Gary Schwitzer writes:
Should publications that post op-ed pieces tell us more about who wrote the op-ed and what their potential biases and conflicts of interest might be? . . .
[Transparency advocate Till Bruckner] tweeted criticism of STAT News for not disclosing more about the author of an op-ed criticizing possible Biden administration plans to evaluate drug therapies. The op-ed delivers a fearful prediction:
Patient advocates should scrutinize any “independent review board” the Biden administration commissions because it will have a conflict of interest and a bias against breakthrough therapies. In this scenario, patients may never receive a new, potentially life-changing therapy because it is rated by a biased review board as not having any “quantifiable benefit” based on its price tag alone rather than the potentially life-changing impact that it could have for patients.
That’s a sweeping and fear-mongering allegation – that any review board the new Administration will appoint will have a bias and conflict of interest.
Schwitzer continues:
What bias or conflict of interest does the op-ed author have? STAT didn’t provide that background. Readers are only told that the author is a visiting fellow at the Boston-based Pioneer Institute. . . . Readers were not told that the author, William S. Smith, according to the Institute’s website, “spent ten years at Pfizer Inc as Vice President of Public Affairs and Policy where he was responsible for Pfizer’s corporate strategies for the U.S. policy environment.”
What is the Pioneer Institute? Its website tells you that it is a public policy research think tank whose mission statement endorses free enterprise and free markets. Who funds them? . . .
Bruckner . . . wrote that “it would have been good (for STAT) to disclose that one of the think tank’s top funders is Pfizer, especially as their op-ed mirrors industry talking points 100%.”
And there’s a history:
HealthNewsReview.org [Schwitzer’s website has criticized STAT’s op-ed policies several times over the past 3.5 years:
In response to some of this criticism, STAT revised its guidelines for op-ed submissions. But the latest example begs the question of whether that went far enough.
To be clear, STAT delivers some terrific journalism. But in the interest of continuous quality improvement, it should re-examine its op-ed and broader conflict of interest policies again.
I clicked through to read the Stats article and check for myself but it was blocked by a paywall, so I found it here on the Internet Archive, and indeed, all it says at the end is, “William S. Smith is a visiting fellow in life sciences at the Boston-based Pioneer Institute,” with no acknowledgement of the Pfizer connection.
But Stat News is far from monolithic. I went to their website right now and their top story is Pharma funded more than 2,400 state lawmaker campaigns in 2020, new STAT analysis finds, by Lev Facher. Another recent story is A landmark Alzheimer’s drug approval would likely deepen racial inequities in dementia care, by Usha Lee McFarling.
So it does not look like Stat News has a bias, exactly. Rather, it seems that pharmaceutical reporting, like many other areas of journalism, is saturated with advocates, including public relations professionals. There’s nothing wrong with P.R. people writing op-eds—that’s part of their job! The burden is on the journalistic outlet to look into conflicts of interest, given that efforts are often made to obscure these connections. In this case, the Pioneer Institute’s funding by Pfizer is no secret, but I doubt the typical reader of the op-ed would do the searching needed to discover it.
And not everyone reads the op-ed; some people just read the title.
The op-ed in question was titled, “The Biden administration needs to look beyond ICER for evaluating drug therapies.” A more descriptive title would be something like, “Pfizer spokesman says the Biden administration needs to look beyond ICER for evaluating drug therapies.” Or, “‘The Biden administration needs to look beyond ICER for evaluating drug therapies,’ says Pfizer-funded lobbying group.”
It’s fine to acknowledge the pharma funding: no shame in that! We recently took our Pfizer vaccines. I like Pfizer! Just be open about it.
P.S. There’s some cynical discussion in the comments, along the lines of that everyone has conflict of interest so we shouldn’t consider it at all. I think that argument is ridiculous. Lots of people in our public discourse do not have such conflicts; and there are lots of other people out there who do have conflicts but can still make useful arguments, in which case they can make their conflicts clear and go on to present their case. Again, there’s nothing wrong with a headline such as, “‘The Biden administration needs to look beyond ICER for evaluating drug therapies,’ says Pfizer-funded lobbying group.” There’s no need to be embarrassed about a conflict of interest; just disclose it and we can move on. If you’re trying to hide the conflict, that makes me suspicious.
