A few weeks ago, Frank Rich wrote a sharp column on the Clinton campaign’s failure to see the danger of peddling her Bosnia bullet-ducking story. Rich argued that at a time when the new, million-headed media made a YouTube video contradiction of her tale almost inevitable, Clinton had failed to see that a politician could not get away with myth-making autobiographical fibs of the sort that would fly only four years earlier. “A new bottom-up culture,” he argued, “is challenging any candidate’s control of a message.”
Rich’s two-month-old column jumped to mind a couple of weeks ago when I came across “Stealth Marketers,” an article in Slate by Shannon Brownlee and Jeanne Lenzer about pharmaceutical industry influence on journalism. Acting on a tip from a long-time source (this exposé’s mechanisms were more Woodward & Bernstein than Web 2.0), Brownlee and Lenzer revealed that The Infinite Mind, a well-regarded and independently produced radio series distributed to over 300 public stations, had failed to disclose potential conflicts of interest during an hour-long show in March called “Prozac Nation: Revisited,” which portrayed media coverage of the risks of violence or suicide from antidepressant use as overblown. All four experts featured on the show have or have had financial ties to antidepressant makers and The Infinite Mind has itself taken funding from the Lilly Foundation, which is tied to pharmaceutical giant Eli Lilly, the maker of Prozac.
The show’s producer, Bill Lichtenstein, immediately went on the defensive. In posts on Slate and in several conversations with me, he explained that in consultation with National Public Radio (which airs The Infinite Mind on one of its Sirius satellite radio channels - and for an interesting trip down the rabbit hole of what qualifies as an “NPR Show” click here), he created guidelines in 1994 designed to limit any industry-related funding and influence of his production company, Lichtenstein Creative Media. His measures seem sensible enough for that time, and Lichtenstein appears to be a well-intentioned and highly capable producer. Yet his aggressive counterattack to the Slate article seems as tone deaf, or perhaps time deaf, as the Clinton campaign’s allegiance to the Bosnia airport story.
For starters, some of Lichtenstein’s counterpunches had a below-the-belt quality. In a response that Slate ran just below to the original article and in the magazine’s reader discussion forum, he suggested that Lenzer and Brownlee wrote their article after he had rejected a pitch from Lenzer to follow up the Infinite Mind show on Prozac with a report on pharma’s influence on journalism. The implication, obviously, is that their Slate piece amounted to revenge. In a retort also posted beneath the story and in the readers forum, Lenzer and Brownlee say, no, that actually it was Lichtenstein who raised the idea of a radio piece when Lenzer contacted him to ask about the undisclosed financial ties in “Prozac Nation: Revisited.”
Brownlee and Lenzer have copies of an email exchange consistent with (if not proving) their account, and it seems a stretch to imagine they needed a pitch rejection to motivate their criticism of The Infinite Mind. Besides, Lichtenstein doesn’t win much sympathy or credibility with another charge in his Slate reply, which is that the British Medical Journal had retracted a 2005 article by Lenzer in which she mistakenly wrote that documents from Eli Lilly had gone “missing” during a 1994 product liability suit. In fact, the BMJ retracted only that statement, not the whole article, as Lichenstein states (a very big difference). Furthermore, given that the journal has continued to publish Lenzer’s work, Lichenstein’s allegation seems even less relevant.
The biggest problem with his response, however, is that it all but ignores the well-documented effects that financial ties to pharmaceutical companies can have on medical opinion, and how central such influence has become to discussions of drug safety. The issue of whether antidepressants increase the risk of suicide is joined at the hip with the issue of whether the drug industry’s full-court “information” campaigns, along with the selectivity with which the industry has released study data, have made it impossible to fairly evaluate suicide and other risks and side effects associated with antidepressant use.
“Prozac Nation: Revisited” ignored this completely. The show never mentioned that industry meddling has compromised everything from the information database to public trust. It focused almost exclusively on industry-friendly views (however well-informed) without revealing that the show’s host, Dr. Fred Goodwin, and two other guests, Drs. Andrew F. Leuchter and Nada Stotland, had or have had consulting, speakers bureau, and/or research grant ties to antidepressant makers. Most egregiously, The Infinite Mind introduced the fourth expert, Peter Pitts, only as a “former FDA commissioner” and not by his present job title: vice-president for the public relations firm Manning, Selvage & Lee, which represents Eli Lilly in addition to more than a dozen other pharmaceutical companies.
In his response to Slate, Lichtenstein acknowledged that Infinite Mind should have noted Pitts’ association, but defended the show by saying that Pitts did not disclose that information. Meanwhile, Pitts - the only involved party to grant Slate an interview - said he revealed his industry ties and told Brownlee and Lenzer he was “surprised” they weren’t mentioned. Regarding the funding the show has received from Eli Lilly as recently as 2006, Lichtenstein defended himself by writing that:
It would be hard to find support from organizations without some kind of substantial interest in the subject matter. The important thing … was to assure listeners and stations that there was an absolute firewall between funding sources and editorial decision-making.
