The FDA's role in telemedicine is still a moving target

Until the Food and Drug Administration (FDA) releases its final guidance for mobile medical applications and clinical decision support (CDS) software, the telemedicine industry is left with trying to figure out what role the agency will play.

Figuring out whether an application will be regulated is a guessing game at best – a point that was well illustrated in a Monday afternoon public policy session.

Bradley M. Thompson, MBA, JD, a partner at Epstein Becker Green, served up two use cases and sought advice from a panel of “expert advisors” to guide him on whether his imaginary applications would be under FDA scrutiny.

After listening to Thompson’s presentations of his “company’s” applications, Robert Harasy, project specialist-customer operations manager for Partners Healthcare’s Center for Connected Health; Yishai Knobel, an mHealth entrepreneur; Stuart Blitz, director of worldwide commercial development for AgaMatrix; Andrew J. Rosenthal, chief strategy officer of Massive Health; and Chuck Parker, executive director of Continua Health Alliance, quizzed Thompson and laid down their judgments.

The first use case involved mobile applications. The proposed July 2011 guidance document breaks down mobile applications that it regulates into three buckets: Applications that act as an accessory to medical devices, applications that transform the mobile platform into a medical device and algorithms that output a patient-specific result, diagnosis or treatment recommendation to be used in a clinical practice.

It does not regulate e-copies of medical textbooks loaded on smartphones, applications solely used to provide clinicians with training or apps solely used for general health and wellness. However, there is a gray area – Footnote 13 – which Thompson pointed out is the “stuff in the middle that the FDA is not sure about.”

That gray area, which includes applications that only automate common medical knowledge, allow individuals to self-manage their disease and automate the common clinician’s diagnostic and treatment tasks using simple general purpose tools, can be confusing.

The first case study involved the development of an application that manages data on a consumer’s weight, calorie and nutrition intake, exercise, sleep and stress. The fictitious company is working with a national diabetes group to create a big marketing campaign aimed at people with diabetes and is anticipating allowing physicians access to the data.

The answers to the question of whether the application should be regulated by the FDA ranged from yes (the application is dealing with a disease state) to no (the application is just automating a manual process) to recommendations of tweaking the product to avoid regulation (step back from some cause-and-effect and correlation claims made by the company). The variety of takes illustrates how confusing the issues are, Thompson said.

Thompson pointed out that while the CDS software guidance document has yet to be written, it has developed categories (preliminary definition, conversion of data and clinical decision that results in patient-specific and actionable result), as well as the degree of regulation associated by high and low risks. High-risk software includes radiation dose calculators, while low-risk includes medication reminders. The FDA has indicated that clinical workflow analytics and financial analysis software and software used for general medical reference or education are not considered CDS software.

The second use case involved the freedom of users to develop their own tools or modify another tool, and the question centered on how much latitude developers have before the FDA regulates their application. The developer in the example was a healthcare system that took another application and changed the user interface for its physicians’ use.

Expert responses to the question ranged from no (if the application was internally developed and used to guide physicians in the practice of medicine) to yes (if the original application was substantially changed and it is given to other physicians outside the healthcare provider). Others were on the fence, but noted that changing the product changed the risk profile, which would keep some up at night and in conversation with attorneys.

The use cases illustrated that there is still “a lot of groping to figure out what to do as we wait for FDA to release final rules,” Thompson said.
 

Comments

ProfessorMac
Great article and a troubling state of affairs. It is troubling if you are being treated for a chronic health condition like hydrocephalus, and your disorder is one of the many hundreds whose care & treatment outcomes would be significantly improved with the availability of mobile phone apps and patient tools. As a CNS shunt patient user (and with a diagnostic medicine bkg), I developed and patented a software monitoring method in 1997 for people like me with hydrocephalus. Sadly, I could never get the required funding, but had I done so I would have been dealing with FDA hurdles. My DiaCeph Test interestingly is quite similar to the IMPACT Test used today in post concussion monitoring, which has made major advances in sport and concussion safety, and that test was also developed around 1997. The FDA needs to get off its XXX and get how these apps and patient tools will revolutionize care today. I've seen three physicians in the last 3 days and in each instance I spent hours going through hundreds of medical records, including, many of which I did not currently have. With mHealth apps, patients would be able to review and organize their key medical tests results, correlate with relevant symptoms and complaints, and leave notes with their treating physician - all before they see the physician. The physician would then be able to go to work each day, review his/her patient case-load, read thru patient notes, and know what to do/say with the patient before the visit. The boost and benefit in care outcomes would be 300-400%, no doubt reducing costs, improving patient productivity at work, and possibly helping to pay down the national debt. So let's see.... is that a good thing? FDA needs to get off its XXX. I spoke by email to Mr. Thompson last year and offered to speak to help facilitate the availability these technologies. Yes, I do have some business interests in the outcome if my DiaCeph Test were ever made available. But, I'm more interested in my health and why I will have to undergo my 11th shunt surgery in 20 years. Clearly, the FDA is in over its head. They do nothing, care next to nothing, when it comes to problems with CNS shunt devices. On each and every instance, they have errored on the side of industry. I believe what makes their role in this mHealth discussion so interesting, is that again they are on the side of the medical industry and the AMA. This "club," comprised of companies and groups which benefit the most from the way medicine is currently conducted, albeit for bad or good, like the way it is - and clearly their lobby is causing FDA to act as it is. If this were a war, if the U.S. was at war and a gov't agency was holding up technology that would positively affect personnel safety and the outcome of the war, that agency would be quickly over-ruled. I am unsure what Mr. Thompson is currently doing to change FDA's mindset, but if I were involved FDA would not get one night of sleep until these barriers are lifted. Let's get this done, and hopefully before I die from complications which would otherwise be more readily treatable, were these technologies be made available. Thank you for listening. Stephen Dolle
CNS Shunt User x 20 years
Inventor, Health Advocate & Neuroscientist
Dolle Communications
Newport Beach, California

Post new comment

* Fields marked with an asterisk are required.
No HTML code or hyperlinks are allowed in comments.
Login or create an account to save your mHIMSS profile.
By submitting this form, you accept the Mollom privacy policy.