Climate change anxiety among young women living in the urban slums of Kampala, Uganda: findings from the baseline assessment of the TOPOWA cohort study

Abstract

Introduction

Climate change is an emerging global health crisis, particularly in low-resource settings like Uganda’s urban slums. These areas face significant challenges in mitigating climate risks, exacerbating the vulnerability of residents. This study aims to assess the prevalence of climate change anxiety and its association with generalised anxiety among young women living in urban slums. Understanding this relationship is crucial for developing community-based mental health strategies and resilience-building initiatives to address the psychological impact of climate change.

Methods

This study utilises baseline data from ‘The Onward Project On Wellbeing and Adversity’ (TOPOWA), a prospective cohort study focused on mental illness and social determinants of health among young women aged 18–24 years living in Kampala’s urban slums. A total of 300 women were recruited from three sites: Banda, Bwaise and Makindye. At baseline, interviewer-administered surveys measured climate change anxiety using the 13-item Climate Anxiety Scale and generalised anxiety using a 7-item scale. Bivariate and multivariate analyses were conducted to determine the associations between climate anxiety, generalised anxiety and demographic factors.

Results

Of the 249 women included in the analysis, 21% reported moderate to severe levels of climate change anxiety, while the remaining 79% experienced mild to minimal levels. Multivariate analysis revealed a strong association between higher levels of generalised anxiety and increased climate change anxiety. Additionally, older age was linked to higher climate anxiety, whereas higher education levels and living in multigenerational households were associated with lower levels of climate anxiety.

Conclusion

One in five young women in Kampala’s urban slums experiences moderate to severe climate change anxiety, closely linked to generalised anxiety. These findings highlight the urgent need for targeted mental health interventions and community-based resilience programmes. Leveraging family support in multigenerational households may also play a role in reducing climate-related anxiety and fostering adaptive coping mechanisms.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Climate change anxiety is recognised globally, mainly in higher-income countries, with little research on low-income urban slums like Kampala.

• Young women in slums face heightened vulnerability to environmental stressors, but their levels of climate change anxiety have not been well studied or documented.

WHAT THIS STUDY ADDS

• It shows young women in Kampala’s slums experience significant climate change anxiety linked to survival struggles and environmental threats.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• The findings call for more research on climate change anxiety in vulnerable populations, particularly in urban slums in low-income countries.

• This study suggests the need for mental health support integrated with climate adaptation strategies and gender-sensitive climate resilience programmes.

Climate change is a significant threat to the global population in the 21st century, affecting the existential well-being of humans, ecosystems and societies.¹ Even so, research on the perceptions of climate change and related anxiety is emerging, particularly from those populations most impacted in low- and middle-income countries. At this time, it is projected that climate change will be responsible for 250 000 additional deaths annually between 2030 and 2050 from malaria, undernutrition, diarrhoea and heat stress alone.² Evidence is increasing that links climate change with population health as documented by studies’ findings.3,7

In response to these dire projections, the 2016 Paris Agreement committed to holding the global temperature rise below 2°C above pre-industrial levels, with efforts to limit the increase to 1.5°C.⁸ However, climate change has already resulted in a range of direct hazards, including unpredictable weather patterns, flooding, extreme temperatures and drought. These climate shocks have frequently resulted in increased migration, further straining local resources and placing people in overcrowded and unstable conditions which foster disease spread.^{2 9 10} The physical circumstances created by climate shocks can have profound psychological impacts, particularly for vulnerable populations, including women and girls in low-income settings.

Though the climate change crisis has been felt worldwide, developing countries are bearing the biggest burden of its impact because of the limited access to climate information, poor infrastructures and inadequate resources to build resilience.² The 2022 report of the Intergovernmental Panel on Climate Change¹⁰ showed that vulnerability to climate shocks was much more pronounced in tropical and subtropical regions that are also experiencing substantial limitations to development, such as poverty, conflicts, limited access to basic services and resources and high levels of climate-sensitive livelihood.10,13 It is approximated that in 2030, sub-Saharan Africa will suffer the greatest burden of mortality attributed to alterations in climate. In fact, a report by the UCL Lancet Commission estimated that about 34% of global disability-adjusted life-years (DALYs, the sum of the years lost due to disability and to early mortality) in sub-Saharan Africa are attributed to the impacts of climate change.¹⁴ West, Central and East Africa have been listed among the global hotspots for high climate change vulnerability.¹³ East Africa, in particular, has shown an increased risk to flooding, heat stress and drought.¹⁵

