Checklist playbook · Design changes
Medical Device Design Change Checklist
Six stages, from change request to closed change, for devices and IVDs
Every check says who owns it, whether it's a requirement or good practice, and where it comes from, in current ISO 13485, QMSR, EU MDR and IVDR terms. Tailor it to your device and markets, and use it on each change, from a supplier's new resin to a software update.
Notes
Scale the effort to the change: a label typo and a new sensor shouldn't get the same paperwork, but both need a documented decision. Run the checklist once per change; ticks are saved in this browser only.[1]
Fit it to your device
Tailor the checklist
Stage 1
Request and classify
What exactly is changing, why, and how big is it?
-
Changes have to be identified before they can be reviewed; a vague request leads to a vague impact assessment.
Read more
- Good practice
- Describe the change in one place, list every affected item by number and revision, and attach the current and proposed versions.
- Rule of thumb
- If two reviewers would picture different changes from the request, rewrite it.
- Common pitfalls
- Requests that say “update drawing” without saying what changes; affected documents found only after approval.
- Evidence to show
- Change request with the affected-items list and before-and-after versions.
-
The source tells reviewers what the change must achieve and how to check it worked.
Read more
- Good practice
- Link the request to the record that triggered it, so the purpose and success measure are clear.
- Rule of thumb
- Every change should answer “what problem does this solve?”.
- Common pitfalls
- Changes with no stated purpose; CAPA actions implemented as changes without the link back.
- Evidence to show
- Change request linked to the CAPA, complaint, supplier notice or project.
-
Significance sets how much verification, validation and regulatory work the change needs.
Read more
- Good practice
- Use a short, defined scale and criteria so the same kind of change gets the same classification every time.
- Rule of thumb
- Classify by what could go wrong for the patient or user, not by how much paperwork you want.
- Common pitfalls
- Everything classified “minor”; significance judged by the person who wants the change.
- Evidence to show
- Significance assessment with the criteria used.
-
A “temporary” deviation renewed for a year is a design change nobody assessed.
Read more
- Good practice
- Use deviations or concessions for planned, time-limited departures with a clear end date, and convert anything permanent into a design change.
- Rule of thumb
- If a deviation has been extended twice, it’s a change.
- Common pitfalls
- Deviations used to avoid change control; no expiry dates; product built under expired deviations.
- Evidence to show
- Deviation or concession records with scope, duration and approval.
-
Impact is usually missed by the function that wasn’t asked.
Read more
- Good practice
- Name one owner and pick reviewers based on what the change touches, using a standard list of functions to consider.
- Rule of thumb
- Ask “who would be surprised by this change?” and invite them.
- Common pitfalls
- Regulatory or operations left out until the end; reviewers who sign without reading.
- Evidence to show
- Reviewer list and their recorded input.
-
A new supplier portal or test procedure can affect product as much as a drawing change.
Read more
- Good practice
- Evaluate process and system changes, such as new software tools, procedures or sites, for their effect on the quality system and on devices before rolling them out.
- Rule of thumb
- A change to how you make or check devices is a change to the devices’ risk.
- Common pitfalls
- New tools introduced without validation; procedure changes not linked to product impact.
- Evidence to show
- Process change records with impact evaluation.
Stage 2
Impact assessment
What does this change touch, inside and outside the company?
-
A change to an output can quietly break the trace to an input or a risk control.
Read more
- Good practice
- Trace the change through requirements, specifications, risk controls and verification, and update each affected link.
- Rule of thumb
- Follow the trace both ways: up to needs and down to tests.
- Common pitfalls
- Changing an output without checking which inputs and tests depend on it.
- Evidence to show
- Updated trace matrix or impact analysis.
-
Devices already in the field may need the change too, or may need a decision that they don’t.
Read more
- Good practice
- List affected parts and assemblies, quantities in process and finished goods, and units in the field, and decide what each needs.
- Rule of thumb
- For every changed part, ask: where else is it used?
- Common pitfalls
- Shared parts changed for one product and breaking another; field units forgotten.
- Evidence to show
- Where-used analysis and inventory and field assessment.
-
The most common audit finding on changes is a risk file that was never opened.
Read more
- Good practice
- Review hazards and risk controls affected by the change, update estimates, and check whether the change introduces new hazards.
