Science and Political Affiliations

August 26th, 2005

Posted by: Roger Pielke, Jr.

The Chicago Tribune has a very interesting article today (Thanks JA!) on the recent study of fetal pain and the political leanings of its authors. Here is an excerpt:

“A research article about when fetuses feel pain is sparking a heated debate over the nexus between science and politics and what information authors should disclose to scientific journals. The report, published Wednesday in the Journal of the American Medical Association, analyzed previously published research and concluded that fetuses probably don’t feel pain until 29 weeks after conception because of their developing brain structures. Undisclosed was the fact that one of the five authors runs an abortion clinic at San Francisco’s public hospital and another worked temporarily more than five years ago for an abortion-rights advocacy group. Several ethicists said they consider those points regrettable omissions that left readers without important information. Other experts consider the authors’ background irrelevant.”

Does it matter what the authors professional or political affiliations happen to be?

The story presents two perspectives. First, no it doesn’t matter.

“”The standard for disclosure in medical and scientific journals is not your politics. There’s no obligation to tell people what your mind-set is … as long as the data is sound and gathered objectively,” said Dr. Alan Leff, a University of Chicago pulmonologist and editor of the Proceedings of the American Thoracic Society . Dr. Philip Darney, chief of obstetrics, gynecology and reproductive sciences at San Francisco General Hospital, defended that decision, saying in a statement: “The research team does not believe that being an abortion provider is a conflict of interest.” Medical journals require authors to disclose financial ties to industry or other funding sources. But there are no standards for disclosing other factors that might influence an author, such as clinical practices or organizational affiliations. Dr. Catherine DeAngelis, the journal’s editor-in-chief, said she wasn’t concerned by Drey’s failure to indicate she performed abortions. “That’s part of [an obstetrician's] scope of practice. They don’t have to reveal that.” …

“As a scientist, if you think I’m wrong, you probe my data, question my findings and do a critical study–not point your finger and talk about my politics,” Caplan said. Rigorous methodology is supposed to minimize the potential for bias in scientific research, he said, “whether studies are done by communists in China or free-marketers in Chicago, whether they’re done by left-wingers in Berkeley or right-wingers at the Wharton School here in Pennsylvania.”"

Second, it does matter:

“Anti-abortion groups say the authors’ affiliations are crucially important. “These are people with years of professional and ideological investment in the pro-abortion cause, not some neutral team of medical professionals,” said Douglas Johnson, legislative director of the National Right to Life Committee. “We think readers and viewers have a right to know who’s filtering the information they’re being presented with.” …

… but certain medical issues are so explosive politically–abortion certainly, and perhaps stem cell research and animal rights–that researchers have a special obligation to inform readers of relevant affiliations. The San Francisco researchers “must have known there would be criticism from the right-to-life people,” said Dr. Arnold Relman, former editor of the New England Journal of Medicine. “In a situation as contentious as this, it seems more disclosure should be the rule rather than less.” Dr. Marcia Angell, a senior lecturer in social medicine at Harvard Medical School, is also a former editor of the New England Journal. “Suppose it were the other way,” she said. “Suppose there were an article that said that [fetuses] do feel pain and it was written by people who were involved in the right-to-life movement. Would I want to know that? I think I would.” With an issue as divisive as abortion, disclosing potentially important associations can only help a journal edit! or, said Sheldon Krimsky, author of “Science in the Private Interest.” “It kind of ratchets up everyone’s attention to the science and makes them that much more vigilant in detecting potential bias,” he said.”

Should political affiliations relevant to a paper_s content be disclosed by authored of peer-reviewed articles?

Is there a justification for limiting conflict-of-interest only to financial considerations?

Tough questions.

2 Responses to “Science and Political Affiliations”

  1. Marlowe Johnson Says:

    Ideally, when a paper is peer-reviewed, it is judged based on its relative merits alone. Whether or not this actually occurs is a separate question. So, for example, if a study survives the peer review process and shows that the global warming potential of CFCs is offset by the destruction of ozone, then it shouldn’t matter if the study was funded by Dupont or not.

    Personally, I don’t think it’s the researcher’s responsibility to disclose any information about who has funded a study, or, for that matter, anything about their particular belief system. Instead, the onus lies with the individual or group that uses that study for advocy purposes to demonstrate bias, or lack thereof, depending on the spin that one wants to put on the issue at hand. Presumably this should be difficult to do– assuming of course that the peer review process has been properly conducted.

    The other problem, of course, is that most people’s views on controversial issues such as abortion, are often nuanced and don’t fit into tidy either/or characterizations. So while it is relatively simple to determine financial conflict of interest, the same cannot be said of the relationship bewteen belief systems and scientific inquiry.

  2. 2
  3. Dylan Otto Krider Says:

    I would say that science is set up with the assumption that people are biased, and (unless Bush gets his way) to remove or at least put a check on those biases as much as possible. A think tank study that is not put up for peer review or for other scientists to repeat the experiments is automatically suspect. A bit of research published in a journal for other scientists to critique will find some scrutiny.

    Patrick Michaels certainly has biases, but being Cato’s resident rent-a-skeptic doesn’t mean he’s not doing good science. As long as he’s submitting his papers through the scientific process, that doesn’t really matter. If his biases get the best of him, you can be sure those with different biases will submit a letter in reply.

    Contrast that with Dr. Brind who discovered a link between abortion and breast cancer after a meeting with a pro-life group. He submitted his research, it got published and summarily rejected. I have no problem with that. What I do have a problem with is now that after he found the scientific community was against him, he moved to Phoenix to air commercials telling young women they will get breast cancer if they have an abortion. His need to drop out of the scientific process tells you all you need to know aobut the validity of his pet project.

    To paraphrase Asimov, I automatically assume a study is bogus if it appears in the NYT before a scientific journal.