Evidence exists that links the impacts of climate change to poor health in Uganda.16,18 Groups disproportionately affected by climate change include the poor, refugees, the elderly, women and children.^{2 19 20} In climate hotspots and insecure environments such as slums, women and children face severe deterioration in sexual and reproductive health, gender-based violence, teenage pregnancies and poor mental health.2020,23 For instance, in Zambia, Rosen and colleagues²⁴ found that as an indirect consequence of drought, female-headed households suffered the most. Additionally, young girls were prematurely forced into marriage or the workforce, and with less income and autonomy, access to healthcare became more limited. In Uganda, weather shocks decreased the household consumption expenditure by 17 percentage points, and the decline was also significantly larger among female-headed households.²⁵

Climate change exerts a substantial impact on the physical and mental well-being in humans; however, the psychological effects have been shown to surpass physical injuries.²⁶ The physical and mental health manifestations of climate change are closely linked and are mediated by social, political and economic determinants of mental health, such as poverty, food insecurity and inadequate housing.27,30 According to WHO, alterations in climate have heightened existing social, environmental and economic predictors of psychosocial well-being,³¹ inducing both short-term effects like anxiety and long-term disorders resulting from factors like displacement or interrupted social cohesion.^{2 32}

Even though research in this area is emerging, the key indicators of climate anxiety appear to span emotional, cognitive, physiological and behavioural dimensions.³⁰ A scoping review by Charlson and colleagues²⁸ indicated that several climate-related exposures were associated with exacerbated mental health issues, including psychological distress and death among people with pre-existing mental conditions. Also, disadvantaged people and residents of informal settlements—especially in developing countries—are disproportionately affected by mental health concerns.²⁷ Factors like poverty, inadequate social services and conflicts compound the situation.^{33 34} Substance use, which has been shown to be strongly associated with poor mental health, has also been found to be associated with climate events.^{35 36} Youths are particularly vulnerable to the psychological impacts of climate change.3437,40 While young people in lower- and middle-income countries (LMICs) appear to suffer more than other age groups from the mental health impacts of climate change, research has shown that those from the higher-income nations worry the most about the future.⁴¹ This may be due to less research done in this area in the LMICs. A global survey among young people aged 16 to 25 indicated that more than half were extremely worried about climate change, and 84% were moderately worried.⁴²

Our previous research underscores that the mental health context for young women residing in poverty in urban Kampala is quite complex as there is a high prevalence of sexually transmitted infections, including HIV,⁴³ high levels of sexual and gender-based violence victimisation and intimate partner violence44,47 and transactional survival sex⁴⁸ and overall a high unmet mental health need reflecting suicidality, depression, anxiety and stress.4749,52 Despite the growing interest in assessing climate change anxiety (CCA), very little is known about this issue among young women living in this setting where the impacts of climate change are expected to be among the most severe. It is within this context that the current study aims to assess the prevalence of CCA, generalised anxiety disorder (GAD) and the associated factors among young women residing in the urban slums of Kampala.

Methods

Study participants

Between July 2023 and November 2023, we enrolled 300 young women aged 18 to 24 years as part of a prospective observational cohort study across three sites in Kampala (Banda, Bwaise and Makindye) to examine the mechanistic pathways of mental illness (TOPOWA study). The study target population was those aged 18 to 24 who self-reported as female, lived within a radius of two kilometres from the Uganda Youth Development Link (UYDEL) vocational training centres and who had attained a minimum of primary five education level. Those with self-reported pregnancy, significant intellectual disability, severe mental illness or substance use requiring hospitalisation were excluded from the study. Among the 495 young women screened, 137 were not eligible for participation and 58 did not appear during the study enrolment. As such, among those eligible, the participation rate was 83%.

In this paper, we present cross-sectional analyses from baseline assessment for the TOPOWA cohort study. Participants were asked to complete a research assistant-administered survey containing a broad range of measures pertaining to demographic and psychosocial characteristics and life experiences, in addition to other study components. The study was conducted in accordance with the ethical declaration of Helsinki.⁵³ It was approved by the Kennesaw State University, Makerere University School of Health Sciences (MAKSHS) Research and Ethics Committee (MAKSHSREC-2023–532) and the Uganda National Council of Science and Technology, UNCST (HS2959ES). All participants provided written informed consent before taking part in the study. Participants also received remuneration for participating in the survey, and the other data collection protocol components. Participants or the public were not involved in the design, or conduct, or reporting or dissemination plans of the research presented in this paper.

