Apples, Arsenic, and Risk - Part 1: One in 500

In the January 2012 Consumer Report's article "Arsenic in your juice," they write:

Our analysis was led by Richard Stahlhut, M.D., M.P.H., an environmental health researcher at the University of Rochester with expertise in NHANES data, working with Consumer Reports statisticians. Ana Navas-Acien, M.D., Ph.D., a physician—epidemiologist at Johns Hopkins University’s Bloomberg School of Public Health, also provided guidance. She was the lead author of a 2008 study in the Journal of the American Medical Association (PDF) that first linked low-level arsenic exposure with the prevalence of type 2 diabetes in the United States.

So Consumer Reports accepts that there is a "link" between type 2 diabetes and low levels of arsenic...which is most likely one of the contributing factors for support of this statement that is also in their report:

But chronic toxicity can result from long-term exposure to much lower levels in food, and even to water that meets the 10-ppb [EPA] drinking-water limit.

as well as printing this quote...

“I suspect there is an awful lot of chronic, low-level arsenic poisoning going on that’s never properly diagnosed.” - Michael Harbut, M.D.

leading to this...

Consumers Union urges federal officials to set a standard for total arsenic in apple and grape juice. Our research suggests that the standard should be 3 ppb. 

Such standards would better protect children, who are most vulnerable to the effects of arsenic... 

Moreover, the EPA should impose stricter drinking-water standards for arsenic...

Ignoring the desire to have zero arsenic in the food and drink we consume, would lowering the level of arsenic to 3 ppb or less make a difference?  Would lowering the amount of arsenic in drinking water to a level less than 10 ppb (as the EPA is proposing) make a difference in preventing chronic toxicity?

Consumer Reports believes it would:

For known human carcinogens such as inorganic arsenic, the EPA assumes there's actually no "safe" level of exposure, so it normally sets exposure limits that include a margin of safety to ideally allow for only one additional case of cancer in a million people, or at worst, no more than one in 10,000. For water with 10 ppb of arsenic, the excess cancer risk is one in 500.

Public Health, especially those that look at chemical exposure risk, have been pigeon-holed by the "no safe level of exposure" for carcinogens statement.  Why?  because it always leads to the "risk of cancer" discussion of one in a million .

How do you think the general public - including medical professionals - understand the statement:

For water with 10 ppb of arsenic, the excess cancer risk is one in 500.

That statement is only true if you accept EPA's model that generated that risk.  And even if you accept their model, there is this little thing called reality that can nullify it.

But the damage has been done.  Drinking water with 10 ppb will bring about one more cancer for every 500 persons according to the EPA.

No...no it will not.  That's what the model predicts, and because the "people demand a number" we have made the model's output sacrosanct.  However, the prediction of one excess cancer risk in 500 for 10 ppb arsenic is nothing more than an answer spit out of an equation:

Source

That risk level - "R" is what we want to see, one in a million, one in 100,000... and that along with the "Cancer Potency" or "q" for the carcinogen in question will tell us how much of that chemical a 70 kg person can be exposed to over a lifetime (70 years) so that out of 1 million or 100,000, or 500 we should not see more than one additional cancer.

Sounds reasonable, until you look at how simple that formula is. The lifetime cancer risk - R - is whatever number we want to put in there.  Same with body weight.  Those numbers are not going to ever come into dispute.  So the "safe" intake, or in this case, the intake that would reduce your lifetime risk to one out of something, is predicated solely on the "theoretical cancer potency estimate for humans" which is also called the "cancer slope factor."

When the EPA produce data that shows arsenic at 10 ppb resulting in "male bladder cancer model outputs" for a "Lifetime Risk" of 3.20E-03 (3.2 per 1000) it is stating that their model's cancer slope factor (CSF) is accurate.  And to know if it is accurate, we need to compare the projected incidence to the actual incidence of bladder cancer in males.

That's reasonable, isn't it?  Well you would think so, but somehow that little bit of common sense seems to be missing.  You know, if the model predicts x is x what we are seeing.  Instead we have a system that has developed a methodology devoid of conformation.

And to top it off, we have lost sight of just what "for water with 10 ppb of arsenic, the excess cancer risk is one in 500" actually means.  That it is based on a theoretical cancer slope factor for one particular type of cancer.

Next post: Apples, Arsenic, and Risk - Part 2: EPA's Proposed Cancer Slope Factor.


.