Is There an (Unbiased) Doctor in the House? Journalists Often Forget That Conflicts of Interest Might Bias the Opinions of Their Expert Sources

Observations

medicine and the media Is there an (unbiased) doctor in the house? Journalists often forget that conflicts of interest might bias the opinions of their expert sources. Jeanne Lenzer and Shannon Brownlee explain how, in a bid to disentangle commercial messages from science, they have compiled a list of around 100 independent medical experts that reporters can turn to

Ho hum, another medical scandal in the try ties but only within the past five years. news. Earlier this month US Senator Chuck Despite their recent commercial ties, these Grassley announced his intention to investi- experts are included in the list because they gate Alan Schatzberg, chairman of the psy- have provided key insights into the inner chiatry department at Stanford University workings of partnerships between physicians and the incoming president of the American and the industry—and thus have bitten the Psychiatric Association, about his multimil- hand that feeds them, in effect. lion dollar interest in Corcept Therapeutics, The reaction to the list, which has been a company that is seeking to market a drug embraced enthusiastically by our fellow that Dr Schatzberg is researching with federal reporters and roundly condemned by sev- funding, and the extent to which he disclosed eral allies of the drug industry, suggests that and was required to disclose that interest to the effect of simply gathering these names Stanford.1 In June the New York Times broke together could well go beyond making life a a front page story about the alleged failure of little easier for our fellow journalists. three top research psychiatrists at Harvard, each of them a proponent of drug treatment Seeking unbiased sources for psychiatric conditions in children, to report Where do the media get their medical stories from? The need for such a resource is evident from that since 2000 they had collectively received studies showing that bias and poor report- more than $4.2m (£2.1m; €2.6m) from vari- the “gotcha” tradition of journalism, the list’s ing on medical topics are widespread in the ous drug companies.2 members are not physicians on the take but popular media. Gary Schwitzer, a professor After ignoring the growing controversy over rather the reverse: they are leading independ- of journalism at the University of Minnesota, conflict of interest for years, journalists now ent experts, many of them sources we have publishes HealthNewsReview.org, a website seem only too happy to expose wrongdoing cultivated over years of reporting. It includes, that reviews healthcare news for balance, in medicine. Yet when it comes to report- from journal publishing, two former editors of accuracy, and completeness. Schwitzer and ing medical news, those self same reporters the New England Journal of Medicine, the former a team of academic researchers analysed often seem to forget that conflicts of interest editor of the Western Journal of Medicine, and 500 stories published in top outlets between might also bias the opinions a senior editor of PLoS April 2006 and April 2008 for two key crite- of their expert sources. The “Because journalists fail Medicine; former advisers ria: did the journalist quote an independent media are filled with happy to seek out sources who to the US Food and Drug expert, someone not involved in the relevant talk about “medical break- can offer an independent Administration; physician research; and did they make some attempt throughs” that is based on perspective, many medical educators; researchers; to report potential conflicts of interest. The information gathered from bioethicists; epidemiolo- result? Half the stories failed to meet these stories in the popular media 3 sources with ties to the indus- are either unbalanced or gists, methodologists, genet- two very basic requirements. try. Yet simply knowing that icists, and clinicians from In another study Alan Cassels, a pharma- conflicts of interest can cre- simply wrong” various specialties; medical ceutical policy researcher at the University of ate bias doesn’t answer the whistleblowers; and several British Columbia, and his colleagues analysed question of which studies we ought to believe. medical journalists. media coverage of five prescription drugs pub- Because journalists fail to seek out sources Those applying to be on the list fill out a lished in 193 Canadian newspapers in 2000.4 who can offer an independent perspective, form affirming that they have not received Cassels, who is on our list, found that the sto- many medical stories in the popular media “any financial support in any form from phar- ries were overwhelmingly positive towards are either unbalanced or simply wrong. maceutical or medical device manufacturers the drugs: all 193 articles included at least In an attempt to disentangle commercial during the past five years” and that they don’t one drug benefit, while 68% (132/193) failed messages from science and to contribute to have other affiliations or financial involve- to mention any potential harm. Two thirds of better reporting we took a drastic step that we ments that would present a conflict of inter- the stories quoted a source by name, but only believe can go to the heart of the problem: we est. A three member board decides whether a scant 3% (5/164) included information about decided to name names. We created a list of to accept applicants. We also maintain a “page conflicts of interest for sources who were not nearly 100 international medical experts in 2” list of experts who willingly disclose their government or industry officials. a wide variety of disciplines. But contrary to conflicts of interest or have ended their indus- In the view of one list member, Arnold

