Does CDS Need a Regulatory Carve Out?

EHR System Technical Functionality vs. UsabilityHouse Bill to Clarify FDA Regulation on Software

COMMENTARY
William A. Hyman
Professor Emeritus, Biomedical Engineering
Texas A&M University, w-hyman@tamu.edu
Read other articles by this author

The acronym gurus were hard at work when they figured out a name for HR3303 (1) which has been introduced in the House to “clarify” FDA regulation of software. The bill’s title is “Sensible Oversight for Technology Which Advances Regulatory Efficiency” and yes indeed, by golly, that creates the acronym SOFTWARE. And before even reading the bill, who could be against “sensible oversight” and “regulatory efficiency”? The FDA has long claimed authority in this arena even though the legal definition of a medical device is, at least to some, ambiguous with respect to its inclusion of standalone software as opposed to software that is imbedded in a more recognizable medical device. If you are a dedicated student of Congressional activity, there is a two hour hearing on this bill at EnergyCommerce.House.gov.

One component of the bill is to define “clinical software”, which includes Clinical Decision Support (CDS), as a distinct category. According to the bill such software “may recommend courses of clinical action, but does not directly change the structure or any function of the body”; and “is intended to be marketed for use only by a health care provider in a health care setting”. Based on this definition and the intent of the bill, CDS would be except from regulation by the FDA. While this might clear-up the current confusion (in some circles) as to what CDS’s are or are not, and whether they should be or don’t need to be regulated, opting for no FDA regulation is unlikely to be the optimum solution.

The language of the bill goes on to assert that such software might advance the goals of enhanced patient safety and continued innovation; hold much promise to lower costs and improve the health of patients; and can improve the quality and efficacy of health care provider services. Therefore “the President and the Congress should work together to develop and enact legislation that establishes a risk-based regulatory framework for such clinical software…that reduces regulatory burdens, promotes patient safety, and fosters innovation”. Thus the bill as proposed would take CDS away from the FDA first, and then direct that new legislation, which has not yet been proposed, be enacted to do all those good things. Of course if it happened in that order there would be at least a period in which such software was not regulated at all. And while there are those that would seem to applaud that, but I am not among them.

This bill and the quoted language apparently overlooks, or intentionally disregards, that the FDA already practices risk-based regulation as reflected by the three classes of medical devices and their respective levels of scrutiny. This was explained (again) by Dr. Shuren of the FDA at the hearing. Similarly the FDA is mandated to take a least burdensome approach, and to foster innovation, and certainly to promote patient safety. In this regard, in its recent Guideline the FDA explained in some detail how it proposed to regulate Mobile Medical Apps (2). Whether mobile or not, the software issue is the same: is it well designed based on a firm knowledge base, and is the advice it provides reasonable correct for a carefully defined population? And in the context of the proposed bill, is there a need to do legislatively what the FDA is already doing administratively, and with far greater flexibility than new law would allow. Even some proponents of the bill at the hearing seemed to acknowledge that more work was needed to craft a bill that actually made sense.

Of course the introduction of a bill is far from it becoming law, especially in Congress as we know it.

1. HR 3303

2. FDA Guideline