The mental health impact of the climate and ecological crisis on vulnerable populations in Africa

ABSTRACT

Scientists have been predicting the severe impacts of climate change for decades, and today these threats have intensified, with climate events pushing temperatures, sea levels and biodiversity loss to record extremes. This review examines the impact of climate change on mental health in vulnerable populations residing in Africa. Medline, Embase, Ovid, APA PsycINFO, Global Health were searched for studies published after 2007 researching the psychological impact of climate change on vulnerable African populations. Studies were screened according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and results were reviewed using a narrative synthesis approach. Twenty-one articles, including 36,425 participants, were analysed. The results suggest that women, young people, farmers, migrants, people living in rural areas and survivors of climate-related traumatic experiences (such as extreme weather events) are all vulnerable to mental health impacts caused by the climate crisis. Future research is needed in African countries which have not yet been studied, including island nations. Further high-quality research is required to establish longitudinal impacts of the climate crisis and to explore the impacts on vulnerable groups which are, as yet, neglected in the research, including, e.g., ethnic minorities, indigenous groups and LGBTQ+ communities.

Introduction

Almost 20 years ago, scientists were predicting the global effects of climate change on extreme weather events, water availability, food security and ecosystems, identifying particularly vulnerable continents, including Asia and Africa (IPCC Climate Change, I, 2007). Now, nearly two decades on, the climate crisis has become the global issue which threatens our livelihood, and ultimately humanity’s existence, leading 40 counties to declare a “climate emergency” (Climate Emergency Declaration, 2024).

Despite global recognition of the crisis, action has not been sufficient to mitigate or slow the effects of climate change. The most recent State of the Climate Report indicated that 2023 marked a critical turning point for the planet, with climate trends reaching levels never recorded before in history, including record global temperatures and rising sea levels (Ripple et al., 2023). Climate change is also interrelated to the ongoing ‘ecological crisis’. Across the world, climate changes are accelerating biodiversity loss, species decline and rising rates of infectious diseases (Pfenning-Butterworth et al., 2024). According to our current trajectory, climate change is expected to cause approximately 250 000 additional deaths per year between 2030 and 2050, from malnutrition, diarrhoea, malaria and heat stress alone (WHO, 2021).

Research indicates that climate change impacts both our physical and mental health, with known effects disproportionately impacting countries who have contributed least to the problem (IPCC Climate Change., I, 2023). Physical health impacts from the Lancet Countdown Report include heat stress and heat stroke, exacerbation of cardiovascular and respiratory disease, exacerbation of respiratory symptoms from wildfire smoke and the spread of vector-borne and water-borne diseases following a flood or drought, malnutrition due to food insecurity, increased migration, mortality (Rocque et al., 2021; Watts et al., 2021). The United Nations Framework Convention on Climate Change provides examples of eco-based adaptation strategies in vulnerable groups, such as those implemented by coffee farmers in Costa Rica (Least Developed Countries Expert Group, 2018): “Café de Costa Rica” produces low-carbon coffee as part of Costa Rica’s Nationally Appropriate Mitigation Action (NAMA), using agro-forestry to achieve carbon sequestration and climate adaptation (Namacafe, 2018).

Impact of the climate crisis on mental health

The psychological effects of climate change, as noted by Cianconi et al. (2020), include both direct and indirect impacts, ranging from Post-Traumatic Stress Disorder (PTSD) caused by extreme weather events (Zuñiga et al., 2019) to persistent mental health conditions, such as Depression and Generalised Anxiety (Lawrance et al., 2022). The widely accepted aetiological link between the climate crisis and poor mental health has led to the conceptualisation of “eco-emotions”, including eco-anxiety, eco-guilt and solastalgia. These terms are used to describe instances in which increased awareness about the global environmental crisis of climate change and its consequences have brought upon distressing emotions such as guilt, sadness and anger (Pihkala, 2020). While some authors debate the usefulness of such terms, arguing that the development of these descriptors unnecessarily pathologizes understandable emotional reactions to the climate crisis, which poses global existential threat (Bhullar et al., 2022), others claim that giving these psychological experiences labels is important as it recognises these emotional experiences as unique phenomena, enabling further research and better understanding (Schipper et al., 2024). For the purposes of this review, eco-emotions were not included in the search terms, meaning that papers included in the review measured universally agreed upon diagnoses and conditions, behaviours commonly associated with mental health conditions, e.g. self-harm, and subjective wellbeing outcome measures.

Psychological interventions are in the initial stages of development, for example Lindhe et al. (2023) successfully piloted a randomized controlled trial (RCT) that aimed to improve climate-related mental health distress, addressing depressive symptoms, anxiety, stress, insomnia and pro-environmental behaviour.

Populations living in tropical and subtropical climates are particularly vulnerable to temperature increases. Temperatures rising appears to have a dose–response relationship with mental health difficulties, with a 1°C temperature rise reportedly associated with a 1% increase in the incidence of suicide (Gao et al., 2019) and a significant increase in mood disorders, organic mental disorders, schizophrenia, neurotic and anxiety disorders (Liu et al., 2021).

Global research into the mental health impacts of climate change has identified that groups of people belonging to certain demographics are more vulnerable to developing mental health difficulties than others. For example, White et al. (2023) found that indigenous people, children, older adults and climate migrants were among the most vulnerable populations to expressing solastalgia, suicidality, depression, anxiety/eco-anxiety, PTSD, substance use, insomnia and behavioural disturbances. Indeed, the significant negative impact of the climate crisis on children and young people is consistently reported in the literature (Burke et al., 2018; Hickman et al., 2021; Sharpe & Davison, 2022).

However, worldwide research, such as surveys measuring climate-related mental health difficulties, has historically been biased in reporting data collected mostly from developed, wealthy or industrialized countries (e.g. Galway & Field, 2023; Hickman et al., 2021; White et al., 2023). Despite being the biggest carbon emitters, developed and westernised countries are not yet directly experiencing the effects of the climate crisis in the way that developing countries in the Global South are (IPCC Climate Change, I, 2023).

Mental health impact in Africa

Africa is a particularly under-researched continent in this field (Atwoli et al., 2022; Deglon et al., 2023). In particular, the lack of longitudinal data collected in African countries makes it challenging to understand the cumulative and long-term impact of direct and indirect effects of the climate crisis.

This is problematic, because changes in the climate pose a significant threat to the African population. Specific predicted threats include a reduction in precipitation which will result in droughts, rising sea levels which will impact coastal cities resulting in migration, increased food insecurity with an expected 18.6% reduction in agricultural produce (Serdeczny et al., 2017).

Preliminary research into mental health impact in Africa indicates that flooding, drought, extreme heat and bushfires are all related to increases in mood disorders, trauma- and stressor-related disorders, suicide, emotion regulation difficulties, disturbed sleep, alcohol use, stress and anxiety (Deglon et al., 2023). The present review seeks to build on this research, by focussing more widely on slower and indirect impacts of the climate crisis (e.g. temperature increases, sea-levels rising) as well as incidents of extreme weather events.

There is some suggestion that certain vulnerable populations within Africa are at significant risk of developing mental health difficulties, including those who depend on a stable environment for their livelihood, such as farmers and pastoralists in rural Ghana who report poor mental health related to livestock losses, linked to the climate crisis (Nuvey et al., 2020). Climate-related migration within the African continent is a likely impact of the climate crisis. People living in Africa who are internally displaced, due to droughts or flooding, are likely to experience malnutrition, mental health problems and gender-based violence (Lindvall et al., 2020).

The United Nations Framework Convention on Climate Change further identified young, old poor remote and indigenous communities in developing countries as uniquely vulnerable to the impact of climate change (The Least Developed Countries Expert Group, 2018). Moreover, women, people living in the direct area and people who experience damage to their homes are also at risk of developing PTSD following natural disasters in developing countries, such as earthquakes (Zuñiga et al., 2019). For the purposes of this review, we use the term “vulnerable group” to refer to any population whose lifestyle is threatened by changes in the climate or who is at risk of developing mental health difficulties due to identity or demographic factors, such as age, pregnancy, socio-economic status and LGBTQ+ identity. The authors acknowledge the potentially controversial nature of the use of the term “vulnerable groups” in this context, with some arguing that this terminology risks characterising these populations as passive victims rather than active communities with access to their own resilience, knowledge and capacity to adapt to climate challenges and mental health challenges (Okpara et al., 2016). The authors have chosen to use the term “vulnerable groups” in the hope that this language accurately conveys immediacy and seriousness, highlighting how the climate crisis disproportionately impacts certain communities, thus making it a powerful tool in communication and mobilization.

The aim of this review is to systematically examine the mental health impacts of the climate crisis among vulnerable populations in Africa, including studies exploring direct climate crisis impacts, such as natural disasters, and slower or more indirect impacts, such as increasing heat levels.

