The trouble with enlightenment, blowing smoke in the media, the evil that employers do, and other matters. THE 30% SOLUTION
By DANIEL SELIGMAN REPORTER ASSOCIATE Robert A. Miller

(FORTUNE Magazine) – Our Nexis search was ''SMOK! W/25 30 PERCENT AND DATE AFT 6/1/94,'' and it instantly hit pay dirt. We were asking for all news stories since June 1 wherein any references to smokers or smoking appeared within 25 words of ''30 percent.'' Nexis found 53 stories that met the criteria, and, as expected, all dealt with the latest big blast against passive smoking. Delivered in the Journal of the American Medical Association (JAMA), its core finding was this: Women who had never smoked but whose spouses were smokers had, on average, a 30% or so greater chance of getting lung cancer than did nonsmoking women whose husbands were also weed-avoiders. Media reportage on the JAMA findings was at once predictably uncritical and curiously incurious. The stories dwelt heavily on that 30% figure and on the ritual denials of the Tobacco Institute, whose bad guys weighed in with a judgment that ''there is a very, very low level of risk if there is any at all.'' Several reporters quoted Stanton Glantz, a passive-smoking expert at the University of California at San Francisco, who served up the arguably clunkiest metaphor of the week. ''The scientific evidence that passive smoking causes lung cancer is pretty ironclad,'' Stanton said, then added, ''This makes the iron quite a lot thicker.'' Nobody seriously denies today that cigarettes kill their smokers in large numbers, and even the Tobacco Institute says ''probably not'' when asked if smoking is healthy. (That was the answer lamely or gamely given UPI by TI spokesman Thomas Lauria the other day.) But Glantz to the contrary, the argument about passive smoking is far from settled, and it keeps getting more bitter as its policy implications come into focus. If smokers are seriously endangering others as well as themselves, then they are deprived of their single most compelling argument, which is that their habit is their own business, so kindly buzz off. The new study led quite a few authority figures to zero in on this implication of the 30% increase in lung-cancer risk. On June 8, Dr. Randolph Smoak of the AMA could be heard proclaiming on CNN: ''Behavior should not be legislated, but when one person's addiction causes harm to yet another person's health, that's where the line must be drawn.'' This brings us to a major weirdity about the JAMA study. In the blizzard of regressions and correlations being laid on the table, it nowhere gets around to quantifying the lung-cancer risks run by nonsmoking spouses of smokers. All it says is that the risk is some 30% greater than it would be if the spouse were also a nonsmoker. But what is that risk? What's the baseline to which we are adding 30%? On the evidence in Nexis, this question evidently occurred to nobody at all in the media. , The answer -- for which we are indebted to Dr. Donald F. Austin of the Oregon Health Division, who cheerfully performed the necessary calculations after getting called by your servant -- is 0.000048. In other words, there are 48 chances in one million that a nonsmoker married to another nonsmoker will get lung cancer in any one year. This figure rises to 0.000061 if the nonsmoker's spouse is a smoker. Stated another way, there is one chance in 20,833 of getting lung cancer in a year if your spouse is a nonsmoker, and the probability increases to one chance in 16,393 if he is a smoker. Another perspective: The chance of your dying from lung cancer because your spouse is a smoker is 1.2 in 100,000, which according to the U.S. National Center for Health Statistics is also the annual risk of death from a ''fall from one level to another.'' Dr. Austin, one of the scholars who worked on the JAMA report, indicated that when you go from annual to lifetime risks, a smoking spouse increases the odds of your dying from lung cancer from one in 304 to one in 236. One hesitates to come down on the same side as the Tobacco Institute, but one inclines to agree that we are looking at ''a very, very low level of risk.'' And a high level of rotten reporting.