Association between climatic conditions and infectious diseases: implications for Central Ethiopia region

Abstract

Introduction

Ethiopia faces heightened climate-health vulnerability due to topography, subsistence farming, and socio-economic limitations.

Objective

This study analyzes the association between climate variables and infectious disease incidence in Central Ethiopia from 2013 to 2024.

Methods

This study investigates climate-sensitive epidemiological trends of enteric fever, malaria, diarrheal diseases, and bacterial meningitis in Central Ethiopia (2013–2024), utilizing hospital-based surveillance and meteorological data with descriptive statistics, Spearman’s correlation, and negative binomial regression to quantify disease burden.

Results

Enteric fever accounted for the highest disease burden (48.5% of 113,456 cases), with a consistent female predominance. Malaria and diarrhea exhibited strong positive correlations with rainfall (ρ = 0.846, ρ = 0.99) and minimum temperature (ρ = 0.625, ρ = 0.482), peaking during wetter months. In contrast, meningitis and enteric fever were negatively associated with rainfall (ρ=–0.951, ρ=–0.554), with meningitis showing high variability (CV = 94.6%). Regression analysis revealed rainfall increased malaria and diarrhea risk but reduced meningitis and enteric fever odds; minimum temperature significantly elevated malaria risk (OR = 1.17, p = 0.015).

Conclusion

This study highlights the intricate relationship between climatic variability and disease burden in Ethiopia, revealing a consistent predominance of female cases in health facility records. The gender disparity necessitates further inquiry into differential health-seeking behaviors, exposure risks, and diagnostic access, ideally through population-based incidence studies. Findings emphasize the urgency of tailored public health interventions, including gender-responsive strategies for enteric fever, climate-sensitive early warning systems for malaria and diarrheal diseases, and non-climatic approaches for meningitis control. Integrating real-time climate data into surveillance platforms is recommended to enhance outbreak preparedness. It is important to note that these findings, based on two hospital datasets in Central Ethiopia, may have limited generalizability to the wider regional context.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-12113-9.

Introduction

Climate change has become a defining global challenge of the 21st century, significantly undermining ecological stability, human health, and socio-economic development [1]. Its far-reaching implications extend beyond conventional environmental concerns, profoundly influencing public health systems by altering epidemiological patterns of climate-sensitive infectious diseases [2, 3]. Variations in temperature, precipitation, and humidity can modify vector habitats, pathogen viability, and host behaviors. These changes collectively reshape the landscape of disease transmission. The Eco-social Disease Transmission Model offers a critical framework for understanding these complexities in Ethiopia, where intertwined ecological and demographic factors modulate disease risks. This model posits that climate-induced shifts in vector ecology, especially in transitional zones between highland and lowland regions, are compounded by population mobility, density, and health system capacity, thus necessitating integrated surveillance and adaptive response strategies [4].

Ethiopia presents a heightened susceptibility to climate-induced health risks due to its diverse topography, agrarian economy, and structural socio-economic challenges [5]. Reliance on rain-fed agriculture amplifies exposure to climatic variability, while limited infrastructure and poverty constrain resilience and adaptive responses [6]. In this context, vector-borne diseases such as malaria and dengue are not merely ecological phenomena but also socio-political issues that intersect with public health governance and equity. There is a pressing need to advance interdisciplinary research that integrates ecological modeling with health system diagnostics, particularly in regions where climatic perturbations intersect with vulnerable human systems [7, 8]. Such inquiry should inform targeted interventions and policy frameworks that enhance adaptive capacity and reduce disease burden across vulnerable populations [9].

Malaria is transmitted through the bite of infected female Anopheles mosquitoes, with transmission dynamics closely linked to temperature, rainfall, and humidity [10]. Warmer temperatures accelerate parasite development within mosquitoes, while rainfall creates breeding sites, making malaria highly sensitive to seasonal and climatic variations [11]. Enteric fever (typhoid and paratyphoid) is primarily spread via ingestion of food or water contaminated with Salmonella bacteria, often exacerbated by poor sanitation [12]. Meteorological factors such as heavy rainfall and flooding can compromise water quality and sanitation infrastructure, increasing transmission risk [13]. Diarrheal diseases, caused by a range of pathogens including viruses, bacteria, and protozoa, are similarly influenced by water quality and hygiene. High temperatures can promote pathogen survival, while rainfall and flooding increase exposure to contaminated sources [14]. Meningitis, particularly the bacterial form prevalent in the African meningitis belt, is transmitted through respiratory droplets. Its outbreaks are strongly associated with dry seasons, low humidity, and dusty conditions, which may damage mucosal barriers and facilitate pathogen spread [15].