Materials and methods

This systematic review was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria (Page et al., 2021). We used a mixed methods scoping review methodology, reviewing both qualitative and quantitative research papers (Baumeister & Leary, 1997; Siddaway et al., 2019). This review methodology was chosen to ensure maximum inclusivity, given the limited research available on this topic in this region. Synthesising both quantitative and qualitative research findings allows for more complete, concrete and nuanced answers to research questions (Heyvaert et al., 2013). The review was registered in advance on PROSPERO, registration number CRD42024445144.

Patient consent statement

This article is a systematic review, meaning that participants were not recruited for the research and therefore consent was not requested. All studies included in the review took informed consent from participants.

Clinical trial details

This review is not a clinical trial and therefore has no associated clinical trial registration number.

Inclusion criteria

1. The paper was written in English.

2. The paper reported on research conducted in Africa, with African residents.

3. The paper reported on research with participants belonging to a vulnerable population (see definition below).

4. The paper reported on the impact of climate disasters (including fires, floods and droughts) and/or climate crisis-related changes in the ecosystem (including extreme temperatures, rising sea levels, biodiversity loss). See Appendix A1 for the full list of included climate impacts.

5. The paper reported on mental health-related outcomes, e.g. changes in diagnoses, symptoms, suicide rates and subjective wellbeing. Search terms were designed to encompass both universally recognized and medicalized mental health indicators (e.g., clinical diagnoses) and behaviours linked to mental health difficulties (e.g., suicide attempts, substance misuse), as well as broader subjective terms such as “mental distress”. See Appendix A1 for full details of search terms designed with the assistance of a librarian.

6. The paper reported on a piece of empirical research (qualitative or quantitative) published in a scientific journal.

7. The paper was published from 2007 onwards (IPCC’s fourth assessment report (AR4), published in 2007, acknowledged the impacts of climate change on human health. This approach was taken in another similar systematic review by Sharpe et al. (2022).

Exclusion criteria

1. The paper reported on the impact of man-made crises, such as war or conflict.

2. The paper did not report novel findings, e.g. narrative reviews, scoping reviews, systematic reviews, meta-analyses, commentaries, editorials, expert opinions, validation studies.

Definition of “vulnerable population”

“Vulnerable populations” is a complex and broad term, which applies to different groups depending on the context. For the purpose of this review, we define “Vulnerable populations” as any group of people who either a) have a livelihood or lifestyle which depends on a stable climate (e.g. farmers, pastoralists, nomadic people) or b) are at risk of experiencing mental health difficulties (e.g. low-income families, children, pregnant women, older adults, people with long-term health conditions, people with pre-existing mental health difficulties, people belonging to LGBTQ+ communities). This definition was adopted and used in a similar systematic review (White et al., 2023).

Search strategies

Papers were identified by searching on the following databases: Medline, Embase, Ovid, APA PsycINFO and Global Health. The search was conducted on 5 September 2024. Search terms, developed via consultation with a research librarian, included keywords such as “Africa”, “climate crisis” and “mental disorder”. Specific mental health diagnoses were selected for inclusion in the search terms based on their frequency of occurrence during the initial scoping searches. Despite their potential relevance, neurodevelopmental conditions such as autism and ADHD were excluded, in order to maintain the focus of the review on mental health presentations. The full search strategy can be found in Appendix A1. Reference lists of previous relevant reviews (Atwoli et al., 2022; Deglon et al., 2023; Sharpe & Davison, 2022), as well as all included studies, were checked to ensure papers were not missed. A manual search was also performed of the 2023 volumes of International Journal of Environmental Research and Public Health, Global Environmental Change, Climatic Change, Journal of Environmental Psychology, Disaster Medicine and Public Health Preparedness, and Environmental Health. Two additional papers meeting the inclusion criteria were identified through manual searching.

Study selection

A PRISMA flowchart (Page et al., 2021) describes the systematic process of study selection. First, all duplicates were removed from the search. All titles and abstracts were reviewed according to the inclusion and exclusion criteria, and a random sample of 20% of papers were double rated. There was excellent agreement between the two reviewers, k = 0.91 (95% CI: 0.780 to 1.034), p < 0.001. Any paper included by either reviewer advanced to the next stage. Manual screening of relevant journals and reference lists took place following this. As shown in Figure 1, 53 papers were retained and included for full-text screening. A random sample of 25% papers was double rated at the full-text screening stage. There was substantial agreement between the two reviewers k = 0.68 (95% CI: 0.288 to 1.064), p < 0.001. Any disagreements about inclusion at the full-text stage were discussed by both reviewers and supervisors for the systematic review were consulted for outstanding differences. For example, Reviewer 2 believed that a paper exploring health perceptions and health-seeking behaviour in climate-driven migrants in Tanzania (Heaney & Winter, 2016) was relevant and should be included. While Reviewer 1 recognised the relevance of the paper, they believed that the paper was not exploring mental health-specific outcomes. After discussion as a team, the reviewers and supervisors agreed that this paper did not fit the inclusion criteria.

Figure 1.

PRISMA flow diagram of study selection process. *Consider, if feasible to do so, reporting the number of records identified from each database or register searched (rather than the total number across all databases/registers). **If automation tools were used, indicate how many records were excluded by a human and how many were excluded by automation tools. Source: Page MJ, et al. BMJ 2021;372:n71. doi: 10. 1136/bmj.n71. This work is licensed under CC BY 4.0. To view a copy of this license, visit

Data extraction and quality assessment

A data extraction spreadsheet was developed and piloted to increase reliability. Data were extracted from each paper meeting inclusion criteria. During the pilot, the reviewers practiced using the table by populating it with data from a small subset of papers (n = 5). Table headings were then refined and clarified based on inconsistencies between the reviewers, enabling greater reliability between reviewers during the data extraction process. Data included number of participants, participant characteristics (sex, age and vulnerability characteristics), climate exposure, mental health outcome measures, design and method of the study, and the results. Fifty percent of the studies extracted were independently checked for accuracy. Due to the heterogeneity of the studies, a meta-analysis was not deemed appropriate, and data were instead reviewed using narrative synthesis (Baumeister & Leary, 1997). Narrative synthesis was chosen for this review to accommodate the diversity of study designs, methodologies and outcome measures present in the existing literature. Additionally, this method facilitates the inclusion of both qualitative and quantitative data, providing a more comprehensive and contextually rich understanding of the topic. Data were synthesised according to the methodology used, i.e. quantitative or qualitative, and then further categorised according to the mental health outcome investigated.

The studies’ methodological quality was assessed using the Standard Quality Assessment Criteria for Evaluating Primary Research Papers from a Variety of Fields, otherwise known as the “Kmet criteria” (Kmet, 2004) to give an overall quality score (see Table 3). This assessment tool was chosen because it is appropriate for assessing both qualitative and quantitative research. Three items on the quality appraisal tool were excluded for quantitative studies as they applied to intervention research (e.g., items assessing random allocation and blinding). See Appendix A2 for the full list of criteria used to assess paper quality. Each paper was individually marked against the checklist to assess risk of bias, and an overall quality score was given. Items on the checklist were rated on a 3-point scale (0 = no, 1 = partial, 2 = yes) with a maximum score of 22 for quantitative and 20 for qualitative. A summary score between 0 and 1 was calculated for each paper by summing the total score obtained across relevant items and dividing by the total possible score, in line with scoring guidance (Kmet, 2004). Kmet (2004) suggests that scores over 0.55 are acceptable for inclusion in systematic reviews. Fifty percent of papers were independently quality assessed by a second reviewer. There was substantial agreement between the two reviewers, k = 0.77 (95% CI: 0.661 to 0.88), p < 0.001. Disagreements were resolved in discussion with the supervisors of the project. Studies were divided into three separate groups based on the quality summary score: (a) low, score of below 0.55; (b) average, score of 0.55 to 0.77; and (c) high, score of 0.77 and above.

Table 3.

Standard quality assessment criteria for evaluating primary research papers from a variety of fields, or the “Kmet criteria”.

