Do I need an IDE to study my device in humans?

It depends on whether your device qualifies for an exemption and whether it is classified as SR or NSR. Lawfully marketed devices studied without intent to change labeled use or intended use are generally exempt. For unapproved devices studied to support a marketing submission, the IDE framework typically applies. If there is any uncertainty, a Pre-Submission meeting with FDA is the most reliable way to get clarity before committing to a study design.

What is the difference between an SR and NSR device determination?

A significant risk device presents a potential for serious risk to subjects, typically because it is an implant, sustains life, or has a novel mechanism of action in a critical body system. An NSR device does not meet that threshold. The sponsor proposes the determination; the IRB reviews it. The practical consequence is that SR studies require FDA IDE approval before enrollment begins; NSR studies require IRB approval only.

Can I promote or sell my device while it is under IDE?

No. 21 CFR 812.7 prohibits promoting or test-marketing an investigational device before FDA has approved it for commercial distribution, prohibits charging subjects or investigators more than cost recovery, and prohibits representing that the device is safe or effective for the purpose being investigated. Promotion and commercialization must wait until the device receives clearance or approval through the appropriate marketing pathway.

What happens if my IDE study generates negative results?

Negative or mixed results from an IDE study do not prevent you from filing a premarket submission, but they must be disclosed. FDA requires complete reporting of study results, including results that are unfavorable to the device. An IDE study with mixed results may lead to an Additional Information request or, for PMAs, a Not Approvable letter. How you handle negative results strategically is a regulatory question; what you cannot do is omit them.

If I run a study without an IDE when one was required, what are the consequences?

Conducting a clinical study on a significant risk device without an IDE is a violation of 21 CFR Part 812. FDA may refuse to accept data from an improperly conducted study as part of a marketing submission, may issue a Warning Letter, and in serious cases may pursue enforcement action. If a study has already been conducted without an IDE, consult a regulatory attorney about how to disclose and address the situation before filing a marketing submission.

How does IDE relate to the Quality System / QMSR requirements?

Investigational devices are subject to the investigational-phase controls in 21 CFR Part 812 (including the manufacturing, processing, and control information required in the IDE application under 21 CFR 812.20) rather than to the full Quality Management System Regulation (21 CFR Part 820), which applies once the device enters commercial distribution following clearance or approval. Companies that implement QMSR-compliant quality systems before commercial distribution are typically better positioned for FDA inspection readiness at launch.

What Is an Investigational Device Exemption (IDE)? A Plain-Language Guide

Direct answer: An Investigational Device Exemption (IDE) is an FDA authorization that permits an unapproved medical device to be studied in human subjects to collect the safety and effectiveness data needed for a future premarket submission. The requirement comes from 21 CFR Part 812. Whether you need FDA's own IDE approval (versus approval from an Institutional Review Board only) depends on whether your device is classified as a significant risk device. Significant risk devices must have both FDA and IRB approval before any subjects are enrolled; non-significant risk devices need IRB approval only. (As of July 2026.)

The Federal Food, Drug, and Cosmetic Act (FD&C Act) generally prohibits the commercial distribution of unapproved medical devices. Clinical research, however, requires studying devices in humans before approval or clearance can be obtained. The IDE framework resolves this tension: it creates a legal pathway for studying unapproved devices in humans, under oversight conditions designed to protect research subjects, while preserving the ability to generate the clinical evidence needed for a regulatory submission.

The statutory basis is Section 520(g) of the FD&C Act. The implementing regulation is 21 CFR Part 812, available at eCFR. Part 812 establishes requirements for sponsors, investigators, and Institutional Review Boards (IRBs) in device clinical studies.

The IDE framework is not just a checkbox. It sets the conditions under which clinical data collected in an IDE study will be acceptable to FDA when submitted as part of a 510(k), De Novo, or PMA application. Data collected outside a properly constituted IDE may be rejected by FDA as not adequately controlled.

The Threshold Question: Do You Need an IDE at All?

Not every device study requires an IDE. 21 CFR 812.2(c) lists exemptions. A study is exempt from Part 812's IDE requirements (including both FDA IDE approval and the abbreviated IDE requirements for non-significant risk devices) when the device is lawfully marketed and the study is not intended to support a change in intended use or significant labeling change that would require a new premarket submission; when the device is a diagnostic device that poses no significant risk and the testing is non-invasive and does not require an invasive sampling procedure that presents significant risk; when the device is a consumer product not intended for direct administration to subjects; or when the study involves veterinary devices or devices studied solely in vitro (outside the human body).

For any device that does not fit an exemption and is being studied in human subjects for the purpose of supporting a premarket submission, the IDE framework applies. This covers the vast majority of significant clinical trials for novel devices.

Significant Risk vs. Non-Significant Risk: The Core Distinction

The most consequential determination under Part 812 is whether your device is a Significant Risk (SR) device or a Non-Significant Risk (NSR) device. This determination drives whether FDA must approve the study before it begins.

What makes a device significant risk?

Under 21 CFR 812.3(m), a significant risk device is an investigational device that is intended as an implant and presents a potential for serious risk; is purported or represented to be for use in supporting or sustaining human life and presents a potential for serious risk; is for a use of substantial importance in diagnosing, curing, mitigating, or treating disease or otherwise preventing impairment of human health and presents a potential for serious risk; or otherwise presents a potential for serious risk to the health, safety, or welfare of a subject.

Examples FDA has historically included in the SR category include cardiac pacemakers, implantable spinal stimulators, orthopedic implants for load-bearing joints, and any device with a novel mechanism of action in a life-sustaining role.

What makes a device non-significant risk?

