- The road to improving medical technology ends not in Palo Alto but in White Oak, Md.
- Since the 1990s the FDA has adopted the bad habit of governing via guidance rather than by establishing clear rules.
- Today, only about 100 mobile medical apps have been reviewed and approved by the FDA. That's a fraction of what's possible.
There's growing frustration among entrepreneurs in Silicon Valley who are finding that the road to improving medical technology ends not in Palo Alto but in White Oak, Md.—at the headquarters of the Food and Drug Administration. "Health is just so heavily regulated," Google co-founder Sergey Brin complained last month to a group of high-tech CEOs, "it's just a painful business to be in."
The pain has become especially acute in the burgeoning field of mobile medical apps and health-care "wearables." Apple is reportedly making a health-information platform a key component of its next iPhone operating system—which is expected to be in the new iPhone it will unveil next month. Other handset makers are following similar paths. But these smartphones will be purposely dumbed down, according to early reports on their features, in order to manage the uncertain risk of unwieldy FDA regulation.
According to the tech publication Apple Toolbox, the FDA recently released (under a Freedom of Information Act request) a document describing its meeting with Apple executives last December. Senior agency officials told Apple that the FDA "would be more likely to regulate the software that puts [a medical] sensor to use, if use of the software alters the device's use to be a medical device." The officials also told Apple that "apps that actively measure something" health-related, like glucose meters used by diabetics, are "diagnostic" and are likely to make the entire tool subject to regulation.
The FDA's regulatory dysfunction is driven by its 30-year failure to establish a coherent approach to regulating medically related software. Last fall the FDA issued guidance purporting to regulate only a limited subset of apps that qualify as "devices" under the law, and only if there is a meaningful risk to patients. The guidance says the FDA will regulate apps—using the same kinds of rules that apply to joint replacements and heart valves and the like—that display, transfer, store or convert patient-specific medical device data from a monitor (for example, a heart monitor) to a mobile platform. But under the guise of enabling innovation, the agency is making an already complex regulatory climate even harder to navigate.
Since the 1990s the FDA has adopted the bad habit of governing via guidance rather than by establishing clear rules. Regulators have also become addicted to addressing novel technologies by reflexively applying their premarket powers. The standard line is "come talk to us," a bureaucratic posture that leaves innovators in limbo while they wait sometimes long periods for a regulatory verdict.
The ambiguity created by the guidance and the agency's premarket review processes forces innovators to seek the FDA's nod for every new launch and every small advance. This slows progress to a crawl. Worse, the lag may be almost entirely unnecessary, as most of these products are not properly regarded as a medical device in the first place.
Many consumer-focused apps that support health are not "devices"—for example, apps that help patients self-manage a health condition. This is clear from the language of the relevant statute and case law. But the FDA's guidance says that they are medical devices but that the agency chooses, at this time, not to actively regulate them. What is an app developer supposed to do with that internally contradictory position?
The agency could help ensure the reliability of mobile health-information tools through clear rules that tech companies can apply on their own, without spending millions on lawyers and consultants to help them talk with regulators and without stymieing progress. To do so, the FDA needs to transform how it regulates.
The agency's regulatory scheme was shaped around an expectation that medical devices could be risky if they are not properly conceived, crafted and labeled—necessitating premarket review. But the consumer devices and medical apps under consideration have little in common with the FDA's usual fare.
These information tools are largely aimed at arming consumers with incremental data such as metabolic and cardiac readings to encourage better-informed health decisions. While faulty information can influence people to make bad decisions, the risks are far lower than those posed by a complex drug or implantable device.
Moreover, a mobile app is likely to be far more helpful in decision-making for many consumers—and certainly can't be worse—than online information sources of varying provenance and quality. There is already ample medical evidence to show that patients make bad decisions when they don't have adequate information about their health.
The FDA should exempt the majority of mobile-health apps from premarket review—the approach it has followed since at least 1997 for low-risk laboratory diagnostic tests. That would give the agency postmarket controls to make sure these tools are meeting expectations, but eliminate a company's need to get permission each time a new product wants to come to market.
The agency also could work with trade groups to establish general technical standards that these products must meet regarding safety, accuracy and data security. This would enable the FDA to draw on broader expertise when it comes to these fast-moving fields. A premarket exemption could hinge on products self-certifying that they meet these specifications.
Apps and smartphone tools are constantly revised, yet under the FDA's existing rules, once a medical app comes under the agency's premarket scheme, each iteration can require formal submissions and in some cases review by the agency before it can be offered to consumers. This can thwart the updating that keeps technology up to date.
Today, only about 100 mobile medical apps have been reviewed and approved by the FDA. That's a fraction of what's possible. The problem isn't merely a regulatory regime ill-suited for reviewing these innovations. It's the products that are never created because mobile-tech entrepreneurs choose to direct their talents elsewhere.