Public Health Risks of Pathogenic Bacteria in Freshwater Bodies: A Review of Quantitative Microbial Risk Assessment Approaches and Applications

Abstract

Freshwater ecosystems play an important role in human survival, ecosystem functioning, and biodiversity conservation, yet industrialisation and urbanisation dump over 80% of untreated sewage into them. This inadequate wastewater management leads to enteric pathogens like Escherichia coli, Salmonella, Shigella, Campylobacter, Vibrio cholerae, Pseudomonas aeruginosa, and Legionella pneumophila that are responsible for a wide range of waterborne human diseases globally with extensive morbidity and mortality. The World Health Organization (WHO) estimates that at least 2 billion individuals drink water contaminated with pathogens, resulting in illnesses like cholera, dysentery, and diarrhoea, and approximately 50,000 diarrheal deaths annually. Classical epidemiology approaches are the basis for determining disease burden in public health, but they are limited in their capacity to predict future health risks. Quantitative microbial risk assessment (QMRA) addresses this by estimating the potential health risks of any exposure to microbial pathogens in any environment using four key elements, which include the identification of the microbial hazards, human exposure to the hazard through diverse activities, dose–response relationships, and the estimated risk of the infection. This review summarises information on freshwater pathogens, their occurrence, sources and health implications. The methodological approaches of QMRA in freshwater systems are reviewed with examples drawn from recreational activities, drinking water, and wastewater-impacted environments. Global QMRA studies indicate a wide range of infection risk estimates, reflecting differences in water sources, pathogens, and exposure conditions. Thus, QMRA is known to be a valuable public health tool for freshwater ecosystems, linking microbial contamination dynamics to health risk estimates that support proactive management and policy-relevant decision-making.

1. Introduction

Freshwater ecosystems (rivers, lakes, wetlands, ponds, glacial lakes, reservoirs and groundwater) are foundational to human survival, providing indispensable services that support civilisation and maintain global biodiversity, despite covering less than 1% of the Earth’s surface [1]. Freshwater continues to be a vital supply of water for drinking, domestic and other purposes in many developing countries, with over 144 million people globally relying directly on surface waters for drinking [2]. Despite their well-known benefits, rapid population growth accompanied by industrialisation and urbanisation severely affects freshwater ecosystems by introducing massive amounts of contaminants, thereby altering the quality of freshwater bodies and posing significant risks to the entire ecosystem [3].

Wastewater treatment plants account for most of the pollutants encountered in freshwater. A large proportion of untreated or poorly treated sewage and wastewater is discharged into water bodies, particularly in low-income countries where wastewater treatment facilities remain inadequate, posing risks to ecosystems and human health [4]. These pollutants include enteric microorganisms like viruses and bacteria, which cause diverse human infections, including skin infections and diarrhoea [5,6]. For example, it is estimated that 80% of infections and 90% of child fatalities worldwide are caused by poor water quality [7]. The United Nations (UN) reports that at least 2 billion people still lack access to safe water sources globally and rely on water polluted with waterborne pathogens, which cause illnesses like cholera, dysentery, and diarrhoea, and over 50,000 diarrheal fatalities per year [8]. Therefore, water quality assessment is essential since it provides proof of the dangers of human exposure associated with various water uses, preventing potential waterborne diseases and enabling investigations into the sources of pollution [9].

Quantitative microbial risk assessment (QMRA) estimates the potential health risks associated with any exposure to microbial pathogens across various environments, including water, food, and occupational settings. This methodology allows the quantification and characterisation of risks by integrating data on pathogen concentrations, exposure scenarios, and dose–response relationships to predict the probability of adverse health effects [9]. QMRA has been extensively applied in the context of water safety management, demonstrating its utility in supporting Water Safety Plans (WSPs) [10].

The United Nations Sustainable Development Goal 6 (SDG 6), which focuses on ensuring everyone has access to clean water and sanitation, depends heavily on Quantitative Microbial Risk Assessment (QMRA) [11]. QMRA provides a systematic approach to evaluate the risk of infection from pathogens, which is essential for making informed decisions about water and sanitation safety. This methodology enables water and wastewater utilities and water and sanitation hygiene (WASH) practitioners to better understand the real risks posed by pathogens, as opposed to relying solely on indicator microorganisms, which can be misleading when making decisions that impact public health and sustainability [12].

Through QMRA, public health risks linked to waterborne pathogens in drinking water and wastewater management can be quantified, allowing stakeholders to implement more effective safety measures and interventions. The methodology has been effectively applied across various scenarios, such as the chlorination of drinking water, surface water affected by wastewater discharge, and the reuse of water for non-potable applications. These applications illustrate the importance of QMRA in ensuring that the treatment levels meet safety criteria, thereby supporting SDG 6 targets [10,12].

Therefore, this review aims to critically examine the occurrence of microbial pathogens in freshwater ecosystems and their major sources of contamination. The review further evaluates the application of QMRA in assessing public health risks associated with pathogen exposure in freshwater environments. In addition, a comparison between classical epidemiological approaches and QMRA is presented, and key uncertainties associated with QMRA are discussed. Finally, the review highlights the public health benefits of QMRA in freshwater risk management while identifying current challenges and potential future research directions. By synthesising current knowledge and methodological approaches, this review provides a framework to improve microbial risk evaluation in freshwater systems and supports evidence-based decision-making for water safety management and public health protection.

