Antibiotics, Animals, and Agency Discretion

Cross-posted from Georgetown Law Faculty Blog.

When an agency defends over three decades of inaction on an important problem by saying that acting would take too long, one hopes a judge reviewing the agency's inaction will see through the pretense.  This is exactly what happened this week, when a federal magistrate judge in New York ruled that the U.S. Food and Drug Administration had acted arbitrarily in citing time and resource constraints in declining to limit the use of antibiotics in animal feed. 

Some 80 percent of the antibiotics used in the United States are given to animals destined to become part of our food supply.  Most of these drugs are given not to sick but to healthy animals; they are used not to treat disease but to promote animal growth or to prevent bacterial infections from occurring.  Both of these purposes are tied to the industrial nature of animal food production in this country.  The economic imperatives of industrial food production reward faster animal growth, and the microbial realities of the modern animal food production facility -- in which animals are tightly confined in stressful, bacteria-rich (to put it nicely), infection-promoting conditions -- encourage herd-wide administration of antibiotics to prevent infections. 

The widespread administration of antibiotics to food animals for non-therapeutic purposes has contributed to the development and spread of antibiotic-resistant strains of bacteria.  When bacteria become resistant to an antibiotic, that antibiotic will no longer treat the infection the bacteria cause.  Antibiotic resistance is to our wonder drugs what Kryptonite is to Superman.

The FDA must approve animal drugs before they are marketed.  The central criterion for approving such drugs is safety: the FDA has said that it will approve a new animal drug only if there is a "reasonable certainty of no harm to human health."  The FDA also may withdraw existing approvals for animal drugs.  Here, too, the central concern is safety.  If the agency finds that an animal drug now on the market is not safe, it must withdraw approval for that drug.

Many antibiotics administered to animals were approved in the 1950s, soon after antibiotics became widely available and before resistant bacteria had begun to outrun them.  By the 1970s, however, the FDA was already actively studying the relationship between herd-wide administration of antibiotics to food animals and antibiotic resistance in the human population.  In 1977, the FDA found that feeding antibiotics to food animals to promote growth and prevent infection was likely to promote antibiotic resistance in humans and harm human health.  The agency therefore gave notice that it would hold hearings on whether to withdraw its prior approvals for certain antibiotics used for these purposes.

The FDA never got around to holding those hearings and thus never got around to withdrawing the relevant approvals.  Yet the agency also never retracted its assertions -- which only grew in urgency and scientific credibility in the intervening years -- that feeding antibiotics to animals destined for human food consumption was not safe for humans.  Indeed, the FDA continued to sound the scientific alarm about the risks posed by widespread, non-therapeutic administration of antibiotics to food animals.  But still it did nothing to take the drugs off the market.  Not surprisingly, after watching the FDA sit still for decades, public health and environmental organizations eventually grew restless, and petitioned the agency to hold the hearings and to withdraw the relevant approvals.

Last November, almost 35 years after finding that using antibiotics in animal feed was linked to antibiotic resistance in humans, the FDA finally answered the petitions.  By denying them.  The agency explained that in order to withdraw the approvals for antibiotics in animal feed, it would need to hold formal hearings.  This would be expensive and take a lot of time, the agency said, and so it would rather just ask the drug manufacturers and users to please practice "judicious" use of antibiotics in animal feed.  In denying the petitions, the FDA did not address the thousands of pages of scientific evidence the public health and environmental groups had offered in support of their claim that non-therapeutic use of antibiotics in animal feed was not safe, beyond saying that a hearing on the science would be a complicated affair.

The FDA should have known better.  Several years ago, the U.S. Environmental Protection Agency found itself in a similar spot.  Environmental groups had petitioned the agency to regulate greenhouse gases from automobiles under the Clean Air Act, asserting that the statutory criterion for regulating -- the endangerment of public health and welfare by air pollution -- was met by greenhouse gases.  The EPA denied the petition, reasoning in part that it preferred to use voluntary rather than regulatory measures and that the science on climate change was complicated.  The matter eventually found its way to the Supreme Court, which held, in Massachusetts v. EPA, that the EPA had acted arbitrarily and capriciously in saying it would refuse to regulate greenhouse gases even if it had the statutory authority to do so.  The Court held that the EPA had erred by justifying its inaction by reference to policy concerns (such as the preference for voluntary measures) rather than by reference to the statutory framework of endangerment.  The Court did not order the EPA to regulate greenhouse gases, but it did direct the agency to take a meaningful look at the scientific evidence on climate change and to tie any decision it made to the legal framework under which it was operating.

Of course, when the EPA eventually did take a hard look at the scientific evidence on climate change, it concluded that greenhouse gases do indeed endanger public health and welfare.  It has since embarked on the regulatory program required by the Clean Air Act for air pollutants that threaten health and welfare.

The FDA's denial of the petitions on antibiotics in animal feed has the exact same flaws as the EPA's denial of the petition on greenhouse gases did.  The FDA grounded its decision not in the underlying statutory framework, but in its own preference for voluntary over regulatory measures.  The agency avoided altogether the scientific evidence on the risks of antibiotics in animal feed.  And, in a particularly ridiculous touch, the FDA complained that taking action on antibiotics in animal feed would just take too darn long.  This, from an agency that had first concluded in 1977 that widespread, non-therapeutic use of antibiotics in animals destined for the human food supply posed a risk to human health.  In 1977, Elvis was still alive, disco ruled, and Jimmy Carter was president.  FDA is an agency in a hurry, clearly.

On Monday, the FDA got its comeuppance.  A federal magistrate judge rejected as arbitrary the agency's explanation of its denial of the petitions to withdraw approvals for the non-therapeutic use of antibiotics for food animals.  Magistrate Judge Theodore Katz held that the FDA erred, as EPA had in Massachusetts v. EPA, by failing to tether its decision to the underlying statute.  The judge also didn't buy for a minute the agency's complaint that the process for withdrawing its prior approvals would take too long.  Noting that the petitions to withdraw the approvals had been pending for many years, the judge observed: "The position that instituting withdrawal proceedings -- what the statute mandates -- is too time consuming is both ironic and arbitrary. Had the Agency addressed the Petitions in a timely fashion, withdrawal proceedings could have been commenced and completed by now."  In a footnote, the magistrate judge added: "One can only wonder what conceding the absence of an effective regulatory mechanism signals to the industry which the FDA is obligated to regulate."  Monday's ruling follows a March decision in which Magistrate Judge Katz held that the FDA had erred in refusing to hold hearings on the non-therapeutic use of antibiotics for food animals.

Magistrate Judge Katz's latest ruling is a bracing reminder that agencies like the FDA cannot simply decline to exercise the responsibilities that have been given to them or refuse to confront the scientific evidence they have gathered.  Rather than compound absurdities by raising ever flimsier arguments in support of its inaction -- among other things, the FDA has argued to the court that its decision on antibiotics in animal feed was not a reviewable agency "action" -- the FDA should put its shoulder to the wheel and do the work the law has assigned to it.