Wednesday, July 25, 2007

Hip Protectors and Conflicts of Interest

Associated Press reported Lindsey Tanner just wrote a story (here, via the Washington Post) about undisclosed conflicts of interest affecting authors of an article about the use of hip protectors to ward off fractures in the most recent JAMA (Kiel DP, Magaziner J, Zimmerman S, Ball L, Barton BA, Brown KM et al. Efficacy of a hip protector to prevent hip fracture in nursing home residents: the HIP PRO randomized controlled trial. JAMA 2007; 298: 413-422.)

A new study showing that padded hip protectors didn't prevent fractures in the elderly has renewed questions about hidden drug industry ties to medical research.

Three of the authors of the study on bone breaks didn't tell editors of an influential medical journal, which is publishing their research Wednesday, that they had consulted for or received research money from the makers of bone-strengthening drugs. That potential conflict was discovered by The Associated Press.

Editors of the Journal of the American Medical Association _ which has tough rules on financial disclosure _ had asked the authors about any conflicts and were told there were none. The researchers said later they didn't believe their industry connections were relevant because the study of hip fractures didn't involve bone drugs and didn't recommend them.

The editor of JAMA agrees. Dr. Catherine DeAngelis said that in this case, the drug company connections didn't violate the journal's detailed financial disclosure policy.

DeAngelis said she believes her journal is being unfairly scrutinized by The AP, which found the researchers' ties to drug companies through searches on the Internet and through a consumer database.

'This has nothing to do with drugs,' she said. 'At what point do you say, come on, is this a witch hunt?'

A close reading of JAMA's guidelines suggests the fracture study authors' ties to drug makers are 'clearly relevant,' said Dr. Michael Callaham, president of the World Association of Medical Editors. 'It's a slam dunk,' he said.

A consumer advocate with the Center for Science in the Public Interest agrees. Readers could easily interpret the study to say that since hip protectors don't work, 'I guess I better take the drugs,' even if that's not what the authors intended, said Merrill Goozner. The consumer advocacy group runs a database on scientists' financial ties, an effort to combat corporate influence on science.

In this case, reporting such ties 'seems to me to be a no-brainer,' Goozner said.

[Study author Dr. Douglas Kiel, a Harvard Medical School researcher] ... said bone-building drugs could be one of several options to hip protectors for older patients, but that they are not ideal for nursing home residents because they take a year or more to work and these patients often are already taking many other drugs. But he also said the drug company money he and his colleagues have received is not relevant to the current study because it wasn't a comparison of both treatments.

'We have no financial disclosures with hip protectors,' Kiel said. 'We were asked repeatedly by the editors whether we were following JAMA's full disclosures and we said yes.'

We have previously noted in discussions of ghost-writing and stealth marketing (e.g., here) that pharmaceutical and other health care companies may make elaborate marketing efforts to position their products. These efforts may include attempts to make the problem for which the product is intended seem more prevalent or more serious, and to make other ways to treat the problem look less effective, more harmful, or otherwise less advantageous. Thus it is easy to see how a study showing that hip protectors do not protect against hip fractures might be useful to those seeking to market drugs meant to protect against fractures.

Thus, in my humble opinion, the authors of such a study should have disclosed any financial entanglements they had with companies making such drugs.

Once again, this case illustrates the enormously tangled web of conflicts of interest that now pervades health care. As I just wrote, each one of these conflicts, no matter how financially advantageous for the parties immediately involved, has the potential to negatively affect health care for the broad population, by raising costs, impeding access, or degrading quality. It is not too late to address this member of the herd of elephants in health care's living room.


InformaticsMD said...

Rule of thumb in determining how far one can "interpret" ethical guidelines:

If your action would sounds bad in a public forum or look bad on the front page on a newspaper, it's probably unethical.

-- SS

Unknown said...

The July 28 issue of The Lancet contains a letter from Gordon Rubenfeld in which he proposes what he call a "conflict vitae." This would be a standardized form that authors would fill out and update as necessary, and which would be publicly accessible through a website. Such a publicly available database would minimize the risk of accidental omission. This proposal deserves serious consideration.

M. Mann