Methodology

This study employed a narrative literature review to synthesise existing knowledge on pathogenic bacteria in freshwater ecosystems and the application of QMRA in evaluating associated public health risks. A broad Google search was first used to scan for existing literature on QMRA in freshwater bodies. Then, relevant articles were retrieved from scientific databases including Web of Science, Scopus, PubMed, and Google Scholar, using combinations of keywords such as “QMRA”, “pathogenic bacteria”, “microbial risk assessment”, “waterborne pathogens”, “E. coli”, “Salmonella enterica”, “Vibrio cholerae”, “Shigella”, “Campylobacter”, “Legionella”, “Aeromonas”, “Pseudomonas”, “Dose-response”, “Exposure assessment”. Boolean operators (AND, OR) were applied to refine results and capture relevant interdisciplinary studies in environmental microbiology, water quality, and public health. Studies were selected based on their relevance to microbial contamination in freshwater environments and the application of QMRA frameworks. The retrieved studies were screened based on relevance to the topic. Inclusion criteria comprised peer-reviewed articles, reviews, and reports addressing microbial pathogens in freshwater systems, their sources, occurrence, transmission routes, or persistence, as well as studies applying or discussing QMRA in water quality assessment. Publications were limited to those available in English, while studies focusing exclusively on marine environments, non-microbial contaminants, or those lacking relevance to freshwater public health risks were excluded. Information from the selected literature was extracted and organised according to major thematic areas, including types of pathogenic bacteria identified in freshwater systems, sources of contamination (e.g., wastewater effluents, agricultural runoff, and urban runoff), human exposure pathways, and the key methodological components of QMRA such as hazard identification, exposure assessment, dose–response modelling, and risk characterisation. The extracted information was synthesised qualitatively to provide an integrated overview of pathogen occurrence, transmission pathways, and the application of QMRA for estimating infection risks associated with freshwater use. Given its narrative nature, this review was not time-bound, and the literature included spans publications from 1960 to 2026, with most studies published after 2010, reflecting the rapid expansion of research on freshwater microbial contamination, antimicrobial resistance, and quantitative microbial risk assessment.

2. Pathogenic Bacteria in Freshwater Bodies

The growing anthropogenic stresses on aquatic ecosystems and the substantial public health hazards associated with waterborne infections have made research on pathogenic bacteria in freshwater an important field of study [13,14]. Although identifying these microbial pathogens represented a major challenge previously, in recent decades, due to developments in molecular and sequencing technology, the identification and description of bacterial pathogens in a variety of freshwater habitats, from natural lakes and reservoirs to urban rivers, has improved [15,16]. Microbial pollution of water sources is responsible for millions of fatalities each year, contributing to the significant worldwide burden of waterborne diseases [17,18]. In addition, the persistence and proliferation of antibiotic-resistant bacteria in freshwater systems amplifies the spread and severity of these infections and raises concerns about treatment efficacy and infection management [19,20].

2.1. Major Bacterial Pathogens in Freshwater

Numerous bacterial pathogens found in freshwater habitats are a serious hazard to human health due to their numerous modes of transmission. These pathogens can be broadly divided into two groups: environmental/opportunistic organisms that naturally persist or grow in aquatic systems (e.g., Legionella, Aeromonas, Pseudomonas, Mycobacterium) and enteric bacteria that are introduced through faecal contamination (e.g., E. coli, Salmonella, Shigella, Campylobacter) (

Table 1. Some major pathogenic bacteria identified in freshwater and their associated diseases.

Bacteria	Primary Source	Transmission Route	Associated Disease	References
E. coli	Human and animal faecal contamination, sewage effluents	Ingestion of contaminated water, direct contact, and recreational exposure	Diarrhoea, gastrointestinal illness, haemolytic uremic syndrome	[14,21,22]
Salmonella enterica	Sewage, Animal waste	Ingestion of contaminated water	Typhoid, fever, gastroenteritis	[13,14,23]
V. cholerae	Warm freshwater, sewage, Human faeces	Ingestion of contaminated water	Cholera, food poisoning	[24]
Shigella spp.	Human faecal waste	Ingestion of contaminated water	Bacillary dysentery	[13,14,24,25]
C. jejuni	Animal faeces (poultry), surface water, agricultural runoff	Ingestion of contaminated water	Campylobacteriosis/gastroenteritis	[26]
P. aeruginosa	Natural waters, Hospital effluents, biofilms	Dermal contact	Endocarditis, osteomyelitis, pneumonia, urinary tract infections, gastrointestinal infections, meningitis, septicaemia, folliculitis and ear infection	[27,28,29]
L. pneumophila	Engineered water systems (biofilms), natural freshwater	Inhalation of aerosols from contaminated water	Legionnaires disease	[29,30]

).

2.2. Sources and Transmission Routes

The continuous introduction of polluted water (regardless of the source) into freshwater bodies is a primary driver of environmental degradation, frequently altering the microbial community structure and significantly increasing the concentration of pathogens. Such pollution introduces high levels of nutrients (e.g., nitrogen and phosphorus) and organic matter, creating conditions that favour the rapid growth of pathogenic bacteria. Enteric pathogens, including opportunists and antibiotic-resistant Enterobacteriaceae, are among the freshwater pathogenic bacteria that come from point (sewage, wastewater effluent, and industrial discharges) and non-point (agricultural runoff, storm water, and wildlife) sources [17,31,32]. Once in the freshwater ecosystem, their persistence, transport, and ultimately, human exposure are determined by critical, interconnected factors, including hydrology, sediments, temperature, and biological stressors [33,34]. While pollution from point sources can be pinpointed and controlled, it is more challenging to create treatment and control techniques for non-point sources of microbiological pollution due to their diffuse nature [35].

2.2.1. Point Sources

Wastewater Treatment Plants (Sewage Effluents)

Unrestricted sewage or wastewater disposal from treatment plants worldwide has created a negative impact on the health of freshwater ecosystems in recent years. Investigation shows how wastewater treatment plants (WWTPs), which are frequently built to manage organic pollutants and nutrients, are typically ill-equipped to deal with pathogens and can lead to the spread of harmful microorganisms into the environment [36]. Many freshwater bodies receive a significant amount of antibiotic resistance genes (ARGs) from hospital and municipal sewage [37]. Therefore, WWTPs of sewage treatment plants (STPs) are considered potential sources of microbial pollutants, including antibiotic-resistant pathogenic bacteria, in freshwater bodies [38]. With rapidly growing human populations and fast-growing cities, protecting freshwater habitats from the effects of such untreated wastewater or sewage input is getting progressively harder [31].

The most prevalent microbiological pathogens in wastewater are bacteria. Listeria monocytogenes is one of the newly discovered wastewater bacterial pathogens that pose a serious threat to public health. Other bacterial pathogens in freshwater include Staphylococcus aureus, Salmonella spp., E. coli, and Clostridium perfringens which can cause infections such as diarrhoea, dysentery, typhoid, human enteritis, legionellosis, melioidosis, stomach ulcer and cancer [39]. A study by Abraham [40] in São Paulo assessed the bacteriological parameters in the Tiete River that passes through the inner part of the city. Water sample analyses revealed the presence of bacterial pathogens like E. coli, Shigella flexneri, and Shigella boydii. A similar study was done by Hamner et al. [41] in the River Ganga, Varanasi, India, and revealed high concentrations and diversity of Enterococcus spp. The study also suggested that the river was responsible for cholera disease in families that were dependent on its waters for bathing or washing clothes.

