QUANTITATIVE MICROBIAL RISK ASSESSMENT IN THE ...

Journal of Environmental Research And Development

J. Environ. Res. Develop. Vol. 8 No. 1, July-September 2013

Review Paper (NS-1)

QUANTITATIVE MICROBIAL RISK ASSESSMENT IN THE MANAGEMENT OF Escherichia coli STRAINS VIA DRINKING WATER Jessen George, Divya L. and Suriyanarayanan S.* Department of Water and Health, JSS University, SS Nagar, Mysore (INDIA) Received June 05, 2013

Accepted September 03, 2013

ABSTRACT Quantitative Microbial Risk Assessment (QMRA) is a process, analogous to chemical risk assessment of estimating human health risks due to exposures to microbial pathogens. Now -a days Escherichia coli strains have emerged as water borne pathogens. It has been involved many water borne outbreaks across the world. The predominant outbreaks caused by the strain Escherichia coli O157:H7.It belongs to the member of the Enterohemorrhagic Escherichia coli. In order to manage Escherichia coli strains and to devise approaches to reduce the public health risk posed, many national and international groups have applied QMRA techniques to model the risk posed by Escherichia coli O157:H7 on drinking water. While multiple sources and routes of transmission of this pathogen are now recognized, contaminated drinking water serves as one of the important vehicle of the pathogen and continue to be linked to outbreaks across both developed and developing countries. This paper reviews these quantitative microbial risk assessments and their application in managing the risk posed by Escherichia coli O157: H7 on drinking water.

Key Words : QMRA, Water borne pathogens, Enterohemorrhagic Escherichia coli O157:H7, Water, Human health risks

INTRODUCTION

  

Microbial Risk Assessment (MRA) is a valuable tool in the management of microbial water safety issues and can provide a systematic approach for the regulatory authorities to control the risk posed by a pathogen. Risk assessment is a scientific process in which the hazards are iden-tified and the risk posed by the particular path-ogen is calculated.1 QMRA has emerged as a tool for deriving estimates of the risk of exposure to pathogenic organisms through drinking water.2, 3

Exposure assessment, Dose-response assessment and Risk characterization.

DISCUSSION Overview of QMRA Quantitative Microbial Risk Assessment (QMRA) is a process, analogous to chemical risk assessment of estimating human health risks due to exposures to microbial pathogens. This technique developed by Haas and coworkers.5 By Using QMRA can easily answer questions such as how many people’s in a society will be infected if part of drinking water treatment fails and how much public health will be affected if disinfection will eliminated.3,6 Escherichia coli strains in human clinical aspects Escherichia coli strain has emerged as waterborne pathogens.7, 8 They involved many waterborne outbreaks in developed and developing countries across the world.9

AIMS AND OBJECTIVES This paper provides a brief review about the Quantitative Microbial Risk Assessment (QMRA) models for pathogenic organisms in drinking water with special reference to gastroenteritis caused by Escherichia coli strains. The principles of risk assessment consist of four stages. Each of the stages is summarized below (Fig. 1).4  Hazard identification, *Author for correspondence

60



Step-1 Problem Formulation and Hazard identification Water source Pathogen concentration(µ) Step-2

Exposure Assessment Distribution Pathogen inactivation(πŏ) Ingress Phatogen concentration(µingress) Treatment Pathogen removal(π)

(V ingress)

Consumption Volume of water consumed (V)

Step-3 Dose-Response

Step-4 Risk characterization

Fig. 1 : General frame work for calculating microbial risk from drinking water4 The predominant outbreaks caused by the strain Escherichia coli O157:H7. It belongs to the member of the Enterohemorrhagic Escherichia coli (EHEC). Escherichia coli is a member of coliform group of bacteria that is naturally found in the intestine of human and warm blooded animals and they belongs to the family Enterobactericia. However certain strains of Escherichia coli are responsible for gastrointestinal illness along with other health problems.10-12 They are summarized in Table 1. The Major strain for gastroenteritis is produced by the species of Enterohemorrhagic Escherichia coli (EHEC) or Verocytotoxigenic Escherichia coli (VTEC). The commonest EHEC strain is Escherichia coli 057:H7.Escherichia coli 057:H7 is a member of the Enterohemorrhagic Escherichia coli (EHEC) and was identified as a pathogen in