Previous climate-health research in Ethiopia has predominantly concentrated on isolated infectious diseases or subjective perceptions of risk, leaving substantial gaps in our understanding of the specific pathways and thresholds through which meteorological variables influence disease emergence and intensification [4, 13, 16, 17]. Central Ethiopia serves as a critical case study in this regard, grappling with limited spatiotemporal analytical capacity, fragmented health and climate data systems, and under-resourced institutional frameworks [17]. Although there is growing recognition of climate change’s health implications, diseases sensitive to climatic fluctuations, such as malaria, enteric fever, diarrhea, and meningitis, remain insufficiently examined in relation to shifting climate patterns and are inadequately integrated into surveillance mechanisms [5]. This deficit is compounded by the region’s socio-ecological heterogeneity, encompassing diverse microclimates across highland, urban valleys, and rift margins, combined with urban expansion, demographic transitions, and evolving disease ecologies [18]. Existing studies, often constrained by coarse geographic scales or narrow disease foci, fail to capture the nuanced vulnerabilities and intra-regional disparities of Central Ethiopia, rendering the region markedly underrepresented in the broader climate-health discourse [19].

The present study distinguishes itself through its geographically targeted scope, centering on the Central Ethiopia Region to enable a more granular and context-sensitive analysis than the national, multi-regional approach adopted by Simegn et al. (2024), which examined six climate-sensitive diseases across Ethiopia. This localized focus permits the integration of fine-resolution climatic and epidemiological data that are typically obscured in national-level aggregations [20, 21]. Furthermore, the eco-social framing of this study shifts from overarching national policy narratives to a focused examination of region-specific determinants, including localized irrigation regimes, patterns of urban-rural migration, and land-use transformations that shape climate-disease interactions factors often homogenized in country-wide retrospective analyses [22]. This refined regional lens offers actionable insights for central government and regional health authorities, whereas broader studies primarily inform national-level public health emergency planning [19].

This study directly addresses these deficiencies by conducting a high-resolution observational analysis of multiple infectious diseases (malaria, diarrhea, enteric fever, meningitis) in relation to key meteorological variables (rainfall, temperature, and humidity) exclusively within Central Ethiopia. Moving beyond single-disease approaches, it provides a composite picture of climate-health vulnerability tailored to the region’s specific realities. Crucially, the research is designed not only to elucidate localized climate-disease linkages but to generate actionable evidence for regional adaptation [23]. It directly supports health bureaus, and disaster risk offices in developing localized early warning systems and guiding targeted investments in climate-resilient health infrastructure [24]. By aligning with Ethiopia’s decentralization efforts, the study offers a replicable model for hyper-local climate-health assessments in similarly complex regions [25].

Materials and methods

Study area

The study was conducted in the Central Ethiopia region, specifically in the Kambata Zone and Tembaro Special Woreda, as shown on the Fig. 1. Geographically, the region is located at approximately 7°40’00"N latitude and 38°0’00"E longitude. According to the National Meteorology Agency, the study area is characterized by three distinct seasons defined by rainfall distribution and three climatic zones determined by elevation. The seasons are: Kiremt (June to September), the summer season marked by heavy rainfall; Bega (October to January), a dry season with frosty mornings, particularly in January; and Belg (February to May), the autumn season with sporadic showers. These seasonal variations are critical in shaping the region’s agricultural and ecological patterns.

Fig. 1.

Geographical locations of the study area. Source: Ethio-GIS, 2023

The climatic zones of Kambata Zone and Tembaro Special Woreda are defined by elevation and associated temperature and rainfall patterns. The Kola zone (Lowland), situated below 1830 m, experiences an average annual temperature of approximately 27 °C and annual rainfall around 510 mm. The Woinadega zone (Midland), encompassing elevations between 1830 and 2440 m, has an average annual temperature of 22 °C and receives rainfall ranging from 510 to 1530 mm. The Dega (Highland) zone, located above 2440 m, is characterized by a cooler climate, with an average annual temperature of 16 °C and annual rainfall between 1270 and 1280 mm. These diverse climatic and geographical attributes influence the region’s environmental and socio-economic dynamics [5].

Study design

A retrospective longitudinal study was conducted to examine the association between meteorological parameters and the incidence of prevalent climate-sensitive infectious diseases in the study area over the past eleven years.

Data collection

Meteorological data encompassing temperature and rainfall records from 2013 to 2024 were obtained from the National Meteorological Agency (NMA) of Ethiopia. These datasets were delivered in digital formats compatible with statistical software, enabling streamlined processing and robust analysis of long-term climate trends. In parallel, retrospective epidemiological surveillance data on climate-sensitive diseases namely malaria, diarrheal disease, meningitis, and enteric fever were collected from facility-based case reporting systems at Dr. Bogalech Gebre Memorial General Hospital, which primarily serves midland and highland populations, and Mudula Primary Hospital, which predominantly covers lowland communities.

To ensure data validity, these hospital records were triangulated with corresponding case reports archived by the Ethiopian Public Health Institute (EPHI), thereby aligning the disease dataset with the meteorological timeframe for integrated analysis. All data acquisitions adhered to institutional ethical protocols. Rigorous data quality checks were conducted to ensure completeness and consistency, with missing values addressed through appropriate interpolation techniques. To capture spatial climate variability, two geographically distinct points at varying altitudes were selected within each district. Monthly averages of climatic variables from these locations were computed and subsequently used to examine associations with monthly disease incidence.

