Ignorant No More

I believe that no doctor should do anything to or put anything into a patient without knowing or studying the long-term outcome.

Seems logical, doesn’t it? Yet, most doctors do not know the outcomes of their interventions. In fact, most of the data they need to do this is not available. Fortunately, this may be about to change—due to a combination of new outcome tools, social media, and the ability to mine enormous data sets from disparate inputs.

First, the unknown “facts.” Only a tiny percentage of all published articles in the medical literature are based on Level 1 studies. Level 1 means that the studies are done well enough, with proper controls and appropriate blinding, so that the conclusions are highly likely (though not necessarily correct). The reliability of the other 90% of studies varies widely, as most of these are based on the clinical experience of the reporting doctor, biased data, or study designs that are not always completely robust.

There is no required reporting on nearly any of the medical devices used in the United States once the implant has been approved by the FDA—and most procedures do not even require FDA approval. The approval protocols for devices are based on small safety trials, and usually on one or two efficacy trials reported from a relatively small (usually less than a few hundred) number of patients.

What could improve the safety, lower the cost, and speed the time to commercial use of medical devices in the United States? For one thing, after the initial safety trial, the FDA could permit sales while simultaneously requiring all manufacturers to do a “post-market approval study.” All doctors who implant a device would then be required to place an app on a patient’s smartphone. The app would permit the doctor and the patient to monitor all outcomes, using standardized, validated outcome questionnaires. The data would be posted publicly, and all physicians, scientists, and manufacturers would be able to see the results as they unfold over months, years, and even decades after implantation. By crowdsourcing the outcomes of medical devices (and potentially procedures), the safety and efficacy of these devices could be revealed, studied, and potentially improved over time. Complications would be spotted early and problem devices could be removed.

In the past, the wide variety of hard-copy patient records and electronic medical records did not cross-communicate with each other. This was a barrier to any kind of global medical record analysis. New clinical outcomes software, that mines various data inputs from a wide variety of medical records, is now available through companies like Lynxcare. These permit the creation of enormous data lakes, which can be fished for the pearls physicians need in order to truly understand their patients’ outcomes.

And tracking down patients through their initial contact information has now been simplified since almost nobody changes their smartphone number.

The day is quickly coming when novel devices will reach the market after safety trials alone. Efficacy will be confirmed by an open examination of the results by a wide variety of experts. Instead of just saying “all my patients do well,” physicians will state, “here is the outcome data from my practice, and the practice of all the surgeons around the world implanting these devices.” We may all be pleasantly surprised by what this spotlight reveals—or at least informed by it.