I am disappointed in the Commission’s December 17, 2014 decision to proceed with rulemaking on phthalates.  We often say we are a data-driven agency, but this decision calls our commitment into question. 

The CHAP’s cumulative risk assessment is the cornerstone of their work.  Their recommendations for banning certain phthalates were all based on the cumulative risk assessment.  But those recommendations were based on old data, not the “likely exposures” required by section 108 of the Consumer Product Safety Improvement Act (CPSIA).

The CHAP used exposure estimates based on 2005-2006 and earlier data.  The Centers for Disease Control releases new exposure data about every two years.  The 2007-2008 data was released just a few months after the CHAP’s second meeting.  The 2009-2010 data was released in September 2012—well before the December 31, 2012 cutoff date, chosen by the CHAP, and more than 18 months before the CHAP’s report was delivered to the Commission.

What drives the cumulative risk assessment is the high exposures to DEHP, which is by far the most potent of the five phthalates included in the cumulative risk assessment. It is critical to note that DEHP was banned by Congress in toys and child care products in August of 2008. While exposures from toys were only a small fraction of the overall exposure to DEHP, particularly for adult women, the ban on DEHP in toys prompted a broader market reaction.  Manufacturers started to replace DEHP in other products, too.  So exposure to DEHP started to drop significantly.

The exposure estimates from the 2005-2006 data cannot possibly reflect post-CPSIA exposure levels of DEHP. Since DEHP was such a dominant factor in the cumulative risk assessment, that post-CPSIA data is critical and relevant information.   

Other scientists have done the same calculation that the CHAP did, but with the more recent exposure data.  They found that if the CHAP had used the 2009-2010 NHANES data—the most recent data available to them prior to the December 2012 deadline —it would have completely changed their results.  The so-called “hazard index” falls below the action level.  No need to ban more phthalates.  No need to continue the interim prohibition on DINP, much less expand it to a larger group of toys.

However, we do not even need to take their word for it.  Our staff has the expertise to do these calculations.  Our staff has access to all the data the other scientists had and probably better access to the CHAP’s methodology and models.   The staff should have conducted this analysis before deciding what actions to recommend to the Commission.  I hope the staff will make their revised analysis available to the public well before the end of the public comment period.

Has the CPSC ever before knowingly proposed a regulation that is based on data that is this far out of date?  I doubt it.  Usually, we update our most important data analyses every year.  We do an annual report for ATVs and for fireworks, for nursery products and for toys.  If we are proposing a mandatory standard, we usually do an additional analysis of deaths and injuries using the latest available data.  Can anyone imagine us proposing a new standard for ATVs or toys based solely on injury data that is 8 or 9 years old?

I know that Congress has imposed a tight time line for our rulemaking and doing the analysis of this more current data prior to rulemaking would impede our ability to meet that mark. We should always respect Congressional deadlines as much as we possibly can. In this instance, however, Congress also said that the CHAP should consider “likely exposures” and the “most recent, best available” scientific data.  The CHAP failed to do that, and so it is up to us to examine the impact of using the more recent exposure data. 

My concern over the data used by the CHAP is merely the threshold issue for me. I have many additional concerns as well.  I am disappointed that the CHAP did not invite public comment on their draft report but only a narrow, private, peer review.  Not only did this decision deny stakeholders an opportunity to engage on the issues, it slowed the report process down significantly.  While there is no explicit CPSIA requirement for a public peer review, it is a common and highly appropriate process for such a highly influential technical and scientific document to undergo a public peer review.  CHAP’s independence notwithstanding, an open peer review would have allowed public comments to be reviewed and analyzed prior to the rulemaking, giving our staff all of the relevant feedback prior to their own assessment of the CHAP’s report and recommendations.

Beyond this, I have questions about the proper legal standard for expanding the scope of an interim ban from mouthable toys to all children’s toys.

I am not convinced that all of animal studies reviewed by the CHAP are applicable to humans.

I question the idea of banning chemicals based on a cumulative risk assessment to which they contribute little or no risk. 

I am troubled by the idea of banning the use of chemicals in toys when the exposures from toys are dwarfed by the exposures in food, cosmetics and other sources outside our jurisdiction. 

And I am greatly concerned that we are considering banning chemicals that have been in use for many years, and whose risks have been studied for a long time, when we know very little about the alternatives that may be used instead.

The bottom line is this: there is a lot we do not know; a lot that perhaps could have been answered by a public peer review and/or data analysis prior to the issuance of an NPR. We must analyze more current data and we must review input from all stakeholders. When we do, the staff’s recommendations may change.  Maybe the staff will see different reasons why we still must ban some of these phthalates.  Maybe their recommendation will be reversed. I do not know what the outcome will be, but what I do know is that if we want the right answer, we must use the right data.