Industrial and Domestic Discharge

The quality of freshwater bodies is degrading daily globally due to increasing population and industrialisation. The disposal of industrial effluents in these water bodies is recognised as a source of waterborne pathogens that cause 80% of all diseases and deaths around the world [42]. When microorganisms and hazardous chemicals from industries and households come into surface freshwater bodies or seep into groundwater resources, it results in a rise in water pollution [43]. Gao et al. [44] used next-generation sequencing to examine the microbial diversity in a small urban river affected by domestic sewage. The findings revealed that microbial communities in rivers that consistently received domestic wastewater exhibited longitudinal and vertical assemblage patterns and included pathogenic species that could pose a serious public health risk. A similar study was conducted by Cui et al. [45] in Changzhou, a member of the Yangtze River delta. Their research offered an extensive understanding of the distribution of bacterial pathogens and the factors that influenced them in urban waterways affected by domestic sewage.

Apart from introducing bacteria into freshwater bodies, industrial waste, particularly nutrient-rich effluents from food-processing industries, also significantly impacts freshwater bodies by drastically increasing organic and nutrient loads, thereby creating conditions that favour the persistence and proliferation of microorganisms, including human pathogenic bacteria [46].

2.2.2. Non-Point Sources

Agricultural Runoff

Water quality may be impacted by bacterial concentrations in freshwater caused by agricultural runoff, especially from manured croplands [47]. Water bodies are contaminated more quickly by pathogens, especially from animal operations, during periods of high rainfall and flooding. Also, farming practices, including spraying chemical fertilisers, applying pesticides, and improperly disposing of human and animal waste, make agricultural runoff a significant source of waterborne infections [48]. Regardless of whether the area was manured or farmed, heavy runoff, particularly in rainy weather, could result in a decline in water quality [49]. Antibiotic-resistant E. coli, Salmonella, and Pseudomonas are among the AMR bacteria that have been found in water affected by agriculture, raising concerns about the ecological spread of antibiotic resistance [50]. Furthermore, the pathogens in these runoffs flourish in the presence of biological waste and fertilisers that are high in nutrients [51].

Stormwater and Overland Flow

Stormwater and overland flow often carry faecal indicator and disease-causing bacteria, which have been identified at varied levels in field studies. The most commonly detected bacteria are E. coli, faecal coliforms, and enterococci in storm water monitoring and are used to indicate faecal contamination in receiving waterways [52]. Hathaway et al. [53] conducted a study in Charlotte, North Carolina, to investigate water quality degradation caused by pathogen contamination, identifying stormwater runoff as a major contributor to the transmission of indicator bacteria from urbanised areas and to set best management practices (BMP). It was found that while some BMPs were effective in treating water containing indicator bacteria, others failed to perform well. Many stormwater BMPs provide damp, nutrient-rich conditions, which allow indicator bacteria to persist, thus making their way into receiving surface water bodies, including freshwaters.

Wildlife and Birds

Determining wildlife levels of contamination is one of the main challenges in microbial pollution evaluation. In rural watersheds, wildlife, including waterfowl, can significantly contribute to faecal contamination (Hubálek [54]. Birds and other wildlife carry numerous pathogenic bacteria, including Salmonella, Campylobacter, pathogenic E. coli, Enterococcus, and Clostridium [55,56,57]. These bacteria enter freshwater through faecal deposition by the wildlife, surface runoff containing their waste, and sediment resuspension when the animals enter freshwater bodies, raising faecal indicators and occasionally leading to high pathogen detections (Khalefa et al. [58]. It is extremely concerning that migratory birds may carry and spread some harmful bacteria. Mohapatra et al. [59] collected faecal samples of poultry birds with the help of cloacal swabs and isolated E. coli. This study helped to find the non-point source of faecal pollution and the development of essential management practices.

Urban Runoff and Mixed Watersheds

Urban runoff, which introduces a mixture of sewage, industrial effluents, rainwater, and pollutants from impermeable surfaces into aquatic systems, is also a major factor in freshwater microbiological contamination. Giardia, Norovirus, Vibrio, and intestinal bacteria like E. coli are among the pathogens frequently associated with urban runoff [60]. These pollutants represent serious risks to human health, especially in areas where irrigation, drinking, and bathing rely on contaminated water supplies [50]. Ibekwe et al. [61] conducted a study in the Santa Ana River watershed area, whose land use varies between urbanisation and agriculture. It was found that urban runoff water (7.94%) and agricultural runoff sediment (6.52%) had the greatest percentages of all potential pathogens. Proteobacteria, Bacteroidetes, and Firmicutes are the three main phyla from which the majority of potentially harmful bacterial sequences were found.

Figure 1 is a diagrammatic representation of the various sources of microbial pollutants into freshwater sources and the associated impact on human health.

3. Epidemiological Approaches Versus QMRA

Broadly defined, epidemiology studies how often diseases and health-related events occur in populations, who is affected, why they are affected, and how to use that information to control health problems, forming the bedrock of public health. It goes beyond individual medicine to look at patterns, risk factors, and interventions for entire communities, using descriptive, analytic, and experimental methods [62]. Although classical epidemiological methods are foundational for public health, they have several inherent limitations that can impact the validity, precision, and application of their findings. Major limitations include challenges in establishing causality, susceptibility to various biases, and logistical constraints [63]. Also, classical epidemiology primarily explains what has happened, meaning it is not suitable for predicting the potential of future outbreaks. However, QMRA estimates what could happen under defined exposure scenarios [64]. Nevertheless, together, classical epidemiology and QMRA provide a comprehensive, risk-informed basis for public health decision-making, including environmental health. Thus, to better understand these limitations and how they are addressed using QMRA, it is worthwhile comparing QMRA against individual classical epidemiological methods.

Descriptive epidemiology focuses on describing the distribution of disease according to person, place, and time using data from surveillance systems, routine health reports, hospital records, and surveys. It typically produces estimates of incidence, prevalence, and temporal trends, helping to identify disease hotspots and emerging patterns [65,66,67,68]. This approach is relatively simple to implement and useful for generating hypotheses about potential risk factors. However, it cannot establish causal relationships, and findings are often affected by underreporting, surveillance bias, and limited ability to attribute disease outcomes to specific environmental exposures. In contrast, QMRA extends beyond descriptive analysis by quantitatively linking environmental contamination, exposure pathways, and the probability of infection or health risk. This is important because it enables more precise estimation of public health risks from environmental pathogens and supports evidence-based decision-making for risk management, policy development, and targeted interventions.

