Math errors and risk reporting
On Sunday evening, 60 Minutes gave an update to their story about Lumber Liquidators importation of lamenant flooring with high levels of formaldehyde. Turns out, the CDC has released a risk-assessment report. But the report has an error -- it fails to correctly convert between feet and meters at one point, leading to a big mis-calculation in the risk of cancer associated with the formaldehyde release. Got to get your units right, everybody! Remember the Mars climate orbiter!!
But the report has an interesting perspective as well. On pages 36 and 37, it states that these floors are conservatively estimated to cause 2-9 more cases of cancer per 100,000 people exposured, and then points out that more an 1/3rd of all people living in the United states will develop cancer. Reading between the lines, the suggestion is that these numbers are very low, and maybe shouldn't be a general concern to the public.
I think it's interesting to juxtapose this with another on-going issue -- the Takata airbag recall. This recall is effecting 34 million airbags, about 1 in every 7 cars in the US, and costing more than 2 billion dollars. So far, around 8 people have died. If we multiply the deaths by a factor of 10 to account for injuries, we're in the ballpark of 20 million dollars per American hurt by their airbag.
Now, airbags -- that's a real immediate issue. Cancer risk from formaldehyde is much less obvious, since it happens over a long time and might be confounded with many other factors. But let's take the (incorrectly) reported numbers of 2-9 cases per 100,000 exposures. We live in a country of around 300 million. Suppose 1 million (1 in 100) of those people get exposed. Then that's about 20 - 90 cases of cancer. At the the cost rate of the airbag recall, it would be reasonable to expect Lumber Liquidators to pay 400 million - 1.8 billion dollars to fix the problem -- quite a pretty penny. The corrected numbers in the CDC model are 6-30 cases per 100,000, so 600 million - 10 billion dolars!
These numbers seem out of wack with reality, one way or the other. I guess it's up to us to figure out which is the right way.
References
The CDC, Atsdr, and National Center for Environmental Health's original Feb 10th version of the report. Here is a Cached version if the official version disappears on us.
Bloomberg: Lumber Liquidators' Flooring Has Low Risk of Cancer, CDC Says
CDC's offical acknowledgement of the error and the need for corrections.
Bloomberg: Lumber Liquidators' Flooring Has Increased Cancer Risk, CDC Says