1982. The illness was characterized by abdominal cramping, watery diarrhea followed by bloody diarrhea with or without fever.11, 12 The Infectivity of EHEC strains is substantially higher than that of the other strains and is an important cause of diarrhea in developing countries, especially in children. Escherichia coli in drinking water Waterborne transmission of pathogenic Escherichia coli has been well documented for both recreational waters and contaminated drinking-water.13,14 These pathogens are transmitted via drinking water are predominantly of fecal-oral route and therefore known as enteric pathogens.15 In all of these cases infection follows direct or indirect fecaloral spread from human or animal origin. Few selected outbreaks of Escherichia coli attributes to contaminated drinking water shown in Table 2. 61



Table 1 : Virulence type and serogroups associated with gastroenteritis in Escherichia coli species Diseases and symptoms

Strain

Serogroup

Enteropathogenic Escherichia coli (EPEC)

O55 H6, NM / O86 H34, NM / O111 H2, H12, NM / O119 H6, NM / 0125 a c H21 / O126 H27, NM / O128 H2, H12 O146 H6 06 H16/ 08 H9/ 011 H27 / 015 H11/ 020 NM 025 H42, NM /027 H7 / 0148 H28 / 0149 H10 0159 H20 / 0173 NM

Enterotoxigenic Escherichia coli (ETEC) Verocytotoxigenic Escherichia coli (VTEC) and Enterohemorrhagic Escherichia coli (EHEC). Enteroinvasive Escherichia coli (EIEC)

Gastroenteritis in infants Diarrhea, vomiting and fever, travelers’ diarrhea.

026 H11/ H32, NM / 055 H7 / 0111 ab Shigella-like dysentery H8, NM 0113 H21 / 0117 H14/ 0157 H7 (stools contain blood and mucus) 028ab NM/ 029 NM/ 0112ac NM /0124 Abdominal cramps, H30, NM/0136 NM /0143 NM /0144 diarrhea, vomiting and NM/ 0159 H2, NM/0164 NM /0167 fever H4,H5, NM

Sources : Rowe., 1983; Salyers and Whitt., 1994; Beutin et al., 1997; Willshaw et al., 1997Bell and Kyriakides., 1998

Table 2 : Summary of selected outbreaks of Escherichia coli attributed to contaminated drinking water Location and Country Oregon (USA) Mediterranean Cruise Missouri (USA) Grampian (Scotland) South Africa and Swaziland Wyoming (USA) New York (USA) Golan heights (Israel) Ontario (Canada) Indonesia Indonesia

Year 1975

E. coli strain

Number Number of cases of deaths

Reference

ETEC

2000

-

Rosenberg et al

ETEC

251

-

O’Mahony et al (1986)

E. coli 0157: H7

243

4

Swerdlow et al (1992)

E. coli 0157: H7

4

-

Dev et al (1991)

1992

E. coli 0157: H7

1000

-

Isaacson et al (1993)

June 1998

E. coli 0157: H7

157

-

Olsen et al (2002)

1999

E. coli 0157: H7

>1000

-

Hilborne et al (1999)

2000

ETEC

175

-

Huerta et al (2000)

2300

7

Payment et al (2000)

