What FDA device classification decides
FDA classification for medical devices is the risk-based system FDA uses to decide what level of regulatory control is needed to provide reasonable assurance of safety and effectiveness. For a manufacturer planning U.S. market entry, classification affects whether a device may be exempt from premarket notification, whether it typically needs a 510(k), whether a De Novo request may be appropriate, or whether a premarket approval application is required.
Classification is not based on marketing preference or product complexity alone. FDA looks at the device type, intended use, indications for use, risks to patients or users, and the controls needed to manage those risks. This overview explains the three FDA device classes, how classification connects to U.S. market entry, and where classification mistakes commonly occur. It is based on FDA device classification materials, 21 CFR Part 860, FDA database descriptions and FDA regulatory control guidance current to August 2026.
The three FDA medical device classes
The FDA system places medical devices into Class I, Class II or Class III. These classes are often described as low, moderate and high risk. In practice, the more precise question is whether general controls, special controls, or premarket approval are needed for a particular generic device type.
| FDA class | Typical risk level | Main regulatory controls | Common market entry implication |
|---|---|---|---|
| Class I | Lowest risk to low or moderate risk | General controls | Many are exempt from 510(k), but other FDA requirements may still apply |
| Class II | Moderate to higher risk | General controls plus special controls | Often requires 510(k) unless the device type is exempt or uses another pathway |
| Class III | Highest risk or insufficient information for lower classification | General controls plus premarket approval requirements | Usually associated with PMA when the PMA requirement applies |
Class I devices
Class I devices are subject to general controls. FDA uses this class when general controls are sufficient, or when the device does not raise the life-supporting, life-sustaining, substantial-health-importance or unreasonable-risk concerns that would require Class III treatment. Class I devices are usually simpler and lower risk than higher-class devices, but Class I does not mean unregulated.
General controls may include establishment registration and device listing, labeling controls, records and reports, medical device reporting where applicable, prohibitions on adulteration or misbranding, and quality requirements unless an exemption applies. A Class I device may also have limitations on exemption, so the exact regulation and product code still matter.
Class II devices
Class II devices require general controls plus special controls. Special controls are additional measures FDA considers necessary for a generic type of device. They may include performance standards, special labeling, bench testing, clinical data expectations, postmarket surveillance, patient registries, guidance recommendations or other controls tailored to the risks of that device type.
Most devices that need more than general controls but do not require Class III treatment fall into Class II. Many Class II devices enter the U.S. market through the 510(k) pathway by demonstrating substantial equivalence to a legally marketed predicate device. Some Class II device types are exempt from 510(k), but exemption must be verified against the specific classification regulation and any stated limitations.
Class III devices
Class III is used for devices that support or sustain human life, are of substantial importance in preventing impairment of human health, may present a potential unreasonable risk of illness or injury, or lack sufficient information to show that general and special controls would be adequate. Class III devices generally face the most demanding premarket review burden.
For many Class III devices, the relevant route is premarket approval, commonly called PMA. PMA is not the same as 510(k) clearance. A PMA application requires scientific and regulatory evidence to support safety and effectiveness for the device. Because some device types have historical classification status and specific regulatory requirements, the exact FDA regulation and database entry should always be checked before assuming the submission type.
Controls are cumulative, but exemption is not deregulation
A common mistake is to treat classification as only a premarket submission question. It is broader than that. Classification determines the level of regulatory control, and many of those controls continue after market entry.
General controls apply across all three classes unless a regulation provides a specific exemption. These baseline controls can include registration, listing, labeling compliance, complaint and adverse event reporting obligations, corrections and removals reporting, and compliance with device quality system requirements. Since February 2, 2026, FDA has used the Quality Management System Regulation, or QMSR, under 21 CFR Part 820. The QMSR incorporates ISO 13485:2016 by reference and replaced the older Quality System Regulation framework for FDA device current good manufacturing practice requirements.
Special controls are added for Class II devices when general controls alone are not sufficient. PMA requirements are associated with Class III devices when FDA determines that approval is needed to establish reasonable assurance of safety and effectiveness. The compliance point is practical: a 510(k)-exempt Class I or Class II device may avoid one premarket submission, but it may still need registration, listing, labeling controls, quality system compliance and postmarket obligations.
How classification connects to 510(k), De Novo and PMA
FDA classification should be determined before a company selects a submission strategy. A business may prefer a shorter or less burdensome route, but the product’s intended use, technology and risk profile drive the pathway.
510(k) clearance
A 510(k) is used when a device must show substantial equivalence to a legally marketed predicate device and is not exempt from premarket notification. Predicate analysis is more than a feature comparison. It must consider intended use, technological characteristics, performance data and whether differences raise new questions of safety and effectiveness.
Many Class II devices and some non-exempt Class I devices use 510(k). However, a device cannot use 510(k) simply because it looks similar to another product. The predicate must be legally marketed and appropriate for the intended use and technological comparison.
De Novo classification
The De Novo pathway is for certain novel devices for which there is no legally marketed predicate, but where general controls alone or general and special controls can provide reasonable assurance of safety and effectiveness. A successful De Novo request classifies the device type into Class I or Class II and can create a new classification regulation and product code. Future devices of the same type may then be able to use that device as a predicate, depending on the circumstances.
