Wrong, Wrong, Wrong.

by hilzoy

The Seattle Times has run a series of stories (the main one is here, the rest are accessible from the sidebar) on doctors being paid by Wall Street analysts to talk about ongoing drug trials:

“Doctors testing new drugs are sworn to keep their research secret until drug companies announce the final results. But elite Wall Street firms — looking to make quick profits — have found a way to harvest these secrets: They pay doctors to divulge the details early.

A Seattle Times investigation found at least 26 cases in which doctors have leaked confidential and critical details of their ongoing drug research to Wall Street firms. The practice involves doctors at top research universities from UCLA to the University of Pennsylvania, and powerful financial firms including Citigroup Smith Barney, UBS and Wachovia Securities. In 24 of the 26 cases, the firms issued reports to select clients with detailed information obtained from doctors involved in confidential studies. The reports advised clients whether to buy or sell a drug stock.

Trading stock based on secret information bought from medical researchers is illegal, say legal experts who were told of The Times’ findings. “That’s a good way to go to jail,” said lawyer Thomas Newkirk, former associate director of enforcement at the Securities and Exchange Commission (SEC).”

The article summarizes some of the problems with this:

“Ordinary investors are victimized when inside information is leaked to select investors. Those who know in advance whether a drug is going to succeed or fail can buy stock low or sell it high to those who don’t know, making quick fortunes by taking advantage of unwitting investors. And there is a broader cost to society: Leaking details about ongoing research can introduce bias into drug trials and possibly halt development of potentially life-saving drugs, biotech executives said.

“It appalls me, I must say,” said Christopher Henney, a Seattle biotech pioneer who co-founded Immunex, now part of Amgen. “It’s absolutely outrageous that they [researchers] would allow themselves to be corrupted in that way.” “The practice is a moral cesspool,” said Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania. “It really just seems to me to be the last straw in the corporatization of American medicine.””

The point about insider trading is straightforward: people who have access to the relevant information get to clean up financially, at the expense of people who don’t. That’s wrong, especially since those who have such access are investors in hedge funds, who are virtually certain to be extremely wealthy, and thus to have such built-in investment advantages as being able to absorb more risk than others. They are just about the last people who need an extra edge.

The point about the effects on medical research is more tricky. Many medical trials absolutely depend on being ‘double-blinded’: neither the researchers nor the patients know which patients are getting which therapies. When an experiment is successfully double-blinded, doctors can’t interpret results in light of their prejudices — say, by interpreting the results to fit their beliefs about whether the drug being tested works — since they don’t know who’s getting the drug to start with. Moreover, since trials are often carried out at a number of different centers, individual doctors are often not in a position to know the trial’s overall results even if they figure out which of their patients are getting the drug, and so the results of the trial overall can’t effect either their interpretation of what they see or the patients they choose to enroll. By removing several sources of bias, these sorts of ignorance make the overall trial a lot more trustworthy, which is important since our knowledge of what works and what doesn’t depends on these trials.

If doctors find out about the overall results of the trial, either from the analysts they talk to or from money managers, this can reintroduce these biasses into the trial. This makes the trial as a whole less trustworthy, and thus less useful. Moreover, it sometimes happens that when a trial starts out, the data (or: the data that one person knows about) look one way, but after a certain amount of time, their trend is quite different. (Imagine flipping coins: your first seven flips might all come up heads, but eventually the heads/tails ratio would settle down at around 1:1.) If the data start out looking bad for a drug, and this information gets out, the company testing the drug can lose its capitalization, which (in turn) can cause it to pull the plug on the study. But since companies generally do not enroll lots more people than they need to establish the results they’re looking for, preliminary data of this kind almost always involve too small a sample size to warrant much confidence. If companies routinely end up losing funding halfway through trials, this will almost certainly mean that some drugs that would be shown to be effective will not make it to market. And that harms us all.

I think that drug companies are not in a good position to stop this. They depend a lot on the doctors who do their research, and will not want to alienate them. I would think that some combination of criminal prosecutions and pressure by the universities they work for would work better. Medical schools, as a general rule, already require that the doctors who teach for them report various financial arrangements — consulting for companies, etc. — and have conflict of interest policies that are (at least where I work) taken quite seriously. They should probably be expanded to deal with this practice. And, as I said, criminal prosecutions as well, for both the analysts and the doctors involved. Because this is just wrong.