- Rule of thumb
- Open the risk file for every design change, even if you close it again with “no impact” and a reason.
- Common pitfalls
- Risk files updated only for “major” changes; risk controls changed without re-verification.
- Evidence to show
- Risk assessment of the change and the updated risk file.
-
A moved button or reworded step can create a use error that wasn’t there before.
Read more
- Good practice
- Map the change to user tasks and use-related risks, and decide whether formative or summative testing is needed for the changed parts of the user interface.
- Rule of thumb
- If a user would notice the change, assess it for use errors.
- Common pitfalls
- Labeling and screen changes treated as cosmetic; no record of why testing wasn’t needed.
- Evidence to show
- Use-related risk review and any usability test results.
-
A new colorant, adhesive or cleaning agent can change the biological evaluation.
Read more
- Good practice
- Check whether the change affects materials, processing or the nature and duration of contact, and update the biological evaluation where it does.
- Rule of thumb
- If anything that touches the patient changes, the biological evaluation needs a look.
- Common pitfalls
- Supplier material changes accepted on a datasheet; process chemicals overlooked.
- Evidence to show
- Biological evaluation update or a documented rationale for no change.
-
Small changes to density, materials or packaging can invalidate sterilization or aging data.
Read more
- Good practice
- Assess the change against the sterilization validation, sterile barrier system and aging data, and plan any repeat work.
- Rule of thumb
- Density, materials and packaging are the three things that most often invalidate sterile data.
- Common pitfalls
- Load configuration changed without assessment; new pouch supplier treated as equivalent.
- Evidence to show
- Sterilization and packaging impact assessment.
-
Validated processes need a defined trigger for revalidation, and design changes are a common one.
Read more
- Good practice
- Check validated processes, equipment settings, fixtures and test methods against the change, and plan revalidation where needed.
- Rule of thumb
- If the process sees anything different, check whether its validation still covers it.
- Common pitfalls
- Parameter windows quietly widened; fixtures modified without re-qualification.
- Evidence to show
- Process impact assessment and revalidation plan.
-
Supplier changes are design changes too when they affect the part you buy.
Read more
- Good practice
- Update purchasing specifications and agreements, and evaluate supplier-initiated changes as you would your own.
- Rule of thumb
- Treat a supplier change notice as the start of a change request.
- Common pitfalls
- Supplier changes discovered at incoming inspection; agreements without change notification.
- Evidence to show
- Supplier change evaluation and updated purchasing information.
-
Translations and artwork are often the longest lead time in a change.
Read more
- Good practice
- List every label, IFU and translation affected, and plan artwork, translation and approval time.
- Rule of thumb
- Count the languages before you promise a date.
- Common pitfalls
- One language updated and the others forgotten; electronic labeling out of step with printed labeling.
- Evidence to show
- Labeling impact list and the updated artwork plan.
-
A change that creates a new version or model usually needs a new device identifier and database update.
Read more
- Good practice
- Use the triggers in the UDI rules for each market to decide whether a new identifier is needed, and record the decision.
- Rule of thumb
- New version or model, new identifier; when in doubt, check the rules rather than habit.
- Common pitfalls
- Changes shipped under the old identifier; database records not updated.
- Evidence to show
- UDI decision record.
-
Software changes ripple; the regression scope should follow the architecture, not the calendar.
Read more
- Good practice
- Identify affected software items and requirements, set the regression scope from the architecture, and check third-party components and security.
- Rule of thumb
- Small code changes can have large effects; let the architecture set the test scope.
- Common pitfalls
- Regression scope chosen by time available; dependency updates not assessed.
- Evidence to show
- Software change impact analysis and regression plan.
-
A change can quietly move the device outside what the clinical data shows.
Read more
- Good practice
- Check whether the changed device is still covered by the clinical or performance data, and update the evaluation where it isn’t.
- Rule of thumb
- If the claim or the population changes, the evidence probably needs to change too.
- Common pitfalls
- Clinical evaluations not reopened after design changes; equivalence assumed to survive the change.
- Evidence to show
- Clinical or performance evidence impact assessment.
-
A change that saves cents per unit can cost more in scrap and requalification.
Read more
- Good practice
- Estimate costs, lead times and inventory effects, and include them in the approval decision.
- Rule of thumb
- Count the scrap before you approve the savings.
- Common pitfalls
- Write-offs and requalification costs discovered after approval.
