How are pharmaceuticals and medical devices regulated?

Under the terms of the Food, Drug and Cosmetics Act (the Act), the Food and Drug Administration (FDA) is required to ensure that new drugs are  both safe and effective before they can be sold in the United States.  The Act gives FDA a great deal of discretion to determine what it means for a drug to be “safe” and “effective,” however, because most medicines are actually toxic if taken at inappropriate doses.  In order to treat or cure many diseases, it is necessary to use drugs that may have substantial side effects, so the Act leaves it up to FDA to balance the need for new medicines against the known side effects of those products.  Most medical devices (a category that includes such things as pacemakers, arterial stents, and diagnostic tools, such as X-ray and MRI machines) are regulated in a similar fashion.

In order to generate data on safety and efficacy, drug manufacturers conduct a broad variety of pre-clinical tests on laboratory animals, and then three phases of graduated clinical trials on human subjects, under supervision of the FDA.  Phase I trials typically include fewer than 100 healthy volunteers, and the tests are designed to test the basic toxicity of the chemical compounds, and to assess the biochemical and physiological effects of the drugs.  If FDA is satisfied that the drugs are not inherently unsafe at a clinically meaningful dose, Phase II trials are conducted on a few hundred patients, and the tests are designed to collect broader safety information and begin assessing how well the drug works.  Then, if FDA gives permission, Phase III trials are conducted on as many as a few thousand patients to generate more statistically significant safety and efficacy data.

The whole process of pre-clinical and clinical testing can take anywhere from a few years to over a decade, with a typical expected time of about seven or eight years.  And, for every 100 drugs that enter clinical testing, approximately 70 will go on to Phase II trials, 30 will make it to Phase III trials, and fewer than 20 will be approved for commercial use.  When the cost of conducting tests on drugs that are not approved are factored in, it can cost well over $800 million to bring a new drug to market.

When deciding whether or not to approve a new drug or medical device, FDA must weigh the known risks of the products against their effectiveness.  And, with products that treat more serious conditions (such as cancer or other life-threatening illnesses) for which there are few available remedies, the agency is willing to tolerate a higher degree of potential side effects and more attenuated effectiveness.  However, while the FDA recognizes that there is no such thing as a drug that has no side effects, the agency is under tremendous political pressure from Congress, the news media, and the general public, not to approve drugs and devices that may later be implicated in identifiable injuries or deaths. 

A dying cancer patient may be willing to risk a 10 percent chance that a new drug will kill her, if there is a 50 percent chance that it will prolong her life by a few months or years.  FDA, however, is rarely willing to approve such drugs.  The reason is that, if the agency approves a drug or device that later is found to be unsafe in some way, the public and politicians blame FDA for the error.  However, if the agency delays when reviewing applications, the patients who need new treatments are worse off – some may even die because the treatment wasn’t available – but neither they, nor their families are likely to be aware that a possible treatment exists, or that it was blocked by the FDA.  Because no political pressure comes to bear for making the latter type of mistake, FDA’s natural tendency is to avoid making the first type of error.  This phenomenon often leads the agency to conduct data reviews of new drugs very slowly, keep good drugs off the market, and withdraw approved drugs when unexpected side effects later become apparent.  But, while individual patients and their doctors can choose not to use approved drugs and devices that are known to have certain safety risks, patients are denied the choice to take a calculated risk with unapproved products.

What should be the role of government in regulating pharmaceuticals and medical devices?

It’s common for critics to refer to the monetary costs to pharmaceutical companies that result from FDA’s long review times and rejection of promising new drugs.  However, the far more important social cost of this agency risk aversion is often prolonged exposure by patients to disease, and sometimes death.  Different patients will have vastly different tolerances for risk.  While some would willingly forgo a promising new treatment when it poses even a remote risk of injury, others would be willing to use a highly risky new product knowing that it poses a substantial health risk.  Unfortunately, while individual patients and their doctors can choose not to use approved drugs and devices that are known to have certain safety risks, patients are denied the choice to take a calculated risk with unapproved products.  Unfortunately, this problem is inevitable in a government-managed safety agency because the way in which FDA balances the risks and benefits of new medicines will necessarily be influenced by the political process. 

The only realistic solution is to permit doctors and patients to use drugs and devices that have not been approved by the FDA.  So long as patients were made aware of their unapproved nature, they and their doctors could evaluate what is known about the safety and efficacy of the products, consider that many risks may still be unknown, and decide for themselves whether the chance of potential benefits outweighed the known and unknown risks.  Those patients and doctors who trusted only the FDA would still be able to wait for an FDA approval before trying new products.  But those who want to have a choice would be free to rely on the evaluations by private or charitable certification bodies, such as doctor or patient groups like the American Society of Clinical Oncology or the Kidney Cancer Association.  Importantly, the choice would be theirs to make, not a government bureaucrat’s.  And, no patient would be forced to take an unapproved treatment.

If the FDA approves a medication that later is shown to have severe side effects, is that a sign that the FDA was careless in their duty?

There is no such thing as a medication with no possible side effects.  Drugs are chemicals, after all, and even the most seemingly innocuous chemicals can be dangerous when used inappropriately.  When deciding whether or not to approve a new drug or medical device, FDA must weigh the known risks of the products against their effectiveness.  And, often, FDA approves new treatments knowing that, even the recommended use can pose a risk of health problems, such as toxicity.  When the agency does so, these risks are typically included on the product’s label, so doctors can be made aware of them and can monitor patients for possible complications. 

In other cases, previously unknown health risks appear only after a drug or device is approved for commercial use.  Even when a product is subjected to very large clinical trials, the drug or device is tested on only a few thousand patients.  When such treatments are then used commercially in patient populations totaling in the millions, some previously unidentified side effects may become apparent.  Critics often charge that, with greater diligence and a lengthier review period, a product’s risks could have been identified sooner.  But, even when a particular side effect crops up in clinical testing, it is rarely clear whether an effect that occurs infrequently is caused by the drug or device or by the patient’s pre-existing condition.  Hindsight is often 20/20.  But, regulators should not be immediately condemned for approving a drug that later turns out to pose identifiable health risks.  As discussed above, being too willing to “err on the side of caution” poses its own risks.