On 7-8 November 2012, more than 600 of the world’s leading food safety experts took part in a high-level scientific conference organised by EFSA to mark its tenth anniversary. Scientific and technological advances as well as the changing legislative framework are pushing the boundaries of current risk assessment practices and approaches. While keynote speakers set out these challenges to the conference on Wednesday morning, experts rolled up their sleeves during five parallel break-out sessions to tackle these issues: identifying and characterising risk, the efficacy of products used in food and feed, environmental impact, exposure through diet, and expressing risks.
Notes & quotes from the five sessions
New trends in identifying and characterising hazards
In relation to biological hazards, the creation of a centralised microbial whole genome sequence database based on the use of common experimental protocols is needed to better predict the outcome of pathogen-host interactions. Also the history of food preservation (e.g heat, acidic conditions, etc.) along with the food constituents need further attention. In chemical risk assessment, problem formulation should drive toxicity testing so that risk assessments are fit-for-purpose. Combined exposure to multiple hazards should also be addressed. The results of the risk assessment should be integrated into an impact assessment.
What are the boundaries? New methods exist for meeting the challenges but who should decide which of those should be used for risk assessment purposes, and when?
Prof. Colin Hill of University College Cork on the implications of host-food interactions for safety: “It is surprisingly difficult to predict the outcome when a contaminated meal is consumed. Some individuals may have no symptoms, some may become ill and others may even die”.
Professor Alan Boobis of Imperial College London: “Hundreds of chemicals have been found in humans in tests but that doesn’t mean they are of concern; the key is to understand which combinations might be of concern.”
Assessing efficacy in food and feed products
The speakers defined “efficacious” in relation to food and feed products as meaning “fit-for-purpose”. However, they also recognised that the “purpose” varies by area depending on the legislative framework. For health claims it refers to a causal relationship between consumption of a product by healthy people and its alleged beneficial effect. For feed additives, normally a conclusion can be reached on the potential for efficacy while for pesticides a risk/benefit analysis is performed for the target crop, much as in the area of human medicines. There was agreement that efficacy and safety assessment should remain separate but that the conclusions should be presented in an integrated way to risk managers.
What are the boundaries? There is not a clear definition of efficacy. Efficacy testing in relation to human medicine testing does not always take into account the interests of the end user.
Prof. Hildegard Przyrembel, formerly of Germany’s Federal Institute for Risk Assessment speaking in relation to the efficacy of health claims as opposed to their scientific substantiation: “You have to define beforehand who can benefit from the use of these products as well as who might be at risk from exposure to them.”
Having an eye for the environment
The Ecosystem Services concept was recognised as a common basis for environmental risk assessment (ERA) of pesticides, GMOs and plant pests. Ecosystem services were defined by Millennium Ecosystem Assessment in 2005 as the benefits people obtain from ecosystems, which include provisioning services (food and water), regulating services (pollination, regulation of floods, drought, land degradation, and disease), supporting services (soil formation and nutrient cycling) and cultural services (recreational, spiritual, religious, aesthetic values and other non-material benefits) . Harmonisation across EFSA’s Scientific Panels work in these areas would be beneficial. An interactive dialogue between the risk assessors and the risk managers is needed to agree on which are the specific protection goals. Important environmental monitoring is currently conducted by national authorities as well as by research projects and a common repository is needed to store and update the data collected.
What are the boundaries? Lack of data remains a critical boundary to make quantitative assessments.
Prof. Tony Hardy Chair of EFSA’s Scientific Committee, on how EFSA took broad protection goals set by EU legislation and crystallised these into specific and measurable targets for environmental risk assessment: “This harmonised approach could have applications beyond pesticides and could offer a way forward to focus risk assessment where current protection goals are not precisely defined”.
University of Coimba’s Prof. Paolo Sousa quoting a statement from the audience and concerning the widening of the dialogue between risk assessors, risk managers and stakeholders: “At the end of the day the risk managers in this area are the farmers.”
Making sense of dietary exposure
There was widespread recognition of how EFSA has broken former boundaries related to the lack of data on dietary exposure assessment at international level. Further work is needed to harmonise and strengthen the quality of data, for example by complementing data from monitoring with those from Total Diet Studies. Biomarkers of exposure can also be used to complement or validate exposure estimates. Screening of data can help to assess and explain uncertainties which help risk managers to manage their expectations about scientific advice. Communication support is essential in explaining complex concepts.
What are the boundaries? Total exposure is not currently taken into consideration. Risk managers need to weigh up the costs of using new techniques against the degree of certainty they require from risk assessors.
Catherine Leclercq of the Italian Agricultural Research Council on new boundaries in the area of exposure assessment: “It’s true that technological advances offer opportunities, but they also come at a cost. Greater resources and more sophisticated knowledge is required to make the most of them. The only real solution to achieve this in Europe is through even further strengthening cooperation and avoiding duplication of work.”
This key session focused mainly on chemical risk assessment with lessons applicable to the whole food safety area. The current “risk assessment paradigm” – hazard identification and characterisation followed by exposure assessment and risk characterisation – will remain, however various elements of the process are expected to evolve, mainly in relation to alternative testing methods, consideration of combined exposures and effects, uncertainty analysis, low dose effects, burden of disease estimates amongst several others. Risk assessments should be relevant and fit-for-purpose for risk management. A proper framing of the question with a consideration of life-cycle and socio-economic aspects and the introduction of step-wise approaches in exposure and hazard assessment will improve fitness-for-purpose. Risk assessment should then be communicated clearly with a consistent use of terminology and using analogies and metaphors in a careful way.
What are the boundaries? Expectations of safety vary and risk assessments should reflect that better. Problem formulation needs to improve through dialogue and consideration of the socio-economic aspects. Science needs to team up with communicators to express risk more effectively.
Dr. Josef Schlatter of EFSA’s Scientific Committee, referring to the risk assessment of substances that are both carcinogenic and genotoxic: “Before EFSA the final outcome was ALARA , which was of no help for risk managers. Now we take greater account of both potency and human exposure in evaluating the potential risk.”
Dr Angelika Tritscher of the World Health Organization commenting in relation to transparency: “Risk assessors have really increased efforts to make their scientific decision-process more transparent. But we are missing such transparency in the decision-making of risk managers – and it is often not clear how the output of the risk assessment was taken into account.”
 ALARA: as low as reasonably achievable