A penal data analysis on the effects of climate variations on malaria incidences among children 0-5 years in Uganda

 Corresponding author: George Robert Okello Copyright © 2025 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0. A penal data analysis on the effects of climate variations on malaria incidences among children 0-5 years in Uganda George Robert Okello 1, *, Robert Wamala 2, Hellen Namawejje 1, Martin Mbonye Kayitale 1 and Herbert Susan Sendege 3 1 Department of Population Studies, College of Business and Management Sciences, Makerere University, P. O. Box 7062, Kampala, Uganda. 2 Directorate of Research, Innovations and Partnership, Makerere University, P. O. Box 7062, Kampala, Uganda. 3 Department of Statistical Methods, College of Business and Management Sciences, Makerere University, P. O. Box 7062, Kampala, Uganda. World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 Publication history: Received on 20 July 2025; revised on 28 August 2025; accepted on 03 September 2025 Article DOI: https://doi.org/10.30574/wjarr.2025.27.3.3122 Abstract Despite global reductions in malaria due to intensified interventions, it remains a major cause of hospitalizations and deaths among Ugandan children under five. Uganda’s changing climate may be influencing malaria incidence in this vulnerable group. This study examined the relationship between climate variations and malaria prevalence among children under five across 13 Ugandan districts, selected for their high disease burden and susceptibility to climate change. Using a retrospective research design and penalized data analysis, district-level random effects were incorporated into a negative binomial regression model. Results showed a positive association between malaria cases and minimum temperature (coefficient = 0.0333, p < 0.0000) as well as vegetation cover (NDVI, coefficient = 0.3240, p < 0.0000). Conversely, maximum temperature was negatively associated (coefficient = -0.0099, p < 0.0000). Rainfall was insignificant in the combined model but significant in separate analyses. Regional variations emerged, with particularly high incidence in West Nile districts Yumbe (214), Adjumani (168), Koboko (150), and Karamoja’s Kotido (141), Nabilatuk (196), and Napak (130). Findings highlight the influence of temperature and vegetation on malaria prevalence, underscoring the need for region-specific malaria control strategies that integrate climatic and environmental factors. Keywords: Climate Variability; NDVI; Malaria; Rainfall; Temperature 1. Introduction Almost half of the global population is vulnerable to malaria. In 2022, around 249 million individuals contracted malaria across 85 nations(1). That year, the disease resulted in roughly 608,000 fatalities(2),(3). Sub-Saharan Africa is responsible for approximately 93% of all malaria-related deaths worldwide(4). In 2020, this region accounted for 95% of malaria cases and 96% of deaths. Children under five represented about 80% of malaria fatalities in this region(5). There are geographical variations in malaria incidence, notably among sub-Saharan African countries, with Nigeria and the Democratic Republic of Congo contributing over 10% of the global malaria cases: Nigeria (31.9%), Democratic Republic of the Congo (13.2%), United Republic of Tanzania (4.1%), and Mozambique (3.8%) (Saba, 2022). Conversely, nations such as Burundi and Swaziland recorded less than 1% of malaria cases(6). Additionally, it is estimated that 96% World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 214 of the malaria deaths in Sub-Saharan Africa are among children under five years(7). Uganda ranks sixth worldwide and third in Africa in terms of malaria mortality, with 16 million cases and over 10,000 deaths reported each year(8). The Malaria Indicator Survey 2018-2019 revealed that 9% of children aged 0-59 months tested positive for malaria parasites (9). The same survey indicated an increase in malaria incidence among this age group from 3% to 12%. Furthermore, malaria incidence is nearly four times greater in rural areas (11%) compared to urban settings (3%). Regionally, Karamoja exhibited the highest malaria incidence at 34%, followed by West Nile at 22% and Busoga at 21%, while Kampala and Kigezi had the lowest incidence, each below 1%(10). Malaria places a substantial socio-economic burden on households in Uganda and on the national economy. Direct expenses, such as consultation fees, medication, transport, and care, are compounded by indirect costs like lost workdays, decreased productivity, and negative educational impacts. An average malaria episode costs households about $26, totaling $78 annually for families encountering three episodes, which is 3% of their income. Poor households in malaria-endemic areas may allocate up to 25% of their income on prevention and treatment, which exacerbates