Press Box

Bill Keller Replies

The editor of the New York Times gives “Press Box” a piece of his mind.

Drugs and money

About a month ago, I knocked a Page One New York Times story titled “At F.D.A., Strong Drug Ties and Less Monitoring.” The magic of the Internet being what it is, Times Executive Editor Bill Keller e-mailed a detailed defense of his reporter’s piece to me at my account the next day. The magic of the Internet being what it is, I never received it—or more likely, the spam filter tossed Keller’s message into my bottomless junk mail pile.

This week, Keller remailed his note, I received it and am happy to publish it and respond.

If you’re coming to the topic belatedly, you needn’t read the original Times pieceor my critique to follow Keller’s argument or my response, which follows, but I encourage you to do so.

From: Bill Keller
Date: Jan. 3, 2005
To: Jack,If our FDA story failed to make its point clearly to a careful reader such as yourself, then that is certainly a strike against it. Maybe I was just better caffeinated than you, but I found the import of the story pretty clear, and it bore little relation to the version you so cutely satirized.First, I think it’s unfair to categorize it in the “underfunded agency” genre. The FDA may be underfunded or it may be overfunded, but the point of this story was that its regulatory mission was seriously distorted by a drastic reallocation of priorities under the 1992 formula—away from monitoring the safety of drugs already on the market and to the accelerated testing of drugs vying for market approval. The story is not about whether the FDA has enough, but about how it spends what it’s got. You at least imply that this is one of those stories in which an agency collaborates in poor-poor-me accounts of the horrors that have befallen the public interest as a result of cuts to its budget. The FDA seemed to understand that this was not one of those stories; it refused to cooperate in even a wink-wink way, and declined to give us the numbers the reporter, Gardiner Harris, sought. Gardiner had to calculate the pre- and post-approval numbers from several different sources, which is probably one reason nobody else has done this piece. Second, you are right that the case study we used to illustrate the point, Seldane, is a bit more complicated than our summary conveys. Your best point is that it took even longer for the (fully funded) FDA to get Seldane off the market than it did for the (not fully funded) FDA to stop Vioxx. But the point of using Seldane was to show that back in the day the FDA, once it had questions about a product, had mechanisms in place to answer those questions rather quickly. Now it does not. Now it depends entirely on voluntary efforts by the drug makers to discover safety issues regarding drugs that have been approved. That is a fundamental change in the way the FDA operates, and not widely understood. Seldane happens to illustrate that point. It is true [the] FDA was unable to force Seldane off the market, but it was able to issue a black-box warning, and—contrary to what you suggest—it had a devastating effect on sales of the drug. At least Seldane’s makers and their market-share-conscious stockholders thought so. Claritin quickly surpassed Seldane to become the biggest seller in the category. Third, you fault the “subdued hysteria” of those who say fast-tracking of drugs produced an avalanche of dangerous new medicines. That’s a fair point—except that our article carefully does NOT make that claim. You’ve made us into a straw man. Some 2,000 words into Gardiner’s piece, there are exactly four paragraphs on the debate over whether fast-tracking lets bad drugs into the marketplace. Three of those paragraphs cite government and industry officials explaining that this has not happened. The point of those four paragraphs was simply to say: most of what you read on regulation of drugs focuses on this inconclusive debate about fast-tracking, but hardly anyone writes about the inadequacies of the system for policing drugs AFTER they get to market. Which is exactly why we thought the issue merited a full page of our paper, and why I think your caricature of it is wrong. I hope your readers will pour coffee, click the link, and read the story for themselves. Best,

Dear Bill,

Sorry about the screw-up. I’ve added your name to my “trusted senders” list so my e-mail program will never mistake your words for spam again.

Let me field your objections one by one.

You reject my characterization of the Times story as an underfunded agency genre piece, insisting that it’s really about the “drastic reallocation of priorities under the 1992 formula. …”

I disagree on a couple of fronts.

To begin with, the formula you refer to was specifically set up to prevent “reallocation.” As your piece reports, the pharmaceutical industry worried that if it started paying the FDA millions in user fees to speed new-drug review, the agency would merely dump the money into its general budget. The formula committed the government to spending the user fees on new-drug review only and to continue supporting new-drug review with taxpayer funds at 1992 (and then later 1997) levels—in other words, it blocked the FDA from reallocating priorities. As the Times article states, “the industry promised to give the agency millions—in the 2003 fiscal year, $200 million—but only if the agency spent a specified level of money on new drug approvals.”

The next sentence supports my notion that the story is really about underfunding: “As Congressional support sank since then, the agency has cut everything else but new drug reviews. In the past 11 years, spending on the reviews has increased to more than four-fifths of the budget of the agency’s drug center from about half.” If the sinking of congressional support doesn’t qualify as underfunding, I don’t know what does.