Case–control studies examine the association between exposure and disease by comparing individuals with a disease (cases) to those without the disease (controls), typically using retrospective exposure histories [69,70,71]. The main output is an odds ratio, which estimates the likelihood that a particular exposure is linked to the disease outcome. This approach is particularly efficient for studying rare diseases and is commonly used in outbreak investigations because it can be conducted relatively quickly and with fewer resources. However, case–control studies are susceptible to recall and selection bias, and they often rely on self-reported or poorly quantified exposure data, which can limit the accuracy of exposure assessment. On the other hand, QMRA avoids recall bias by using environmental and experimental data to model exposure and can evaluate transmission pathways that may not be captured through self-reported information. This is important because it enables a more quantitative and mechanistic understanding of infection risks associated with environmental contamination.

Cohort studies assess the risk of disease following exposure by tracking groups of exposed and unexposed individuals over time using longitudinal exposure and health outcome data [72,73,74]. The main outputs are incidence rates and relative risk estimates, which help determine whether exposure increases the likelihood of disease, providing stronger causal inference than many observational approaches because it establishes temporal relationships between exposure and outcome. Despite their strength, cohort studies are often expensive, require long follow-up periods, and may be difficult to implement in low- and middle-income countries due to logistical and resource constraints. Contrarily, QMRA can estimate health risks using environmental contamination data and modelled exposure pathways without the need for long-term human follow-up, allowing faster and more cost-effective assessment of potential health risks, particularly in settings where long-term epidemiological studies are impractical.

Randomised controlled trials (RCTs) evaluate the effect of interventions by randomly assigning exposures or interventions to study participants and comparing outcomes between groups [75,76,77]. The main outputs include measures of risk reduction and intervention efficacy, making RCTs the gold standard for establishing causal relationships and informing policy decisions. Their controlled design minimises confounding and bias, providing strong evidence on the effectiveness of public health measures. However, RCTs are often costly, logistically complex, and may raise ethical concerns, particularly when withholding beneficial interventions. They are also difficult to apply in many environmental exposure contexts where randomisation is not feasible. Quantitative Microbial Risk Assessment becomes advantageous in this case because it can evaluate potential interventions and exposure scenarios through modelling without requiring experimental manipulation of populations. This enables risk evaluation and intervention planning in situations where controlled trials are ethically or practically impossible.

Ecological studies are broad and examine population-level relationships between exposures and health outcomes using aggregated regional or national data [78,79,80]. These studies typically produce correlations that suggest potential associations between environmental factors and disease patterns. Because they rely on existing datasets, ecological studies are relatively low-cost and useful for identifying large-scale trends or generating policy-relevant signals. The cons with ecological studies are that they are limited by the ecological fallacy, where associations observed at the population level may not hold true for individuals, and they are often affected by confounding variables. To address these, QMRA focuses on individual-level exposure pathways and microbial dose–response relationships, providing a more mechanistic and exposure-specific understanding of risk that cannot be inferred from aggregated population correlations alone.

In many cases, disease occurrence is not spatiotemporally static. Thus, time-series and spatial epidemiology investigate how disease patterns change over time and across geographic areas by analysing longitudinal health data alongside environmental or climatic variables [78,79,80]. These approaches typically produce outputs such as trends, clusters, and statistical associations that help identify emerging disease hotspots and potential environmental drivers. They are valuable for early warning systems and for understanding how environmental factors such as climate variability influence disease dynamics. The drawback with these approaches is that these analyses are largely correlational and often rely on indirect measures of exposure, which can limit their ability to establish causal mechanisms. Contrasting this approach, QMRA explicitly models exposure pathways and microbial dose–response relationships, which link environmental contamination to measurable infection risk, enabling more precise evaluation of health impacts.

Molecular methods have gained considerable ground in microbial studies due to the many limitations associated with culture-based approaches. Therefore, molecular epidemiology investigates pathogen transmission pathways using genomic and molecular tools such as whole-genome sequencing and molecular typing [81,82,83]. The primary outputs include identification of pathogen lineages and transmission links, which allow high-resolution tracking of outbreaks and antimicrobial resistance dissemination. This approach is particularly powerful for understanding how pathogens spread between hosts, environments, and regions, despite being resource-intensive and its inability to directly quantify the probability of infection or exposure risk. To eliminate these challenges, QMRA focuses on estimating the likelihood of infection based on exposure levels and dose–response relationships. Combining molecular insights with quantitative risk estimates can better inform public health interventions and environmental risk management.

Syndromic surveillance systems monitor symptom-based data, such as clinical reports, pharmacy sales, or emergency department visits, to detect early signals of disease outbreaks [84,85,86]. The main outputs are alerts or signals that may indicate emerging public health threats. These systems are particularly valuable in settings with limited laboratory capacity because they enable rapid detection of unusual disease patterns. However, this epidemiological approach often lacks specificity, as symptoms may be caused by multiple pathogens, and it does not identify the specific causative agent. In comparison, QMRA focuses on specific pathogens and quantifies the risk associated with exposure to these specific pathogens, providing pathogen-specific risk estimates that can support targeted risk mitigation strategies.

Transmission modelling examines how pathogens spread within populations by using epidemiological parameters to simulate epidemic dynamics [87,88]. The outputs often include metrics such as the basic reproduction number (R₀), epidemic curves, and projections of intervention impacts. These models are valuable for predicting outbreaks and testing potential control strategies under different scenarios, although they rely heavily on assumptions and parameter estimates that may be uncertain, and they often provide limited detail on exposure pathways. Compared to this approach, QMRA focuses on modelling the exposure process from environmental contamination to infection. This is important because it provides detailed insight into how specific exposure routes contribute to disease risk, complementing population-level transmission models.

Burden of disease modelling estimates the population-level health impact of diseases by integrating multiple epidemiological data sources [89,90,91]. The main outputs include metrics such as disability-adjusted life years (DALYs), years of life lost (YLLs), and years lived with disability (YLDs), which enable comparison of health impacts across diseases and risk factors. These estimates are highly relevant for policy and resource allocation. However, while burden of disease models often rely on numerous assumptions and may be limited by sparse or incomplete data, particularly in low- and middle-income countries. QMRA estimates infection risks associated with specific environmental exposures, which is critical for providing exposure-based risk estimates that help inform and refine broader burden of disease assessments.