206 BMJ | 26 July 2008 | Volume 337 Observations

The list John Abramson, clinical instructor, Harvard Medical School Jerome Hoffman, professor of medicine and emergency University of British Columbia, and Therapeutics Initiative, Marcia Angell, former editor in chief, New England Journal medicine, University of California, Los Angeles Canada of Medicine John P A Ioaniddis, professor and chairman, Department Thomas L Perry, clinical assistant professor, Department David Antonuccio, professor, Department of Psychiatry of Hygiene and Epidemiology, University of Ioannina of Anesthesiology, Pharmacology and Therapeutics and and Behavioral Sciences, University of Nevada School of Medicine, Ioannina, Greece, and Institute for Department of Medicine, University of British Columbia Clinical Research and Health Policy Studies, Department Bruce Psaty, professor of medicine and epidemiology, Michael J Barry, chief of general medicine unit, of Medicine, Tufts-New England Medical Center, Tufts Massachusetts General Hospital, Harvard Medical School University of Washington Cardiovascular Health Research University School of Medicine Unit Ken Bassett, professor of family practice, pharmacology, Peter Juni, head of division, Institute of Social and and therapeutics, University of British Columbia Arnold Relman, former editor in chief, New England Journal Preventive Medicine, University of Bern, and director, of Medicine Lisa Bero, professor, University of California, San Francisco Clinical Trials Unit, Bern University Hospital David Rind, senior deputy editor, UpToDate, and assistant Stephen Bezruchka, Department of Health Services and Jon Jureidini, head, Department of Psychological Medicine, clinical professor of medicine, Harvard Medical School Department of Global Health, School of Public Health Children’s Youth and Women’s Health Service, Adelaide, and Community Medicine, University of Washington, and associate professor, disciplines of psychiatry and Charles Rosen, clinical professor of surgery, University of Seattle paediatrics, University of Adelaide California, Irvine, and founding director, US Association for Ethics in Spine Surgery Laura Boylan, assistant professor, Department of Scott Kim, assistant professor of psychiatry Neurology, New York University Haya Rubin, director, research and evaluation, Palo Peter D Kramer, clinical professor of psychiatry and human Alto Medical Research Institute, California, and adjunct Phil Brewer, university medical director, Quinnipiac behaviour, Brown University, Providence, Rhode Island professor of medicine, Johns Hopkins University, Baltimore University, Connecticut; and past medical safety fellow, US Barnett Kramer, associate director for disease prevention, Larry Sasich National Highway Traffic Safety Administration US National Institutes of Health John Schumann, assistant professor of medicine, Howard Brody, director, US Institute for the Medical Sheldon Krimsky, Tufts University, and Council for Humanities University of Chicago, and MacLean Center for Clinical Responsible Genetics Medical Ethics, Chicago Steven R Brown, Banner Good Samaritan family medicine Stefan Kruszewski, Stefan P Kruszewski and Associates residency, University of Arizona College of Medicine Lisa Schwartz, Dartmouth Institute for Health Policy and Richard A Lange, professor of medicine, Johns Hopkins Clinical Practice, Lebanon, New Hampshire Daniel Carlat, assistant clinical professor of psychiatry, Hospital, Baltimore Tufts University School of Medicine, and editor in chief, The Gary Schwitzer, director, health journalism, MA Carlat Psychiatry Report Jeffrey Lacasse, assistant professor, Department of Social programme, University of Minnesota School of Journalism Work, College of Human Services, Arizona State University and Mass Communication Alan Cassels, pharmaceutical policy researcher, University at West Campus of Victoria, British Columbia Vera Hassner Sharav, Alliance for Human Research Dara K Lee, staff cardiologist, Presbyterian Heart Group, Protection, US Robert Cook-Deegan, director, Center for Genome Ethics, Albuquerque, and vice president, Medical Staff Affairs, Allen Shaughnessy, professor, Tufts University School of Law and Policy, Duke Institute for Genome Sciences and Presbyterian Hospital, Albuquerque Policy Medicine, Boston, Massachusetts Gretchen