- Evidence to show
- Business impact summary.
Stage 3
Regulatory decision
Can this change be made under the current authorization, or does it need a submission or notification?
-
A new 510(k) is needed when a change could significantly affect safety or effectiveness, or is a major change in intended use.
Read more
- Good practice
- Work through FDA’s decision flow for each type of change, record each answer and the rationale, and compare against the device in the most recent clearance.
- Rule of thumb
- Compare to the last cleared device, not to last month’s version.
- Common pitfalls
- Decisions recorded as a one-line conclusion; cumulative changes never assessed.
- Evidence to show
- Documented 510(k) decision with the flowchart answers and rationale.
-
A change in complexity changes which laboratories can run the test.
Read more
- Good practice
- Check whether the change affects test complexity, and contact FDA if categorization may change.
- Rule of thumb
- If the steps a user performs change, check the categorization.
- Common pitfalls
- Categorization assumed to carry over.
- Evidence to show
- Categorization review record.
-
Software changes have their own decision flow in the US.
Read more
- Good practice
- Work through the software decision flow for each change, including cybersecurity patches, and record the outcome.
- Rule of thumb
- Assess each release, including small patches, and record the outcome.
- Common pitfalls
- Bundled releases assessed as one change when parts of them are significant.
- Evidence to show
- Software change regulatory assessment.
-
Only changes described in the authorized plan avoid a new submission.
Read more
- Good practice
- Follow the authorized plan’s protocols for development, validation and implementation exactly, and document conformance.
- Rule of thumb
- If the change isn’t in the plan, it isn’t covered by the plan.
- Common pitfalls
- Changes stretched beyond what the plan describes.
- Evidence to show
- PCCP conformance record.
-
A significant change ends the device’s right to rely on its legacy certificate.
Read more
- Good practice
- Assess the change against the guidance on significant changes for legacy devices before implementing it.
- Rule of thumb
- For a legacy device, every design change is a question about the certificate.
- Common pitfalls
- Changes made without checking legacy status.
- Evidence to show
- Significant-change assessment.
-
A change can move a device into a class that needs a notified body.
Read more
- Good practice
- Update the technical documentation and declaration of conformity, and confirm the classification hasn’t changed.
- Rule of thumb
- Recheck the classification rules for every change to intended purpose or technology.
- Common pitfalls
- A self-certified device drifting into a higher class.
- Evidence to show
- Updated technical documentation and declaration.
-
Some changes need the notified body’s approval, and its certificate may need a supplement.
Read more
- Good practice
- Check your notified body agreement and the conformity assessment annex for what must be notified, and notify planned substantial changes before implementing them.
- Rule of thumb
- If in doubt, ask the notified body before the change, not after.
- Common pitfalls
- Changes implemented before notified body approval; certificate scope not updated.
- Evidence to show
- Notification to the notified body and its response.
-
A change can take a device outside its exemption and into needing a 510(k).
Read more
- Good practice
- Check the changed device against the classification regulation and its limitations of exemption.
- Rule of thumb
- Exempt today doesn’t mean exempt after the change.
- Common pitfalls
- New technology or intended use added to an exempt device without checking.
- Evidence to show
- Exemption review record.
-
Changes that affect safety or effectiveness need FDA approval before they are made; others go in the periodic report.
Read more
- Good practice
- Decide the supplement type from the regulation and FDA’s guidance, and file before implementing when approval is needed.
- Rule of thumb
- For a PMA device, assume FDA needs to know until you’ve shown otherwise.
- Common pitfalls
- Changes implemented under internal control that needed a supplement.
- Evidence to show
- Supplement decision record and FDA correspondence.
-
The same change can be a note to file in one market and a submission in another.
Read more
- Good practice
- Assess each market’s rules separately, record the decision and rationale, and have regulatory approve it before implementation.
- Rule of thumb
- One change, one decision per market, all on record.
- Common pitfalls
- EU impact assessed only for the US decision; decisions made after implementation.
- Evidence to show
- Regulatory assessment for each market.
-
Many countries require notice of changes to registered devices, sometimes before implementation.
Read more
- Good practice
- Keep a register of registrations by country with their change notification rules.
- Rule of thumb
- If you don’t know what a country requires, ask your local representative before the change ships.