Sample size, statistical power determination and sampling

The power calculation used a two-tailed alpha level of 0.05 and was estimated using R’s powerlmm package and Optimal Design. Based on a sample of 150 youth receiving Socioeconomic Targeted Training (SeSTT/Intervention) and 150 youth in the control (comparison) group, the overall power to detect a moderate-to-large effect size (Cohen’s d of 0.35) of the impact of SeSTT on mental health outcomes longitudinally, after accounting for an anticipated 10% cumulative attrition, was >0.80.

Participant recruitment was conducted at three different UYDEL sites. For both the intervention and comparison groups, 50 women were selected per site.

Measures

Demographic characteristics

A demographic survey questionnaire was used to assess the background characteristics of the study population. The survey collected information that included age (≤20 years, >20 years), education level, the status of participants’ parents (both parents alive, one parent alive or no parents alive), living status (lives with parents, lives with children) and parenting status (has biological children).

Climate change anxiety

To assess CCA, the Climate Anxiety Scale (CAS), a 13-item measure of CCA, was used, which consists of two subscales that assess cognitive impairment (eight items) and functional impairment (five items).⁵⁴ Despite the fact that the developed CAS scale showed satisfactory psychometric properties in the original study by Clayton and Karazsia,⁵⁴ the 13-item CAS has not been validated in this population or similar settings. However, the subsequent validation studies have shown that the CAS two-factor scale has a satisfactory and good fit to the data.^{55 56}

The items included ’Thinking about climate change makes it difficult for me to concentrate’ and ’My concerns about climate change make it hard for me to have fun with my family or friends’. The individual item scores were scored on an ordinal scale ranging from 1 (never) to 5 (almost always). In the present sample, the 13 CCA scale items were treated as a single scale, following the results of a previous study.⁵⁷ Hence, the sum score of the CCA scale ranges from 13 to 65, with higher scores indicating higher climate-related anxiety⁵⁷ and has satisfactory psychometric properties. The Cronbach’s alpha value in this study was α=0.86.

Generalised anxiety disorder

The GAD-7,⁵⁸ a valid and efficient 7-item anxiety tool for assessing GAD, was used. The items include ‘feeling nervous, anxious, or on edge’ and ‘feeling afraid as if something awful might happen’. The questionnaire asked participants how often they were bothered by each symptom. Response options were ‘not at all’, ’several days’, ‘more than half the days’ and ‘nearly every day’, scored as 0, 1, 2 and 3, respectively. Confirmatory factor analyses confirmed the one-dimensional structure of the GAD-7.⁵⁹ The GAD-7 scale score ranges from 0 to 21, with a higher score reflecting greater anxiety severity. The sum score of GAD-7 was categorised to form the level of anxiety severity as follows: minimal (0–4), mild (5–9), moderate (10–14) and severe.15,2158 The Cronbach’s alpha value was α⁷ = 0.75 in this study.

Statistical analysis

Demographic characteristics were presented as counts and percentages. To assess CCA prevalence, CCA was measured as a mean score, which was later categorised to present group percentages by recoding the CCA mean (M) score as ‘mild’ (1.00≤M≥ 2.33), ‘moderate’ (2.34 ≤M≥ 3.66) and ‘severe’ (3.67≤M≥ 5.00) in line with previous CCA research.⁶⁰ The last two groups were combined into ‘moderate to severe’ to reduce variation across the groups. The mean score was used to develop categories for CCA prevalence since the high values of Cronbach’s alpha indicated that the items of the CCA scale were sufficiently inter-correlated and grouping them provided a valid measure of CCA.⁶¹

The association between CCA, GAD and demographic characteristics was assessed using χ² tests, and the effect of global warming on women was also examined using the same method. Bivariate analysis was conducted by fitting linear regression models between CCA and independent variables to determine variables to be included in the multivariable model. The cut point for variables in the bivariate analysis to be selected in the multivariate analyses was p<0.25.⁶² The multicollinearity test was conducted using a correlation matrix to rule out highly correlated independent variables. Only variables with correlation coefficient values below 0.4 were considered not as highly correlated and hence included in the multivariable model. Multicollinearity was confirmed present if the variance inflation factor was greater than 5.⁶³ All variables significant at the bivariate level were included and maintained in the multivariable model.