An NSR device is any investigational device that does not meet the SR definition. FDA has given examples including daily-wear contact lenses (not extended-wear or implanted), certain diagnostic ultrasound devices at established power levels, and Foley catheters. The common thread is that the device, even if unapproved, does not present a potential for serious risk to subjects in the context of the proposed study.

Who makes the determination?

The sponsor makes an initial SR/NSR determination. That determination is then presented to the IRB reviewing the study protocol. The IRB reviews the determination and either agrees or disagrees. If the IRB disagrees with the sponsor's NSR determination and finds the device is SR, the sponsor must treat the device as SR and obtain FDA IDE approval. If both the sponsor and IRB agree the device is NSR, the study may proceed under the abbreviated NSR requirements without a formal FDA IDE submission.

When there is genuine uncertainty about SR vs. NSR classification, sponsors can submit a Pre-Submission (Q-Sub) to FDA asking for the agency's view before committing to a study design.

Requirements for Significant Risk Device Studies

If your device is SR, you must submit an IDE application to FDA and receive approval before enrolling any subjects. The IDE application must include (under 21 CFR 812.20) the name and address of the sponsor; a report of prior investigations, including any animal studies, laboratory testing, and prior clinical experience; an investigational plan, including the protocol, risk analysis, and description of how informed consent will be obtained; a description of the device, including technical specifications; manufacturing information sufficient for FDA to evaluate device safety; a monitoring plan for the study; copies of all investigator agreements, IRB approvals obtained to date, and informed consent documents; and a financial disclosure statement.

FDA's review timeline for IDE applications

Under 21 CFR 812.30, an investigation of a significant risk device may not begin until FDA approves the IDE by order, or until 30 days after FDA receives the application unless FDA notifies the sponsor within that window that the investigation may not begin. In practice FDA acts within the 30-day window, and significant risk studies proceed on the basis of an affirmative FDA approval order rather than on the lapse of the clock alone. For devices in the Breakthrough Devices Program, IDE applications receive priority review.

FDA may approve an IDE with conditions, disapprove it, or place a clinical hold (halting enrollment) at any point during the study if new safety concerns emerge.

Requirements for Non-Significant Risk Device Studies

NSR device studies do not require FDA IDE approval. They must still comply with abbreviated requirements under 21 CFR 812.2(b), which include IRB approval before initiating the study, labeling that identifies the device as an investigational device, monitoring of the study, records and reports as specified in Part 812, and the prohibitions on promotion and commercialization in 21 CFR 812.7.

The NSR abbreviated requirements exist because even non-significant risk devices studied in humans require oversight. The difference is that IRB oversight is considered sufficient without FDA's independent review.

Key Sponsor and Investigator Obligations Under Part 812

Whether the device is SR or NSR, Part 812 distributes responsibilities across sponsors (typically the device manufacturer) and investigators (typically the clinicians conducting the study).

Sponsor obligations include: selecting qualified investigators and providing them with adequate information about the device; monitoring the study to ensure compliance with the protocol and Part 812; withdrawing investigators who fail to comply; maintaining records and submitting required reports to FDA (for SR studies); ensuring the device is labeled correctly as investigational; and observing the promotion and commercialization prohibitions in 21 CFR 812.7. That section prohibits promoting or test-marketing an investigational device before FDA has approved it for commercial distribution, prohibits commercializing it by charging subjects or investigators more than the cost of manufacture, research, development, and handling, and prohibits representing that an investigational device is safe or effective for the purpose being investigated.

Investigator obligations include: obtaining informed consent from each subject; following the investigational plan and FDA regulations; controlling the investigational device; and maintaining required records and submitting reports to the sponsor.

The prohibition on promotion in 21 CFR 812.7 is a common compliance gap. Sponsors sometimes begin commercial activities, including advertising or sales conversations, before clearance or approval, relying on IDE status as a shield. It is not. 21 CFR 812.7 prohibits promoting or test-marketing an investigational device before FDA has approved it for commercial distribution, prohibits charging above cost recovery, and prohibits representing that an investigational device is safe or effective for the purpose being investigated.

How IDE Clinical Data Feeds a 510(k), De Novo, or PMA Submission

The point of an IDE study is not the study itself. The point is the data that study generates, which must be sufficient to support the premarket submission that follows.

510(k)

Most 510(k)s do not require clinical data because substantial equivalence is established through comparison to a predicate using bench and performance data. However, when a device's technological differences from the predicate raise questions that bench testing cannot resolve, clinical data may be required. In those cases, an IDE study generates the data that closes the equivalence argument.

De Novo

De Novo submissions establish new special controls for novel Class II devices. When clinical performance data is needed to establish the appropriate special controls, an IDE study provides that data.

PMA

PMA submissions require valid scientific evidence, which for Class III devices typically means clinical data from one or more well-controlled studies. IDE studies are the mechanism by which that clinical data is generated under conditions that FDA will accept. A PMA clinical study that was not conducted under an IDE (or that deviated materially from its IDE-approved protocol) may result in FDA declining to rely on the data.

The earlier in the IDE planning process that sponsors discuss their submission strategy with FDA, the better aligned the study design will be with what FDA will actually need to see in the marketing submission. This is one of the key topics for a Pre-Submission (Q-Sub) meeting. See the companion article on FDA Pre-Submission meetings.

The Role of the IRB

An Institutional Review Board is an independent committee required by 21 CFR Part 56 to review and approve research involving human subjects. Every IDE study, whether SR or NSR, must have IRB approval. The IRB's function is independent of FDA's: the IRB protects subjects; FDA protects the public health and the integrity of the clinical data.

For SR studies, IRB approval is required in addition to FDA IDE approval. Obtaining one does not substitute for the other. For NSR studies, IRB approval is the primary oversight mechanism.