	Article	1	2	3	4	5	6	7	8	9	10	11	Total	Summary Score	Quality Rating Category
Quantitative	Njeru et al. (2022)	2	1	1	2	0	1	2	1	0	2	2	14	0.64	Average
	Abunyewah et al. (2024)	2	2	2	1	2	2	2	2	0	2	2	19	0.86	High
	Libey et al. (2022)	2	2	1	1	2	2	2	2	1	1	2	18	0.82	High
	Ndetei et al. (2024)	2	2	1	1	1	1	2	0	0	1	2	13	0.59	Average
	Nothling et al. (2024)	2	2	1	2	2	1	1	2	2	0	2	17	0.77	High
	Obaniyi et al. (2023)	1	0	1	0	0	0	0	1	0	0	1	4	0.18	Low
	Prencipe et al. (2023)	2	2	1	2	1	2	2	2	2	1	2	19	0.86	High
	Taukeni et al. (2016)	2	2	2	1	2	2	2	1	0	2	2	18	0.82	High
	Goodman, Raimer-Goodman et al. (2023)	2	2	2	1	2	1	2	2	2	0	2	18	0.82	High
	Goodman, Sharma et al. (2023)	2	2	2	2	2	1	2	1	1	2	2	19	0.86	High
	Abass et al. (2022)	1	2	1	1	1	1	1	2	2	0	2	14	0.64	Average
	Abu et al. (2024)	1	1	1	2	1	1	2	1	0	2	2	14	0.64	Average
	Clayton et al. (2023)	2	2	1	1	2	1	2	2	0	2	2	17	0.77	High
	Kabunga et al. (2022)	2	2	2	2	1	2	1	2	0	2	2	18	0.82	High
	Tomita et al. (2022)	2	2	2	2	1	1	2	2	1	1	2	18	0.82	High
Qualitative	Babugura (2008)	2	2	2	1	1	2	1	1	2	0	N/A	14	0.7	Average
	Beyeler et al. (2023)	2	2	2	1	1	2	2	2	2	0	N/A	16	0.8	High
	Cooper et al. (2019)	2	2	2	2	2	1	1	0	2	0	N/A	14	0.7	Average
	Kadio et al. (2024)	2	2	2	1	2	2	2	2	2	0	N/A	17	0.85	High
	Shoko Kori (2023)	2	2	2	1	2	2	2	0	2	0	N/A	15	0.75	Average
	Adams and Nyantakyi-Frimpong (2021)	2	2	2	2	1	2	2	2	2	2	N/A	19	0.95	High

Results

Description of results

The database search initially yielded 1120 papers, 389 of which were duplicates (Figure 1). As seen in Tables 1 and 2, after screening via two reviewers, 21 studies met the inclusion criteria and included a total of 19,448 participants (sample size range of 20 to 2652. NB: one included article had a total sample size of 10,000, however only 1,000 participants were from an African country).

Table 1.

Study characteristics and baseline patient demographics

	Article	Sample Size	Country	Description of Participants (e.g. sex, age range, occupation)	Reason for Vulnerability	Climate Crisis Exposure	Study Design
Quantitative	Njeru et al. (2022)	400	Kenya	Smallholder crop farmers in Embu and Meru counties. Age, sex not reported.	Farmers	Multiple Impacts of Climate Change	Cross-sectional survey
	Abunyewah et al. (2024)	507	Ghana	Farmers. Male (63.1%) and female. Age range: 18–63 and above.	Farmers	Drought	Cross-sectional survey
	Libey et al. (2022)	469	Ethiopia	Rural People. Mean age = 35 years. Both male and female.	Rural population	Water scarcity	Cross-sectional survey
	Ndetei et al. (2024)	2652	Kenya	High School Children. Male (66.6%) and female. Mean age = 16 years old, range: 13–23 years.	Children and Adolescents	Multiple Impacts of Climate Change	Cross-sectional survey
	Nothling et al. (2024)	69	South Africa	Women, low-income. Age range: 18–44 years. Mean age: 26.3	Women	Flood	Cohort design, two time points (pre and post flood)
	Obaniyi et al. (2023)	120	Nigeria	Rural Farming Households. Age, sex not reported.	Rural population	Multiple Impacts of Climate Change	Mixed methods: structured questionnaires and interview schedules
	Prencipe et al. (2023)	2053	Tanzania	Young people. Age range: 18–23 years old. Male (55%) and female.	Children and Adolescents	Multiple Impacts of Climate Change	Cross-sectional survey
	Taukeni et al. (2016)	480	Namibia	Children (8–18 years old). Female (63%) and male.	Children and Adolescents	Flood	Cross-sectional survey
	Goodman, Raimer-Goodman et al. (2023)	362	Kenya	Adults participating in a community-based empowerment program. Mean age = 41.7 years old. Female (92%) and male.	Rural population	Resource scarcity	Cohort study (2 time points)
	Goodman, Sharma et al. (2023)	152	Kenya	Adults participating in a community-based empowerment program. Female (97.3%) and male. Mean age: 45 years old.	Rural population	Resource scarcity	Cohort study (2 time points)
	Abass et al. (2022)	767	Ghana	Flood-prone community-dwelling households. Mean age: 47.3 years. Males (61.4%) and females.	Survivors of Climate Trauma	Flood	Cross-sectional
	Abu et al. (2024)	505	Ghana	Relocated and non-relocated individuals living in Volta Delta region. Age range: 18–60+. Females (58.9%) and males.	Migrants	Rising sea levels	Cross sectional survey
	Clayton et al. (2023)	10,000	10 countries, 1000 participants per country: Australia, Brazil, Finland, France, India, Nigeria, Philippines, Portugal, the UK and the USA.	Young people aged 16–25. Mean age: 20.82 years old. Male (51.4%) and female.	Children and Adolescents	Multiple Impacts of Climate Change	Cross-sectional survey
	Kabunga et al. (2022)	276	Uganda	Landslide survivors in Bududa district. Female (67.3%) and male. Age range: 18–40+.	Survivors of Climate Trauma	Landslide	Cross-sectional study
	Tomita et al. (2022)	17,255	South Africa	Men and women (52.1%) living in South Africa, depression-free at baseline. Aged between 15 and 65+ years old.	Survivors of Climate Trauma	Extreme Weather Disasters	Prospective cohort study
Qualitative	Babugura (2008)	30	Botswana	Rural young people (10–18 years). Female (60%) and male.	Children and Adolescents	Drought	Interviews, picture drawing and story telling.
	Beyeler et al. (2023)	40	Kenya	Farmers living with HIV. Age range: 23 and 58 years old. Male (50%) and female.	Farmers	Multiple Impacts of Climate Change	Semi-structured interviews
	Cooper et al. (2019)	64	Ethiopia	Pastoralists. Age range: 20–70 years old. Men and women.	Farmers	Water scarcity	Focus Groups and Semi-structured interviews
	Kadio et al. (2024)	150	Burkina Faso	Pregnant/Post-Partum women, community representatives and healthcare professionals. Age range: 20–80 years old. Females (% not specified) and male spouses.	Women	Temperature increases	Focus Groups and Semi-structured interviews
	Shoko Kori (2023)	54	Zimbabwe	Farmers. Male (82%) and female. Age range: 31–80 years old.	Farmers	Multiple Impacts of Climate Change	Exploratory Decriptive Qualitative Research (Semi-structured interviews)
	Adams and Nyantakyi-Frimpong (2021)	20	Ghana	Adult residents of largest informal settlement in Accra (Ghana capital city). Age range: 18–55 years old. Male (75%) and female.	Migrants	Flood	PhotoVoice qualitative research methods (participants use photography and written narratives to represent their community’s concerns).

Table 2.

Main findings of studies.