A multivariable linear regression model was fit to the data to assess the relationship between CCA, GAD and demographic characteristics. The outcome variable was CCA, modelled as a sum score and the main independent variable was the GAD and the models controlled for the participants’ demographic characteristics.

Missing data were handled by excluding non-responded items and estimating total scores as the mean of completed items multiplied by the total number of items, an approach recommended when missingness is minimal.⁶⁴ For example, the total score of the GAD was computed as (GAD item mean)×(7).

Stata 15.0 (StataCorp, College Station, TX) was used for all analyses, and the level of statistical significance was less than 5% for all tests.

Results

Sample demographics

This study analysed cross-sectional baseline data of 249 women who had ever heard about climate change or global warming out of the 300 women enrolled in the cohort. Of these, the majority (70.7%) had attained at least some secondary education, 51.8% were older than 20 years of age, 61.5% had ever given birth and 49.4% were living with their children. Most of the women (65.5%) had both of their parents alive. Details of the results are in table 1.

Table 1. Demographic characteristics for the women aged 18–24, from the baseline assessment of the TOPOWA cohort study, Kampala, Uganda, 2023.

Characteristic	Category	Count (n=249)	Percentage (%)
Education level	Primary or lower	73	29.3
	Some secondary or higher	176	70.7
Age	18–20 years	120	48.2
	21–24 years	129	51.8
Has biological children	No	96	38.6
	Yes	153	61.5
Household size	≤4	143	57.4
	>4	106	42.6
Parents’ living status	Both parents alive	163	65.5
	At least one parent is not alive	86	34.5
Lives with parents	No	163	65.5
	Yes	86	34.5
Lives with her children	No	126	50.6
	Yes	123	49.4
Household generations	One	43	17.3
	Two	122	49.0
	Three or more	84	33.7

Climate change awareness and impact

Of the 249 women who had ever heard about climate change or global warming, about 90% believed it was happening, and 69.5% were knowledgeable about it. Of those knowledgeable about climate change, more than half (56.7%) perceived it as bad, 38.2% thought it was neither good nor bad, while the rest thought it was good. More than half (52.6%) thought global warming was due to a naturally occurring process, while 44.5% thought it was due to people’s actions.

Within the past year, global warming was reported to impact 46.6% of all women in the study; 12.9% by drought, 24.5% by flood, 41.4% by storms/too much rain, 34.1% by heat waves/high temperatures and 9.2% by unpredictable weather patterns.

Prevalence of CCA

Generally, the level of CCA across the 13 items was somewhat low (figure 1). Notably, CCA was higher among women who presented with higher levels of GAD (figure 2). About 21% of the women presented with moderate to severe CCA, while the rest presented with minimal to mild CCA.

Figure 1. Mean scores on climate change anxiety scale items as reported by women aged 18-24 at the baseline assessment of the TOPOWA cohort study. CCA, climate change anxiety.

Figure 2. Climate Change Anxiety mean scores by generalised anxiety disorder severity as reported by women aged 18-24 at the baseline assessment of the TOPOWA cohort study. CCA, climate change anxiety.

Also, table 2 indicates a significant association between CCA and GAD, education level, age and the number of generations in a household. Also, there was a significant association between global warming impact and education level, living with parents and the number of generations in a household.

Table 2. Associations between climate change anxiety, global warming impact and generalised anxiety disorder, by demographic characteristics among women participants, aged 18–24 from the TOPOWA cohort study baseline assessment, Kampala, Uganda.