Translation: taxpayers should pay for patients to get very expensive and probably dangerous moonshot NNT > 100 treatments.
Where do people think the the cutoff should be? 1/1000 chance of benefit?
I’d say odds of 1/2 (NNT = 2) are somewhat reasonable, but that is a very select set of interventions.
Well I think it is too bad no one is actually discussing the merit of the arguments:
The more I think about disclosure of conflicts of interest, in science, investing, politics (e.g., who funds political advertisements etc) and whatever else, the more I think it’s a waste of time if not a destructive distraction.
For example, let’s say someone’s advocating for some investment – Tesla, for example. Who do you trust more: an analysis from someone with nothing on the line, or an analysis from someone that has skin in the game? There are rational reasons to put your trust in either person, assuming all other things are equal. That’s why in the end the trick is: don’t decide who you should trust. Trust isn’t relevant. Make your own decision, independent of trust, based on the data.
We face the same conundrum with academic science vs. industry science. On the one hand, industry has an obvious profit motivation. On the other hand, academia (and regulators) has every motivation to reject every technological advance: they get nothing if it succeeds, and they lose nothing if an important technology is suppressed. But if they advocate **for** a technological advance and it fails in the near term, they risk getting heat. Hence the extreme anti-industry bias in academia. So the perception that academics have no bias is false. They have a strong anti-growth bias because that’s what benefits them. In effect, they’re analogous to the financial advisor with no skin in the game. So when academic and industry scientists are contending over an issue, the trick is: don’t trust. Trust isn’t the way to solve the problem. Make a decision on the data, independent of trust.
Here’s the reality of disclosure: it takes the focus away from whether the thing under discussion is any good or not and puts the focus on who benefits from it and what their supposed motivations are. It’s a great way to avoid the real issue and get into an unresolvable discussion about motivations. And its a great tool for people who aren’t bright enough to evaluate the topic on its merits, but do have the modest capability of picking out which team to throw stones at.
Last but not least even if one must resort to some level of trust, there are obvious signs as to an individual’s competence and trustworthiness that any sensible person should be able to pick out. For example, in this editorial, the person claims that **all** boards or panels **will** be biased. This is clearly false, and a pretty good indication that the writer isn’t fairly evaluating the potential outcomes, regardless of associations. Or, if for example someone is advocating for TSLA stock, while the company has many merits, if the person is claiming that it’s a screaming buy at any price, this is clearly false. No stock is a screaming buy at **any** price, as plenty of not-so-savvy investors have discovered. It’s an indication that the advocate is not providing a fair evaluation
Anon:
I think it’s fine for people who have “skin in the game” to write op-eds. I think it is also a good idea for journalists to identify what that “skin” is. I don’t want to see a man-on-the-street interview that doesn’t mention that the man being interviewed is a political operative.
In summary, your comment makes no sense to me. Disclosure is more information. I don’t buy your claim that disclosure “takes the focus away from whether the thing under discussion is any good or not and puts the focus on who benefits from it and what their supposed motivations are.” I think it’s the opposite: hiding disclosure turns things into a cat-and-mouse game. Let’s get the disclosure out in the open and then move on. If a Pfizer employee or investor wants to make a pro-Pfizer case, let him do so openly and we can evaluate his argument directly without having to first sleuth out what’s going on.
I tend to agree with Andrew here, but Anon has one point that concerns me. Given the complexity of the world, difficulty with statistical matters, and extreme polarization of everything, disclosing the potential biases may well result in people viewing op-eds as mere parroting of positions devoid of any content. I would still favor disclosure, but I think there is a very real possibility that op-eds will just become part of the echo chambers. Perhaps that is an improvement – I tend to ignore op-eds myself. I even ignore any media pieces that are labeled “Analysis,” preferring to look at things purported to be “facts.” But, as our skepticism increases (which it needs to), it isn’t clear whether op-eds should be at all meaningful.
Some people who write op-eds are truly knowledgeable and may have important things to say. I guess that means they should start blogging (in my mind, Twitter has always been out, media “analysis” is out, and now “op-eds” can join the “information sources” I will ignore).
+1 Its healthier to avoid op-eds – and time-saving.
Does the observation to avoid op-eds go back to a conversation between Socrates and Plato and reference earlier philosophers? It seems like it should.
