Q&A with Kim Klausner: Challenging the "normalcy" of pharma ghostwriting

Career archivist Kim Klausner takes her roles as a historian and as a public health advocate equally seriously. As the Industry Documents Digital Libraries Manager for the University of California-San Francisco, she is in charge of the Drug Industry Documents Archive, a collection of thousands of records that shine a light on practices by Wyeth, Pfizer, Abbott and other Big Pharma companies.

Klausner earned her master's in U.S. History from San Francisco State University and wrote her thesis on Tima Ludins, a New York City teacher and union activist who was a member of the Communist Party and refused to name fellow party members when forced to testify in front of the Senate Internal Security Subcommittee. I asked her about naming names when I reached her at her office in San Francisco. The first part of my interview with Klausner was posted last Friday. The second part is below. It has been edited for space and clarity.

Q: What are your plans for the archive over time?

A: I want to add millions of documents. The more documents you have, the richer the research will be. That's my goal. But we don't have anything along the lines of the master settlement agreement, which was the impetus for the tobacco industry to make their documents public. We don't have an analogous thing with pharma.

Q: So how do you generate millions of new documents?

A: It's a question of educating people about DIDA and why it benefits everybody to have one place to research these documents. We're trying to educate attorneys and judges about why judges should unseal documents and provide them to the archive. The other thing I want to do is develop the software that runs DIDA. The search and retrieval mechanisms need to be expanded so that researchers have adequate tools. There is no sort mechanism, for example, sort by document date. There is no book bag where you can keep track of your citations once you find good documents.

Q: You can save a copy to your hard drive, right?

A: Yes. You can save the PDF, but you have to open it up and save it. In our tobacco archive you can click next to the document and save the citation, without ever opening the document itself or putting it on your hard drive. Then you have a permanent URL that you can use to go back to those documents again and again. And you also can cite the URL when you are putting together a report and when you click on it you can get right back to the document.

Q: Similar to a PubMed search?

A: Exactly.

Q: Are you doing anything to reach out to plaintiff's attorneys?

A: I've started to do a little bit of that. The group formerly known as the American Trial Lawyers Association, now the American Association for Justice, has a working group of pharmaceutical plaintiffs' lawyers, and we have been in touch with them. We are sort of networking around. An attorney gave us some Vioxx documents, for example. My sense now is that now it makes more sense for a third party like journalists or advocacy groups or libraries like us to go to the judge rather than the plaintiffs' attorney. If you go as PLoS and The New York Times and say, "Unseal these documents for the public good," the judge is going to be more sympathetic. The people at Public Justice who represented PLoS said they would be interested in doing that in the future. I am hoping that is the case. They've succeeded here, and it should get easier to do that in the future.

Q: Do you have any concern about how this effort is perceived? If the industry starts to see the library as a foe and not as an unbiased repository, might judges start to look differently at the prospect of unsealing these records?

A: I would think that the appeal is based on overriding public interest, and we are a library that is serving the public interest. We are advocating for the free flow of information. But who knows how the pharmaceutical industry will respond?

Q: What are your big barriers to expanding the database?

A: The problem is we don't really have the capacity to add even thousands of documents because of the metadata creation. It costs money to have that done. There's a company that we use to create metadata to create our tobacco documents, and they charge 20 cents a page to do the metadata. It can be quite costly to do thousands of pages of documents. One of the things we're looking into now is possibly having the public create metadata to create a method by which people can do that sort of voluntarily, whether they be students in the history of medicine or library school students or people interested in looking at the documents. That's not the greatest, but if it gets the job done we would go with that.

Q: Why not pay a nickel a page to some volunteer types to add some incentive but still save on costs?

A: It's partially the bureaucracy here. We can get a contract with a vendor, but to start paying hundreds of people a nickel for each page, then you get into the 1099s and everything. Writing a check for $26 to 100 different people would just be too much.

Q: This is sort of the Wikipedia approach.