	Article	Research Aim/Objective	Outcome Measure	Main Findings	Overall Quality Assessment
Quantitative	Njeru et al. (2022)	Aim: To investigate the effects of climate change on the mental health of smallholder crop farmers in the Embu and Meru Counties of Kenya	Self-Reporting Questionnaire 20-item (SRQ-20)	1. Prevalence of mental health issues (psychosomatic discomfort, anxiety, depression) among smallholder crop farmers in Embu and Meru is at 35.2%. 2. Strong positive correlation between climate change and increased mental disorders.	Average
	Abunyewah et al. (2024)	Aim: to investigate the effects of drought on the mental health–specifically, depression, anxiety and stress–of farmers (pastoralist, crop grower and agro-pastoral) in the Talensi district, Ghana.	Drought impact*, Depression, anxiety, stress, social capital	1. Drought impact has a positive statistically significant relationship with depression (β = 0.51, p < 0.001), anxiety (β = 0.24, p < 0.05) and stress (β = 0.36, p < 0.001). 2. Personal social capital was found to be a moderator between drought impacts and mental health outcomes.	High
	Libey et al. (2022)	Aim: to test whether the installation of motorized boreholes in this context contribute to improved water security and emotional well-being.	Water service level (joint monitoring programme), household water security (12 questions), water-related emotional distress (choosing emotion words)	1. Emotional distress from a household’s water situation is related to the frequency of activity interruptions from water deprivation experienced. 2. More than 6 hours of borehole useage is associated with greater household security and emotional wellbeing.	High
	Ndetei et al. (2024)	Aim: to determine the association of climate change with adolescent mental health and suicidality as reported by Kenyan high school students.	Climate-related worry/emotions*; Strengths and Difficulties Questionnaire; Suicidality Questions (5 questions)	1. Girls were more worried and afraid of climate change than boys. 2. Direct positive relationship between the severity of climate change experiences and emotional symptoms (SDQ). 3. Climate change experiences were positively associated with suicidal thoughts, plans and attempts.	Average
	Nothling et al. (2024)	Aim: to determine the extent of damage, loss, injury and death resulting from floods that occurred in and around the city of Durban, South Africa, in April 2022, and associated changes in mental health pre- to post-floods in a low-income setting.	Kessler Psychological Distress Scale; Centre for Epidemiologic Studies Depression Scale (CES-D); Generalised Anxiety Disorder 7; Harvard Trauma Questionnaire (HTQ); Household Hunger Scale (HHS); Childhood Trauma Questionnaire–Short Form (CTQ-SF); Intimate Partner Violence (IPV) and Non-Partner Sexual Violence (NPSV); Composite Trauma Load Score; Flood Impact Questionnaire*	1. Loss of infrastructure (lacked access to drinking water, electricity, fresh food, could not travel to work, had to stay in a shelter and could not get hold of friends or family) was a predictor of post-flood change in levels of emotional distress and anxiety. 2. Higher levels of prior trauma exposure were associated with higher post-flood levels of emotional distress. 3. Higher pre-flood food insecurity was also associated with higher post-flood anxiety.	High
	Obaniyi et al. (2023)	Aim: to examine the impacts of climate change on the health of rural farming households in Kwara State, Nigeria	Effect of climate change on human health*; Adaption to climate and health challenges.	1. Major effects of climate change on the participants were heat stress, decrease in productivity, malaria, fever, depression, hunger and death. 2. Adaptation strategies included personal hygiene and tree planting programme for soil protection and mitigation of climate change, practicing sanitation, prayer to god. 3. Age and educational level of the rural household influenced adaptation strategies to climate change	Low
	Prencipe et al. (2023)	Aim: to measure self-perceived climate change distress among young people living in rural, low resource communities of Tanzania, and to identify whether climate-sensitive risk factors and climate distress were associated with worse mental health.	Depression (Centre for Epidemiological Studies Depression Scale); Cimate change knowledge (Gallup World Poll question: “How much do you know about global warming or climate change?”); Distress about changing weather patterns*(1 idiosyncratic question)	1. Higher education, religiousity, being female, younger and unmarried were linked to greater climate change awareness and climate distress. 2. Region and religiosity were key factors in predicting distress even after controlling for awareness. 3. Working in extreme temperatures, severe food and water insecurity were linked to higher depression.	High
	Taukeni et al. (2016)	Aim: to assess post-traumatic stress disorder (PTSD) in school children as a result of floods in Namibia.	PTSD (Child Trauma Screening Questionnaire)	1. 55% children aged 12 and below and 72.8% aged 13 and older reported re-experiencing symotoms of trauma, 2 years post-flooding event.	High
	Goodman, Raimer-Goodman et al. (2023)	To investigate the temporal relationships between food insecurity, anxiety and depression among adult participants in a community-based empowerment program in Meru County, Kenya.	Food insecurity (Household Food Insecurity Access Scale), anxiety (GAD-7), depression (BDI-II).	1. Food insecurity predicted higher subsequent depression and anxiety at T2. 2. Anxiety at T1 predicted higher subsequent food insecurity.	High
	Goodman, Sharma et al. (2023)	To investigate temporal relationships between food insecurity, water insecurity and generalized anxiety within rural Kenyans during a period of erratic rainfall.	Food insecurity (Household Food Insecurity Access Scale), Household water insecurity* (Household Water Insecurity Access Scale); anxiety (GAD-7),	1. Food insecurity (T1) was more strongly predictive of water insecurity (T2) than water insecurity (T1) predicted food insecurity (T2). Baseline better food security may reduce the impact of future water instability. 2. generalized anxiety may reduce the capacity to adapt to future disruptions to water security.	High
	Abass et al. (2022)	Aim: to examine associations of flood exposure with Psychological Distress among flood survivors in a representative sample in Ghana and to evaluate whether age and sex differences modify this association.	Psychological distress, measured using the Kessler Psychological Distress Scale (K − 10).	1. Strong association between exposure to floods and psychological distress: the prevalence of PD increased with increasing levels of flood exposure. 2. Loss of economic assets and life support systems, disruption to household businesses, damage to personal properties, injuries, and health problems occasioned by flood events increased distress. 3. Females and younger participants were more vulnerable to psychological distress when exposed to flood.	Average
	Abu et al. (2024)	Aim: To explore the outcomes of planned relocation on well-being in the Volta Delta region of Ghana.	Idiosyncratic survey* with items measuring the following concepts: Social identity, Attachment to place and home, Community-based self-efficacy, Safety, Well-being, anxiety.	1. Individuals who experienced planned relocation as a result of sea-level rising had worse well-being, anxiety and lower perceptions of safety when compared to a community that is equally exposed but has not moved. 2. Relocated community members reported significantly lower levels of attachment to the local area and home, significantly lower levels of community-based self-efficacy, and significantly lower levels of overall community-based identity.	Average
	Clayton et al. (2023)	To examine differences in climate related emotions associated with gender and age (important predictors of vulnerability to the impacts of climate change).	Survey* which included questions on: climate-related worry, climate-related functional impact, climate-related emotions (presence of 14 positive and negative key emotions about climate change), climate-related thoughts, experience of being ignored or dismissed when talking about climate change, beliefs about government response to climate change, emotional impact of government response to climate change (reassurance/betrayal).	[Findings on Nigeria only reported, main findings of the study are not applicable to this review]. 1. In India, the Philippines and Nigeria, more than two-thirds reported functional impact; by contrast, US, UK and Finland reported the least functional impact. 2. Nigeria had among the lowest percentages of people reporting angry, guilty, ashamed, despair and powerless. 3. Participants were most likely to report being ignored or dismissed in India, Nigeria, Brazil and the Philippines.	High
	Kabunga et al. (2022)	Aim: To assess the prevalence and correlates of post-traumatic stress disorder among Bududa landslide survivors.	PTSD measured using the checklist—civilian version (PCL-C-5)	1. 46.8% of participants had PTSD symptoms. 2. Male gender, widowhood, lack of counseling and lesser duration since the landslide were associated with an increased likelihood of screening for PTSD in landslide survivors.	High
	Tomita et al. (2022)	Aim: To assess the impact of natural disaster incidents on the onset of depression.	Center for Epidemiologic Studies Depression Scale (CES-D)	1. Exposure to a disaster was associated with significant depressive symptoms. 2. Women, black african, lower educational attainment and lower income groups were more likely to experience depression.	High
Qualitative	Babugura (2008)	Aim: to explore vulnerabilities of children and youth during drought disasters.	Thematic analysis	1. Emotional distress 2. increased workload 3. vulnerability shaped by demographics	Average
	Beyeler et al. (2023)	Aim: to identify potential social and economic pathways through which climate change impacts mental and emotional wellbeing, focusing on a vulnerable population of Kenyan smallholder farmers living with HIV.	Thematic analysis	1. High degrees of stress; fear and concern about the future; and sadness, worry, and anxiety from losing one’s home, farm, occupation, or ability to support their family. 2. Climate-related economic insecurity is a main driver of emotional distress. 3. Widespread economic insecurity disrupts systems of communal and family support, which is an additional driver of worsening mental health.	High
	Cooper et al. (2019)	Aim: to explore relations between emotion, wellbeing and water security among pastoralist communities in Afar, Ethiopia.	Thematic analysis	1. Participants associated water situation (drought) with negative emotional wellbeing, e.g. extreme worry and fatigue during dry season.	Average
	Kadio et al. (2024)	Aim: to investigate how women perceive the impacts of heat on their physical and mental health, in addition to their social relationships and economic activities.	Thematic analysis	1. Extreme heat affects women’s functionality and well-being. 2. Heat undermines a woman’s ability to care for themselves and their child and interferes negatively with breast feeding. 3. Heat negatively affects abilityto work and to maintain relationships. 4. Cultural practices such as ataboo on taking the baby outside before the 40th dayexacerbates the negative consequences of heat.	High
	Shoko Kori (2023)	Aim: to establish the implications of the psychosocial impacts of climate change on sustainable development among smallholder farmers in Chirumanzu	Grounded Theory	Themes identified: 1. Threat of climate change, 2. Feeling humiliated/embarassed at having to ask for money/food from others, 3. Negative feelings, thoughts and emotions.	Average
	Adams and Nyantakyi-Frimpong (2021)	Aim: to examine how recurrent flooding interacts with gendered vulnerability, social differentiation, and place-related historical and structural processes to produce unequal physical and mental health outcomes.	Thematic analysis	1. Five forms of fear were particularly worrying to residents: unpredictable timing of rains, fear of death from drowning or electrocution, fear of property destruction or theft, fear of sicknesses, and fear of fire outbreaks; differentiated and gendered vulnerabilities; and structural and socio-political concerns about place. 2. Women and children are disproportionately more vulnerable to flooding and its consequent health effects. As well, social markers such as housing type, class and wealth, family size, and social relations shape vulnerability. 3. Structural and socio-political concerns about place, specifically community helplessness, feeling neglect by authorities, and threatened sense of place and identity.	High

*Subjective measures of wellbeing, including eco-emotions.

