Ghostwriters, Data Manipulation and Dollar Diplomacy: How Drug Companies Pull the Strings in Clinical Research

Abstract

A massive study of hypertension medication that enrolled more than 42,000 patients concluded earlier this decade with good news: The cheapest of the 4 drug types tested, diuretics, were the best option for most people. At the time the number 1 brand name drug to treat hypertension, the calcium channel blocker Norvarsc, cost $1.50 to $2 for a one-a-day pill. Generic brands of diuretics cost about 10 cents a day.The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial, or ALLHAT study, epitomizes all that is good about clinical trial research –gather the leading drugs for a various ailment, test them, and determine the best option for public health.Such trials are, alas, an anomaly in the current landscape of clinical trial research.That's because drug companies hold undue influence in the conduct of research trials, which may lead to misrepresentation of research data, and manipulation of clinical studies as well as review articles. Such influence often remains cloaked, but not always. Most recently, during litigation involving Merck's drug Vioxx, a pair of examples were brought into clear view. Reported in April by the Journal of the American Medical Association, one of 2 articles outlines the extensive practice of ghostwriting employed by Merck, and the second shows how the company appears to have misrepresented Vioxx's mortality risk to the US Food and Drug Administration.“The profession of medicine, in every aspect–clinical, education and research–has been inundated with profound influence from the pharmaceutical and medical device industries,” the journal's editor in chief, Catherine DeAngelis, MD, thundered in a JAMA editorial. “This has occurred because physicians have allowed it to happen, and it is time to stop.”A growing body of scholarship has cataloged the magnitude of drug company influence on clinical trials. This has not only proven embarrassing for the medical profession, and here emergency physicians often find themselves on the front lines of patient interaction, but it also may have begun to undermine public confidence in the practice of medicine. The latest bombshell landed in the New York Times recently with revelations that a prominent Harvard child psychiatrist received $1.6 million from drug makers between 2000 to 2007, much of it unreported, according to a Congressional investigation. Joseph Biederman's research, often industry funded, has been instrumental in broader use of antipsychotics in children.Now, with the Merck examples coming to light, some physicians are re-examining what they, professional societies, medical schools and the government can do to ensure that there are fewer conflicted clinical trials, which serve the aims of profit rather than evidence-based medicine, and more ALLHAT-like studies.Pharma Funding More StudiesThe proportion of clinical studies funded by industry has risen steadily in recent decades. Physicians cite various reasons–the National Institutes of Health has come to view drug trials as the financial responsibility of drug companies, cash-strapped medical schools are seeking new sources of funding, the drug companies themselves view studies as an opportunity to extend their marketing reach. None dispute the overall trend, however. A 2006 article in the British Medical Journal by Nikolaos Patsopoulos and colleagues found that, for the most frequently cited papers in medicine from 1994 to 2003, 65 of the 77 articles had funding from industry, and that the proportion increased over time. For frequently cited trials after 1999, 31 of 32 were funded by industry, 18 exclusively so.Studies funded by drug companies, in general, produce more favorable results for the company that bankrolled the study than an independently funded study would have. One analysis, led by Joel Lexchin and colleagues and appearing in a 2003 British Medical Journal study, found no difference in the quality of studies funded by pharmaceutical companies, and those by independent sponsors. However, the researchers found that the pharmaceutical company-sponsored studies were 4 times more likely to return results favorable to the sponsor than were studies with other sponsors. The medical literature is littered with similar findings.A particularly startling analysis, conducted by Lisa Bero and others, and published last year in PLoS Medicine, reviewed 200 statin drug comparison studies conducted between 1999 and 2005. The researchers found that if the reported results favored the test drug, the trial was some 20 times more likely to be funded by the statin maker rather than the comparison drug company. Further, they found that if the study author conclusions or interpretation of the test drug were favorable, the trial was 35 times more likely to be sponsored by the statin maker rather than the comparison drug.“This finding is simply too big to dismiss by chance,” said Lisa Bero, PhD, a University of California, San Francisco professor of clinical pharmacy and health policy studies.So how do the drug companies do it? Dr. Richard Smith, MD, a former editor-in-chief of the British Medical Journal, says they employ a variety of techniques. Among them: Conduct a trial of their drug against a treatment known to be inferior; trial their drugs against too low a dose of a competitor; use multiple endpoints in the trial and publish those that give favorable results; and conduct multi-center trials, and publish the results only from centers that are favorable.