Clearance and approval routes in the United States
In the United States a device's intended use fixes its risk class, and the class decides its route to market and the evidence that route asks for: premarket notification ending in a clearance, premarket approval, a de novo request, a humanitarian device exemption or an exemption from premarket notification. Clearance, the route most devices take, is not the FDA's assertion that a device is safe and effective.
The FDA and the device classes
For much of the twentieth century medical devices were largely unregulated and not subject to regulatory review. To address this gap and prevent the distribution of dangerous and ineffective devices, Congress passed the Medical Device Amendments to the Federal Food, Drug, and Cosmetic Act in 1976, which created a comprehensive premarket review mechanism and a three-tier scheme that stratified devices into classes by their potential risks. US device regulation dates from 1976, whereas drug regulation has existed since 1938. Devices marketed legally in the United States before the amendments are preamendments devices, and those that entered the market afterwards are postamendments devices.
Drugs and therapeutic biologics are regulated by the FDA's Center for Drug Evaluation and Research, while devices are primarily regulated by its Center for Devices and Radiological Health, which approves devices in the United States.
The regulatory control of a device is essentially based on risk: class I devices are low risk, class II devices medium risk and class III devices high risk, and the risk class drives the regulatory controls. The greater the risk, the higher the class and the more stringent the controls. The risk category is determined by the intended use. To determine a device's class, the FDA considers its intended use, labelling, indications and associated risks. Intended use is the general purpose or function of the device and encompasses the indications in its labelling; indications for use are the disease or condition the device will diagnose, treat, prevent, cure or mitigate, and include a description of the patient population for which it is intended. One device can have more than one risk category: medical support stockings are class I for general medical purposes and class II for preventing pooling of blood in the legs. The manufacturer initially proposes the risk category, and the FDA can concur with the proposal or override it. In 2013 there were approximately 780 low-risk, 800 medium-risk and 120 high-risk devices on the US market.
Class I devices are the least risky, and they include surgical masks and dental floss. They are not intended to support life or prevent health decline, and they do not pose unreasonable risks of injury or illness. To ensure their safety and effectiveness the FDA applies general controls, which include registration of both product and manufacturer, regulation to ensure that good manufacturing practice standards are upheld, and labelling according to strict guidelines. Class II devices, such as wheelchairs and infusion pumps, are riskier and more complex, and general controls are usually not sufficient to ensure their safety and effectiveness. So they are also subject to special controls, which may include particular labelling requirements, patient registries, postmarket surveillance and FDA-recognised performance standards. The majority of all devices fall in class II. Class III devices, such as HIV diagnostic tests and heart valves, make up the smallest proportion of devices on the market. They pose an unreasonable risk of illness or injury, are intended to sustain and improve life or prevent impairment of human health, and are subject to both general and special controls. An unclassified device is a preamendments device for which no classification regulation has been promulgated, and a not-classified device is a postamendments device for which the FDA has not yet reviewed a marketing application or made a final decision on one.
By potential for harm, a tongue depressor is a class I device, with the least; a battery-powered wheelchair is class II, with intermediate risk; and an implantable cardiac pacemaker is class III, with the highest. Class I devices present minimal potential for harm; examples are manual stethoscopes, adhesive bandages, crutches and tongue depressors. Class II devices carry a higher risk than class I devices; examples are syringes, pregnancy test kits, platelet-rich plasma separation kits and electric wheelchairs. Class III devices sustain or support life, are implanted, or present a potential unreasonable risk of illness or injury; examples are intraocular lenses, artificial heart valves, pacemakers and implanted prosthetics.
Reclassification
Reclassification of a device may be initiated by the FDA or in response to a petition from an outside party. Both pre- and postamendments devices can be reclassified, provided the FDA or the petitioner gives sufficient evidence that the device meets the appropriate class definition, presents new publicly available information on the device type, and identifies sufficient valid scientific evidence that the safety and effectiveness of the device type can be assured through the regulatory controls of the new class. A reclassification order must highlight the public health benefits of reclassification, the risks associated with the device type and how often they occur. When a preamendments device is reclassified from class III to class II, the order must provide evidence that the general controls are sufficient to ensure safety and effectiveness. If the reclassification is approved, an FDA-published order describes the reasons for the approval and the health risks posed by the device.