poverty levels. For children under five, the estimated annual economic impact is around $614 million, of which $57.7 million is attributed to direct medical costs(11) . In addition to affecting individuals and households, malaria hinders national development by lowering agricultural and industrial output and deterring foreign investment(12). Severe cases of malaria can impair cognitive development in children by as much as 60%, thereby weakening human capital and adversely affecting Uganda's universal education initiatives(13). Campaigns to fight malaria through interventions such as insecticide-treated nets (ITNs), indoor residual spraying (IRS), and intermittent preventive treatments have made some headway. For example, the national malaria parasite incidence fell from 42% in 2009 to 9.1% in 2018, and the 2024 campaign distributed over 30 million ITNs nationwide(14). Nonetheless, malaria cases have been on the rise in more than 70 districts, indicating significant gaps in control strategies. Environmental factors, including increased temperatures, humidity, and modified rainfall patterns, further exacerbate malaria transmission dynamics, undermining the potency of existing interventions(15). Climate change is increasingly acknowledged as a major factor in the revival and spread of malaria in specific areas. Changes in environmental aspects such as temperature, rainfall, humidity, and the frequency of severe weather events create situations that can either promote or hinder malaria transmission(16). These alterations in environmental conditions can significantly affect the life cycle of Anopheles mosquitoes and the development of the Plasmodium parasite within them, ultimately impacting malaria rates and the geographical distribution of the disease(17). This study intends to explore the connection between climate variations and malaria occurrences in children aged 0-5 years in Uganda over an eight-year period from 2015 to 2022. 2. Materials and Methods 2.1. Study Design The research employed a retrospective longitudinal design. This was centered on verified malaria cases in children under five years old, along with climate factors such as maximum temperature, minimum temperature, and vegetation cover. 2.2. Data Sources The study used secondary data from January 2015 to December 2022 as follows. • Uganda National Meteorological Authority (UNMA), where we got data on climate variables such as rainfall and temperature • Uganda Ministry of Health (MoH), where we got data on confirmed malaria cases among children under five years. • Satellite Data (Vegetation Indices: NDVI), where we got data on vegetation cover The above date can be obtained from the following links. World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 215 Table 1 Data Sources Variable Access Link Vegetation (NDVI) NASA Earth Data, Copernicus Climate (Rainfall, Temp) CHIRPS, WorldClim, ESGF Malaria Incidence WHO Malaria Data, DHIS2 Land Use/Cover NASA EOSDIS 2.3. Study area The research focused on four areas: West Nile, Karamoja, Ankole, and South-central. The choice of these regions was informed by the findings from the Uganda malaria indicator survey conducted in 2019, which indicated that West Nile and Karamoja experienced the highest rates of malaria. Additionally, the survey revealed that South Buganda and Ankole had the lowest rates of malaria among children under five years. The study considered thirteen districts in all four regions. These were selected based on the load of the malaria cases as reported by the Ministry of Health. Three districts were selected from each region: two with high loads and one with a lower load. As illustrated in Figure 1. Figure 1 Area of study World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 216 Figure 2 Distribution of Health facilities under study Figure 2 illustrates the distribution of health facilities by their respective levels. The facilities are categorized according to the levels defined by the Ministry of Health. The analysis focused on four categories of health facilities: Hospitals, HCII (Health Center II), HCIII (Health Center III), and HCIV (Health Center IV). These facilities are responsible for reporting monthly health indicators to the Ministry of Health via DHIS2. Each level signifies a different tier within the healthcare system, offering distinct capabilities and services. As indicated in Table 2.1, the distribution includes 34 (2.9%) hospitals, 721 (51.7%) HCII facilities, 336 (39.5%) HCIII facilities, and 39 (5.8%) HCIV facilities. 