You can avoid op-eds, but can you avoid people informed by op-eds? And people informed by people informed by op-eds?
Individual behaviour just doesn’t scale when you consider the information ecosystem we live in.
《 I don’t want to see a man-on-the-street interview that doesn’t mention that the man being interviewed is a political operative.》
Why not assume everybody is, and then focus on the actual content of what they are saying, not who they are? Why does the term “ad hominem” come to mind?
Rsm:
You ask, why not assume that everybody is a political operative? The answer is, because the vast majority of people aren’t political operatives! It’s kind of silly to assume something that is so far from the truth.
The reason is that much (most?) of the stuff published is not individually fact checked, and the majority of readers are not expected to interrogate the claims and verify them logically and scientifically. Criticism also generally has much less exposure than the source article. Websites like Stat news operate on some degree of trust – trust that a published article is by some kind of authority, that they have some basis to make whatever claims they make, and that they are not acting in bad faith.
The political operative pretending to be a “man-on-the-street” is not the victim of ad hominem because he’s already exploiting a disguise to push a false message (in this case, that a neutral person thinks this without monetary motivation)
“I don’t want to see a man-on-the-street interview that doesn’t mention that the man being interviewed is a political operative.”
Who gives a rat’s bum what one “man-on-the-street” thinks anyway? The MOTS interview is just a tool for journalists to shade the side of the issue they oppose. I read an interesting piece the other day that was written entirely to throw shade on one side of an issue. Guess what? The lone MOTS interview was “OMG, that’s terrible! They shouldn’t be able to do that!” Hilarious. They don’t even give the question they asked, so who knows how it was phrased.
“Disclosure is more information.”
You missed a word: insert “irrelevant” between “more” and “information”.
Let’s take, for example, Big Tech contracting with Department of Defense. How do Big Tech employees feel about this? Does their employer – and hence paycheck – determine their views? Suppose two Google employees present arguments for and against the company contracting with DOD. Please explain to me how their employer influences their views. What you probably think – admittedly or not – is that the anti-DOD argument is more powerful because the person is ostensibly arguing against their own interests. But of course this has nothing to do with the argument itself, it’s just you exposing your own bias.
Here’s your chance to develop a method to distinguish the quantitative amount that a person’s view is influence by their paycheck or what job they do.
Anon:
1. If you find that conflict of interest is irrelevant information, feel free to ignore it. Many people find this information to be relevant. Indeed, sometimes advocates go to some trouble to obscure these conflicts.
2. Maybe there should be no man-on-the-street interviews everywhere. What I’m saying is that, if a political operative is being interviewed, I’d like that to be reported. Similarly, if the newspaper is running an article about Columbia University and they interview a Columbia executive, I’d like that to be reported.
3. How about you don’t go around saying “What you probably think” to me. In your example, yes, if someone is publicly writing about what they think Google should do, and they work for Google, or for a competitor, I think they should share this conflict of interest. It’s not a statement that an argument is “more powerful”—I don’t know what that means. I’m funded both by Big Tech and the Department of Defense, so I’m conflicted in both directions, and I think it’s good for this information to be available.
In any case, I guess that, in this respect, the world of journalism is pretty much what you want it to be. News organizations run press releases without reporting the source of the claims, they interview conflicted people without disclosing the conflicts, etc.—it happens all the time. And then people like Till Bruckner, Gary Schwitzer, and I will complain about it. Fortunately for you, lots of news organizations will continue to not reveal conflicts of interest in the people they’re quoting, so often you will not be bothered by this information that you consider to be irrelevant.
“Here’s the reality of disclosure: it takes the focus away from whether the thing under discussion is any good or not and puts the focus on who benefits from it and what their supposed motivations are. It’s a great way to avoid the real issue and get into an unresolvable discussion about motivations.”
It is not a way to avoid the “real issue” unless you make it so. If your point is that the argument should be considered anyway, no one is disagreeing.
Have you never encountered a disingenuous argument? What tools do you have in your skeptic’s toolbox to ferret them out?
I suspect that you are just trolling, again. I bet that you take conflicts of interest into account every day and respond to them just like anyone else. If you need to set down your hamburger, why not let the dog hold it?
