Climate change impact on disease emergence and antimicrobial resistance: A perspective on transmission and detection

Summary

Climate change is increasingly disrupting environmental and ecological systems, creating conditions that facilitate the emergence, transmission, and persistence of infectious diseases and the spread of antimicrobial resistance (AMR). This scoping review analyzed 427 articles published over the past five years, retrieved from major scientific databases. Findings show that climate variables significantly influence pathogen growth, reproduction, and transmission through vectors. Strong associations were identified between climatic changes and the spread of bacterial, viral, fungal, and protozoal diseases. Temperature fluctuations and anoxic conditions can promote AMR development. Climate-driven migration of birds and bats further contributes to cross-border pathogen transmission. The review also highlights emerging AI tools for rapid pathogen detection, diagnosis, and surveillance. Overall, climate change poses significant public health risks through increased infectious and non-infectious disease burden, zoonotic spillover, and AMR spread, emphasizing the need for integrated climate-health policies, AMR stewardship, AI-enabled surveillance, and interdisciplinary One Health research.

Graphical abstract

Environmental science; Ecology

Introduction

Climate change refers to long-term alterations in climatic variables resulting from both natural and anthropogenic activities. The average global temperature is rising at approximately 0.2 °C per decade,^1,2,3 and continued warming is projected to reach 1.5 °C between 2030 and 2052.⁴ Moreover, by 2100, climate projections estimate that global mean temperatures could increase by 1.1°C–6.4°C.⁵ Due to the gradual increase in average global temperature, changes in ocean current speeds (an increase of about 15% was observed between 1990 and 2013) contributed to altered air circulation, increased humidity, and changes in precipitation patterns.⁶ According to the IPCC, sea levels rose significantly, at rates of 1.3 mm/year from 1971 to 1990, 1.9 mm/year from 1990 to 2006, and 3.7 mm/year from 2006 to 2018.⁷ As a result, the drastic effects can cause flooding and the flow of saltwater into surface and groundwater, leading to coastal erosion.⁸ Rising ocean temperatures reduce CO₂ absorption, potentially leading to increased atmospheric CO₂ levels.

Global warming may influence pathogen breeding and can be linked to disease progression or suppression. Factors influencing infectious disease can be ecological, socioeconomic, and anthropogenic, including overpopulation, travel, intensive animal husbandry, reduced immunity, and changes in food production. Extreme weather events caused by climate change, such as melting polar ice caps, sea level rise, landscape changes, and wildfires, contribute to the spread of infectious diseases, which are responsible for millions of deaths annually. Moreover, excessive rainfall, floods, changes in wind patterns and velocity, droughts, heatwaves, and humidity alter the survival, pathogenicity, and dissemination of communicable or infectious organisms.^9,10 Excess rainfall and floods may also increase organic matter availability and create optimal conditions for growth and breeding sites for disease vectors. In contrast, decreased precipitation causes drought, increasing suspended particulate matter (PM) in the air and potentially increasing airborne infections in communities.¹¹ Global warming may also alter bird migration patterns, which contribute to the spread of avian-associated pathogens. Figure 1 depicts the impacts of various climatic parameters on disease transmission dynamics.

Figure 1.

Impact of climate change parameters on the emergence and spread of diseases

Vector-borne diseases, such as dengue, malaria, and West Nile virus (WNV) fever, are exacerbated by high temperature, relative humidity, and changes in rainfall patterns. These diseases are further exacerbated by drought and flooding, which favor mosquito breeding. A study suggests that, due to climate change, the distribution of vector-borne disease-endemic areas in China has shifted northward. Additionally, the older population faces challenges posed by rising temperatures and infectious diseases. Globally, in 2021, dengue fever was responsible for 29,200 deaths and 2.08 million disability-adjusted life years (DALYs). Similarly, in the same year, Malaria caused 7,49,000 deaths; among them, 57% were children under the age of 5 years¹²

Infectious or communicable diseases are caused by pathogenic organisms such as bacteria, viruses, fungi, or parasites and are transferred from person to person through infected animals, contaminated surfaces, bodily fluids, blood products, or direct contact. These diseases are the leading causes of global mortality.¹³ Infectious diseases are further classified as emerging and re-emerging. Re-emerging infectious diseases are those that were once major public threats, had reduced substantially, but are now recurring and posing a serious public health crisis. Emerging infectious diseases are those that have recently surfaced and had not previously occurred, for instance, severe acute respiratory syndrome (SARS), COVID-19, Lyme disease, and pathogenic Escherichia coli (E. coli) infections. 76.9% of these diseases originate from wildlife, 60.3% are zoonotic, and 54.3% are induced by bacteria.¹⁴

Non-communicable diseases (NCDs) are chronic and non-contagious medical conditions, also referred to as an “invisible epidemic”. These diseases are generally caused by environmental factors, such as exposure to toxins, ultraviolet (UV) radiation, extreme weather events, and air pollution. Several conditions, including changes in the growth and multiplication rates of existing pathogens, the emergence of novel pathogenic organisms, over- or under-breeding of disease vectors, and the emergence of new or intensification of previously existing NCDs, may be directly or indirectly linked to climate change.¹⁵

This review explores the complex relationships between climate change variables and their influence on the prevalence of infectious diseases and NCDs. In addition, the study highlights how changes in bird migration patterns due to climate change may facilitate the spread of pathogens across biogeographical barriers. The study also attempted to explore how various climatic variables are linked to the spread of AMR in the community. Furthermore, the review outlines the future prospects for the application of artificial intelligence and machine learning (AI/ML) tools in the early detection, diagnosis, and prediction of diseases and pathogens. These emerging technologies and algorithms can be used for early detection by analyzing large datasets of clinical records, genomic information, and environmental indicators. This can improve diagnostic accuracy and enable rapid pathogen identification. In addition, AI/ML tools can monitor climate-related factors, such as changes in temperature, precipitation patterns, and humidity, which increase disease emergence and transmission. Climate change, often called the climate crisis, is an escalating global problem. Its impact on human health remains underexplored and needs deeper investigation.

Methodology

The current study was conducted as a scoping review to synthesize existing literature on the effects of climatic variables on infectious diseases, NCDs, vector-borne diseases, AMR, and the application of AI/ML in disease detection and diagnosis. The review is reported in accordance with the PRISMA workflow.

A literature search was conducted across four scientific databases: Google Scholar, PubMed, Web of Science, and Scopus. The search focused on peer-reviewed articles published within the last five years. Keywords such as “climate change variables,” “vector ecology,” “pathogen transmission,” “artificial intelligence,” “machine learning,” “antimicrobial resistance,” and “infectious and non-communicable diseases” were used in various combinations.

The inclusion criteria were articles relevant to climate change, disease dynamics, vector ecology, pathogen transmission, AMR, and AI/ML tools for disease diagnosis and surveillance. Articles that were non-English or irrelevant to the article’s theme were excluded.

A total of 427 articles were included based on titles and abstracts. After irrelevant studies were removed, articles were retained for further consideration. Of these, 138 articles were included in the final synthesis for this review. The remaining 289 articles were used specifically for country-wise analysis of AI/ML applications in pathogen detection and diagnosis to generate a graph. The study selection process is illustrated in Figure 2.

