March 22, 2019 – In this Policy Forum, Gretchen Goldman and Francesca Dominici raise concerns over recent developments at the U.S. Environmental Protection Agency (EPA) that “stand to quietly upend the time-tested and scientifically backed process the agency relies on to protect the public from ambient air pollution.”

Goldman and Dominici cite proposed changes to how the EPA handles causality between air pollutants and health effects. An alternative approach to causality in this context, called manipulative causation, is being promoted by Tony Cox, the current chair of the EPA Clean Air Scientific Advisory Committee (CASAC). The CASAC provides independent advice to the EPA Administrator. As CASAC Chair, Cox will oversee the committee’s development of a scientific recommendation to the EPA administrator on whether a new ambient particulate matter standard should be set in order to protect public health in the U.S.

Cox’s proposed approach here, the authors explain, argues that in order to inform causality, data must pass the test of manipulative causation; the assumption of this framework is that implementing a new, more stringent air quality standard must measurably change health outcomes, and unless there is direct evidence of this, the regulatory action shouldn’t be taken. The framework further argues that epidemiological studies only provide evidence of an association between exposure to air pollution and health effects – that they don’t show causation. Cox’s approach calls for several tests of causality, including those that rely on machine learning.

“Although, in principle, the concept of manipulative causation can be viewed as attractive,” the Policy Forum authors write, “in reality, this requirement is unattainable in the context of observational data of air pollution and health.”

The authors cite several reasons why, also suggesting the rate at which this approach is trying to be standardized is concerning. They conclude by saying the philosophy proposed is contrary to the overarching idea that to protect public health, we should err on the side of protecting people. “Importantly,” they conclude, “if the particulate matter and ozone standards are loosened, people will suffer the consequences.” The matters discussed in the Policy Forum will be explored at a meeting on 28 March.

Journal Article: