With the advance of AI in the medical device field, questions have risen as to whether medical device makers may risk future liability following recent FDA efforts. AI medical devices are innovative and groundbreaking. However, potential liability questions center on the nature of these devices and their unique learning abilities.
Currently, the approval process set by the U.S. Food and Drug Administration (FDA) provides broad legal protections for manufacturers. These protections come in the form of preemption from product liability claims based on pre-market approval from the FDA. However, the learning abilities of AI medical devices could potentially remove their preemption status. As such, experts are now considering if AI medical devices could face a barrage of future product liability claims.
What is Preemption?
Federal Preemption is based upon the Supremacy Clause of the Constitution which declares that all Constitutional valid Federal laws are the “supreme law[s] of the land” and that “the judges in every state shall be bound thereby, anything in the constitution or laws of any state to the contrary notwithstanding.” In simplistic terms, Federal Law preempts State Law and can be express or implied.
Express preemption arises when Congress adopts a statute that explicitly displaces state law. Implied preemption arises when Federal law occupies the entire regulatory field leaving no place for state law. Situations, when state law thwarts a Federal statutory scheme, can also result in an implied preemption situation.
In 1976, Congress enacted the Medical Device Amendments (MDA) to the Food, Drug and Cosmetic Act (FDCA). Specifically, 21 U.S.C. Section 360 K (a) granted the FDA the authority to regulate medical devices. The provision expressly preempts any claim that imposes a state law requirement. The end result is an express preemption. In turn, Section 360 K(a) provides a broad preemptive effect with a Congressional goal of ensuring innovative medical devices would be readily available to treat patients. It does, however, only apply to Class III devices that support or sustain human life or present a potential unreasonable risk of illness or injury.
The Process for FDA Medical Device Approval
Current manufacturers of AI medical devices such as Philips follow the same process as in the past. Depending on intended use and risk, manufacturers could apply for pre-market approval (PMA) or one of the FDA’s two other approval designations. This includes clearance under Section 510(k) of the FDA Act for devices substantially similar to other devices on the market, or de novo classification for new, low-to-moderate risk devices.
Generally, any medical device makers that win pre-market approval from the FDA can assert preemption from product liability claims. However, the current preemption legal position in the United States is still firm and encompassing even when facing the newest technology innovations from manufacturers.
AI Medical Devices as an Exception
Because AI medical devices are capable of learning and making improvements over time, the abilities of such devices could change significantly from what was originally approved by the FDA. In turn, plaintiffs could potentially bring lawsuits on the grounds that the defective AI medical devices no longer have immunity under any FDA based preemption and could be subject to state regulations and product liability case for defective or negligently created medical devices.
Riegel v. Medtronic, Inc.
Currently, Riegel v Medtronic, Inc. (2008) still remains the leading case in this field. The Supreme Court noted that once a device has received PMA status, the manufacturer is forbidden “to make, without FDA permission, changes in design specifications, manufacturing process, labeling, or any other attribute, that would affect safety or effectiveness.” Riegel, 531 U.S. at 319. This is a twofold standard. It requires manufacturers to ensure that no changes are made to a device if they wish to maintain preemption status. Riegel also reaffirms the Congressional position that preempts any state law claims. This is a powerful broad doctrine that favors manufacturers. It allows defense counsel the ability to successfully defend cases that challenge design or manufacturing claims.
Does AI Technology Change Legal Analysis?
Despite a recent Bloomberg article on the subject, experts have argued that the Riegel decision is still sound even when facing current AI medical devices. Noted Attorney and Blogger, James Beck, Esq. reassured any readers of the Bloomberg article that assuming the FDA continues to use the framework for AI medical devices, plaintiffs relying on state law for their claims are still not able to second-guess the FDA’s decision to regulate a device as requiring PMA versus any other regulatory path to marketing. He notes that at least since Buckman Co. v. Plaintiffs Legal Committee, 531 U.S. 341 (2001), there is no case in which a court allows a plaintiff to second-guess an FDA decision. This includes challenging the PMA status for a medical device. Thus, the preemption remains intact to the benefit of the manufacturer.
Beck dismisses that the idea that every algorithmic alteration created by AI medical devices themselves could affect preemption. This would turn into a seemingly impossible standard for determining pre-market approval. He notes that the FDA approves “devices” not “uses.” as a process. Therefore, once PMA is obtained, preemption exists under Riegel. This is true regardless of which particular “use” or “intended use” a device is put to in any given case.
A New Approach For the FDA?
There is no doubt that AI medical devices are the future of medicine. They have the potential to transform how doctors diagnose and treat diseases and they can ensure patients get the right treatment at the right time. But analysts are wondering if the FDA could take a new approach to AI medical device technology. When facing such innovative technology, the FDA could develop a new form of pre-market device approval which would require a substantially new form of review to address the learning capabilities of AI medical devices.
Presently, the FDA does not appear to be changing courses. AI medical devices will continue to fall under the current PMA framework. Thus the Riegel preemption standard will continue to apply for years to come. Unless evidence can be shown to the contrary which fundamentally alters the current FDA framework, manufacturers can take comfort in knowing the liability risks associated with their production of medical devices.