While harder to discern, another EPA risk evaluation severely understates risk, this time for methylene chloride

Richard Denison, Ph.D., is a Lead Senior Scientist.

When EPA released the draft of its risk evaluation for methylene chloride at the end of last month, some were surprised that EPA had identified numerous unreasonable risks presented under a variety of the chemical’s conditions of use.

In an earlier post, EDF provided some context, noting how dangerous the chemical is and raising initial concerns that EPA was once again excluding known uses and exposures, making unsupported assumptions, and applying inappropriate risk benchmarks that were once again leading it to significantly understate the actual risks posed by methylene chloride.

Four weeks later, EDF has confirmed these concerns in spades.  Last night we filed 84 pages of comments on the draft risk evaluation, for consideration by EPA’s Scientific Advisory Committee on Chemicals (SACC), which will meet next week to peer review the draft.

EDF’s deep dive into the draft demonstrates that EPA has employed a host of unwarranted and unsupported assumptions and methodological approaches that lead it to either avoid identifying unreasonable risk when it should have, or to understate the extent and magnitude of the unreasonable risks it did identify.  Below we summarize some of the major concerns, which are addressed in detail in our comments. 

Underestimation of occupational risks:  Of particular concern is the extent to which EPA has underestimated occupational risks.  As part of its review, EDF conducted extensive analyses of each of the hundreds of individual risk estimates EPA has made in this draft risk evaluation.  Along with our written comments, we submitted a spreadsheet to share the results with the SACC and others.  Our analyses reveal and quantify five major ways in which EPA has underestimated occupational risks, including through: its unsupported assumptions regarding worker use of personal protective equipment in many scenarios; its use of a cancer risk benchmark level for workers that fails to protect them as a vulnerable subpopulation as required by TSCA; its failure to consider combined exposures of workers from multiple sources; its failure to identify unreasonable risks for the most highly exposed, and hence most vulnerable, of occupational non-users (ONUs); and dismissal of numerous unreasonable risk findings by invoking “uncertainty” or unwarranted use of PPE, or without any explanation at all.

To give you a sense of what we found, let me provide two examples:

First, EPA’s assumption that workers would always wear effective respiratory protection under many conditions of use dramatically reduced how often it found risk and how much risk it found.

For the nearly 30 conditions of use where EPA assumed workers always wear respirators, that assumption led EPA to downgrade 74% of its risk estimates for the various health effects it examined.  Absent that assumption, EPA would have both:  identified more unreasonable risks than it did; and found those unreasonable risks it did identify to be much higher.

Second, EPA’s decision to use a very high cancer risk benchmark for workers of 1 in 10,000, instead of the 10- or 100-fold more protective population-level standards EPA virtually always uses, also radically altered the outcome of EPA’s inhalation risk characterizations.

Across all of methylene chloride’s 65 conditions of use entailing worker exposures, EPA found that its inhalation risk estimates exceeded EPA’s chosen cancer risk benchmark of 1 in 10,000 a total of 88 times.  In contrast, our analysis of EPA’s own risk estimates found that, relative to EPA’s 1 in 10,000 cancer risk benchmark:

  • a cancer risk benchmark of 1 in 100,000 was exceeded an additional 146 times, for a total of 234 exceedances; and
  • a cancer risk benchmark of 1 in 1,000,000 was exceeded an additional 214 times, for a total of 302 exceedances.

These data show that EPA has dramatically understated the risks to worker from inhalation of methylene chloride.

Other major areas of concern we address in our comments include:

  • Exclusion of known uses and exposures: Once again, EPA has abdicated its responsibility under TSCA to identify and evaluate the risks the chemical presents to the general population, by excluding from its risk evaluation conditions of use and exposures that are known or reasonably foreseen, including exposures from releases of methylene chloride to air, water, and land.
  • Insufficient consideration of vulnerable subpopulations: EPA has not met its mandatory duty under TSCA to thoroughly identify and evaluate the risks to vulnerable subpopulations. These include subpopulations that are genetically susceptible to methylene chloride exposure; the developing fetus who may be exposed through placental transfer of the chemical; and consumers and others who may be at risk of cancer from acute exposures.
  • Dismissal of epidemiological evidence: EPA has sought to downplay or dismiss epidemiological evidence through a series of unsupported, misleading arguments and the application of flawed, biased systematic review criteria that do not represent best practice.
  • Failure to appropriately account for uncertainty: EPA has neither acknowledged nor addressed the major uncertainties in the available hazard data, including by not applying or underestimating necessary uncertainty factors when deriving its benchmark risk values.  Ironically, EPA invokes uncertainty as an unwarranted basis for ignoring risks it has identified to the environment and to ONUs, and for not accounting for combined exposures to methylene chloride.
  • Failure to use its authority to address data gaps and uncertainties: Even as it invokes lack of data and uncertainty as reasons to avoid finding risks, EPA has utterly failed to utilize the enhanced authorities Congress granted it in 2016 to ensure that it has or obtains robust information on methylene chloride’s uses, hazards and exposures.

The SACC needs to thoroughly assess the scientific consequences of EPA’s unsupported assumptions, selection of benchmarks, and methodological choices.  The fact that EPA found unreasonable risks more often than in other risk evaluations cannot be used to obscure the fact that EPA has significantly underestimated methylene chloride’s risks to workers, the general population, consumers, vulnerable subpopulations, and the environment.

If not corrected, this will have a real impact on human health and the environment.  If EPA’s approaches erase unreasonable risks, then EPA will not regulate the chemical under TSCA and will forgo its only opportunity to ensure this chemical’s risks are mitigated.  And if EPA understates the extent and magnitude of the unreasonable risks it does identify, then any subsequent regulation EPA promulgates under TSCA will be under-protective.




This entry was posted in Health policy, Health science, Regulation, TSCA reform, Worker safety and tagged , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.