Figure 2.

Prisma flowchart of the selected papers for this scoping review

Relevant data from the included studies were extracted and organized into thematic categories, including climate variables, disease transmission patterns, AMR trends, and AI/ML applications. A descriptive and thematic synthesis approach was used to summarize the findings and identify key research trends and gaps across the literature.

Impact of climate variables on infectious diseases

Climate change parameters, such as temperature, precipitation, and relative humidity, influence the dynamics of pathogens, including replication rate and disease transmission. Extreme weather events, such as heavy rainfall and floods, can contaminate drinking water sources. Ocean warming may contribute to sewage overflows from drainage systems, which can clog sources and contaminate drinking water. These environmental changes thereby increase the seasonal prevalence of waterborne and foodborne diseases, especially in warmer climates. In addition, ocean warming ultimately contributes to the clogging of sewage drainage systems and their overflow, contaminating drinking water. Air pollution further enhances climatic effects by increasing susceptibility to infectious diseases, impairing the immune system and compromising respiratory barriers.¹²

Bacterial pathogens

Several bacterial pathogens may be associated with climatic factors, such as Vibrio cholerae, which is linked to a mean summer temperature of 30.3 °C and precipitation of 550 mm during the monsoon.¹⁶ Non-cholera Vibrios (NCVs) are pathogenic bacteria and have been found in water as cold as 12 °C, but can grow above 20 °C. Estuarine beaches with reduced salinity are prone to developing high NCV concentrations. Infection occurs when open wounds come into contact with contaminated water.¹⁷ Furthermore, Legionnaires’ disease is a form of pneumonia transmitted through aerosols, typically from evaporative cooling systems or hot tubs, and domestic water systems. Temperatures between 20 °C and 45°C provide favorable conditions for their growth.¹⁸ Campylobacter is generally transmitted through the consumption of undercooked meat, poultry, and contaminated milk or water. Quail meat (68.4%) has the highest contamination rate, followed by chicken (56.1%), turkey (27.4%), and ostrich meat (11.7%).¹⁹ It has been found that increased temperatures and heavy rainfall can increase the rate of Campylobacter infection. During summer, a 1 mm increase in precipitation causes a 38% increase in the incidence. Additionally, a 1 °C rise in temperature is linked to a 4.1% increase in salmonellosis cases, and a 1 unit increase in extreme precipitation leads to a 5.6% increase.²⁰

Viral pathogens

Climatic conditions and air quality significantly influence viral disease dynamics. For instance, COVID-19 infections have been found to correlate significantly with ambient temperature, humidity, and air pollution. It is reported that a 1% rise in relative humidity reduces daily COVID-19 deaths by 0.5%, while a 1 °C increase in temperature decreases the number of cases by 3%. Conversely, a 10 μg/m³ increase in PM_2.5 is associated with a 66% increase in infections.²¹ Climatic conditions and air quality are closely related. It is estimated that about 90% of the world’s population is exposed to health risks due to air pollution. Pollutants, such as nitrogen dioxide, PM, ozone, and diesel exhaust particles, increase human susceptibility to infections by disrupting immune responses, altering cell-surface expression, and compromising airway epithelial barriers. Furthermore, chickenpox incidence increases by 1.99% for every 1 °C drop in temperature, and a 1 mm increase in precipitation (above 4.9 mm) causes a 1.36% increase in cases, while a 1 °C increase in relative humidity (above 57.18%) correlates with a 2.05% increase in cases.²² Seasonal influenza A and B viruses can grow at 33 °C; similarly, respiratory viruses, such as rhinoviruses, also efficiently replicate at 32 °C–33°C.²³ Dust storms contribute to the spread of diseases such as the common cold and flu.²⁴ Likewise, Gastroenteritis, a digestive infection caused by norovirus, is more severe at low temperatures between −6.6 °C and 20°C and at relative humidity between 10% and 66%.²⁵ Rotavirus incidence is reduced by 10% per 1 °C rise in temperature, 1% reduction per 1 cm increase in rainfall, and 3% reduction per 1% increase in relative humidity.²⁶ Intense weather events, such as hurricanes and heavy rainfall, facilitate the spread of infections and drain the pathogenic viruses into water bodies.

Fungal pathogens

Traditionally, fungi are not able to grow at a human’s normal body temperature, due to “Thermal Restriction”. However, as the temperature rises, so do the elevated atmospheric CO₂ levels, acidity, salinity, and nutrient availability, which affect the microbiota environment and allow fungal pathogens to evolve and adapt quickly. Fungi are saprotrophic and produce infectious spores in enormous quantities, which they spread via aerosols, so infrequent fungal diseases become prevalent. Landfill areas with airborne fungal concentrations above 10⁴ CFU/m³ are highly contaminated. The highest concentration of airborne fungi is found at temperatures between 25 °C and 30°C, relative humidity of 60%–70%, and wind speed below 1 m/s²⁷ Physiological data reveal that most Aspergillus fumigatus strains tolerate temperatures above 50 °C, whereas Cryptococcus can survive and replicate at 37 °C, allowing it to adapt to the human body and act as a pathogen in an immunocompromised host. During infection, Cryptococcus neoformans can adapt to varying pH levels to survive; for instance, the Rim101 protein, a pH-reactive transcription factor, responds to neutral and alkaline pH. Cryptococcus cells shield themselves from UV radiation through melanization, as melanin insulates the cells from oxidative damage and phagocytosis.²⁸ Blastomyces can grow in the environment as filamentous mold and hyphae at 22°C–25°C and as yeast at 37 °C in human tissues.²⁹

Protozoal and parasitic pathogens

Protozoal infections, including giardiasis, cryptosporidiosis, amebiasis, and toxoplasmosis, may be associated with environmental contamination of water sources. Physical footprints (resource use) and carbon footprints (greenhouse gas, or GHG, emissions) both contribute to the spread of diseases, thereby increasing the endemic range of parasitic diseases. Parasite spread patterns are influenced by climatic factors, including host density, ambient temperature, and moisture levels. Fasciola hepatica (liver fluke), a causative agent of fasciolosis, has an optimal temperature for egg hatching between 20 °C and 30°C, and its cercarial shedding is enhanced at 27 °C by snail hosts, and intermediate snail hosts show their growth highest at 25 °C.³⁰

Overall, the above data strongly support the idea that different climatic conditions and variables, including temperature, relative humidity, and precipitation, affect a wide range of pathogens (bacterial, viral, fungal, and protozoal) and diseases in distinct ways, reinforcing the need for an integrated One Health perspective.