Participatory epidemiology involves gathering community-reported disease patterns and local knowledge to identify health risks and disease trends [92]. Data sources often include interviews, community reporting systems, and participatory mapping, producing qualitative or semi-quantitative insights into disease dynamics. This approach is particularly useful in settings where formal surveillance systems are weak and can improve community engagement and local disease reporting. Yet, difficulties in standardising data and potential reporting biases remain significant challenges with this method. Quantitative Microbial Risk Assessment overcomes these challenges by relying on structured environmental and microbiological data to model exposure and risk, and provides standardised, quantitative estimates of infection risk that can complement community-based observations and guide evidence-based public health interventions.

A summary of each of the classical epidemiological approaches, including their main characteristics, strengths, and limitations, is provided in

Table 2. Summary of Classical Epidemiological Approaches, their descriptions, strengths and limitations.

Approach	Primary Focus	Typical Data Used	Main Output	Strengths	Limitations	References
Descriptive epidemiology	Distribution of disease by person, place, time	Surveillance, routine reporting, surveys	Incidence, prevalence, trends	Simple; identifies hotspots and trends; hypothesis-generating	No causal inference; underreporting is common; weak exposure attribution	[65,66,67,68]
Case–control studies	Association between exposure and disease	Retrospective exposure histories	Odds ratios	Efficient for rare diseases; outbreak investigations	Recall and selection bias; poor exposure quantification	[69,70,71]
Cohort studies	Risk of disease following exposure	Longitudinal exposure and outcome data	Relative risk, incidence	Stronger causal inference; temporal clarity	Expensive; long follow-up; often impractical in LMICs	[72,73,74]
Randomised controlled trials (RCTs)	Effect of interventions	Experimentally assigned exposures/interventions	Risk reduction, efficacy	Gold standard for causality; policy-relevant	Ethical, costly, and logistically difficult; limited environmental applicability	[75,76,77]
Ecological studies	Population-level exposure–outcome patterns	Aggregated regional or national data	Correlations	Low cost; useful for policy-level signals	Ecological fallacy; confounding	[78,79,80]
Time-series & spatial epidemiology	Temporal and spatial disease dynamics	Longitudinal health and environmental data	Trends, clusters, associations	Early warning; integrates climate/environment	Correlational; exposure often indirect	[78,79,80]
Molecular epidemiology	Pathogen or AMR transmission pathways	Genomics, WGS, molecular typing	Lineages, transmission links	High resolution; strong for outbreak tracing	Resource-intensive; no direct risk quantification	[81,82,83]
Syndromic surveillance	Early outbreak detection	Symptom reports, pharmacy data	Alerts, signals	Rapid; useful where labs are limited	Low specificity; no agent identification	[84,85,86]
Transmission modelling	Pathogen spread dynamics	Epidemiological parameters	R0, epidemic curves	Predictive; intervention testing	Parameter uncertainty; weak exposure detail	[87,88]
Burden of disease modelling	Population health impact	Multi-source epidemiological data	DALYs, YLLs, YLDs	Policy-relevant comparisons	Heavy assumptions; sparse LMIC data	[89,90,91]
Participatory epidemiology	Community-reported disease patterns	Local knowledge, reports	Qualitative/quantitative trends	Context-sensitive; improves surveillance	Standardisation and bias issues	[92]

.

4. Methodological Approaches in QMRA

The QMRA framework generally includes four essential components, which include identifying the microbial hazards, evaluating human exposure through various activities, establishing dose–response relationships, and estimating the infection risk [93].

4.1. Hazard Identification

The first step in freshwater QMRA involves recognising and describing microbial hazards that may be present in natural or treated water systems. Various pathogens are commonly identified as significant hazards in freshwater sources, especially those impacted by human activities such as recreation or where untreated sewage may enter the water systems [94]. Thus, the purpose of hazard identification is to identify the potential hazards to humans caused by certain pathogens, such as bacteria, protozoa, or viruses [95]. These include bacteria (e.g., E. coli, Salmonella spp., V. cholerae), protozoa (e.g., C. parvum, G. lamblia), and viruses (e.g., enteroviruses, noroviruses, adenoviruses). Such organisms are widely documented as causative agents of waterborne infections across diverse freshwater environments [96]. Since not all human enteric pathogens can be included in QMRA, reference pathogens are typically chosen based on how significant they are to the exposure pathways, offering a conventional paradigm for risk assessment. Accordingly, it is presumed that other significant pathogens within each class would also be under control if the reference pathogen is controlled. Local conditions, such as identifying exposure pathways, source water properties, prevalence, and severity of waterborne disease, should be considered while selecting the reference pathogen [97].

4.2. Exposure Assessment

Exposure to contaminated freshwater poses significant health risks. Exposure assessment in QMRA for freshwater recreation, for example, involves estimating the intensity, frequency and duration of exposure to pathogens by humans, which could be via inhalation, ingestion of water, or by dermal exposure, and evaluating the potential health risks associated with it [98]. The exposure assessment includes identifying the volume of water ingested per event and the concentration of pathogens present in the water.

For a daily exposure, for example, the daily dose (D) will be defined as the total number of microorganisms ingested or contacted per day, and this is calculated by multiplying the volume of the medium consumed or contacted (I_v), usually in litres or mL, by the concentration of the pathogen in that medium (M), usually in microbial count (or colony forming units [CFU] per litre or mL) [99]. Thus, the daily dose of the pathogen will be calculated according to the following equation:

D = I_v × M

(1)

For example, in a study by Ngo et al. [94] on freshwater recreation in Ontario, Canada, the exposure assessment considered the volume of water ingested by recreators as a critical factor in determining the risk of illness. Another study in Bangladesh by Islam and Islam in 2020 [100] evaluated the risk of illness from pathogens such as E. coli O157:H7 and Cryptosporidium during river bathing and highlighted a high risk of illness due to the ingestion of water contaminated with untreated sewage.

4.3. Dose–Response

The dose–response assessment describes the connection between the quantity of pathogens consumed (dose) and the probability of an unfavourable outcome, such as an infection, a disease, or death [101]. The most common dose–response models based on infectivity include the exponential model and the beta-Poisson model [102,103].

