Secondhand Smokescreen

Researchers reported this week that nonsmokers living with smokers are exposed to tobacco smoke. That's obviously not news. So that's not how the study was touted by the researchers and reported by the media.

"Study: Wives of smokers absorb cancer chemicals from smoke," alarmed an Associated Press headline.

Dr. Stephen Hecht and other University of Minnesota researchers compared urine samples from 23 women who lived with smokers with urine samples from 22 women who lived with nonsmokers.

Hecht reported that the women who lived with smokers had blood levels of two chemicals — NNAL and NNAL-Gluc — about five times higher than the women who lived with nonsmokers.

The chemicals are produced when the body metabolizes a chemical called NNK, a component of tobacco smoke. Laboratory experiments indicate that massive doses of NNK — on the order of the NNK exposure from smoking two packs of cigarettes per day for 40 years — increase lung cancer rates in rodents.

Based on finding the byproducts of NNK in the women exposed to secondhand smoke and NNK being associated with cancer in lab animals, Hecht concluded to the Associated Press, "A number of studies have shown a connection between environmental tobacco smoke and lung cancer. Our study provides the first biochemical support for this data."

If spin were science, Hecht would win a Nobel Prize.

Biochemistry aside, Hecht's grossly misrepresented the state of the science on secondhand smoke and lung cancer. A credible link between secondhand smoke and lung cancer remains elusive despite more than 40 published studies.

The largest-ever study on secondhand smoke and lung cancer, published in 1998 by the World Health Organization's International Agency for Research on Cancer, reported no statistically significant increase in lung cancer risk associated with exposure to secondhand smoke.

That result was no surprise. It was the result the U.S. Environmental Protection Agency should have reported in its notorious 1993 secondhand smoke risk assessment — the study that greatly accelerated efforts to ban smoking in public places.

At the time of the EPA study, there were about 30 studies from around the world involving human populations exposed to secondhand smoke. Some studies reported weak statistical associations between exposure to secondhand smoke and lung cancer. The vast majority of studies reported no statistical association.

None of the studies were very good. All were statistical, not scientific in nature. All lacked data on how much secondhand smoke study subjects were exposed to.

But since the EPA already had pre-determined that secondhand smoke caused lung cancer — issuing guidelines for banning workplace smoking in 1989 — something had to be done to whip the science into shape.

The EPA statistically combined the results from the 11 published studies of U.S. populations. The agency hoped that statistical magic could be worked on the pooled results to produce the "correct" answer.

Alas, there was still no joy for the EPA. The statistical combination produced yet another a weak association between secondhand smoke and lung cancer. The association was not statistically significant, meaning that the agency could not rule out that the association occurred by chance.

More bad news arrived. Two more studies were published of U.S. populations exposed to secondhand smoke. Neither associated secondhand smoke with increased lung cancer risk.

Back to the drawing board in panic, the EPA brazenly abandoned standard statistical practices. The agency released a fudged result as its final product, concluding that secondhand smoke was a lung carcinogen that caused 3,000 deaths per year.

The tobacco industry challenged the EPA in court. A federal judge vacated the EPA's main conclusions stating that,"EPA disregarded information and made findings on selective information; ... deviated from its [standard procedures]; failed to disclose important findings and reasoning; and left significant questions without answers. EPA's conduct left substantial holes in the administrative records."

The ruling should have been a devastating blow to the hysteria surrounding secondhand smoke, except that it came more than five years after the EPA issued its report. The anti-tobacco industry exploited that time to convert the EPA's secondhand smoke junk science into conventional wisdom.

Now researchers like Hecht unabashedly cite the nonexistent EPA report to support the unsubstantiated assertion that secondhand smoke causes lung cancer.

Without the EPA report, after all, Hecht's new study is merely biochemical support that nonsmokers living with smokers are exposed to tobacco smoke.

Did taxpayer dollars need to be spent to prove that?

— Steven Milloy is a biostatistician, lawyer and adjunct scholar at the Cato Institute and publisher of