178

4

Bali health agency

1500

4

Jakartha Post

1986 Dec 1989 to Jan 1990 1990

May E. coli 0157: H7 2000 March E. coli 2008 April to E. Coli May 2009 62



and cause damage can lead to bleeding. This can happen not only in the intestine, but also to the small blood vessels of other organs, such as the kidney or brain. The incubation period for the infection with Escherichia coli O157:H7 is between 2 -12 days and most commonly around three or four days.22 Control Within a water safety plan, control measures that can be applied to manage potential risk from Enterohemorrhagic Escherichia coli include protection of raw water supplies from human and animal waste, adequate treatment facility and protection of water during distribution.13 QMRA model for Escherichia coli 0157:H7 Few number of Quantitative risk assessments have been developed for Escherichia coli O157:H7 on drinking water and some of these models are reviewed below. Using Quantitative risk analysis techniques, a risk assessment model can be determined for Escherichia coli 0157:H7 and this model can be used to assess and manage the risk posed by pathogen via drinking water.23-26 Hazard identification To describe a particular pathogen (Microbe) and the disease it cause, including their symptoms severity of the infection and the disease associated with each specific pathogen. This also includes the pathogenicity and the virulence of the microorganisms. The first step in the QMRA process is to define which pathogens will be modeled and what conditions will be investigated. Selecting the reference pathogen for risk assessment is the first important stage to consider.2, 26, 27 Studies done by Howard et al on drinking water system in Kampala, Uganda he selects Pathogenic Escherichia coli as one of the reference organism.2 Escherichia coli 157:H7 is a well known waterborne pathogen in developed countries and has been involved outbreak of disease such as that at Walkerton, Ontario.16 This pathogen is one of the selected reference pathogen for QMRA.14 Exposure assessment It is an attempt to determine, the frequency and magnitude of pathogens exposure through contaminated drinking water. Exposure assessment is a quantitative estimation of the

A well publicized waterborne outbreak of illness caused by Escherichia coli O157:H7 and Campylobacter jejuni occurred in Ontario, Canada. The outbreak took place in May 2000 and led to 7 deaths and more than 2300 illness. 16 Sources Humans are the major reservoir, particularly of EPEC, ETEC and EIEC strains. Animals are the major reservoir of Escherichia coli 0157:H7, with cattle and sheep are the principal reservoir of this pathogen.17 Other sources such as goats, pigs and chickens also reservoir of these EHEC strains. While multiple sources and routes of transmission of this pathogen are now recognized, contaminated drinking water serves as one of the important vehicle of the pathogen and continue to be linked to outbreaks across both developed and developing countries.9 Studies have shown that cattles are generally asymptomatic carriers of Escherichia coli O157 with illness only reported in young calves.18 Colonization of cattle was short term (1-2 months) and reported concentration of the pathogen in naturally infected cattle of up to 6.7×105 and 1.6×106 CFU g-1.19 Transmission Infection is associated with person-to-person transmission, contact with animals, food and consumption of contaminated water. The unprotected drinking water supply was contaminated by rainwater runoff during extreme climatic conditions that carry cattle excreta. Escherichia coli O157:H7 is spread to humans by ingesting material contaminated with fecal sources.11,14,20 The Studies done by Nataro and Kaper developed the infectious dose was considered to be very low and ingesting just few organisms, possibly between 10 and 100 can cause illness.10 The water have been implicated as vehicle for the transmission of these pathogens.21 Survival and infection After ingestion via contaminated drinking water the bacteria travel through the gut and attach to the surface of the large intestine. The bacteria produce toxins and that destroy the cells lining that colon and cause inflammation of the intestine. The toxin may be absorbed and travel in the blood stream to other parts of the body. The toxin can damage cells lining blood vessels 63



