Written and maintained by CASRAI Editorial Board
Last updated
Last verified: August 30, 2026, against the electronic Code of Federal Regulations (eCFR), Title 21, Part 812 (ecfr.gov/current/title-21/chapter-I/subchapter-H/part-812). This is a regulatory-orientation guide, not legal or regulatory advice — confirm any specific determination with your IRB and, where relevant, with FDA’s Center for Devices and Radiological Health (CDRH) or a qualified regulatory affairs professional.
What an Investigational Device Exemption is
An Investigational Device Exemption (IDE) is the FDA mechanism that allows a medical device that has not yet been cleared or approved for commercial use to be shipped and used in a clinical study, so a sponsor can collect the safety and effectiveness data FDA needs to eventually clear or approve it. The regulation itself is 21 CFR Part 812, “Investigational Device Exemptions.”
The name describes exactly what it does: normally, distributing an unapproved medical device would violate FDA’s premarket requirements. An approved IDE exempts the sponsor from those requirements for the specific, IRB-and-FDA-overseen study at hand — it is a permission to study the device, not an approval to sell it. Once the study is complete, the resulting data becomes part of the evidence package for the device’s eventual marketing pathway — typically 510(k) clearance or PMA approval.
It is worth separating two related but distinct ideas that share the “investigational” label: an investigational device is the device’s regulatory status (it is currently the object of a study, not in ordinary commercial use); the Investigational Device Exemption is the mechanism that makes studying it lawful. A device doesn’t have to be brand-new hardware to be investigational — an already-cleared device studied for a new indication, population, or use outside its cleared labeling becomes investigational again for that study.
Why it exists
FDA’s premarket clearance and approval pathways exist to keep devices with unproven safety or effectiveness off the market. But that same requirement would make it impossible to generate the clinical evidence those pathways need in the first place — you cannot prove a device is safe and effective for its intended use without studying it in people first. The IDE regulation resolves that: it lets a defined, monitored, IRB-and-informed-consent-governed study proceed under controlled conditions, without treating the study itself as commercial distribution.
Significant risk vs. nonsignificant risk devices
Not every device study needs a full FDA-reviewed IDE application. 21 CFR 812.3(m) sorts device studies into two tracks:
- Significant risk (SR) studies — the device is an implant, supports or sustains life, is substantially important in diagnosing/treating/curing/mitigating disease, or otherwise presents a potential for serious risk to a subject. SR studies require a full IDE application submitted to CDRH under 21 CFR 812.20, and the study cannot begin until FDA approves it (or, for most non-banned devices, 30 days pass after FDA receives the application without a hold).
- Nonsignificant risk (NSR) studies — everything that doesn’t meet the SR criteria. These follow abbreviated IDE requirements: IRB approval of the investigation, device labeling under 812.5, monitoring, recordkeeping, and informed consent — generally without a separate formal application to FDA, unless FDA has specifically told the sponsor otherwise.
The sponsor makes the initial SR/NSR call, but the reviewing IRB confirms it, and FDA retains authority to overrule an NSR determination. This distinction, and the full criteria behind it, is covered in depth in IDE and the Significant Risk Determination (21 CFR 812).
IDE vs. IND: the device/drug parallel
An IDE is the device-side counterpart to an Investigational New Drug (IND) application: both let a sponsor study an unapproved product in humans before it reaches the market, both exist under Title 21 of the CFR, and both are reviewed by an FDA center — CDRH for devices under 21 CFR Part 812, CDER or CBER for drugs and biologics under 21 CFR Part 312. The two pathways diverge in structure past that point (IDE’s SR/NSR tiering has no direct IND equivalent; IND has its own phased-trial and safety-reporting framework). See IND vs. IDE: FDA’s Drug and Device Pathways Compared for the full side-by-side.
IDE vs. 510(k) and PMA
These answer different questions in the device lifecycle. An IDE governs the study that generates evidence about a not-yet-marketed device. A 510(k) or PMA submission is what a sponsor files afterward (or, for lower-risk devices, sometimes without a device study at all) to actually market the device commercially: a 510(k) shows the device is substantially equivalent to an already-legally-marketed predicate device, while a PMA is the more rigorous pathway for higher-risk (Class III) devices without a suitable predicate. In short: IDE gets you permission to study; 510(k)/PMA gets you permission to sell. Many SR-device programs move through both in sequence — IDE-supported clinical data feeding directly into the PMA submission.
The IDE application, briefly
For SR studies, the application itself is a defined package under 21 CFR 812.20(b): sponsor information, a report of prior (nonclinical and any earlier clinical) investigations, the investigational plan, manufacturing information, investigator agreements, IRB information, labeling, and informed consent materials, among other items. FDA notifies the sponsor of the date it received the application and then has 30 days to approve, approve with conditions, or disapprove it. The full mechanics — what has to be in the package and how FDA’s review clock works — are covered in IDE Application (Investigational Device Exemption Application).
Once a device has an approved IDE, FDA also assigns it a Medicare coverage category (A or B) that determines whether Medicare will pay for the device itself, not just routine care costs, during the study — a distinction that matters directly for site budgeting. See Category A vs. Category B IDE Studies: Medicare Coverage Differences.
Who deals with this in a research organization
IDE questions typically land with several roles at once, not one:
- Research administrators / clinical trial office staff track which studies at their institution involve an IDE, coordinate the IRB submission alongside (not instead of) the FDA application for SR studies, and manage the site-level budgeting implications of Category A vs. B coverage.
- IRB staff and reviewers make or confirm the SR/NSR determination for studies that come through their board, and hold ongoing review responsibility for NSR studies that never go to FDA at all.
- Regulatory affairs staff (whether centralized or embedded in a department) typically own drafting and submitting the IDE application itself and tracking FDA’s 30-day review clock and any required supplements.
- Sponsors and study coordinators at device-study sites need to know a study is running under an IDE because it changes what can be shipped, how the device must be labeled, and what supply/tracking discipline applies before the site can enroll its first subject — see Clinical Trial Site Startup: Supply Checklist for how investigational-device supply and accountability fits into broader site-startup logistics.
Frequently asked questions
Does every device study need an IDE application?
No. Only significant risk (SR) studies require a full IDE application submitted to and approved by FDA. Nonsignificant risk (NSR) studies proceed under abbreviated requirements overseen by the IRB, without a separate formal FDA application in the typical case.
Is an IDE the same thing as FDA approval to sell the device?
No. An IDE only authorizes using the device in a specific, defined clinical study. Commercial distribution requires a separate 510(k) clearance or PMA approval, which typically comes after the IDE-supported study is complete.
How is an IDE different from an IND?
They’re parallel mechanisms for different product types: IDE (21 CFR Part 812, reviewed by CDRH) covers medical devices; IND (21 CFR Part 312, reviewed by CDER or CBER) covers drugs and biologics. See the full comparison for how their processes diverge from there.
Who decides whether a study is significant risk or nonsignificant risk?
The sponsor makes an initial determination, the reviewing IRB confirms or disputes it, and FDA retains authority to overrule an NSR determination if it disagrees.