A mixture of both quantitative (15/21) and qualitative (6/21) research methodologies were implemented in the papers included in this review. Studies took place in Kenya (n = 5), Ghana (n = 4), Ethiopia (n = 2), Nigeria (n = 2), South Africa (n = 2), Tanzania (n = 1), Namibia (n = 1), Uganda (n = 1), Botswana (n = 1), Burkina Faso (n = 1), Zimbabwe (n = 1). Countries studied in the articles included in this review are highlighted on a continent map in Appendix A4. Many African countries have not explored the link between the climate crisis and mental health in vulnerable populations, including largely populated countries such as Egypt, Democratic Republic of the Congo, Sudan and Algeria. Some geographical themes are evident across the research, e.g. studies based in Ethiopia explored the impact of drought in rural communities, whereas Kenyan-based studies tended to explore the impact of resource scarcity.

10/15 quantitative studies used cross-sectional designs for the research (Abass et al., 2022; Abu et al., 2024; Abunyewah et al., 2024; Clayton et al., 2023; Kabunga et al., 2022; Libey et al., 2022; Ndetei et al., 2024; Njeru et al., 2022; Prencipe et al., 2023; Taukeni et al., 2016). Three of the quantitative articles included were cohort studies (Goodman, Raimer-Goodman et al., 2023; Goodman, Sharma et al., 2023; Nothling et al., 2024). One used mixed methodologies, interviewing participants for additional and complimentary qualitative findings, however in both studies the quantitative outcomes were the central focus of the research (Obaniyi et al., 2023).

Two qualitative studies used semi-structured interview techniques alone (Beyeler et al., 2023; Shoko Kori, 2023). Two qualitative studies used image-based data collection, to complement semi-structured interview techniques (Adams & Nyantakyi-Frimpong, 2021; Babugura, 2008). The final two articles reported on studies which collected qualitative data from both individual semi-structured interviews and focus groups (Cooper et al., 2019; Kadio et al., 2024). All qualitative studies used thematic analysis to synthesize the data, with the exception of one study which used Grounded Theory (Shoko Kori, 2023).

Most studies recruited both men and women, and the ages of participants ranged from 8 to 80 years old, though several of the studies focused on young people and no studies were specifically researching the older population. The vulnerable populations studied include women, young people, farmers, migrants, people living in rural areas, survivors of climate-related traumatic experiences (such as extreme weather events). The climate-related exposure variables studied in the articles found for this review include drought, water scarcity, flooding, landslides, temperature increases and climate change more broadly. Mental health outcomes measured in the studies include anxiety, depression, climate-related emotions (such as eco-anxiety, eco-guilt), PTSD, psychological/emotional distress, suicide rates and water-related emotional distress (defined as the emotional response that individuals experience in relation to their water security, particularly in terms of coping with water scarcity, having adequate water for needs and the challenges of water access across seasons).

Quality assessment

The Standard Quality Assessment criteria (Kmet, 2004) was utilised for all included studies (Appendix A3). Papers were assessed as follows: <0.55 = low quality, 0.55 to 0.77 = average, >0.77 = high. The total summary scores ranged from 0.18 to 0.95. The average quality score was 0.74 and one paper scored considerably lower than this, at 0.18 (Obaniyi et al., 2023) which is below the 0.55 level of acceptability suggested by Kmet (2004). Compared to other papers, Obaniyi et al. (2023) provided limited descriptions or justifications for the study methodology, outcome measures used, subject characteristics or analysis methods used. They also did not report about any attempts to control for confounding in their study.

The majority scored highly on the initial items of the criteria indicating strengths in the quality of the methodological aspects of the design such as the description of the research question, explanation of outcome measures used and the description of the recruitment strategy. All studies drew conclusions which were supported by the results.

Sufficient relevant demographic information clearly characterizing the participants was not provided adequately in most papers (9/14 quantitative studies included in this report were assessed as “partially” fulfilling this quality assessment criteria – Kmet, 2004). Studies varied regarding the quality of reporting on measures taken to control for confounding variables and a score of 0 was most frequently assigned due to there being no evidence for this occurring in data analysis. Regarding reporting on an appropriate sample size, most studies were assigned a score of 1 due to there being limited evidence of a priori power calculations. Qualitative papers were most likely to lose points on item 10 which required papers to provide evidence of reflexivity. This was due to absent records of reflexivity within the articles, which would normally take the form of a section within the paper acknowledging the authors’ own positionality, subjectivity and biases.

Excluded studies

There were a number of studies that were highly relevant to the present research question that were excluded from inclusion in the review due to ineligibility. For example, research by Tfifha et al. (2021) was excluded from the review due to the article being abstract-only, the research having been presented at a conference. A full assessment of research quality was therefore not possible given the limited information available from the abstract.

Two papers were excluded because they did not directly investigate climate crisis exposures, as defined in the inclusion criteria, though findings related to the climate crisis may be relevant (Prencipe et al., 2021; Myers et al., 2022). A further paper was excluded because the research explored the mental health impact of migration, caused by both war and conflict, but also by climate change, making it impossible to delineate the mental health impact of the climate crisis from the findings (Ali et al., 2023).

One relevant paper was excluded as it did not meet the inclusion criteria for the research having taken place in an African country (Di Giorgi et al., 2020). However, the paper did explore the mental health of African migrants in Italy and linked migration to climate change.

Finally, two relevant papers were excluded for not meeting the inclusion criteria for measuring mental health outcomes (Balakrishnan et al., 2024; Militao et al., 2022). Outcomes measured in the studies (coping strategies, socio-ecological measures) are relevant and linked to mental health outcomes but were felt to be distinct from the mental health outcomes specified in the inclusion criteria.

The following section synthesizes findings from the reviewed articles, categorizing them by group identity: farmers and pastoralists, young people, rural communities, women, migrants and survivors of climate-related trauma. This approach was taken as there was greater convergence in terms of group identity when compared with the mental health outcomes measured in the studies, which were highly heterogenous.

Farmers and pastoralists

Farmers and pastoralists were a vulnerable group which was most frequently studied in the articles included in this review. The first quantitative study used a cross-sectional design to survey a Kenyan small-holder crop farming community that reported a 35.2% prevalence of mental health difficulties (psychosomatic discomfort and symptoms of anxiety and depression were measured using the 20-item Self-Reporting Questionnaire, Andreou et al., 2011) (Njeru et al., 2022). Regression analyses in Njeru’s study found a positive correlation between the poor mental health of smallholder farmers and climate changes (R = 0.126, p-value = <.005). This high-quality study clearly outlines the methodology used to conduct the study and the statistical analysis, making replication of the methods possible. A second, high-quality study explored the specific impact of drought on the mental health of farmers in Ghana through a cross-sectional survey design (Abunyewah et al., 2024). Through Exploratory Factor Analysis, Abunyewah and colleagues found that drought impact had a statistically significant, positive relationship with depression (β = 0.51, p < 0.001), anxiety (β _ = 0.24, p < 0.05) and stress (β _ = 0.36, p < 0.001), and that personal social capital was a moderator between drought impacts and mental health outcomes.

Three qualitative studies included in this review, exploring the views of farmers, took place in Kenya, Ethiopia and Zimbabwe. The highest quality study used thematic analysis on transcripts of semi-structured interviews to find that, almost universally, smallholder Kenyan farmers living with HIV report some level of climate change impact on emotional health, including high degrees of stress; fear and concern about the future; and sadness, worry and anxiety from losing one’s home, farm, occupation or ability (Beyeler et al., 2023). Shoko Kori (2023) corroborated these findings, reporting that Zimbabwe farmers felt humiliated and embarrassed at having to ask for resources from others as a result of climate changes, including resource scarcity (NB: Average quality study). Similarly, Ethiopian farmers reported negative emotional wellbeing as a result of climate changes, such as feeling threatened by climate change, extreme worry and fatigue, especially during times of drought or the dry season (Cooper et al., 2019; NB: average quality study).

Young people

Young people were a vulnerable group commonly studied in the articles found in the present systematic review, with three high-quality quantitative research articles (Clayton et al., 2023; Prencipe et al., 2023; Taukeni et al., 2016) and one average quality study (Ndetei et al., 2024). A qualitative study also explored the impact of climate change in this vulnerable group (Babugura, 2008).