presence of contaminant (pathogen) during the done by Regli et al conducted on drinking water, consumption of water. This method is to a water consumption of 2L per person per day estimate the number of people exposed and the has been assumed.26 The studies done by categories of the people affected. Exposure can Roseberry and Burmaster reported the average be considered as a single dose of pathogens consumption of water was 0.96L per day.30 that a consumer ingest at a certain point in time Teunis et al obtained data on cold tap water or the total amount over several exposures(e.g. consumption in Netherland for use in QMRA over a year).Exposure is determined by the reported drinking water consumption 0.15L per concentration of pathogens in drinking water day.31 and the volume of water consumed. 14 Dose response The studies done by Howard and group showed The relationship between the dose received and that the reference pathogen selected for the resulting health effects. Dose assessment pathogenic Escherichia coli, the start point of the exposure model was the prevalence and aims at presenting a mathematical relationship concentration of Escherichia coli in cattle faeces. between the dose and the probability of the This pathogen particularly important in Uganda. infection or illness in exposed persons. By Few studies have shown that a significant different mathematical methods dose response 31,32 proportion of microbial contamination is derived models can be fitted to experimental data. from animal faeces.28 The reference pathogen Dose Response modeling is the key to selected for pathogenic Escherichia coli In Microbial Risk Assessment as it provides a link undertaking the QMRA, the risk from all between exposure dose and the probability of pathogenic Escherichia coli was calculated based infection. Prior to dose-response relationships, on the impact expected from Escherichia coli human feeding experiments were only used to estimate infectious doses such as ID50 or O157:H7.29 When assessing exposure to pathogens through Minimum Infective Dose (MID).Some selected drinking water, both the concentration of Dose-Response models of Quantitative pathogens and the volume of drinking water Microbial Risk Assessment (QMRA) are consumed are important parameters. The studies reviewed below in Table 3. Table 3 : Selected dose-response models of Quantitative Microbial Risk Assessment (QMRA) S/N Year Country

Reference pathogen

QMRA Model Dose Response Model

1

1986

Canada

E. coli 0157: H7

2

2000

USA

E. coli 0157: H7

3

2000

USA

E. coli 0157: H7

4

2004

USA

E. coli 0157: H7

5

2005

UK

E. coli 0157: H7

QMRA Title and author

Dose Response Model

Experimental infection of infant rabbits with verotoxin-producing Escherichia coli. Infection and Immunity (Pai et al., 1986) Development of a dose-response relationship for Escherichia coli O157:H7 (Haas et al.,2000)

Dose Response Model Dose Response Model

Dose-response envelope for Escherichia coli O157:H7 (Powel et al.,2000) Transmission and Infectious Dose of Escherichia coli O157:H7 in Swine (Cornick et al.,2004)

Dose Response Model

Dose response modeling of Escherichia coli O157 incorporating data from food borne and environmental outbreaks

64



Netherland of 0.2% for infection with Escherichia coli O157:H7.For the studies in Kampala water system used the mortality rate quoted by Kotloff et al. Since, it is more likely to reflect the higher mortality among children in developing countries. Risk characterization The final stage in the process estimates the adverse public health effect or risk of exposure to the hazard. In this stage, to integrate the information from the above three stages in to a single mathematical model to calculate risk. The probability of an outcome like infection, illness or death. This may be a prediction of illness per typical serving or calculated an annual risk of illness. The probability of infection is multiplied by 365 to calculate the probability of infection per year. 33 The studies done by Howrad et al in Kampala water system, Uganda showed that the assessment of E. coli is based on the data obtained for thermotolerant coliforms from monitoring programmes. In this assessment the susceptible fraction is based on known rates of access to piped water supply in Kampala, using a direct household connection rate of 20% of the population and estimated 70% of the unserved population using taps.2 One of the major issue facing the application of risk assessment models in developing countries is how to take assess exposure among the population that commonly use more than one type of drinking water source. This risk assessment is based on a several sets of data, including the status of Kampala water system, annual water quality data from two treatment work. The risk from water in the distribution system is an annualized risk based on average contamination as indicated by thermotoleranat coliform concentrations. It is assumed that 95% of these are Escherichia coli and that 8% of all Escherichia coli are pathogenic. The average from the surveillance data are therefore multiplied by 0.95 and then by 0.08.The figures in the risk assessment are expressed in organisms per litre and therefore this figure is multiplied by 10 in order to gain a final figure. The dose- response is based on shigella, which is estimated has a 1.0×10-3 risk of infection from exposure to a single

