Tuesday, November 22, 2016

Junk science is no way to win a jackpot: Column

Lawyers in the hunt to strike it rich are ignoring scientific nuance and misleading juries.

Alex Berezow and Hank Campbell

A jury in St. Louis awarded a woman over $70 million last month because her lawyers convinced a jury that talcum (baby) powder caused her ovarian cancer. This is the third jackpot verdict issued by a jury in that city against Johnson & Johnson. Since sharks are smelling blood in the water, surely more lawsuits will follow.

This madness must stop. There is no convincing evidence to support a link between talcum powder and ovarian cancer.

The majority of the data that supports a possible link is based on case-control reports, observational studies that compare diseased people with healthy controls to determine previous exposures. For instance, in May 2016, the journal Epidemiology published a paper that concluded that use of talcum powder increased the likelihood of ovarian cancer by 33%. But hold on.

That might sound like a lot, but consider that tobacco smoke increases the risk of lung cancer in men by 2,300%. Asbestos increases the risk of mesothelioma, on average, by 300% to 700%. By comparison, a 33% increase in ovarian cancer hardly constitutes “slam-dunk” evidence.

Cohort studies, on the other hand, often eliminate this problem by tracking volunteers based on current or recent exposures. Indeed, a new cohort study in Epidemiology assessed women who had used or had not used talcum powder in the past year. The researchers followed up with the women roughly six years later, and they concluded that there was no link between talcum powder and ovarian cancer.

They did, however, find a possible link between douching and ovarian cancer. If true, that might suggest that studies which show a link between talcum powder and ovarian cancer are actually confounded by douching. (In other words, women who douche may also be using talcum powder.)

The American Cancer Society remains skeptical of a link. The organization writes, “For any individual woman, if there is an increased risk, the overall increase is likely to be very small.”
Such nuance, of course, is purposefully ignored by lawyers seeking a jackpot verdict. Nuance and statistical uncertainty, which are inherent to the complex field of epidemiology, are thrown overboard in the hunt to strike it rich.

Numerous examples of such overreach abound. Former senator and presidential candidate John Edwards ignored most expert advice and found the few experts who would agree with him to conduct his personal injury cases. That's unethical and unscientific. The 2000 Julia Roberts movie Erin Brockovichmeanwhile, glorified a $333 million settlement against an electric company for allegedly causing cancer in the citizens of a California town that had hexavalent chromium in its drinking water — but a study 10 years later did not find a disproportionately high number of cancers in the area.

Such perversions of science by the judicial system are why we are in dire need of medical tort reform. Juries of 12 untrained laymen have proven over and over that they are not capable of logically analyzing biomedical and epidemiological evidence. Medical cases ought to be decided by a jury of experts.

Unlike in law or politics, where demagogues insist on black-and-white answers, science is full of gray areas. Unfortunately, we have a legal system that is incentivized to prey upon scientific uncertainty in order to line the pockets of unscrupulous lawyers with multimillion-dollar lawsuits. That won’t stop unless the system is reformed.

Alex Berezow, senior fellow of biomedical science at the American Council on Science and Health, holds a Ph.D. in microbiology and is a member of USA TODAY's Board of Contributors. Hank Campbell is president of ACSH. Follow them on Twitter @AlexBerezow and @HankCampbell.
You can read diverse opinions from our Board of Contributors and other writers on the Opinion front page, on Twitter @USATOpinion and in our daily Opinion newsletter. To submit a letter, comment or column, check our submission guidelines

No comments: