Risk assessment of enteric viruses along the food chain and in the population

Abstract

Food‐borne microbial illness contributes up to one third of global disease burden. The largest category of food‐borne illness is gastroenteritis, the majority of which is caused by enteric viruses. Viruses like these are transmitted to food either by waste‐contaminated waters, or by handling and transfer during processing. An important tool for reducing or controlling food‐borne microbial risk is risk analysis. This framework has been adopted globally to manage risks associated with microbial contamination in food. Several hundred microbial risk assessments (MRAs) have been published by different national and international organisations, for different food‐hazard combinations. The use of MRAs in controlling and understanding virus risk has, to date, been limited, compared with the efforts made on bacterial pathogens. Given the large disease burden that viruses are responsible for, this disparity should be addressed. The main reasons for the relative lack of risk assessments are the difficulty in detecting and monitoring viruses compared with bacteria. This means less data on prevalence, concentration and inactivation, and allows viruses to remain silent contributors to global disease. There are also key conceptual differences between virus risk assessment and bacterial risk assessment. This project aimed to assess the current state of the art for food‐borne virus risk assessment, then to progress the field further by using the data available to produce risk rankings and risk assessments. This was done by a combination of literature reviewing and various risk assessment tools. The result was an assessment of the overall evidence base in the literature, a semi‐quantitative ranking comparison between the viruses and foods of most concern, and a survey of inactivation methods, leading to a quantitative ranking of the effectiveness of each in reducing and managing food‐borne virus risk.

1. Introduction

1.1. Background

Viruses are a major food‐borne hazard, with enteric viruses causing over half of the food‐borne disease burden (FAO and WHO, 2008; Hall et al., 2014; Havelaar et al., 2015). Unlike bacteria, viruses will not grow on food once contaminated and pose little risk of spoilage. Because the most common viruses can be infectious at low doses, it is important to ensure as much as possible that they are removed or inactivated before consumption (Bradshaw and Jaykus, 2016).

Food safety risk management should be based on solid scientific evidence and on transparent and reproducible processes and procedures that use available knowledge to arrive at public policy decisions. The importance of risk assessment lies not only in its ability to estimate public health risk, but also in its use as a framework for organising data and allocating responsibility for analysis. Risk assessment is a process that can include a variety of methods and models to reach its conclusions. The FAO and WHO distinguish risk assessment from risk management and communication but recognise that risk assessment and risk management have a number of significant interfaces. For example, establishing priorities and policies for risk assessment often includes input from risk management consideration (FAO and WHO, 1995).

1.2. Microbial risk assessment

Microbial risk assessment (MRA) focuses on biological hazards, mainly pathogenic bacteria. However, adverse health effects of other biological agents such as food‐borne viruses and parasites are increasingly considered for assessment. Among MRA methods, quantitative microbial risk assessment (QMRA) is a tool used in analysing and managing the risk of food‐borne infectious disease quantitatively. The same principles of QMRA apply for all pathogens, but there are important reasons to treat viruses as a separate category of hazard (FAO and WHO, 2008; CAC, 2012; EPA/USDA‐FSIS, 2012). Microbial risk assessment emerged as a separate discipline to chemical risk assessment by the mid‐1990s (Foegeding et al., 1994; Bradshaw and Jaykus, 2016; Nauta, 2021). Since 2008, the FAO and WHO have formally acknowledged viruses in food as a hazard category of importance (FAO and WHO, 2008). Food safety measures for bacteria are insufficient in dealing with virus risk, due to some key difference in persistence and infectivity (FAO and WHO, 2008; Bradshaw and Jaykus, 2016). This can lead to diverging standards for virus food safety criteria. Some of the earliest examples of published MRAs focused on viruses as a hazard (Haas, 1983; Gerba and Haas, 1988; Haas et al., 1993; Gerba et al., 1996). Since then, however, most published risk assessments have been for bacterial hazards. Virus hazards do not require a new paradigm of risk, just some differences in detail, similar to the separation between chemical and microbial hazards (Havelaar and Rutjes, 2008; Bradshaw and Jaykus, 2016). Rather, the problem is a lack of data on certain key question, and the difficulty in obtaining this data by present detection.

The hazard identification step for virus risk assessment is a qualitative description of a food product, a virus and the effect of both on a population of concern. Viruses are estimated to cause more than 50% of the world's food‐borne disease burden (Koopmans and Duizer, 2004; Havelaar et al., 2015). The most significant viruses, as identified by the FAO/WHO, are hepatitis A, norovirus and rotavirus, with emerging viruses like hepatitis E also of particular interest (FAO and WHO, 2008). The foods most associated with virus outbreaks are shellfish, fresh produce and ready‐to‐eat or preprepared foods (Duizer and Koopmans, 2008; Le Guyader and Atmar, 2008).

The exposure assessment stage for virus risk assessment has a specific set of concerns and challenges that are important to consider. Since most viruses of concern do not grow outside of a human or animal host, the main factors to consider will be transmission, detection, inactivation, survival or persistence, and consumption. Much data for exposure modelling falls into these categories (FAO and WHO, 2008; Bradshaw and Jaykus, 2016).

