Microbial Risk Assessment

bacteria monitored in microbial risk management
Photo by qimono, c/o Pixabay

Microbial risk assessment (MRA) is a scientific method of analysis of the safety risks posed by pathogens in a complex food system. It determines the likelihood and severity of an adverse health effect in a population exposed to a certain pathogen and food combination.  Put another way, it helps us understand the hazard posed by pathogens and how extensive the risk of exposure is. Food recipes, processing and handling and the environment a microbe lives in are all part of the assessment. 

Probably the best interpretation to be found comes from  a chapter on the subject of microbial food safety risk assessment in a book ‘Foodborne Infections and Intoxications‘ (Lammerding, 2013) which states:-

“Microbial risk assessment (MRA) is a systematic approach to aid our understanding of complex food systems and to translate the potential presence of pathogens in the food production, processing, and preparation environments into statements of the likelihood and magnitude of a food safety risk defined in terms of adverse public health outcomes.”

A good starting point for conducting a microbiology risk assessment can be found in the monograph by the Codex Alimentarius Commission (CAC, 1999).

When Did It Start ?

In the past it was felt that microbiology food safety was too complex for formal risk assessment techniques to be applied. From 1994, risk assessment began to be treated much more seriously and in 1998 the first ‘formal’ assessment was made in the USA. One of the first risk assessments was issue of Salmonella enteriditis in eggs which began in 1994 and ended in 1998. It proved extremely complex to investigate but set the ground rules and illustrated to all the issues in conducting such a risk assessment.

What Makes Microbial Risk Assessment Difficult ?

A number of features make a microbial risk assessment difficult to conduct. Microbial growth populations show tremendous variations in their virulence in causing both disease and the susceptibility of their hosts. All microbial populations alter rapidly due to differing levels of growth and inactivation. Likewise there is a large variation in growth and survival characteristics of microorganism. Contamination of food is also intermittent and sporadic. Whilst there is plenty of data covering various adverse effects of pathogens, the dose-response relations are generally not available.

What Is A Microbial Risk Assessment ?

A quantitative microbial food safety risk assessment is a raft of modelling techniques allowing for a researcher to describe the relationships between the presence of a hazard in a food and the likelihood that the hazard will lead to an adverse public health effect consequence.

A general risk assessment is generally made up of four phases.

  1. Hazard Identification
  2. Exposure assessment
  3. Hazard Characterization which is a combination of a dose-response relation coupled to a severity assessment.
  4. Risk characterization

The general risk assessment covers three types:- chemical, physical and microbial. Whilst we are interested in the microbial hazard, the former two hazards should be regarded with equal priority.

Specific techniques for microbial assessments vary with the type of food system being examined. 

1. Hazard identification

The first step is the identification of biological, chemical and physical agents capable of causing adverse health effects which may be present in a specific food or group of foods. 

In a microbial risk assessment, one of the key features is identifying the microbial species or group of species involved. Plenty of data exists on what factors influences growth and survival, how the microbe is distributed, its type of transmission, symptoms of illness, whether it is foodborne etc. Whilst we know of only a limited number of outbreaks, it is the situation that a number of bacterial pathogens in food remain to be identified. 

2. Exposure Assessment

Exposure assessment is an investigation of pathogen concentration in food at the time of consumption. It covers whether it is present on or in a food, how it is distributed in that food and what its concentration is. The other element concerns the consumption of food. What is the frequency of consumption and how much is consumed.

The changes in populations of bacteria are affected by many complex interactions of factors. These include:-

  • ecology of the bacterial pathogen of concern
  • the processing, packaging and storage of the food
  • cultural factors relating to consumers
  • preparation steps which include cooking which can inactivate bacterial agents

3. Hazard Characterization

The qualitative and/or quantitative evaluation of the nature of the adverse health effects associated with the hazard. In a microbial risk assessment (MRA) all the concerns relate to the microorganism and the toxins they produce.

The characterization covers the pathogenesis of the microorganism. Hopefully, models exist which predict the growth of a microbe. In many cases there are dose-response models now available. In many instances Monte Carlo Simulation is used to generate values of investigation.

4. Risk Characterization

The final stage of a risk assessment.

It is the process of determining the qualitative and quantitative estimation of the probability of occurrence and severity of known or potential adverse health effects in a a given population. It also includes uncertainty. The characterization is based on the previous three factors. It involves bringing together the exposure and the dose-response assessments. This integrated approach makes it possible to mathematically estimate the overall probability of the effect on public health.  

Uncertainty analysis is a method used to estimate the uncertainty associated with model inputs, assumptions and structure or form. A variety of mathematical models are now available using stochastic approaches to help put context into this analysis. 

Conclusion

Having conducted a quantitative microbial risk assessment (QMRA), the next step is passing through the steps associated with risk management. Another key step is risk communication which is an interactive process of exchanging opinions and information amongst risk assessors, risk managers, stakeholders and the general public.

