Wednesday, April 26, 2006

Harvard Law School Professor Takes Issue With Being Listed as an Example of a "Junk Scientist" by Anti-Smoking Group

In a letter to the editor published in the just-released issue of the American Journal of Public Health, a Harvard Law School professor, Dr. Kip Viscusi, takes issue with his research being characterized as being "junk science" in a previous AJPH article written by several tobacco control researchers. Dr. Viscusi was an expert witness in the Minnesota tobacco litigation as well as in the Mississippi tobacco case.

In the original article, Dr. Viscusi's research was cast as being "junk science." In a section entitled "The Tobacco Industry's Own Use of Junk Science," Dr. Viscusi's work is provided as one of four examples. He is referred to as a "professional witness" and a statement is made to the effect that a professional witness "spends more time in the courtroom than in the laboratory."

The evidence of the potential inappropriateness of allowing Viscusi to testify is that he was alleged to have conducted research that was commissioned by the law firm representing R.J. Reynolds specifically for the purposes of its defense at the trial. Apparently, Dr. Viscusi was not allowed to testify in Minnesota; according to the article, the reason was that his "testimony was based on surveys done in anticipation of litigation (a reason to exclude such testimony under Daubert II), commissioned by Jones Day Reavis and Pogue, the law firm representing RJ Reynolds Tobacco Company."

In responding to the characterization of his research as being junk science, Viscusi notes that: "the studies I have relied upon in my expert testimony in tobacco cases have been published by the leading peer-reviewed university presses, such as Oxford University Press and University of Chicago Press, and by some of the most distinguished peer-reviewed economics journals, such as the Review of Economics and Statistics and the Journal of Political Economy. To suggest that these groups, considered preeminent in their field, would publish work based on 'tainted surveys' and 'junk science' defies reason and experience. This research and my other work on decisions involving health and safety risks are among the most widely cited publications in the economics field."

The Rest of the Story

I am familiar with Dr. Viscusi's work, and I find it unfair, inappropriate, and inaccurate to suggest that his research represents junk science.

I think what is going on here is that three very different issues are being somewhat conflated, leading to a presentation that attempts to present the facts in a simple way, but which perhaps becomes misleading because the issues are more complex.

There are three very different issues at work here.

First is the issue of junk science, which is "faulty scientific data and analysis used to promote special agendas." The key here is that the science is faulty - there is something terribly wrong with it. It is not valid science.

The second issue is bias in research. This bias could be created because someone is being paid to testify, because research was commissioned by the law firm representing a client, and/or because the research was conducted for the specific purpose of producing data that could be used at trial. The key here is that it does not necessarily mean that the science is invalid; there just may be bias present that needs to be disclosed and accounted for.

The third issue is the legal issue of what constitutes permissible expert testimony. This is where Daubert comes in. Each of the first two issues comes in to play here, as a judge must consider whether research is faulty and whether there is an inherent bias that would preclude the testimony from being admissible.

The most important point I want to make is that just because there is an inherent bias in a piece of research and even allowing that a judge makes a decision to bar testimony under Daubert, this does not necessarily mean that the research represents junk science. It is possible, for example, that the judge ruled that the inherent bias (such as the preparation of the research specifically for trial) was enough to preclude it under Daubert.

The reason why I think Dr. Viscusi has a legitimate grievance with the way his work was portrayed in the AJPH article is that the article casts his research as being "junk science," not simply as being research that was ruled inadmissible in court because of the nature of its preparation (i.e., the alleged fact that some of the relevant research was "commissioned" by the law firm representing R.J. Reynolds and that it was allegedly disallowed under Daubert). Note, by the way, that Dr. Viscusi's letter takes issue with the contention that his testimony was actually disallowed under Daubert).

The same arguments being used by the authors to discredit Dr. Viscusi's work and cast it off as "junk science" could be used to cast off my own research or that of innumerable witnesses against the tobacco companies.