Findings show that 46% of young people (18–23 years old) surveyed in Tanzania reported some distress about climate change, with climate awareness, female gender, higher education, religiosity, older age and working in extreme temperatures, identified as predictive factors (Prencipe et al., 2023). Depression, measured using the Centre for Epidemiological Studies Depression Scale (CES-D10) was also found to be 23 percentage points (95% CI 17–28) higher among youth who had severe water or food insecurity. Though a lower quality study, Ndetei et al. (2024) found similar results in Kenyan high school students, after using a cross-sectional survey in 10 schools in 3 regions of Kenya. Female students reported an increased risk of mental health difficulties related to climate change, such as emotional symptoms and suicidal thoughts and ideation. Ndetei and colleagues attempted to sample in a representative way, selecting schools within three regions of Kenya to ensure representation from across the socioeconomic spectrum within the country, however they do not describe trying to control for any other demographic variables. The worldwide survey exploring climate-related emotions by Clayton et al. (2023) corresponds with these findings, with results indicating that more than two-thirds of Nigerian youth feel a functional impact of the climate crisis, whereas young people from the US, UK and Finland were unlikely to report functional impact. Nigerian respondents, along with young people from India, Brazil and the Philippines, were most likely to report being ignored or dismissed when they tried to talk about climate change.

A high-quality study on the effects of extreme weather events on young Namibians found that 2 years after a flooding event, 55% of children aged 12 and under, and 72.8% of those aged 13 and older, reported re-experiencing trauma symptoms (Taukeni et al., 2016).

A lower quality qualitative research article described the impact of drought on Botswanan participants, aged 10–18 years old, with descriptions of the emotional impact of seeing wildlife dying during drought, their increased workloads during times of drought and anxiety about food insecurity, separation from loved ones and losing opportunities for education reported in the article (Babugura, 2008).

Rural communities

Four quantitative studies explored the impact of the climate crisis on mental health in rural communities. A survey of 469 Ethiopian heads of households established that water insecurity was positively correlated with water-related emotional distress (0.57, p < 0.01), and that lower income and being female increased this risk (Libey et al., 2022; NB: high-quality study). Similarly, food and water insecurity was identified as a predictor for the onset of anxiety and depression in rural Kenyans with diagnoses of HIV (Goodman, Raimer-Goodman et al., 2023; Goodman, Sharma et al., 2023; NB: high-quality studies).

The lowest quality study included in the review investigated the impact of the climate crisis on Nigerian people living rurally, belonging to farming households (Obaniyi et al., 2023). Nigerian participants reported heat stress, malaria, depression, hunger and death were common impacts of the climate crisis on health. Despite being of low quality, due to lack of clarity about outcome measures used, data analytic methods and the lack of described attempts to control for confounding variables, this study represents an important overlap between vulnerable groups identified in the present review, as farmers and pastoralists are likely to live in rural communities, thus making this population at a specific increased vulnerability due to their dependency on a stable climate for financial income and potential risks related to living in a small, rural community.

Women

Two studies included in the review reported on the impact of the climate crisis on women specifically.

Qualitative research conducted in Burkina Faso describes the specific impacts of extreme heat in pregnant women: women reported feeling irritable, sad and anxious during pregnancy (Kadio et al., 2024; NB: high-quality study). Women additionally reported relationship difficulties, difficulties in self-care and caring for their child, e.g. breastfeeding. Moreover, one high-quality study found that, in the wake of extreme weather events such as flooding in South Africa, 38.2% women belonging to low socio-economic groups were found to have above clinical levels of PTSD symptoms, which is an increase of 19.1% from rates prior to flooding (Nothling et al., 2024; NB: high-quality study). Women affected by poverty, food insecurity and a history of trauma were at greater risk of the additive adverse mental health effects of floods, including PTSD, depression and anxiety.

Migrants

Migration is a predicted social impact of the climate crisis. Abu et al. (2024) investigated the mental health impact of migrating on Ghanaian coastal communities exposed to extreme sea-level rising and coastal erosion. Using an idiosyncratic measure assessing concepts such as social identity, attachment to place and community-based self-efficacy, authors found that relocated individuals reported significantly lower levels of overall wellbeing, higher levels of anxiety and lower perceptions of safety, compared to non-relocated community members, despite both groups being exposed to the same climate changes (NB: average quality study). The authors understood these outcomes as being related to the disruption of community connection, identities and feelings of efficacy.

Adams and Nyantakyi-Frimpong (2021) conducted a qualitative study in Ghana, exploring the impact of flooding, which received the highest quality rating of all included studies in this review. Findings from a large informal settlement, made up of displaced individuals, indicate that residents fear: unpredictable timing of rains, fear of death from drowning or electrocution, fear of property destruction or theft, fear of sicknesses and fear of fire outbreaks. Women and children are disproportionately more vulnerable to flooding and its consequent health effects. Housing type, class and wealth, family size and social relations shape vulnerability. Finally, respondents reported structural and socio-political concerns about place, specifically community helplessness, feeling neglected by authorities and threatened sense of place and identity.

Survivors of climate-related trauma

The final vulnerable community researched in the studies identified was survivors of disasters caused by the climate crisis. Following repeated landslides in the Bududa district of Uganda, killing close to 1000 people, 46.8% of survivors surveyed in the study reported symptoms of PTSD, even after at least 2 years after the most recent event (Kabunga et al., 2022; NB: high-quality study). Female gender, widowhood, lack of counselling, low social support and duration since the landslide were significantly associated with PTSD.

In Ghanaian adults, exposure to urban flooding was significantly and positively associated with psychological distress (β = 0.030; p < .005), among male participants (β = 0.019; p < .05) and female participants (β = 0.048; p < .001), this was significant in young adults more so than in older adults (Abass et al., 2022; NB: average quality study). “Exposure to flooding” was measured using an idiosyncratic measure which included items assessing the damage the respondent experienced as a result of flooding (damage to property, loss of livestock, death of loved ones). These findings remained true after controlling for socio-demographic factors, including age, gender, education level, employment status and income level. The article by Abass and colleagues lacked transparency in its reporting of sampling and statistical analysis methods, and results reported lacked detail.

Finally, a high-quality piece of South African research implemented a prospective methodology to follow-up participants who were healthy at baseline, finding that cumulative exposure to natural disasters was significantly associated with the onset of depression (Tomita et al., 2022). Symptoms of depression were particularly common in those exposed to natural disasters who were also female, black African, of lower educational attainment and on a lower income.

Discussion

The aim of this systematic review was to build on the existing understanding of how the climate crisis is impacting mental health, but specifically examining the impact in African nations. This review provided updated evidence in an area of research which is rapidly evolving, while also focusing on vulnerable populations, which has not been done in previous reviews (such as in Atwoli et al., 2022). A total of 21 papers of low to high methodological quality were identified and included in this review, including a total of 36,425 participants.

The 21 studies included in this review span 11 African countries: Kenya, Ghana, Ethiopia, Nigeria, South Africa, Tanzania, Namibia, Uganda, Botswana, Burkina Faso and Zimbabwe. Given that Africa comprises 54 countries, this research cannot be assumed to be representative of the entire continent and does not encompass the diversity of experiences across all African populations. Notably, no studies were identified from island nations, such as Madagascar, which is significant as these nations may be especially vulnerable to the adverse impacts of the climate crisis.

Findings from the review indicate that the climate crisis, both in terms of immediate climate impacts (e.g., natural disasters) and slower or more indirect effects (e.g., rising temperatures), increases symptoms of depression, anxiety, PTSD and incidents of suicide. Vulnerable groups studied include farmers, young people, women, migrants and survivors of climate disasters. Drawing from this narrative review, we may hypothesise that farmers and rural communities face lifestyle threat, economic instability and income loss due to climate disruptions (drought, floods, unpredictable weather), which may trigger anxiety, depression and other mental health distress. Demographics and personal traits can place people at pre-disposed increased risk for the development of poor mental health in response to climate change exposure (Lawrance et al., 2022). Young people may experience heightened emotional distress, including existential dread, due to the anticipation of a compromised future. Women face social and economic inequalities which can place them at specific risk for mental health difficulties. Responsibilities such as caregiving and securing household resources become more challenging under environmental stress, increasing vulnerability to mental health conditions like depression and anxiety. Disrupted community networks further exacerbate mental health risks (Lawrance et al., 2022), particularly for migrants, who may experience cultural dislocation and isolation, compounding the emotional toll of displacement. Lastly, survivors of climate disasters are exposed to immediate and direct impacts of the climate crisis and may develop mental health difficulties due to physical injury, observed loss of life, loss or damage to infrastructure (e.g. homes).

In line with previous reviews, studies included in this review measured the following mental health outcomes: symptoms of anxiety, depression, PTSD and climate-specific emotions (such as eco-anxiety, food-anxiety and water-stress). A similar review, by Sharpe and Davison (2022), which reviewed research from various Lower- and Middle-Income Countries (LMICs) found that PTSD and depression were the most common mental health outcomes, whereas this review found anxiety and depression to be the most common. Similarly to previous reviews, the present review did not identify papers measuring other relevant mental health outcomes, such as eating disorder prevalence or symptomatology, despite the identified impact of the climate crisis on this mental health presentation (Rodgers et al., 2023). This could reflect limitations in the search terms used, though more likely reflects an absence of research in this area. Sharpe and Davison (2022) similarly found that outcome measurements used in included studies were heterogeneous.