2.4. Variables and their measurement The research focused on confirmed malaria cases in children under five years as the dependent variable, while Vegetation Cover, Rainfall, and Temperature were treated as the independent variables. Table 2 Summary of Variables and their Measurements Variable Measurement Method Data Source Units Spatial Resolution Dependent variable Malaria Incidence Confirmed malaria cases among children under five years per health facility Local Health Centers Confirmed cases Independent variables Vegetation Cover Derived from satellite reflectance (NIR and RED bands) MODIS NDVI (-1 to +1) 250m Rainfall Measured via satellite-based precipitation estimates and ground stations CHIRPS, Weather Stations mm (month) 0.05° Temperature Measured via satellite or groundbased meteorological stations CHIRPS and Weather Stations °C, minimum and maximum 0.1/9km NIRNear-Infrared, REDRed region of the electromagnetic spectrum, MODISModerate Resolution Imaging Spectroradiometer, NDVINormalized Difference Vegetation Index; CHIRPS-Climate Hazards Group InfraRed Precipitation with Station data 2.5. Data Preparation and Standardization Before merging the datasets, both the climate and malaria datasets were formatted to ensure compatibility regarding the same time frames and geographical areas. The climate and malaria information were synchronized to include the same duration, with both datasets compiled into monthly intervals across the study period from 2015 to 2022. World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 217 Additionally, both datasets were aligned to a unified coordinate system (latitude/longitude) to ensure precise spatial integration. 2.6. Data Analysis This research employed panel data analysis. Panel data is characterized by information gathered from multiple subjects (districts) observed over a period(18). This methodology facilitates the examination of both time-related changes and variations across different sections. The primary outcome variable for this study was the aggregate number of confirmed malaria cases in children under the age of five. The panel data analysis framework accommodates variations within districts across time and differing characteristics between districts, providing a comprehensive perspective on malaria dynamics throughout the districts. 2.7. Descriptive Statistics Descriptive statistics offered a deeper understanding of the data characteristics. Univariate analysis focused on examining the distribution, central tendency, and variability of individual variables. For climate-related variables such as temperature and rainfall, the mean, standard deviation, minimum, and maximum values were calculated. Trend analysis revealed seasonal patterns or temporal changes. With regards to malaria incidence, descriptive statistics and line graphs illustrated trends and distributions. Pearson correlation analysis evaluated the relationships among variables, yielding insights into their interdependencies and interactions. 2.8. Exploratory Analysis A vital element of the negative binomial model is the dispersion parameter, which assesses the extent of variability or dispersion in malaria cases. A significantly greater than zero dispersion parameter signals overdispersion, supporting the choice of the negative binomial model instead of the Poisson model. By analyzing the dispersion parameter, researchers gain a clearer perspective on the inherent variability in the data, essential for precise modeling and prediction. The findings from this exploratory analysis highlighted which variables played the most critical roles in forecasting malaria cases, guiding the selection of variables for the final modeling phase. The final model is expected to include these significant variables, facilitating more accurate predictions or inferences concerning malaria incidence. The fitting of a basic negative binomial model during the exploratory phase was an essential step. It not only enhanced the understanding of the distribution and variability in malaria cases but also informed the choice of key variables for the final model. This comprehensive methodology guarantees that the concluding model effectively reflects the data and is capable of yielding meaningful insights for public health strategies and policy-making. Inferential Analysis At the multivariate level, both Poisson and negative binomial models were constructed using confirmed malaria cases in children under five years old. This was carried out through random effects estimation. Initial tests, such as overdispersion using an auxiliary model and Hausman tests, guided the selection of the model. A divergence test was conducted to compare Poisson and negative binomial models to identify the best fit. Visual comparisons of observed data versus modeled distributions validated the chosen model, ensuring the reliability and validity of malaria dynamics analysis. 