Check out this Magnum Photos gallery of cows.
This was good news for grass-fed beef: Eliminate grain from a cow's diet and you'll keep its intestines from getting too acidic and spawning dangerous, acid-resistant bacteria. There was only one catch. The authors of the Science piece never specifically tested for E. coli O157:H7. Instead, they guessed that the pattern of O157:H7 growth and induction of acid-resistance would mirror that of E. coli strains that are always living in the colons of cattle. If this assertion were true, E. coli O157:H7 would reach dangerous levels only in gastrointestinal tracts of grain-fed cows.
But between 2000 and 2006, scientists began to take a closer look at the effect of diet on E. coli O157:H7 specifically. A different set of findings emerged to indicate that this particular strain did not, in fact, behave like other strains of E. coli found in cattle guts. Most importantly (in terms of consumer safety), scientists showed in a half-dozen studies that grass-fed cows do become colonized with E. coli O157:H7 at rates nearly the same as grain-fed cattle. An Australian study actually found a higher prevalence of O157:H7 in the feces of grass-fed rather than grain-fed cows. The effect postulated (and widely publicized) in the 1998 Science report—that grain-fed, acidic intestines induced the colonization of acid-resistant E. coli—did not apply to the very strain of bacteria that was triggering all the recalls.
What might explain this discrepancy? Scientists wondered whether there could be two subtypes of E. coli O157:H7 with varying degrees of acid-resistance. By that logic, the microbes from the grass-fed guts would be less resilient—and therefore less dangerous—than the ones that were growing up in the cows reared on grain. So they started running tests to find out.
In 2003, a research team from the University of Idaho reported no difference at all in the levels of acid resistance between E. coli O157:H7 from grass- and grain-fed cattle. (In both cases resistance was high.) Their conclusion stands in direct contrast to the broad claims about grain diets that have been made in the popular press since 2006. It must be that some other factor or factors were responsible for the development of E. coli O157:H7.
We don't yet know what these might be. But four studies, published between 2003 and 2005, have developed an intriguing hypothesis. Maybe, some reasoned, E. coli O157:H7 behaves differently from other strains because it develops in a different part of the cow's intricate digestive system. Sure enough, O157:H7 turned out to have a strong tendency to congregate in the recto-anal junction, whereas most other E. coli tend to gather primarily in the colon. Given that, we might presume that the production of E. coli O157:H7 depends more on its unique location than on what its cow host happens to be eating.
The point in dredging up these studies—ones the media never covered—is not to play gotcha with advocates of grass-fed beef. (As mentioned above, grass-fed beef may be healthier than conventional beef over all, and kinder to the animals.) Instead, it's a warning that advocacy for a trendy food choice might result in a public health hazard. Such a fear is confirmed by consulting the cooking directions provided by many purveyors of grass-fed beef. The home page for one major producer explains that "cooking 'real food' is not the same as cooking concocted food. … Grass-fed meats are best when raw (steak tartar), rare, or medium rare." Given that the FDA recommends cooking ground beef to 160 degrees to guarantee safety from E. coli, this eat-it-undercooked advice could be dangerous.
When it comes to the intricacies of our food system—and especially the meat industry—what's true one day can be less true the next. A case in point involves the final FDA report (PDF) on the source of the 2006 E. coli O157:H7 outbreak that motivated Planck to write her seminal Times op-ed. Released in March 2007, it suggests that the spinach wasn't contaminated by grain-fed, industrial cattle. Rather, the culprits were more likely to have been wild pigs or pastured (i.e., grass-fed) cattle—animals that were, of course, doing nothing more than eating what they were meant to eat.