Journalism and Astroturfing

by Henry Farrell on March 27, 2014

Back when Nick Confessore broke the Tech Central Station scandal, another journalist wrote a “very good piece”:–98649 about the problems that you got when journalism merged into astroturfing.

bq. For years — literally years — I’ve been writing about Astroturf organizing and that trendsetting operation in the trade, DCI — home of that Johny Appleseed of the plastic and the green, Tom Synhorst. Simply put, Astroturf organizers are in the business of creating phony grassroots support, or rather the appearance of grassroots support, for this or that cause. You got the money and the cause? They’ll bring the front groups, the push-polls, the oped payola, you name it. …The secret of ‘turf is a simple one. Advertisements and paid spokesman may influence us to some degree. We hear their opinions, see them on TV and such. But because they’re paid, because they’re essentially advertisements, we also tend to tune them out, or at least bracket them off in our minds. … For years, the trendsetter in Astroturf has been DCI. And a couple days ago, if you were watching really closely, a tiny sentence changed on an out-of-the-way page on the TechCentralStation website. The sentence that read … “Tech Central Station is published by Tech Central Station, L.L.C.” now reads … “Tech Central Station is published by DCI Group, L.L.C.” It wasn’t an accident. It was because this article — ‘Meet the Press’ by Nick Confessore — was about to be published by The Washington Monthly.

That journalist was Joshua Micah Marshall of Talking Points Memo. Today, as part of its “very cool new section … which is being sponsored by the Pharmaceutical Research and Manufacturers of America,” Talking Points Memo published this piece about “the data sharing effort to cure cancer.”

bq. Imagine a world where the cures for diseases are discovered with the speed of software development. … Recognizing that mortality rates for cancer have not improved in 40 years and that the research to find cures is too slow, the CEO Roundtable on Cancer’s Life Sciences Consortium, a group of leaders from life science organizations, have created a data-sharing platform resembling open source tech sites. Known as The Project Data Sphere initiative, the platform is designed to provide one place where the research community can broadly share, integrate, and analyze historical, patient-level cancer phase III comparator arm data from commercial and academic organizations. … However, not all data is created equal. Data shared between developers is rarely a threat to individual rights, but sharing clinical trial data runs the serious risk of violating the privacy of a person’s health. Last summer, the Pharmaceutical Research and Manufacturers of American (PhRMA) and the European Federation of Pharmaceutical Industries and Associations (EFPIA) adopted joint Principles for Responsible Clinical Trial Data Sharing.

Sounds kind of awesome, right? After all, cancer doesn’t exactly have many fans. The problem is that the piece is, under the very kindest interpretation, a little unforthcoming about the actual politics behind behind the Project Data Sphere Initiative and the Joint Principles for Responsible Clinical Trial Data Sharing. Under a less kind interpretation, the article might be read as a deliberate effort to provide covering fire for just the kind of astroturfing exercise that Josh Marshall used to go after. This “Guardian article”: provides a good backgrounder on the politics behind the Initiative and Joint Principles.

bq. The pharmaceutical industry has “mobilised” an army of patient groups to lobby against plans to force companies to publish secret documents on drugs trials. … Under proposals being thrashed out in Europe, drugs companies would be compelled to release all of their data, including results that show drugs do not work or cause dangerous side-effects. … The strategy was drawn up by two large trade groups, the Pharmaceutical Research and Manufacturers of America (PhRMA) and the European Federation of Pharmaceutical Industries and Associations (EFPIA), and outlined in a memo to senior industry figures this month, according to an email seen by the Guardian. … The email describes a four-pronged campaign that starts with “mobilising patient groups to express concern about the risk to public health by non-scientific re-use of data”. Translated, that means patient groups go into bat for the industry by raising fears that if full results from drug trials are published, the information might be misinterpreted and cause a health scare. The lobbying is targeted at Europe where the European Medicines Agency (EMA) wants to publish all of the clinical study reports that companies have filed, and where negotiations around the clinical trials directive could force drug companies to publish all clinical trial results in a public database. … A recent review of medical research estimated that only half of all clinical trials were published in full, and that positive results were twice as likely to be published than negative ones. A source in the European parliament, who is close to the negotiations over the clinical trials directive, said he had experienced intense lobbying from patient groups.

bq. The EFPIA told the Guardian it had been working with PhRMA on a “commitment to enhance sharing of clinical data” to researchers and the public, and intended to make an announcement this week. “Knowing that some people want all data to be made available to everyone, EFPIA is engaging with stakeholders to share concerns with harmful ‘re-use’ of data. We will engage not only with patient groups, but also with the scientific community,” it said.

