“And, I can say without qualification, the victims have been right and the experts wrong in every significant story I’ve covered.”

He did not say “every story I have covered” but “every significant story I have covered.” Perhaps in Mr. Higgs mind the only significant ones were the ones in which he found the experts to be wrong. Conformation bias is almost explicit in such a comment.

Each year, 17000 non-smokers in the US develop lung cancer. Even if 10% are due to second hand smoke, that’s nothing to laugh at.

But 10% are NOT due to second hand smoke. Statistically, there is a greater likelihood of dying from a stray bullet. The odds of dying in a car crash are significantly higher, and that is not semantics.

I don’t assume, I look at case studies of these exact things. Tobacco and AGW denial campaigns are prime examples. Exponent and PR consultants are brought in to do damage control once the shit hits the fan.

In a similar fashion, HIV doesn’t cause AIDS, flu doesn’t cause flu symptoms, and a fatal gunshot wound doesn’t cause death. Let’s see someone argue that in court. “The bullet didn’t kill him, the blood loss did.” If you take the semantics to the extreme, the direct cause of most deaths is hypoxia.

“The odds of getting cancer from second hand smoke are 12.5 in 1M. The odds of getting cancer in the absence of second hand smoke are 10 in 1M. There is your 25% ‘increased risk’.”

Each year, 17000 non-smokers in the US develop lung cancer. Even if 10% are due to second hand smoke, that’s nothing to laugh at.

But here’s something to laugh at, speaking of bias.
“Why review articles on the health effects of passive smoking reach different conclusions”
http://www.ncbi.nlm.nih.gov/pubmed/9605902

In multiple logistic regression analyses controlling for article quality, peer review status, article topic, and year of publication, the only factor associated with concluding that passive smoking is not harmful was whether an author was affiliated with the tobacco industry.

It just isn’t rational to assume that any large company would go to such great expense to push a bogus or dangerous product onto the market, then go to further expense buying bogus research. Business is risk management, and the risk involved in your implication is too great. Remember that they also have shareholders to answer to. Public scandal, even if cleared afterward, is extremely damaging to a publicly traded company. Bad news causes investors to bail out, driving stock prices down. They might come back if enough traders take positions while the price is low, but those traders will only do that with companies that show strong fundamental analysis, in which case they have enough revenue to ride out the storm anyway.

The “danger of second hand smoke” is just plain laughable. Just about every argument out there is based on a bogus EPA study that was thrown out by a federal court. The popular figure of “25% increased risk from exposure” is deceptive. The odds of getting cancer from second hand smoke are 12.5 in 1M. The odds of getting cancer in the absence of second hand smoke are 10 in 1M. There is your 25% “increased risk”.

“Maybe they don’t ask a question they don’t already know the answer to.”

So they already knew before doing the research that secondhand smoke doesn’t cause cancer, that a double hull wouldn’t have prevented the Exxon Valdez spill, that Suzuki Samurai wouldn’t tip over during a turn, that Toyota’s sudden acceleration isn’t caused by electronics, etc.
Is their intuition so good that they knew all the conclusions before doing the research, or do they just deliver the conclusion their client pays for?