Doctors who are using or recommending mobile apps can breathe a bit easier. The U.S. Food and Drug Administration's new draft guidance, published this morning, excuses a whole raft of lower-level health apps from the strictest Class II and III regulations.
That generally includes online medical reference apps, such as Epocrates or Physicians' Desk Reference, general fitness and wellness apps, EHR or PHR apps, billing and administrative apps for healthcare providers, and any generic tools such as an iPhone magnifying glass or LED light that aren't marketed as performing a medical function.
What that leaves is a small but growing slice of the mobile pie: apps that actually affect a patient's care and treatment.
In most cases, the guidance includes apps that perform a medical function, such as reading blood glucose test strips, or apps that analyze patient data and recommend clinical treatments.
The FDA expects the "mobile medical apps" grouping to be a "small subset" of the wider health app market, and is focusing "only [on] those mobile medical apps that present the greatest risk to patients when they don't work as intended," Jeffrey Shuren, director of the FDA's Center for Devices and Radiological Health, said in an announcement.
The FDA said that the highest-risk apps (and those most likely to hit the Class II and III thresholds) include those that:
1. Transform the mobile platform into a medical device. In most cases, this is through attachments or sensors, such as transducers that turn a mobile device into a stethoscope, apps that read blood glucose strips, or ECG electrodes that allow a smartphone to measure and display ECG signals.
It is crucial to note that decision-support apps will be on FDA's radar. This is an important distinction, and one that many unfamiliar with FDA regulation might miss, Bradley Thompson, a medical device attorney and FDA expert with Epstein Becker Green in New York City, told FierceMobileHealthcare. Thompson also is a member of the mHealth Regulatory Coalition.
2. Control an existing medical device's use, function, modes, or energy source. This includes apps that control the function of a medical accessory, such as inflating a BP cuff or controlling the delivery of insulin through an insulin pump. It also includes apps that that display data from bedside monitors and medical images from a PACS server, the guidance explains.
FDA's thinking: The app should be covered because it is connected to, or controls, an already FDA-regulated device, Bakul Patel, Policy Advisor for Center for Devices and Radiological Health with the FDA, said on a conference call. Added Thompson: "In most cases, the level of regulation will directly correspond to the classification of the primary medical device."
Overall, these are the two types of apps likely to be placed in Class II or III, requiring more significant controls. Which class they'll fall into, exactly, is still in question, but using the FDA's current list of medical devices, users can make a solid guess, Thompson said. For example, electronic stethoscopes are considered Class II devices, so any app that controls or is connected to them would also be Class II. By the same token, most cardiac devices, such as defibrillators or pacemakers, are Class III, so any app that links to them would automatically be considered Class III devices, he pointed out.
Thompson feels that this particular issue needs more clarity, but says that will only come about through full rule-making, with the FDA officially placing different mobile medical apps into specific classifications. Because the proposed draft is only a guidance document, FDA can't use it to change official classifications. The mHealth Regulatory Coalition, which created its own software classification document earlier this month, plans to submit its comments for consideration on this score.
Of less concern are apps that simply display, sort or transmit patient-specific medical device data in its original format, FDA officials said. Those will be considered Class I and, as already indicated in earlier guidance, are devices that just store or transmit patient data, and aren't considered high risk. For example, apps that record or track fitness information would be Class I mobile medical apps.
Another area that Thompson believes still needs a more granular description is the "intended use" issue concerning health/wellness apps vs. disease-specific apps. FDA officials indicated they will regulate apps that provide disease- and patient-specific recommendations--such as a smartphone app that tracks a diabetic person's glucose levels, diet and exercise routine and recommends a caloric intake for the day, Thompson explained.
As for apps that essentially do the same thing--for example, tracking weight, diet, exercise, etc.--to give the user a general recommendation on ways to stay healthy? Those aren't likely to be cleared. " We need a level of specificity beyond that," Thompson said.
For hospitals developing their own clinical apps, there's also some confusion over who is a mobile medical app manufacturer--and thus on the hook for FDA regulation--and who isn't. It appears that app stores like iTunes and BlackBerry App World won't be considered manufacturers because they only sell a one-time download of the app. Companies that develop, market, support and maintain apps and their underlying software, however, will be considered manufacturers.
From a business and a consumer perspective, Paul Sonnier, vice president of partner development with the Wireless-Life Sciences Alliance, believes the draft guidance ultimately will prove helpful.
"I think maybe it will open the doors a little bit for investors and companies that wanted to get going and said 'well, maybe there's some uncertainty here. Now this makes it a little bit less risky,'" Sonnier told FierceMobileHealthcare. "Ultimately, this type of regulatory clarity by FDA [also] benefits consumers...[as such] apps may become more plentiful. FDA-cleared apps will be able to differentiate [which ones are the best], perhaps gaining a competitive advantage."
The FDA is asking for public comment on the guidance by Oct. 19.
Dan Bowman contributed to this article.