This legislation is up for a vote in the European Parliament next week. It’s hard – no, actually, it’s impossible – not to see this piece as part of that broader lobbying effort. It is specifically touting the two key elements of Big Pharma’s proposed alternative to actual binding legislation, which form part of a much larger campaign involving flacking and astroturfing on a massive scale. And Talking Points Memo – former foe of astroturfing in all its forms – is part of it.

I don’t think that Josh Marshall is a bad actor – from my limited online interactions with him, I’ve found him to be decent and honest. I would guess – while knowing absolutely nothing about TPM’s internal decisions – that this move is the product of need, not choice. Again, TPM is revealing PhRMA’s sponsorship (and the post in question has author credit given to ‘PhRMA.’ None of this changes the fact that this sponsored section goes against all the principles that made Talking Points Memo attractive and admirable in its heyday. It’s particularly egregious that this is part of a lobbying effort to prevent pharmaceutical companies from having to disclose publicly valuable information that would reveal their wrongdoing. You simply can’t run an investigative journalism outfit with your right hand, and take money from lobbyists who want to shut down information disclosure with your left one. And I hope that Josh Marshall figures this out, sooner rather than later. [HF – post lightly edited to remove ambiguities]



Josh Marshall 03.27.14 at 9:02 pm

For those who are interested, I responded in Henry’s first post on this topic. The link is here.


Dominick 03.27.14 at 10:35 pm


Would you consider publishing a post that directly takes issue with the characterization of PhRMA’s activities in their sponsored post on the subject? Would that violate your agreement with PhRMA? If such a post were to violate your agreement, it would seem that Henry has a point here. You are taking PhRMA’s money in exchange for not investigating their practices AND promoting their lobbying efforts.

If not, and you were perfectly able and willing to publish content that contradicts PhRMA’s claims (assuming it met your other editorial criteria) then you seem to be on much firmer ground.


Andrew Burday 03.27.14 at 10:56 pm

The link in “another journalist wrote a very good piece” appears to be broken, unless you’re doing some ironic self-referential thing that is going over my head.


Kevin Donoghue 03.28.14 at 12:01 am

Quick Googling suggests the correct link is:–98649

Feel free to delete this comment when the link is fixed.


Main Street Muse 03.28.14 at 12:03 am

That’s the bitch about capitalism – one must find funders for all ventures.

What journalism outlets are free of the pressures of sponsors? Thank Ben Day, the founder of the penny press. Cost of news production far exceeds the costs we pay for a news source (paper, digital, radio, TV.) There’s been a golden tether between news and advertisers for more than 150 years. (Cannot bring myself to use AP Style’s new rule on “over = more than!)

Sponsored content is different than ad-supported journalism. Sponsored content is when the sponsor calls all the shots about the content. TMP says this is not the case. However, it’s clear that the Data Sharing story is not written by a reporter – and there is a disclaimer that calls this “sponsored content.” If you click on the “what is this?” button, it states that TPM editorial staff did not participate in the development of this content. So I guess taking PhRMA’s money means you have to take their “news story” ads too.

Here’s a section that smacks of content by committee:

“…but sharing clinical trial data runs the serious risk of violating the privacy of a person’s health.”

Unclear how the privacy of a person’s health is violated by the sharing of clinical trial data. Perhaps the privacy of a person would be violated… this language is very, very flack-like in tone and content.

(The Atlantic came under fire for doing this a while ago:

Big pharma spends significantly more on marketing than on R&D. Like David Koch with PBS, PhRMA will stop funding this “Impact” section, should TPM’s reporting uncover any pharmaceutical boo boos. But if this Data Sharing “sponsored” article is any indication, TPM has bought the PhRMA bait, hook, line and sinker.


Barry 03.28.14 at 12:21 am

First Nate Silver, now Joshua Marshall.

I guess in April Paul Krugman will get an endowed ‘Professor of Acme Corp.’ chair.


rageahol 03.28.14 at 7:38 am

this is, i’d guess, all part of a coordinated media push around the Trans-Pacific Partnership agreement, a big free trade deal that would also happen to shut out the developing world or global sount (choose your term) from their rights to invalidate drug patents on public health grounds. a whole lot of ground was won in the early 2000s on this because of the efficacy of HIV drugs, but it stands to get rolled back.