: )
“Have you never encountered a disingenuous argument? What tools do you have in your skeptic’s toolbox to ferret them out?”
Data and reason. What tools do you use?
:)
That would be alternative data and zero-th order reasoning? Good tools for political operatives with scant regard for social norms, but not for others, I think.
I’m struggling to find a context where that makes sense. In my experience (biomolecular/biomedical) academic scientists welcome technological advances and often work in collaboration with industrial scientists to develop technologies. The advances in NMR/MRI and Cryoelectron Microscopy were made as a result of academic scientists embracing tecnological breakthroughs designed and manufactured in industrial setting by incorporating design features made through discussion with academic (amongst other) end users. Novel technologies are embraced by academic scientists since they recognize that their work can only be enhanced by incorporating these into research efforts. It’s not obvious anyway that there are two seperate entities as “academic science” and “industry science” since science is science and there is continual back and forth flow of information and methodologies between academic groups and industry groups. I’m not seeing this “extreme anti-industry bias” at all! Maybe you could give some examples.
One example might be the race to determine the human gene sequence that started out with some conflict between industry and public science but ended up as a collaboration (or at least was presented as such in the interests of diplomacy!). But I wouldn’t call that an example of “extreme anti-industry bias” on the part of academic scientists but only a recognition that it would be good if the human genome sequence was publicly accessible and not subject to monetisation. In any case there was no way that anyone was trying to “suppress” an “important technology”.
…and “strong anti-growth bias”???
Exactly. Maybe because my department is in an engineering school, that bit about “extreme anti-industry bias” sounded like it came from another planet. But same for my wife’s stats dept in an LAS school. Or my dad’s chem dept in an LAS school. Maybe opposing dumping toxins in water supplies counts as “extreme anti-industry bias”? Or maybe it’s just about the usual, marginal tax rates.
‘“extreme anti-industry bias” sounded like it came from another planet. ‘
Michael:
“academia” “chemistry” (or “engineering”, or “biomed”). If you look up the number of degrees granted, chemistry doesnt even register. I guess it’s included with “Physical Science and Science Technologies” – as a group producing less than one-tenth the number of business degrees. Engineering and bioscience together roughly equal business, but together still comprise fewer degrees than “social science and history” and “health professions”.
Having myself come from a sci/tech school with strong connections to industry, I’m aware that there are pockets of pro-industry sentiment in academia. But even in the sciences it’s hardly the majority of people.
“In my experience (biomolecular/biomedical) academic scientists welcome technological advances and often work in collaboration with industrial scientists to develop technologies.”
Yes, of course that’s true. But as I pointed out to Michael below, “biomed” “academia”. What percentage of faculty work in these disciplines? 2019-2020 data indicates biomed is 8% of top ten degrees, significantly less of all degrees. It’s not academia, not even close.
But you’re employing the same fallacy that I argued against above to claim that biomed is “pro-growth”. The fact that people work in the pharma industry doesn’t make them pro-growth on a majority of economic issues. It doesn’t even make them “propharma”. Thousands of Google employees protested against Google providing services to Department of Defense. What – are they lying?
Independent of issues with conflict of interest disclosure, the “First Opinion” section of Stat News has some fairly poor quality articles.
This is probably the most memorable one for me – “An Omicron oddity: The number of cases doesn’t predict the number of deaths” – https://www.statnews.com/2021/12/22/omicron-oddity-case-numbers-dont-predict-deaths/
It plots daily cases (including negative ones) and daily deaths and draws lines through them and concludes “at the moment in the U.K., Covid-19 daily cases no longer meaningfully link to deaths.”
There *is* a fairly clear disclosure at the bottom about the authors and their interests so that’s good!
Do we have to disclose every place that we “used to” work? I have worked for pharma in the past. So what? I don’t work there now. Also, there are groups who give my employer money. Good! But they are not giving it to me based on my writings. How far do these rectal disclosures have to go?
Rodney:
As with most things there is no sharp line. I think that “spent ten years at Pfizer Inc as Vice President of Public Affairs and Policy where he was responsible for Pfizer’s corporate strategies for the U.S. policy environment” is a big enough deal that it should be disclosed.
Put yourself on the witness stand and imagine the cross-examination. If it jeopardize your standing (perception as well as legally) as an expert witness, then it needs to go that far.