Findings from a review based in Africa on the mental health impacts of extreme weather events aligned with our results in certain aspects, such as the observed increase in mood disorders, trauma symptoms and suicide rates (Deglon et al., 2023). However, this review did not identify research measuring disturbed sleep, difficulties in emotion regulation or alcohol use, areas that were examined in Deglon and colleagues’ review. This discrepancy is likely due to the lack of research focused on these specific outcomes within vulnerable African populations, and further research should bridge this gap, to further knowledge regarding these mental health outcomes.

The vulnerable populations identified in this review map onto the vulnerable populations referenced in a South African-based review, where women, fishing communities, farmers and people living in informal settlements were identified (Chersich et al., 2018). This review finds additional research with young people, rural communities and survivors of climate disasters.

The vulnerability of young people has been previously established globally (Lawrance, 2024), with social media engagement on climate and war crises being implicated as compounding psychological distress. Our findings correspond with this, in showing that young people in Tanzania, Namibia, Nigeria, Kenya and Botswana experience symptoms of depression, PTSD and suicidal thoughts which are related to extreme weather events, temperature increases, resource scarcity and climate crisis awareness. Research with children reported in this review ranged in methodological quality, with average to high-quality ratings given. No studies identified explored the role of social media and how its use in young African people might mediate or moderate negative mental health impacts.

Farmers and pastoralists have previously been identified as a vulnerable community to the effects of climate change, with previous literature outlining how farmers in Kenya, Somalia and Tanzania often have to leave their land and livelihoods, leading to mental health difficulties, e.g. feelings of hopelessness, and increased alcohol and substance abuse (Atwoli et al., Sheriff & Mash, 2022). Longitudinal research with Australian farmers indicates that, regardless of financial hardship, drought worry or adverse farming events, farmers had worse mental health than non-farm workers and were less likely to visit a GP or mental health professional (Brew et al., 2016). Further research into help-seeking behaviour in farmers in African countries should be conducted, to establish whether farmers in Africa need support to overcome barriers to accessing available mental health support.

Our findings suggest that extreme weather has a particularly significant impact on the mental health of women, with relationship difficulties, child-rearing and PTSD symptoms all specifically identified. Notably, findings from the present review on the impact of the climate crisis on women are from high-quality research papers. One paper identified in the review also found a connection between high temperatures and suicide attempts, particularly suicide by hanging, in women (Tfifha et al., 2021). However, this research could not be included in the study as the research is only available in conference abstract form. Results in this review correlate with a review of global literature on women’s mental health in response to the climate crisis (Rothschild & Haase, 2023), where women, especially those in marginalized or low-income communities, are found to be disproportionately affected by socioeconomic impacts, leading to heightened eco-anxiety and negative effects on parenting and child development. The authors recommend integrating mental health strategies into climate policies and emphasize the need for gender-sensitive climate action to address these challenges, particularly for women of colour and those in developing nations. Gender-sensitive interventions described include facilitating women’s re-connection with and valuation of nature, as well as working collaboratively with obstetricians/gynaecologists to ensure that patients are supported in making the right reproductive decisions for themselves amidst a changing environment.

Limitations

There are some limitations of this systematic review that need to be considered when interpreting the results presented. First, there is a large variation in the mental health outcomes reported in the studies and how studies reported these outcomes. Depression, anxiety, water-related emotional distress, climate-worries, PTSD symptoms and diagnoses and behavioural difficulties were just some of the outcomes reported in the studies. In papers where outcomes were similar (e.g. depression in Goodman, Raimer-Goodman et al., 2023; Nothling et al., 2024), different measures were used (Beck’s Depression Inventory vs Centre for Epidemiologic Studies Depression Scale), making meta-analysis impossible and in-depth comparisons of outcomes difficult.

Secondly, the term “vulnerable communities” is difficult to define. Although we use a previously adopted definition (White et al., 2023), it was still challenging to determine whether certain groups studied met inclusion criteria. The researchers opted to be inclusive, meaning that, for example, studies with participants from rural populations were included as their vulnerability to the climate crisis was assumed, even if their specific vulnerability was not outlined in the research paper. The study excluded papers where stakeholders or professionals were participants, as the inclusion criteria stated that participants should be African residents from vulnerable populations. This means that some potentially relevant studies were excluded (e.g. Lindvall et al., 2020; Sorgho et al., 2020). It is likely that the authors’ lack of geographical knowledge will have prevented the identification of certain types of vulnerability specific to the local area, for example groups may be vulnerable due to the local provision of mental health care (or lack of), local societal/political events or differences across the continent in the exposure to indirect impacts of climate change (e.g. heat levels rising).

The impact of the climate crisis is also difficult to determine. For example, studies investigating the impact of war or conflict were excluded, though some may argue that war can oftentimes be linked to climate change (Koubi, 2019). Migration also was not included as a specific search term, despite the recognised link between the climate crisis and migration (Trummer et al., 2023), so studies where migration is studied as a climate change impact were only included if the authors themselves linked migration to climate change (and therefore used terms such as “climate change” in the manuscript). This means that some relevant papers may not have been included. Similarly, resource scarcity was included as a climate change exposure variable when authors linked the conditions causally to the climate crisis.

Due to practical constraints, this review includes only studies published in English, as access to professional translation services was not available. While this may limit the inclusion of relevant research in other languages, it ensures the accuracy and consistency of data extraction and analysis. Future reviews should prioritise access to translation services to enable the inclusion of papers written in the languages spoken in Africa. Moreover, accessing and including grey or unpublished literature was also outside of the scope of this review. Although this increases the likelihood of publication bias, it also increases the likelihood that included studies are of a higher methodological standard, with peer-reviewed studies being required to include full methodological disclosure.

Finally, this review focuses on the adverse impacts of climate change on mental health in Africa and negates research into coping and adaption. Research with indigenous African communities reveals that coping strategies including weather-forecasting systems based on changes in animal behaviour, livestock and crop diversification, cattle stress-management techniques and division of labour (to allow for acclimatisation to drought challenges) are effective practices in use (Leal Filho et al., 2021).

Future research directions

Although many of the papers included in this review were deemed to be of a high quality, much research was of average to low quality. Higher quality research in this field is required, and the use of longitudinal methodologies, rather than cross-sectional surveys, is recommended to enable greater understanding of the long-term impact of climate changes and temporal relationships between mental health factors. Barriers to improving research quality include limited funding, which constrains study design and scope; the absence of standardized methodologies for measuring climate-related mental health impacts, which could lead to inconsistent research approaches; and mental health stigma, leading to underreporting and challenges in accessing representative populations.

No research studies were found in our search investigating the mental health impact of climate change in the following vulnerable groups within Africa: those with pre-existing health conditions, historically marginalised or persecuted communities, e.g. ethnic minorities, indigenous groups, LGBTQ+ communities. Future research should work with these vulnerable groups to understand their unique experiences. Future research would benefit from the use of mixed methods, using interviews or focus groups to compliment survey data, to understand how pre-existing risk factors (such as belonging to one of these vulnerable groups) interact with direct or indirect exposure to climate changes, resulting in possible mental health difficulties.

The studies included in this review occurred in a total of 12 African countries, with the majority taking place in Ghana and Kenya. Given the size of the African continent, and the differing impacts of the climate crisis across the various eco-systems within it, it is vital that future research studies the unique impacts of the climate crisis on mental health in other African countries, especially in island communities who are likely to feel the impact of the climate crisis keenly. Future research in island African nations could use longitudinal methods to monitor exposure to the direct impacts of the climate crisis and mental health outcomes. Future research should also investigate impacts on individuals with neurodevelopmental conditions, such as autism and ADHD, who may have sensory needs that are affected by climate changes (Georgiou et al., 2025). Additionally, research could investigate resilience and protective factors in these nations to better understand personal or community resources that enable island populations to withstand unstable climate changes. Currently, there is a significant bias in the climate-change psychological literature, with research predominantly taking place in the Global North, and this imbalance needs to be addressed, via funding and research attention from within the African content.

Finally, future pilot intervention-based studies should work with farmers and people living in rural areas in Africa. Farmers and people living in rural communities were consistently identified as particularly vulnerable to the impacts of the climate crisis in this review. The impact of improved mental healthcare access in rural areas should be measured following small-scale pilots. Mental health interventions may include a focus on resilience, strengthening community and activism. Mental health interventions could be implemented alongside farming-based adaptions which focus on modifications to farming methods which maintain production stability despite unstable climate changes.