2.9. Negative Binomial Model The analysis regarded the total count of confirmed malaria cases among children under five years old as count data, typically modeled via the Poisson distribution. However, in situations of overdispersion (where variance is greater than the mean), more sophisticated models are necessary. The Negative Binomial (NB) model, which introduces a parameter for additional variability, is often applied to over-dispersed count data. Alternative models include Quasi-Poisson, which modifies standard errors for over-dispersion, and Zero-Inflated models, which accommodate a surplus of zeros. Since there were no instances of zero malaria cases in any district throughout the months, zero-inflated models were disregarded. Considering the diverse climate conditions across districts, the Negative Binomial model was chosen for its capacity to manage over-dispersion. This study adhered to the model specifications proposed by noted authors in the field. From the general representation of a classic linear model 𝑌 = 𝑋Β +ϵ......................................2.10 Where 𝑌 = (𝑦1,𝑦2,...,𝑦𝑛)𝑇 is the n dimensional vector if the observed outcome variable; 𝑋, is the 𝑛×𝑝 design matrix where p is the number of parameters in the model; Β = (𝛽1,𝛽2,...,𝛽𝑝)𝑇 is a 𝑝 dimensional vector of parameters including the constant term; and ϵ = (𝜖1,𝜖2,...,𝜖𝑛)𝑇 is an n dimensional vector of the error term which represents the World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 218 random variation in the model. The model specification described above is based on several assumptions about the error term and the dependent variable, such as normality and constant variance. Violations of these assumptions necessitate modifications to the model, particularly in cases where the outcome variable is a count, as in the current study. In such situations, the concept of generalized linear modeling offers a valuable solution for the modeling process. The generalized linear model concept is hinged on three key components, that is The outcome variable comes from the distribution of the exponential family with 𝐸(𝑌) = 𝜇, and 𝑉𝑎𝑟(𝑦) = 𝛼(∅).𝑣𝑎𝑟(𝜇), where ∅ 𝜀 is a scale parameter and 𝛼() is a function of ∅ that explains the random component. 𝜂 = 𝑋Β, which is the systematic component examining the effect of all the covariate information on the response variable 𝜂 = 𝑔(𝜇) Where 𝑔(.) is known as the link function The algebraic representation of the generalized linear model then becomes 𝑔(𝑌) = 𝜂 = 𝑋Β ......................................2.11 The generalized model framework assumes that the outcome variable Y follows a distribution that belongs to the exponential family, which can be represented as below 𝑓𝑌(𝑦,𝜃,∅)= 𝑒𝑥𝑝{[𝑦𝜃−𝑏𝜃] 𝛼(∅)+ 𝑐(𝑦;∅)}……….2.12 In this case, ∅ it is referred to as the dispersion and 𝜃 as the canonical parameter. This analysis in this study assumed a modeling procedure of this nature, assuming a negative binomial link. There are several ways that can be embedded in the above modeling procedure to cater for the variation of malaria cases across districts, these include accounting for either fixed or random effects during the analysis. Although both terms are crucial for handling the inherent heterogeneity in the data, fixed effects account for variations at the parameter level, estimating a specific parameter for every individual observation in the data. This, however, restrains the possibility of examining variations across the different groupings in the data. Accounting for random effects in the model presents a solution to this, as it enables estimation of variation across groups of interest (districts in this study). Including this in the estimation procedure in equation 2 generates equation 3 below, which was adopted at the modeling stage in this study 𝑔(𝑌) = 𝜂 = 𝑋Β + 𝑍𝑈......................................2.13 From Equation 3 above, 𝑍𝑈 defines the variation across districts and is referred to as the random effects term in the analysis. 3. Results Table 3 Distribution of health facilities by region Regions Health facilities Ankole Karamoja South central West Nile Total Hospitals 1 2 28 3 34 Health Centre IV 7 2 27 3 39 Health Centre III 43 17 220 56 336 Health Centre II 83 24 548 66 721 Total 134 45 823 128 1130 World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 219 Table 3 presents the allocation of health facilities among the four regions. The South-Central Region leads with the highest number of facilities, totaling 823, which includes 28 hospitals, 27 Health Centre IVs, 220 Health Centre IIIs, and 548 Health Centre IIs. Conversely, the Karamoja Region contains the smallest number of facilities, with a total of 45, consisting of 2 hospitals, 2 Health Centre IVs, 17 Health Centre IIIs, and 24 Health Centre IIs. The Ankole Region has 134 facilities, which comprise 1 hospital, 7 Health Centre IVs, 43 Health Centre IIIs, and 83 Health Centre IIs, while the West Nile Region has a total of 128 facilities, including 3 hospitals, 3 Health Centre IVs, 56 Health Centre IIIs, and 66 Health Centre IIs. Altogether, the total number of health facilities across all regions is 