LeFever, director of patient safety and Sam S Dahr, Retina Center of Oklahoma Anthony So, programme on global health and technology performance excellence, Sentara, US access, Terry Sanford Institute of Public Policy, Duke John M Davis, Gilman professor of psychiatry, University of Trudo Lemmens, associate professor, Canada University, Durham, North Carolina Ilinois at Chicago Jonathan Leo, associate professor of neuroanatomy, US Robert C Solomon, American College of Emergency Raymond De Vries, professor, bioethics programme, Physicians, and medical editor in chief, ACEP News, US University of Michigan Medical School Joe Lex, emergency physician, US Joel Lexchin, professor, School of Health Policy and Des Spence, general practitioner, Glasgow, and UK Richard Deyo, Kaiser Permanente professor of evidence spokesman of No Free Lunch based family medicine, Department of Family Medicine, Management, York University, Toronto Oregon Health and Science University Abby Lippman, professor, McGill University, Montreal Sydney Z Spiesel, clinical professor of paediatrics, Yale University School of Medicine, and regular commentator Kay Dickersin, director, US Cochrane Center Peter Lurie, Health Research Group at Public Citizen, for Slate and US National Public Radio United States Mark Ebell, deputy editor, American Family Physician, and Alex Sugerman, attorney, Prescription Access Litigation, US professor, University of Georgia William K Mallon, associate professor of clinical Leonore Tiefer, New View Campaign, and New York emergency medicine, Keck School of Medicine at the Carl Elliott, University of Minnesota Center for Bioethics University School of Medicine University of Southern California, and director, Division of David J Elpern International, LAC+USC Medical Center, Los Angeles Alexander Tsai, residency training programme, Department Margaret Ewen, Health Action International, Netherlands of Psychiatry, University of California at San Francisco Peter R Mansfield, director, Healthy Skepticism, Australia Anne Rochon Ford, coordinator, Women and Health Jennifer Washburn, journalist, US Linda Marsa, freelance journalist, US Protection, Canada H Gilbert Welch, Dartmouth Institute for Health Policy and Adriane Fugh-Berman, professor, Department of Charlea Massion, Center for Education in Family and Clinical Practice, Lebanon, New Hampshire Physiology and Biophysics, Georgetown University Community Medicine, Stanford University School of Michael Wilkes, professor of medicine and director of Medical Center, and director, PharmedOut.org Medicine, and member of board of directors, American College of Women’s Health Physicians global health, University of California, and former vice dean Joseph Glenmullen, clinical instructor in psychiatry, of education and former editor in chief, Western Journal of Charles Medawar, Social Audit, UK Harvard Medical School Medicine, University of California, Davis Steven Miles, professor of medicine, Center for Bioethics, Robert Goodman, founder and director of No Free Lunch Sidney Wolfe, director, Health Research Group of Public University of Minnesota and general internist at Montefiore Medical Center, New Citizen, US York Barbara Mintzes, assistant professor, Department Steven Woloshin, Veterans Affairs Outcomes Group Merrill Goozner, director, Integrity in Science, US Center for of Anesthesiology, Pharmacology and Therapeutics, Alastair Wood, US Science in the Public Interest University of British Columbia Steffie Woolhandler, associate professor of medicine, Peter Gøtzsche, director, Nordic Cochrane Centre, Denmark Steven Morgan, associate professor and associate director, Centre for Health Services and Policy Research, Harvard University Mark E Helm, medical director, EBRx, Arkansas Evidence- School of Population and Public Health, University of James Wright, managing director, Therapeutics Initiative, Based Prescription Drug Program, and assistant professor, British Columbia Canada College of Pharmacy, University of Arkansas for Medical Sciences Ray Moynihan, journalist, Australia Gavin Yamey, senior editor, PLoS Medicine, US David Himmelstein, associate professor of medicine, Vijaya Musini, assistant professor, Department of The list can also be viewed at www.healthnewsreview.org/ Harvard University Anesthesiology, Pharmacology and Therapeutics, independentexperts.php

BMJ | 26 July 2008 | Volume 337 207 Observations

“Industry knows that buying doctors is an effective marketing tool . . . far more effective than the dollars they spend on drug representatives” List member Peter Gøtzsche

Relman, former editor in chief of the New for at least the past five years. Beliefs about on news releases from the industry’s public England Journal of Medicine and professor certain drugs or treatments were not criteria relations departments, and some even use emeritus of medicine and of social medi- for inclusion or exclusion. Indeed, the list releases as the sole source of information on cine at Harvard Medical School, such bias includes experts who sit at opposite poles of experts to interview. By offering an alternative fails to serve the public good. “People who the spectrum of beliefs on certain issues. list of highly credible, independent experts, have a financial stake in the results of clini- the industry may fear that its paid key opin- cal research can well be biased in the way Backlash and honour ion leaders7 and the professional societies research is conducted, in the way they report Within days of our announcing that we whose favour they cultivate will no longer be it, and what they say about it when inter- would make our list available to reporters the first source of medical news. viewed by the media.” the requests began pouring in. Thus far we Peter Gøtzsche, director of the Nordic have sent a copy of the list to 105 reporters, Cochrane Centre and a member of the Changing the status quo authors, and editors from such media outlets Danish group Doctors Without Sponsors, The question is why reporters seem unable as the New York Times, Newsweek, Forbes, described why he joined the list: “Industry to grasp the connection between the large Fortune, Bloomberg News, the Washington knows that buying doctors is an effective body of evidence showing that financial Post, US News & World Report, the Cana- marketing tool . . . far more effective than the conflicts of interest create bias in medical dian Broadcasting Corporation, Medscape, dollars they spend on drug representatives. research and the need for the media to seek and many other publications across the US This leads to less than optimal health care out independent sources. To be fair, journal- and several other nations. Senators and a state for patients.” ists face a daunting task when trying to sift attorney general have also requested it. Beyond the list’s usefulness to journalists, through medical research, and many are as Surprisingly, we are also receiving requests we hope that it will also be used by govern- yet unaware of the profound influence the from recognised experts who wanted to be on ment agencies, medical journal editors, and drug industry has over research results and the list. Being a member, it seems, is a badge professional societies as they seek out experts the ways in which the industry shapes medi- of honour, say several of the list members to serve as editorialists and members of clinical “truths.” Many reporters also fail to realise we interviewed for the BMJ. Others, like list cal guideline and advisory panels. The FDA, that the individuals and organisations they member Barnett Kramer, want to improve for example, has a copy of the list. We would turn to for expert commentary, such as pro- the quality of medical reporting. Kramer, a be pleased to send it to other agencies and fessional groups and charities, professional medical oncologist and associate director for professional societies. guideline authors, federal advisory panellists, disease prevention at the US National Insti- Readers can decide for themselves whether and patients’ groups, often depend financially tutes of Health, said, “Working journalists can our list of independent experts includes any on the industry. Thus there is a self reinforc- be overwhelmed by PR.” experts with “something worth saying.” ing process in which commercially sponsored The other surprise came after the publica- Jeanne Lenzer is a medical investigative journalist, researchers, whose prominence is enhanced tion of a story we wrote in the online maga- New York [email protected] by the industry’s public relations machine, are zine Slate that mentioned the list.6 Within Shannon Brownlee is a senior fellow at the New dubbed “experts,” while independent sources days bloggers were furiously accusing us of America Foundation, Washington, DC 1 Garber K. Committee questions a top psychiatrist. are cited less often. everything from biased, sloppy reporting to US News & World Report. www.usnews.com/ From informal conversations with col- being members of the Church of Scientol- articles/news/national/2008/06/26/committee- leagues we also know that other factors are ogy (which is opposed to psychiatric drugs). questions-a-top-psychiatrist.html. 2 Lenzer J. Review launched after Harvard at work when reporters fail to take conflicts Many of our critics—virtually all of them psychiatrist failed to disclose industry funding. BMJ of interest into account. Some confess that backed by the industry—opined that our list 2008;336:1327. 3 Schwitzer G. How do US journalists cover they hesitate to ask sources about any poten- was undoubtedly filled with experts who treatments, tests, products, and procedures? An tial conflicts for fear that the source will were on the payrolls of plaintiffs’ attorneys. evaluation of 500 stories. PLoS Med 2008;5:e95. take umbrage and refuse to be interviewed. (A few have testified in court cases, and those 4 Cassels A, Hughes MA, Cole C, Mintzes B, Lexchin J, McCormack JP. Drugs in the news: an analysis of Others assume that if a study appears in a who have been paid for their testimony have Canadian newspaper coverage of new prescription peer reviewed journal it must be valid. disclosed it for the list.) This venom was unex- drugs. CMAJ 2003;168:1133-7. 5 Shuchman M, Wilkes MS. Medical scientists and One of the solutions to the problem of pected, as we imagined that the list would health news reporting: a case of miscommunication. biased news reporting, in the view of Michael be viewed as a positive step towards helping Ann Intern Med 1997;126:976-82. Wilkes, professor of medicine and director reporters identify doctors and other experts 6 Brownlee S, Lenzer J. Stealth marketers: are doctors shilling for drug companies on public radio? Slate of global health at the University of Califor- who can address thorny and complex medi- www.slate.com/id/2190775/. nia, Davis, is greater transparency.5 We think cal issues without having competing financial 7 Moynihan R. Key opinion leaders: independent experts or drug representatives in disguise? BMJ the list is a step in that direction. The chief interests. Now we think we understand the 2008;336:1402-3. requirement for membership, besides a rec- backlash a little better. Cite this as BMJ 2008;337:a930 ognised area of expertise, is that the expert One of the problems recognised by Schwit- A longer version of this article is available on bmj.com. must not have taken any industry funding zer is that many journalists rely for story ideas For a related blog, visit http://blogs.bmj.com/bmj

208 BMJ | 26 July 2008 | Volume 337 observations

Yankee Doodling Douglas Kamerow Should we screen for childhood dyslipidaemia? The obesity epidemic raises the stakes in the issue of statins for children

