The Blue Pill Or The Red Pill?President Obama once famously quipped, "If there's a blue pill and a red pill, and the blue pill is half the price of the red pill and works just as well, why not pay half price for the thing that's going to make you well?" Increasing the use of cheaper, but still effective treatments, medicines and procedures is one of the central ideas behind a new science called comparative effectiveness research.
When it comes to meds, cheaper is not always better
January 15, 2010
Thanks in part to cheerleading from the White House, Congress is poised to pass health care reform legislation that devotes dozens of pages to creating new agencies, commissions and standards for conducting comparative effectiveness research (CER) in the hopes of cutting billions in unnecessary spending.
CER is a good idea but can be harmful when done through centralized methods. Fiscal pressures mean that sooner or later CER will drive reimbursement decisions for public programs like Medicare and Medicaid. Some cheaper medicines may become first-line treatments, even if they offer worse outcomes for many patients.
Before we deny patients access to expensive new medical innovations, we need to understand CER's limitations and learn how to overcome them. The blue pill vs. red pill comparison is reductive. The same meds that work well for some patients may not work for others; indeed they may make others worse. Yet such "one-size-fits all medicine" will likely increase under centralized CER.
Take antipsychotics, one of the biggest drug classes in Medicaid. In the 1950s, researchers created a class of drugs called "typical" antipsychotics, which were one of the first treatments for schizophrenia. These drugs were effective, but after prolonged exposure many patients developed a side effect called tardive dyskinesia (TD)—jerky, involuntary facial or limb movements that could become severe and permanent. This serious side effect can stigmatize patients, leading them to stop taking effective medicines.
By the 1990s, drug companies had developed a new class of drugs called atypical antipsychotics that had fewer side effects, including TD. As a result, the atypicals quickly dominated the market, accounting for 90% of prescriptions. Costs for these drugs, however, have as a consequence gone up.
In 1999, the National Institute of Mental Health funded a $42.6 million study called Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) comparing the effectiveness of one first generation antipsychotic (perphenazine) with all of the atypicals currently available in the U.S. One of the heavily publicized findings from the report was that the newer medicines didn't seem to be any more effective than the older, cheaper, generic ones. This lead to widespread calls to reduce the use of the newer medicines, and about 40% of states did so through policies that discouraged doctors from such use.
In a recent study with Anirban Basu, however, I tested whether a policy in favor of generic perphenazine vs. three newer drugs would really improve health and lower costs.
If policymakers set out to simply substitute perphenazine for the three newer drugs in the Medicaid program, poor patients couldn't afford to switch to a newer medicine. Based on the CATIE trial, about a third of perphenazine patients would stay on the drug long term. Others would quit the drug but relapse, resulting in new hospital visits. However, for those patients who would have responded better on a non-reimbursed treatment, there would be a loss of health and increased spending. Thus, although the two treatments were deemed equally effective on average, it turned out that not reimbursing the newer more expensive drugs would lead to a loss valued at about a tenth of spending in this drug class. This occurred because those who did not respond well to the cheaper drugs incurred larger hospital costs and worse health outcomes when not having access to the more expensive new drugs.
Because of widespread variation in patient response, the more restrictive the payment policy the higher the lost value in human life and productivity from excess side effects or non-responsive drugs, even when the cheaper drug seems to be a good bargain. In fact, it is difficult to think of a single drug class in which one medicine is always the best option for every patient. Instead of considering CER a substitution of the blue for the red pill, we have to focus on finding the right treatment for the right patient on a case-by-case basis; for whom is the blue vs. red pill the best?
Unfortunately, the science of CER is only beginning to tackle this problem. The current standard for evaluation, the randomized clinical trial, is not always as useful when we have to sequence first-, second- and third-line therapies across a diverse population.
Clinical trials typically only tell us which treatment is better when administrated once to the overall population, when what we really need to know is the response of patients to subsequent treatments after they fail to respond to the first option—as many patients often do.
With CER set to take center stage in our cost-control efforts, policymakers must proceed cautiously and improve the science before shifting to new reimbursement policies. Choosing the blue pill now may give us a bigger headache tomorrow.
Tomas J. Philipson, Ph.D., is the Daniel Levin Professor of Public Policy at the University of Chicago and the chairman of Project FDA of the Manhattan Institute. In 2003-04, he served as the Senior Economic adviser to the commissioner of the FDA and in 2008 as a senior health care adviser to the presidential campaign of John McCain.