SEPTEMBER 1999 · VOLUME 20 · NUMBER 9

Dr. Sidney M. Wolfe has been the director of Public Citizen's Health Research Group (HRG) since 1971. Since 1995 he has been an adjunct professor of internal medicine at the Case Western Reserve University School of Medicine. His awards include receiving the MacArthur Foundation Fellowship in 1990. HRG promotes research-based, system-wide changes in health care policy and provides advice and oversight concerning drugs, medical devices, doctors and hospitals and occupational health. It has been a major player in the successful efforts to ban or provide warning labels on numerous drugs and medical devices. Major HRG publications include Pills That Don't Work, Over the Counter Pills That Don't Work; Worst Pills/Best Pills, "Unnecessary Cesarean Sections: How To Cure A National Epidemic" and monthly newsletters entitled Health Letter and Worst Pills Best Pills News.

Multinational Monitor: What are the origins of the U.S. drug regulatory system?
Dr. Sidney Wolfe: The modern era of food and drug regulation begins with investigative reporters focusing on filthy meat, notably Upton Sinclair's The Jungle. In the wake of the scandals involving the Chicago stockyards, the Pure Food and Drug Act was passed in 1896. In terms of drugs, that law really didn't do anything other than define or redefine some authority that had to do with FDA's being able to take a drug off the market if enough people are killed or injured.

That was all the authority the FDA had on drugs until 1938, when another crisis was precipitated by one of the first antibiotics, the sulfa drugs that were used before penicillin was invented. One called elixir sulfanilamide was contaminated with an organic solvent, diethylene glycol which caused fatal kidney disease, and killed more than 100 people, many of them children. This crisis generated support for the idea that drugs should be checked for safety before they were approved rather than waiting until bodies start piling up.

The 1938 FDA law was focused on safety, almost exclusively. It had some provisions on advertising, but it had nothing in it about whether drugs had to be efficacious.

The major step forward in the 1962 law was the first-time requirement for evidence, from well-controlled clinical trials, that a drug be effective. All the drugs that came on the market before 1962 did not have to be shown to be effective. Many of them were not.

We published a national best-seller in 1981, Pills That Don't Work, based on data from the reviews of these pre-1962 drugs that were still on the market because the FDA hadn't moved very quickly on these drugs even though they had not been shown to be effective. Almost all of them are now off the market.

For medical devices, though, the situation was the same as it was before 1938, namely there was no requirement at all to test devices for safety. Two events precipitated the passage of the 1976 Medical Device Amendments. First, several heart valves which had not been rigorously tested, failed and killed people. Second, the Dalkon Shield caused sterility in more than 10,000 women because of pelvic inflammatory infections caused by the device. There were some further amendments in 1990 and 1992, mainly strengthening the device law. That was it for positive, consumer-protective drug and medical device regulations.

MM: But in recent years, FDA's authority has been diminished and regulatory standards weakened?
Wolfe: The industry onslaught began in the early 1990s. First came the Quayle Council on Competitiveness report - an industry wish-list recommending explicitly to weaken the FDA's authority. It did not get anywhere because of the Democratic control of the Congress.

However, in 1994, Congress passed the Dietary Supplement, Health and Education Act (DSHEA), which stripped FDA of the authority to require evidence of safety and effectiveness in herbals or other food supplements. Under DSHEA, anything without an explicit health claim isn't a drug, and therefore doesn't require testing.

Meanwhile, building on the Quayle Council on Competitiveness, using some of the recommendations of Gingrich's Progress and Freedom Foundation, as well as Boyden Gray's Citizens for a Sound Economy, the drug and device industry drafted and organized hearings around FDA deform legislation. The legislation, ironically entitled the FDA Modernization Act, eventually passed and was signed into law in 1997.

One provision of the law privatized testing of devices. If you are a device company you can go to a private, for-profit device reviewing company instead of the FDA and get your device approved.

The concern that this will happen to drugs as well as devices is certainly at the top of the list as to why the drug division of the FDA has behaved so miserably. Things at the FDA are by far the worst they have been in the last three decades.

So after 80 years of strengthening the FDA, it has gone really backwards. Apart from the legislative rollback, there is all the intimidation and worry that if the FDA doesn't perform better (from industry's point of view), then even more deform legislation will be enacted.

