 |
|
 |
|
|
|
 |
|
Manufacturing Uncertainty?
Wednesday, June 22, at 1 p.m., U.S. Eastern time
Watchdogs have sounded alarms over the influence of drug companies on medical studies, but they have given far less scrutiny to corporate influence on research in occupational and environmental health. Critics say, however, that industrial interests increasingly drive research on public-health dangers in the workplace and the environment -- challenging studies that paint companies in a bad light or even suppressing such findings.
What should academic scientists do? Government support for research in the field is slender, and industry sponsorship can give academic researchers access to a facility's workers or to crucial data that would otherwise be unavailable. So scientists find themselves testifying on opposite sides in lawsuits that drag on for years, and the regulatory process often grinds to a halt in the face of conflicting analyses.
Can unbiased science be done with industry sponsorship? If so, what safeguards are necessary, both in conducting the research and publishing it? Do researchers have a responsibility to avoid corporate work, or should they simply go where the data lead them?
» Occupational Hazards (6/24/2005)
David Ozonoff is a professor of environmental health at the Boston University School of Public Health and is one of two editors in chief of a new online journal, Environmental Health. His research centers on the health effects in communities of hazardous-waste sites and the use of scientific evidence in court. He directs the Superfund Basic Research Program at Boston University.
Lila Guterman (Moderator):
Welcome to The Chronicle's live online chat. Today we'll be discussing a thorny issue: industry influence on academic research in occupational and environmental health. I'd like to thank a source for my article and our guest today, David Ozonoff, from Boston University. Let's get started.
Question from Lila Guterman, Moderator:
Let's begin with your telling us briefly your own take on the issue, Dr. Ozonoff. Do you ever consult for industry? Do you think research with industrial ties can be done while maintaining scientific integrity?
David Ozonoff:
I don't consult with industry. Most of my research has involved epidemiological studies of communities exposed to hazardous substances and I have worked closely with these groups. My funding comes almost entirely from government sources (state or federal).
In legal cases I have testified for plaintiffs almost exclusively. The reason is not because defendants are always wrong and plaintiffs always right, but because defendants, who are usually large companies with effectively unlimited financial resources, can afford any expertise they wish to have, while plaintiffs are much more limited in their choices. Since I am primarily a researcher and a teacher, my time available for this is also very limited, so I save what time I have for plaintiff's cases I feel are meritorious and which my expertise is appropriate for (which means I turn down cases as well). Thus I have testified often but about a quite narrow range of chemicals or topics.
Research can always be done with integrity. There are excellent scientists in the employ of industry. The question is about the incentives and disincentives to act in one way or another or to adapt the expression of your science to different norms and requirements. The latter often seems harmless to a scientist, who feels he or she is saying the same thing but in a different way, without realizing the use to which it will be put in a different setting and context. There is also the serious but larger structural issue of the distortion of the scientific agenda caused by other objectives, such as exonerating a chemical that might be the source of significant liability. This is a complex issue which I am sure we will touch on as this colloquy continues.
Question from Patricia Buffler, U. of California at Berkeley:
While it is extremely unfortunate that federal and state funding to support the needed research on occupational and environmental hazards is grossly inadequate, this is a funding situation that existed in the mid -1980’s and still exists two decades later. Obviously, we must press for more public research support, but we also must try to develop research relationships that will support the conduct of high quality research to address the pressing environmental and occupational health questions that continue to arise. Contrary to the opinions of some of my colleagues, I think it would be a mistake to ignore the opportunity to do significant important public health research with private funding because of the concerns regarding a few bad examples. My experience, and the experience of many of my colleagues, is that it is possible, with most private sponsors, to negotiate research relationships that honor the traditions of science and academia.
The issues raised by the problematic examples cited by some of the interviewees suggest the need for an open collegial discussion as to how investigators can successfully and responsibly participate in industry-sponsored research and, when necessary, participate in the legal arena, so as to advance the health of the populations we serve and not become partisans in nonproductive and acrimonious debates. The resulting attacks on the professional and personal behavior of individuals do not advance the knowledge needed to develop sound health programs and policies nor do they foster the confidence of the public in the work that we do. What type of forum is needed to foster this dialogue? What professional groups should provide leadership for such discussions?
