About the book

Post-Publication Notebook: Two Unanticipated Paths

The withdrawal of Vioxx nine days after Overdosed America was published catapulted me into the public arena. Less than twenty-four hours after Merck’s announcement, I was sitting on a couch talking with Katie Couric as if we were chatting in my living room, except for the wall of television cameras in front of us and the millions of viewers. Television and radio interviews took up a lot of my time in the following weeks. Three months later, a study showed that Celebrex might increase the risk of cardiovascular complications too, and there was another flurry of media activity and another appearance on the Today show.

But still I didn’t have a plan for what I would do after the post-publication hoopla died down. Just when I was feeling the need to seek an academic position or apply for a fellowship to fund ongoing research, two unanticipated paths, kind of the yin and the yang of my work, opened up.

“I was sitting on a couch talking with Katie Couric as if we were chatting in my living room.”

During my first trip cross-country to give a lecture and do publicity for the book, the woman sitting next to me pulled a medical-looking article out of her briefcase. I couldn’t help but notice that the title had something to do with using predictive modeling to contain health care costs. I told her I had some experience with health issues and asked her what the article was about. She said that predictive modeling involved using artificial intelligence to analyze medical and prescription drug claim data to identify people at greatest risk of becoming seriously ill in the following year. Once identified, she explained, specially trained nurses could contact these people and their doctors to head off impending medical disaster at the pass. Consistent with her metaphor, the woman told me that she worked for a subsidiary of Wells Fargo, and that predictive modeling was part of a package of health management services they offered to self-insured companies with the goal of controlling medical costs and improving employees’ health.

As we flew across the western half of the United States, we talked about the opportunity to combine predictive modeling, wellness interventions, and health information free of commercial bias to provide optimal health care. More than two years of work later, we made a site visit to our first client, a hospital and health center serving a Native American community in Arizona. This community has major health challenges, and provides an opportunity for us to facilitate workplace, family, and community-based interventions informed by predictive modeling and non-commercially biased health information.

“As we flew across the western half of the United States, we talked about the opportunity to combine predictive modeling, wellness interventions, and health information free of commercial bias.”

At about the same time, a lawyer in Houston who read Overdosed America invited me to present a seminar about Vioxx and Celebrex to his firm. This was the cat’s meow for me—the lawyers were interested not just in an overview, but in the smallest statistical details of my research showing that these drugs are not nearly as safe as doctors reading their trusted journals like the New England Journal of Medicine and the Journal of the American Medical Association had been led to believe. When I was doing the original research on Vioxx and Celebrex for the book, I often felt as if I were writing a legal brief. But I never imagined that lawyers would actually pay me to show how the drug companies were using distorted science to convince doctors to prescribe their drugs.

In the fall of 2006 I was asked to make a similar presentation about Celebrex and Bextra, this time at a special “science tutorial” for a federal judge and two state judges who would soon be hearing cases involving these drugs. The judges, sitting in the federal courtroom in San Francisco, listened intently and interrupted my presentation frequently with very thoughtful questions. They clearly wanted to understand the science and the marketing of these two drugs. What a thrill: having found the truth about Vioxx and Celebrex on the FDA Web site while I was still a practicing family doctor, having spent three years piecing together how the drug companies produce science that is favorable to their products (sometimes going further than just bending the rules of science), and now sharing my findings with judges playing a pivotal role in litigation that will determine just how far the drug companies can go in exaggerating the benefits and minimizing the risks of their drugs before they risk having to pay a prohibitive price.

In addition to these projects, I continue speaking around the country to medical students, doctors, and general audiences, and continue with academic writing and research. Though I had no preconceived plan, I could not have asked for a more engaging or productive combination of projects. What began as a passion and an intellectual challenge has now become an ongoing mission. Image

“The judges, sitting in the federal courtroom in San Francisco, listened intently and interrupted my presentation frequently with very thoughtful questions”

Medical Developments An Update

VIOXX

Merck pulled Vioxx from the market after its own APPROVe study was stopped prematurely because the people taking Vioxx were developing almost twice as many heart attacks, strokes, and blood clots as people taking placebos. Merck was praised for withdrawing Vioxx as soon as it learned about the interim results of the APPROVe trial in September 2004. But reports now available to the public on the FDA Web site show that exactly the same situation had existed in November and December 1999 in Merck’s VIGOR study. In this study (discussed in chapter 3), Vioxx was being tested against naproxen (sold over the counter as Aleve) in patients with rheumatoid arthritis. Just like in the APPROVe study, the committee monitoring the safety of the people participating in the study was informed that there were “excess deaths and cardiovascular adverse experiences,” nearly twice as many, in the group that was obviously taking Vioxx. But unlike the APPROVe trial, not only was the VIGOR study not stopped, but the New England Journal of Medicine article that reported its results didn’t even report that there were more serious cardiovascular complications caused by Vioxx than there were serious GI complications prevented. Instead, the article presented just the results for heart attacks.

“Not only was the VIGOR study not stopped, but the New England Journal of Medicine article that reported its results didn’t even report that there were more serious cardiovascular complications caused by Vioxx than there were serious GI complications prevented.”

Even so, the article’s claim that Vioxx significantly increased the risk of heart attack only in people who already had cardiovascular disease turned out not to be true. In December 2005 the editors of The New England Journal of Medicine issued an “Expression of Concern,” stating that the November 2000 article presenting the results of the VIGOR trial had failed to include three heart attacks that occurred among people taking Vioxx. Why did three heart attacks matter so much? They changed the statistical analysis dramatically: with the three heart attacks included, Merck could no longer claim that Vioxx did not increase risk of heart attacks for healthy people without a history of cardiovascular disease. In other words, the VIGOR trial showed that treating people with Vioxx instead of naproxen significantly increased everyone’s risk of heart attack, whether or not they had a previous history of cardiovascular problems.

Had these true results been published in The New England Journal of Medicine in November 2000, Merck probably would have sold a lot less Vioxx. And Merck would probably now be facing a lot fewer than the 27,000 lawsuits that have been filed by people who suffered cardiovascular complications while taking Vioxx.

STATINS

On February 2, 2007, Judy Norsigian, the executive director of Our Bodies Ourselves, and I had an op-ed piece in the Atlanta Journal-Constitution. The piece started out:

Today’s “Go Red for Women Day” campaign to increase awareness of heart disease and stroke should be called “See Red for Women Day.”

Despite being endorsed by some of our most trusted institutions like the American Heart Association and the National Heart, Lung, and Blood Institute, this event isn’t just about protecting your health. It’s also about exploiting your concerns about health.

The purpose of the “Go Red” campaign was, ostensibly, to “educate” women that heart disease, while usually silent, is their number-one killer (scary stuff) and to encourage them to “know their numbers,” including their cholesterol numbers. The truth is that below the age of seventy-five, cancer claims seventy-eight percent more lives of women than heart disease. And it also turns out that—as described in chapter 9 of Overdosed America— there still has never been a clinical trial showing that statins are beneficial to women of any age or men over sixty-five who do not already have heart disease or diabetes.

Dr. Jim Wright, the director of the Therapeutics Initiative of British Columbia, and I explained this in a commentary published in The Lancet in January 2007. And it’s not even as if there are no data about primary prevention in women. The results from almost 11,000 women without heart disease who participated in clinical trials of statins have been published. The benefit of statins in these 11,000 women? None.

“The results from almost 11,000 women without heart disease who participated in clinical trials of statins have been published. The benefit of statins in these 11,000 women? None.”

The take-away message has not changed: if you want to decrease your risk of heart disease the first things to do are exercise routinely, eat a healthy Mediterranean diet, don’t smoke, drink alcohol in moderation, maintain a healthy body weight, and control the stress in your life. The scientific evidence shows that each of these is more important than cholesterol levels for the vast majority of people. (This holds even for people for whom statins have shown a definite benefit—those who already have heart disease or diabetes, or men at greatly increased risk who don’t yet have heart disease.) Don’t allow yourself to be tricked by the drug companies into believing that lowering your cholesterol level with drugs is the primary way to prevent heart attacks. And don’t let your doctor get tricked into believing this and imposing it on you.

NEXIUM

After cholesterol-lowering drugs, Nexium and its cousins (Prevacid, Protonix, and Aciphex) remain one of the most frequently prescribed classes of drugs. These drugs now cost about five dollars per pill. There is little to no evidence that they are any more effective than over-the-counter Prilosec, which costs seventy-one cents per pill. Perhaps even more important, a recent study shows that taking these drugs for a year increases the risk of hip fracture by forty percent and even more with longer use. This probably happens as a result of decreased absorption of calcium from the stomach when these powerful drugs suppress acid. And there still hasn’t been a good study to determine how many of the people taking these drugs could switch down to less powerful acid-blocking drugs (like Zantac, Tagamet, and Axid—all available over the counter as generics), or even use old-fashioned antacids when needed.

CELEBREX

After the withdrawal of Vioxx and Bextra, Celebrex is the last of the COX-2 selective inhibiting anti-inflammatory drugs remaining on the market. Despite having been found to more than double the risk of cardiovascular complications compared to placebo (albeit at twice the dose recommended for the treatment of osteoarthritis), despite having been found by the FDA to be no less likely than other NSAIDs to cause serious stomach problems, and despite being no more effective at relieving arthritis symptoms and pain than older, less expensive anti-inflammatory drugs, Celebrex remains a blockbuster drug, with sales of about $2 billion in 2006.

FOOD AND DRUG ADMINISTRATION (FDA) OVERSIGHT OF DRUG SAFETY

As a result of the FDA’s questionable handling of drug safety issues such as Vioxx, investigations by two of our most trusted institutions were initiated. The Government Accountability Office (GAO) report, requested by Senator Charles Grassley (R-IA) and Representative Joe Barton (R-TX), was completed in March 2006. The FDA, according to the report,

lacks a clear and effective process for making decisions about, and providing management oversight of, postmarket drug safety issues. The process has been limited by a lack of clarity about how decisions are made and about organizational roles, insufficient oversight by management, and data constraints.

“Despite being no more effective at relieving arthritis symptoms and pain than older, less expensive anti-inflammatory drugs, Celebrex remains a blockbuster drug, with sales of about $2 billion in 2006.”

The Institute of Medicine (IOM) report, undertaken at the request of the FDA, was issued six months later. The IOM found the FDA’s system of tracking post-approval safety issues “outdated and inefficient.” The IOM also called attention to the budget of the FDA division responsible for approving new drugs and overseeing drug safety. Half of this budget, according to the IOM, is paid for by the drug companies themselves, as a consequence of which the FDA’s reporting requirements

are excessively oriented toward supporting speed of [new drug] approval and insufficiently attentive to safety.

One would think these reports, along with the public’s eroding confidence in the FDA, would stimulate swift change. A front page article in the Wall Street Journal on March 3, 2007, shows just the opposite. Reporting on the FDA’s bungled attempt to update its system for tracking drug safety issues, the Journal article identifies the culture within the FDA as the “root cause” of the failure to improve drug safety monitoring. The executive director of the company that had been hired by the FDA to evaluate its attempt to improve its safety monitoring system told the Journal that the FDA asked him to “gut the report, and I refused to do it.” Hardly cause for optimism. Image