As follow on to its Chemical Action Plan on Bisphenol A (BPA), EPA published an Advance Notice of Proposed Rulemaking July 26, 2011, to develop environmental effects and exposure testing. EPA is not addressing human health effects in this notice as there is ongoing testing work already with other agencies. (See a pointed Trevor Butterworth blog about that work here.)
Related to my earlier blog on GLP, it's nice to see the following acknowledgement regarding effects "data" that has been generated by academia:
It is difficult to interpret this information in a regulatory context, because the scientific methods employed in individual academic settings to test a hypothesis are not necessarily geared toward meeting or establishing generally applicable guidelines for evaluating ecotoxicity and setting corresponding regulatory limits or controls. In terms of environmental toxicity, EPA considers the currently available research as evidence that BPA has the potential to interact with the estrogen hormone system.
In order to be useful to an investigation of potential environmental risks posed by BPA, environmental testing must be representative and of known quality. To accomplish this, data should be collected using approved or recognized sampling, preparation, and analytical techniques. Appropriate quality assurance and quality controls also should be incorporated in the protocols for collection and analyses.
EPA is soliciting comment on "whether and on what basis they believe the current data are sufficient to determine whether BPA does or does not present an unreasonable risk of injury to the environment." This is the threshold for a TSCA section 6 control regulation, whereas a section 4 test regulation requires only a "may present" hurdle. So maybe the point is to skip any additional testing and go straight to a ban. Sorry, EPA, get some good, reliable data first.
Not that there aren't challenges to getting environmental monitoring data, especially when it's off-site. There should be some creative work done to figure out how to get manufacturers (and processors?) to pitch in to cover monitoring by EPA or some third party. Yikes - do it through some sort of SIEF process?
Getting some quality, meaningful data to support risk evaluation and potential restrictions is a valid and doable Action Plan under today's TSCA .