The journalistic language that signifies underfunding peppers the piece. Your reporter writes of the “slashing of laboratories,” of laboratories “starved” of new equipment or “shut down.” Programs have “gotten squeezed” or “ended,” and financial constraints have forced the agency to cancel collaborations with academic drug-safety experts. We learn that one drug-safety project got funded only because the “agency has raided furniture and travel budgets,” that new-drug reviews receive ample funding for travel and training but the drug-safety division got “two thirds less, which keeps most at home,” and that “Congressional financing has lagged the agency’s escalating payroll costs.” Your reporter quotes an FDA report that laments that the FDA isn’t adequately funded to collect reports of problems with drugs on the market.

Your reporter repeatedly links the 1992 formula to the slashing, starving, shut downs, squeezes, and raids—but he produces no evidence or describes no mechanism to explain how the one has caused the others; he only asserts. For example, at one point he writes, “None knew that the reason was that money had to be shifted out of their [safety] programs into new drug reviews to satisfy the requirements of the agreement and industry demands.”

For a piece always kvetching about money, it is woefully short on hard budget numbers. In fact, the only hard number offered is FDA’s current budget ($1.8 billion). The curious reader would logically want to know if the FDA budget gone up since 1992, stayed level, or gone down.

I’ve got the answer. According to the FDA press office, Congress appropriated $746 million in FY 1993 to the agency, and user fees kicked in another $36 million. In FY 2004, Congress appropriated $1.67 billion, and almost $250 million in user fees were collected. Even after you correctthe appropriated funds for inflation, the budget increase over the last decade is dramatic: $746 million in 1993 dollars translates into $949 million in 2003 dollars. FDA starved, slashed, or squeezed? I don’t think so.

I’m glad you brought up Seldane. I won’t dispute your claim that the agency had superior “mechanisms” in place in 1990 to answer safety questions, but isn’t that merely a way of saying the safety division was better-funded back then?

You mischaracterize Seldane’s decline and departure from the drug market. In 1996, Seldane was the second-most popular allergy drug in the United States. Not bad for a drug that the FDA had been pummeling as “unsafe” since the early ‘90s. You’re right about Claritin overtaking Seldane, but I don’t think health concerns were the main reason. As late as 1997, it was still considered safe enough to be sold by prescription in the United Statesand over the counter in several European countries, including the U.K. I credit Claritin’s rise to a couple of factors. It was a new drug, approved by the FDA in 1993. It was a very effective drug. And it was a heavily marketed drug. Advertising Age reports (March 16, 1998) that $200 million in direct-to-consumer advertising was spent on Claritin between 1993 and 1998. Nor was Claritin the only new antihistamine introduced in the 1990s. Allegra, manufactured by Seldane’s maker, Hoechst Marion Roussel, started digging into Seldane’s market share when it was approved in 1996. According to Advertising Age (July 29, 1996), Hoechst switched its marketing support from Seldane to Allegra that summer in hopes of wresting the No. 1 antihistamine spot back from Claritin.

This Congressional Budget Office study suggests Hoechst started weaning itself from Seldane in 1996 because the drug’s patent protection was set to expire in three years, making it a generic drug. It was in Hoechst’s best interest for Seldane to be banned before it could go generic, especially if Allegra caught on, which it did. Now that Claritin is sold over the counter, Allegra runs neck and neckwith Zyrtec as the most frequently prescribed antihistamine. (See this article in Pharmacy Times and the accompanying chart.)

On your third point, I yield. I wish I could go back and rewrite that one sentence accusing the piece of the sort of “subdued hysteria” that would lead readers to believe fast-tracking has resulted in a flood of potentially dangerous drugs. That assertion is not in the article.

Still, I defy anybody to read the 3,500-word investigation and miss its gist that the Prescription Drug User Fee Act of 1992 is directly responsible for the reduction of monitoring programs, a move which has endangered the nation’s health. That’s the clear message in the story’s first two paragraphs. They read:

When federal drug officials suspected in 1992 that a popular allergy pill might cause heart problems, they turned to their own scientists. Their trial confirmed the danger, and the drug was pulled from the market.

Eight years later, similar worries surrounded the arthritis pill Vioxx. But by then, the Food and Drug Administration had shifted gears, slashing its laboratories and network of independent drug safety experts in favor of hiring more people to approve drugs, changes that arose under an unusual agreement that has left the agency increasingly reliant on and bound by drug company money. Discovering Vioxx’s dangers would take four more years.

I, too, hope readers brew a fresh pot, click the link, and reread the article after reading our exchange. And because I’ve blabbed on so long here, I offer you the opportunity of the last word if you’ve got more to say. I’ll append it below and make sure the Slate cover page runs a “refer” pointing to it. [Addendum, Jan. 10:It proved too unwieldy to append Keller’s last word here, so click this link to read his final word and a comment from reporter Gardiner Harris.]




Did you know that the Prescription Drug User Fee Act of 1992 won the Ford Foundation’s “Innovations in American Government” award in 1997? Send e-mail to and let’s all hope it doesn’t end up in my junk folder. (E-mail may be quoted by name unless the writer stipulates otherwise.)