“In a way, you can't blame the pharmaceutical companies,” Dr. DeAngelis said in an interview. “They are in business to make money. And if a drug company is out there funding studies, more often than not they're going to get positive results. So they're going to sell more drugs, whether patients need them or not. They're not doing this for the scientific good. If anyone tells you otherwise, they're lying.”If physicians must wrest the profession of medicine back from the drug companies, as Dr. DeAngelis contends, it will not be an easy fight.Pharmaceutical companies have a stake in nearly every group that could bring about change in the way clinical trials are funded and conducted. They buy ads in journals–spending about half a billion dollars a year on medical journal advertising. Drug companies also sponsor “supplements,” single-topic, stand-alone volumes, as well as purchasing large numbers of article reprints. Through their generous support of medical journals, drug companies generally fund a substantial proportion of medical society activities. Drug companies also sponsor large numbers of continuing medical education.Blurring the Lines“They are financially involved in every aspect of drug development,” said Sheldon Krimsky, PhD, a Tufts University professor who wrote the book Science in the Private Interest. “They have so many controls over the research and the marketing of drugs that I would say the drug companies are like a vertically integrated research company.”In addition to their role on the medical side, pharmaceutical companies have deep roots on Capitol Hill, where the drug manufacturing industry has had one of the largest lobbying efforts for years. Prior to the soft money ban, following the 2002 election cycle, pharmaceutical companies donated about $20 million to Congress and Presidential contenders every 2 years. Following campaign finance reform, the total has fallen to about $10 million every 2 years, according to the Center for Responsive Politics, which tracks political contributions.Marcia Angell, MD, the former editor-in-chief of the New England Journal of Medicine, and now a senior lecturer at Harvard Medical School, summed up drug company influence, and the possibility for clinical trial reform, succinctly: “The short answer is that not much will change as long as the pharmaceutical industry owns Congress.”The most recent problems with Merck's influence on clinical studies, reported in JAMA, became public only as a result of litigation involving Vioxx after Merck removed the drug from the market in 2004.The first of the 2 studies, led by Joseph Ross of the Mount Sinai School of Medicine, studied Merck's ghostwriting practices using about 250 court records from the lawsuits filed against the company. The researchers found that, for 16 of 20 early clinical trials on Vioxx, a Merck employee wrote the first draft, but the final paper carried an academic scientist as lead author. Such an author leaves an imprimatur of sound, rigorously conducted science.A Merck official said the company adhered to its explicit policies for authorship, and that the lead authors of papers in the clinical trials were “intimately involved in the studies.”The second JAMA article, led by Bruce Psaty, MD, of the University of Washington, found that a Merck report to the FDA may have sought to minimize the mortality of Vioxx by using an “as-treated” analysis rather than a more proper “intention-to-treat” analysis.“They didn't honestly report the trials to the FDA and the public,” Dr. Psaty said. “There were 2 trials back in April, 2001, and they showed an increase in total mortality. So there was a finding and a replication, and they blew it off to the FDA as a chance fluctuation.”Merck said a “full, unbiased evaluation of the Merck documents shows significant errors in the conclusions put forward by the authors of the JAMA articles.”Medical professionals who have studied the conflict-of-interest issue say, if change is to come, it will have to come from all sectors–journals, medical societies, medical schools and the government.Medical journals have sought reform for some years. The International Committee of Medical Journal Editors has, for example, sought to remove ghostwriting practices and pushed for full disclosure from authors.“We will continue to monitor and improve our policies on this,” said Michael Callaham, MD, Annals' editor-in-chief. “As the JAMA editorial points out, this is a huge problem that spreads all across medical research and practice. So far the measures used against it have been mostly trivial. It will continue to be a problem until the structure of payment of support to researchers, medical schools, and others is radically changed.”The changes enacted so far, which have been modest, attempt to increase the disclosure of ties between researchers and industry. But these efforts rely upon honesty from researchers, and good faith from industry. There's evidence this may not always happen, however, as sometimes the funding sources for research and review articles may be obscured. In 2007 Boston University researcher Michael Holick wrote a review article in The New England Journal of Medicine that suggested a few minutes in the sun or tanning booth would help avert the consequences of vitamin D deprivation. Under funding sources, the “UV Foundation” was mentioned. The foundation is an offshoot of the Indoor Tanning Association.