MM: Why has the Clinton administration agreed to the weakening of the FDA?
Wolfe: One of the reasons that the administration and the Congress, even many of the progressive elements, were willing to accept the FDA Modernization Act was that they thought it was going to be just one shot. If we could get a bill that isn't too bad now, they hoped, then we will avert the possibility of much worse legislation passing in the future. Unfortunately, that has not been the case. There are already rumblings of more legislation. And unlike the legislative hearings in the 1960s, 1970s and 1980s, where the FDA was criticized from a progressive perspective - where FDA officials were subpoenaed and FDA was asked why it made this or that mistake, now the hearings criticize the FDA for being too hard on the industry.

MM: What are the problems that arise from an inadequate drug regulatory system?
Wolfe: There are four general ways in which problems arise with prescription drugs. Initially, before the new drug application is even submitted, is the failure to adequately test products, or the failure to test them ethically. Second, even if the tests are done right, there can be a failure to properly report the results to the government. Third, even if the tests are ethically and adequately conducted and reported, the government may fail to regulate, usually for reasons having to do with pressure from industry. The fourth problem is failure to accurately and promptly inform patients and doctors of risks.

Finally, not really an FDA regulatory problem, is the issue of price-gouging. We did a study looking at three anti-psychotics and five anti-depressants in 18 countries. On average, we found prices are twice as high in the United States as everywhere else. We clearly have price gouging unlike any other country, and it is because, unlike any other developed country, we do not have national health insurance. However, even without national health insurance, the Department of Defense and the Veterans Administration are able to negotiate low prices, because they have some clout.

MM: What are examples of the failure to adequately and ethically test products?
Wolfe: One issue is the questionable use of surrogate markers. For example, short-term weight loss as a surrogate marker is meaningless if in the long term the person's weight doesn't stay down. Which it doesn't. There is no evidence from any weight-reduction drugs that people using them have any long-term health benefit, and yet over and over again these drugs get approved for short-term use. The longest period of time in this category for which a weight reduction drug has ever been approved is a year for Redux (now taken off the market because of its dangers). That is one of the reasons it sold so much; none of the other drugs have ever been studied for so long.

Similarly, calcium channel blockers are drugs for reducing high blood pressure. They are approved, as are all high blood pressure drugs, based on their ability to lower blood pressure, which doesn't necessarily have anything to do with a health benefit. If the drugs reduce heart attacks and strokes, then they have a health benefit. If they don't, there's no reason to take them for high blood pressure. None of the calcium channel blockers except one has ever been shown to reduce clinical endpoints such as heart attacks and strokes.

A key issue in ethical testing is the use of placebos in cases where an already tested and available product is known to provide benefits. A study on a calcium channel blocker done in Eastern Europe and China, for example, recruited people into the study with very high blood pressure and gave them placebos, even after we knew that available diuretics reduced heart attacks and strokes as well as high blood pressure. In tests on HIV transmission to children, HIV-positive pregnant women in Africa were getting placebos instead of AZT after it was known that AZT will reduce perinatal transmission. Based on a recent seminar at the FDA on the use of placebos, it is clear the FDA has no policy at all and still believes it is okay to give placebos to people who are psychotic or depressed.

To the extent that FDA welcomes New Drug Applications which are based on such studies, the companies are perfectly happy to do studies that way. Their drug is always going to look better against a placebo than against a competing drug that has some positive effects and which might outperform it.

MM: What have been the historic consequences of companies failing to report test results?
Wolfe: Bjork-Shiley's heart valve, Selacryn, an anti-hyperintensive, the arthritis drug Oraflex and the anti-depressant Merital are four examples - one heart valve, three drugs - of products that would not have come on the market had the company notified the FDA before approval of what the FDA discovered after approval. None of those products are around anymore. All of the manufacturers were subject to criminal prosecution.

The three drug manufacturers were successfully prosecuted - these three instances are a complete list of successful criminal drug prosecutions.

The government brought suit for the heart valve a little late under the statute of limitations, and instead settled for $25 million under the False Claims Act.

There is a larger number of other instances in which criminal prosecution should have been brought. For instance, the FDA should have brought a criminal case against Johnson & Johnson for withholding information about a high blood pressure drug called Suprol.

If one believes the Wall Street Journal, a criminal case will be brought in the Fen-Phen case against Wyeth-Ayerst and American Home Products, as it should be, for having failed to promptly send in information about heart valve damage and about primary pulmonary hypertension. Unlike systemic hypertension, pulmonary hypertension is fatal about half the time.

At least in these instances, one can't fault the FDA for making drug approval decisions unknowingly based on inadequate information. But one can fault them for not more vigorously criminally prosecuting those companies.

MM: What have been the historic consequences of FDA's failure to regulate sufficiently?
Wolfe: Raw milk, aspirin and Reye's Syndrome, tampons and toxic shock, Phenformin and lactic acidosis, Posicor and Duract are all instances where there was clear, unequivocal data at the FDA showing harm, but where the FDA failed to either ban a product or require necessary labeling.

FDA failed to ban interstate commerce in raw or unpasteurized milk, more than 100 years after Pasteur discovered that cooked milk is safer than uncooked milk. This was largely because of the influence of the biggest raw milk dairy in the world, Alta Deena, located in a Los Angeles suburb called City of Industry. After we filed a petition to ban raw milk and then sued the FDA, a federal district court judge ordered the FDA to ban raw milk within 30 days. And it has been banned since then.

In the case of aspirin and Reye's Syndrome, there were four case control studies that showed clearly that children who had chicken pox or flu who were given aspirin were at much greater risk of Reye's Syndrome. Reye's Syndrome is frequently fatal and 30 to 50 percent of the time it causes brain damage in those whom it doesn't kill. The CDC, although not a drug regulatory authority, warned that parents should not be giving their children aspirin if they had chicken pox or flu. But the FDA didn't seem to be interested in putting a warning on aspirin labels.

After we filed a petition in early 1982 to put warning labels on aspirin for Reye's Syndrome, we sued the FDA. Congressional hearings put further pressure on the FDA, and after a voluntary labeling campaign had been a complete failure, in 1986 a warning label was finally put on aspirin for Reye's Syndrome.

The same thing happened with tampons and toxic shock. In 1982, there was clear evidence from case control studies that those women who used high-absorbency tampons were much more likely to get toxic shock than women who used low-absorbency tampons. But because the tampon industry didn't want labeling, the FDA didn't do anything. We filed a petition. The FDA's response was to convene an informal group of consumer groups, industry, and some scientists to try to work something out and get a label. We boycotted that group, which predictably failed to do anything at all. After a couple of years, we sued the FDA, and again in a fairly strongly worded opinion, the judge ordered the FDA to put a warning label on tampons for toxic shock. And they've been on since then.

MM: You recently surveyed FDA officers and found widespread concern over current regulatory failures and improprieties.
Wolfe: Fifty-six drugs that came on the market in the United States, the United Kingdom, Germany and France between 1990 and 1992 were later banned for safety reasons. None of the 56 drugs were an important breakthrough drug. I studied how the United States did during that interval compared to those other countries. And the answer was that 31 drugs had been approved and subsequently were taken off the market in France, 30 in Germany, 23 in the UK and only nine in the United States for safety reasons. The United States kept a lot of the drugs off the market altogether. In those days, as opposed to now, when there was a safety question, the FDA did not approve a drug.

Earlier this year, we did a study of physicians at the FDA whose primary responsibility was looking at new drug applications. There are 172 of those physicians. We sent a detailed questionnaire out and 53, or 31 percent, responded. Although it was anonymous, we asked the physicians if they were willing to identify the divisions of the FDA they were in. Responses came from 11 of the 15 reviewing divisions, which are divided by therapeutic category. So it is a fairly broad look at the FDA. The FDA physicians who responded cited 27 instances between late 1995 and 1998 in which a drug was approved which shouldn't have been approved, but only six in which a drug should have been approved but wasn't.

Of the subset of those who have been at the FDA long enough to know what things were like three years earlier, 17 thought the standards were lower, whereas only six thought the standards were higher. This, of course is a critical issue - if the standards for safety and efficacy are lowered, more drugs are going to reach the market that wouldn't have been on the market before.

These FDA medical officers also identified 25 drugs that have been approved too fast. A larger number may have been approved only because of post-marketing studies. In the case of post-marketing studies, problems arise prior to approval but instead of not approving a drug, which should be the response, the FDA says let's do a study after the drug is approved, after it has gone on the market. These studies are rarely finished on time, and some of them are never finished at all. They are a sorry excuse for approving a drug that was otherwise too dangerous to be approved.

