Item #26: Was the IOM Report Biased?

In March 2004, Fishbein began seeking whistle-blower protection. He met with congressional staff and attracted enough attention on Capitol Hill to force the NIH to agree to a study by the National Academy’s Institute of Medicine (IOM). The terms of that inquiry were skewed from the outset, however, and the nine-member panel decreed that it would not deal with any questions of misconduct. The panel ignored Fishbein’s evidence that DAIDS had covered up the study’s failures and relied on testimony from the HIVNET investigators and NIH officials. Not surprisingly, it found that HIVNET’s conclusions were valid.

Farber says that the terms of the IOM study were skewed from the start because the IOM would not look at issues of misconduct.

Investigating misconduct is not the role of the IOM and it was not asked to do so in this case either. The IOM was asked to examine scientific issues. It concluded that the science underlying the HIVNET 012 was sound. It also found that the trial largely conformed to internationally accepted ethical standards. Issues of misconduct are investigated by the NIH’s Office of Research Integrity and/or by the Department of Health and Human Services’ Office of the Inspector General, if they are justified. We are not aware that any such investigations have been initiated.

The reaction of the modern scientific community and its funding partners in government and industry is quite commonly shamefully slow and inadequate. Often more effort is spent trying to attack the whistleblowers than to deal with the actual problems.
This is described in great detail in books such as Horace Judson’s “The great betrayal: Science in the culture of fraud.” (Harcourt. 2004) and Bell’s “Impure Science: Fraud, Compromise, and Political Influence in Scientific Research” (John Wiley. 1992).
Recent articles, editorials and letters in the July 30, 2005 issues of both the British Medical Journal and Lancet illustrate the difficult in dealing with scientific fraud, even when there are well documented suspicions that a series of papers in respected journals over many years have been totally fabricated.
The absence of an investigation is not proof of absence of misconduct or bad science. As Judson points out the focus should not be on fraud in any case. The question is whether the science was good or bad. Discovering whether the bad science is due to fraud, sloppiness or sabotage is important, but secondary to the main task.
The main aim should be to remove recommendations based on unreliable data. And what has been discovered so far is that the Nevirapine data is unreliable and should not be used for decision making. Just because we don’t know exactly how bad the situation is does not mean that it is alright to continue relying on the data.