Commentary: When Medicine And Money Don't Mix

Do drugmakers have too much control over lab data?

GlaxoSmithKline PLC is the drugmaker in the hot seat this month. New York State Attorney General Eliot Spitzer sued the company on June 2, alleging that Glaxo fraudulently concealed data suggesting that its antidepressant Paxil might trigger suicidal thoughts and acts in kids under 18. British regulators, meanwhile, have been investigating whether Glaxo improperly withheld similar data on how the drug affects children. Glaxo says it has acted responsibly in conducting studies on pediatric patients and in disseminating that data. On June 14, Glaxo posted full pediatric study data on Paxil on its Web site.

The Paxil case, unfortunately, highlights a growing problem: Over the past two decades big drugmakers have exerted increasing control over medical research. Now they face broad criticism that they often spin the resulting data for commercial gain.

In some cases, drug companies downplay or delay the release of negative data. Critics also accuse drugmakers of designing their studies to improve the odds that their drugs will outshine competing products. And to spread the word, some pharma companies have hired consultants to write positive reviews of trials and paid doctors to sign on as authors.

Even the revered National Institutes of Health has drawn congressional scrutiny into financial ties between its researchers and companies. "The threat is to the objectivity of scientific research," says Sheldon Krimsky, a science policy expert at Tufts University and author of Science in the Private Interest. "It is reaching crisis proportions."

Pharmaceutical companies don't bury all unfavorable data. In early March, for example, Bristol-Myers Squibb announced that a study comparing its cholesterol-lowering drug to Pfizer's had shown the rival's product to be superior.

But such moves are all too rare. Take the case of the arthritis treatment Celebrex. A big product for Pharmacia in the 1990s, the drug is a $3 billion blockbuster. It got a boost in 2000 when a trial showed that it was linked to fewer ulcers than two older nonsteroidal anti-inflammatory drugs (NSAIDS). The problem was, Pharmacia's published data showed how patients fared after just six months, even though the company had gathered data for 12. The figures for the full year revealed less of a safety advantage for Celebrex -- and the Food & Drug Administration continues to require that Celebrex' label carry the same warning about stomach side effects as older NSAIDS.

A spokesman at Pfizer Inc. (PFE ), which bought Pharmacia in 2003, says a committee of medical experts determined that the 6-month data was more meaningful because a large number of patients dropped out of the study at the half-year mark. In any case, he says, the full results were presented at scientific meetings. But critics argue that all that data should be publicized, so that doctors can reach their own conclusions. "The information prescribing physicians get on the safety and efficacy of drugs is grossly inadequate," contends Dr. Sidney Wolfe, director of the health research group at Ralph Nader's watchdog body, Public Citizen.

Yet even if all trial information is made public -- a step being pushed by a variety of groups, including the International Committee of Medical Journal Editors and the American Medical Assn. -- other problems will persist. That is because corporate funding now permeates most drug-related research. One Yale University study notes that in 1980, 32% of biomedical research and development in the U.S. was funded by industry. By 2000, the figure had soared to 62%.

As the industry has tightened its grip on research, the traditional emphasis on independence and ethics in medical science seems to have eroded. The Yale study found that in industry-funded research the odds are 3.6 times higher that the results will buttress the sponsor's product than in studies by independent groups such as the NIH and various foundations.

The reason: Drug companies can design trials in ways that give their products an edge. Dr. Cary P. Gross, assistant professor of medicine at the Yale School of Medicine, says drugmakers may, among other things, select dosages that are likely to show their product in a positive light.


Independent researchers are supposed to provide a counterbalance, thwarting the drug industry's tendency to turn research studies into marketing grist. But some drugmakers appear to have found ways around that. A lawsuit brought by a whistle-blower against Warner-Lambert in 1996 alleged that the company brought in consulting firms to "ghostwrite" articles on its epilepsy drug Neurontin, and then paid physicians to sign on as the authors. Pfizer, which bought Warner in 2000, agreed to pay $430 million to settle a government suit triggered by the whistle-blower allegations. A spokesman for Pfizer says the company did not admit to the allegations in its settlement, and points out that all the activities at issue in the case occurred before Pfizer bought Warner.

It's impossible to know how prevalent ghostwriting is. But Dr. Drummond Rennie, professor of medicine at the University of California San Francisco and deputy editor of the Journal of the American Medical Assn., believes it is "pervasive, deceptive, and disgraceful."

In fact, Big Pharma's money and influence touch almost every corner of the medical industry. Consider that the NIH was compelled to set up a commission to look into conflicts of interest at the agency last February. An NIH spokesman now concedes that 118 employees at the Institutes have 196 ongoing consulting deals with corporations, primarily pharmaceutical and biotech companies.

So what's the right prescription for the questionable practices that have infected medical research? The first logical step is to establish a public registry that would carefully archive the results of all clinical trials. Drug companies have not supported such a move to date. And even if they bend, it won't be enough simply to disclose results as they are analyzed by the trial sponsors. Independent scientists -- enlisted by citizen groups, medical journals, or the government -- must have access to the raw data to ensure that conclusions are fair and accurate.

At the same time, the federal government should make more money available for follow-up trials on drugs that are already on the market. Such tests, uncontaminated by corporate funding, would provide valuable comparative data on how well drugs work -- especially in "off-label" treatments that don't involve FDA approval. Why bring in the government? Because drugmakers may resist doing expensive trials if they believe the drug may fare poorly.

The discouraging fact is that for many pharmaceutical companies, the current system is working beautifully. Few will acknowledge that major reforms are necessary. Of course, Wall Street analysts took exactly the same stance when they first came under legal fire. Big Pharma would do well to take their traumatic experiences to heart.

By Amy BarrettWith Kerry Capell in London and Susann Rutledge in New York

    Before it's here, it's on the Bloomberg Terminal.