Characteristic	Climate change anxiety		Impacted by global warming
	Minimal to mild (n=197)	Moderate to severe (n=52)	Yes (n=116)	No (n=50)	N/A (n=83)
GAD
Minimal	46 (23.4)	5 (9.6)	25 (21.6)	10 (20.0)	16 (19.3)
Mild	83 (42.1)	19 (36.5)	45 (38.8)	19 (38.0)	38 (45.8)
Moderate	52 (26.4)	18 (34.6)	33 (28.5)	15 (30.0)	22 (26.5)
Severe	16 (8.1)	10 (19.2)	13 (11.2)	6 (12.0)	7 (8.4)
Education level
Primary or lower	48 (24.4)	25 (48.1)	29 (25.0)	12 (24.0)	32 (38.6)
Some secondary or higher	149 (75.6)	27 (51.9)	87 (75.0)	38 (76.0)	51 (61.5)
Age
18–20	106 (53.8)	14 (26.9)	54 (46.6)	22 (44.0)	44 (53.0)
21–24	91 (46.2)	38 (73.1)	62 (53.5)	28 (56.0)	39 (47.0)
Has biological children
No	81 (41.1)	15 (28.9)	44 (37.9)	20 (40.0)	32 (38.6)
Yes	116 (58.9)	37 (71.2)	72 (62.1)	30 (60.0)	51 (61.5)
Household size
≤4	108 (54.8)	35 (67.3)	73 (62.9)	30 (60.0)	40 (48.2)
>4	89 (45.2)	17 (32.7)	43 (37.1)	20 (40.0)	43 (51.8)
Parents' living status
Both parents alive	133 (67.5)	30 (57.7)	75 (64.7)	31 (62.0)	57 (68.7)
Some parent not alive	64 (32.5)	22 (42.3)	41 (35.3)	19 (38.0)	26 (31.3)
Lives with parents
No	123 (62.4)	40 (76.9)	85 (73.3)	35 (70.0)	43 (51.8)
Yes	74 (37.6)	12 (23.1)	31 (26.7)	15 (30.0)	40 (48.2)
Lives with her children
No	101 (51.3)	25 (48.1)	58 (50.0)	27 (54.0)	41 (49.4)
Yes	96 (48.7)	27 (51.9)	58 (50.0)	23 (46.0)	42 (50.6)
Household generations
One	27 (13.7)	16 (30.8)	23 (19.8)	14 (28.0)	6 (7.2)
Two	97 (49.2)	25 (48.1)	70 (60.3)	12 (24.0)	40 (48.2)
Three or more	73 (37.1)	11 (21.2)	23 (19.8)	24 (48.0)	37 (44.6)

Bold values indicate statistically significant results.

GAD, generalised anxiety disorder; N/A, nonresponse.

Multivariable analysis of factors associated with CCA

Results from table 3 indicate that women with moderate (β=3.77; 95% CI 1.14 to 6.40) and severe (β=6.54; 95% CI 3.05 to 10.03) GAD scores had significantly higher CCA levels. Similarly, age 21 -24 years (β=3.18; 95% CI 1.34 to 5.02) was linked to higher CCA. The presence of two (β=−2.95; 95% CI −5.54 to –0.36) and three (β=−4.03; 95% CI −7.02 to –1.05) generations in a household was linked to lower CCA.

Table 3. Multivariable regression model of climate change anxiety-associated factors among women aged 18–24, from the TOPOWA cohort study baseline assessment, Kampala, Uganda.

Characteristic	β	95% CI	P value
GAD
Minimal	Ref.
Mild	1.91	−0.56 to 4.38	0.129
Moderate	3.77	1.14 to 6.40	0.005
Severe	6.54	3.05 to 10.03	<0.001
Age
18–20 years	Ref.
21–24 years	3.18	1.34 to 5.02	0.001
Education level
Primary or lower	Ref.
Some secondary or higher	−1.11	−3.12 to 0.91	0.282
Household size
≤4	Ref.
>4	−0.01	−2.09 to 2.08	0.995
Household generations
One	Ref.
Two	−2.95	−5.54 to −0.36	0.026
Three or more	−4.03	−7.02 to −1.05	0.008

Bold values indicate statistically significant results.

GAD, generalised anxiety disorder.

The conceptualisation of the CCA scale as a two-dimensional construct was considered by independently assessing factors associated with cognitive and functional impairment. Results of this indicated that predictors of CCA were the same as those for the two forms of impairment. This finding suggests that similar factors may predict both cognitive and functional impairment.

