FDA Medical Device Classification System

The FDA medical device classification system assigns every device sold in the United States to one of three regulatory classes based on the level of risk it poses to patients and the controls necessary to provide reasonable assurance of safety and effectiveness. Understanding this framework is foundational for manufacturers, importers, and healthcare procurement professionals because the assigned class determines which premarket pathway — if any — a device must complete before it can reach the market. The FDA Medical Device Classification framework directly governs more than 1,700 distinct device types across 16 medical specialty panels (FDA Device Classification Database).


Definition and scope

The classification system is codified under 21 CFR Parts 862–892, which enumerate specific device types organized by medical specialty — from cardiovascular and orthopedic to ophthalmic and neurological devices. The Food, Drug, and Cosmetic Act (FD&C Act), as amended by the Medical Device Amendments of 1976 and later by the Safe Medical Devices Act of 1990, establishes the three-class structure and grants FDA authority to reclassify devices as evidence of risk accumulates (FDA: Overview of Device Regulation).

Scope extends to devices intended for human use: diagnostic equipment, surgical instruments, implants, in vitro diagnostic tests, and software that meets the statutory definition of a device. Combination products — items that blend a device with a drug or biologic — fall under separate coordination rules managed by the FDA's Office of Combination Products, described further on the FDA Combination Products Regulation page.


How it works

Classification is triggered when a device type is formally recognized, either through the original 1976 panel classification process, a 510(k) substantial equivalence determination, a De Novo authorization, or a Premarket Approval (PMA) order. Once a device type is classified, all subsequent devices of the same type inherit that classification unless a manufacturer petitions for reclassification.

The three classes operate as follows:

  1. Class I — General Controls
    Devices that present minimal potential for harm and are subject only to general controls: registration, listing, labeling, good manufacturing practices (Quality System Regulation/QSR), and prohibition against adulteration or misbranding. Approximately 47% of all device types fall into Class I (FDA Device Classification Database). Most Class I devices are exempt from premarket notification; examples include elastic bandages, examination gloves, and tongue depressors.

  2. Class II — Special Controls
    Devices for which general controls alone are insufficient to provide reasonable assurance of safety and effectiveness, but for which there is enough information to establish special controls — such as performance standards, post-market surveillance requirements, or guidance documents. Class II devices typically require a 510(k) clearance before marketing. Examples include powered wheelchairs, infusion pumps, and surgical drapes.

  3. Class III — Premarket Approval
    Devices that support or sustain human life, are of substantial importance in preventing impairment, or present a potential unreasonable risk of illness or injury. These require a Premarket Approval (PMA) application, the most stringent premarket pathway. Examples include implantable pacemakers, cochlear implants, and certain heart valves.

A device that cannot be classified through substantial equivalence to a predicate — and for which low-to-moderate risk is demonstrated — may pursue the De Novo device pathway to establish a new Class I or Class II classification with special controls.


Common scenarios

Scenario 1: A manufacturer introduces a new digital blood pressure monitor.
Blood pressure monitors are classified under 21 CFR § 870.1130 as Class II devices subject to special controls, requiring 510(k) clearance. The manufacturer must identify a legally marketed predicate device and demonstrate substantial equivalence in intended use and technological characteristics.

Scenario 2: A startup develops an AI-driven diagnostic software tool.
Software as a Medical Device (SaMD) is evaluated using FDA's risk-based framework aligned with the International Medical Device Regulators Forum (IMDRF) guidance. Depending on the significance of the information provided and the severity of the condition, the software could be Class I, II, or III. FDA's approach to digital health and software regulation addresses these evolving product types specifically.

Scenario 3: An importer brings in a novel bone-anchored hearing aid.
This device type carries a Class III designation because it is implanted and supports sensory function. The importer's foreign manufacturer must hold a valid PMA, and the importer must comply with FDA import regulations before the devices clear customs.


Decision boundaries

The most consequential boundary in the classification system is the line between Class II and Class III, because it separates the 510(k) pathway — which does not require independent clinical trials — from the PMA pathway, which requires valid scientific evidence, typically including clinical data (FDA: PMA Submissions).

Key decision factors at each boundary:

For regulated entities seeking a comprehensive orientation to FDA's authority and the full breadth of its regulatory reach, the FDA Authority overview provides structural context across all product categories the agency oversees.