June 30, 2002
by David Kopel
As Fort Collins considers banning smoking in restaurants and all other public indoor places, The Denver Post's Northern Colorado Bureau did a good job interviewing Fort Collins residents on all sides of the issue: smokers, anti-smokers and property owners (June 16). But the Post's coverage of the science regarding secondhand smoke was a litany of misinformation.
Making several serious errors in a single sentence, the Post asserted: "The Environmental Protection Agency classifies secondhand smoke as a cancer-causing substance with no safe level of exposure, causing 60,000 deaths each year in the United States."
First of all, the EPA's classification of secondhand smoke as a carcinogen was declared void in 1998 by a federal District Court. The court found that the EPA "manipulate\[d\] the Agency's standard scientific methodology," acted in "complete disregard of statutory procedure," engaged in "circular" reasoning, appeared to have " 'cherry picked' its data", "deliberately refused to assess information," evaded review by outside experts, "changed its methodology" without explanation in the middle of the study, relied on contradictory and shifting scientific theories, and rigged the report to support predetermined political conclusions. (4 F.Supp. 2d 435).
The Post's "no safe level of exposure" statement is technically true, but, in the context of the article, quite misleading. It's almost impossible to establish a safe level of exposure for anything. No one has ever proven that standing outside in the sunshine for just 100 hours over the course of your life won't give you skin cancer. Likewise, nobody has proven that inhaling secondhand smoke for 100 hours over the course of a lifetime won't give you cancer. But there is no scientific evidence showing that minimal exposure to sunlight or to secondhand smoke is dangerous, either.
The Post's sentence implied that the EPA believes that "cancer-causing" secondhand smoke causes 60,000 deaths annually. In fact, the EPA never made such an estimate. The 60,000 figure was created by a California state agency, which suggested that most of the deaths were due to heart disease, not cancer. Skeptics, including the author of an article published in The New England Journal of Medicine in 1999, argue that these estimates are far too high: since passive smokers inhale only 1 percent as much as active smokers, their risks should be about 1 percent of the risk faced by active smokers.
The Post then cited a U.S. Surgeon General report "that secondhand smoke could cause fatal lung cancer and other respiratory illnesses, especially among young children." The Post's sentence structure implied that young children could get fatal lung cancer from secondhand smoke. To the contrary, the Surgeon General warned about lung cancer for adults and respiratory illness for children.
Reassuring restaurant owners worried about lower sales, the Post wrote: "A 1997 study conducted by researchers at the University of California at San Francisco, published in the American Journal of Public Health, found that smoke-free ordinances in California and Colorado did not hurt restaurant or bar sales."
Actually, the study was published in 1994, based on data from 1986-91. A follow-up study conducted by Northwestern University economics professor Michael Evans found that the previous study had misclassified and misdescribed the laws in most of the cities studied, had failed to distinguish fast-food restaurants (which are mainly carry-out) from sit-down restaurants, and had made a variety of technical econometric errors. When the various flaws were corrected, the data showed a significant decline in restaurant business, which grew worse over time, Evans said. (The full texts of both studies are available on a Philip Morris Web site, www.pmoptions.com .) The 1994 authors, by the way, argue that their study is still valid, and that the identified errors are minor. (See www.no-smoke.org .)
Two good antidotes for media-induced panic over other people smoking are the books Passive Smoking: The EPA's Betrayal of Science and Policy (Vancouver: Fraser Institute), and For Your Own Good: The Anti-Smoking Crusade and the Tyranny of Public Health (N.Y.: Free Press). Jacob Sullum, author of the latter book, pointed me to some of the research cited in this column.
A letter writer (June 10) complained that the Post had allowed an opinion columnist to use the word "gypped," which, the letter writer alleged, is a slur against gypsies. Actually, the Oxford English Dictionary describes the origin of "gyp" as "unknown," rather than as derived from "gypsy." In any case, if a word was once an ethnic slur but no longer has that connotation, it is pointless for newspapers to worry about giving offense. Otherwise, papers are going to have to give up calling anyone a "piker" (originally, a Californian slur against newcomers from Pike County, Missouri) and newspapers will never be able to write about the destruction wrought by vandals - for fear of offending descendants of ancient Germanic tribesmen. Oops, I should have said "tribes-persons."