Statistical analysis

The study employed a multifaceted analytical approach for the climatic and disease data. Descriptive statistics, including measures such as mean, median, standard deviation, and frequency distributions, were used to systematically summarize the data. Advanced statistical software, including Microsoft Excel, IBM SPSS Statistics v25 and, R software ensured precision and reliability in data analysis. Correlation analysis was conducted to examine the relationship between climatic variables and disease incidence, using Spearman correlation coefficients. Furthermore, trends in climatic conditions and disease incidence over an 11-year period were analyzed through linear regression models and time-series analysis techniques. To address potential confounding factors and assess the independent effects of climatic variables on disease outcomes, negative binomial regression was applied, tailored to the distribution of the data.

Data analysis

Rainfall Variability.

Coefficients of Variation (CV) were calculated to evaluate the variability of the rainfall. A higher CV value indicates larger variability, and vice versa, which is computed as:

Where CV is the coefficient of variation,

σ is standard deviation and µ is the mean precipitation. According to Hare (2003), CV is used to classify the degree of variability of rainfall events as less (CV < 20), moderate (20 < CV < 30), and high (CV > 30).

Effect of climate change on infectious disease

Descriptive statistical analyses like mean, standard deviation, and total numbers were conducted for the overall eleven years of the study period. In addition, a graphical presentation displaying trends in temperature and infectious disease cases over time were done. Spearman correlation coefficients were employed to analyze the association between the incidence of infectious diseases and each of the meteorological variables (MAT, MTmax, MTmin, and MCP). The lag effect of the meteorological variables on the incidence were considered in the correlation analysis. Thus, if the effect of temperature and precipitation are lagged by one month based on the assumption that a month’s average temperature and precipitation may be related to the number of cases in the following month a P < 0.05 was considered as statistically significant.

Negative binomial regression (NBR) is a statistically rigorous approach widely used to model overdispersed count data, where the variance exceeds the mean a common feature in epidemiological datasets. Its relevance is particularly pronounced in the analysis of climate-sensitive diseases such as malaria, diarrhea, meningitis, and enteric fever, where disease incidence is influenced by environmental variability. Unlike Poisson regression, which assumes equidispersion, NBR incorporates a dispersion parameter to account for unobserved heterogeneity, thereby yielding more reliable and interpretable estimates. This methodological advantage enables researchers to assess the influence of climatic factors (temperature, rainfall, and humidity) on disease dynamics, supporting evidence-based public health interventions and climate adaptation strategies [26, 27]. In this study, lag variable selection was guided by the Cross-Correlation Function (CCF) between monthly detrended and deseasonalized disease incidence and meteorological variables, with the most significantly correlated lag (up to six months) retained to reflect biologically plausible latency periods [28]. Missing data were addressed using Multiple Imputation with Chained Equations (MICE), under the assumption of Missing At Random (MAR), and final estimates were derived by pooling five imputed datasets using Rubin’s rules. Model selection was further refined by validating the use of the negative binomial family over the Poisson.

Results

As shown in Table 1, Enteric Fever was the most prevalent disease, accounting for nearly half (46.1%) of all reported cases across the eleven-year period underscoring its epidemiological dominance and need for targeted public health intervention. Diarrheal disease and malaria follow as substantial contributors to the overall disease load, comprising 28.3% and 25.5% of cases respectively, suggesting persistent challenges in water sanitation and vector control. Bacterial Meningitis, though clinically severe, represents only 0.03% of cases, potentially reflecting underreporting or episodic outbreaks rather than systemic endemicity.

Table 1.

Distribution of diseases according to gender (2013–2024)

Disease	Male patients	Female patients	Total patients
Malaria	13,712 (47.3%)	15,273 (52.7%)	28,985 (100%)
Diarrheal Disease	15,296 (48.0%)	16,778 (52.0%)	32,074 (100%)
Bacterial Meningitis	18 (47.0%)	20 (53.0%)	38 (100%)
Enteric Fever	25,342 (48.4%)	27,027 (51.6%)	52,359 (100%)
Total	54,368(47.9%)	59,088 (52.1%)	113,456 (100%)

Table 2 reveals distinct variability across both infectious diseases and meteorological indicators from 2013 to 2024, underscoring complex, potentially interlinked ecological and public health patterns. Malaria demonstrates considerable fluctuation (CV: 41.9%), with its highest prevalence occurring in 2024, a year marked by the lowest average temperature (16.67 °C). This may indicate a sensitivity to cooler conditions, challenging conventional assumptions about malaria’s climatic thresholds. Diarrhea showed its peak incidence in 2013 aligned with the year of maximum rainfall suggesting potential associations between heavy precipitation and waterborne diseases. However, by 2023, diarrhea cases dropped to their lowest, possibly reflecting improvements in water sanitation or adaptive public health measures.

Table 2.