Discussion

In this cross-sectional analysis of baseline data of the TOPOWA study, we sought to determine key demographic risk factors associated with CCA. The findings of this study highlighted a significant level of awareness and understanding of climate change among the women surveyed, with 90% acknowledging its occurrence and 69.5% demonstrating knowledge on the subject. Despite this awareness, perceptions of the impact of climate change varied, with the majority viewing it negatively. Interestingly, a considerable proportion of respondents attributed global warming to natural processes, while a substantial minority recognised human activities as contributing factors. Over the past year, nearly half of the participants reported experiencing direct effects of global warming, predominantly in the form of storms, heat waves and floods. Whereas slums are often situated in environmentally precarious locations from climate hazards with no formal infrastructure such as drainage systems, residents further face intersecting vulnerabilities of low income, insecure tenure and limited access to health and education services. These exacerbate the overall impact of climate change on their mental well-being.65,67 Our observations re-emphasise the critical need for targeted educational initiatives to enhance comprehension of climate change’s causes and effects and to better equip communities to mitigate and adapt to its diverse impacts.

There were several important findings to highlight. Participants with severe CCA exhibited higher levels of GAD. Notably, 19.2% of those with severe CCA had severe GAD, compared with only 8.1% among those with minimal to mild CCA. The current study’s findings echo evidence provided in a factor analysis by Clayton and Karazsia,⁵⁴ which found a strong positive association between the CCA response (particularly cognitive and functional impairment) and general anxiety and depression. Also, among participants impacted by global warming in our study, varying levels of GAD were observed. Specifically, 11.2% of these participants had severe GAD, while 38.8% had mild GAD.

Also, intriguingly, lower education levels were linked to more severe CCA. Nearly half (48.1%) of participants with moderate to severe CCA had an education of primary or lower, compared with 24.4% with minimal to mild CCA. These results seem to contradict findings by Clayton and Karazsia,⁵⁴ who found higher education levels were associated with higher levels of functional and cognitive impairment due to CCA. However, there was limited variation in our sample: the vast majority (about 70%) of the 249 women analysed had some secondary or higher education level, which may have hindered our ability to fully examine the relationship between CCA and education across a wider range of educational attainment. Also, research has shown that while the youths in developed countries worry more because of increased knowledge (higher literacy) on climate change, those in developing countries like Uganda suffer the most from the direct effect of climate change which disproportionately affects the vulnerable such as the less educated living in slum areas.4041 68,71 Whereas informal settlements in high-income countries also face climate risk, their governments typically have the capacity and have put up mitigation measures including early warning systems, robust infrastructure and faster recovery and adaptation, unlike many low-income countries.^{67 72}

We also found that participants with severe CCA were more likely to live in smaller households (67.3%) and with their parents (69.2%). Similarly, smaller household sizes (four or fewer members) were associated with more severe CCA (67.3%). Also, women in single-generation households reported higher levels of severe CCA (30.8%), whereas women living in multi-generational households were more likely to have minimal to mild anxiety (49.2% in two-generation households). The rationale or explanation for this finding is unclear but worth exploring in future research to determine if there are any key confounders to explain these associations. Additionally, participants with both parents alive tended to have minimal to mild anxiety (67.5%), while those with one parent alive showed higher levels of moderate to severe anxiety (34.6%). This finding is probably driven by the level of hardship experienced overall but would contradict the findings that a higher percentage of participants with severe anxiety lived with their parents (69.2%) compared with those with minimal to mild anxiety (57.9%). Again, these complex findings will be important for future studies to examine.

In the current study, having biological children was associated with more severe CCA, with 71.2% of those with severe CCA having children, compared with 58.9% with minimal to mild CCA. Climate change may present an additional stressor for parents as they worry about their children’s future. These findings are in line with previous studies. For example, a systematic review of qualitative studies found that worry for future generations was a prominent theme among participants, particularly among parents who worry over their children’s well-being in the face of climate change.⁷³ Ekholm and Olofsson⁷⁴ also found parenthood was a significant determinant that serves to increase worry over climate change.

In multivariable analyses, we found several important associations between various factors and the levels of CCA among the participants. Higher levels of GAD were significantly associated with more severe CCA, with severe GAD showing a strong positive correlation. Age also played a significant role, with participants over 20 years old exhibiting higher levels of anxiety related to climate change. Higher education levels trended towards lower CCA, though not all associations reached statistical significance. Household structure, particularly the number of generations in the household, was also significant, with multi-generational households showing lower levels of CCA. These findings underscore the multifaceted nature of CCA and highlight the importance of considering mental health, demographic factors and household dynamics in addressing and mitigating CCA in communities.