The problem is that Infinite Mind did not make those assurances about its relationship with Eli Lilly before airing the show about Prozac. After the fact, Lichtenstein laid out a series of its existing funding guidelines, including the provision that his production company would accept no more than 15 percent its budget from “any one industry sector.” In the wake of the controversy, and as recommended by NPR’s ombudsman, his production company also added a page on its Web site explaining its underwriting policies and one clarifying Pitts’ background.
At one point in my conversations with him-actually, at several points-Lichtenstein offered variations of the “we’re all conflicted” argument. I suppose he has a point. We all have interests. We have people we like, people we want to please, people or institutions that have paid us in the past, or pay us now, or might pay us in the future. I encountered such a situation early in this story when I realized that Brownlee is a senior fellow at the New American Foundation, where I am currently applying for a fellowship.
This hoped-for affiliation with the foundation that supports Brownlee is a clear interest for me-one that some might think creates at least the perception of a conflict of interest for me when assessing an article written by Brownlee. I should also disclose that in 2005 I wrote six stories for Slate, earning about $3,000. Perhaps also that my parents were both doctors; that I admire good doctors, and that I am sincerely grateful that drugs ranging from antibiotics to antidepressants have saved the lives of people enormously dear to me.
So, yes, we all have interests. But I think we also realize that there’s a difference between the particularized, often indirect, and piecemeal interest that most of us encounter in our daily lives and the kind of heavily financed and highly sophisticated interest between funder and fundee, and in particular, the kind of interest quite effectively and intentionally generated by the “information” programs of industries like Big Pharma. It’s not that such campaigns corrupt every doctor they touch. But the covert nature of so many of them has clearly corrupted public trust in drug safety information. If journalists like Lichtenstein want the information they present to the public to be taken as credible, they need to err on the side of transparency, presenting not only the voices, but also the relevant financial interests of the experts they feature. Failing to do so only damages message and messenger alike. But in the wake of the repeated scandals about drug-company concealment of drug-trial data, it’s strange that I have to spell this out.
I asked Lichtenstein if he would be changing his show’s disclosure policies. He said the show would not invite guests with significant conflicts of interest, and if there were a compelling reason to include someone with a vested interest, he would disclose that on air and on the show’s Web site. But he also said this was not a change in policy, and that while he had simply blown it in the case of Pitts the PR man, he would not, if he were doing this over again, reveal the economic ties of the others on “Prozac Nation: Revisited.” He should.
Over the last five years or so, peer-reviewed medical journals, recognizing the danger for conflicts of interest generated by pharma funding, have mandated that authors declare all financial interests related to their research-grants, speaking and consulting contracts, patents, etc. This is a changing landscape. Brownlee told me she didn’t consider such issues when writing about medical issues five or ten years ago, and I must admit I did not either. But it’s clear, at least to most of us, that times have changed.





I find myself deeply troubled by this article, and the reason is, I suspect, linked to why most of my friends in the medical profession despise journalists and the news media. Now let me declare my affiliations first: I've written about the ethics of this dispute at STATS.org, where I found both sides culpable, but that Slate's authors were guilty of poorer journalistic ethics. How can this be so?
First, "The Infinite Mind" should have disclosed the associations and should have given some nod to the reality of adverse reactions to antidepressants; that being said, the program addressed the reality that the black box warning on suicidal ideation, and the media's hyping of the story, led to more people actually committing suicide.
Ms Lenzer and Brownlee are simply uninterested in this fact; for them pharma money corrupts, so much so that they created a list of supposedly uncorrupted scientists. This is not only ethically problematic (in the sense that people who may need antidepressants may be convinced by their anti-pharma jeremiad into forgoing medication) but it is simply unrealistic to think that medicine can function without industry support.
Think of it this way, their rubric for excluding money-tainted "experts" would rule out most if not all practicing neurosurgeons who work on spinal injuries in the U.S. Why? Because this field relies on manufacturers to work in a symbiotic way to create the parts needed to rebuild a spinal column. Many surgeries are even performed with a rep from a medical equipment manufacturer present to take feedback as the operation is in progress. The result is that people today can beat terrible diseases and accidents.
The point is that there's good data and bad data. Focusing on who funds what is a get out from having to do the hard work of discovering what makes study x good or bad. Good studies - massive multigenerational studies - are enormously expensive, but there are objective criteria for determining whether their results are reliable.