Premarket approval
PMA is the most stringent premarket pathway and is generally associated with Class III devices. It is used when FDA needs valid scientific evidence to evaluate safety and effectiveness for a high-risk device. PMA review usually involves a deeper evidence package than 510(k), and post-approval requirements may also apply. See also: Implants.
The pathway decision is therefore not a simple choice among three forms. It begins with classification, then moves to the regulatory controls and submission type that apply to the identified device type.
How to classify a device in practice
For a manufacturer, importer or regulatory team, classification should be recorded as a structured rationale rather than an informal database search. A practical approach includes the following steps.
- Define the device and its intended use. FDA classification depends heavily on intended use and indications for use. Claims in labeling, instructions, promotional materials and sales communications can affect this analysis.
- Identify the generic device type. FDA has classified approximately 1,700 generic types of devices across 16 medical specialty panels in 21 CFR Parts 862 through 892.
- Search FDA classification resources. The Product Classification Database identifies device names, product codes, review panels, regulation numbers, device class and other regulatory information. FDA states that product code files are updated weekly.
- Read the classification regulation. The regulation usually provides the device identification, class and relevant notes. Do not rely only on a product code name if the device has different technology, indications or risk characteristics.
- Check exemption status and limitations. Some Class I and Class II devices are exempt from 510(k), but exemptions can be limited. If the device exceeds the limitation of exemption, a 510(k) may still be required.
- Review potential predicates. If the device appears to be non-exempt Class I or Class II, assess whether a legally marketed predicate supports a 510(k) strategy.
- Consider De Novo or 513(g) when classification is uncertain. De Novo may apply to novel low- to moderate-risk devices without a predicate. A 513(g) request can be used to ask FDA for information about classification and applicable requirements.
Teams working under U.S. device regulation should maintain a written classification memo that connects claims, intended use, product code, regulation number, class, exemption status and pathway rationale. For broader regulatory updates, see the Regulatory section.
Common mistakes and edge cases
Assuming Class I means no FDA obligations
Class I often carries fewer requirements than Class II or Class III, and many Class I devices are 510(k)-exempt. That does not automatically remove registration, listing, labeling, reporting or quality system obligations. The device-specific regulation and exemption notes must be checked.
Classifying by technology instead of intended use
Two devices may use similar hardware or software but fall into different classifications if their intended uses differ. A tool used for wellness tracking may be treated differently from a tool intended to diagnose, treat or guide clinical decisions. Classification analysis should start with the proposed medical purpose, not the engineering platform.
Using an unsuitable predicate
A predicate device must be legally marketed and relevant to the same intended use and technological comparison. If differences create new safety or effectiveness questions, a 510(k) may not be appropriate without additional evidence, and De Novo may need to be considered for a novel low- or moderate-risk device type.
Overlooking accessories
Accessories can have their own classification analysis rather than automatically inheriting the class of the parent device. FDA has continued to examine distinct classification of accessories. On August 17, 2026, FDA announced a request for public comment on a proposed list of existing medical device accessories that may or may not be suitable for distinct classification into Class I, with comments due by October 16, 2026. That activity shows why accessory classification should be assessed based on the accessory’s own intended use and risk controls.
Confusing FDA cleared, approved and authorized
In U.S. device regulation, terminology matters. A 510(k) device is generally described as cleared, a PMA device as approved, and De Novo as granted or authorized. Using the wrong term in labeling, investor materials or product pages can create regulatory and credibility problems.
Frequently asked questions
What is the main purpose of FDA classification for medical devices?
The main purpose is to assign the level of regulatory control needed to provide reasonable assurance of safety and effectiveness. Classification affects premarket pathway, evidence expectations, applicable controls and postmarket obligations.
Does every Class II device need a 510(k)?
No. Many Class II devices require 510(k), but some Class II device types are exempt from premarket notification. The classification regulation, product code database and exemption limitations should be reviewed before deciding that a submission is or is not required.
Can a novel device avoid Class III if it has no predicate?
Possibly. A novel device without a predicate may be eligible for De Novo classification if it is low to moderate risk and general controls or general plus special controls can provide reasonable assurance of safety and effectiveness. If those controls are not sufficient, Class III and PMA may be required.
Is product code the same as device class?
No. A product code identifies the generic category of a device for FDA tracking and review purposes. The product code is linked to classification information, but teams should still review the regulation number, device description, class, review panel, exemption status and submission type.
When should a company consider a 513(g) request?
A 513(g) request may be useful when classification or applicable FDA requirements remain uncertain after reviewing FDA regulations and databases. It is not a substitute for a complete regulatory strategy, but it can provide formal FDA information about classification and requirements for a particular device.
Key takeaway
FDA device classification is the starting point for U.S. medical device market planning. It links the product’s intended use and risks to the controls FDA considers necessary. A reliable classification analysis should identify the generic device type, product code, regulation number, class, exemption status and likely pathway before a company commits to testing, labeling or submission planning.