The Food and Drug Administration is at it again. An agency that has plenty of problems performing its current duties now wants to take on a dubious new challenge: regulating clinical decision-support systems as medical devices.
Healthcare executives shouldn't make the mistake of thinking the FDA's ill-conceived plan is only of concern to vendors. It should worry the heck out of administrators, too. Under newly proposed regulations, the agency could attempt to regulate providers who develop clinical software in-house, even if they give it to another hospital or clinic. Telemedicine transmissions across state lines also could be regulated.
Proposed regulatory language says "an organization that develops a software device and distributes that device to multiple establishments, whether or not corporately linked," is a "manufacturer subject to the full range of regulatory controls, unless exempted by the regulation."
We think the FDA should forget about regulating clinical software and stay within its jurisdiction: i.e., regulating instruments intended for use in the diagnosis or treatment of disease. To us, that plainly means such devices as infusion pumps, magnetic resonance imaging machines, diagnostic X-ray systems, clinical lab instruments and even software that is a component of such devices.
The FDA doesn't need to insinuate an unwieldy government bureaucracy into a fledgling effort by providers and suppliers to respond in a timely manner to the need for products that provide better clinical data.
Current FDA drug- and device-approval processes are marked by unacceptable bureaucratic delays. But instead of getting their house in order, agency leaders are talking about taking on regulation of tobacco and contemplating this new challenge. We wonder how they intend to fund such ambitious new regulatory undertakings.
Healthcare executives who support current industry efforts to generate creative responses to the needs of a changing marketplace should mount a concentrated effort to tell the FDA to stick to its knitting.