Plenty of civil society groups are against this, for obvious reasons. the EFF and Public Citizen both have pages, I’m just going to let readers find them themselves.

the US Chamber of Commerce has been pushing their pro-TTP promoted tweets lately. There have been pieces in NYT. and yes, it looks like Marshall has been bought. can’t say i’m surprised.


rageahol 03.28.14 at 7:38 am

dammit, “global south” and “TPP”.


aspergum 03.28.14 at 11:35 am

This is all pretty simple, seems to me: that ad at TPM looks just like one of the site’s regular posts, except the author’s name is different:

Finally, a reason to stop reading what had become a rather bland and toothless place. Again, it’s simple: Josh sold his journalistic integrity to a rapacious, greedy corporation.

Which means I’ll spend even more time on Counterpunch (despite the lack at times of careful editing).


Barry Freed 03.28.14 at 1:53 pm

Too bad about TPM, now where else can I find my Daily Show and Colbert Report clips?


Abby 03.28.14 at 3:49 pm

As per Dominick@2’s question, If Josh is willing to run actual investigative reporting pieces on the pharmaceutical industry, as well, I don’t see this clearly identified white propaganda from PhRMA as much more troubling than the Russiyskaya Gazeta section in the New York Times I perused today (“Russia’s annexation of the peninsula has brought the region’s history into the spotlight”). If Josh doesn’t run investigative pieces that take issue with PhRMA positions in the future, TPM will be suspect and I might have to turn to the Wall Street Journal (?!?) for independent views on clinical trials data availability:

Could the FDA make a difference?
It’s vital to also keep regulators in our sights. The fact is that the FDA is in possession of more clinical trial data than any single company, and policies allowing easy third-party access to the data in FDA’s holdings could single-handedly create sweeping changes. That is what is happening in Europe, with the European Medicines Agency’s new transparency policies, and we need to see the FDA, which holds far more data than EMA, take a similarly active interest in data transparency.


Barry 03.28.14 at 4:26 pm

” Recognizing that mortality rates for cancer have not improved in 40 years …”

I thought that they had? I did a quick search, but couldn’t find any simple figures.


JW Mason 03.28.14 at 5:23 pm


The CDC gives cancer death rates per 100,000 as 194 in 1950, and 178 in 2007. (Table 24 here.) So some improvement, but not much, compared with other causes of mortality. Heart disease death rates, for example, fall by more than 50% in the same period.

Those are age-adjusted rates, using the fixed 2000 age distribution. If you just look at raw mortality rates, cancer was 187 per 100,000 in 2007, up from 154 in 1955, and up from 64 in 1900. (I have no idea how accurate 1900-era cancer diagnoses were.) So yes, overall mortality from cancer is going up, not down, but that’s because other causes of mortality have been dealt with so successfully and people still have to die of something.

(This is one of my hobbyhorses, but it is really striking to look at the leading causes death at the beginning, middle and end of the 20th century. In the first half of the century, some of the biggest causes of death, like tuberculosis and diphtheria, were completely eliminated. Over the second half of the century the list hardly changes at all. It’s hard to deny that the era of great medical breakthroughs — as for many other technologies — was the early 20th century, not today.)


JW Mason 03.28.14 at 5:33 pm

(To be fair, the trend in cancer mortality has a lot to do with the fact that cigarette smoking became common in the 1920s and only really took off in the 1930s. So in the middle part of the century whatever other improvements were happening were counterbalanced by a rising proportion of smokers in older cohorts.)


JW Mason 03.28.14 at 5:42 pm

Re TPM, yes, the sponsored posts are indistinguishable from regular posts. Especially since, at the moment, the post right above the advertorial one is a wire service piece with the author listed as “Associated Press.” So the distinction between bylined journalistic content and non-bylined advertising that Josh claims to be drawing, does not exist. Nothing on the site tells you that “by the Associated Press” means something very different from “by PHRMA”; the whole design conveys the opposite impression.


Marc 03.28.14 at 8:06 pm

The message in question has a large banner at the top saying
“sponsored content”. I don’t know if it always looked like that, but it was obvious to me that it was different.

The post here is a tu quoque exercise. I guess it’s supposed to illustrate…something, and I suppose that in principle bad outcomes are possible. But he just told us (in a comment that Henry conveniently ignored) that his PollTracker in 2012 was sponsored by Big Oil. If that wasn’t the end of the world (and I sure didn’t see any evidence of softness towards the oil industry), why is this?


aspergum 03.28.14 at 9:12 pm

” Nothing on the site tells you that “by the Associated Press” means something very different from “by PHRMA”; the whole design conveys the opposite impression.”

