Creekstone loses in court

A federal appeals court has ruled against Creekstone Farms and in favor of the government. Quick summary: Creekstone wanted to go beyond USDA regulations and test all of its cows for mad cow disease. The USDA, beholden to the interests of larger meat packing companies who don’t want to compete on safety, told them they couldn’t. A lower court had ruled in favor of Creekstone, but now it looks like the company won’t get the chance to market their product with greater assurances of safety. Thanks, USDA!

Previously:
Paul Roberts and I debated food safety in the L.A. Times here and here.

Comments

  1. Joshua Macy says:

    It would have been a laugh riot if Creekstone had pursued their strategy, and then had cows falsely test positive for Mad Cow disease (apparently quite likely if they’re using the approved test). While I’d let them go ahead and do it, I can’t feel too badly that they’re not going to be able to play on the public’s irrational fears and statistical ignorance or contribute to the no risk, any time, anywhere, for any reason culture. Those people are generally not on the side of the angels when it comes to personal freedoms.

  2. Ben says:

    Well, but Joshua, I fail to see how a forbidding the company from trying to be safer is a victory for personal freedom.

  3. Joshua Macy says:

    Because it’s fake safety. Every binary test can have two kinds of errors: false negatives (condition is true but is reported as false), and false positives (condition is false but is reported as true). The measures of those are called the sensitivity and specificity of the test. It’s a simple fact of statistics that if the condition you’re testing for is unlikely to be present in the population then most of the positives you see are going to be false positives; by performing the test all you’re really doing is generating false alarms.

    It so happens that in this particular case, the chance that the disease is actually present is particularly low (only 3 cases in the US ever), and the test has a particularly high rate of false positives. The combination statistically guarantees that Creekstone would eventually turn up cases and cause a panic, but be wrong. And if they have anybody halfway competent advising them on this, they have to know that. So far from trying to be safer, they’re trying to sell snake-oil, preying on the public’s ignorance of the actual risks.

    As I said originally, if it were up to me I’d let them go ahead anyway, but I’m pretty laissez-faire about attempts to hoodwink the public, viewing the costs of preventing it as often being worse than the disease. Witness the fact that the government’s argument as to why they can forbid this is particularly bogus, turning on a blatant miscontrual of what it means to treat instead of test an animal rather than straightforwardly calling it for what it is: a form of “This Way to the Giant Egress!” fraud. But I’m not going to shed any tears over Creekstone not being able to gin up any hysteria and make people even more fearful of non-risks than they already are.

  4. Joshua Macy says:

    On further reflection, I’m probably too ready to dismiss the possibility that Creekstone is just a bunch of ignorant cattle ranchers who don’t get why what they want to do is a bad idea. I have to keep reminding myself that even MDs, who are required to pass courses in statistics and whose daily jobs demand they interpret the conditional probabilities of such tests have a pretty dismal record when it comes to understanding statistics.

  5. Jacob Grier says:

    Hi Joshua,

    I agree with you that this probably isn’t the best use of resources. As a consumer I’d rather have beef companies putting safety efforts into more common food borne illnesses than into something rare like mad cow.

    There are a couple other issues at work here though that I should have researched better. One is that some major foreign markets, like Japan, already use the test on all their cattle and the country used to require importers to do so as well. Rightly or wrongly, not being able to meet the same standard kept them out of those markets.

    I’m also not sure if it would create panics. Finding false positives after the beef is released to the market certainly would, but finding them early and either re-testing or taking the cows out of the cycle before they’re slaughtered might not. I assume that’s what ranchers in Japan are doing.

  6. Joshua Macy says:

    Actually, it seems that Creekstone’s position is even worse than I thought. Apparently the test they want to use won’t even detect the disease in cattle that have it, because they slaughter them younger than the minimum age where the test can detect the presence of the disease. And they know this, because the FDA told them so. So even if the cattle had the disease, they’d pass the test. So I really don’t see that their claim that they just want the freedom to make their beef safer holds any water.

Leave a Comment

*