Proposed bills to reform TSCA have argued to put the burden on chemical manufacturers to prove that there is “reasonable certainty that the substances pose no harm.

But what does this mean, particularly in light of recent EPA positions on dioxin risk and endocrine disruptor tiered screening? [Covered in Feb 23, 2011,'s Toxics Regulation News]

In the dioxin IRIS draft, EPA appears to be using a standard that there is no safe level of exposure to dioxin based on linear modeling. EPA cancer risk assessment guidelines indicate its assessors should use the linear modeling when it is unknown how a chemical causes cancer, and when it causes cancer by mutagenic means. Otherwise, a non-linear, or threshold model can be used, which assumes that there is some level of exposure below which no health effects are anticipated to occur.  Therefore, if a chemical has any positive result in a cancer study (at any dose) and a manufacturer can't prove how, his chemical will be deemed to fail the proposed new TSCA safety standard.  A PMN will be denied and a ban of an existing chemical will be initiated.

Similarly, if the Endocrine Society has it's way, any effect at any dose seen in any of the eleven Tier I endocrine disruptor screening assays (including in vitro tests) would argue that there is no safe exposure because for EDCs there is no such thing as a threshold effect and even tiny changes can indicate significant effects later appearing in life.  "No harm" would be impossible to prove for chemicals with positive results in these assays - leading to bans.

I am not expert in the pesticide tolerance arena from where this language was lifted.  I've heard Richard Denison, EDF, say that "no" doesn't really mean zero risk.  (Not the least because that is about pesticides that are intended to harm pests or you can't claim them to be pesticides.)  Well, if "no" doesn't mean "no," the language should not be "no harm."  Actually, the Senate bill did offer a definition:

The term ‘reasonable certainty of no harm’ means, in establishing whether a chemical substance or mixture meets the safety standard under this subchapter, that aggregate exposure and cumulative exposure of the general population or of any vulnerable population to the chemical substance or mixture presents a negligible risk of any adverse effect on the general population or a vulnerable population.

So "no" means "negligible."  "Negligible" means "so small or unimportant or of so little consequence as to warrant little or no attention." [Merriam-Webster]  So are dioxin and EDC's risks "negligible"?  Not from all the blogging I've seen and the bans on BPA cropping up everywhere.

Today the TSCA standard is "presents or will present an unreasonable risk" - also not explicitly defined, but considerations under TSCA section 6 are laid out: effects and exposures, use the least burdensome but still "adequate" control measures,  the availability of substitutes and economic consequences.  Still doesn't get at how certain and how big the impacts of the chemical uses are and how to determine when we've reached acceptable risk and to whom.  That's what rulemaking is for: EPA publishes its facts and rationale for action, the public comments, EPA responds and maybe somebody sues.

If risk is always going to be unreasonable to someone, is there some place between no (as in zero), negligible and unreasonable? How about "safe enough"  to allow us/society to benefit from a chemical's use?  Oops - showed my bias.  I think we do need to consider both the costs and benefits in deciding whether a chemical use (emphasis on use and not the inherent properties of a chemical all by itself) is safe enough.  We do need to look at whether appropriate use of risk management (including safety practices, design and training) is happening so that it stays safe enough.  But just because there is a default linear model or a maybe-might-be effect based on a simplistic in vitro assay  shouldn't mean there is certainty of harm.  I've argued before that there is no such thing as life with no risks.

It's easy to say we should make manufacturers "prove" their products are "safe" - not so easy to really say what that means in practice.  It's time for stakeholders to quit talking in vague generalities or demanding the impossible and try to work out a rational program that supports beneficial chemical uses and manages risks in practical ways.

EPA says TSCA section 6 is too hard.  It's too hard to have to explain why the agency is doing what it is doing and too many people complain.  It's called democracy.


Leave a reply