Descriptive analyses of infectious diseases and meteorological data (2013–2024)

Infectious diseases	Mean	Std. deviation	Max	Year (Max)	Min		Year (Min)	CV
Malaria	2415	1013	5127	2024	1371		2015	41.9%
Diarrhea	2672	801	4167	2013	1829		2023	29.9%
Meningitis	3.17	3	9	2017	0	2021, 2023, 2024		94.6%
Enteric Fever	4363	556	5013	2015	3344		2013	12.7%
Meteorological data
Rainfall (mm)	104.2	30.75	156.34	2013	71.41		2015	29.5%
T Ave (°C )	18.26	0.77	19.24	2015	16.67		2024	4.2%
T Min (°C )	6.54	1.064	8.34	2022	5.07		2015	16.3%
T Max (°C )	30.95	4.18	34.19	2016	21.99		2023	13.5%

Notably, meningitis exhibited the highest relative variability (CV: 94.6%), with zero recorded cases in 2021, 2023, and 2024. This sporadic pattern may point to strong dependency on localized factors such as population mobility or microclimatic shifts rather than broader climate trends. Meanwhile, enteric fever presented the most stable disease profile (CV: 12.7%), peaking in 2015 coinciding with maximum average temperature hinting at a more predictable seasonal behavior potentially tied to food safety or water quality issues.

Among meteorological variables, maximum temperature displayed pronounced variability (CV: 13.5%), reaching extreme highs in 2016, a year not marked by peak disease prevalence indicating that temperature spikes alone may not be sufficient drivers. Rainfall, while “highly variable” due to its ecological implications and seasonal amplitude (CV: 29.5% is slightly below the “high” threshold), lacked direct annual alignment with most disease peaks, suggesting that temporal lags or cumulative climatic stressors may play more critical roles. Together, these findings highlight the nuanced, disease-specific interactions with meteorological forces, emphasizing the need for tailored, climate-responsive surveillance and intervention strategies.

Seasonality significantly influences disease dynamics. Malaria displays pronounced peaks aligned with temperature and rainfall (Fig. 2), yet its unexpected surge during cooler 2024 conditions challenges temperature-based transmission models. This anomaly necessitates further investigation into vector ecology, parasite thresholds, and microclimatic refugia in highland Ethiopia to enhance predictive accuracy. Diarrhea also follows seasonal trends, likely shaped by sanitation and water quality, while enteric fever remains relatively stable with minimal environmental correlation. Meningitis emerges sporadically, potentially linked to overcrowding or socio-economic stressors. Among climatic variables, average and maximum temperatures show strong positive correlations with malaria and diarrhea, suggesting warmer conditions promote transmission. Minimum temperature exerts a weaker influence but may affect cold-sensitive diseases. These findings highlight the intricate climate disease relationship.

Fig. 2.

Monthly prevalence trends of four major diseases (malaria, enteric fever, diarrhea, and pneumonia) derived from hospital records (2013–2024), highlighting intra-annual variability and peak transmission periods

Malaria, diarrhea, enteric fever, and meningitis exhibit distinct patterns of occurrence over time. Malaria demonstrates a cyclical pattern, with peaks occurring approximately every 6–12 months, and the highest peak of 623 cases. Diarrhea also follows a cyclical trend, though the peaks are less pronounced, with the highest peak of 747 cases. In contrast, enteric fever presents a more erratic pattern, marked by occasional spikes, with the highest peak of 699 cases. Meningitis shows a relatively low incidence throughout the period, with the highest peak of 5 cases.

Figure 3 illustrates the monthly prevalence of four diseases (malaria, diarrhea, enteric fever, and meningitis) over a year (January to December), revealing distinct seasonal patterns. Malaria exhibits a pronounced seasonal trend with peaks in between September and December, maintaining a relatively low prevalence throughout the year. Diarrhea demonstrates a bi-modal distribution with peaks in May and August, although its overall prevalence is higher than that of Malaria. Enteric fever shows a clear seasonal peak in January, February and December but remains relatively low throughout the year. Meningitis, in contrast, displays a subtle seasonal variation with a slight peak in January, and February and its prevalence remains minimal.

Fig. 3.

Seasonal patterns of disease incidence aligned with meteorological fluctuations (2013–2024), illustrating potential climate-sensitive transmission cycles

The seasonal patterns observed underscore the influence of environmental and behavioral factors on disease transmission. Malaria’s consistently high prevalence, particularly during its peak months, highlights its substantial public health burden. The bi-modal trend of diarrhea suggests multiple causative factors, potentially linked to variations in weather, food hygiene practices, and water quality. Meanwhile, the lower and less pronounced seasonal prevalence of enteric fever and meningitis may be attributed to factors such as effective vaccination programs and improved sanitation measures, emphasizing the importance of preventive strategies in mitigating these diseases.

Figure 4 below, which spans the study period and includes precipitation (mm), relative humidity at 2 m (RH2M, %), maximum temperature (T2M-MAX, °C), and minimum temperature (T2M-MIN, °C) the climatic trends reveal marked variability across parameters, with discernible seasonal and temporal patterns: Precipitation exhibits pronounced fluctuations, with peaks exceeding 500 mm followed by sharp declines to near-zero levels, reflecting the cyclical nature of rainfall and potential monsoonal influences. Relative humidity aligns closely with precipitation patterns, reaching highs near 89% during wetter periods and dipping to below 40% during dry spells, indicating strong hydro-meteorological coupling. Maximum temperatures show variability ranging from 21 °C to over 33 °C, with elevated temperatures often preceding or trailing low humidity phases, suggesting possible links to dry season intensification. Minimum temperatures remain more subdued, spanning 5 °C to 14 °C, yet they too mirror seasonal shifts with slight lags compared to maximum temperature spikes.

Fig. 4.

Temporal trends of key meteorological variables (rainfall, temperature, humidity) based on EMA data (2013–2024), showing interannual variability and potential climate anomalies

Overall, the dataset suggests a climate system characterized by distinct wet and dry periods, each impacting both humidity and temperature dynamics. This interplay of variables may carry significant implications for agro-ecological planning, climate-health vulnerability assessments, and resilience strategy development in Ethiopia.

Potential correlations between diseases and climatic factors were observed. For instance, malaria and diarrhea show a plausible association with Rainfall, as peaks in disease cases coincide with periods of increased rainfall, particularly in 2014, 2017, and 2019. However, the relationship between temperature trends and disease patterns is less clear, as the temperature data indicate a steady rise, while disease incidences exhibit varied trends. These observations underscore the complex interplay between climatic conditions and public health, necessitating further investigation to better understand the underlying dynamics.

The analysis of meteorological and disease data reveals significant correlations, as detailed in Table 3. Rainfall demonstrates a strong positive correlation with malaria (ρ = 0.84, p = 0.00) and a very strong positive correlation with diarrhea (ρ = 0.99, p = 0.00). However, these exceptionally high coefficients likely reflect strong seasonal patterns and temporal autocorrelation rather than a direct causal link. They indicate a powerful statistical association that may be influenced by unmeasured confounding variables and should not be interpreted as proof of causation. Conversely, meningitis incidence shows a strong negative correlation with rainfall (ρ = -0.951, p = 0.00). Similarly, enteric fever exhibits a moderate negative correlation (ρ = -0.554, p = 0.00), suggesting that increased rainfall is associated with fewer cases of both meningitis and enteric fever.

Table 3.

Spearman’s correlation between diseases and meteorological factor (2013–2024)

Meteorological factor	Malaria		Diarrhea		Meningitis		Enteric fever
	ρ	p-value	ρ	p-value	ρ	p-value	ρ	p-value
Rainfall(mm)	0.84	0.00	0.99	0.00	-0.951	0.00	-0.554	0.00
Average Temp. (°C)	0.20	0.016	-0.24	0.004	0.080	0.338	0.231	0.005
T2M Max (°C)	-0.56	0.00	-0.44	0.00	0.312	0.00	0.405	0.00
T2M Min(°C)	0.62	0.00	0.48	0.00	-0.428	0.00	-0.452	0.00

The high magnitude of Spearman’s correlation coefficients reported in Table 4 such as ρ = 0.99 for rainfall and diarrhea, and ρ = − 0.951 for rainfall and meningitis suggests a very strong statistical association. However, these values should be interpreted with caution to avoid overstating causality. Such extreme correlations may reflect underlying confounding effects or temporal autocorrelation, rather than direct causal links. For instance, seasonal patterns in meteorological variables often coincide with seasonal disease incidence, which can inflate correlation measures. Similarly, long-term trends such as gradual climate shifts or changes in public health interventions over the 2013–2024 period might introduce spurious associations. Without accounting for these temporal structures or potential confounders like population density, sanitation levels, or socioeconomic factors, the observed relationships could misleadingly imply causation where none exists. Therefore, further analyses using time-series adjustment methods or multivariate models are needed to clarify the nature of these associations.

Table 4.

Negative binomial regression analysis of meteorological variables on disease incidence (2013–2024)

Disease	Meteorological variable	Coefficient (Estimate)	P-value	IRR	95% C.I. for IRR		Omnibus Test (x2)
					Lower	Upper
Malaria	Rainfall (mm)	0.007	0.000	1.007	1.004	1.009	91.361
	Average Temp. (°C)	0.127	0.500	1.127	0.795	1.598
	Max Temp.(°C)	-0.151	0.058	0.860	0.736	1.005
	Min Temp. (°C)	0.157	0.015	1.170	1.031	1.328
Diarrhea	Rainfall (mm)	0.009	0.000	1.009	1.006	1.012	85.409
	Average Temp. (°C)	-0.060	0.741	0.941	0.659	1.345
	Max Temp.(°C)	-0.009	0.916	0.991	0.844	1.164
	Min Temp. (°C)	0.087	0.188	1.091	0.958	1.243
Meningitis	Rainfall (mm)	-0.131	0.002	0.877	0.808	0.952	114.385
	Average Temp. (°C)	-0.254	0.751	0.776	0.161	3.727
	Max Temp.(°C)	0.146	0.713	1.157	0.532	2.519
	Min Temp. (°C)	0.094	0.857	1.098	0.396	3.042
Enteric fever	Rainfall (mm)	-0.006	0.000	0.994	0.992	0.997	32.203
	Average Temp. (°C)	0.062	0.729	1.064	0.749	1.512
	Max Temp.(°C)	-0.019	0.816	0.981	0.834	1.154
	Min Temp. (°C)	-0.029	0.677	0.971	0.847	1.114

