From Jacob Grier at Slate:
We Used Terrible Science to Justify Smoking Bans
For three anti-smoking advocates—local physicians Richard Sargent and Robert Shepard, and activist and researcher Stanton Glantz from the University of California at San Francisco—this sudden drop in heart attacks was proof that smoking bans usher in extraordinary benefits for public health. “This striking finding suggests that protecting people from the toxins in secondhand smoke not only makes life more pleasant; it immediately starts saving lives,” said Glantz in a press release sent out by UCSF.
Newspapers ran with the story, credulously assuming that the correlation had been truly caused by the smoking ban. “The bottom line of Helena’s plummeting, then soaring, heart attack rate is painfully obvious,” warned an op-ed in the New York Times. “Secondhand smoke kills.” The BBC projected that “[banning] smoking in public places could prevent hundreds of deaths from heart disease.” Wire services carried the result around the globe, and even the conservative Wall Street Journal cited the result as an important finding.
When the Helena study and its heirs were originally published, a few scientists noted that the results were wildly implausible and the methodologies deeply flawed. So did a handful of journalists, including Jacob Sullum writing for Reason (to which I am also a contributor) and Christopher Snowdon in England. Yet their criticism was generally ignored. Studies reporting miraculous declines in heart attacks made global headlines; when better studies came along contradicting those results, they barely registered a blip in the media. …
There were good reasons from the beginning to doubt that smoking bans could really deliver the promised results, but anti-smoking advocacy groups eagerly embraced alarmism to shape public perception. Today’s tobacco control movement is guided by ideology as much as it is by science, prone to hyping politically convenient studies regardless of their merit and ostracizing detractors.
This has important implications for journalism. More.
Yes, the winds were favourable for pom pom-waving, unfortunately.
Of course, if science communicators keep getting away with it, they will keep doing it. Science really can be high tech voodoo using numbers. And then it must be kept in place by fear, not trust.
On a personal note: I found the anti-second hand smoke campaign rolled out across Canada some years ago disturbing. One government-sponsored ad featured women who were dying of cancer, whose husbands had smoked. The clear implication was that the wife got cancer (implication terminal) because the husband smoked.
The problem is, we don’t usually know that. Smoking greatly increases a person’s chance of lung cancer but not all lung cancers are caused by smoking. And second-hand smoke damage necessarily depends on many factors, including how much time a person spends involuntarily inhaling how much smoke.
I have never smoked, disapprove of the practice for many reasons, and support smoking bans in public places, as well as crackdowns on sales to minors and smuggling.
But I am concerned about using science to pretend we know more than we do, using voodoo numbers. That could lead to later family problems: = Dad killed Mom because he smoked. Science PROVES it! = We hate Dad. = [that’s just great when Dad is old and sick and needs family support]
Please. The world is full of problems for which we have a much clearer trail of cause and effect. – O’Leary for News