measured disease burden, using standard mortality and morbidity metrics. The relationship between temperature and premature mortality was assessed using years of life lost [62,63], burden estimates of morbidity [64,65,66], and disability-adjusted life years (DALYs) [67]. DALYs were also used in four studies to estimate the overall burden of HAPs on health [18,67,68,69].

3.5.9. Mental Health Conditions

Mental health conditions were notably understudied (n = 12) in the article set. These studies evaluated the impacts of extreme weather events and climate shocks on economic status, mental wellbeing, and psychological distress [70,71,72,73].

3.5.10. Heat Exposure and Skin Conditions

Nine studies investigated the impacts of temperature extremes on mortality measures and found significant positive associations between exposure to low temperature and mortality in Nairobi populations [62,63,74,75]. An additional two studies reported on vulnerability indices [76] and hydration in agropastoralists [77] and a single study described podoconiosis distribution and risk prediction in relation to different scenarios of environmental suitability [78].

4. Discussion

Our review compiled literature on EEs and HOs in Kenya over a twenty-five-year period, and in doing so, identified broad trends and emerging gaps in evidence. The research on environmental determinants of health in Kenya has seen a near six-fold increase in output between 2000 and 2024, with studies increasingly accounting for social determinants, occupational and gendered health impacts, and other equity considerations. Even so, geographical disparities remain, with fewer than half of all studies undertaken in the climatic-fragile northern regions of the country. With meteorological projections indicating warmer temperatures, changing rainfall patterns, and increasing flood and drought events [79,80], the findings of this review are relevant to researchers and policymakers aiming to establish the climate risks to Kenya’s population health.

4.1. Research Trends

We identified a diverse range of research on environmentally mediated HOs, with the highest proportion of literature on VBDs, particularly studies that evaluated clinical malaria and malaria vector abundance. In comparison, funding in the previous decade for environmentally mediated neglected tropical diseases (NTDs) pales in comparison to other diseases; examples of this can be seen with VBDs such as Trypanosomiasis, Leishmaniasis, and Lymphatic filariasis, highlighted in this review as having minimal pertinent evidence [81,82,83]. Given international funding uncertainties at present, there is a risk that underfunded NTDs will remain understudied and insufficiently controlled [81,82]. In addition, a comparatively low amount of literature used a One Health approach to evaluate interconnected relationships between human, animal, and environmental health pathways. This may be due to a lack of data, difficulties in establishing ecological dependencies, and challenges in the use of integrative and multisectoral approaches to support sustainable control efforts [83].

Most of the literature in this review was published by collaborations of international authors and supported by international funders, raising questions about inequities in global health research and funding structures in Kenya, particularly given the prominence of Kenyan actors in CCH research in Africa [84]. More equitable funding and support for Kenyan-produced research outputs, utilising models similar to the Advancing Research for Climate and Health (ARCH) initiative for East African regional climate–health research hubs, could drive contextually relevant research agendas and better amplify local voices [84,85,86].

4.2. Shifting Narratives

Our evaluation of key themes in environmental and health research in Kenya suggests a broadening of interest over the last decade towards social science, health equity, and policy research from more conventional environmental health topics such as VBDs. This shift reflects a larger pattern of change towards improvements in attribution methods alongside cross-disciplinary approaches, including those in social sciences, occurring in public health research [87,88,89]. Subjects relevant to health equity were identified in several articles focused on vulnerable populations, including pollution exposure studies of informal settlements, water scarcity conflicts in pastoralist communities, and extreme weather effects on intimate partner violence, cognitive development, and education access for young girls. A shift was also seen towards studies which explored contextual experience of CCH pathways, assessed through qualitative techniques. Perceptions of how climate change impacts health provide insights into the need for improved public health promotion in Kenya and underscore calls for targeted capacity building as well as climate change training for healthcare workers and the wider public health sector [90].

4.3. Research Gaps

Mental health conditions were understudied in our article set and limited in scope, despite growing global evidence of linkages to both extreme weather and slow-onset climatic factors, including rising ambient temperature and drought [91,92]. Similarly, heat exposure was understudied, despite climate model projections that indicate that parts of Kenya, alongside other SSA countries, will experience the greatest increase in heat stress days globally [76].

Recent research from West Africa has confirmed a link between heat impacts and adverse birth outcomes related to foetal strain [93]; however most research on this topic is based on data from high-income countries, possibly due to scarcity of temperature ground monitoring and health data [94]. Given the high fertility rate in Kenya and relatively high rate of neonatal mortality [95,96], the low volume of gender-oriented articles points to an important research gap. Accordingly, we found a need to amend health outcome categories that were based on a WHO framework on climate-sensitive health risks to specifically include adverse birth and pregnancy outcomes [97], given the recognised gendered inequities in climate impacts on health [98].

There was a gap in research on malnutrition and foodborne disease, a principal risk factor of deaths and disabilities in Kenya, despite evidence that climate change contributed to the Horn of Africa drought that caused 20,000 excess child deaths in 2022 [99,100]. In infants and children, malnutrition can have long-lasting health impacts which is relevant to Kenya’s young population [51,101]. In the northern ASALs of Kenya where pastoralist communities reside, there is a need for greater exploration of the impacts of worsening drought patterns [102]. These EEs can cause secondary effects in communities in the form of tribal conflict over water, food, and livestock resources or gendered violence [26,30]. Ongoing conflict and insecurity in northern areas may further exacerbate the geographic disparities seen in the research output, resulting in a much higher density of publications in the southwestern region, home to nearly 90% of the population in an area of less than 20% of the Kenyan land mass [2,102,103].

4.4. Limitations

As environmental health research is a rapidly growing area of study, our search terminology aimed to encompass a wide variety of exposures and outcomes in line with the WHO-defined pathways. We aimed to minimise missing articles by using broad terminology in our search and screening criteria, incorporating a grey literature search, and using recognised frameworks for health categorisation.

A key limitation was that this scoping review did not set out to conduct a risk of bias assessment and therefore cannot evaluate the quality of evidence, in line with the methodological framework for scoping reviews. However, some constraints and potential sources of bias were notable in the review and data extraction process. For instance, seasonality was a frequent variable used to define Kenya’s rainy and dry seasons and was often applied to differentiate timing of sampling. However, a number of studies applied seasonality as a categorical variable (i.e., samples taken during the wet or dry season) rather than being based on climate data, potentially introducing misclassification bias given daily weather variability. Similarly, the high number of identified studies reporting vector abundance, often used as a proxy for infection and risk, may misinform exposure and outcome relationships and require caution when interpreting. Moreover, some studies did not present statistical analysis of temporal health outcomes preventing meaningful and unbiased assessments of relationships between exposures and the onset of health outcomes. Overall, there were also few studies that incorporated long-term (decadal) health datasets, precluding the trend analysis necessary for health impact attribution to anthropogenic climate change. Lack of climate attribution research restricts the opportunity for evidence-led negotiations and litigation for loss and damage financing [104]. Confounding remained a concern given the nature of the climate–health relationships. These limitations highlight the need for more rigorous methodological approaches to be applied in future studies.

5. Conclusions

This review provides a baseline analysis of the scale and scope of evidence describing environmental impacts on health in Kenya. Its diversity illustrates the wide range of health pathways that have been studied in Kenya and identifies trends in institutional collaborations, funding patterns, and research priorities. Greater attention needs to be paid to vulnerable groups, geographical disparities in research, and the complex relationships between environmental determinants and less frequently studied HOs to ensure equity of the growing research. Targeted capacity building, funding reform, and enhanced support for local and regional institutional networks are necessary steps to build the evidence base and safeguard population health in the face of Kenya’s changing climate.