4.3.1. Exponential Model

This model is used when limited experimental data exist and when infection events are independent. In this model, the probability of infection is measured using the formula:

P_inf = 1 − e^−rd,

(2)

where P_inf = probability of infection (per exposure), r = infection probability per organism (pathogen-specific infectivity rate) and d = number of pathogens ingested (dose).

To determine the link between the dosage and the infection risks, Chen et al. [104] employed the exponential dose–response model for S. aureus bioaerosol. Another study by Ranjdoost and Owrang [105] used the exponential model for Legionella pneumophila. Ngo et al. [94] also used this model for Salmonella during freshwater recreation in Ontario, Canada. The risk of infection associated with pathogenic E. faecium (ingestion exposure), K. pneumoniae (ingestion exposure), and P. aeruginosa (ingestion exposure) in the environmental water samples for different exposure scenarios has also been evaluated using the exponential dose–response model [106].

4.3.2. Beta-Poisson Dose Response Model

The Beta-Poisson model is used in QMRA for pathogens where individual variability in susceptibility or infectivity is significant, especially for bacteria and viruses, to estimate infection risk from low doses in food, water, and wastewater, by better fitting experimental data than simpler models like the exponential; it is particularly useful when extrapolating to low-dose exposures is needed [107,108]. It is favoured for its plausible representation of microbial infection processes, capturing the ‘single-hit’ mechanism (although argued otherwise by some researchers), where each organism has a chance to infect, and is widely applied due to good performance with real-world data, despite its mathematical complexity [103].

The Beta-Poisson model calculates the risk of infection using the simplified formula:

P_inf = 1 − (1 + D/β)^−α

(3)

where P_inf is the probability of infection from a single exposure, D is the ingestion dose, and α and β are shape factors.

Different dose–response models have been used by different researchers for various organisms (

Table 3. Dose–response models for key pathogenic bacteria.

Organism	Dose–Response Model	Values	Reference
Salmonella spp.	Exponential model	r = 0.00752	[94,109]
Salmonella spp.	Beta-Poisson model	α = 0.21 β = 49.78	[101,110,111]
Shigella dysenteriae	Beta-Poisson model	α = 0.265 β = 1480	[101,110,112]
V. cholerae	Beta-Poisson model	α = 0.250 β = 243	[93,101,110]
E. coli	Exponential model	r = 9.7 × 10−9	[106,113]
E. coli E. coli O157:H7	Beta-Poisson model Beta-Poisson model	α = 0.395 β = 2.473 α = 0.0571, β = 2.2183	[110,114] [115,116]
C. jejuni	Beta-Poisson model	α = 0.145 β = 7.59	[117]
P. aeruginosa Enterococcus faecium	Exponential model Exponential model	r = 1.87 × 10−8 r = 2.19 × 10−11	[106,118,119] [106,119,120]

).

4.4. Risk Characterisation

The final step of the QMRA is risk characterisation, where the likelihood of an infection (represented as likely numbers of infections per 10,000 people per year) is estimated for each target organism and the related exposure scenarios, using the equation:

P_{inf (x)} = 1 − [1 − P_inf/day]ⁿ

(4)

where P_inf is the probability of infection from n exposure events per time x (e.g., x could be yearly or every six months).

The probability of illness is then calculated using the equation:

P_ill = P_{inf (x)} × P_ill/inf

(5)

where P_ill/inf is the probability of illness per infection.

This method has been used in several QMRA studies to estimate public health hazards associated with exposure to waterborne pathogens such as E. coli O157:H7 [115,116], Cryptosporidium spp. [121], norovirus [122], and rotavirus [123] in recreational and occupational settings (

Table 4. Some studies that used QMRA in Freshwater Systems to estimate the probability of infection globally.

Country	Type of Water Sample	Pathogen Studied	Model Used to Estimate the Probability of Infection	Results	Reference
South Africa	River water	E. coli and enterococci	Beta-Poisson dose response model	Pathogenic E. coli and enterococci were detected at high concentrations, indicating substantial faecal contamination across sites. QMRA showed a significantly higher infection risk from upstream water use and repeated exposures, particularly for workers and recreational users.	[126]
Bangladesh	River water	E. coli O157:H7, Cryptosporidium spp., norovirus and rotavirus	Exponential and Beta-Poisson models (pathogen-specific)	The QMRA revealed unacceptably high illness risks from accidental ingestion of sewage-impacted river water, exceeding USEPA guideline levels. Children faced higher infection probabilities (9–19%) than adults (7–16%), emphasising the need for improved risk management interventions.	[100]
Pakistan	Groundwater and surface water	E. coli, Salmonella spp., Shigella spp., V. cholerae	Beta-Poisson dose response model	The QMRA showed widespread microbial contamination of school drinking water, with around half the samples positive for E. coli (49%), Shigella (63%), Salmonella (53%), and Vibrio cholerae (49%), leading to high predicted risks of illness in school children. Southern Sindh, particularly Karachi, had the highest annual illness probabilities, with school children facing about 70% risk from Campylobacter and 22.6% from Rotavirus, indicating an urgent need for school water safety and management interventions.	[101]
South Africa	River water	E. coli, Cryptosporidium	Exponential model (approximating Beta-Poisson model) with Beta parameters.	QMRA showed high Cryptosporidium and pathogenic E. coli infection risks in the uMsunduzi River exceeding South African/WHO guidelines for drinking, swimming, and canoeing. SWAT identified pollution hotspots from wastewater, mines, and farming for targeted management.	[127]
USA	Creek water and river water	Cryptosporidium, Giardia, Norovirus, E. coli O157:H7, and Salmonella	Exact Beta-Poisson model (for E. coli and Norovirus), and Exponential model (for Salmonella, Giardia, and Cryptosporidium)	QMRA showed recreation during Combined Sewer Overflow (CSO) events (<24 h) increased Acute Gastrointestinal Illness (AGI) risk by 39–75% compared to non-CSO conditions (>24 h), with high risks persisting for some activities. CSO-impacted sites may contribute 1–8% of Philadelphia’s salmonellosis, cryptosporidiosis, and giardiasis cases, supporting targeted CSO reduction strategies.	[128]
Nigeria	Drinking water (boreholes, spring water source, sachet water)	E. coli, Salmonella, Giardia	Comprehensive susceptibility dose–response model	The modified QMRA model predicted high daily infection risk (mean 0.236 ± 0.056) and diarrhoea risk (mean 0.039 ± 0.016) from contaminated borehole water in Nigeria. Predicted diarrhoea rates correlated strongly (r = 0.74) with observed prevalence, validating the model against existing QMRA approaches for developing country contexts.	[129]
Germany	Ornamental water fountains	E. coli, Enterococci, and Salmonella, P. aeruginosa	Exponential model (P. aeruginosa) and Beta-poisson (E. coli, Enterococci, and Salmonella)	High microbial contamination occurred in ornamental fountains (E. coli: 1.6 × 101–6.1 × 102 MPN/100mL; Enterococci: 1.2 × 10–1.2 × 103 MPN/100mL; Salmonella: 8.6 × 103–3.1 × 105 CFU/100mL). QMRA revealed that children’s gastrointestinal and dermal infection risks exceeded USEPA benchmarks (36 illnesses/1000) at multiple fountains, particularly for Enterococci, Salmonella, and P. aeruginosa.	[130]
Canada	Creek water and river water	Giardia, Cryptosporidium, Campylobacter, E. coli O157:H7, Norovirus, Salmonella	Probabilistic model using pathogen-specific dose–response (mostly Beta-Poisson)	QMRA estimated AGI risk from freshwater recreation at 0.8–36.7 cases per 1000 swimmers (5th–95th: 0–226.3), aligning with Lake Ontario studies. Upper estimates exceeded Health Canada guidelines (<20 cases/1000), demonstrating QMRA’s utility for risk assessment without large epidemiological data.	[94]
USA	Lake water	Rotavirus, norovirus, adenovirus, Cryptosporidium spp., Giardia lamblia, Campylobacter jejuni, E. coli O157:H7, and Salmonella enterica	Exponential model, Beta-poisson and Hypergeometric	QMRA predicted 49 AGI cases per 1000 swimmers at East Fork Lake, with human enteric viruses as primary contributors, exceeding U.S. EPA Recreational Water Quality Criteria. Illnesses occurred even at low E. coli levels.	[131]
Iran	Tap water	Legionella pneumophila	Exponential model	Legionella counts were highest in warm water during summer, with positive correlation to pH and negative correlations to chlorine and temperature. QMRA showed the annual infection risk exceeded the WHO/USEPA acceptable limits, indicating inadequate hospital water management practices and a need for enhanced seasonal control strategies.	[105]
India	Treated wastewater samples from 11 Effluent Treatment Plants (ETPs)	E. coli, Salmonella spp., Cryptosporidium spp. and Giardia spp.	Exponential model (Cryptosporidium spp. and Giardia spp.) and Beta-Poisson (E. coli, Salmonella spp.)	QMRA revealed high infection risks from treated wastewater reuse in Delhi, with pathogenic E. coli (100%), Salmonella (63%), and Cryptosporidium (81%) persisting post-chlorination for irrigation, toilet flushing, and industrial uses. Adults faced 1.24 × higher annual infection probability than children.	[132]

