N of One, part II

A.
Self magazine--which advertises itself as "the first-ever magazine of total well-being, merging beauty and health, fitness and nutrition, and happiness and personal style together in one package...for the woman who wants to stay informed, get inspired, grow and achieve her personal goals"--recently ran an in-depth story about generic medicines entitled "Bad Bargain." The subtitle captures the essence of the piece: All of us want cheaper medicine—but not if it costs us our health. Troubling reactions and a series of recalls are making some doctors wonder, Are generic drugs as safe as the FDA says they are? SELF investigates.

I am not a journalist but I'll give the piece a generally good grade for doing in-depth research, taking great pains to explain in detail how the generic medicine industry is regulated, and even devoting a few paragraphs to those who are skeptical about the claims that generics aren't safe. As articles on medicine and industry go I've seen and read a lot worse. But despite some impressive homework by the Self staff they still make at least one and possibly two major errors of contextualization that, from my standpoint, casts doubt on their claim that generics are scary. Boo.

The first problem, the more minor of the two, shows up at the beginning of the article, when Self describes the medical odyssey of Beth Hubbard, a woman whose life came undone when she was switched from the brand-name anti-depressant Wellbutrin to the generic Budeprion. The article devotes many paragraphs to Hubbard's story (nearly 20 percent of the article), and her discovery that the switch to generic was the likely cause of her decline. During her trials she ended up being prescribed a large number of other medications (more on this anon), but eventually sorted out that the problem lay somehow in the generic formulation, and Self moves on from there to discuss the problem of generics.

Health journalists love using this particular conceit, which is a classic "N of One" scenario. If a journalist looks hard enough, he or she will find all number of strange tales of odd reactions and failures to respond to medications. How much can one conclude from such isolated stories? Not much. The important question is whether or not such a story is representative of something larger. In this case Self at least tries to use the story as an entree to discuss larger trends, and while it makes a few key points to support its claim I remain skeptical that they've demonstrated that this is a massive systemic problem. You be the judge; I'd love to hear feedback.

The second and much more troubling problem is that they pluck the issue of generic medicine regulation out of context and discuss it as if there are no problems with brand-name drugs. The article clearly implies that buying generics is more dangerous than buying brand-name drugs, yet does no comparison of the relative safety of brand-name versus generic drugs (that could be done easily enough). They note that the part of the FDA charged with oversight of generics is overwhelmed and understaffed--a claim that seems all too plausible to me--but don't bother to point out that the same situation applies to the FDA as a whole, especially including regulation of patented and new drugs. In fact, in one of the richest bits in the article, the writer notes that as part of Ms. Hubbard's trials, she took the drug Zelnorm to help cure her symptoms. Zelnorm, made by Novartis Pharmaceuticals, was approved in 2002 to treat Irritable Bowel Syndrome (about which much more could be written, but not today), and not long after approval advertisements on television started popping up showing women revealing their happy midriffs to the world, gleeful with satisfaction from being relieved of their IBS. Alas, the drug was removed from the market in March 2007 at the FDA's request because "data from 29 clinical studies with more than 18,000 patients showed that patients taking Zelnorm were eight times as likely to have heart attacks and strokes compared with those taking placebos." That's your brand-name industry working for you! At its height, Zelnorm prescriptions numbered two million annually. Not so great if you're a 30-something lady suffering from intermittent constipation and suddenly find yourself suffering from a heart attack.

Might there be real problems with the regulation of generics? Indeed, and Self does a reasonably good job for making that case. Might there be real problems with the regulation of all medications, thus making the attack on generics seem weird? I think so. Might the fact that brand-name drug makers advertise heavily in television and print media, while generic manufacturers advertise much less, have some influence on this asymmetric attack on generics? Hmmm.

B.
For my part, I think that if you had to boil down all useful political punditry, you couldn't do much better than being a regular reader of just two writers: Paul Krugman and Bob Somersby of The Daily Howler. Paul, the more famous of the two due to his perch at the NY Times and that prize he got in Stockholm recently, uses his pulpit not only to discuss economics in a manner intelligible to the educated layperson, but also delves into various political topics. A few weeks back he did the latter, and to my mind made a fairly egregious N of One type mistake.

After the shooting of Dr. George Tiller in Witchita, Kansas, Krugman wrote an article entitled "The Big Hate," in which he provocatively asserted "right-wing extremism is being systematically fed by the conservative media and political establishment." Krugman noted that the shooting at the Holocaust Memorial Museum by white supremacist/anti-Semite James W. von Brunn, in addition to the Tiller killing, represented a definite trend that had been predicted by an internal report for the Office of Homeland Security--a report that was met with howling indignation by conservatives.

Do these two examples show that the Homeland Security report was right? Maybe. One counter example that pops a bit of a hole in the right-wing violence theory is the "left-wing" killing of Private Quinton Ezeagwula in Little Rock AK by "Muslim convert" Abdul Hakim Mujahid Muhammad earlier this month. (Unclear if "left-wing" is a useful designation here, but Muhammad's actions are certainly motivated by political beliefs very different from those of the other shooters, and as such doesn't really fit into the scenarios discussed by the Homeland Security report.)

Is Krugman accurate in his assertion that the conservative media feed the kind of hatred that results in such killings? I think yes--although I think he mixes his examples and ends up with a sticky argument. Sometimes an N of One makes a certain amount of sense, if and only if you can demonstrate it to be the end result of a causal chain. Can one trace the killing of Dr. Tiller to the wide exposure he received on Bill O'Reilly's show, where he was frequently called out by name as a "baby-killer" and one who ran a "death mill"? Definitely maybe--at least it's more than reasonable to see if one can trace a line from O'Reilly to the shooter. (Politifact highlights the number of times that O'Reilly singled out Tiller for abuse here.) But can von Brunn's actions be lumped in with this, and are they attributable to the increasingly violent rhetoric of the right-wing media? I think not; von Brunn was a nut, and his hatred of any form of "establishment" media, whether right or left, was uniform. He threatened conservative publications as much as liberal ones. This was no O'Reilly fan, and while O'Reilly and his ilk certainly should be taken to task for the light and heat of their rhetoric, they can't be blamed for von Brunn's actions.
--br