David Ozonoff:
Professor Buffler is correct that the underfunding for occupational and environmental health research is not new but is a chronic problem that has faced the field since the outset. Indeed initially, most of the research was done by industry for industry purposes (prior to the OSHA Act of 1970). So for a couple of decades we have enjoyed at least some federal support for this area, although not nearly enough.
But there are now many more demands than previously and many gaps in knowledge that urgently need to be filled. If cooperation with industry is the only recourse, it gives that partner an unusual advantage in the direction and substance of the research, so this is a real problem.
Regarding open and collegial discussion of the problem, I couldn't agree more, and this colloquy and the article that spurred it is an excellent opportunity to do this. The subject has been broached in other contexts and several people or groups have tried to fashion "codes of conduct" for researchers who work on environmental and occupational health research (e.g., the Industrial Epidemiology Forum has a very general code and Colin Soskolne, who is quoted in the Chronicle article, has tried to extend and refine that in the International Society for Environmental Epidemiology). Such codes already exist for clinicians in occupational health (Association of Occupational and Environmental Clinics and the American College of Occupational and Environmental Medicine), where the same kinds of conflicts as to "who the client is" (the patient or the company) exist.
But they need to be embodied in explicit and legal binding agreements, on paper, to handle ahead of time the requirements of transparency (right to publish whatever the outcome), notification of workers, etc. Scientists should also refuse to sign confidentiality agreements about scientific results and the companies should know this ahead of time. The need for these codes has come from obvious problems and conflicts. The examples discussed in the article are not at all rare but unfortunately quite common.
Regarding attacks on the professional and personal behavior of individuals, I, too, agree this is unfortunate. Some of this has been produced by legal requirements that not only encourage but require one scientist to attack another for having flawed methods or bad science. This has been a favored tactic of industry in court cases (not of plaintiffs). Most scientific disagreements are differences of opinion (sometimes purely scientific, sometimes related to source of support, but at least defensible on scientific grounds). They are rarely about one side or another not following accepted scientific method, although that is the form the lawyers require for a challenge of this type. Scientists should not engage in this, even if the lawyers tell them it is required. This would be a big step in restoring the collegiality whose loss Prof. Buffler and I regret.
Question from Ilise Feitshans GWU SPHHS:
Isn't the problem really that we have lost perspective? The field of occupational health has failed to justify itself across the recent decades since the passage of the OSH Act of 1970. Rather than banning industrial sponsorship of studies, which does seem inevitable, why not simply encourage studies from a broad spectrum of grass roots groups who have a public interest but have no funding? I should like to point out also that organized labor has not done its job in this field either. Why not develop a system of parallel funding for grass roots studies every time someone accepts industry sponsorship?
David Ozonoff:
There is at least one other model here, that your colleague David Michaels at GWU mentions in the article. The Health Effects Institute is a hybrid funded by USEPA and the auto industry to do independent research on air pollution. Its first director was Archibald Cox of Watergate fame, succeeded by the current head, Dan Greenbaum. They have garnered the respect of scientists for their work and are a role model of sorts. However it is not clear how easy it would be to reproduce this in other areas. The current administration has shown no compunction about involving itself politically in matters of science and even a joint venture between a federal agency and industry might be suspect because of the blatant realigning of those agencies with industry interests. The Health Effects Institute has the advantage of a long history with stable and very competent leadership and management.
The notion of providing "matching funds" for grassroots groups to support their own research is interesting, although it is hard to see exactly how it could be carried out. In theory, NIH investigator initiated grants could do this, but they are now severely limited and never really responded to the needs of communities. EPA Technical Assistance Grants are also meant for this purpose, but are so small that they do little to balance the resources of industry. Ideas about how this might be accomplished would be welcome. The failure of organized labor to respond is well taken. In fact they are marching in the opposite direction. The AFL-CIO is drastically cutting its health and safety operation.