Note: The dependent variable is the count of disease cases. IRR = Incidence Rate Ratio; an IRR > 1 indicates a positive association with the disease count, while an IRR < 1 indicates a negative association. C.I. = Confidence Interval. The meteorological variables are defined as follows: Rainfall (total monthly precipitation in mm), Average Temp. (mean monthly temperature in °C), Max Temp. (mean monthly maximum temperature in °C), Min Temp. (mean monthly minimum temperature in °C)

Regarding temperature, the average temperature exhibits a weak positive correlation with malaria (ρ = 0.2, p-value = 0.016), indicating a slight association between higher average temperatures and increased malaria. For diarrhea, there is a weak negative correlation (ρ = -0.24, p-value = 0.004), suggesting a slight decrease in diarrhea cases as average temperatures rise. Meningitis shows a very weak positive correlation (ρ = 0.080, p-value = 0.338), which is not statistically significant. Enteric fever has a weak positive correlation with average temperatures (ρ = 0.231, p-value = 0.005), indicating a slight association between higher average temperatures and increased enteric fever cases.

The maximum temperature (T2M Max) demonstrates a moderate negative correlation with malaria (ρ = -0.560, p-value = 0.00), indicating a significant association between higher maximum temperatures and increased malaria. Diarrhea, in contrast, shows a moderate negative correlation (ρ = -0.44, p-value = 0.00), suggesting that higher maximum temperatures are associated with fewer diarrhea cases. Meningitis shows a weak positive correlation (ρ = 0.312, p-value = 0.00), indicating a slight increase in meningitis cases with higher maximum temperatures. Typhoid has a moderate positive correlation with maximum temperature (ρ = 0.405, p-value = 0.00), suggesting a significant association between higher maximum temperatures and increased enteric fever cases.

For minimum temperature (T2M Min), malaria shows a moderate positive correlation (ρ = 0.62, p-value = 0.00), indicating a significant association between higher minimum temperatures and increased malaria. Diarrhea also exhibits a moderate positive correlation (ρ = 0.48, p-value = 0.00), suggesting that higher minimum temperatures are associated with more diarrhea cases. Meningitis and enteric fever both show moderate negative correlations (ρ = -0.428, p-value = 0.00 for meningitis and ρ = -0.452, p-value = 0.00 for enteric fever), indicating that higher minimum temperatures are linked to fewer cases of these diseases.

The regression analysis on the Table 4 offers a multidimensional view of how meteorological variables are associated with disease incidence across four health outcomes: malaria, diarrhea, meningitis, and typhoid, demonstrating variable-specific climate sensitivities. Rainfall consistently emerges as a statistically significant predictor (p < 0.01) across all diseases, yet its directionality varies: it increases the risk for malaria and diarrhea, with an incidence rate increase of approximately 0.7% and 0.9% per unit rise, respectively, while conversely reducing the likelihood of meningitis and enteric fever by 12.3% and 0.6%. These contrasting probability shifts suggest rainfall acts as both a vector-amplifier and a dilution mechanism depending on disease ecology, enhancing transmission for waterborne and vector-related illnesses, but potentially suppressing airborne or contact-spread diseases.

Temperature effects are more nuanced. Minimum temperature has a positive and significant association with malaria (p = 0.015), increasing its odds by 17%, aligning with literature on elevated nighttime temperatures enhancing mosquito survival. In contrast, temperature variables show weak or statistically insignificant influence on diarrhea and enteric fever, reinforcing rainfall’s primacy in their transmission pathways. Meningitis appears uniquely sensitive, with only rainfall yielding significance, pointing to its seasonal pattern potentially aligning with dry periods rather than thermal thresholds.

Comparatively, malaria and meningitis demonstrate the highest model fitness (χ² = 91.361 and 114.385), signifying strong explanatory power of meteorological variables, while enteric fever lags in predictive strength. This divergence underscores the need for tailored climate-health surveillance strategies malaria, meningitis, and diarrhea may benefit from meteorologically driven early warning systems, whereas enteric fever might require integrated approaches addressing broader socio-environmental determinants.

Discussion

Unlike previous studies in Ethiopia that have either concentrated on a single infectious disease or relied predominantly on public perceptions of climate change and disease transmission, this research represents an observational analysis that systematically examines multiple infectious diseases in relation to three key weather variables [29–32]. Anchored in the Eco-social Disease Transmission Model, the study elucidates how intertwined environmental, social, and infrastructural determinants shape infectious disease burdens within the Central Ethiopia Region [33]. The model underscores that health outcomes arise from dynamic interplays among socio-political structures, ecological stressors, and behavioral practices a perspective particularly salient for resource-limited settings facing evolving climatic conditions. Through an eleven-year dataset, the study reveals that enteric fever, diarrhea, malaria, and meningitis are not isolated biomedical events but manifestations of deeper systemic vulnerabilities rooted in inadequate water sanitation, gendered disparities, and climate variability. These findings resonate with prior research conducted in Bangladesh [34], reinforcing the transnational relevance of eco-social frameworks in understanding climate-sensitive disease burdens.

