How exactly does a medicine get approved? (Part 4 of 5)

Episode 4: The One With the Bummer Ending

[Ok, so I couldn’t get the articles to properly align, so we get the dark ending an episode early!]

Here are links to Parts 1, 2, and 3. Now on with the show!


Clinical Testing

As mentioned earlier, novel drugs and devices are generally subjected to one or more clinical trials in order to demonstrate safety and efficacy. These investigations are highly regulated and subject to independent oversight in order to ensure study quality.

First, the sponsor (often the manufacturer, but occasionally the inventor, a clinician, or an academic institution) must obtain regulatory approval for the trial; since the trial will be using an unapproved product for medical care, it must be granted exemption from the approval process for a limited time and use. Trials in the US, EU, Canada, Australia, and many other nations must be registered on public websites (,, or others; Phase I safety trials are generally exempt from this requirement) to ensure open visibility.

Second, an Independent Review Board (IRB) must review the study sites, study protocol, study population, and risk/benefit profile to ensure that all necessary safety protocols are in place and that study subjects will be adequately informed and protected from unnecessary risks. This includes close monitoring of side effects and complications; if a treatment begins to demonstrate an unexpected rate of complication, the study may be terminated early for the safety of all participants. Similarly, a study could be terminated early if the treatment demonstrates unequivocal superiority early on, as continuing the study would imply withholding better care from the control population.

Third, the study sites and resulting data are subject to regular independent monitoring. The FDA refers to this as Biomedical Research Monitoring (BIMO). Sites can and are audited at random intervals with minimal warning to ensure that the investigators are complying with the protocol, watching for adverse effects, and maintaining proper data integrity and security. Any site found insufficient can be cited with a Warning Letter, which is then made public via the FDA website.

Two central tenets of all clinical trials are the concepts of Informed Consent and Clinical Equipoise. In the first case, the subject must be fully capable of understanding the risks and benefits of participating in a study and free of undue influence to participate. Thus prisoners, military personnel, developmentally/intellectually disabled individuals, and children must be considered with extreme caution to protect their rights. Clinical Equipoise means that all investigators must be genuinely undecided about which treatment is superior; this generally prevents clinicians with financial or other interest in the product from being involved in the study.


Physician Interactions

Since physicians will be intimately involved with the use of a medical product, their input is vital in the development process. Physicians are commonly inventors on initial technology or consultants on design, for which they receive financial compensation in the form of royalties and hourly fees. This offers potential for conflict of interest, as a physician could be prescribing a product in which they hold a financial interest. The Physician Payments Sunshine Act of 2010 attempted to alleviate this issue by creating a public database with mandatory reporting of industry-physician payments.

There was a time when industry-physician training was rather famous. “Educational” programs were held on cruise ships, at golf courses, in box seats, and even adult-oriented clubs. Millions were spent over the decades to wine and dine physicians into using a particular product. I personally recall a late-90s trade convention where one company was promoting its novel cast system; in order to see how easy it was to apply, the manufacturer had a booth full of bikini-clad models for hands-on testing. In the early 2000s the US Department of Justice launched a wave of investigations and prosecutions that largely shut down these practices. Companies are now much more careful with how they interact with physicians, and governments worldwide continue to watch them carefully.


The Dark Side

No matter how careful the analysis, how definitive the clinical data, or how detailed the quality controls, occasionally bad products to get into the market. In fact one of the most tragic cases gave birth to the modern era of medical device regulation in the US.

Prior to 1976, devices were loosely regulated as “drugs” under US law, but they sat in a legal gray area with minimal review and little oversight. All that changed with the disastrous Dalkon Shield IUD.

The Dalkon Shield was an Intrauterine Device (IUD) for contraception. A small head was placed into the uterine wall, and a filament hung from there into the cervix to facilitate retrieval. Since the product was essentially unregulated, no bench or clinical testing was required to prove safety and efficacy of the product. Unfortunately, the Dalkon filament was designed as a braided filament (think conventional thread) instead of a monofilament (think fishing line), and acted as a very efficient wick for fluids and bacteria. Thousands of women fell ill as a result of the design, including at least five fatalities. While IUDs are in general a very safe and effective form of contraception, this particular design was a disaster, both in its immediate health impact and in its long-term impact on the use of IUDs in the US.

Largely in response to the Dalkon Shield, Congress passed the 1976 Medical Device Amendments. This act formally placed medical devices under FDA’s responsibility and created the new 510(k) and PMA pathways.

FDA oversight certainly helped to standardize and regulate medical devices, but problems still occur. Sometimes it’s due to breakdowns in a company’s quality systems, wherein faulty products are accidentally released. Sometimes it’s due to intentional fraud; for a particularly heinous example, I refer you to CIR’s excellent reporting on fake pedicle screws. And sometimes it’s due to changing understanding of the underlying science and engineering, as in the case of metal-on-metal hip prostheses.

It’s worth noting that manufacturers are legally bound to maintain significant postmarket surveillance, primarily in order to catch problems as soon as possible. Every employee is bound to report any complaint that they identify, even in casual conversation. That’s right: if you walked up to me and said you had a problem with a medical device (even one from a competitor), I would be legally required to pass that information on to my employer’s complaints department for review*. Device-related complaints are archived on the FDA’s searchable Manufacturer and User Facility Device Experience (MAUDE) database.

Whatever the cause, faulty drugs and devices can be disastrous for patients and their families, and should be minimized as much as possible. However for every call for more extensive testing, there is a call for more rapid development of new products. This is the challenging balance that every inventor, manufacturer, researcher, and regulatory body faces: speeding innovation while maximizing safety.

And we haven’t yet considered the complications that child health brings to the issue.


*Now before you flood the comments forum with faux ‘complaints’ to test this, realize that there has to be information identifying the actual product used.


Topher is a nerd, scientist, bioengineer, photographer, costumer, dog trainer, beer brewer, and a lot of other nouns. He has too many degrees and too many hobbies because it’s all fun and interesting, and … squirrel!

Related Articles

Leave a Reply