The US Environmental Protection Agency's proposal this week to all but eliminate sulfur in highway diesel fuel revives an interesting standard in the analysis of health risk: premature death.
Now, death is a sensitive subject for mortals. And the idea of premature death is utterly horrifying. If pressed on the subject, in fact, most people would probably judge their own prospective deaths to be premature, whatever their timing.
The philosophical difficulties notwithstanding, EPA tosses the concept around at every opportunity. For example, it linked 15,000 premature deaths/year with small-particulate pollution - soot - when it tried to toughen emission standards for soot and ozone smog last year. That initiative ran into trouble in the courts.
This week, EPA made a similar claim in support of the proposal to reduce the cap on the sulfur concentration of diesel to 50 ppm from 500 ppm. EPA Administrator Carol M. Browner asserted that the action would reduce emissions of nitrogen oxides by 95% and soot by 90%, then quantified the health effects like this:
"Soot and smog pollution are scientifically associated with 15,000 premature deaths annually and a million cases of respiratory problems each year. They are also responsible for some 400,000 cases of asthma attacks every year, including thousands of aggravated cases of asthma, especially in children."
A careful reading leaves a lot of room for those claims to wiggle. What, for example, does "scientifically associated with" mean?
The phrase certainly doesn't mean "cause." If it did, an agency head with a program to sell would have used the more concise and powerful term.
In similar fashion, the term "respiratory problem" covers much intellectual ground-from minor coughing to the inability to breathe. To get to "a million" instances of these problems, Browner must be taking in quite a number of the minor variety. How, then, does it become possible to scientifically associate those episodes with smog and soot strongly enough to distinguish them from ordinary coughing and thus to include them in the accounting?
Of course, citizens casually coming across Browner's statement about premature death aren't likely to fret much over those distinctions, which are trivial by comparison. She's talking about premature death, after all.
But what exactly is that? Somewhere, EPA must have a technical definition for "premature death."
Browner chose not to trouble the masses with that definition when she cited the scientific association of 15,000 annual episodes of premature death with smog and soot in the context of her sulfur rule.
Absent the clarification, the citizenry has no ready reason not to think that what she meant was that 15,000 healthy folks each year fall over in their tracks and die after inhaling smoggy or sooty air.
Extrapolation to the worst imaginable case no doubt comes easily to a great number of people, given the natural horror that surrounds the very idea of such "premature" death. To them, slashing the sulfur content of highway diesel must seem not only sensible but overdue, whatever the cost. And that is just not so.
EPA should define "premature death" whenever it uses the phrase in defense of its increasingly aggressive and extremely costly regulatory initiatives.
And once it has described what "premature deaths" represent in the context of its risk analyses, the agency should explain how it counts them, which must be fascinating.
