This morning, the New York Times’s Virginia Heffernan, who writes a column that is basically on the theme of what to make of modern media (and who, btw, I generally quite like and read regularly) offered an overly simplistic assessment that basically said, trust the Mayo Clinic’s website, not WebMD.
The analysis includes the most common misguided version of “follow the money”: WebMD gets money from Big Pharma, while the Mayo Clinic is a “nonprofit medical-practice-and-research group that started as a clinic” and has a “storied past as the country’s premier research hospital, in Rochester, Minn., and its storied present as one of Fortune’s ‘100 Best Companies to Work For'”. Setting aside the irrelevancy to ethics and honesty of being a good place to work (I am sure Goldman Sachs employees are quite delighted with their employer), major research hospitals and those affiliated with them make a mint in pharma money. So both of these entities, like most other purveyors of information, face the question of how they are going to trade off reporting what they think is right against pleasing those who help pay their bills.
But even if there were such an asymmetry in funding as Heffernan implies, it still does not tell us who to trust. It is naive, though common, to think that corporations (or those who work closely with them) are unique in having interests that may run contrary to telling the unbiased truth (using “truth” as a simple shorthand for purposes of communication – I am not going into the question of whether there is such a thing existentially). If you think about it, most not-for-profit entities exist explicitly for the purpose of advocating something, and so have worldly interests that are fairly similar to those of corporations. With regard to Mayo Clinic specifically, I have documented how they have actively lied to the public to try to discourage tobacco harm reduction. They have been doing so since I started documenting anti-THR disinformation campaigns (e.g., here) and are still doing it. This is not something that falls into the realm of honest mistakes – they are clearly engaged in a marketing effort every bit as aggressive and narrowly-interested as any corporate marketing effort. (Indeed, they would be legally forbidden from publishing the falsehoods they have written if they were doing it to sell a pharmaceutical product.)
I suspect, based on knowing a great deal about this one topic, that there are ten or a hundred other areas where Mayo are such aggressive advocates of a particular view that their information is no less biased than corporate marketing. Even apart from narrow special interests, there is the issue of broad special interests: Mayo and its employees have interests in supporting the current regime in health care and health research. These interests can influence all manner of what they choose to communicate.
Heffernan also tries to suggest that Mayo is more concerned with protecting its reputation, by not producing bad information, than is WebMD, but the better analysis seems closer to the opposite. WebMD has nothing other than its reputation for providing useful information on its website; its entire existence depends on it. It may well be, as Heffernan argues, that it has blown its credibility so badly that everyone should “block” it (not sure exactly what that means – some web trick I do not know, I guess). But if that is true, and people take such advice, WebMD is doomed. By contrast, the likes of Mayo, universities, and the government create “truth” by their actions and are thoroughly anchored such that they do not have to fear important repercussions from a damaged reputation. Thus they can distribute misinformation with impunity. I am fairly confident in estimating that the U.S. government provides more misinformation to support its own worldly interests than any other source of English language information, but is still considered the most authoritative source of health information. Indeed, that government might provide more misinformation about health specifically than any other Anglophone source, and its health claims are still treated as fact.
If you are going to try to learn something about the credibility of health information by considering the worldly interests of the author (or their employer), looking only for corporate interests is bound to be a failing strategy. Also, once you do identify someone’s worldly interests (corporate or otherwise), it takes a bit of thinking to figure out which of their claims might be affected. Simply thinking “they have a worldly interest in this topic area so their conclusions are suspect” is just as overly simplistic as saying only corporations and those they fund have worldly interests. I will expand on that point in a post on some later weekend.
In the Mayo vs. WebMD comparison, Heffernan does offer some valid substantive analysis. Searching on the subject of headaches, she observes that WebMD is likely to drive you toward hypochondria (the major theme of the column) because they emphasize nastier drug-requiring causes of headache, while Mayo’s article on the topic emphasizes common tension headaches. This comparison tells you far more about their relative credibility than any issue of funding, for-profit status, desire for credibility, or being a “best place to work”. But what it means is that a physician organization is good at the “dial a nurse”-style communication of generic clinical advice. This is unsurprising good news, but physicians have their limits. They are not particularly good at assessing research and new information, let alone dealing with unusual situations (Greg House notwithstanding) or sorting out controversial subjects. I will grant that the Mayo Clinic’s website is not as rich a source of unhealthful news as either WebMD or Heffernan’s own New York Times, but that is because they do not try to report so much news. I am still pretty confident I could, if I wanted, write good additions to this series for weeks based on what I found at Mayo’s site.
Since I was recently reminded of the value of periodic quizzes/tests as an educational tool, I will conclude with the following:
You can trust what I am saying here is accurate to the best of my ability because:
(a) I must be telling the truth because no one is paying me to write this.
(b) Sitting in my living room while watching the Superbowl is a great place to work.
(c) I am a research hospital.
(d) I am entirely dependent on my reputation for offering useful honest information.
(Quiz answers will be provided at the conclusion of the series.)
I will try the web trick and 'block' websites from now on. Not quite a black or white issue.