Case Studies 

In the USA a number of risk assessments have been conducted on key pathogens such as:-

  •  Clostridium botulinum,
  • the public health impact of Escherichia coli O156:H7 in home-cooked ground beef for hamburgers (Cassin et al., 1998; Marks et al., 1998) 
  • risk to public health of foodborne Listeria monocytogenes in various types of Ready-To-Eat foods.
  • Bacillus cereus presence in pastuerised milk (Notermans et al., 1997).
  • Campylobacter jejuni in chickens and pieces of chicken (Fazil et al., 1999)

Modeling Risks

The traditional approach to modeling risks is to generate steps in the model. This uses algebraic models based on standard probability theory. The method generates precise values which produces exact algebraic solutions to these equations. The algebra is extremely complex. The solution is only achieved through numerical integration or making approximations to the real solution for the assumptions made. It is also difficult to check for errors and in the process making a decision is weakened by the other difficulty of having confidence in the data.

A more sophisticated method is to employ software and hardware to generate Monte Carlo simulation of the problem. Classic Monte Carlo simulation products use Palisade’s @RISK® and Decisioneering’s Crystal Ball®.

Listeria monocytogenes is a foodborne pathogen that causes roughly 500 deaths every year. The Food and Drug Administration (FDA) and the Food Safety and Inspection Service (FSIS), working collaboratively, developed a quantitative microbial risk assessment for this microbe that compared the risk of listeriosis among twenty – three categories of ready-to-eat (RTE) foods. This risk assessment was completed in 203. It concluded that deli meats posed the greatest risk for listeriosis and accounted for about 1,600 illnesses every year. Following the stages of MRA:-

  1. Measured the prevelance and level of L. monocytogenes in deli meat and poultry in retail. What type of packaging was used ? – prepacked or retail-sliced ?
  2. The growth of the microbe was determined from retail purchase to consumption. They used an exponential growth rate and estimed the action of growth inhibitors.
  3. They assessed how consumers ate the deli meats by determining storage time and temperature, the serving size and other factors.
  4. A dose-response relationship was produced based on age of consumer. 

The output from the model was annual illness and deaths, the annual mortality by age and the risk of death per serving.

The data was used to estimate deaths and illness by slicing location and growth inhibitor use. They compared their figures with other food  groups. The sensitivity analyses considered consumer storage times and temperatures, the shelf-life and the total number of deaths.

A typical and more recent example using the @Risk software was an examination of the microbial quality of drinking water from water filtration dispenser toll machines (DFTMs) (Wibuloutai et al., 2019). In this example, a large number of samples of water were collected from the dispensers on a random basis. Key microorganisms associated with water quality such as E.coli and Staph. aureus were analysed to generate prevalence data. A risk calculation using the @Risk program based on drinking water consumption of 2L/day produced a probability of exposure. The program also generated values for the probability of illness due to either of these tow microbes and the risk of illness.  Other factors were considered such as the general sanitation status and drinking water quality which could be affected by other microorganisms because of poor hygiene in the dispensers and  inappropriate consumption of water.

Resources

ISO 22000:2005 Food safety management systems – requirements for any organization in the food chain.

World Health Organization have a web-site for microbiological risks in food.

JEMRA

Codex Alimentarius

References

Brown, M. & McClure, P. (2006) 6 – Microbiological risk assessment for emerging pathogens. In: Emerging Foodborne Pathogens. In: Emerging Foodborne Pathogens. Woodhead Publishing Series in Food Science, Technology and Nutrition. Pp. 130-152  https://doi.org/10.1533/9781845691394.1.130

CAC (Codex Alimentarius Commission) (1999) Principles and guidelines for the conduct of a microbiological risk assessment. FAO, Rome. CAC/GL-30.

Cassin, M.H., Lammerding, A.M., Todd, E.C.D., Ross, W., McColl, R.S., (1998) Quantitative risk assessment for Escherichia coli O157:H7 in ground beef hamburgers. Int. J. Food Microbiol. 41, pp.  21–44

Caswell, J.A. (2013) Chapter 4 – Development of Risk-based Food Safety Systems for Foodborne Infections and Intoxications. In: Foodborne Infections and Intoxications (4th Edt.) Academic Press pp. 53-64  https://doi.org/10.1016/B978-0-12-416041-5.00004-4

Fazil, A., Lowman, R., Stern, N., Lammerding, A.M., (1999) Quantitative risk assessment model for Campylobacter jejuni in chicken. Abstr. CF10, pp. 65, 10th Int. Workshop on CHRO. (Campylobacter, Helicobacter and Related Organisms) Baltimore, MD

Lammerding, A. (2013) Chapter 3 – Microbial Food Safety Risk Assessment. In: Foodborne Infections and Intoxications (4th Edt.) Academic Press pp. 37-51  https://doi.org/10.1016/B978-0-12-416041-5.00003-2

Marks, H.M., Coleman, M.E., Lin, J.C.T., Roberts, T., (1998) Topics in microbial risk assessment: dynamic flow tree process.  Risk Anal. 18, pp. 309–328

Notermans, S., Dufrenne, J., Teunis, P., Beumer, R., Giffel, M.T., Weem, P.P., (1997) A risk assessment study of Bacillus cereus present in pasteurized milk. Food Microbiol. 14, pp. 143–151

Wibuloutai, J., Thanomsangad, P., Benjawanit, K., Mahaweerawat, U. (2019) Microbial risk assessment of drinking water filtration dispenser toll machines (DFTMs) in Mahasarakham province of Thailand. Water Supplyhttps://doi.org/10.2166/ws.2019.016

Be the first to comment

Leave a Reply