- Common pitfalls
- Distributors in smaller markets left unaware of changes.
- Evidence to show
- Country registration register with notification decisions.
Stage 4
Verify and validate
Have we shown the changed device still meets its requirements and its users’ needs?
-
Skipping a test is fine when the rationale is written down; skipping it silently isn’t.
Read more
- Good practice
- Plan the tests needed for the change based on risk and impact, and write down why other tests don’t need repeating.
- Rule of thumb
- Write the rationale for not testing as carefully as the test plan.
- Common pitfalls
- Full retesting by default, or none by default; rationale missing.
- Evidence to show
- Change verification and validation plan with rationale.
-
Regression testing shows the change didn’t break what already worked.
Read more
- Good practice
- Verify changed requirements and run regression tests on what could be affected, with predefined acceptance criteria.
- Rule of thumb
- Test the change, then test what the change could have broken.
- Common pitfalls
- Only the changed feature tested; failures dismissed as unrelated without investigation.
- Evidence to show
- Verification and regression test reports.
-
New crevices or materials can defeat a validated cleaning process.
Read more
- Good practice
- Revalidate cleaning and disinfection or sterilization when design features, materials or instructions change.
- Rule of thumb
- New crevices mean new cleaning validation.
- Common pitfalls
- Instruction changes made without revalidation.
- Evidence to show
- Reprocessing revalidation reports.
-
MR conditions depend on the exact materials and geometry.
Read more
- Good practice
- Re-evaluate MR safety for changes to materials, size or shape, and update labeling and implant cards.
- Rule of thumb
- Any change to an implant’s metal content or geometry is an MR question.
- Common pitfalls
- Labeling not updated after material changes.
- Evidence to show
- MR safety assessment and updated labeling.
-
A component swap that looks minor can fail an emissions test.
Read more
- Good practice
- Assess whether the change affects safety or EMC test results, and retest where it could.
- Rule of thumb
- Treat component swaps on power or radio circuits as test triggers.
- Common pitfalls
- Alternate components approved on datasheets alone.
- Evidence to show
- Electrical safety and EMC impact assessment and test reports.
-
Small reagent or algorithm changes can shift results across the measuring range.
Read more
- Good practice
- Re-verify the performance characteristics the change could affect, using the same methods as the original studies where possible.
- Rule of thumb
- Check results at the clinical decision points, not just the average.
- Common pitfalls
- Reagent supplier changes accepted without bridging studies.
- Evidence to show
- Bridging or re-verification study reports.
-
Shelf-life and in-use claims rest on the exact formulation and package tested.
Read more
- Good practice
- Check whether existing stability data still applies, and start new stability studies where it doesn’t.
- Rule of thumb
- If the formulation or package changes, the stability claim needs a fresh look.
- Common pitfalls
- Claims carried over without justification.
- Evidence to show
- Stability assessment and any new study plans.
-
If the change touches what users experience, validation is part of the change.
Read more
- Good practice
- Repeat validation activities, including usability and clinical evaluation, where the change could affect user needs or intended use.
- Rule of thumb
- Validation follows the user, not the drawing.
- Common pitfalls
- Validation skipped for labeling or software changes that affect how users act.
- Evidence to show
- Validation reports or a rationale for not repeating validation.
-
Product made on a changed process before revalidation is a nonconformity waiting to be found.
Read more
- Good practice
- Revalidate affected processes before product made on them is released.
- Rule of thumb
- Revalidate before release, not after.
- Common pitfalls
- Product shipped while revalidation was pending.
- Evidence to show
- Process revalidation reports.
-
Results can change risk estimates; record that they did, or that they didn’t.
Read more
- Good practice
- Update the risk file with verification and validation results, and confirm overall residual risk is still acceptable.
- Rule of thumb
- Close the loop: assessment before, results after.
- Common pitfalls
- Risk assessment at the start, never updated with results.
- Evidence to show
- Updated risk management file.
-
The review is where someone asks the question the change owner didn’t.
Read more
- Good practice
- Hold a review with the functions concerned to confirm the change is complete, verified and ready for approval.
- Rule of thumb
- A good review asks what the change might have broken elsewhere.
- Common pitfalls
- Reviews done by email with no record; same reviewers every time.
- Evidence to show
- Change review record.
Stage 5
Approve and implement
Is the change approved, and will it reach production, labeling and the field in a controlled way?