“In this case, disclosure really didn't work very well,” Dr. Krimsky said.Journals Shed Light on the IssueMedical journals have probably made the strongest effort to address the problem. They have published hundreds of studies outlining the effects of drug company funding on study outcomes, and have devoted thousands of pages to developing policies to curb that influence. But they could do more, Dr. Krimsky and others say. The pharmaceutical industry has taken great heat for not publishing negative clinical trials, but such results are not something journal editors are wild about either, said David Korn, MD, a senior vice president for biomedical and health sciences research at the Association of American Medical Colleges, and the former Stanford University vice president dean of medicine.The burden falls as much on medical societies and medical schools, Dr. Krimsky said. Societies could adopt policies that establish greater firewalls between drug companies and academic evaluations of their products, and then lobby Congress furiously for their support. And medical schools could do more in their research contracts between academic scientists and drug companies.“Universities are hungry, and that's a mild term, to find new sources of funding,” Dr. Krimsky said. “The competition between universities can be very intense, and it's very difficult for universities to give up funding. There have to be some counter forces so that, if they do something that allows undue pharmaceutical influence on a study, professional societies and the government will look down upon them.”In the end, then, serious reform of clinical trials will likely have to involve some form of government intervention.The FDA Modernization Act of 1997 required the National Institutes of Health to establish a registry of clinical trials, ClinicalTrials.gov, for both federally and privately funded trials “of experimental treatments for serious or life-threatening diseases or conditions.” Some studies have found the database to be far from complete, however.And in 2007 Congress created the Reagan-Udall Foundation to, among other tasks, identify and address the safety and effectiveness of FDA-regulated products. Some advocates hope this foundation could help reform clinical trials. The organization, however, has already come under question from such groups as the Union of Concerned Scientists and the Consumer Union for the potential of undue industry influence in regulatory decisionmaking. In other words, Congressional stabs at reform have not thus far been radical.“I think we'll get to a point where any clinical trial has to be made public,” Dr. Korn said. “There's just an ethical imperative, and I think in the end we'll get there. But I don't think we're there yet, and I don't think we're going to get there very fast.”Several medical scholars who have studied the conflict of interest issue say such full disclosure and clinical trial reporting goals are relatively low-hanging fruit. The big prize, they say, is somehow disentangling clinical trials of a particular drug completely from the drug's maker.Proposed Federal OversightThe proposals for such a system vary, but most suggest the creation of a new federal agency, such as a National Institute of Drug Testing. Pharmaceutical companies would submit their drugs for testing to the agency, which would screen researchers for conflicts of interest, and determine the best method to test the new medicine. Such an agency would likely place a premium on ALLHAT-like studies. It is, indeed, no coincidence that ALLHAT was funded in large part by the National Heart Lung and Blood Institute.Where would funding for such an agency come from? Perhaps a tax on drugs, or pharmaceutical companies could be required to pay funds into a general pot of money every time they submit a drug for testing to the agency. Under the current system of drug company influence, most academic researchers admit such a system may be a long shot.“I don't think, politically, it's likely to happen, although many people think it's the logical move at this time,” Dr. Krimsky said.Still, a similar model has been adopted in Italy, said Dr. Bero, the UC-San Francisco researcher. Italy's Italian Pharmaceutical Agency (AIFA) charges a fee to all companies that sell medicine in the country, and uses the funds for head-to-head studies and meta-analyses that provide meaningful clinical information. The program is only about 3 years old, Bero said, but Spain has also recently adopted the system.A key factor that made the development palatable to companies selling in Italy, she added, is the country's national health care system.As the United States lacks such a system, it's not hard to call a national drug testing agency a long shot. At least not until there are more Merck-like revelations about the state of clinical trial research, and public condemnation of the current practice, said Howard Brody, MD, director of the Institute for Medical Humanities at the University of Texas Medical Branch, and author of the book Hooked: Ethics, the Medical Profession, and the Pharmaceutical Industry.“An NIH-style federal agency assumes that the industry will foot the bill, mostly because the US taxpayer does not choose to pony up the funds needed to be sure that science is truly conducted in the public interest and not by special commercial interests,” Dr. Brody said.“Sometime down the road I would hope that we taxpayers come to understand that you get what you pay for.” A massive study of hypertension medication that enrolled more than 42,000 patients concluded earlier this decade with good news: The cheapest of the 4 drug types tested, diuretics, were the best option for most people. At the time the number 1 brand name drug to treat hypertension, the calcium channel blocker Norvarsc, cost $1.50 to $2 for a one-a-day pill. Generic brands of diuretics cost about 10 cents a day. The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial, or ALLHAT study, epitomizes all that is good about clinical trial research –gather the leading drugs for a various ailment, test them, and determine the best option for public health. Such trials are, alas, an anomaly in the current landscape of clinical trial research. That's because drug companies hold undue influence in the conduct of research