). For example, in river bathing scenarios, the risk of illnesses ranged from 7% to 19% depending on the pathogen and population type, indicating notable cumulative annual risks that exceeded permissible criteria set by the United Nations Environmental Protection Agency (USEPA) [100]. According to Ashbolt et al. [124] and Carducci et al. [125], risk characterisations in wastewater, recreational beaches, and occupational exposures employed dose–response models to estimate per-exposure infection probabilities, which were then scaled by exposure frequency to determine annual infection and illness probabilities.

Although QMRA has proven to be a valuable tool in public health, the approach is not without challenges. Quantitative Microbial Risk Assessment involves several sources of uncertainty that can influence the accuracy and reliability of risk estimates. These uncertainties arise from several factors, including limitations in data, model assumptions, and variability in environmental and human factors. These uncertainties, their sources and possible ways of reducing them have been summarised in

Table 5. Types of Uncertainties in QMRA, their sources and ways to address them.

Type of Uncertainty in QMRA	Main Sources	Approaches to Address/Reduce Uncertainty
Pathogen occurrence and concentration	Limited spatial and temporal environmental sampling; variability due to rainfall, temperature, and sewage discharges; detection limits and recovery inefficiencies in analytical methods [133]	Increase monitoring frequency and spatial coverage; use standardised sampling protocols; apply improved detection and quantification methods; incorporate probabilistic distributions in models
Exposure assessment	Uncertainty in ingestion rates, exposure frequency and duration, and human behavioural patterns; reliance on assumptions or data from other regions [134].	Conduct local behavioural studies; use observational or survey data; apply sensitivity analysis and probabilistic exposure modelling
Dose–response relationships	Models derived from limited laboratory or clinical studies; differences in pathogen strains and host susceptibility [135,136,137].	Use updated or pathogen-specific dose–response models; incorporate uncertainty ranges in parameters; compare multiple models where possible
Pathogen viability and infectivity	Detection of genetic material rather than viable organisms; environmental factors affecting pathogen survival (e.g., UV radiation, salinity, temperature) [138,139]	Use viability-based detection methods (e.g., culture-based or viability PCR); incorporate pathogen decay models and environmental survival data
Model structure and parameter uncertainty	Simplified assumptions in modelling environmental transport, decay, and exposure pathways; limited or uncertain parameter values [140,141]	Apply sensitivity and uncertainty analysis; refine models with empirical data; validate models using field or epidemiological data
Variability vs. uncertainty	Difficulty distinguishing natural variability in environmental conditions and human behaviour from a lack of knowledge about parameters [142,143]	Use probabilistic modelling (e.g., Monte Carlo simulation) to separate variability from uncertainty; collect additional data to reduce knowledge gaps
Data scarcity (especially in LMICs)	Limited environmental monitoring, pathogen surveillance, and exposure data; reliance on surrogate data from other regions [144]	Strengthen surveillance and environmental monitoring systems; conduct local studies; use adaptive modelling approaches with region-specific data where possible
Scenario and assumption uncertainty	Assumptions regarding exposure scenarios, treatment efficiencies, and environmental conditions [145,146]	Develop realistic exposure scenarios; test multiple scenarios; perform scenario and sensitivity analyses to evaluate their influence on risk estimates

below.

