The two of us are not always on the same side of contentious legal questions, but a recent move by the Food and Drug Administration to expand its jurisdiction to regulate laboratory-developed testing services as if they were medical “devices” has brought us together. The FDA’s assertion of this massive new authority without going to Congress for new legislation bears a family resemblance to other examples of administrative overreach, but there is one key difference. This overreach poses a threat not just to the separation of powers, but to the practice of medicine, and ultimately to the health of every American. For this reason, the American Medical Association and leading health care providers have urged FDA to withdraw its proposal.

Not everyone is familiar with the term “laboratory-developed testing services,” but nearly all have benefitted from them. When our doctors order blood tests, for example, clinical labs use their equipment and expertise to generate the results that our physicians need to make basic diagnostic and treatment decisions. These laboratory services range from routine tests to advanced and innovative molecular and genetic sequencing tests for cancer, heart disease, and rare and infectious diseases.

For decades, clinical labs have been comprehensively regulated by the states and by the Centers for Medicare and Medicaid Services (CMS) under a 1988 federal statute, the Clinical Laboratories Improvement Act, specifically tailored to clinical labs. Most labs also obtain accreditation by the College of American Pathologists or the Joint Commission for Oversight.

This past fall, the FDA announced its intent to regulate clinical labs and their testing services based not on a new grant of statutory power, but rather based on the agency’s baffling proclamation that it has enjoyed an unrecognized and unexercised authority to regulate in this area for nearly forty years. Specifically, FDA’s theory is that laboratory-developed testing services qualify as medical “devices” under 1976 amendments to the federal Food, Drug & Cosmetics Act.

But testing services are not “devices” in any normal or legal sense of the word. The testing services are no more medical devices subject to FDA regulation than are the services of a radiologist. Both laboratories and radiologists use medical devices to provide their services, but in neither case do the services somehow become devices. “Devices” are tangible articles of commerce. Laboratory-developed testing services are services performed by medical professionals in labs using proprietary methodologies. Those services (as opposed to the test results) never leave the lab and are not distributed in interstate commerce.

FDA’s illogical assertions are worse than trying to fit a square peg into a round hole. We are expected to swallow the notion that a federal agency disregarded its own regulatory authority for almost four decades. Accepting FDA’s position means that Congress was profoundly mistaken when it acted to fill a perceived regulatory gap in 1988 by giving CMS authority over clinical labs.

To legal scholars, FDA’s unilateral power grab raises important questions of administrative law and the separation of powers. It poses a real threat to public health as well. Clinical labs are already closely regulated by the states and CMS. Subjecting them to duplicative FDA regulation would eliminate the very features that make their services so valuable: their flexibility and adaptability in embracing new scientific and technical developments and responding in real time to unmet patient needs. Clinical labs have developed tests critical to the diagnosis and treatment of emerging infectious diseases, such as HIV, SARS and H1N1. They are constantly updating their tests to keep pace with rapidly evolving knowledge. It defies logic and threatens the progress of modern science to force a lab to obtain prior approval from Washington in order to modify a cutting-edge test for the benefit of patients.

There is also a complete mismatch between the FDA’s resources and the volume of innovative testing services available. More than 11,000 laboratories are currently permitted to use and develop an estimated 100,000 different tests. By comparison, in 2013, the FDA approved just 21 pre-market submissions for all medical devices. Seemingly recognizing the magnitude of that task, the FDA has given itself nine years to implement its proposed regulatory scheme, a timeline that highlights one of the many problems with bypassing Congressional action. The FDA’s unilateral action must be taken with existing funding that the FDA has repeatedly told Congress is insufficient for even its traditional responsibilities. If Congress really wanted the FDA (and not CMS) to have this authority, it presumably would have granted the FDA significant new resources to pay for this new layer of regulation.

There is a right way. If the CLIA statute needs strengthening, as some have argued, Congress can enact the required legislation and appropriate the requisite funds. Such action would be more tailored to whatever needs exist and would avoid both administrative overreach and a new layer of duplicative and conflicting regulation. At stake is nothing less than protecting the integrity of the law and the practice of medicine — on both of which every one of us depends.

Laurence H. Tribe is a university professor at Harvard University and a professor of constitutional law at Harvard Law School and Paul D. Clement is a former solicitor general and a partner at Bancroft PLLC. Both attorneys have been retained by the American Clinical Laboratory Association for representation on matters relating to the FDA’s recently issued draft guidance to regulate laboratory developed tests as medical devices.