Risk assessment of dietary supplements

Risk assessment of dietary supplements shares many of the requirements of that for other chemicals, although there are some important differences. Amongst these is the essential nature of some nutrients so that it may be necessary to balance the need to minimize toxicological risk with the need to avoid deficiency. There may also be limitations on experimental design, in that high doses may not be achievable for nutritional reasons and available human data on toxicological hazard is likely to be very limited. Prior to embarking on a risk assessment the problem needs to be formulated. This involves risk assessors, risk managers and relevant stakeholders. A key decision is whether a risk assessment is necessary and, if so, what is required of the assessment. This will shape the nature and output of the assessment. Risk assessment itself is a scientific process comprising four steps, hazard identification, hazard characterization, exposure assessment and risk characterization. Hazard identification involves determining the range of toxicological effects that might be caused by the substance, whilst hazard characterization establishes dose-response relationships, toxicological and species relevance of the findings and establishes health based guidance values. Exposure assessment involves predicting or measuring the level, pattern and duration of intake of the substance by exposed individuals. This may require dietary consumption data. Finally, risk characterization is the process whereby all of the prior information is integrated to reach conclusions in a form appropriate to the question posed. The nature of the output can take several different forms, and may be qualitative or quantitative. There are some cross-cutting issues in risk assessment, primarily on uncertainty and variability. The sources of uncertainty at each step of the risk assessment should be clearly identified and quantified to the extent possible. Variability requires that the risk assessment should take into account all relevant subpopulations and groups, where there might be differences on the basis of intake or sensitivity. This would very often include different life stages, but may also include gender, ethnic or genetic differences. The report of the risk assessment should be systematic and transparent, identifying all key assumptions and defaults used.