There has also been inappropriate pressures on medical officers in the last three years. The survey respondents reported 23 instances of inappropriate phone calls, largely, we suspect, from industry. The medical officers cited 14 instances where they were instructed not to present data or their own opinion, or an advisory committee might unfavorably influence their approval of the drug; and finally 18 said they were asked to change their opinion to agree with their supervisor.

These results explain why drugs like the high blood pressure drug Posicor and the pain-killer, Duract - despite clear evidence of dangers before approval - were approved.

Another example is a drug called Trovan -Trovafloxacin - which is in the same therapeutic category of antibiotics as Cipro, or Ciprofloxacin. There is no therapeutic advantage of Trovan over any of the other drugs in this family, and before it was approved there was evidence that 10 percent of men in the trial where it was being used to treat prostatitis developed abnormal liver tests. On that basis alone, the drug should have been ditched. It was approved despite this evidence.

Eventually enough people were damaged - 140 cases of serious liver damage occurred, including a number of deaths, patients who required liver transplants - that action was required. It was banned in a number of other countries, but the FDA couldn't quite let go of it - it would have added another drug to the list of drugs that they had made a mistake on. We petitioned in June to ban the drug, and a week later they banned it except in nursing homes and hospitals. So if you're in a nursing home or hospital, you can unfortunately still get this very dangerous drug because the FDA wanted to avoid taking it off the market.

Another drug which stayed on the market for a number of years after it should have been banned is Seldane, a widely prescribed antihistamine. For two or three years after it was crystal clear that this drug caused fatal cardiac arrhythmia when used with other drugs, it stayed on the market because the FDA was waiting for the company that made it to get another drug approved which had the benefits but not, by and large, the risks. During that period, a lot of people were taking it and a number of cases of cardiac arrhythmia and death occurred.

A more recent example of a drug for which there was clear evidence it shouldn't be approved before it was approved is the new flu drug Relenza, made by Glaxo Wellcome. You'll see massive ads for it as the flu season starts. The FDA advisory committee voted 13 to 4 not to approve that drug because it doesn't work. On a scale of 1 to 15, this drug reduced the symptom severity by one point compared with a placebo. It did not allow people to return to work significantly more quickly compared with a placebo.

The people who are of most concern in terms of flu epidemics are people with serious lung disease and heart disease and they should get flu shots. There is no evidence that this drug works at all in them. There is no evidence that it keeps people out of hospitals, and there is no evidence it reduces the serious complications of flu. There is, however, evidence that there are more adverse reactions occurring among those with serious lung disease. The implication in Relenza's advertising is that this drug will make a difference. It will, because it will make those people who are more susceptible more likely to get sick.

MM: What are the consequences of the surge in prescription drug advertising directed at consumers?
Wolfe: Consumer advertising of drugs is illegal in every country of the world, except for the United States. It didn't exist much in the United States in the 1980s. It is a serious problem because patients put pressure on their doctors after they've seen such ads, which similar to the ads directed at doctors, are misleading or false about half the time. The FDA has found something like 40 to 50 percent of all the direct-to-consumer prescription television ads violate the law and regulations, but after 30 million people have seen one of them, it is too late. It is in their mind. And they haven't exactly run corrective advertising that's of any significance. It is really a deadly form of misinformation. Another growing problem involves prescription sales over the internet, with people able to get prescriptions without a doctor or pharmacist being around.

MM: Absent a renewal of FDA's authority, how should people protect themselves from unsafe medications?
Wolfe: Because the FDA has become such a sieve, no physician should prescribe any new drug for at least five years after its been introduced on the market, unless it is a breakthrough drug. And there are not that many breakthrough drugs. In the first five years, there is a reasonable likelihood that a drug is going to be taken off the market or it is going to have to have a serious change in its warning label. Until then, doctors should not make guinea pigs out of their patients.

MM: Is there any sign that the industry sees things as going too far - that they're going to lose the benefit of an FDA seal of approval that the public trusts?
Wolfe: A little bit, but it is only retarding further legislative assaults. I think the Pharmaceutical Research and Manufacturers Association (PhRMA) probably does not want to privatize the drug-approval process. It is the gold standard, though it has been tarnished, to say the least.

I have never heard more favorable comments from the industry about the FDA than in the last year or two. I was on a podium with the FDA commissioner six months ago, and I was attacking the FDA and some guy from PhRMA was saying just how pleased they are with the FDA. That is a serious problem.