Substantial Equivalence for US FDA 510(k) Clearance
Substantial equivalence is the FDA finding that your new medical device is as safe and effective as a legally marketed predicate device: it has the same intended use, and any differences in technology do not raise different questions of safety and effectiveness. We build that argument, and the test evidence behind it, for your 510(k).
What is substantial equivalence in a 510(k)?
Substantial equivalence (SE) is the standard the US FDA uses to clear a medical device through the 510(k) premarket notification pathway. Instead of proving safety and effectiveness from scratch, as a PMA does, you show that your device is at least as safe and effective as a device that is already legally on the US market.
The concept comes from section 513(i) of the Federal Food, Drug, and Cosmetic Act and is applied through 21 CFR 807.100(b). FDA explains how reviewers assess 510(k) substantial equivalence in its 2014 guidance, The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications. The device you compare against is called the predicate device.
SE does not mean identical. A new device can use a different material, a new software feature or a different design and still be found substantially equivalent, as long as those differences do not raise different questions of safety and effectiveness and your test data show the device performs as well as the predicate.
In one sentence: SE = same intended use + same technology, or different technology that raises no new safety and effectiveness questions and is backed by data.
Most Class II devices and some Class I devices reach the US market this way. To check whether your device needs a 510(k) at all, start with its FDA device classification and product code.
What are the substantial equivalence criteria in 21 CFR 807.100?
Under 21 CFR 807.100(b), FDA finds a device substantially equivalent when every one of the conditions below is met. Missing any one of them leads to a not substantially equivalent (NSE) decision.
A legally marketed predicate
The predicate must be legally marketed in the US and must not have been removed from the market by FDA or found adulterated or misbranded by a court.
Same intended use
Your device must have the same intended use as the predicate. Indications for use can differ slightly, provided the general purpose and function stay the same.
Same technological characteristics
If the materials, design, energy source and other features match the predicate, the device can be found SE on that basis, supported by appropriate data.
Or different, without new questions
If the technology differs, the differences must not raise different questions of safety and effectiveness than the predicate does.
Data showing it is as safe and effective
Where technology differs, your submission must include information, usually performance testing, showing the device is as safe and effective as the legally marketed device.
Acceptable scientific methods
The test methods used to compare the devices must be scientifically sound, ideally FDA-recognised consensus standards or methods FDA has accepted before.
Technological characteristics in 21 CFR 807.100 include the materials, design, energy source and other features of the device. A significant change in any of these is a “different technological characteristic” that must be justified with data.
How does FDA evaluate substantial equivalence step by step?
To decide 510(k) substantial equivalence, FDA reviewers follow the decision-making flowchart in the 2014 guidance. Your submission should answer each question in the same order, so the reviewer can tick through it without sending an Additional Information request.
Is the predicate legally marketed?
We confirm the predicate’s 510(k) number, De Novo number or pre-amendment status in the FDA 510(k) database and check that it has not been removed from the market.
Same intended use?
FDA compares the intended use and indications for use. A new indication that changes the purpose of the device, such as a new disease or patient population, can end the review here.
Same technological characteristics?
If the materials, design and energy source match, FDA moves to the data review. If not, it asks whether the differences raise new questions.
Different questions of safety and effectiveness?
A change such as moving from a mechanical to a software-driven function may raise a different question. If it does, the device is NSE.
Acceptable methods?
FDA checks whether there are scientific methods to evaluate the differences, such as recognised standards, and whether your protocols follow them.
Do the data demonstrate SE?
Finally, FDA reviews your bench, biocompatibility, software and, if needed, clinical results to confirm the device is as safe and effective as the predicate.
The full submission flow, from classification to eSTAR upload, is explained on our FDA 510(k) submission page.