Implications for policy

Policies and interventions across Africa are currently focused on minimizing the physical and societal impact of climate-related disasters and extreme weather events. For example, the World Meteorological Organization (WMO) announced an “Early Warnings for All Action Plan” for Africa in 2023, which aims to build infrastructure to allow timely and accurate information about natural hazards and impending disasters to reach all segments of African society, particularly the most vulnerable (World Meteorological Organization, 2024). However, policy-makers also need to consider pre-emptive resourcing for interventions to encourage “mental health disaster preparedness” (dos Santos, 2022).

Greater action by policymakers in Africa is required to prevent the impacts of climate change on mental health by addressing the root causes of climate change; proactively address the mental health impacts outlined in this review as well as predicted mental health impacts; and innovate in response to these challenges through evidence-based interventions for policy and practice (Grantham Institute, 2021). Policy changes that address the climate crisis may also reduce adverse mental health impacts, in a “win-win” policy strategy, e.g. reducing burning fossil fuels will reduce climate change and the impact on warming temperatures could have a positive effect on mental health, as research described in this paper has linked rising heat levels with poor mental health outcomes.

Conclusion

This systematic review analysed articles that investigated the mental health impact of the climate crisis in vulnerable populations living in Africa.

The main findings indicate that the climate crisis, both in terms of extreme weather events and more gradual impacts (such as sea level and temperature rises), is severely impacting the mental health of farmers, young people, women, migrants and people who have survived climate-related traumatic events. The most consistent findings were in relation to symptoms and rates of depression, anxiety and PTSD. Future high-quality research is needed to establish longitudinal impacts of the climate crisis and to explore the impacts on vulnerable groups which are, as yet, neglected in the research, including, e.g. ethnic minorities, indigenous groups, LGBTQ+ communities.

Appendix A1. Search terms

African Countries AND	Subject heading term: “Africa” (explode)
	Specific Terms: “North Africa” or Egypt* or Libya* or Libia* or Tunisia* or Algeria* or Morocco* or “West Africa” or Benin or “Burkina Faso” or Cameroon or “Cape Verde” or Chad or “Canary Island” or “Cote d Ivoire” or “Ivory Coast” or Gambia* or Ghana* or Guinea* or Guinea-Bissau or Jamahiriya or Liberia* or Mali or Mauritania* or Niger* or Nigeria* or Benin* or Senegal* or “Sierra Leone” or Togo or “Central African Republic” or Congo* or “Democratic Republic of Congo” or Zaire or “Equatorial Guinea” or Gabon* or “Sao Tome and Principe” or “East Africa” or Eritrea* or Ethiopia* or Somalia* or Djibouti* or Sudan* or “South Sudan” or Uganda* or Kenya* or Tanzania* or Rwanda* or Burundi* or “The Comoros” or Mauritius or Seychelles or Madagasca* or “South Africa” or “St Helena” or Angola* or Botswana* or Lesotho* or Malawi* or Mozambique* or Namibia* or Swaziland or Zambia* or Zimbabwe* or “cabo verde” or eswatini* or Liberia or “sub-saharan”.ti,ab.
Changes in the climate AND	Subject heading term: “climate change” and related terms
	Specific terms: “Climate change” or climate* or “climate crisis” or “natural disasters” or avalanche* or “cyclonic storm” or drought* or flood* or landslide* or tornado* or wildfire* disaster* or weather* or “extreme weather*” or humidit* or “rain fall” or “change in temperature” or “temperature change” or lightning* or ice or “greenhouse effect” or “greenhouse gas” or “global warming” or biodiverse* or “particulate matter” or “air pollution” or “air pollutants” or “water quality” or “water pollutant” or “soil pollutant” or “weather events” or “climatic variability” or “ecological change” or “heat wave” or “land use” or “extreme temperature” or “cold wave” or snowstorm* or frost* or freez* or fog* or storm* or derecho* or hail* or thunderstorm* or rockfall* or mudflow* or “wave action” or “rogue wave” or “glacial lake outburst” or “forest fire” or “land fire” or “brush fire” or “bush fire” or “pasture fire” or “habitat loss” or “resource scarcity”.ti,ab.
Mental Health Difficulties	Subject heading term: mental illness and related terms
	Specific terms: “mental health” or “mental illness” or “mental disorder” or “mental distress” or “psychological disorder” or “psychiatric disorder” or psychopatholog* or anxiet* or depress* or “post-traumatic stress” or OCD or “obsessive compulsive disorder” or “posttraumatic stress” or PTSD or suicide* or “self harm” or “self injury” or “self injuries” or “substance use” or “substance abuse” or “alcohol use” or “alcohol misuse” or “alcohol abuse” or “drug abuse” or “drug misuse” or “mood disorder” or depression* or anxiet* or psychos* or schizophreni* or bipolar* or “bipolar disorder”.ti,ab.

Appendix A2. Included questions from kmet quality assessment tool

Quantitative

Q1. Question / objective sufficiently described?

Q2. Study design evident and appropriate?

Q3. Method of subject/comparison group selection or source of information/input variables described and appropriate?

Q4. Subject (and comparison group, if applicable) characteristics sufficiently described?

Q5. Outcome and (if applicable) exposure measure(s) well defined and robust to measurement / misclassification bias? Means of assessment reported?

Q6. Sample size appropriate?

Q7. Analytic methods described/justified and appropriate?

Q8. Some estimate of variance is reported for the main results?

Q9. Controlled for confounding?

Q10.Results reported in sufficient detail?

Q11. Conclusions supported by the results?

Qualitative

Q1. Question / objective sufficiently described?

Q2. Study design evident and appropriate?

Q3. Context for the study clear?

Q4. Connection to a theoretical framework / wider body of knowledge?

Q5. Sampling strategy described, relevant and justified?

Q6. Data collection methods clearly described and systematic?

Q7. Data analysis clearly described and systematic?

Q8. Use of verification procedure(s) to establish credibility?

Q9. Conclusions supported by the results?

Q10. Reflexivity of the account?

Appendix A3. Quality assessment table

Standard Quality Assessment Criteria for Evaluating Primary Research Papers from a Variety of Fields, or the “Kmet criteria”

	Article	1	2	3	4	5	6	7	8	9	10		11	Total	Summary Score	Quality Rating Category
Quantitative	Njeru et al. (2022)	2	1	1	2	0	1	2	1	0	2		2	14	0.64	Average
	Abunyewah et al. (2024)	2	2	2	1	2	2	2	2	0	2		2	19	0.86	High
	Libey et al. (2022)	2	2	1	1	2	2	2	2	1	1		2	18	0.82	High
	Ndetei et al. (2024)	2	2	1	1	1	1	2	0	0	1		2	13	0.59	Average
	Nothling et al. (2024)	2	2	1	2	2	1	1	2	2	0		2	17	0.77	High
	Obaniyi et al. (2023)	1	0	1	0	0	0	0	1	0	0		1	4	0.18	Low
	Prencipe et al. (2023)	2	2	1	2	1	2	2	2	2	1		2	19	0.86	High
	Taukeni et al. (2016)	2	2	2	1	2	2	2	1	0	2		2	18	0.82	High
	Goodman et al. (2023)	2	2	2	1	2	1	2	2	2	0		2	18	0.82	High
	Goodman et al. (2023)	2	2	2	2	2	1	2	1	1	2		2	19	0.86	High
	Abass et al. (2022)	1	2	1	1	1	1	1	2	2	0		2	14	0.64	Average
	Abu et al. (2024)	1	1	1	2	1	1	2	1	0	2		2	14	0.64	Average
	Clayton et al. (2023)	2	2	1	1	2	1	2	2	0	2	2		17	0.77	High
	Kabunga et al. (2022)	2	2	2	2	1	2	1	2	0	2	2		18	0.82	High
	Tomita et al. (2021)	2	2	2	2	1	1	2	2	1	1	2		18	0.82	High
Qualitative	Babugura et al. (2008)	2	2	2	1	1	2	1	1	2	0	N/A		14	0.7	Average
	Beyeler et al. (2023)	2	2	2	1	1	2	2	2	2	0	N/A		16	0.8	High
	Cooper et al. (2019)	2	2	2	2	2	1	1	0	2	0	N/A		14	0.7	Average
	Kadio et al. (2024)	2	2	2	1	2	2	2	2	2	0	N/A		17	0.85	High
	Shoko Kori. (2023)	2	2	2	1	2	2	2	0	2	0	N/A		15	0.75	Average
	Adams et al. (2021)	2	2	2	2	1	2	2	2	2	2	N/A		19	0.95	High

Appendix A4. Map of african counties studied in research papers included in review

Funding Statement

This work was funded by the Oxford Institute of Clinical Psychology Training and Research.

Ethical approval statement

Due to the nature of this research, ethical approval is not required. The article is a review of previously published literature, meaning no primary data collection took place during this research.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/28324765.2025.2500747.

SRL Appendices