Yet although I acknowledge that the behavior of the physician researchers described in the Seattle Times series does not look too pretty, I contend that this story may be a bit murkier than it first seemed. Some additional questions to consider are:
- Were There Really "Drug Secrets?" - I previously posted about a New England Journal of Medicine article that described how medical school administrators entered into research contracts with commercial research sponsors. Half or more of schools permitted contracts that allowed the commercial research sponsor to "own" the data and to control of the writing of articles describing the research findings; and required the school to keep research results and the provisions of the research contract itself confidential. Such contracts appeared to be in conflict with medical schools' scientific and humanitarian mission. In particular, allowing research sponsors to suppress research results violates the implicit promises made to research subjects that their data will advance science. Thus, clinical research on humans ought not to produce any "drug secrets." Whether researchers who work for schools which agreed to these contracts ought to be bound to keep such "drug secrets" is unclear. Whether it is unethical to violate these contracts' confidentiality agreements is similarly unclear. Although disclosing research data to investment analysts does not give the physician researchers the moral high ground, establishing the confidentiality agreements within the contract does not give the research sponsors or the medical schools any moral high ground either.
- Of What Value is the Information Revealed by the Physicians? - The physicians in the Seattle Times series generally claimed that the information they revealed included general impressions of how the research project was going, and their "gut feelings" about which drugs involved in the comparisons was likely to be better. Consideration of some principles of evidence-based medicine suggest that this information is unlikely to accurately to predict the results of the clinical trials: 1) Double-blinding should have prevented the physicians from telling which patients got which drug. 2) In multi-center trials, results from one center may not predict results from the whole trial. 3) In long-term trials, short-term results may not predict final results. Although the stock analysts cited by the Seattle Times claimed that they could predict trial results based on the information revealed to them, they supplied only anecdotal data about their performance. In fact, their predictions may have been no better than chance alone. So the Wall Street analysts had no claim to the moral high ground either.