Intended use vs technological characteristics
Every SE argument rests on two comparisons. Intended use is a yes-or-no gate; technological characteristics are where most of the evidence work happens.
| Aspect | Intended use | Technological characteristics |
|---|---|---|
| What it covers | The general purpose of the device: what it does and the condition it addresses, taken from labelling | Materials, design, dimensions, energy source, software, sterilisation and other features |
| Can it differ from the predicate? | No. It must be the same; minor differences in indications for use may be acceptable | Yes, if the differences raise no different questions of safety and effectiveness |
| How it is shown | Side-by-side indications for use statements and labelling | Comparison table plus performance testing |
| Typical risk | A new patient population, anatomical site or disease | A new material, energy type or software algorithm |
| Where it sits in the flowchart | Decision point 2 | Decision points 3 to 6 |
Our predicate device selection page explains how we compare indications for use statements word by word before a predicate is locked in.
How do you choose a predicate to prove substantial equivalence?
A legally marketed predicate is a device that was on the US market before 28 May 1976 (a pre-amendments device), a device reclassified from Class III to Class II or I, a device already found substantially equivalent through a 510(k), or a device granted marketing authorisation through De Novo. FDA’s page on how to find and effectively use predicate devices sets out these routes.
Signs of a weak predicate
- Different intended use or patient population
- Different technology that needs clinical data to bridge
- Cleared decades ago with methods no longer accepted
- Linked to a design-related recall or safety issue
- Combines with a second device as a split predicate
Signs of a strong predicate
- Same intended use and same product code
- Same principle of operation and energy source
- Cleared recently using current standards
- Public 510(k) summary with useful test details
- Clean recall and adverse event history
No split predicates: FDA’s 2014 guidance does not accept taking the intended use from one device and the technology from another. You may cite several predicates, but each must share your device’s intended use. A reference device can support test methods, but it is not a predicate.
In its September 2023 draft guidance on best practices for selecting a predicate device, FDA encourages manufacturers to choose predicates cleared with well-established methods, that meet or exceed the performance of current devices, and that have no unmitigated use-related or design-related safety issues or design-related recalls. That guidance was still in draft at the time of writing, but reviewers already apply its thinking.
What testing does FDA expect to see?
Most 510(k)s prove equivalence with non-clinical performance data. The tests you need depend on the differences between your device and the predicate and on the FDA-recognised consensus standards for your device type.
Performance testing
Mechanical, dimensional and functional tests run side by side against the predicate or against recognised standard limits.
Biocompatibility
Required for patient-contacting materials. See our biocompatibility testing service for the endpoint plan.
Electrical safety and EMC
For active devices, electrical safety testing and EMC reports show the device is as safe as the predicate.
Software and cybersecurity
Software documentation and, for connected devices, cybersecurity documentation that FDA now expects in every relevant 510(k).
Sterility and shelf life
Sterilisation validation and shelf-life studies for sterile or packaged devices.
Clinical data, when needed
Only when differences in technology or indications cannot be resolved by bench testing. We plan this early so it does not surprise your timeline.
Human factors work, such as a usability engineering file, and risk management records also help show that design differences do not raise new questions.
Example: a 510(k) comparison table
Every 510(k) includes a side-by-side comparison of the subject device and the predicate. Below is a simplified example for a non-invasive blood pressure monitor. Each “Different” row must point to the test report that closes the gap.
| Characteristic | Subject device | Predicate device | Comparison |
|---|---|---|---|
| Intended use | Non-invasive measurement of systolic and diastolic blood pressure and pulse rate in adults | Same | Same |
| Measurement principle | Oscillometric | Oscillometric | Same |
| Cuff material | Nylon with TPU bladder | Nylon with PVC bladder | Different: ISO 10993 testing |
| Power source | Rechargeable Li-ion battery | 4 x AA batteries | Different: IEC 60601-1 and battery testing |
| Connectivity | Bluetooth Low Energy | None | Different: wireless, software and cybersecurity documentation |
| Accuracy | Validated to ISO 81060-2 | Validated to ISO 81060-2 | Same |
None of the differences above changes the purpose of the device, and each has an accepted test method, so the reviewer can follow the argument to an SE decision. Our 510(k) cost calculator helps you estimate the testing budget behind a table like this.
What happens after FDA reviews your 510(k)?
All 510(k)s must now be submitted through FDA’s eSTAR template. Under the MDUFA performance goals, FDA aims to reach a decision within 90 FDA review days; the clock stops whenever FDA is waiting for your answers. Eligible device types can also go through FDA’s 510(k) Third Party Review program. You will receive one of three results.
