Mobile healthcare, telemedicine, telehealth, BYOD

Apps & Software News

CDS Coalition Floats Guidelines for Mobile Health App Design

The CDS Coalition has released a draft set of guidelines designed to help healthcare providers and mHealth developers create mobile health applications with clinical decision support capabilities.

Source: ThinkStock

- A coalition of software developers and healthcare providers is pushing for new guidelines that would help certain mobile health applications avoid FDA regulation.

The CDS Coalition has issued a 26-page document designed to help developers create software with clinical decision support (CDS) capabilities. This would include mHealth app and digital health platforms that offer CDS tools.

The impetus for the guidelines is the 21st Century Cures Act, passed and signed into law in December 2016, which specifies that the U.S. Food and Drug Administration won’t regulate CDS software as long as it meets two key conditions: providers can independently review the basis for the app’s recommendations, and users aren’t compelled to rely primarily on the app to make clinical recommendations.

That bill capped a multi-year debate between the FDA and the healthcare community over how to regulate mobile health devices and software, with the FDA ultimately deciding to focus on devices and software that have a high risk of causing harm to a patient if used improperly. Healthcare entrepreneurs have long sought to limit FDA regulation, saying it hinders innovation, and seem to have found an ally in the Trump Administration.

Bradley Merrill Thompson, an attorney for the Washington, D.C. firm of Epstein, Becker & Green and legal counsel for the CSD Coalition, says the 21st Century Cures Act gives the healthcare industry the opportunity to regulate itself.

“Congress took this action in part because it is consistent with the dividing line between FDA regulation of medical devices, and state oversight of the practice of medicine,” he wrote in a commentary accompanying the proposed guidelines. “When the software takes over the decision-making, FDA takes over the regulation.  On the other hand, when the healthcare professional remains firmly in control, state oversight of the professions is enough, without FDA intruding.”

With more than 260,000 mHealth apps on the market, mHealth developers have often walked a fine line between consumer-facing health and wellness content and products requiring FDA review. The healthcare industry, meanwhile, has been hesitant to embrace apps, primarily because of the reliability of data collection and analysis tools.

The result: the an oft-mentioned “Wild Wild West” of mobile health apps, some of which have been risky and deceptive enough to prompt the FTC and even state attorneys general to step in and sanction their developers.

Thompson says it’s up to the industry to police CDS software by helping developers and providers plan and design their products. This, in turn, will give providers the confidence to use and prescribe mHealth apps, and will give consumers the confidence to use them.

“The coalition’s goal is to develop a framework that will provide consistency in user empowerment across software developers,” he wrote. “The key for this unregulated portion of CDS is ensuring that the healthcare professionals can gather and analyze the information needed to make the best recommendations to their patients.”

Thompson said the coalition’s guidelines focus on four factors that need to be addressed to determine if the software can evade FDA review:

  1. The user must be a healthcare professional who is “educated, trained and licensed to make the decision that the software is supporting.”
  2. The user must be able to review the information and the basis for which that information was used to make a recommendation. For software that makes a recommendation concerning a particular patient, it has to have information about that patient. It must also access general clinical knowledge, through channels like artificial intelligence tools and “expert systems.” “Either way, the program acquires clinical knowledge that forms the basis for its recommendations,” Thompson said. “Like the patient specific information, that clinical knowledge needs to be identified and available the user.”
  3. The user needs adequate time to analyze the information.
  4. The developer needs to understand how the user will access information outside of the software. “If software is used to simply crunch a few numbers that aid the doctor in decision-making, but the doctor also has a variety of other information such as direct physical examination, radiological images, laboratory test results and other such information to supplement the assessment by the software, the doctor will be much less reliant on the software,” Thompson said. That process will also be more complex id machine learning is included.

The CDS Coalition has set a July 1 deadline for comments on its draft guidelines.

X

Join 20,000 of your peers

Sign up for our free newsletter to keep reading our articles:

Get free access to webcasts, white papers and exclusive interviews.

Our privacy policy

no, thanks