Impacts of climate variability on infectious disease outcomes in India

Studies support that mosquito life cycle stages, such as biting and blood-feeding rates, are significantly influenced by temperature and rainfall. In India, particularly during the summer and monsoon season (June to September), rainfall lowers temperatures, creating water habitats for mosquito breeding.³¹

India’s Epidemic-Climate Dataset (EpiClim) connects climate and public health data, enabling climate forecasts to inform epidemic modeling. The dataset provides the first district-level, weekly record of major epidemics in India since 2009, sourced from the Integrated Disease Surveillance Program (IDSP). The data indicate that acute diarrheal disease is most widespread, followed by dengue and cholera. Among the Indian states, West Bengal records the highest number of cases, approximately 178,830, followed by Delhi (80,933), Uttar Pradesh (60,173), Maharashtra (53,575), and Karnataka (39,532).³² In 2017, there were 219 million cases of malaria and 435,000 deaths globally. Meanwhile, cases decreased by 80% between 2015 and 2023, with 2 million cases and 83 deaths reported in 2023.³³

Northeast India, especially the Imphal Valley, has seen a significant emergence of dengue cases from 2020 to 2022. In 2021, there were 168 cases, and in 2022, the number increased to approximately 452, with three deaths. From 2020 to 2022, the average summer temperature in Imphal Valley ranged from 32.1 °C to 33.2°C, the average winter temperature was 1.5 °C–2.3°C, and the annual rainfall was 1640 mm–1810 mm. The findings suggest a link between climate variation and the spread of dengue.³⁴

Overall, these observations highlight the growing influence of climate variability on infectious disease across India. However, robust conclusive scientific evidence linking climatic factors to disease incidence and hospitalization rates remains limited. Comprehensive longitudinal studies and enhanced surveillance data are required to establish causal associations, and forthcoming large-scale surveys are expected to provide more definitive insights.

Climate change and vector-borne diseases

Climate change parameters associated with vector activities, including reproduction, maturation, and disease vector migration, show varying responses. New ecological niches, such as water puddles, serve as better breeding sites for mosquitoes.³⁵ Additionally, sea level rise, increased groundwater salinity, and greater coastal influx create favorable conditions for vector proliferation and disease transmission. Table 1 illustrates the impact of climate change on vector-borne diseases.⁴⁸

Table 1.

Impact of climatic factors on the emergence and spread of vector-borne diseases

Diseases	Pathogens	Vectors	Climate change parameters impacting vectors and diseases	References
Dengue	Dengue virus (DENV)	Aedesaegypti	vector egg hatching at 25 °C; the survival rate of female Aedesaegypti is at 27.1 °C, and the blood-feeding rates are highest at 30 °C	Doeurk et al.36
Zika fever	Zika virus	Aedesaegypti	transmission temperature ranges between 22.7 °C and 34.7°C	Tesla et al.37
Chikungunya	Chikungunya virus	Aedesaegypti, and Aedes albopictus	blood-feeding rates are highest in Aedesaegypti at 30 °C, but 25 °C for Aedes albopictus	Doeurk et al.36
West Nile fever	West Nile virus (WNV)	Culex quinquefasciatus	mean annual summer temperature (between 24 °C and 28°C), spring temperature (>15 °C), relative humidity (<77%)	Farooq et al.38
Malaria	Plasmodium falciparum	Anopheles gambiae	the optimal growth temperature is at 24 to 28 °C	Suh et al.39
Yellow fever	Flavivirus	Aedes or Haemagogus species	transmission temperature between 24 °C and 30°C	Gaythorpe et al.40
Filariasis	Wuchereria bancrofti, Brugia malayi	Culex quinquefasciatus and Mansonia annulifera or M. uniformis	mean daily temperature ranges between 24.5 °C and 26°C, 200 m elevation, daily precipitation 2.6 to 3.8 mm for transmission	Kwarteng et al.41
Lyme disease	Borrelia burgdorferi	Ticks (Ixodes species)	warmer temperatures and fall in altitude (for every 1000 m fall in altitude, there is 6.5 °C rise in temperature)	Gilbert42
Monkey fever or Kyasanur Forest Disease (KFD)	KFD virus	Ticks (Haemaphysalis spinigera)	the annual mean temperature is between 23 °C and 26.2°C, and precipitation of 500–650 mm	Pramanik et al.43
Leishmaniasis	Leishmania species	Female phlebotomine sand flies	air temperature (between 20 °C and 30°C) and air humidity (60%)	Trájer44
Sleeping sickness	Trypanosoma bruceigambiense	Tsetse fly (Glossina species)	temperature is between 20.7 °C and 26.1°C, and relative humidity (0–99%)	El-Sayed and Kamel45
Chagas disease	Trypanosoma cruzi	Triatomine bugs	warmer and wetter environments, Temperature is between 20.7 °C and 26.1°C, and relative humidity (0–99%)	El-Sayed and Kamel45
Schistosomiasis	Schistosoma species	Aquatic snails	favorable temperature for development (20 °C–27°C) and temperature for transmission (27 °C–33°C)	Tabo et al.46
Ebola	Ebola virus (EBOV)	Bats (Mops condylurus)	the body temperature range of vectors between 27 °C and 37°C causes the growth of the pathogen	Bokelmann et al.47

Vector biology and ecology

Vectors are highly sensitive to environmental conditions, particularly temperature and humidity. Mosquitoes, being poikilothermic, depend on environmental temperature for survival and development; slight fluctuations can affect their activity. From egg hatching to adult survival, everything depends on the specific temperature for every stage. For example, the Aedes aegypti egg hatching rate is elevated, yielding 97.7% at 25 °C, whereas for Aedes albopictus, it is 20 °C, achieving 90.63%. Larval stages cannot survive at 40 °C. In addition, development is decreased at 35 °C for immature stages. The survival rate of females of Aedes aegypti is 27.1 °C, while Aedes albopictus is 24.5 °C. Moreover, the blood-feeding rates are highest at approximately 61% in Aedes aegypti at 30 °C, but at 25 °C for Aedes albopictus, which is about 52.5%. This suggests that Aedes aegypti survives better at higher temperatures than Aedes albopictus and has a major impact on dengue fever prevalence.³⁶

Similarly, ticks can also be affected by climate change. Altitude and temperature are important factors in the tick population; for every 1000 m fall in altitude, there is approximately a 6.5 °C rise in temperature. A maximum temperature between 10.5 °C and 26.3°C and relative humidity greater than 78.8% are associated with the prevalence of Lyme disease.⁴⁹

In the case of leishmaniasis, generally, air temperatures of about 25 °C and 60% humidity are favorable for the distribution of several sand fly vector species. The activity of the vector, Phlebotomus neglectus, is most active in a temperature range between 20 °C and 30°C.⁴⁴ Temperature can influence the time of blood meal digestion, development, and survival time of vectors. As the temperature drops below 20 °C, the activity of adult and larval development also slows rapidly; however, global warming resulting from anthropogenic activity provides more favorable conditions for the spread of leishmaniasis infection.⁵⁰

Moreover, the tsetse fly, Glossina pallidipes, a vector of the disease trypanosomiasis (sleeping sickness), is heat-sensitive; however, male fertility is more affected than female fertility. At 40 °C, 100% mortality is recorded, but at 38 °C, only 8% acute mortality has been shown. Exposure to 38 °C delays offspring production by 1 week and results in a female-biased sex ratio.⁵¹

Human and animal health implications

Vector ecology shifts are closely linked to patterns of disease transmission in humans and animals. For instance, Aedes aegypti’s greater tolerance to higher temperatures gives it a larger role in spreading dengue than Aedes albopictus. Tick-friendly climates elevate Lyme disease risk, while increasing temperatures are widening the distribution of leishmaniasis and sleeping sickness. They are attributed to more than 17% of all infectious diseases and are responsible for approximately 700,000 deaths annually.⁴⁸

Collectively, vectors and pathogens take advantage of the warmer, wetter environments that climate change creates. There are interconnected effects across environmental, vector, and health systems. Warmer temperatures and altered ecosystems enhance vector survival and reproduction, thereby increasing the risk of disease transmission to humans and animals. This relationship highlights the importance of a One Health approach in understanding and managing vector-borne diseases.