SE order
FDA issues a substantially equivalent letter. Once your establishment registration and device listing are in place, and your quality system meets the FDA QMSR, you can market the device in the US.
Additional Information request
FDA puts the 510(k) on hold and lists deficiencies. You have 180 calendar days to respond, or the submission is considered withdrawn.
NSE decision
The device is placed in Class III by default. Options are a new 510(k), a De Novo request or a PMA.
Common reasons for a not substantially equivalent decision
- New intended use, such as a new disease or patient group
- Different technology that raises different questions of safety and effectiveness
- Missing or inadequate performance data for a known difference
- A predicate that is not legally marketed or is a split predicate
- No response, or an incomplete response, to the Additional Information request
How I3CGLOBAL builds your SE argument
Our 510(k) consultants treat SE as the backbone of the submission, not a table added at the end. This is how we work.
Confirm the pathway
We confirm the classification regulation, product code and whether your device is 510(k) exempt, needs a 510(k) or fits De Novo better.
Shortlist predicates
We search FDA databases, read 510(k) summaries and check recall and adverse event history for each candidate.
Gap analysis
We compare intended use and every technological characteristic, and flag each difference that needs evidence.
Test plan
We map each difference to an FDA-recognised standard and build a test plan with your labs, so no test is missed or repeated.
Write the SE section
We write the comparison table and discussion in eSTAR, linking every claim to a test report.
Handle FDA questions
We answer Additional Information requests and interactive review questions until the SE letter is issued.
Unsure about your predicate before testing starts? A Pre-Submission meeting with FDA can confirm the approach. See our 510(k) services and FDA 510(k) user fees for planning. For in vitro diagnostics, see our 510(k) IVD submission workflow, and if you are outside the US, you will also need a US FDA agent.
Get your 510(k) right the first time
Send us your device description and intended use. We will shortlist predicates, highlight the gaps and give you a fixed 510(k) proposal.
Frequently asked questions
What is substantial equivalence?
Substantial equivalence is the FDA finding that a new medical device is as safe and effective as a legally marketed predicate device. The device must have the same intended use as the predicate and either the same technological characteristics, or different characteristics that do not raise different questions of safety and effectiveness and are supported by data.
Does substantial equivalence mean the devices must be identical?
No. Substantial equivalence does not require identical devices. Differences in design, materials, energy source or software are allowed, provided they do not raise different questions of safety and effectiveness and your performance data show the new device is as safe and effective as the predicate.
Which regulation defines substantial equivalence?
Substantial equivalence is defined in section 513(i) of the Federal Food, Drug, and Cosmetic Act and applied through 21 CFR 807.100(b). FDA explains how it evaluates it in the 2014 guidance The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications.
What is a legally marketed predicate device?
A legally marketed predicate is a device that was on the US market before 28 May 1976 (a pre-amendments device), a device reclassified from Class III to Class II or I, a device already found substantially equivalent through the 510(k) process, or a device granted marketing authorisation through De Novo classification.
Can I use more than one predicate device?
Yes. You can use multiple predicates, but each one must have the same intended use as your device. FDA does not accept a split predicate, where intended use comes from one device and technology from another. A reference device can support scientific methods, but it cannot serve as the predicate.
What happens if FDA finds my device not substantially equivalent?
A not substantially equivalent (NSE) decision places the device in Class III by default, so it cannot be marketed through that 510(k). Your options are a new 510(k) with a better predicate or new data, a De Novo classification request for a low-to-moderate-risk novel device, or a Premarket Approval (PMA).
How long does FDA take to decide substantial equivalence?
Under the MDUFA performance goals, FDA aims to reach a 510(k) decision within 90 FDA review days. The clock stops while FDA waits for your answers to an Additional Information request, so total calendar time is usually longer.
Is clinical data needed to prove substantial equivalence?
Most 510(k) submissions prove substantial equivalence with bench, biocompatibility, electrical safety, software and other non-clinical testing. FDA may ask for clinical data when differences in technology or indications cannot be resolved by non-clinical testing alone.
