The Safety Assessment Paradigm

A Method for Setting Standards for Pesticides, Food Additives, and Chemicals in the Environment

The development of the Safety Assessment Paradigm took place at the same time as the promulgation of new regulations for pesticides in 1954 and food additives in 1958.  Its basic form was described in an obscure journal by two USFDA authors (Lehman and Fitzhugh, 1954). It was also adopted at about the same time by World Health Organization/Food and Agriculture Organization Joint Expert Committee on Food Additives (JECFA; FAO/WHO, 1958).  Both of the new regulations required pesticide and food additive producers to get approval from the USFDA before the chemical could be used.  In addition, if the chemical were to be approved at all, it was necessary to determine how much could be used.  The acceptable daily intake (ADI), as it came to be called, is a dose expressed on a body weight basis (typically mg/kg). The ADI was then calculated using two additional numbers:  (1) the highest dose from a laboratory animal study with a No Observable Effect Level (NOEL) and (2) a safety factor of 100.  That is: 

ADI = NOEL/100

The safety assessment paradigm has evolved since its inception.  The Benchmark Dose was developed as a replacement for the NOEL (Crump, 1984).  The original safety factor of 100 has been subdivided and supplemented, with new rationales for various applications.  The methodology has been transferred to other regulatory agencies, and the terminology has been changed.   Yet some essential characteristics remain the same:

• Level Setting.  The primary and nearly sole purpose of Safety Assessments is to identify levels of exposure that will be considered to be safe or acceptable by a regulatory agency.
• Technocracy.   The SA Paradigm generally relies on the opinion of experts who are qualified by virtue of their scientific training or experience.
• Arbitrary numbers.  Even though they have a basis that is partly scientific, safety assessments generally contain one or more traditional arbitrary numbers or factors.  For example, as originally conceived, the ADI calculation used statistical significance (i.e p<0.05) to define a NOEL, and a safety factor or 100.   Although these factors often have some underlying scientific basis, they are also codified and justified by regulatory policy.
• Precautionary.  Instead of trying to characterize risk, Safety Assessment seek to establish a level of exposure where the risk is negligible.  Safety Assessments do not estimate risk.

The widespread adoption of the safety assessment paradigm is attributable to the fact that it is a relatively simple process and is transparent in the sense that there is formal and visible decision process.  When the pesticide regulation program was transferred from the FDA to the EPA, the Safety Assessment Paradigm went too, and other regulatory programs in the EPA began to use it too.  Because the law was patterned after food additive and pesticide regulation, it was an especially natural fit for implementing the 1976 Toxic Substances Control Act that required the chemical industry to get approval from the EPA before new chemicals could be released into the environment.  But other programs that did not involve premarket began to use it too, many of which did not involve premarket approval.  That didn’t work as well.  It turned out that different programs were developing different ADIs for the same chemical.  So, in the mid 80’s, a committee was developed to standardize the assessment across agency programs, at which point the ADI became the Reference Dose (Barnes and Dourson, 1988).  That didn’t work out especially well either. 

Safety Assessment and Contaminants

When safety assessment paradigm is used for purposes other than premarket approval, it is often not a good fit for several reasons:

• Practical Limitations:  Exposure to contaminants above the acceptable level may occur may occur at levels that cannot be avoided.  Therefore, a precautionary level may not be achievable.  In addition, essential nutrients may be required at intake levels that exceed an ADI or RfD calculated using standard methodology.  
• Lack of Information.  The Safety Assessment Paradigm was developed by toxicologists who were also regulators.  But as the use of SA Paradigm has spread, the safety assessor and the regulator are usually different people, who often reside in different offices, institutions, or geographic locations.  The safety assessor will often not know how the assessment will be used, and the regulator will get no other information regarding the toxicity than the ADI/RfD.  In particular, there will be no characterization of risk as a matter of degree.
• Excessive Reliance on Level Setting.  The dominance of the Safety Assessment paradigm in food safety and environmental regulation often prevents consideration of regulatory strategies that do not involve setting levels.

Even though they also adopted the Safety Assessment Paradigm for food additives, the WHO/FAO Joint Expert Committee of Food Additives (JECFA) often uses less standardized procedure when evaluating contaminants.  Even though they also call levels for food additives Acceptable Daily Intakes, JECFA levels for contaminants are called Tolerable Daily Intakes (TDIs) instead.  Early evaluations for major contaminants like arsenic simply identified a TDI without giving a rationale.  More recent JECFA evaluations for arsenic and lead declined to identify a TDI at all.

All that said, the Safety Assessment Paradigm is still of some use for evaluating contaminants found in food and the environment.  JECFA does use it for lesser contaminants.  In addition, it is useful for setting “screening levels”.  Like FDA guidance levels, the idea is that as long as concentrations are below a level that would permit approval for intentional use, then an unintentional contaminant should be OK as well.  The best known application of screening level strategy is the issuance of “Minimal Risk Levels” by the Agency for Toxic Substances and Disease Registry, which are used by the EPA for determining Superfund cleanup levels:

Following discussions with scientists within the Department of Health and Human Services (HHS) and the EPA, ATSDR chose to adopt a practice similar to that of the EPA's Reference Dose (RfD) and Reference Concentration (RfC) for deriving substance specific health guidance levels for non-neoplastic endpoints. An MRL is an estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse non-cancer health effects over a specified duration of exposure. These substance specific estimates, which are intended to serve as screening levels, are used by ATSDR health assessors and other responders to identify contaminants and potential health effects that may be of concern at hazardous waste sites. It is important to note that MRLs are not intended to define clean up or action levels for ATSDR or other Agencies.

So, if the MRL is exceeded, it doesn’t mean that regulatory action is warranted.  But, it does mean that an issue has been identified that deserves more thought.

References

Agency for Toxic Substances and Disease Registry (2014).  Minimal Risk Levels (MRLs).

Barnes DG and Dourson ML (1988).
  

Reference Dose (RfD): Description and Use in Health Risk Assessments.  Regul Pharmacol Toxicol 8:471-486.  Also at http://www.epa.gov/IRIS/rfd.htm

Lehman AJ and Fitzhugh OG (1954).  100-Fold Margin of Safety.  Quarterly Bulletin of the Association of Food and Drug Officials 18:33-35.

World health Organization (1958). Procedures for the Testing of International Food Additives to Establish their Safety for Use; Second Report, FAO Nutrition Meetings     Report Series, No. 17

Official Post Soundtrack

Osborne, Joan (2000).  Safety in Numbers.  In: Righteous Love, track 3.

Post Notes

Thesis #12.  This part of the Regulatory Toxicology thread; it links up with the NOAEL and BMD discussion posted earlier.  No youtube for the soundtrack, bummer.