Yes, especially on my phone’s browser. Not sure why the link didn’t work before, so I’ll try again–


faustusnotes 03.28.14 at 11:27 pm

I also immediately thought that second sentence about trends in mortality was wrong, and I was right. Trends in the USA are broadly declining from a peak in the 1990s, and the only reason that they aren’t going down much faster is the huge increase in female smoking (see Figure EMO3).

This is also seriously misleading because mortality rates don’t adjust for improved detection, and changes in risk factors. A better measure of the effectiveness of pharmaceutical treatments for cancer is cancer survival, which appear to have doubled since the 1970s. Five-year breast cancer survival in the UK is now 80%, just below the rest of Europe, for example.

I wonder if TPM checked the facts in their puff-piece?


Witt 03.29.14 at 1:14 am

We don’t have to argue about if a publication’s editors and writers will be influenced by this sort of arrangement.

There is an array of research on pharmaceutical companies’ influence of doctors that affirms how humans do get influenced by small things.

Importantly, small gifts (like pens or notepads) can be just as influential as large gifts — because people don’t like to think their behavior can be purchased, especially not cheaply. So they rationalize changing their behavior by convincing themselves they changed because of learning new info (science!) rather than the notepad or mug.

As one study concluded: ” Subtle exposure to small pharmaceutical promotional items influences implicit attitudes toward marketed products among medical students.”

A NY Times write-up of the study (a bit more readable to the layperson):

Another study stated: “Considerable evidence from the social sciences suggests that gifts of negligible value can influence the behavior of the recipient in ways the recipient does not always realize.”

As a reader, I don’t worry about obvious payola or blatant quid pro quo in a case like this. I’m more concerned with why an editor would believe he can somehow stay immune to the array of influencing factors that humans in general are subject to when wooed by a sponsor/advertiser/marketer.


x.trapnel 03.29.14 at 6:01 am

I think Alex asked one key question in the last thread, which never got answered: will TPM use its robots.txt file, etc, to make sure search engines treat these pieces as ads rather than articles? I’m guessing the answer is no–It shows up now for me, looking very much like an article based only on the snippet shown in search results. By the time I’ve clicked through and noticed the banner, I’m already a page away from any non-AstroTurf results: mission accomplished.


Alex 03.30.14 at 11:42 am

x: it got answered in my blog post, by me. the sitemap doesn’t mention idealab.

i am yr linucks: curl | zgrep idealab-impact


Witt 03.30.14 at 2:59 pm

FYI, Alex, I linked your post at the Nieman Lab journalism site:


Jerry Vinokurov 03.31.14 at 3:37 pm

It’s hard to deny that the era of great medical breakthroughs — as for many other technologies — was the early 20th century, not today.

Maybe it’s more accurate to say that the low-hanging fruit have been picked, and the discoveries that remain are much harder to come by.


fzsguitar 03.31.14 at 6:55 pm

I’ll never forget that after the 2008 election, Mr. Marshall attended a dinner at the White House along with a number of other prominent “liberal” “left” bloggers. Since then, TPM has been the premier destination, in addition to MSNBC/Daily Show/HuffPo/NYT, for a good dose of sycophantic blabber (check out the sockpuppets in the TPM comments section if you ever question Our Dear Leader and his illegal wars/defense of the 1%).


john medcalf 03.31.14 at 8:00 pm

The same thing that took TPM off my book marks pre-Obama continues to make it near irrelevant – too much filler. Who dilutes their day on TPM anyway ?


liberal 04.01.14 at 3:52 am

@18: is there anything in that paper which shows they managed to control for lead time bias?


JW Mason 04.01.14 at 6:01 am

Maybe it’s more accurate to say that the low-hanging fruit have been picked, and the discoveries that remain are much harder to come by.

Sure. What’s the difference?


faustusnotes 04.01.14 at 8:33 am

liberal I don’t know, I’m guessing they didn’t adjust for lead time bias, but you’re unlikely to find that information in a Guardian article. In any case, it would be a big issue only in diseases with significant screening programs (i.e. cervical and breast cancer). For diseases primarily detected through symptoms (all of the rarer cancers, prostate cancer in most countries) then I imagine lead time bias is not a big factor.

Comments on this entry are closed.