Regarding the question of "perspective" and failure to justify occupational health and safety, I couldn't agree more. We have been suffering a serious lapse in leadership and the mistaken strategy of trying to play by rules other than those for which we are best suited. For example, the argument that "prevention pays" may or may not be always correct (maybe it is chaper to let someone die, for example). But if prevention doesn't "pay" does it mean we don't want to do it? I think not. We wish to do it because it alleviates premature death, disability and suffering and makes our communities better places to live. But there has been too little hard thinking about what we are about and consequently ineffective advocacy. We have bought into the industry line to our detriment.
Question from Ralph Allan, University of California, Irvine, CA.:
Considering that research through the years seems to conclude, generally, that a healthy workforce is a more productive and efficient workforce, what are the specific factors that contribute to the allegation that corporate sponsorship does not necessarily equate to "good science"? It would seem that the interest in"good health"and "general preventive health" is an interest to all parties involved.
David Ozonoff:
Parsing the question a bit, there seem two elements here: what is the evidence industry sponsored science isn't good science; and why isn't it good science if a healthy workforce is beneficial to them. These each demand longer answers than I can give in this context, but let me briefly indicate the outlines of an answer.
i. Is industry sponsored science "bad?" There is no general answer to this question. There are a lot of ways to do both good and bad science and if you have shot the arrow and want to paint the target around it that's not hard to do and there are unfortunately many examples. But the more subtle issue is the way large sums of money can alter and distort the research agenda or mischievously be used to cast doubt about what should be established issues (global warming being a good example). Money also can be used to direct good scientists away from certain areas inconvenient for industry and toward others that work better for them, either by creating doubt or doing work on profitable topics.
ii. Isn't a healthy workforce beneficial to industry? Not necessarily, certainly not if they make more money on balance from ignoring safety and environmental expenses than they pay out in lost productivity or health care costs. And if they can prevent those costs from ever being assessed against them they are better off still. Additionally, mid level managers who make and implement many of these decisions are rewarded for their performance at the end of the quarter, not the end of a quarter century. So the incentives and disincentives are all wrong.
Comment from Paul English, CA Dept of Health:
In response to Dr. Ozonoff's previous comment on the NIH, two recent RFP's (one on community participation in research and one on health disparities from the National Center on Minority Health and Health Disparities) required community partnerships as a condition of the grant. This might be a sign of the future...
Question from Paul D. Thacker, ES&T:
It's obvious that industry sponsored scientists would probably have some sort of bias in getting findings that support their industry funders. But what about academics? Are there any studies that have found that academics might have a bias toward getting results that show problems? This, of course, would provide them with fodder to get more grants to study the problems further.
David Ozonoff:
Yes, it is obvious about industry. And it should be obvious that academics have motives external to the "search for truth," among them the search for grants, the search for promotion, the search for reputation and the search for consulting money (much of which comes from industry). However it doesn't mean they are necessarily rewarded for showing positive results in matters of occupational and environmental health. Since many of our methods are so insensitive (epidemiology, for example; I am an environmental epidemiologist), non-positive results are much easier to obtain than positive ones and they are just as easy to publish. So the incentives here run both ways. We shouldn't forget, as the question implies, that motives are often very mixed. That is why mechanisms like codes of conduct and peer review, while imperfect, help everyone.
Question from Anonymous:
Is there a method to do a double-blind financing of research? For example: A company wanting something researched shouldn't know to which research lab their money is going, and a researcher or lab shouldn't know from where the financing for their research is coming from. This necessitates some sort of impartial third-party to broker the deal. Of course, the researcher may have a clue who is financing their research by the subject being researched.
Also, ALL data from the research should be made publicly accessible in some way. Either through the usual peer-reviewed publication, or some other presentation.
Is this at all feasible?
David Ozonoff:
What this would entail, it sounds like, is putting out RFPs (Requests for Proposals) to do work, essentially under contract, to some independent body. The Health Effects Institute intra and extramural program is a bit like that, although the source of funds and who the researchers are is known. It is an interesting idea but I can imagine a lot of practical difficulties in carrying it out. But academics (like me) are always like that. Our first impulse is to explain why something cannot be done. It would take a bit of thought, ingenuity and an unknown source of money to actually implement this idea. Also, there may be times when it would be undesirable, since knowing what the object of investigation is might be important.