The eco-social model represents a transformative leap in public health research by bridging ecological drivers with nuanced social vulnerabilities. In the case of enteric fever, the model linked environmental exposure contaminated water sources with gendered social roles, revealing that women and girls may face disproportionate risk due to their responsibilities in water collection and sanitation. This insight moved beyond geographic targeting to advocate for gender-sensitive interventions. Similarly, for rainfall-sensitive diseases like malaria and diarrhea, the model enabled ecological data (e.g., meteorological patterns) to inform timely public health actions, such as early warning systems and preventative campaigns, demonstrating a shift from passive correlation to active forecasting. Crucially, the model also identified when ecological factors were not the dominant drivers, as seen with meningitis, where climate was ruled out, prompting a pivot toward non-climatic strategies like vaccination and infrastructure improvements. Altogether, the eco-social framework redefined disease modeling from a static environmental analysis into a dynamic, context-aware tool for precision public health policy.

The high prevalence of enteric fever and diarrheal disease, comprising over 70% of total cases, signals chronic failures in water management and food hygiene, exacerbated by episodic rainfall spikes which aligns with previous findings [17, 34]. Rainfall’s strong positive correlation with both diseases (ρ >0.84) underscores its role as a key ecological trigger. Drawing from comparative insights in South Asia and the Sahel, these findings mirror regional evidence that poor drainage and inadequate household-level water storage amplify fecal-oral transmission post-precipitation [35]. The eco-social lens calls for intersectoral interventions particularly the integration of climate-resilient WASH (Water, Sanitation, and Hygiene) systems tailored to seasonal flood-prone zones [33].

These findings on climate-disease relationships strongly align with studies like Simegn et al. (2024), which also identified rainfall and temperature as key drivers for malaria and diarrheal diseases; the strong positive correlation with rainfall is a consistent national finding [19]. Conversely, the pronounced gender disparity in enteric fever incidence contrasts with some local studies, such as Alemayehu et al. (2024), which may not have reported such a significant skew, suggesting potential regional or socio-behavioral variations in disease risk factors across the country [36]. Furthermore, while the strong positive link between rainfall and both malaria and diarrhea aligns with established previous findings Simegn et al. 2024, the observed negative correlation for meningitis (ρ=−0.951) and enteric fever is a notable contrast, highlighting critical disease-specific climatic sensitivities and potential disparities with existing studies like Alemayehu et al. (2024) [19, 36].

Malaria’s seasonal and ecological sensitivity especially its peak during cooler, wetter months challenges conventional assumptions tied to warmer climates. The moderate positive association with minimum temperature (ρ = 0.625) and a 17% increase in odds per unit rise in minimum temperature highlight a vector ecology responsive to nighttime warmth and residual moisture, aligning with observations from highland Kenya and southern Rwanda [37]. Targeted vector-control campaigns should be restructured to anticipate low-temperature windows with high humidity, rather than relying solely on rainfall calendars. Furthermore, predictive climate-disease models could inform early warning systems, incorporating local meteorological data and community risk perception [38].

The observed gender disparity in disease case distribution marked by a consistently higher prevalence among female patients across all diagnostic categories underscores the influence of socio-structural determinants that extend beyond biological susceptibility [39]. Interpreted through an eco-social lens, this pattern reflects the compounded effects of gendered caregiving responsibilities, restricted mobility, and systemic barriers to healthcare access [40]. In the study region, cultural norms frequently constrain women’s autonomy in navigating public spaces, while inadequate transportation infrastructure particularly in rural settings exacerbates these limitations. Security concerns, including fear of harassment and gender-based violence, further deter women from seeking medical care or participating in community health initiatives [41]. Structurally, healthcare facilities are often geographically distant and misaligned with the temporal and logistical demands of caregiving roles. The absence of gender-responsive services such as maternal health provisions, female healthcare personnel, and privacy safeguards compounds these challenges. Economic precocity and limited decision-making power within households further inhibit timely and appropriate healthcare utilization [39]. These findings collectively highlight the urgency of gender-sensitive interventions, including the development of local water infrastructure, targeted hygiene education, and culturally appropriate health communication strategies that address gendered risk perception and health-seeking behavior.

The sporadic occurrence of meningitis’ and negative correlation with rainfall (ρ = -0.951) indicate climate-linked seasonality often tied to dry, dusty conditions that facilitate airborne transmission. This finding aligns with existing evidence linking meningitis outbreaks to dry and dusty conditions, often observed in the meningitis belt of Africa [19, 42]. In Central Ethiopia Region, the lack of consistent cases yet occasional spikes suggests localized outbreak-prone pockets that may align with micro-climatic zones or population mobility corridors [42]. Surveillance systems should integrate socio-demographic profiling and localized meteorological forecasting to inform vaccine deployment and public health preparedness.