Climate change impacts bird and bat migration and the geographical distribution of disease vectors and pathogens

Climate change variables can influence the spread of infectious diseases across distant lands through the biology of vector species. They are increasingly changing the distribution, behavior, and ecology of avian species, increasing the possibility of transmission. Birds, being highly mobile, play a significant role in the transmission of zoonotic and vector-borne diseases as hosts. Birds migrate long distances across continents, and such migratory populations are likely to carry diseases, such as WNV, Lyme disease, avian influenza A (H5N1), and many others, such as Japanese encephalitis. Several aquatic bird species are known carriers of avian influenza.⁵² Rising temperatures, irregular rainfall patterns, and habitat degradation are reshaping bird migration patterns and the risk of exposure to vectors, such as mosquitoes, ticks, and other invertebrates. Diseases from migratory birds often spread to poultry populations when they come into contact with them. The prevalence of Campylobacter in wild birds, its potential to impact domestic bird populations, as well as its high rate of spread during climate-oriented fluctuations, have been documented.⁵³ Widespread culling of birds, both domestic and wild, has the potential to disrupt the economy and carries a risk to human health.

Disease vectors such as mosquitoes, which spread avian malaria to birds, are another issue. Such arthropods have more favorable breeding conditions and can spread further due to global warming. Bird populations have faced the consequences of anthropogenic influences on climate change in the Anthropocene era. There has been a noticeable drop in these populations due to habitat degradation and climate change. As habitat loss increases and population sizes decline, inbreeding among migratory populations may further increase their vulnerability to disease and its spread. This has further consequences for species populations and public health.

Bats work as hosts for several viral pathogens. Just as bird migration patterns change, the changing climate also changes bat migration patterns. This is of concern across Western and Eastern Africa, including Central Africa, where the Ebola virus caused an outbreak due to bats,⁵⁴ which are the hosts, coming in close contact with humans. Bats are known to migrate, and climate variability pushes these animals to move closer to human habitation.⁵⁵ Similarly, extreme weather events can displace human communities into forests in search of better land, thus increasing the exposure to infections from wildlife. The Kyasanur Forest disease (KFD) in India was, until recently, endemic in the Shimoga district of Karnataka. In the Indian states of Maharashtra, Goa, and Kerala, a study attributed these increasing cases to the ticks, the disease vector, and high precipitation in the months preceding the outbreak.⁵⁶ Figure 3 illustrates the climate change-linked disease transmission routes through birds, bats, and arthropod vectors to the animal and human populations.

Figure 3.

Climate change-linked disease transmission through birds, bats, and arthropod vectors to animal and human populations

Impact of climate variables on non-communicable diseases

NCDs are chronic medical conditions that are growing global health emergencies and may pose a serious challenge to human health due to their high mortality rate. The main factors responsible for NCDs are genetic, physiological, behavioral, and environmental.⁵⁷ Environmental factors include air pollution, chemical exposure, noise, urbanization, and climate change. Figure 4 shows the effect of climate change on NCDs.

Figure 4.

Effect of climate change on non-communicable diseases

Skin Diseases are primarily linked to UV radiation, which causes DNA damage by producing ROS and pyrimidine dimers, suppresses DNA repair mechanisms, and induces apoptosis.⁵⁸ UV light is beneficial for synthesizing vital vitamins and nutrients. Their overexposure poses significant human health risks, including eye diseases, photodermatoses, skin cancers, and premature aging.⁵⁹ Climate change may contribute to the release and spread of airborne allergens such as pollen and fungal spores, which trigger allergic diseases, including atopic dermatitis. It is influenced by environmental factors such as temperature fluctuations, humidity, air pollution, and wildfire.⁶⁰

Climate change may have both direct and indirect impacts on diabetes. Heat, relative humidity, and PM_2.5 from wildfire smoke, open waste burning, and vehicle exhaust make individuals with diabetes more prone to dehydration, impaired thermoregulation, and hemoconcentration, which in turn increases blood glucose concentrations. Long-term heat exposure increases the release of counter-regulatory hormones, such as cortisol, glucagon, adrenaline, and growth hormones, thereby making body tissues less sensitive to insulin.⁶¹ Furthermore, climate change can play a pivotal role in the prevalence of neurological disorders, particularly Parkinson’s, Alzheimer’s, and motor neuron diseases.⁶² It can alter the risks of multiple sclerosis, headache, migraine, and epilepsy. Rising global temperature and changes in rainfall patterns influence the growth and proliferation of algal blooms in eutrophic water bodies. Cyanobacteria and microalgae synthesize aerosol or waterborne neurotoxins (cyanotoxins) as secondary metabolites, contributing to neurodegenerative disease development.⁶³

Cardiovascular diseases are associated with air pollutants, such as carbon monoxide, nitrogen oxides, sulfur dioxide, ozone, and PM.⁶⁴ The smoke released from biomass burning, vehicular exhaust, and wildfires contains hazardous air pollutants that exacerbate several cardiovascular diseases by weakening the body’s immune system, damaging blood vessels, increasing blood pressure, and hardening the arteries.⁶⁵

Moreover, climate change is also negatively linked with respiratory health.⁶⁶ Elevated levels of PM may lead to inflammatory and oxidative stress, damage lung epithelium, and aggravate lung injury, while nitric oxide causes airway inflammation in chronic obstructive pulmonary disease (COPD) patients. Ozone can oxidize the phospholipids and proteins in the lungs, damaging the lung epithelium. Air pollutants exacerbate several airway diseases by interfering with the lungs’ defense mechanisms, increasing the incidence of chronic cough, triggering bronchial inflammation, and causing oxidative stress.⁶⁷

Impact of climate change on the spread of antimicrobial resistance

Microbial physiology may be impacted by climate variables such as rising temperatures, changes in precipitation, anoxia, and hypoxia.⁶⁸ One of the least explored impacts of climate change on human health is the development of AMR. It has been reported that extreme weather events, including temperature fluctuations, heavy rainfall, drought, and heatwaves, are associated with the spread of AMR in the community.⁶⁹ Higher temperatures increase bacterial growth and reproduction, while floods spread antibiotic-resistant bacteria (ARB) and antimicrobial-resistant genes (ARGs) through broken aquaculture and water supply systems, enhancing gene transfer.⁷⁰ Lower temperatures mitigate AMR development, but lower precipitation increases the rate of AMR, while higher precipitation reduces the incidence of AMR.⁶⁹