It is common practice for "anti-tobacco" witnesses to prepare specific research at the request of plaintiff's attorneys. In my case, the California Attorney General's office supported research I conducted into the exposure of youths to cigarette advertising in magazines (the state provided the data for my analyses). This research was partially the basis for my testimony in California's lawsuit against R.J. Reynolds for allegedly violating the Master Settlement Agreement by targeting youths in its magazine advertising. This work was not commissioned, since I would have done this research anyway, but it was supported by the state of California, which ended up being the plaintiff in the lawsuit in which I testified.

In its defense at trial, R.J. Reynolds did indeed attempt, unsuccessfully, to use this information to disallow my testimony, but the company's attorneys showed me the respect of not suggesting that my work was complete junk, a courtesy apparently not afforded to Dr. Viscusi here.

It seems to me that an assumption is being made that research used to support the tobacco industry's defense at trial is junk, while research conducted under quite similar conditions but used to support the plaintiffs, is automatically valid.

Even the argument about Dr. Viscusi being a "professional witness" doesn't necessarily hold water. Without even getting into whether the characterization is accurate or not, the same thing could be said about me or a number of other tobacco control practitioners who have testified in a large number of tobacco cases. I think I'm up to about 10 different depositions or testimonies at trial now. Does that mean that I'm a "professional witness" and that my research is "junk?" (My research may be junk, but I assure you that's not the reason!)

I think we have to be careful here, because some of the same arguments that we appear to be using against tobacco industry witnesses could easily be used against us.

For example, the critical (and only) research that I am aware of that supports the key contention, made at trial in the DOJ tobacco lawsuit, that the American Legacy Foundation's "truth" campaign is extremely effective in reducing youth smoking, is a paper that was authored by the American Legacy Foundation. You can bet that if Philip Morris wanted to present research at trial on the effectiveness of its youth smoking prevention programs and that work turned out to be written by Philip Morris, we would be demanding that the research be thrown out. We would probably be calling it "junk science." And maybe it would be. But the knife cuts both ways.

Let me say, however, that the biggest reason why I believe it is unfair, inappropriate, and inaccurate to suggest that Dr. Viscusi's research represents junk science is that it is clearly not. Dr. Viscusi's work on risk perception represents seminal work and is highly regarded in prestigious academic journals. The research is valid and reputable and provides important insights that could actually help inform tobacco control interventions if we paid some heed to it rather than ignore it and immediately cast it off as junk science.

Now understand that I am saying this as someone who has directly confronted some of Dr. Viscusi's work in the courtroom and who has been in a position of arguing that smokers do not have an accurate perception of their risks despite Dr. Viscusi's finding that smokers tend to actually overestimate the absolute risk of disease. The fact that I may disagree with some of the interpretation of these findings and with the ultimate conclusion regarding whether smoking is a behavior that is done with a fully-informed understanding of risk does not mean that the research itself is not reasonably valid.

I see no scientific reason why tobacco companies should not be allowed to let Dr. Viscusi present his research in the courtroom and allow the jury to consider his work as well as that of scientists who have reached different conclusions. Let the jury see all the data, let them hear the differing interpretation of the data, and then let them make their decision. Isn't that what trial by jury is supposed to be about?

I would hasten to add here that I think Dr. Viscusi's characterization of the state litigation against the tobacco companies which resulted in the Master Settlement Agreement is dead on -his description of the MSA being about money rather than public health accords with dozens of posts I have written over the past months.

Obviously, there are some elements of Dr. Viscusi's opinions that I don't agree with, but there's no question that his research is not junk science, and that characterizing it as such is unfair as well as inaccurate and inappropriate.

I understand that the legal battle between the plaintiffs attorneys and the tobacco companies is a fierce one and both sides are going to do everything they can to win, but it would be a shame if individual researchers ended up taking unfair hits to their character and the quality of their research because of it. It appears to me that this is what has happened to Dr. Viscusi (not to mention a few other researchers on the anti-tobacco side who I will not mention here).

The rest of the story is that Dr. Viscusi's research has been inaccurately and unfairly cast as being junk science. The issues of potential bias in research and the utter faultiness of research appear to have been conflated, resulting in an unfair attack on the character and research quality of an individual.

We don't need to do this. We can win in the courtroom, and we can win with a little more courtesy and dignity than attacking individuals and their character and reputations based not on the actual quality of their work, but on the side that they take in the Tobacco Wars.

No comments: