Redux, a Deadly Cascade
In April of 1996, Redux (also known as Dexfenluramine), was FDA approved to help combat obesity. In September of 1997, less than 17 months later, Redux was recalled for causing two potentially deadly heart and lung conditions. How could such a flawed drug have made it past not only the FDA, but into millions of doctor filled prescriptions? An American obesity epidemic, previous Redux approval in Europe, and an information cascade led to the disaster.
A Time Magazine cover story on Redux (see picture below) stated that “Some 58 million citizens, nearly a fourth of the nation’s population, are clinically obese-at least 20% above their ideal body weight.” The enormous prevalence of obesity surely played a role in the premature adoption of the drug. In addition, Redux had already been in use for ten years in Europe. These two factors made a cascade effect all the more likely.After being tested on only 500 patients for one year, Redux was put before the FDA for approval for human use. A glaring lack of studies and evidence of significant brain damage in animals “from mice to baboons” (Time) as well as increases in pulmonary hypertension, an often fatal lung condition, was not enough to deter the FDA from approving the drug. Nevertheless, final consumers were theoretically still protected by one last layer: the doctors.
Since Redux was not approved as an over the counter drug, doctor prescriptions were required to obtain the product. It would seem to make sense that some of the most knowledgable medical professionals in the country who had been through years of grueling instruction in medicine would not trust a barely-studied, yet undoubtedly dangerous drug. Unfortunately this was not the case; by October, 1996 estimates range from 85,000 weekly prescriptions (Time) to 407,000 monthly prescriptions (Physicians News) being filled. Some doctors even set up special redux clinics. The country, in desperate search of a remedy to obesity, had found its holy grail. Not only did redux facilitate significant weight loss, but it did so with the utmost rapidity. Only in September 1997 did the FDA ask the manufacturer to stop selling the drug due to overwhelming evidence of severe heart and lung damage. How does the Redux craze fit an information cascade model?
Easley and Kleinberg state that “… an information cascade has the potential to occur when people make decisions sequentially, with later people watching the actions of earlier people, and from these actions inferring something about what the earlier people know.” (484) in their joint book titled Networks, Crowds, and Markets. The decision to approve Redux in Europe was arguably the first step that precipitated the cascade. The FDA chose to trust Europe’s decision; this was the crucial second step. The conditions were then appropriate for the cascade to start (acceptances-rejections=2). Perhaps some doctors thought that the FDA must have had solid evidence to approve the drug, and so they began giving prescriptions. As doctors saw more and more of their peers following the cascade, they apparently cast aside any inhibitions and surmised that those prescribing the drug were doing so because they thought it was safe. Eventually, this led to thousands of doctors all having the same amount of (relatively little) information about redux yet all believing that each of their peers had ample reasons to prescribe.
Easley and Kleinberg conclude a section in their book with the following warnings about cascades: “Cascades can be wrong…”, “Cascades can be based on very little information…” and “Cascades are fragile” (503). The decision to prescribe redux was definitely wrong –it led to two potentially fatal diseases for those who took it. As previously mentioned, the decision was based on very little information – only one study had been done with 500 people subjected to the drug. Finally, the cascade was necessarily fragile; as soon as Redux was removed from the drug market, the doctors were forced to stop prescribing it. As other bloggers have shown, humans are extremely prone to following information cascades. Whenever possible, we must strive to (taking from the red/blue ball example) peek inside the bag and base our actions on solid evidence, and not on assumptions.
Sources:
* The Time cover article: http://www.time.com/time/subscriber/article/0,33009,985187-6,00.html
* An FDA site giving information about the Redux saga: http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm180078.htm
* More information about the Redux (and related fen-phen) disaster:
http://www.physiciansnews.com/law/498.html
http://www.phentermine.com/phentermine_history.htm
http://www.pbs.org/wgbh/pages/frontline/shows/prescription/hazard/fenphen.html