trials, which may lead to misrepresentation of research data, and manipulation of clinical studies as well as review articles. Such influence often remains cloaked, but not always. Most recently, during litigation involving Merck's drug Vioxx, a pair of examples were brought into clear view. Reported in April by the Journal of the American Medical Association, one of 2 articles outlines the extensive practice of ghostwriting employed by Merck, and the second shows how the company appears to have misrepresented Vioxx's mortality risk to the US Food and Drug Administration. “The profession of medicine, in every aspect–clinical, education and research–has been inundated with profound influence from the pharmaceutical and medical device industries,” the journal's editor in chief, Catherine DeAngelis, MD, thundered in a JAMA editorial. “This has occurred because physicians have allowed it to happen, and it is time to stop.” A growing body of scholarship has cataloged the magnitude of drug company influence on clinical trials. This has not only proven embarrassing for the medical profession, and here emergency physicians often find themselves on the front lines of patient interaction, but it also may have begun to undermine public confidence in the practice of medicine. The latest bombshell landed in the New York Times recently with revelations that a prominent Harvard child psychiatrist received $1.6 million from drug makers between 2000 to 2007, much of it unreported, according to a Congressional investigation. Joseph Biederman's research, often industry funded, has been instrumental in broader use of antipsychotics in children. Now, with the Merck examples coming to light, some physicians are re-examining what they, professional societies, medical schools and the government can do to ensure that there are fewer conflicted clinical trials, which serve the aims of profit rather than evidence-based medicine, and more ALLHAT-like studies. Pharma Funding More StudiesThe proportion of clinical studies funded by industry has risen steadily in recent decades. Physicians cite various reasons–the National Institutes of Health has come to view drug trials as the financial responsibility of drug companies, cash-strapped medical schools are seeking new sources of funding, the drug companies themselves view studies as an opportunity to extend their marketing reach. None dispute the overall trend, however. A 2006 article in the British Medical Journal by Nikolaos Patsopoulos and colleagues found that, for the most frequently cited papers in medicine from 1994 to 2003, 65 of the 77 articles had funding from industry, and that the proportion increased over time. For frequently cited trials after 1999, 31 of 32 were funded by industry, 18 exclusively so.Studies funded by drug companies, in general, produce more favorable results for the company that bankrolled the study than an independently funded study would have. One analysis, led by Joel Lexchin and colleagues and appearing in a 2003 British Medical Journal study, found no difference in the quality of studies funded by pharmaceutical companies, and those by independent sponsors. However, the researchers found that the pharmaceutical company-sponsored studies were 4 times more likely to return results favorable to the sponsor than were studies with other sponsors. The medical literature is littered with similar findings.A particularly startling analysis, conducted by Lisa Bero and others, and published last year in PLoS Medicine, reviewed 200 statin drug comparison studies conducted between 1999 and 2005. The researchers found that if the reported results favored the test drug, the trial was some 20 times more likely to be funded by the statin maker rather than the comparison drug company. Further, they found that if the study author conclusions or interpretation of the test drug were favorable, the trial was 35 times more likely to be sponsored by the statin maker rather than the comparison drug.“This finding is simply too big to dismiss by chance,” said Lisa Bero, PhD, a University of California, San Francisco professor of clinical pharmacy and health policy studies.So how do the drug companies do it? Dr. Richard Smith, MD, a former editor-in-chief of the British Medical Journal, says they employ a variety of techniques. Among them: Conduct a trial of their drug against a treatment known to be inferior; trial their drugs against too low a dose of a competitor; use multiple endpoints in the trial and publish those that give favorable results; and conduct multi-center trials, and publish the results only from centers that are favorable.“In a way, you can't blame the pharmaceutical companies,” Dr. DeAngelis said in an interview. “They are in business to make money. And if a drug company is out there funding studies, more often than not they're going to get positive results. So they're going to sell more drugs, whether patients need them or not. They're not doing this for the scientific good. If anyone tells you otherwise, they're lying.”If physicians must wrest the profession of medicine back from the drug companies, as Dr. DeAngelis contends, it will not be an easy fight.Pharmaceutical companies have a stake in nearly every group that could bring about change in the way clinical trials are funded and conducted. They buy ads in journals–spending about half a billion dollars a year on medical journal advertising. Drug companies also sponsor “supplements,” single-topic, stand-alone volumes, as well as purchasing large numbers of article reprints. Through their generous support of medical journals, drug companies generally fund a substantial proportion of medical society activities. Drug companies also sponsor large numbers of continuing medical education.