Transmission for food‐borne viruses occurs in stages. First, food comes into contact with a virus reservoir, then the virus attaches to the food matrix, and finally, the virus persists on the food matrix until consumption (Le Guyader and Atmar, 2008). The principal transmission routes identified by the original FAO/WHO committee are wastewater contamination, unhygienic handling by infected handlers, and animal or zoonotic sources (FAO and WHO, 2008). The main foods of concern for virus transmission are fresh produce, ready‐to‐eat food, shellfish, with pigmeat an emerging concern for zoonotic viruses like hepatitis E virus. For fresh produce, virus is adsorped onto the surface of fruits and vegetables and has also been observed getting taken directly into the plant tissue (Katzenelson and Mills, 1984; Chancellor et al., 2006).

The problem of virus risk assessment has been known for decades. However, only recently has the opportunity to meet this challenge begun. Better detection methods will lead to more information on prevalence, concentration and control. Given the contribution of viruses to the food‐borne disease burden, greater attention should be put on them in the years ahead. The output rate of virus risk assessment will continue to increase as greater data becomes available. The control of virus hazards will be improved by greater awareness of virus attachment and virus inactivation especially the variability possible. Microbiological control criteria are always needed, and with the next generation of tools and data, the future of QMRA can correctly address the role of enteric viruses as significant food‐borne hazards.

2. Description of work programme

2.1. Aims

The work programme covered the topic of risk assessment of food‐borne viruses, divided into three general categories.

First, the fellows were to gain experience and knowledge of two topics: risk assessment of food‐borne viruses and risk management of the same. The training was based on the EU‐FORA training modules, and on additional workshop activities, detailed in Table 1. It also included comprehensive literature reviews on the two topics, and networking activities with other fellows and risk assessment professionals.

Table 1.

Summary of courses, conferences and seminars attended by the fellows

Course/seminar	Title	Date
Training sessions provided by EFSA	Introduction training of the European Food Risk Assessment Fellowship Programme	30.8–17.9.2021
	Module 1 training of the European Food Risk Assessment Fellowship Programme – nutrition, animal welfare and health, animal health, regulated products, GMO	22–26.11.2021
	Module 2 training of the European Food Risk Assessment Fellowship Programme – risk communication and crisis response	21–25.3.2022
	Module 3 training of the European Food Risk Assessment Fellowship Programme – emerging risks, AOP, MoA, applications of OMICs in RA, risk ranking	6–10.6.2022
	Module 4 training of the European Food Risk Assessment Fellowship Programme – data collection and reporting	22–25.8.2022
Other training sessions	Training sessions at University of Cordoba – working with Prof. Antonio Valero, international expert on predictive microbiology	11–22.7.2022
Conferences and workshops	Workshop on predictive microbiology, held at University of Burgos	15–16.11.2021
	IAFP'S European Symposium on Food Safety, Munich, Germany Organisation of Symposium titled ‘New Hazards and Old Threats; Foodborne Viruses and Risk Assessment in Food Safety’ Presentation: The next frontier in risk assessment in food: quantitative viral risk assessment (KH) Presentation: Control of Foodborne Virus Risk in the Context of Risk Assessment (MT)	4–6.5.2022
	ONE – Health, Environment, Society –Conference 2022 Attended virtually	21–24.6.2022
	FoodMicro2022 Next Generation Challenges in Food Microbiology, Athens, Greece Oral presentation: Quantitative microbial risk assessment for foodborne viruses: past, present, and future (KH) Poster presentation: The control and management of foodborne virus hazards, from 2022 and beyond (MT)	28–31.8.2022
Other activities	National workshop for EU‐FORA fellows based in Spain, satellite meeting	23–25.2.2022

The tools and experience gained during training were to increase the capacity for the second aim: the research activities. Having established the current state of the art of food‐borne virus risk assessment, the fellows would use this knowledge to carry out original research, of benefit to the broader research community, relating to both risk assessment and risk control.

The third and final aim for the project was communication, relating to the earlier activities. The fellows were to organise presentations of their work at conferences and other venues, and wrote manuscripts intended for publication.

2.2. Activities/Methods

2.2.1. Training and capacity building

2.2.1.1. Training modules

Training in general risk assessment concepts and tools was done as part of the EU‐FORA training programme. In addition, a 2‐week workshop was arranged in collaboration with the University of Córdoba in Spain, which covered practical aspects of predictive microbiology and quantitative risk assessment in more depth. This included tutorials in the MicroHibro software, whose consortium is based in Córdoba. These activities were tailored towards the questions of viruses in food where possible.

2.2.1.2. Literature reviewing

Comprehensive literature reviews were prepared by both fellows in collaboration on the two main topics of the research program: (1) risk assessment of viruses in food, and (2) control and inactivation of food‐borne viruses.