-
Implementing first and approving later is one of the most common change-control findings.
Read more
- Good practice
- Approve the change after verification, validation and any regulatory clearance or acceptance, by people with the authority to approve.
- Rule of thumb
- No approval, no implementation.
- Common pitfalls
- Implementation before approval; approvers without the needed expertise.
- Evidence to show
- Approval record with dates.
-
Without a clear cut-in, old and new versions mix on the line and in records.
Read more
- Good practice
- Agree the cut-in point with production and suppliers, and record which serial numbers or lots are affected.
- Rule of thumb
- Write effectivity so anyone can tell which version a unit is.
- Common pitfalls
- Mixed old and new parts; effectivity unclear in records.
- Evidence to show
- Effectivity record and cut-in confirmation.
-
The fastest way to undo a change is a superseded drawing still on the shop floor.
Read more
- Good practice
- Release updated documents and software, and remove obsolete versions from points of use.
- Rule of thumb
- Walk the floor after release and look for old versions.
- Common pitfalls
- Obsolete documents still in use; ERP bills of materials not updated.
- Evidence to show
- Release records and obsolete document removal.
-
Users need to know which version they have and what changed.
Read more
- Good practice
- Release software under configuration management with version, release notes, known anomalies and clear update instructions.
- Rule of thumb
- Users should always be able to tell which version they have.
- Common pitfalls
- Silent updates; release notes missing known anomalies.
- Evidence to show
- Release records and release notes.
-
An SBOM that doesn’t match the release can’t support vulnerability monitoring.
Read more
- Good practice
- Update the software bill of materials and threat model for every release.
- Rule of thumb
- If a component changed, the SBOM changed.
- Common pitfalls
- SBOMs frozen at the original submission.
- Evidence to show
- Updated SBOM and threat model.
-
Old stock needs a decision: use, rework, scrap or quarantine.
Read more
- Good practice
- Decide what happens to work in process, finished goods, supplier stock, field units and spares, and record the decision.
- Rule of thumb
- Every affected item needs a decision, even if the decision is “use as is”.
- Common pitfalls
- Old stock shipped by default; service spares forgotten.
- Evidence to show
- Disposition records.
-
Train on the released documents, before the first changed unit is built.
Read more
- Good practice
- Train the people affected on the released documents before the change takes effect, and record it.
- Rule of thumb
- Train before the first changed unit, not after the first mistake.
- Common pitfalls
- Training on drafts; temporary staff missed.
- Evidence to show
- Training records.
-
Label mix-ups often happen at changeovers.
Read more
- Good practice
- Remove old labels from the line, verify new labels against approved artwork, and reconcile label quantities.
- Rule of thumb
- Clear the line of old labels before the first new one is printed.
- Common pitfalls
- Old and new labels side by side at the line.
- Evidence to show
- Line clearance and label verification records.
-
Records should describe the device you now make, not the one you made last year.
Read more
- Good practice
- Update the medical device file, technical documentation and design and development file to describe the changed device.
- Rule of thumb
- Your files should describe what you make today.
- Common pitfalls
- Technical documentation updated months after the change.
- Evidence to show
- Updated file indexes with the change reference.
-
Database records have to match what is on the label.
Read more
- Good practice
- Update device records in each database once the change takes effect.
- Rule of thumb
- Update the database the same day the new label ships.
- Common pitfalls
- Database records showing the old version.
- Evidence to show
- Database update confirmations.
-
The published summary has to match the device on the market.
Read more
- Good practice
- Update the summary of safety and performance when the change affects it.
- Rule of thumb
- Keep the summary in step with the performance evaluation.
- Common pitfalls
- Summaries left out of date.
- Evidence to show
- Updated summary.
-
The published summary has to match the device on the market.
Read more
- Good practice
- Update the summary when the change affects safety or clinical performance information, and send it to the notified body for validation.
- Rule of thumb
- If the change touches the summary’s content, the summary changes too.
- Common pitfalls
- Summaries left out of date in EUDAMED.
- Evidence to show
- Updated and validated summary.
-
The first lot after a change is where implementation mistakes show up.
Read more
- Good practice
- Inspect the first units or lots made after the change and confirm they meet the updated criteria.
- Rule of thumb
- Check the first lot as carefully as a new product.