However there is only limited dose-response data for Escherichia coli 0157:H7.Hunter and group notes that the ID 50 (the number of organism required to infect 50% of people exposed) ranges between 102 and 106.Haas et al and Strachan et al have reported disease outbreak resulting from doses estimated to have been around 102 organisms. A number of dose-response models have been used in Quantitative Risk Assessment to describe the relationship between the level of microbial exposure (i.e., the dose or number of organisms ingested) and likelihood of occurrence of an adverse consequences (i.e., illness).33 The most commonly used models are single hit models. The simplest form of this exponential model was prescribed by Haas et al.4 The data used in generating dose-response models are derived from a variety of sources including human clinical trials, epidemiological studies based on waterborne outbreaks, animal clinical trials, in vitro studies using cell lines.31,34 However, due to the severity of the diseases human dose response studies cannot be carried out Surrogate models from Shigella fed to human and Escherichia coli O157:H7 to rabbits have representatives for human beings.32 The dose-response data for E. coli O157:H7 was obtained from a study done by Pai and coworkers.35 These workers studied the pathogenesis of diarrheal diseases due to Escherichia coli O157:H7 in New Zealand white infant rabbits. The studies done by Haas et al used a dose-response model for Escherichia coli O157:H7 from rabbit inoculated through an oral catheter. Doseresponse data of Escherichia coli O157:H7 from the results of clinical trials have been reported in the literature of Powel et al.36 Teunis et al have recently published a dose response for Escherichia coli 0157:H7 based on a food-borne outbreak in Japan.31 In undertaking the QMRA, the risk from all pathogenic Escherichia coli was calculated based on the impact expected from Escherichia coli 0157:H7.The studies done by Howard and group adopted Havelaar and Melse model as dose response studies. It was a used dose response estimate for Shigella to provide a generic assessment of risk from waterborne bacteria. In contrast, Havelaar and Melse have estimated in mortality rate in the 65


organism.32,33 The risk of developing illness once infected is more problematic as there is limited consolidated data available. From the studies of Howard et al used the treatment effect was calculated based on two assumptions.2 First of all, the raw water at one treatment plant (Gaba 1) had one result of 16cfu/100ml or 160 cfu/litre. As there were no organisms in the final water, indicates at the log reduction must have been at least between 102 and 10. The assessment data suggest a final risk as a disease burden of 1.45E-05 for Gaba 1 and 4.34E-06 for Gaba 2, with a likelihood of Gastroenteritis of 1.46E-04 and 4.38E-05 respectively. The risk assessment for pathogenic Escherichia coli WHO (2004) suggests that a reference level of risk 10E-06 is tolerable.

2.

3.

4.

CONCLUSION 5.

The Quantitative Microbial Risk Assessment (QMRA) appears to be feasible in developing countries. One such study done by Howard et al in Kampala water supply system. WHO have suggested that a reasonable reference level of risk from pathogen in water is 10-6 risk of infection. For Escherichia coli O157:H7, there appears to be relatively little risk from water leaving the treatment works. The assessment preformed by Howard and group in Kampala watersystem suggests that the risk exceeds the WHO reference level. The water quality problems are more associated with poor distribution systems than failures in treatment works in developing countries and have led to outbreaks in developed countries. This review studies concludes that QMRA can be used to provide objective quantitative input for the water safety plans and it can contribute to efficient and effective management of drinking water safety. This paper reviews the developments in QMRA and considers its potential for use in determining the robustness of water company risk assessments, as a tool to support decision making, to evaluate the effectiveness of different control strategies and to assess the impact on water quality from extreme events or changing conditions.

6.

7.

8.

9.

10.

11.

