A scientific article published online August 14, 2013 highlights the absence of appropriate data for determining safety of direct and indirect food additives. The study entitled “Data gaps in toxicity testing of chemicals allowed in food in the United States” by scientists of The Pew Charitable Trusts and colleagues appeared in the peer-reviewed scientific journal Reproductive Toxicology (Neltner et al. 2013). Notably, of over 8,000 substances known to be present in food, merely 21.6 percent had suitable toxicological data needed to determine safe levels in food. In this context, the study authors defined “suitable” as animal data from feeding studies. Developmental or reproductive toxicity data was available for less than 7 percent of food additives in FDA’s own database.

The absence of actual in vivo toxicity data for 71 percent of direct and indirect food additives is partially due to a lack of legal requirements. For example, in the U.S., substances without toxicological data may be used in food contact if their estimated exposure levels are considered insignificant; for slightly higher exposure thresholds only in vitro genotoxicity data are required. The scientific justification of these exposure thresholds however continues to be questioned and debated, especially for endocrine disrupting chemicals. Another reason for the lack of toxicity data is the Generally Recognized As Safe (GRAS) notifications process. GRAS determinations may be based on expert opinion, rather than actual data.

The study’s authors compiled publicly available information on food additives from the U.S. Food and Drug Administration (FDA), the U.S. Environmental Protection Agency (EPA) (for pesticides also used as food additives) and industry (for GRAS self-determinations submitted to FDA). They then compared the chemicals in their database to toxicological information available from various sources, including FDA’s database and Accelrys “Toxicity” database. In particular, they were interested in feeding studies where test animals were orally dosed; both chronic and acute toxicity data were considered.

Finally, the scientists discuss why these data gaps exist, and how they can be filled. The U.S. FDA is aware of shortcomings in their food additives safety assessments; however, the agency lacks authority to amend this issue. According to the study’s authors, “a program is needed to effectively and efficiently fill the significant information gaps to ensure public health is protected”. The scientists further recommend that FDA aims at filling the data gaps by applying modern toxicological tools like computational (in silico) toxicology and high-throughput in vitro testing, like in their “Tox21” program (previously reported on by the FPF).


Neltner, T. et al (2013). “Data gaps in toxicity testing of chemicals allowed in food in the United States.” Reproductive Toxicology (published online August 14, 2013).