Quantitative Risk Assessment

Food safety control in food chains

Food borne infections and intoxications remain relevant sources of illness, hospitalizations, and even deaths. These risks cannot be reduced to zero, but by intelligent interventions and good food safety management systems, one can reduce these risks. Over the years, food safety management systems have become more and more structured. One started with end product testing and using basic principles (Good Manufacturing Processes GMP). Later one introduced the Hazard Analysis Critical Control Point (HACCP) system, in order to organize food safety measures in a structured manner, focussing on built-in safety. Nowadays, quantitative risk assessment (QRA) is introduced in order to make the decisions based on more solid information. In QRA the risks of food borne hazards are determined. QRA consists of hazard identification (in which the relevant hazards are identified), hazard characterization (in which the probability and severity of the response is determined as function of the ingested dose), exposure assessment (in which the exposure is estimated to which the consumer is exposed) and risk characterization (in which the result of the former three steps are combined to estimate the risk).

Often extended risk analyses are meant for governmental activities (decision making, legislation, priorities). However, these are also of relevance for industry: they can enrich, justify, criticize parts of HACCP (CCPs, specifications), they can help to support decisions, they can help to evaluate quantitatively effects of various intervention scenarios, one can quantify risks and one can determine comparative risks (relative risk). Of course, industrial risk analyses will be less extended than governmental RA (sometimes of the order of 20 man-year for one product group). Both governmental and industrial risk assessments have the great advantage that they follow a structured transparent way of determining the risk. The reports of these analyses are important sources of information and the information is often already digested and structured: data concerning growth, inactivation, temperatures in distributions chains, pathogenicity, etc can be found in these reports.

From a public health perspective an Appropriate Levels of Protection (ALOP) can be derived that can be translated to a Food Safety Objectives (FSO) that can be used to set criteria for products and processes for various parts of the food chain (for example based on risk assessment). The implementation of these criteria can be done by applying the HACCP procedure.

Practical use of risk assessment

Although RA supplies large amounts of data, the difficulties with these are:

• accuracy/uncertainty: often data are not very accurate and large uncertainties remain. One could say that one should wait until one has accurate information, but the problem is that one will never have accurate information. Quantifying with the information available, although not totally accurate is better than to base one's decision only on qualitative information, or worse, wait until specific reliable information becomes available.

• interpretation: Since the whole analysis is often rather extended and can be mathematically complex, the analyses might be difficult to interpret, even for experts (a probability of a probability of an issue). Hopefully, in future, one pays more attention to a good information transfer to communicate the results of rather complicated analyses, so that risk managers can use the information.

In order to carry out these QRA, quantitative knowledge of the various phenomena occurring is necessary (quantitative microbiology), as are methods to monitor and verify the process (detection). Furthermore, the use of epidemiology and ecology are of eminent importance. It is with this information that one can determine the relative importance of specific organisms and specific food product groups.

After the analysis one can determine appropriate interventions, that can be based on the quantitative outcomes or the results of the detection methods, but one should also use specific targeted studies. For example studies concerning ecophysiology, specific effects of fermentation or processing effects. It is a challenge to find those specific interventions that can reduce the risk, without too large economic costs, and without effecting too much the taste and nutritional value of the products.

Aims of research

The aims of our research in this field are:

• The development of quantitative models for the various phenomena like growth, inactivation, and recontamination.
  
• The creation of tools using these models in combination with databases that provide quantitative answers within risk analyses.
  
• To develop the field of "analysis of risk analysis": how to carry out the risk assessment and how to present the results.