5. Public Health Benefits of Using QMRA for Freshwater Ecosystems

Quantitative microbial risk assessment is rapidly accumulating recognition as the most practical method for assessing the risks associated with microbial contamination of water [129]. QMRA has been increasingly applied to natural and managed freshwater systems to inform public health decision-making. QMRA, coupled with risk management of water quality, is an effective tool for scientific analysis of drinking water systems and has been incorporated in water safety plans. It can estimate how safe the water is, how much the safety varies and how certain the estimate of safety is. This can be used in the Water Safety Plan (WSP) system assessment to determine whether treatment is meeting health-based targets with the required level of certainty [147]. It is used to evaluate the scenarios for pathogen contamination using analysis, detection methods, decision-making and surveillance to protect public health from infectious disease [10,148]. Quantitative microbiological risk assessment can be used by regulatory agencies and drinking water authorities to quantify the health risks from microorganisms for raw water sources, thus informing treatment options.

5.1. Translating Environmental Contamination into Quantitative Health Risk

QMRA enables the conversion of microbial water quality measurements into health-relevant risk metrics, such as probability of infection or disability-adjusted life years (DALYs). This provides a more meaningful interpretation of surface water quality data than traditional indicator counts alone. By incorporating environmental measurements and dose–response models, QMRA estimates the likelihood of adverse health outcomes under specific exposure scenarios, bridging the gap between environmental detection and public health impact [149].

5.2. Supporting Preventive and Risk-Based Water Safety Management

Unlike traditional approaches that respond to disease incidence after outbreak occurrence, QMRA offers a predictive, preventive tool for managing waterborne disease risks. By estimating risks under current and future conditions, QMRA can inform proactive interventions, such as water treatment upgrades, source protection strategies, and behavioural recommendations for water users [150]. The WHO explicitly promotes QMRA as part of water safety planning to support risk-based management across drinking water, recreation, and reuse contexts [150].

5.3. Informing Standards and Regulatory Targets

QMRA provides a scientific basis for deriving health-based targets and risk thresholds, which can inform regulatory standards for freshwater quality. Risk estimates from QMRA allow comparison with benchmark tolerable risks (e.g., 1 infection per 10,000 exposures per year), facilitating evidence-based criteria development for recreation and potable uses. For example, QMRA studies on freshwater recreational exposure have been used to benchmark gastrointestinal illness risks against guideline targets, illustrating the utility of QMRA in translating microbial data into policy-relevant outcomes [94].

5.4. Prioritising Public Health Interventions

By quantifying risk attributable to different pathogens and exposure pathways, QMRA helps prioritise intervention strategies for freshwater ecosystems. Whether evaluating the impact of sediment disturbance on pathogen release or comparing risks under seasonal variability, QMRA highlights the exposure scenarios with the greatest health implications. In river systems where resuspended sediments elevate pathogen loads, QMRA quantified substantially higher infection probabilities, signalling the need for targeted risk mitigation [110].

5.5. Enabling Resource-Efficient Decision Making

In settings where large epidemiological studies are impractical due to resource constraints, QMRA provides an economical alternative for estimating disease risk from environmental exposures. It utilises existing environmental data and dose–response relationships to estimate risk without requiring extensive clinical surveillance, making it particularly valuable in low-resource or data-scarce contexts [151].

5.6. Enhancing Public Health Communication

By expressing risk in quantitative terms (e.g., annual probability of infection), QMRA aids risk communication with stakeholders, policymakers, and affected communities. Communicating risk in these terms supports transparent, evidence-based discussion of water safety issues and fosters informed decision-making about freshwater use and interventions [124,152,153].

Table 6. Different application areas of QMRA with their purpose.

SN	Application Area	Purpose	Pathogens Considered	Exposure Pathway	Population at Risk	References
1	Drinking water security	Evaluating microbial pathogen risks	E. coli, Salmonella spp., Shigella spp., Campylobacter, Giardia lamblia and C. parvum	Ingestion	General population, children	[129]
2	Emerging pathogens and antimicrobial resistance	Risks from antibiotic-resistant pathogens	Antibiotic-resistant E. coli, Klebsiella pneumoniae	Ingestion rate in swimming events	Community users	[154]
3	Aquaculture	Identified direct use of animal manure as a major contributor to the faecal contamination of pond water, as well as the skin of cultured fish	E. coli	Ingestion, contact	Aquaculture workers, consumers	[155]
4.	Wastewater effluent management	Helps in evaluating bacterial contamination that affects wastewater treatment plant (WWTP) workers and communities after exposure to waterborne pathogenic bacteria in a WWTP and its associated surface water.	E. coli and Enterococci	Ingestion, occupational	WWTP workers and communities	[126]
5.	Recreational Waters	used to estimate illness risks from recreational activities and to evaluate indicator performance to assign source-attributed risk for management responses	Giardia, Cryptosporidium, Campylobacter, E. coli O157:H7, norovirus, and Salmonella	primary water-contact activities (swimming and wading) and secondary contact activity (fishing)	Recreators	[94]

provides some examples where QMRA has been applied to natural and managed freshwater systems to inform public health decision-making.

6. Conclusions and Future Directions

Quantitative Microbial Risk Assessment (QMRA) has emerged as a critical tool for linking microbial contamination in freshwater systems to public health outcomes. Its principal strength lies in its ability to translate environmental microbiological data into quantitative estimates of infection or illness risk, thereby complementing traditional epidemiological surveillance and water quality monitoring. In freshwater ecosystems, characterised by multiple uses, dynamic hydrology, and complex contamination sources, QMRA provides a predictive, preventive framework that supports risk-based decision-making rather than reliance on reactive outbreak detection.

However, despite its growing application, several challenges constrain the widespread use of QMRA in freshwater systems, particularly in low- and middle-income countries (LMICs). These include limited pathogen monitoring data, high uncertainty in exposure and dose–response parameters, simplified representations of complex freshwater dynamics, and weak integration with public health surveillance systems. Furthermore, the technical and interdisciplinary expertise required to implement QMRA remains unevenly distributed, limiting local ownership and routine application.

Looking forward, the value of QMRA will depend on advances in environmental monitoring, improved dose–response models for a broader range of pathogens, and tighter integration with hydrological, epidemiological, and One Health frameworks. The incorporation of high-frequency environmental data, climate and land-use information, and molecular surveillance will enhance the realism and relevance of QMRA outputs. Equally important is the alignment of QMRA with policy and regulatory processes, ensuring that risk estimates are translated into actionable public health interventions. Capacity building in LMICs will be essential to embed QMRA within national water safety, sanitation, and AMR action plans.

In this context, QMRA is not merely a modelling exercise but a strategic public health instrument, one that can help prioritise interventions, optimise resource allocation, and strengthen preventive management of freshwater-related health risks under conditions of uncertainty and limited surveillance.