REFERENCES 1. Geraldine Duffy, Enda Cummins, Padraig Nally, Stephen O’Brien and Francis Butler, 66


A review of Quantitative Microbial Risk Assessment (QMRA) in the management of Escherichia coli 0157:H7 on beef, Meat Sci., 74(1),76-88, (2006). Guy Howard, Steve Pedley and Sarah Tibatemwa, Quantitative Microbial Risk Assessment (QMRA) to estimate health risks attributable to water supply : Can the technique be applied in developing countries with limited data, J. Wat. heal., 04(1), 4965, (2006). Uk.vir.org-Report., Review of Quantitative Microbial Risk Assessment (QMRA) to drinking water, Drinking wat. Qual.-Heal., (2011). Jeffrey A. Soller and Joseph N. S. Eisenberg, An evaluation of parsimony for Microbial Risk Assessment models, Environ. 19(1), 61-78, (2008). Krantzberg G. and Tanik A., Book Recommendation: Advances in water quality control, J. Wat. Res. Prot., 2(12), 995-996, (2010). Susan Petterson, Ryan Signor, Nicholas Ashbolt and David Roser, QMRA methodology, Report Micro Risk., 123, (2006). Sharma S., Sachdeva P. and Virdi J. S., Emerging waterborne pathogens, Appl. Microbiol. Boitechnol., 61(1), 424-428, (2003). Health Canada Guidelines for Canadian Drinking Water Quality: Guideline Technical Document, Bacterial Waterborne Pathogens, Current and Emerging Organisms of Concern., Heal. Canada; Ottawa, ON, Canada, (2006). Nicholas John Ashbolt., Microbial contamination of drinking water and disease outcomes in developing regions, Toxicol., 198(1), 229–238, (2004). James B. Kaper, James P. Nataro and Harry L. T. Mobley., Pathogenic Escherichia coli, Nat. Rev. Microbiol., 2(1), 123-140, (2004). Alfredo Caprioli, Stefano Morabito, Hubert brugeri and Eric Oswald, Enterohaemorrhagic Escherichia coli : Emerging issues on virulence and modes of transmission, Vet. Res., 36(1), 289–311, (2005).


12. Hugh Pennington, Escherichia coli, Lancet., 376(1),1428-35, (2010). 13. WHO (World Health Organization) Guidelines for Drinking-water Quality, Incorporating 1st and 2nd Addenda, Volume 1, Recommendations, 3rd Edn. 1-668, WHO; Geneva, Switzerland, (2008). 14. WHO Guidelines for drinking water Quality Volume 1 Geneva : World Health organization, 3rd Edn. ,1-540, (2004). 15. Health Canada, Escherichia coli in drinking water, www.hc-sc.gc.ca/ewh-semt/consult/_ 2011/ecoli/draft-ebauche-eng.php, (2011). 16. Hrudey S. E., Payment P., Gillham P. M. and Hrudey E J., A fatal waterborne disease epidemic in Walkerton, Ontario : Comparison with other waterborne outbreaks in the developed world, Wat. Sci. Technol., 47(3), 7-14, (2003). 17. Witold A. Ferens and Carolyn J. Hovde., Escherichia coli O157:H7: Animal reservoir and sources of human infection, Food Borne Pathog. Dis., 8(4) 465-487, (2011). 18. Pohlenz J. F., Winter K. R. and DeanNystrom E. A., Shiga-Toxigenic Escherichia coli-inoculated neonatal piglets develop kidney lesions that are comparable to those in humans with Hemolytic-Uremic Syndrome, Infe. Immun., 73(1), 612-616, (2005). 19. Matthews L., Low J.C., Gally D. L., Pearce., Mellor D. J. and Heesterbeek J. A. P., Heterogenic shedding of E. coli 0157:H7 in cattle and it’s implications for control, PNAS., 103 (3),547-552, (2006). 20. Joseph N. S. Eisenberg., M Alan Brookhart., Glenn Rice., Mary Brown and John M Colford Jr., Disease transmission models for public health decision making : Analysis of epidemic and endemic conditions caused by waterborne pathogens, Environ. Heal. Perspect., 110 (8), 1-2, (2002). 21. João P. S. Cabral., Water microbiology., bacterial pathogens and water, Int. J. Environ. Res. Pub. Heal., 7(1), 3657-3703, (2010). 22. Phillip I. Tarr., Carrie A Gordon and Wayne L. Chandler., Shiga- toxin- producing Escherichia coli and haemolytic uraemic syndrome, Lancet, 365(1), 1073–86, (2005).