Lila Guterman (Moderator):
We're about halfway through the hour. If you've got more questions for David Ozonoff, or comments you'd like to make, please send them in.
Question from ralph allan, university of california, irvine:
Is it time, again, to review the benefits and limitations of a science court to assist in narrowing the science issues before litigation "snowballs"?
David Ozonoff:
We used to hear a lot about "science courts" but they have fallen out of favor because they are essentially unworkable. Virtually every issue of Science or Nature has fifteen serious disagreements registered in it. Thus there is likely to be a lot of "sample error" connected with exactly who is selected. Finding the truly "unbiased" scientist would be difficult because they would be the ones who knew the least about the subject. Science courts are a bad idea (in my opinion) whose time has thankfully gone.
Question from MBC, large research university:
Up to what point are these concerns part of a larger problem? That is, universities in general, and public universities in particular, are becoming increasingly more dependent on private support for their survival, thus isn't it irrelevant whether support for research or expert testimony goes directly to the scientist of the university? Isn't the problem that universities in general are now part and parcel of "big business" and operate with a business (and conservative) mindset? - research design and results are always subject to interpretation. Who has a voice on those interpretations is the larger issue.
David Ozonoff:
Our colleague Sheldon Krimsky has written extensively on this topic and I would refer you to his latest book, Science in the Private Interest. Making large gifts to universities has the effect of calling the research agenda. World-class scientists are not going to fudge their data because Monsanto has paid for it, but Monsanto now gets to direct what they are and aren't interested in. Academics who "just want to do their science" become part of that strategy, often unwittingly.
The question of who "interprets" the science depends on the purpose. In this field, it is done by policy makers and politicians, hence it is a political act and influenced accordingly (thus industry gets another bite at the apple).
Question from Paul English, CA Dept of Health:
I believe that the School of Public Health at UC Berkeley recently voted that faculty should not accept research funding from the Tobacco industry. What is your take on this? Does this restrict academic freedom? Or alternatively, should this be expanded to other sources (e.g. the asbestos industry?)
David Ozonoff:
We had that debate here at our School. Personally, I am not in favor of restricting allowed fields of research or sources of funding. I am in favor of having adequate safeguards on how this is done. No one here takes Tobacco Industry money. But if we apply the same principle to other sources (e.g., involving reproductive rights) it could have some very untoward effects. Better to live with some of the bad effects and not get on a slippery slope with unintended consequences. But that may be a minority opinion when it comes to the tobacco funding issue.
Question from Lila Guterman, Moderator:
Dr. Ozonoff, my article discussed pressure on academics from industry. Do you think epidemiologists working for the government experience the same types of pressure? And how have industry interests affected OSHA and other regulators?
David Ozonoff:
In the past my impression is that government epidemiologists were reasonably independent. That has changed in this administration in several ways. A tight grip on releasing information, a general atmosphere that is hostile to labor and consumer interests and friendly to industry, a lack of aggressive leadership and support at the highest levels that make it more difficult to pursue questions that might be politically sensitive. This is certainly true at the federal level and in many states (Massachusetts is an example). The tendency is to encourage epidemiologists to "make the problem go away" which is relatively easy to do technically in this field. At the same time there is little encouragement to be tenacious and innovative and with staff cuts little opportunity to be so.
OSHA is now a lapdog agency, not a watch dog agency. It is a shell of its former self.
Question from ralph allan university of california irvine:
Shouldn't there be more critical peer review of published material (corporate or academic) to reduce the seemingly excessive publish or perish syndrome (or publish to win syndrome) and help fine tune the issues of significance?
David Ozonoff:
To ask the question in this way is virtually to answer it. But there is an elaborate system of incentives that encourage the "least publishable unit" practice. Editors often don't rebel because they need manuscripts to publish. Hence . . .
Peer review works well sometimes, but often fails. Our colleague Charlie Poole used to say, real peer review begins after publication. I guess I agree.
