The FDA assigns every medical device sold in the United States to one of three regulatory classes based on the risk it poses to patients and users. Class I covers the lowest-risk devices, Class II covers moderate-risk devices, and Class III covers the highest-risk devices. The higher the class, the more evidence a manufacturer must provide to prove the device is safe and effective before it can reach the market.
How the FDA Classifies Devices
The classification system is built around one central question: how much regulatory control is needed to guarantee a device is safe? Every medical device, regardless of class, must meet a baseline set of requirements called General Controls. These include things like proper labeling, manufacturing standards, and registration with the FDA. For higher-risk devices, additional layers of oversight are added on top of that baseline.
The FDA recognizes thousands of generic device types, and each one is slotted into a class. When a company develops a new device, it identifies which generic type the product fits into (or whether it’s truly novel) and follows the regulatory pathway that matches.
Class I: Lowest Risk
Class I devices are simple, well-understood products where the chance of harming a patient is minimal. Think tongue depressors, elastic bandages, manual stethoscopes, examination gloves, and handheld surgical instruments like scalpels. These devices don’t enter the body in complex ways, and failures are unlikely to cause serious injury.
The regulatory burden for Class I is the lightest. Manufacturers must follow General Controls, which cover proper labeling, good manufacturing practices, and adverse event reporting. Most Class I devices are also exempt from the premarket notification process (known as a 510(k) submission), meaning manufacturers can bring them to market without filing a formal application with the FDA beforehand. They still need to register their facility and list their device, but the review step is skipped entirely for most products in this class.
Class II: Moderate Risk
Class II is the largest category and covers devices that carry more risk than a tongue depressor but don’t rise to the level of, say, a heart valve. Examples include powered wheelchairs, pregnancy test kits, clinical electronic thermometers, infusion pumps, surgical drapes and gowns, contact lenses, and certain diagnostic imaging systems.
These devices must meet General Controls plus an additional layer called Special Controls. Special controls are tailored to each device type and can include specific performance standards, post-market surveillance requirements, patient registries, or special labeling guidelines. The idea is to address risks that General Controls alone can’t cover.
Most Class II devices require a 510(k) premarket notification before they can be sold. In a 510(k), the manufacturer demonstrates that the new device is “substantially equivalent” to a device already legally on the market (called a predicate device). This doesn’t require clinical trials in most cases. Instead, the company typically submits bench testing, engineering data, and sometimes limited clinical evidence showing the device performs comparably to its predicate. Some Class II devices are exempt from the 510(k) requirement, but they still must comply with all applicable Special Controls.
Class III: Highest Risk
Class III devices support or sustain human life, play a critical role in preventing serious health impairment, or present a potential unreasonable risk of illness or injury. Heart valves, implantable pacemakers, coronary stents, implantable defibrillators, and deep brain stimulators all fall into this category. So do some high-risk diagnostic tests and novel technologies that don’t have a comparable device already on the market.
Because the stakes are highest, Class III devices go through the most rigorous regulatory pathway: Premarket Approval, or PMA. A PMA application requires the manufacturer to submit extensive non-clinical lab data (covering areas like biocompatibility, toxicology, stress testing, and shelf life) along with full clinical trial results demonstrating safety and effectiveness in humans. The clinical section must include study protocols, patient-level data, adverse event reports, device failure rates, and statistical analyses. The FDA reviews all of this before deciding whether to approve the device.
This is a fundamentally different standard than what Class II devices face. A 510(k) asks “is this device substantially equivalent to something already on the market?” A PMA asks “does the clinical evidence prove this device is safe and effective?” That distinction matters: PMA is an approval, while 510(k) is technically a clearance.
Cost and Timeline Differences
The gap between the two pathways shows up clearly in both fees and review times. For fiscal year 2026, the FDA’s standard application fee for a 510(k) is $26,067, while a PMA application costs $579,272. Small businesses get reduced rates of $6,517 for a 510(k) and $144,818 for a PMA, but the cost difference is still dramatic.
Review timelines reflect the difference in complexity. The FDA’s target for total time to decision on a 510(k) is 112 calendar days as of 2025. For an original PMA, the target is 285 calendar days. These are goals, not guarantees, and they include time for both FDA review and manufacturer responses to questions. In practice, a PMA can stretch considerably longer if the FDA requests additional data or if the clinical evidence raises questions.
Exemptions and Edge Cases
The system isn’t as rigid as it first appears. Most Class I devices and some Class II devices are exempt from the 510(k) requirement entirely. Exemption means the FDA has determined that a premarket notification isn’t necessary to ensure safety and effectiveness for that particular device type. However, exemptions only apply to devices that match the characteristics of products already on the market within that generic type. If a manufacturer makes a novel change to an otherwise exempt device, they may still need to file a 510(k).
Even exempt devices must comply with all other applicable regulatory controls. Skipping the 510(k) doesn’t mean skipping regulation altogether.
Software as a Medical Device
Software adds a layer of complexity to the classification framework. The FDA uses a separate risk categorization system for Software as a Medical Device (SaMD) that considers two factors: how serious the patient’s health condition is, and how significant the software’s output is to a clinical decision.
Software that directly treats or diagnoses a critical condition falls into the highest risk category, while software that merely informs clinical decisions about non-serious conditions sits at the lowest. For example, an app that analyzes cardiac imaging to diagnose a life-threatening arrhythmia would face far more scrutiny than one that tracks mild skin conditions for a dermatologist’s reference. The FDA maps these risk levels onto its existing Class I, II, and III framework to determine which regulatory pathway applies.
How the EU System Compares
Outside the United States, classification works differently. The European Union uses a four-tier system under its Medical Device Regulation: Class I, IIa, IIb, and III. The EU system splits the FDA’s broad Class II into two subcategories based largely on how long a device stays in the body. Class IIa devices are generally placed in the body for short periods (between 60 minutes and 30 days), while Class IIb devices are typically implanted for 30 days or longer. EU Class III, like FDA Class III, covers the highest-risk devices.
This means a device classified as Class II in the U.S. could land in either IIa or IIb in Europe, with different regulatory requirements for each. Companies selling devices internationally need to navigate both systems, and a device’s classification in one region doesn’t automatically determine its classification in another.