For journalists to simply forgo that obligation to back, no matter how slight, the insinuation of corruption with data is just rotten journalism. Ms Brownlee and Lenzer appear to have determined a bias and miss their own in the process. What happened with Ms Lenzer and the British Medical Journal (a retraction not a mere correction) show the power of that bias in journalism - and how it ill serves the public.
Posted by Trevor Butterworth
on Wed 21 May 2008 at 02:27 PM
One more thing, simply disclosing that you're applying for a fellowship at an organization where one of the principals in your story happens to be a senior figure, and then congratulating her on her work is just a tiny bit rich, don't you think?
Posted by Trevor Butterworth
on Wed 21 May 2008 at 02:51 PM
Trevor Butterworth says he is "deeply troubled" by my article, which he suggests is of the sort "despised by" the medical profession. Yet his comment scarcely addresses my article at all; he uses the space instead to attack Brownlee and Lenzer, to toss out complaints that sidestep the article's main subject, which is the need for transparency in declaring relevant financial or other potentially conflicting ties in quoted medical experts, and finally, to take a potshot at me.
Butterworth worries that "[Brownlee and Lenzer's] rubric for excluding money-tainted 'experts' would rule out most if not all practicing neurosurgeons who work on spinal injuries in the U.S." I can't speak for Brownlee and Lenzer, but -- if I can return to the article on which he is supposedly commenting -- I am not proposing that doctors or researchers with financial ties to industry be excluded from press stories or public discussion of medicine. That would be destructive. I am arguing that the relevant financial ties of quoted experts should be disclosed so that the public can consider those in considering their positions. Butterworth appears averse to this idea. But it's one that the leading medical journals have embraced, and it seems to have raised rather than eroded faith in the studies published in those journals.
Butterworth also says, "The point is there's good data and bad data." Well, yes, but as every alert researcher and student of science knows, it's not quite that simple. There are also many types of bias in data reporting, as is well-documented in many studies. And the selective reporting of data about drug risks, including the risks of suicide and suicidality examined in "Prozac Nation: Revisited," has created great doubt among many psychiatrists about the integrity of that data. The media coverage of these risks has indeed often been sloppy and sensationalistic, as "Prozac Nation: Revisted" reported. But the the psychiatric and research communities are waging a vigorous scientific debate over the integrity of the risk and side-effect data and struggling to figure out out to improve it. Smart, well-informed, reasonable people who know the field disagree on the integrity of that data.
One is University of California at San Francisco Medical School professor and Journal of the American Medical Association deputy editor Dr. Drummond Rennie. In 2004, when I was doing a story about the SSRI-youth suicide controversy (see http://daviddobbs.net/page2/page11/page11.html), Rennie, referring to a practice of selective study release that drug companies routinely followed with FDA knowledge, "If a company does ten trials on a drug and two show it helps but eight show it works no better than Rice Krispies, I'm not exactly getting a scientific view if they publish only the two positive studies.... [H] can we practice sophisticated medicine if the drug companies are hiding their results? That's not science. That's marketing." FDA rules and journal policies have sharply reduced that practice since then. But the database, and the faith of both the public and the medical profession in it, remains compromised. This makes it all the more sensible to reveal financial ties to industry. The public wants to trust its doctors and experts. That trust is harmed when ties to industry are hidden. Disclosing ties increases that trust.
Finally, I am ALMOST amused at Butterworth's cheap shot at me for being "a tiny bit rich" in disclosing that I am applying for a fellowship at the New America Foundation, where Brownlee is a senior fellow. This confuses me: Just when Butterworth had me thinking that financial interests real, past, or potential do NOT corrupt opinion, he seems feel quite certain that my own potential such interest does corrupt. I should explain -- this was in the original story but was condensed for space's sake -- that "Finish application to New American Foundation" has been on my to-do list for several months, so I can assure readers that both the intended application far precedes this story, as does my interest in conflict-of-interest and data-integrity issues (see story link above) and the nature of empiricism as a foundation of good science. Does my still-incomplete application to a fellowship at New American Foundation override those other interests and fire a mercenary interest in this subject? I don't think so. But in declaring the potential conflict I leave it to readers to judge for themselves.
In the meantime Butterworth should take his cheap shots a bit more judiciously. Butterworth's characterization of Brownlee's position at New America Foundation as a "senior figure" is itself a tiny bit rich. Brownlee is a senior fellow (one of a couple dozen or so), but if the NAF (about which I know very little) works like most places, that just means she gets more money than junior fellows, not that she's high on the decision-making chain and passes judgment on fellowship applications, as Butterworth's use of the term "senior figure" would suggest. And while I called the Slate piece a piece of "nifty reporting" in my blog, nowhere in the CRJ story did I "congratulate" Ms. Brownlee. Let's try to get the data right.
Posted by DaveD
on Thu 22 May 2008 at 06:44 AM