A: Exactly. The Guardian in London is doing something like this with members of Parliament's expenses. You get retired people who are interested in reading their MPs expenses, and I could see someone here being interested in reading about these drug companies. The Australian Newspapers also have a really neat site where they have people correcting the OCR text.

Q: Would you be concerned about someone manipulating the process, as has been a problem with Wikipedia?

A: We would have to do some quality assurance. People would have to log in to do it, and if we saw somebody entering in really incorrect data than we would just bar them from being able to log in or something. Once it was created it would get added and nobody could do anything to it. You either have the high school students who think it is funny to do bad metadata or people with political motives. I'd be more concerned about sloppiness.

Q: Has there been anything surprising in what you've found while going through the documents?

A: I was a little bit surprised by the normalcy of the whole endeavor. These professors who are going to become the authors act like it's entirely normal. Maybe they're used to having graduate students do this writing and they think, "What does it matter who does the research and the first draft?" It just sort of seems like nobody has any regrets or misgivings. Except for this one South Korean doctor: There's an email back and forth between him and Karen Mittleman, who works for DesignWrite. He asks Karen Mittleman, "Why don't you be the co-author with me?" He was recognizing that she had been key to the article, but, of course, that's not the way the whole process works.

Q: What do you think of Karen Mittleman?

A: She's an interesting character. I actually have just received the full depositions, and I'm impressed with the resources that the company paid for thiis work. This was a big operation, and it was really important to these companies in terms of their marketing.

Q: You wrote your master's thesis about a teacher who refused to name names during the Communist blacklisting days. Now you are in charge of an archive where lots of names are named. Does that seem odd to you at all?

A: Well, I really do feel like it's apples and oranges. People in the McCarthy era were asked to name names as a way of humiliating them and sort of as a way to control people. The investigators already had the names. They didn't need names from the people being interviewed. Instead, it was a ritualized public and political humiliation. In this case, people don't have the names or know what went on. These documents are providing new information. And the documents speak for themselves. Nobody is being asked to name names. It is true that people did write these documents not knowing that they would be made publicly available. And, you know, that is kind of a shame for them. But, on the other hand, there is the overriding public good of sharing this information.

Q: But you have these examples of people in stories, like Gloria Bachmann or this McGill professor, who all of a sudden have their entire careers reduced to one bad headline. I play a role in that, too. And I wonder if it were not for that one article, would these people be remembered differently decades from now? Do you ever feel bad about that?

A: I don't really. It seems to me in the case of that person, if she wants to defend what she did, then she can do that. And if she wants to say, more like that fellow in Australia you wrote about, "Yeah, I don't think it was a good decision on my part," then that's great. People are able to spin what they did. I don't feel bad at all. In fact, I feel really good to be exposing things that have been covered up and have become practices that people don't comment on. That's part of the problem. As I said before, all of this has become so normal.

Q: I spoke with Tom Sullivan at Rockpointe Corporation, a medical education company. They don't write ghostwritten articles for journals, but they do something similar with medical education. They work with drug companies on designing training for doctors, although, it seems, their level of disclosure is a little higher. He sympathizes with these researchers who have been caught with ghostwriting on their resumes. He says that just by labeling it "ghostwriting," you are sort of standing in judgment and taking a self-righteous position above the researcher or, even the company that did the work.

A: I don't look at it as a self-righteous. I thought about it when this first came out. "What is really wrong with what they're doing?" On the one hand, you can say these companies were doing a literature search and providing this information to somebody and asking them to sign their names. But it really isn't just that. It's very different from a graduate student doing a literature search and then putting it into an outline, and then having the professor think about it and then do the actual writing. This is where the company has a goal to sell more products, and they are contracting for a piece of work that they think will sell more products. Instead of paying for advertising, or in addition to that, they are finding a way to work that marketing message into a journal article with the appearance of unbiased science. That is the only reason the article gets written. It's not to disseminate research, although it may help do that. The purpose is to sell more of the product. And it's disingenuous, because it's not labeled as such. I feel like what they're doing is wrong, and it hurts people.