Temperature fluctuations and anoxic conditions may influence horizontal gene transfer (HGT) and promote AMR spread by increasing bacterial cell membrane permeability, intercellular adhesion, pilus growth, plasmid replication, and conjugation tube formation.^71,72 Excessive ROS under heat causes cellular damage and may induce the SOS response.⁷³ Additionally, over-precipitation and floods mix antibiotics and other pharmaceuticals with nearby water sources, which may lead to eutrophication and exaggerate ARG expression and mutagenesis. Droughts lead to poor sanitation, which is responsible for various diseases and increases antibiotic use.⁷⁴

A 10 °C rise in temperature increases antibiotic resistance in E. coli, Klebsiella pneumoniae, and Staphylococcus aureus by 4.2%, 2.2%, and 2.7%, respectively.⁷⁵ The mean temperature in Europe is linked to methicillin-resistant Staphylococcus aureus (MRSA), multidrug-resistant E. coli (MREC), and carbapenem-resistant Klebsiella pneumoniae (CRKP). Warmer temperatures and increased humidity intensify vector-borne bacterial disease prevalence, which may result in hospitalizations and antibiotic use, exerting pressure on resistance. Bacteria acquire resistance through modifications in cell-wall structure, the production of resistant enzymes, HGT,⁷⁶ target gene mutations, antibiotic degradation, and biofilm formation.⁷⁷

Impact of temperature fluctuations on horizontal gene transfer

HGT is the movement of genetic information between microorganisms through mechanisms such as conjugation, transduction, and transformation. It plays a vital role in bacterial and archaeal evolution. Several factors influence the HGT mechanisms, including pollutants, microbial physiological mechanisms, quorum sensing, oxidative stress, and environmental factors.⁷⁸ Climate change is an emerging factor contributing significantly to AMR by altering environmental conditions that help bacteria survive, reproduce, spread, and exchange resistant genes. Global warming promotes HGT, stress-induced mutagenesis, and efflux pump expression, with a 10 °C increase correlating with a 4%–17% rise in bacterial resistance. Increased UV-B radiation helps microbes uptake fragmented extracellular DNA and increases plasmid conjugation by 2 to 10-fold.⁷⁹ While global warming, in combination with UV irradiation, accelerates the spread of Shiga toxin (stx) genes in E. coli by enhancing Stx prophage induction and Stx phage-mediated gene transfer.⁸⁰ In Klebsiella pneumoniae, conjugative transfer of carbapenemase genes, such as blaKPC (Klebsiella pneumoniae carbapenemase) and blaNDM (New Delhi metallo-β-lactamase), is found to be temperature-dependent.⁸¹ Studies also show that conjugation of R27 conjugative plasmid genes (tra genes and IncHI plasmids) in Salmonella and other Enterobacteriaceae is thermosensitive.⁸² Conjugative transfer frequency of IncP-1 plasmid pKJK5 to wild E. coli strains varied significantly with a decrease in temperature and nutrient availability.⁸³

High temperatures generate ROS, may result in bacterial SOS response, and generate pilus channels. Temperature stimulates ATP production and facilitates conjugation. Recent studies have provided experimental evidence indicating the effect of increased temperatures on the transfer of antibiotic resistance genes. Where the study shows that different temperatures (17 °C, 27 °C, 37 °C, and 42 °C) affect the conjugation of ARGs under residual chlorine, with transfer frequencies of 1.07–2.43, 1.20–4.80, and 1.24–2.82, respectively.⁷² Conjugation transfer of plasmid-mediated ARGs under two different temperature ranges was found, and at 5 °C, the transfer rate was 17.16 × 10⁻⁷, which increased to 2.18 × 10⁻⁵ at 30 °C.⁷³ Oxygen availability also influences HGT; anaerobic and/or thermophilic bacteria and archaea exhibit higher gene transfer rates.^84,85,86 Another study found that a 10 °C increase in temperature is associated with approximately 5% increases in AMR for E. coli, 3.5% for Klebsiella pneumoniae, and 3% for Staphylococcus aureus.⁸⁷

In addition to this laboratory-based experimental evidence, an ecological analysis was conducted to clarify the actual effect of temperature on AMR. Data were obtained on AMR of E. coli from 31 provinces in China over the period of 2014–2020. They estimated that a 1 °C increase in average ambient temperature was associated with a 2.71% increase in the prevalence of E. coli resistance to cephalosporins, a 32.92% increase in carbapenems, and a 1.81% increase in quinolones.⁸⁸ The limitations of such associations depend on various factors, including healthcare services, environmental and socioeconomic factors.

Figure 5 illustrates changes in HGT in microbial communities due to climate change.

Figure 5.

Role of climate change in horizontal gene transfer

Artificial intelligence tools and their role in disease prediction and diagnosis

A perfect AI system has a “learn from experience” feature. It can grasp and respond to the external environment. In 1956, John McCarthy introduced the concept of AI at the Dartmouth Conference, mentioning that it imitates human intelligence.⁸⁹ In 1959, Arthur Samuel introduced the field of ML, a subfield of AI. Since then, ML has been applied to predict and diagnose diseases and to forecast outbreaks. These advancements have contributed to improved early disease detection and more efficient public health responses, potentially reducing disease burden and mortality.⁹⁰

Recently, various AI-powered algorithms, including convolutional neural networks (CNNs), deep learning (DL), linear regression (LR), support vector machines (SVMs), the Internet of Things (IoT), remote sensing, and data analytics, have been developed for disease surveillance and prediction. These approaches have diverse applications across multiple domains, including epidemiological modeling, physiological data analysis, and contact tracing.⁹¹ They have several promising applications in healthcare, including medical image analysis (e.g., X-rays and MRI), microscopic image detection, real-time ECG analysis, personalized medicine, medical robots, and electronic health records (EHRs). However, challenges persist regarding data quality, interpretability, and real-world implementation.

This section focuses on the application of AI/ML in climate-sensitive disease prediction and diagnosis, particularly the integration of environmental and epidemiological data to identify high-risk areas and enable early warning systems. Applications in clinical diagnosis and broader environmental modeling are discussed selectively where they directly inform climate–health linkages.