Significant-risk devices and investigational exemptions
The FDA treats every device as having either significant or nonsignificant risk. A significant-risk device, such as a cardiac pacemaker or an orthopaedic implant, has the potential for serious risks to the health, safety or welfare of patients; all other devices, such as contact lenses and ultrasonic dental scalers, are nonsignificant risk. The degree of risk and the amount of historical information determine the level of data needed to move a product to the US market.
To initiate clinical trials of a significant-risk device, the manufacturer or sponsor must submit an Investigational Device Exemption (IDE) and receive FDA approval before shipping via interstate commerce, an approval comparable to the Investigational New Drug application for pharmaceuticals. A nonsignificant-risk device does not require an IDE, but its study does require approval by an ethics panel, such as an institutional review board, before it can begin. A pivotal study of a significant-risk device may be performed under an FDA-approved IDE.
Routes to market
Once a device is classified, its manufacturer must submit one of two types of premarket submission to the FDA: the premarket notification, or 510(k), and the premarket approval (PMA). The 510(k) is named after the section of the Federal Food, Drug, and Cosmetic Act that defines it. Intended use is one criterion that determines whether a device must be evaluated for marketing through the 510(k) route, the PMA route or, if appropriate, a de novo request. A 510(k) notifies the FDA of a manufacturer's intention to market a device and demonstrates its substantial equivalence to a previously marketed or predicate device, and some class I, most class II and certain unclassified devices require one.
A 510(k) must show that the device is substantially equivalent to a predicate legally marketed in the United States before 1976 or previously cleared by the FDA. Predicates are usually 510(k)-cleared, but any legally marketed device can serve as one, as long as substantial equivalence to it can be established. Substantial equivalence means that the new device has the same intended use as the predicate and either the same technological characteristics or different ones that do not raise other questions of safety and effectiveness relative to it. The information submitted to the FDA must also demonstrate that the device is at least as safe and effective as the legally marketed device. A device substantially equivalent to a previously classified device falls into the same class as its predicate.
In 2013, approximately 90% of the devices available in the United States receive premarket notification 510(k) clearance. Laboratory testing is mandatory to obtain it. Clinical trial data are usually not required. The remaining 10% are approved by PMA, which is analogous to the new drug application for drugs. A PMA is required for class III high-risk devices with the potential to cause illness or injury (serious illness or injury, in a 2013 account), and for products found not substantially equivalent through the 510(k) route. Approval rests on sufficient valid scientific evidence that the device is safe and effective under the conditions of its intended use.
Because of the nature of their class, class III devices are subject to a premarket review different from that of class I or II devices, and without PMA approval a class III device cannot be legally marketed in the United States. In a PMA the sponsor must submit valid scientific evidence, based on clinical trials, that directly establishes device safety and efficacy, and must show that the benefits of the device outweigh its potential risks and that a large portion of the population will benefit from it.
Under certain circumstances a device may be exempt, as are the majority of class I and certain class II devices, and preamendments devices that have not been changed substantially and have no published requirements for additional regulatory control. Exemptions help free up FDA resources so that more time can be spent on the submissions that will significantly benefit public health. A device that qualifies for 510(k) exemption may go to market immediately but must meet the general controls.
A humanitarian-use device is a device intended to treat or diagnose rare diseases or conditions. The humanitarian device exemption application is similar to a PMA: scientific evidence must show that the benefits of the device outweigh its risks and that it will not expose patients to an unreasonable risk of injury or illness, but unlike a PMA it does not have to prove efficacy. In support of either a PMA or a humanitarian device exemption application, a device may receive an investigational device exemption that allows its use in clinical studies.
New types of devices that have no predicate are automatically classified as class III, and a manufacturer may submit a request for de novo classification to reclassify such a product into a lower class. De novo classification is for some low- to moderate-risk devices for which there is no legally marketed predicate. It may also be used if a product has undergone a 510(k) review without demonstrating substantial equivalence because no predicate exists, has a new intended use, or has technological characteristics that differ from its predicate's and raise questions about safety and efficacy. To be eligible, the general and special controls must be sufficient to ensure safety and efficacy, and the risks of the device must be clearly understood and have distinct ways of being mitigated.
The FDA's Breakthrough Devices Program, with its guidance document of December 2018, is meant to expedite the market availability of novel devices while retaining the statutory premarket approval standards. It gives manufacturers access to FDA experts to ascertain how to study the device and which regulatory path to use, and a Breakthrough Designation request can be sent to the FDA at any time before a marketing submission.
The risk class of a device determines its path to market. Class I devices are 510(k) exempt or need a 510(k); class II devices need a premarket notification 510(k) or a de novo classification, with or without Breakthrough Designation; and class III devices need premarket approval or a humanitarian device exemption, with or without Breakthrough Designation. The humanitarian device exemption is for patients with rare diseases or conditions.
What a clearance does and does not assert
In one 2013 account, politics and policy have shaped the regulation of the US medical device industry, with results that can at times put patients at risk or limit their ability to sue when harmed by new devices. The same account notes that higher-risk devices might logically be required to have some form of randomised controlled trial to prove their safety and effectiveness, but that for some devices this is not the case, and it gives policy and politics as the two reasons.
In the same account, unlike drugs and biologics, some higher-risk devices are regulated by the FDA with a grandfathering loophole: if a manufacturer can prove that a device is substantially equivalent to a similar device made before 28 May 1976, the effective date of the Medical Device Amendments of 1976, the FDA clears the device for market, no matter how great the patient risk. In this account, a substantially equivalent device must be at least as safe and effective as its pre-1976 predecessor. Devices on the 510(k) route are cleared for marketing, not approved. A 510(k) clearance does not mean that the FDA is asserting that a device is safe and effective; it means the FDA clears it for sale.
The same account gives three examples of problems the FDA's device policies have created. A metal-on-metal hip implant that the FDA cleared in 2008 through the 510(k) route had had no clinical trials, and its substantial equivalence rested on three pre-1976 devices, each with a different equivalent characteristic; each of the three had itself had high revision rates, meaning patients had to return for corrective surgery caused by problems with the designs, and the new hip was also beset with problems. A stent for intracranial stenosis was approved by the FDA in 2005 through a humanitarian device exemption on preliminary data in 45 patients; in the account's words, "they expected it to be so good, they wanted to get it out there quickly"; a later trial of 451 patients was stopped early because patients who received the stent had a significantly higher risk of stroke or death than patients who did not (15% against 6%, P = 0.002). The account's lesson is that unless proper clinical testing is done with these apparently innovative implantable devices, there may be problems lurking around the corner. A defibrillator lead approved through premarket approval, the FDA's most stringent approval process, proved susceptible to fracture and malfunction, and hundreds of patients' lawsuits over these problems were thrown out because the law governing that process prohibits lawsuits in state courts for PMA-approved devices.
The major device lobby, the Advanced Medical Technology Association, has as its primary goal strengthening the market by encouraging innovation. The FDA Modernization Act of 1997 directed the FDA to consider the least burdensome means of showing device effectiveness. A manufacturer directly pays the FDA a user fee for each 510(k) filed. The same account holds that these constraints on the FDA encourage rapid clearances for new products, paid for in large part by manufacturers. In 2013 the same account judged the 510(k) clearance process seriously flawed but likely to persist for some time.
In 2011 the FDA began implementing changes intended to address critical challenges to the 510(k) program and to enhance 510(k) decision-making.
Writing a submission for its reviewer
One practitioner advises the following when writing a device submission.
- Consider the reviewers: use the FDA website to find the names of likely reviewers, information on recent recalls and recommendations for similar devices.
- Consider the goal: is it to establish safety and substantial equivalence, or to prove safety and efficacy?
- Remember the audience, since regulatory environments vary around the world: the FDA uses "substantial equivalence" and "safe and effective", while the European Union uses "safety and performance".
- Evaluate the documents carefully, looking for misunderstandings and inconsistencies.
- Monitor the timeline for submissions: track it in a spreadsheet, educate the team about it, and be specific about what is needed from each team member to avoid redundancy.
- Stick to the facts: avoid unsubstantiated claims and make sure the documents do not present information in a biased manner.
- Remember the story: ask what the reviewer is to be brought to do and whether the device is equivalent or better, and state the same message consistently across the documents.