1,130, highlighting a substantial concentration of resources in South Central and a more restricted availability in Karamoja. Table 4 Distribution of study variables Variable No. Mean Std. Dev Min Max Skewness Confirmed Malaria 4,024,028 94.25 81.27 4 350 0.77 Rainfall 4,580,845.31 91.47 20.21 51.75 136.83 -0.01 Temperature Min 796,028.42 16.24 2.26 9.66 20.34 -1.21 Temperature Max 1,370,410.84 29.02 3.74 17.87 34.68 -0.92 NDVI 2,3031.8 0.50 0.07 0.33 0.65 -0.42 Table 4 presents summary statistics for various variables associated with malaria and environmental factors. It displays the count of observations, mean, standard deviation, minimum, maximum, and skewness for each variable. Confirmed Malaria shows an average of 94.25 cases, with a standard deviation of 81.27, indicating significant variability in the number of malaria cases, and a positive skewness of 0.77, which points to a long right tail. Rainfall averages at 91.47 mm and has a standard deviation of 20.21 mm, with a skewness of -0.01, indicating that the data is nearly symmetrical. The minimum and maximum temperatures have averages of 16.24°C and 29.02°C, respectively, with negative skewness values of -1.21 and -0.92, suggesting that their distributions are more concentrated at the higher end. Lastly, NDVI records a mean of 0.50 with a standard deviation of 0.07 and a skewness of -0.42, indicating a mild negative skew, with values mainly clustered towards the upper range of the vegetation index. 3.1. Descriptive statistics Table 5 Summary Statistics for Analysis Variables by District Average Malaria cases Average Total rainfall amounts Average Minimum Temperature Average Maximum Temperature Average NDVI Adjumani 168 99.313 18.702 32.95 0.514 Bukomansimbi 28 98.255 16.710 26.98 0.556 Gomba 27 103.371 16.485 28.058 0.580 Isingiro 34 78.456 14.533 27.61 0.504 Kiruhura 16 78.702 14.738 27.239 0.515 Koboko 150 112.091 17.738 30.639 0.480 Kotido 141 64.08 16.367 30.716 0.351 Lwengo 36 97.415 16.491 27.539 0.487 Nabilatuk 196 82.832 17.131 32.34 0.435 Napak 130 81.135 16.675 31.329 0.407 Rwampara 6 61.739 10.294 19.749 0.411 Wakiso 60 118.267 18.106 27.896 0.464 Yumbe 214 102.461 18.576 32.199 0.499 World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 220 Error! Reference source not found.5 shows considerable variation by districts: Yumbe and Nabilatuk have high malaria cases (214 and 196, respectively, while Rwampara has the lowest (6). Rainfall also varies, with Wakiso receiving the most, while Kotido has the least. Temperature ranges are broad, with higher temperatures generally observed in areas with more malaria cases, possibly indicating a relationship between these factors. Table 6 Summary Statistics for Analysis Variables by year Average Malaria cases Average Total rainfall amounts Average Minimum Temperature Average Maximum Temperature Average NDVI 2015 77 91.879 16.105 30.019 0.523 2016 85 83.76 16.207 30.244 0.509 2017 84 91.899 16.213 30.284 0.512 2018 75 104.899 15.941 29.58 0.519 2019 88 108.216 16.504 29.678 0.528 2020 94 111.034 15.876 27.989 0.534 2021 87 87.04 17.973 26.876 0.51 2022 95 95.201 17.998 26.931 .4094 Table 6 presents summary statistics by year. From the table, the average number of malaria cases fluctuates, with a general increase observed from 2018 to 2022. Rainfall amounts also show variability, peaking in 2019 and 2020. The minimum temperature remains relatively stable, while the maximum temperature shows a gradual decline from 2016 onward. The NDVI values suggest slight changes in vegetation health, with a peak in 2020 followed by a decrease in subsequent years, potentially indicating variations in environmental conditions over time. 3.2. The relationship between rainfall, minimum temperature, maximum temperature, vegetation cover, and confirmed malaria cases among children under five years. Here, we conducted a correlation analysis to assess the relationship between confirmed malaria cases among children under five years and rainfall, minimum and maximum temperature, and Vegetation cover, as shown in Table 7 Correlation Results7. Table 7 Correlation Results Malaria cases rainfall Minimum Temperature Maximum Temperature Veg cover Total Malaria cases 1.000 Total rainfall amounts 0.080*** 1.000 Minimum Temperature 0.194*** 0.328*** 1.000 Maximum Temperature 0.175*** 0.019*** 0.702*** 1.000 Vegetation cover 0.048*** 0.355*** 0.217*** 0.130*** 1.000 The data presented in Table 7 reveal noteworthy yet generally low correlations among the variables, emphasizing positive connections between climatic and environmental factors and instances of malaria. Malaria Cases exhibit weak positive correlations across all variables, with the strongest association being with Minimum Temperature (0.194), indicating a slight rise in malaria cases as minimum temperatures increase. The correlations with Maximum Temperature (0.175) and Rainfall (0.080) are less pronounced, while the link with Vegetation cover is the weakest (0.048), suggesting a minimal direct effect of vegetation health on malaria instances. Rainfall shows a moderate correlation with Vegetation cover (0.355) and Minimum Temperature (0.328), underscoring the impact of rainfall on World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 221 supporting vegetation health and affecting minimum temperatures. Minimum Temperature has the most robust correlation with Maximum Temperature (0.702), demonstrating their natural interdependence, along with a moderate correlation with Vegetation cover (0.217), highlighting its significance for vegetation health. In summary, while environmental factors influence malaria dynamics, the weak correlations indicate that other unmeasured factors could play a more substantial role. Overall, the correlations among the variables range from low to moderate, with the highest correlation found between Minimum and Maximum Temperatures (0.702). The relatively weak relationships of Malaria Cases with environmental factors like rainfall, temperatures, and Vegetation cover imply that these variables each contribute to explaining malaria dynamics independently, without redundant overlaps. 3.3. Climate variation and malaria using panel analysis Table 8 Relationship between rainfall amounts and malaria cases Malaria cases Coef. St. Err. t-value p-value [95% Conf Interval] Sig Total rainfall amounts 0.0007 0.0001 11.05 0.0000 0.0005 0.0007 *** Constant 0.0625 0.0085 7.35 0.0000 0.0458 0.0791 *** ln(alpha) -0.0192 .345 -0.6953 0. 6571 ln(s) 3.3672 .44 2.5046 4. 2298 *** Mean dependent var 85.850 SD dependent var 135.895 Number of obs 44624 Chi-square 122.101 Prob > chi2 0.000 Akaike crit. (AIC) 454433.758 *** p<.01, ** p<.05, * p<.1 The positive coefficient (0.0007) suggests that as the total rainfall amounts increase, the number of malaria cases also increases. The relationship is statistically significant (p-value = 0.0000) at the 5% level of significance. The dispersion parameter (ln(alpha) = -0.0192) suggests that 2% of the variations in total malaria cases, but this variation was not significant. Overall, the regression analysis indicates a significant positive relationship between rainfall amounts and malaria cases, with a statistically significant model fit. Table 9 Relationship between minimum temperature and malaria cases Malaria Cases Coef. St. Err. t-value p-value [95% Conf Interval] Sig Minimum Temperature 0.025 0.001 17.09 0.0000 0.022 0.027 *** Constant -0.281 0.025 -11.33 0.0000 -0.329 -0.232 *** Constant 0.052 0.347 -0.629 0.732 Constant 3.483 0.436 2.628 4.339 Mean dependent var 85.850 SD dependent var 135.895 Number of obs 44624 Chi-square 292.199 Prob > chi2 0.000 Akaike crit. (AIC) 454232.716 *** p<.01, ** p<.05, * p<.1 Table 9 presents a regression analysis concerning malaria cases, indicating that the minimum temperature has a significant impact on malaria incidence, shown by a positive coefficient of 0.025 (p < 0.01), signifying that higher minimum temperatures correlate with an increase in malaria cases. The overall significance of the model is validated by a chi-square value of 292.199 (p < 0.001). The average number of malaria cases reported is 85.85, and the standard deviation is 135.895, demonstrating considerable variability in the dataset. The Akaike Information Criterion (AIC) World Journal of Advanced Research and Reviews, 2025, 27(03), 213-228 228 [22] Chaves LSM, de Sousa TM, Penha LCF, Hacon SS. Malaria in the Amazon Basin: how climate change and natural disasters create new challenges for an old disease. In: Planetary health approaches to understand and control vector-borne diseases. Wageningen Academic; 2023. p. 54–91. [23] Orondo PW, Wang X, Lee MC, Nyanjom SG, Atieli H, Ondeto BM, et al. Habitat diversity, stability, and productivity of malaria vectors in irrigated and nonirrigated ecosystems in western Kenya. J Med Entomol. 2023;60(1):202– 12. [24] Tsegaye A, Demissew A, Hawaria D, Abossie A, Getachew H, Habtamu K, et al. Anopheles larval habitats seasonality and environmental factors affecting larval abundance and distribution in Arjo-Didessa sugar cane plantation, Ethiopia. Malar J [Internet]. 2023;22(1):1–12. Available from: https://doi.org/10.1186/s12936-02304782-1 [25] Abigail Nana Afia Yeboah. Investigating the impact of changing environments on mosquito-borne diseases, through the lens of alterations in vegetation and human-driven landscape modifications. World J Adv Res Rev. 2023;18(2):1440–54. [26] Sertić Perić M, Dražina T, Žutinić P, Rubinić J, Kuczyńska-Kippen N, Zhang C, et al. Ecological Dynamics and Conservation Strategies for Mediterranean Salt Marshes: Insights from a Pilot Study of Biodiversity and Environmental Drivers in the Palud Marsh, Croatia. Sustain. 2024;16(23).