It’s not an accident that evidence based have correlated autopsy findings with have heart attacks, investigators have guidelines more or less began with dyslipidaemia in children.1 2 So it would looked at the effects of statins on clinical preventive services. Unlike be nice to try to identify children who endothelial dysfunction and carotid treatment for problems that produce are at risk of developing heart disease, intimal medial thickness, early markers symptoms, preventive medicine is assuming that we could find them and for atherosclerosis in adults. Controlled optional. We have the luxury of time actually do something that would make studies in children show that, in to gather and evaluate the evidence a difference when they are adults. But comparison with control children, for the efficacy and even effectiveness there are a number of problems. statins improve these.14 15 So it looks as of screening tests and counselling. One is that lipid measurement in though statins can make a difference, at When someone rushes into your children is not a perfect marker for least in the short term. surgery bleeding or doubled over in present or future heart disease. Lipid But what about evidence that dozens pain, it would hardly be acceptable to concentrations vary during childhood, of years of statin treatment in children send them away untreated to await especially around puberty.3 4 They also with raised lipids will actually improve the results of a randomised controlled vary with sex and race.3 5 6 And they This is no longer cardiac outcomes in adulthood? That, of trial for their problem. But that is just don’t “track” into adulthood perfectly: “just a discussion course, is the holy grail, and such data what we do when people want to know somewhere between 30% and 50% of what to do are not available. It is likely that they whether they should undergo computed of children with raised cholesterol with a very small never will be, at least for the foreseeable tomography to screen for lung cancer concentrations won’t have them as group of children future. And that is a big problem. 7-10 or be given vitamins to prevent heart adults. with an autosomal I think the obesity epidemic really disease. “Sorry,” we say, “insufficient A further problem is the treatment raises the stakes in this discussion. This dominant evidence.” for children with raised lipids. Exercise is no longer just a discussion of what to And this is even truer for children—at and diet management work, but only genetic disorder do with a very small group of children least when the question is whether to in research settings. It’s very hard in that virtually with an autosomal dominant genetic screen them for early signs or symptoms the real world to get an individual child guarantees disorder that virtually guarantees of adult diseases. Firstly, we need to to eat better, exercise more, and lose disastrous cardiac disastrous cardiac outcomes as know whether the problem will even weight—and to maintain all of that until outcomes as adults adults. Now we are moving to mass persist into adulthood. Secondly, do adulthood. And, as usual, long term population screening and treatment we have a safe and effective treatment? studies that follow such children until of a rapidly increasing number of fat And most importantly, does treating the they are old enough to have cardiac children. Most of them will not have problem in childhood have any effect on related health outcomes are almost familial hypercholesterolaemia, and we clinical outcomes in adulthood? impossibly difficult to do. ” really don’t know what we are doing by Which brings us to the case But the real controversy behind treating them for 50 years with statins. of dyslipidaemia in childhood. these new guidelines is drug treatment. The obesity epidemic is real. We New guidelines for screening and In a striking departure from previous don’t have to just stand by and watch it treatment from the American Academy recommendations the American progress. We can and should improve of Pediatrics have caused a lot of Academy of Pediatrics endorses the use many things, including food and controversy in the United States (BMJ from the age of 8 of statins for children exercise policies in schools, the built 2008;337:a813). The paediatricians who have raised lipid concentrations environment, and families’ diets and recommend screening with a fasting that haven’t responded to diet, weight physical activity. But I’m very wary lipid profile every three to five years for reduction, and exercise. Admittedly this of committing a generation of obese all children aged 2 to 10 years who are will be a small subset of all children; but children to a lifetime of drug treatment overweight or have diabetes or a family commitment to what is likely to be at on the basis of pathological markers for history of cardiovascular disease. It’s least 50 years of statin treatment raises possible future disease. reasonable to ask, especially as the many questions. This is preventive medicine, after all. epidemic of childhood overweight and Do statins lower lipid concentrations Without good evidence, rather than say, obesity has increased the number of in children? Yes, they do. Short term “Don’t just stand there—do something,” children who will be screened, what this clinical trials of children with familial I’d advocate the opposite. screening will accomplish. hypercholesterolaemia have found Douglas Kamerow is chief scientist, Does heart disease start in statins to be safe and effective in RTI International, and associate childhood? It probably does, as lowering concentrations of low density editor, BMJ [email protected] autopsies of children who die from other lipoprotein (LDL).11-13 What about Cite this as: BMJ 2008;337:a886 causes have found. And some studies clinical outcomes? As children don’t References are on bmj.com

BMJ | 26 JUly 2008 | Volume 337 209