Comparative analysis with other regions reveals the resilience dividend of long-term, climate-informed public health investments. For instance, in Tanzania’s rural districts, incorporating rainfall data into malaria control strategy reduced vector density by 30% [43]. Ethiopia’s surveillance and response frameworks could benefit from similar data-to-action platforms, integrating disease incidence, meteorological data, and spatial mapping. Tailored dashboards for district health offices may enhance real-time decision-making and cross-sectoral collaboration, especially with local agriculture and water authorities [44].

While enteric fever exhibited epidemiological stability, its moderate association with elevated temperatures and slight attenuation during periods of heavy rainfall indicates a nuanced interaction among food safety, urban sanitation, and seasonal behavioral patterns. This observation corroborates earlier research [45, 46]. In Ethiopia’s Amhara Region, rapid urbanization and the proliferation of informal food markets may elevate vulnerability during warmer months [13]. A comparable trend was documented in peri-urban Ghana, where heightened street food consumption during hot seasons intensified enteric fever transmission risk [47]. To mitigate seasonal exposure, regulatory authorities should adopt climate-responsive food hygiene inspections and targeted public health education initiatives within densely populated trade corridors.

This study has several methodological and contextual limitations that warrant consideration. Its reliance on data from only two hospital facilities may limit generalizability, as these sites may not capture the demographic and epidemiological diversity of Central Ethiopia. The retrospective design and dependence on routine hospital records and EPHI surveillance data introduce risks of diagnostic inconsistencies, underreporting, and temporal variability in case definitions. Additionally, the analysis did not account for key confounders such as urbanization, WASH interventions, and vaccination campaigns, which may have independently shaped disease patterns. Given the observational nature of the study and potential temporal autocorrelation, correlation should not be interpreted as causation. To improve causal inference and policy relevance, future research should integrate multi-source data, control for contextual confounders, and adopt prospective designs that better capture the complex interplay between climate variability and health outcomes.

The study underscores how an eco-social lens shifts our understanding of disease incidence from isolated clinical cases to indicators of systemic vulnerability and resilience deficits, particularly in the Central Ethiopia Region. By embedding localized epidemiological insights into surveillance, community engagement, and cross-sectoral planning, health systems can evolve from reactive care to forward-looking, climate-resilient strategies. Findings reveal that temperature and rainfall significantly influence the occurrence of climate-sensitive infectious diseases particularly malaria and diarrhea, signaling the need to enhance early warning systems and response mechanisms. Although constrained by a short timeframe, retrospective data from two hospitals, and limited funding, the study’s triangulation with Ethiopian Public Health Institute (EPHI) data lends credibility to its conclusions. These insights serve as a foundational platform for broader cohort studies, encouraging future research to integrate behavioral epidemiology and predictive modeling to track feedback loops between climate variability, perception, and ecological exposure. Ultimately, understanding disease-specific climate interactions will empower stakeholders to develop targeted interventions that mitigate health risks in Ethiopia and comparable global contexts.

Conclusion

This study elucidates the multifaceted associations between climatic variables and infectious disease incidence across the Central Ethiopian region during the period 2013–2024. Enteric Fever was the most prevalent condition, with disproportionately higher incidence among females. Statistically significant and seasonally patterned correlations were observed between malaria and diarrhea vis-à-vis rainfall and minimum temperature, while meningitis cases were associated predominantly with arid conditions.

These findings underscore the imperative to incorporate eco-social frameworks into climate-health research, thereby accommodating the inherently nonlinear and context-contingent nature of climate-disease interactions. Policy-relevant implications include the development of climate-responsive and gender-sensitive disease surveillance systems tailored to local epidemiological patterns. Specifically, this entails deploying rainfall-based early warning mechanisms for malaria and diarrhea, implementing gender-responsive intervention strategies for enteric fever, and prioritizing non-climatic determinants in the targeted management of meningitis.

Future research directions should focus on disentangling temporal lag effects of climatic stressors, advancing the granularity of microclimatic disease mapping, evaluating institutional preparedness for climate-health integration, refining vulnerability modeling using socio-demographic variables, and enhancing culturally grounded climate communication strategies to bolster community-level resilience and inform precision public health responses.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

Ethiopian

Central Statistical Agency

EMA

Ethiopian Meteorological Agency

CC

Climate change

IPCC

Intergovernmental Panel on Climate Change

WHO

World Health Organization

Author contributions

I.H. contributed to the study by conceptualizing and designing the research, preparing materials, collecting and analyzing data, and drafting the original manuscript. K.E. played a significant role in reviewing and editing the manuscript as well as engaging in scientific discussions. G.T. provided critical reviews of the scientific content, contributed to manuscript editing, and supervised the preparation of the final version. All authors participated in reviewing and providing feedback on previous versions of the manuscript and gave their approval for the final version.

Funding

Not applicable.

Data availability

The datasets utilized and/or analyzed in this study are accessible from the corresponding author upon reasonable request, subject to written permission from the original data providers via the official email: ephi.gov.et.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and received ethical approval from the Jimma University Institutional Review Board (Ref: IHRPGD/1204/22; Date: 20/06/2022). It also complied with the national ethical standards of the Ethiopian Science and Technology Commission (ESTC). Written informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.