While research in this area is emerging rapidly, it is clear to date that CCA reveals varying levels of awareness and concern across different regions. A study conducted in Rwanda, the Democratic Republic of Congo (DRC) and Morocco found that only 11.64% of respondents reported frequent eco-anxiety.³⁹ This finding aligns with a systematic review by Soutar and Wand,⁷³ which highlighted that climate-related anxiety is often limited by people’s knowledge of climate change. While residents of sub-Saharan Africa experience the direct and indirect effects of climate change, many do not recognise these impacts as being related to the climate crisis.⁷⁵ For example, a study in Ghana on knowledge, attitudes and adaptation to climate change revealed that only 43.9% of participants could define what climate change is.⁷⁶ Similarly, research in Zimbabwe indicated that communities facing poverty have limited access to reliable information about climate change, which may contribute to their lack of awareness.⁷⁷

Study limitations

This study has several important limitations that should be acknowledged when interpreting the findings. First, the cross-sectional design limits the ability to establish causality between the observed associations and CCA. We will be able to examine these issues prospectively when the data becomes available for this prospective cohort study. These planned analyses will provide more robust insights into how these factors influence anxiety over time. Second, the reliance on self-reported data may introduce bias, as participants might underreport or overreport their anxiety levels and experiences with climate change impacts due to social desirability or recall bias. Additionally, the study sample may not be fully representative of the broader population, as it focuses specifically on women living in poverty across three slums in Kampala. This sample may limit the generalisability of the findings to other populations or regions with different demographic and socio-economic profiles. The study also does not account for other potential confounding variables, such as access to mental health services, social support networks or individual coping mechanisms, which could influence the levels of anxiety reported. Moreover, while the study assesses the perceived impact of climate change, it does not objectively measure environmental factors or verify participants’ experiences with climate-related events. Future research should integrate objective environmental data to corroborate self-reported impacts. Moreover, the CAS deployed in this survey has not been specifically validated in this population; hence, this may affect the cultural validity of our measurements. Additionally, the categorisation of education levels and household dynamics may oversimplify the complex socio-economic factors that contribute to CCA. More nuanced measures and a broader range of socio-economic indicators could provide a deeper understanding of the underlying causes of anxiety in relation to climate change, particularly in low-resource settings. Finally, in the urban slums of Kampala, climate anxiety is likely intensified by slum-specific stressors such as chronic poverty, inadequate housing and exposure to violence, which collectively heighten young women’s sense of vulnerability and powerlessness in the face of environmental threats. While these contextual factors warrant comparison with other settings, such analyses were beyond the scope of this study, as all participants in this study resided in urban slums. Despite these limitations, the study provides valuable insights into this emerging research by highlighting factors associated with CCA and areas for further investigation for young women living in poverty. Also, while research and new approaches are emerging in this area, research indicates that climate change can be identified and reliably measured.⁵⁴

Conclusion

This baseline assessment of the prospective TOPOWA cohort study of young women identified several demographic characteristics associated with CCA. The findings revealed a high level of awareness and understanding of climate change among participants, yet perceptions of its impact varied significantly. Severe CCA was linked to higher levels of GAD, lower education levels, smaller household sizes and having biological children. The study also highlighted the complexity of familial influences on anxiety, with multi-generational households providing a mitigating effect.

These insights emphasise the need for targeted educational initiatives and community-based interventions to address and mitigate the impacts of CCA. Future research should focus on prospective assessments or mixed-method approaches and establish causal relationships and incorporate objective environmental data to validate self-reported experiences. Additionally, exploring the role of social support networks and access to mental health services will be crucial in developing comprehensive strategies to support those most affected by climate change and may also be a fruitful area for new and innovative technology including mHealth.⁷⁸ As Rother and colleagues⁷⁹ point out, there is a concerning scarcity of mental health research focused on climate change and its impact on young people in urban areas of sub-Saharan Africa, including Uganda, even though these populations are expected to bear the brunt of its negative consequences.⁸⁰ As such, this emerging area of research needs to be prioritised.

As the next steps, we recommend the implementation of educational programmes aimed at increasing understanding of climate change’s causes and effects, particularly in lower-education communities. Developing support systems within multi-generational households and enhancing mental health services access will also be vital. By continuing to investigate these areas, we can better equip communities to adapt to the diverse impacts of climate change and improve overall mental health resilience.

Data availability statement

Data are available upon reasonable request.