Question from Ilise L Feitshans GWU SPHHS:
As another commenter noted, this one-sided approach to funding is chronic in occupational health. In the 1930's and 1940's, it was because only industrial sources such as corporate medical directors possessed the information, and not until there was a disaster of the magnitude of Gualey Bridge were problems known. In the 1970's, popular pressure for MSHA and OSH Act came on the heels of legislation to protect the environment, such as TSCA and later Superfund. The public knew that problems existed, but then only indistry had the money for major studies.
So my question is, do we accept this as a chronic feature of the field, embraced in our working assumptions (as did organizations such as the early AIHA meetings) or do we think about ways to pro-actively change the embedded nature of industry's role in our research models?
The ACOEM Code of Ehtics is rather weak in this regard. First of all, it is only about 13 sentences long; there is no textual interpretation by its authors or the ACOEM governing boards to support detailed analysis of its terms. The one relevant sentence (7) "Communicate to individuals and/or groups observations and recommendations concerning their health and safety' is arguably outweighed by the duty of confidentiality in sentence 5." No disucssion provides official guidance about balancing these competing goals.
So my question is: Do we need to recognize this embedded feature of our field, and if so is there a creative approach that can change it? I think this will ultimately lead us back to questioning why the field exists, which by the way, my answer would be three words: work+health=survival. No society exists without work and no worker survives without good health. Our field is therefore, in my opinion, essential to the perpetuation of civilization and posterity.
David Ozonoff:
Your question strikes me as one that is rather fully theory laden (for example, the term embedded nature), so an adequate answer would have to engage the theory involved. I can think of a number of theoretical pespectives here and I am not sure which you subscribe to (and I'm not sure which one I subscribe to, either!). So rather than give you a superficial answer to your deep question, let me acknowledge what I think is a major failing of our field, the lack of deep thinking about what we are about and what we are doing. We just have not been willing to do the hard theoretical labor we need to do.
Not an adequate answer to a hard question but that's my take on it.
Question from Paul D. Thacker, ES&T:
I'm wondering if there are lessons to be learned from Europe. My impression from speaking with researchers is that they do not have as much industry influence as here in the US. Any thoughts?
David Ozonoff:
My European colleagues say that the pressures and the evolution are much the same there as here but they have not reached the same level of development and are less intense. But the problems remain the same. For example, the EPIC study (on obesity and cancer) now housed at the International Agency for Research on Cancer (IARC) is being moved, at industry request, to a university they believe will be more friendly and less independent.
Comment from Paul English, CA Dept of Health:
Responding to the issue of state scientists, state health departments are rapidly moving to extramural sources of funds also (e.g. our branch has lost 50% of our state staff over the last ten years; now about 70% of our staff is extramurally funded) but they are usually NIH, EPA, or CDC funding.
The public is there to protest if the state health dept takes industry funds, I think there is a higher standard of transparency and the state in general tries to stay away from issues that will cause media controversy. Academia in some ways doesn't have a mandate to be so accountable to the public - the highest goal is academic freedom, and this must be preserved over all else, even potential conflicts of interest.
Since the public is free to critizice the goverment as inefficient and unresponsive, and industry is free to critizice the govt as making too many burdensome regulations that hamper economic progress, the state science system is slowly being degraded. The public then votes to reduce revenues (by not approving taxes) that end up resulting in further layoffs of state employees.
Essentially, the result of this is that the public contributes to the degradation of an institution which really could be the best advocate for the best unbiased science for the public. There are no natural advocates for public employee scientists in this atmosphere.
Question from Ralph Allan, University of California, Irvine:
If the issue is the scientists themselves, i.e., money influencing the results presented by scientists, science court selection problems,expert witness bias, et al, are there any mechanisms to moderate the problem?
David Ozonoff:
I wouldn't say the problem is the scientists themselves as much as the nature of science itself. It is not a majority rules enterprise. Moreover, the function of science in the courtroom is not the same as it is in our disciplines. The function in the courtroom is largely to help with the assignment of blame or responsibility (although we don't say this to the witnesses or the jury).
I think the mechanisms will vary. For court cases, I am in favor of letting the jury hear both sides. The cross-exam is the adversarial version of peer review.
Lila Guterman (Moderator):
That's all the time we have. Thanks to everyone who joined our discussion today, and thanks especially to David Ozonoff for taking these questions.