AI for climate-sensitive disease prediction and diagnosis

Dengue fever risk can be predicted based on climatic variables, such as maximum and minimum temperature, humidity, and precipitation, as well as socioeconomic factors. The highest predictive accuracy was achieved using a hybrid mathematical framework integrating spatial analysis techniques and multiscale geographically weighted regression (MGWR) with LSTM-based neural network architectures. This model performs better in short-term prediction. It is also used to estimate case prevalence at the state or provincial level.⁹² Across studies, hybrid models combining statistical approaches (e.g., LASSO, DLNM) with DL architectures, such as LSTM, consistently outperform single-method models, particularly for short-term forecasting. Incorporating spatial heterogeneity via geographically weighted regression or clustering further improves predictive accuracy.⁹³

RF and eXtreme gradient boosting (XGBoost) algorithms help predict hand, foot, and mouth disease (HFMD).⁹² HFMD is a pediatric infectious disease that requires accurate forecasting for effective control. Autoregressive integrated moving average (ARIMA) and Moirai are traditional models that perform similarly for short-term predictions, while TimesFM-500M performed best for 5-step forecasts and TimesFM-200M for 10-step forecasts, with shorter training inputs. Overall, foundation models show strong forecasting potential but require greater computational resources, with zero-shot capability eliminating the need for retraining.⁹⁴

In the case of Malaria, the LSTMSeq2Seq model predicts their re-emergence with an accuracy of 87.3%.⁹² Some computational models, including LSTM, gated recurrent units (GRU), support vector regression (SVR), RF, XGBoost, and ARIMA, were analyzed to predict malaria prevalence. The models were selected to compare traditional and modern forecasting approaches. LSTM and GRU, based on recurrent neural networks, capture long-term temporal dependencies in time series data. SVR, RF, and XGBoost are robust traditional ML models, while ARIMA serves as a widely used statistical baseline for comparison. Among them, the RF model consistently outperformed the others, with the lowest root mean squared error (RMSE) and mean absolute error (MAE). Furthermore, incorporating K-means clustering enabled the models to better represent spatial heterogeneity, resulting in more accurate predictions through localized adaptation. This hybrid approach improves malaria surveillance and supports more effective public health strategies.⁹⁵ Despite high reported accuracy, many diagnostic models rely on curated datasets and may not perform similarly in real-world clinical settings.

AI for infectious disease diagnosis

Advanced DL models, such as EfficientNetB3, are used to detect tuberculosis (TB). Segmented images of the region of interest in chest X-rays (CXRs) have been added to the model. By using this advanced method, healthcare professionals can distinguish between TB and non-TB cases with an average accuracy of 98.7%.⁹⁶ A recent study demonstrates that a convolutional autoencoder neural network and a multi-scale CNN with deep-layer aggregation, combined into an ensemble learning architecture, outperform existing classifiers. The ensemble model achieved strong generalization, effectively identifying active pulmonary TB on CXRs in a worldwide cohort with 99% sensitivity and 94% specificity.⁹⁷

In AI algorithms, physiological data analysis enables the detection of subtle changes in vital signs and biomarkers, which can help identify infections and disease outbreaks.⁹⁸ During the COVID-19 period, early diagnosis involves collecting data from numerous sensors, including potentiometers, blood pressure sensors, graphene field-effect transistor (G-FET) sensors, and electrochemical sensors. After collection, the data were preprocessed, and the African buffalo optimization (ABO) method was used for feature extraction. Then, the shuffle shepherd optimization-based generalized deep convolutional fuzzy networks (SSO-GDCFN) method was used to analyze the features, resulting in 99.99% accuracy in diagnosing COVID-19. During that period, aerial vehicles with thermal imaging cameras also identified individuals without masks and recognized abnormal body temperatures in crowds. This technology operated autonomously to detect potential COVID-19 patients.⁹¹ Additionally, NCDs can be diagnosed using an AI/ML approach, such as a deep neural network framework based on deep SHapley additive explanations (DeepSHAP), which includes a feature selection technique to predict NCDs.⁹⁹ This model can contribute to a medical decision support system and provide practical requirements.

We analyzed papers published between 2020 and 2024 that discussed the use of AI/ML tools in infectious diseases to assess the status of AI/ML applications worldwide. Surprisingly, we found that China consistently produces the most AI-related research outputs each year, followed by the USA, India, the UK, and Italy. The dominance of high-income countries in AI research output highlights global inequalities in data availability and model development, potentially increasing the risk of bias. Figure 6 provides a detailed illustration of the use of AI/ML tools in infectious diseases.

Figure 6.

Overview of research articles published worldwide on the line of artificial intelligence use in infectious diseases (the references have been provided in supplementary materials)

The application of AI tools in various fields has made significant progress, enabling more accurate disease diagnosis. A wide range of diseases, including bacterial, viral, fungal, and NCDs, can now be detected and diagnosed using different AI/ML algorithms. Details are summarised in Table 2 for infectious diseases and Table 3 for NCD_s. These methods are cost-effective, rapid, and efficient, and eliminate the need for culture-based colony isolation, costly equipment, and specialized human resources.

Table 2.

AI/ML tools for the detection, prediction, and diagnosis of infectious diseases

AI/ML tools and algorithms	Application in detection, prediction, and diagnosis of infectious diseases	Model accuracy	References
Bacterial diseases
RF, SVM, and XGBoost	predict Antimicrobial Phenotype Resistance in Staphylococcus aureus	–	Wang et al.100
XGBoost model	predict LTFU (Loss to Follow-up) risk in TB patients	82.5%	Chen et al.101
You Only Look Once version 4 (YOLOv4)	accurately identify the presence of E. coli at the microcolony stage in food products	94%	Ma et al.102
Tuberculosis segmentation-guided diagnosis model (TSSG-CNN)	detect TB	98.75%	Kim et al.103
DeepLysin	identify antibacterial proteins	–	Zhang et al.104
DL-based AI-assisted lesion method	identify pulmonary infection in CT images	99.76%	Wang et al.105
Deep neural network antibiotic resistances (DnnARs)	identify antibiotic-resistant strains of E. coli bacteria	97%	Nayak et al.106
DL-Xception model	detect TB accurately in chest X-ray images	99.29%	Sharma et al.107
TrioFusionNet (ResNet-50, InceptionV3, and EfficientB4) + Ensemble of AMS Optimization (EAMSO) model	early detection of TB through chest X-rays	98.80%	Dhivyaa et al.108
EfficientNetB3	detection of TB from chest X-rays	99.1%	Nafisah and Muhammad96
Light gradient boosting machines (LGBM)	rapid detection of CZA resistance in Klebsiella pneumoniae	95%	Lin et al.109
Viral diseases
ViRAL	identification, classification, and lineage assignment of SARS-CoV-2 variants	99.8%	Jahshan et al.110
Passive aggressive (PA) classifier	predicting severe COVID-19 cases	70-80%	Bhardwaj et al.89
Shuffle-Shephard optimization-based generalized deep convoluted fuzzy network	early diagnosis of COVID-19	99.99%	Alam et al.111
Deep CNNs	detect COVID-19 pneumonia from digital chest X-ray images	99.7%	Chowdhury et al.112
LASSO, DLNM with DL architectures, such as LSTM	predict dengue	–	Cheng et al.93
RF and XGBoost algorithms	predict hand, foot, and mouth disease (HFMD)	–	Santangelo et al.92
LSTMSeq2Seq model	predict malaria	87.3%	Santangelo et al.92
Bi-directional long short-term memory (Bi-LSTM) and RF	predict influenza cases	–	Zhu et al.113
VIBRANT	virus identification and characterization from metagenomic and genomic sequences	99.15%	Kieft et al.114
Deep-AVPpred	DL classifier for the discovery of Antiviral Proteins (AVPs) in protein sequences	94%	Sharma et al.115
Fungal diseases
Hybrid learning-based neural network classifier (HLNNC)	rapid identification of the Black fungus	99.5%	Fang et al.116
Electric Nose (E-Nose) using sensor arrays and pattern recognition systems	detect odors or flavors of fungus	–	Fang et al.116
In vivo confocal microscopy (IVCM) DL algorithm	identification of fungus species based on the presence or absence of fungal hyphae	96%	Fang et al.116
Regional convolutional neural network (RCNN) and SVM integrating with modified neural networks	detection and diagnosis of Mucormycosis (Black Fungus) in COVID-19 patients	81.65%	Annie Grace Vimala et al.117
DeePathology® STUDIO (fungi detection algorithm)	identifying skin fungal infection	93%	Rappoport et al.118
DL classifier model using IVCM image	identification of Fusarium and Aspergillus	Fusarium 81%, Aspergillus 75%	Tang et al.119
Meta-learning-based DL architecture named MeFunX	early detection of fungal infection from microscopic images	92.49%	Rawat et al.120
CNN	automatic fungal detection	98.03%	Shakir and Mahdi121
Deep neural network, VGG16, using microscopic images	DIAGNOSIS OF BRONCHIAL ASTHMA	99.84%	Shakir and Mahdi121

Table 3.

AI/ML tools for the prediction, detection, and diagnosis of non-communicable diseases

AI/ML algorithms	Application in prediction, detection, and diagnosis of non-communicable diseases	Model accuracy	References
DNN	diagnosis of bronchial asthma and COPD	–	Ijaz et al.122
LR	COPD and congestive heart failure	–	Ijaz et al.122
CNN and MobileNet	asthma and COPD	94.9%	Ijaz et al.122
CNN	early detection of precancerous lesions	96%	Ali et al.123
AI smartphone applications using neural network vision algorithm	medication reminders and dosing instructions	–	Babel et al.124
mPulse Mobile	SMS-based reminder solution for older patients with NCD	–	Babel et al.124
RF	early diagnosis of Chronic NCD	96.16%	Costa and Gouveia125
DeepSHap-based deep neural network framework	NCD prediction and diagnosis	–	Davagdorj et al.99
DCNN	early detection of Eye diseases like Diabetic Macular Edema	82%	Ramirez-Bautista et al.126
ANN	prediction of type 2 diabetes mellitus	98.16%	Ramirez-Bautista et al.126
Multilayer perceptron with 6 input neurons	diagnosis of bronchial asthma	91%	Ramirez-Bautista et al.126
XGBoost ML algorithm	diagnosis of ovarian tumor considering features from blood tests, patient history, and imaging tests	80%	Ramirez-Bautista et al.126
Neural network with Levenberg-Marquardt learning	breast cancer classification	99.68%	Ramirez-Bautista et al.126
KNN using real-time ECG analysis	predict cardiac arrhythmias	99.50%	Ramirez-Bautista et al.126
ANN	predict diabetes	92.15%	Shimirwa et al.127
SVM	predict heart disease	92.59%	–
CNN and capsule neural network-based method	analyze brain tumor	–	Surianarayanan et al.128
Two-dimensional deep convolution encoder linked to a neural network	detect seizures in children	–	Surianarayanan et al.128
Autoencoder-based DNN	detect autism by analyzing fMRI data	–	Surianarayanan et al.128
DNN	detect Schizophrenia (SCZ) by analyzing functional connectivity patterns in fMRI data	–	Surianarayanan et al.128
3D neural network architecture	detect Alzheimer’s disease (AD)	–	Surianarayanan et al.128
K-NN	early and accurate diagnosis of Parkinson’s Disease (PD)	96.07%	Lima et al.129

DNN; deep neural network, LR; logistic regression, CNN; convolutional neural network, DeepSHap; deep shapley additive explanations, ANN; artificial neural network, RF; random forest, SVM; support vector machine, DCNN; deep convolutional neural network, K-NN; K nearest neighbors, XGBoost; eXtreme gradient boosting algorithms.

AI for environmental and climatic forecasting

The AI tools listed in Table 4 are used to predict climatic factors, including solar radiation, daily temperature, relative humidity, water turbidity, rainfall, wind speed, and forest fires. Rainfall prediction techniques, including conventional seasonal autoregressive integrated moving average with exogenous factors (SARIMAX), an ML-based LSTM model, and a hybrid model integrating variational mode decomposition (VMD) with LSTM, have been applied recently.¹³⁴ Other AI tools can assess the quality of drinking water, monitor soil health, and measure pollutant concentrations, such as O₃, NO_x, and CO.

Table 4.

AI/ML tools for the prediction of climatic factors

AI/ML tools and algorithms	Application in the prediction and analysis of climatic conditions	References
Microbial potentiometric sensor (MPS) signals and ML tools	predict the canal surface water turbidity, conductivity, chlorophyll, blue-green algae (BGA), dissolved oxygen (DO), and pH	Saboe et al.130
IoT and ML in the SVM	predict waste generation, classification, route optimization, and estimation of methane emissions	Joshi et al.131
Artificial neural network-multilayer perceptrons (ANN-MLPs)	predict weather conditions like solar radiation, air temperature, relative humidity, and wind speed	Dahmani et al.132
RF and boosted regression tree (BRT)	predict forest fire	Olawade et al.133
LSTM	predict the concentration of O3, PM2.5, NOx, and CO	Olawade et al.133
ANN neuro fuzzy regression and DL-LSTM	analyze and forecast air pollutants	Olawade et al.133
Adaptive neuro-fuzzy inference system (ANFIS) and Multi-layer perceptron neural network (MLP-NN)	forecast water level change	Olawade et al.133
AdaBoost and XGBoost	test the drinking water quality	Olawade et al.133
CNNs and RNNs	analyze soil health monitoring and detect nutrient deficiency	Olawade et al.133
Wavelength artificial neural network (WANN)	predicting Rainfall	Dahmani et al.132
Seasonal autoregressive integrated moving average with eXogenous factors (SARIMAX), an LSTM model, and a hybrid model integrating variational mode decomposition (VMD) with LSTM	predicting rainfall	Singh and Choudhary134

Overall, this section strengthens the application of AI/ML approaches across diverse domains of disease prediction and systematically outlines various tools relevant to both infectious and NCDs within the context of climate change. However, most datasets are strongly biased toward high-income regions, potentially introducing systematic error and reducing performance for underrepresented populations. This raises concerns about equity and reliability when applied in endemic, data-scarce regions. Performance is often evaluated on internal or similar datasets, limiting evidence of external validity. Models trained in one geographic or climatic setting may not generalize to other settings due to differences in disease ecology and reporting systems. This creates a gap between model development and real-world feasibility.

Conclusion

Climate change, driven by both natural processes and anthropogenic activities, is raising global mean temperatures, melting glaciers, causing heatwaves, altering precipitation patterns, increasing humidity, and warming oceans, with profound consequences for ecosystem stability and human health. Human health and disease dynamics are directly related to these ecological and environmental disturbances. The continued alteration in these variables created favorable conditions for the survival and proliferation of existing pathogenic organisms and their vectors, including mosquitoes, ticks, sand flies, and snails. It also created conditions for the emergence of new pathogens or the transformation of environmental microorganisms into pathogens. Their geographic distribution has changed, and many may have transitioned from endemic to epidemic patterns. The climate-induced changes directly influence disease transmission and contribute to the emergence, re-emergence, and intensification of diseases worldwide.