23. Rory Coffey D., Enda Cummins., Martin Cormican., Vincent O Flaherty and Stephen Kelly, Microbial exposure assessment of waterborne pathogens, Hum. Ecol. Risk Assess., An Int. J., 13(6), 1-9, (2007). 24. Kenza jaidi., Benoit Barbeau., Annie Carriere., Raymond Desjardins and Michele Prevost., Including operational data in QMRA model: Development and impact of model inputs, J. Wat. Heal., 07(1), 77-95, (2009) 25. Ashbolt N. J., Petterson S R., Roser D R., Westrell T., Ottoson J and Stenström T A., Microbial Risk Assessment tool to aid in the selection of sustainable urban water systems. In 2nd IWA Leading-Edge on Sustainability in Water-Limited Environments. IWA Publ., London,1-9, (2006). 26. Jeffrey A. Soller., Use of Microbial Risk Assessment to inform the national estimate of acute gastrointestinal illness attributable to microbes in drinking water, J. Wat. Heal., 04(2), 165-186, (2006). 27. Gunther F. Craun., Rebecca L. and Calderon., Workshop summary : Estimating waterborne disease risks in the United States, J. Wat. Heal., 4(2), 117-128, (2006). 28. Pedley S., Yates M., Schijven J.F., West J., Howard G. and Barrett M., Pathogens: health relevance, transport and attenuation, World Health Organization, Protecting Groundwater for Health: Managing the Quality of Drinking-water Sources, 1-32, (2006). 29. Naik Jigna and Desai Pratibha, Detection of entero-haemorrhogic Escherichia coli (Ecoli 0157:H7) and it’s drug resistance pattern, J. Environ. Res. Develop., 7(1), 5155, (2012). 30. Cummins E., Kennedy R. and Cormican M., Quantitative Risk Assessment (QMRA) of Cryptosporidium in tap water in Ireland, Sci. The Total Environ., 408(4) 740–753, (2010). 31. Teunis P., Takumi K. and Shinagaw K., Dose response for infection by Escherichia coli 0157:H7 from outbreak data, Risk Anal., 24(2), 401-407, (2004). 32. Teunis P. F. M., Ogden D. and Strachan N. J. C., Hierarchical dose response of E. coli O157:H7 from human outbreaks 67



35. Signor R. S. And Ashbolt N. J., Comparing probabilistic microbial risk assessments for drinking water against daily rather than annualized infection probability targets, J. Wat. Heal., 07(4), 535-543, (2009). 36. James Withee, Michael Williams, Terry Disney., Wayne Schlosser., Nate Bauer and Eric Ebel, Streamlined analysis for evaluating the use of preharvest interventions intended to prevent, Escherichia coli O157:H7 Illness in Humans, Foodborne Pathog. Disea., 6(7), 1-2, (2009).

incorporating heterogeneity in exposure, Epidemiol. Infect., 136(1), 761–770, (2008). 33. Norval J. C. Strachan., Michael P Doyle, Fumiko Kasuga, Ovidiu Rotariu and Iain D Ogden, Dose response modeling of Escherichia coli O157 incorporating data from food borne and environmental outbreaks, Int. J. Food Microbiol., 103(1), 35–47, (2005). 34. Graham McBride, Tom Ross and Al Dufour, Comparative risk analysis, Animal Waste, Water Quality and Human Health, World Health Organization, 1-2, (2012)

68