For the sake of health, governments should uncouple marketing from science in environmental politics. A good place to start is the US Environmental Protection Agency.
Defending the decision not to fix a regulatory mistake by her predecessor on sulfur in diesel fuel, new EPA Administrator Christie Todd Whitman co-opted a regrettable sales pitch. A Feb. 28 press statement, borrowing from the final rule adopted at the end of the administration of President Bill Clinton, said the action would eventually reduce emissions of nitrogen oxide by 2.6 million tons/year and of particulates by 110,000 tons/year. Then there's this:
"An estimated 8,300 premature deaths, 5,500 cases of chronic bronchitis, and 17,600 cases of acute bronchitis in children will also be prevented annually. It is also estimated to help avoid more than 360,000 asthma attacks and 386,000 cases of respiratory symptoms in asthmatic children every year. In addition, 1.5 million lost work days, 7,100 hospital visits, and 2,400 emergency room visits for asthma will be prevented."
Enumerated horrors
These are compelling numbers. Who wouldn't support a regulation promising to prevent thousands of premature deaths? The phrase itself is creepy enough to make most people suspend doubt and crave rescue. In the marketing of policy, enumerated horrors are very effective.
But do they mean anything?
The numbers in EPA's press statement derive from complex risk assessments essential to policy-making. The assessments themselves come from studies that statistically associate elevated levels of air pollution with elevated rates of mortality and phenomena related to respiratory disease. They are not really estimates but rather extrapolations from estimates of relative risk, which may or may not be significant and which are, in any case, very uncertain.
Numbers like these can breed spurious conclusions. If death rates increase as a function of measured ozone levels, for example, ozone seems to be killing people. But such analysis ignores other factors. Ozone, which certainly aggravates respiratory problems, forms in sunlight and therefore follows weather patterns. So is it ozone or heat, or some consequence of rising temperature, that causes the apparent elevation of death risk? And is the risk change large enough to be meaningful?
EPA's final rule contains a long section confessing to such uncertainty. At one point, citing problems not only of assessing the risks of premature mortality but also of assigning economic values to them, it asserts, "Premature mortality risks from air pollution tend to affect the very old more than the working age population."
Indeed. Statistical ambiguity, however, doesn't stop EPA from pretending to be able to count premature deaths and from citing the 8,300/year estimate frequently in its rule and press releases.
To highlight the uncertainty at work here is neither to denigrate the studies behind the numbers nor to scoff at the problem of air pollution. It is vitally important to know how air pollution affects health and to respond effectively. But the response needs to account for all dimensions of the problem and apply in the broad context of public welfare.
With diesel sulfur, as well as with its associated lurch on fine particles and ozone standards, EPA asserted a problem, assumed the worst about it, mandated the costliest remedy, then sold its aggression with numbers that imply more specificity than exists about complex matters over which scientists remain very much in doubt. This is irresponsible. It might divert policy from more-effective remedies. Worse yet, overselling of an aggressive but possibly ineffective response might lead people with respiratory problems into dangerous complacency.
As was argued in this space last week, EPA's rule cutting sulfur in highway diesel by 97%, instead of the 90% recommended by the refining industry, threatens fuel supply (OGJ, Mar. 12, 2001, p. 19). The prices of goods carried in trucks, including items related to health such as food and medicine, are going to rise.
Price increases
Dollar amounts of the likely increases are just estimates and don't need to be recited here. But the supply reductions and price jumps are much more certain than the 8,300/year premature deaths and other quantified ills that EPA claims its diesel-sulfur fiat will prevent when fully in force.
Whitman and the Bush administration can do nothing greater for health and nature than to remove this sort of statistical sleight of hand from environmental regulation.