This review highlights the association between climatic variables and the rising burden of infectious diseases, particularly vector-borne diseases such as dengue, malaria, Ebola, and West Nile fever. Alterations in temperature, rainfall, flooding, and drought not only modify vector breeding habitats but also affect pathogen viability and host susceptibility. Climate-driven changes in bird migration patterns further facilitate cross-border disease transmission, breaking traditional biogeographical barriers. These changes can spread avian influenza viruses and other pathogens across regions, altering the epidemiology of bacterial and fungal diseases.

Similarly, environmental stressors, including air pollution, suspended PM, and exposure to extreme weather events, contribute to the rising prevalence and severity of NCDs. Wind velocity also plays a role in dispersing contaminated aerosols, toxic gases, and fumes, further affecting community health. Climate-driven water contamination further contributes to the spread of waterborne diseases within the community, while drought increases the density of suspended PM in the air, elevating the risk of respiratory and airborne infections. Furthermore, studies suggest that temperature rise and anoxic conditions play a critical role in HGT and AMR development by altering bacterial cell membrane permeability, intercellular adhesion, pilus growth, plasmid replication, and conjugation tube formation. Excessive ROS production also causes cellular damage and induces the SOS response. Despite growing evidence, the underlying mechanisms and long-term effects remain poorly understood. Furthermore, the increasing use of AI tools across various fields enables rapid and accurate disease detection. A wide range of diseases, including bacterial, viral, fungal, and NCDs, as well as climatic surveillance, can now be detected and diagnosed using widely implemented AI algorithms.

According to the United Nations Sustainable Development Goals (SDGs), these findings have a strong interconnection between SDG 13 (Climate Action) and SDG 3 (Good Health & Well-being). Achieving SDG 3 is dependent on advancing SDG 13, and vice versa. Combining climate and health policies is essential to achieving these SDGs and to reinforcing stronger protection against new climate change-related health risks. To execute these goals, we need to focus on strengthening prediction models with climate data and epidemiological surveillance. AI/ML-supported and big-data analytics accurately forecast disease incidence and AMR hotspots. Early diagnosis plays an important role in developing preventive measures that can significantly reduce loss of life. Additionally, interdisciplinary research is crucial to developing impactful solutions, while global collaboration and policy integration can help to mitigate these issues.

Overall, climate change represents a growing global health crisis with complex, interconnected impacts on infectious diseases, NCDs, and AMR. Addressing these challenges requires integrated, interdisciplinary research and policy approaches that align climate adaptation strategies with public health preparedness. Strengthening surveillance, advancing predictive models, and fostering global collaboration within a One Health framework are essential to mitigate future health risks and enhance resilience in the face of ongoing climate change.

Limitations of the study

Surveys of literature and studies of climatic variables indicate that several diseases are correlated with climate change, but extensive data on these aspects are lacking. There is a critical need to better understand climate-induced physiological changes in pathogens and vectors and their impact on incidence. We need to focus on how changes in climatic variables, such as temperature, precipitation, and extreme weather events, impact the likelihood of floods and droughts and the spread of diseases and pathogens in emerging vulnerable regions. How do these changes contribute to the increase and multiplication of vectors and their hosts in newly affected areas? Furthermore, how are the excess concentrations of GHGs, such as CO_2, CO, N₂O, and CH_4, affecting human health? How does the flux of these gases impact community health and the types of diseases they may trigger? These links are not explored in detail and require more attention.

Literature surveys indicate that the most recent studies are biased toward bacterial pathogens and neglect other groups. Furthermore, only a few factors such as the impact of warming on disease spread and development of AMR has been extensively studied in detail, while the impact of other factors, such as wind velocity, humidity, salinity, flood, drought, deforestation, forest fires, alternation in land use, rapid urbanization and mixing of contaminated water and pathogens from various sources are not yet studied in detail. Planning extensive studies using different pathogens, such as viruses, fungi, protozoa, and helminths, and examining the impacts of climatic variables will yield more in-depth information and assist in mitigation and management. In addition, AMR is an emerging challenge for clinicians and researchers, and a more in-depth understanding of the impacts of climate change on the development and dissemination of AMR is the need of the era. Emerging zoonoses, fungal infections, and behavioral issues are also of interest and should be addressed in future studies. The discipline is emerging, and systematic data on a specific theme are lacking; sporadic studies conducted across variable geographical locations exhibit heterogeneity in sample sizes, the nature of selected organisms or pathogens, and the impact of climatic variables as potential confounders. However, the outcomes of these studies indicate that warming and other climatic variables play a significant role in the emergence and spread of human diseases, and more systematic studies with larger sample sizes are necessary to obtain conclusive data for designing the management and mitigation strategies in different climatic zones. The application of AI/ML tools for disease detection and diagnosis in the context of climate change is constrained by infrastructure limitations, poor health system integration, and limited technical expertise, particularly in low-resource settings, where inconsistent data and funding gaps further limit their real-world application.

Important terminology

Climate change: Climate change refers to a change in the state of the climate that can be identified (e.g., by using statistical tests) by changes in the mean and/or the variability of its properties and that persists for an extended period, typically decades or longer.⁷

Infectious or communicable disease: An illness caused by an infectious agent or its toxins, and spreads from one individual to another by direct or indirect contact.¹³

Non-communicable disease or Non-infectious: Chronic medical conditions that are not transmitted from one person to another and are associated with long durations and slow progress.⁵⁷

Vector-borne diseases: Diseases that are caused by infectious pathogens and transmitted from infected humans, plants, and animals to others via vectors such as mosquitoes and ticks.¹³⁵

Antimicrobial resistance: The ability of microorganisms to survive and grow in the presence of antimicrobial agents.^136,137

Horizontal gene transfer: The process of transfer of genetic material from one organism to another by routes other than parent-offspring transfer.¹³⁸

Acknowledgments

Ujjwala Waghmare is obliged to Symbiosis International (Deemed University) for providing the Doctoral Fellowship to support this work.

Author contributions

A.A. contributed to the writing and editing of the manuscript and prepared the first draft. U.W. helped with writing and editing the manuscript. R.W. reviewed the manuscript. P.R. reviewed and gave suggestions on structuring the manuscript. A.M. helped in writing the manuscript. R.Y. reviewed the manuscript. O.P. participated in the design, writing, and editing of the manuscript. All authors read and approved the final manuscript.

Declaration of interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this manuscript, the authors utilized the AI-based language tool ChatGPT (OpenAI, USA) solely to correct English-language and grammatical mistakes, thereby improving the clarity and readability of the manuscript. The tool did not contribute to the scientific design, data collection, analysis, reporting, interpretation of results, or image generation. All authors reviewed and edited the AI-generated suggestions and take full responsibility for the content of the work.