2.2.1.3. Networking activities

As well as the EU‐FORA training activities, and the research visits to Córdoba, the fellows based in Spain were hosted for a 1‐week visit to AESAN, the Spanish Agency for Food and Nutrition in Madrid, Spain. The activities of AESAN as a regulatory agency were explained, and the relationship to the overall EU food safety framework. All training and networking activities are listed in Table 1.

2.2.2. Research activities

The research activities built on the earlier training and reviewing, extending current knowledge based on systematic reviewing of the literature. They were guided by the risk‐ranking workflow descriptions in EFSA opinions and guidance documents.

2.2.2.1. Risk assessment and risk ranking

The risk assessment contribution of the research was a risk ranking exercise for food‐borne viruses in the EU. Following guidance and examples from institutions like the FAO, WHO and EFSA, the fellows identified (1) the most significant food‐borne viruses present in the EU, (2) potential emerging viruses, as defined by EFSA guidance, and (3) the quantitative criteria most useful in ranking the impact of each virus. Systematic reviewing of the literature provided a map of available evidence for estimating each criteria, and the most significant data gaps. This exercise provides a survey for the state of the art for food‐borne virus risk ranking, and profiles of each virus that will be useful in guiding future laboratory work or expert knowledge elicitation efforts.

Where data was available, the most significant virus‐food combinations were ranked semi‐quantitatively. This was done using the risk ranking tools covered in earlier training, whose different approaches could be compared and contrasted. Sources of uncertainty were considered throughout the process. This ‘top‐down’ approach to risk ranking gives an overview of the most significant food‐virus combinations and the data available for risk assessment.

2.2.2.2. Risk management and inactivation modelling

After the significant food–virus combinations were identified, and the amount of information in the literature mapped, the next research contribution was a review of quantitative predictive models for inactivation of viruses during processing. The literature was systematically reviewed for data on inactivation and reduction by thermal and non‐thermal technologies. These models were presented in tabular format, with key parameters extracted for each inactivation method.

With this information, a ‘bottom‐up’ fully quantitative risk ranking exercise was carried out, comparing the impact of each intervention method on consumer health, and discussing the costs and benefits of each method. This collection of inactivation models could be incorporated in online software tools and databases like MicroHibro, Combase and RAKIP.

2.2.3. Communication and outreach

2.2.3.1. International conferences

The fellows organised a full symposium at the International Association for Food Protection in Munich, Germany, comprising three 30 min presentations on food‐borne viruses, food‐borne virus risk assessment and food‐borne virus control. This was done in collaboration with two other fellows from the same cohort. The fellows will also present results of their research activities at FoodMicro2022, in Athens, Greece, as poster and oral presentation (Table 1).

2.2.3.2. Other formal meetings

The fellows presented themselves and their work in other venues during the course of the fellowship, both internally at the host organisation, and during the visit to the Spanish Agency for Food and Nutrition (AESAN) (Table 1).

2.2.3.3. Publication outputs

The results of the initial literature reviewing during the training formed two book chapters in a forthcoming book on Food and Environmental Virology (Hunt and Trząskowska, 2022; Trząskowska and Hunt, 2022). Other manuscripts relating to the research activities are in preparation. For this reason, results of the research activities are not being presented in this summary.

3. Conclusions

3.1. Summary of work

Food‐borne virus risk assessment is an important topic in food safety, and is becoming more relevant as new detection methods are developed. Although most microbial risk assessments have focused on bacterial hazards, and on modelling growth, there are compelling reasons to consider viruses as a separate subcategory for risk assessment purposes. The prevention and control of viruses during production in particular should be carefully considered in the future, and new or emerging technologies assessed for their inactivation potential.

3.2. Future directions

The work done by the fellows can form the basis of a new and comprehensive risk‐ranking for food‐borne viruses in the EU. This has been carried out for parasitic hazards and bacterial pathogens in the past, and could be repeated for viruses using data from the literature and expert elicitation.

Software and databases designed for predictive microbiology can and should incorporate more data on virus inactivation, and consider the impact of the food matrix on this inactivation.

The map of available evidence assembled during the systematic review process indicates that more laboratory experiments are needed for modelling virus inactivation throughout the production process of uncooked foods.

Given the persistence of viruses in the environment, more data and models for transfer between surfaces and equipment would also be of use.

Finally, better detection methods for infectious viruses at low levels of concentration or prevalence should continue in development. Virus risk assessment is as critical and urgent as bacterial, and has been neglected to date due to the difficulties in detecting and quantifying. With these barriers eliminated, more and more virus risk assessments will continue to be published in the future.

Abbreviation

AOP

adverse outcome pathway

FAO

Food and Agriculture Organization

KH

Kevin Hunt

MoA

mode of action

MRA

microbial risk assessment

MT

Monika Trząskowska

QMRA

quantitative microbial risk assessment

WHO

World Health Organization

Suggested citation: Trząskowska M, Hunt K and Rodríguez‐Lázaro D, 2022. Risk assessment of enteric viruses along the food chain and in the population. EFSA Journal 2022;20(S2):e200918, 9 pp. 10.2903/j.efsa.2022.e200922