- Common pitfalls
- First lots shipped without extra inspection.
- Evidence to show
- First article or first lot inspection records.
-
A change that fixes a safety problem in new units raises the question of units already in use.
Read more
- Good practice
- Ask whether the change fixes a risk in devices already in use, and if so evaluate it under the field action and reporting rules.
- Rule of thumb
- If you wouldn’t ship the old design today, ask what to do about the ones already shipped.
- Common pitfalls
- Safety-related changes made quietly in production only; reporting not considered.
- Evidence to show
- Field action assessment and any reports.
-
Users notice changes; telling them first avoids complaints and confusion.
Read more
- Good practice
- Tell customers, distributors and service teams about changes that affect how the device looks, works or is serviced.
- Rule of thumb
- If users will notice the change, tell them before they do.
- Common pitfalls
- Service teams discovering changes from returned units.
- Evidence to show
- Customer or distributor notifications.
Stage 6
Close and follow up
Did the change work, and is the record complete?
-
Changes made to fix a problem need evidence that the problem is fixed; for other changes it’s good practice to confirm the benefit.
Read more
- Good practice
- Check the change achieved its purpose using the measure defined at the start.
- Rule of thumb
- Define the success measure when you open the change, not when you close it.
- Common pitfalls
- Closing CAPA-driven changes without effectiveness evidence.
- Evidence to show
- Effectiveness check record.
-
Some effects only show up in field data months later.
Read more
- Good practice
- Watch complaints, service data and trends for effects of the change, and compare with the period before.
- Rule of thumb
- Look at field data three and six months after a significant change.
- Common pitfalls
- No follow-up after implementation.
- Evidence to show
- Post-market data review for the change.
-
An auditor should be able to see what changed, who reviewed it and what was done.
Read more
- Good practice
- Close the change only when every action is complete and the evidence is linked.
- Rule of thumb
- If an auditor picked this change, could they follow it without asking anyone?
- Common pitfalls
- Changes closed with open actions; evidence held in email.
- Evidence to show
- Closed change record with linked evidence.
-
Several small changes can add up to one that needs a submission.
Read more
- Good practice
- Review accumulated changes against the last authorized design at a set interval.
- Rule of thumb
- Small changes add up; check the total, not just the latest.
- Common pitfalls
- Many minor changes, never assessed together.
- Evidence to show
- Cumulative change review record.
-
A growing change backlog is a signal about the design, the suppliers or the process.
Read more
- Good practice
- Report change volumes, cycle times and backlog to management review, and act on trends.
- Rule of thumb
- A rising change rate after launch usually points to an upstream problem.
- Common pitfalls
- Backlog invisible to management.
- Evidence to show
- Management review records with change metrics.
Related reading
Changes before launch follow the same discipline. For the full path from concept to first shipment, see our medical device NPI checklist.
References
- ISO 13485:2016, Medical devices — Quality management systems — Requirements for regulatory purposes. Incorporated by reference into 21 CFR Part 820 since 2 February 2026. www.iso.org/standard/59752.html
- ISO 14971:2019, Medical devices — Application of risk management to medical devices. www.iso.org/standard/72704.html
- IEC 62366-1:2015+AMD1:2020, Medical devices — Part 1: Application of usability engineering to medical devices. webstore.iec.ch/en/publication/67220
- ISO 10993-1:2025, Biological evaluation of medical devices — Part 1: Requirements and general principles for the evaluation of biological safety within a risk management process, November 2025. www.iso.org/standard/84512.html
- ISO 11607-1:2019+Amd 1:2023, Packaging for terminally sterilized medical devices — Part 1: Requirements for materials, sterile barrier systems and packaging systems. www.iso.org/standard/70799.html
- 21 CFR Part 820, Quality Management System Regulation: § 820.35 (control of records) and § 820.45 (device labeling and packaging controls). www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-820
- U.S. FDA, Unique Device Identification System (UDI System): 21 CFR Part 830, including § 830.50 (changes that require a new device identifier), and the GUDID. www.fda.gov/medical-devices/device-advice-comprehensive-regulatory-assistance/unique-device-identification-system-udi-system
- Regulation (EU) 2017/745 on medical devices (EU MDR), including Articles 10, 19, 29, 32, 51–52, 61, 89 and 120, and Annexes VI, IX and X. eur-lex.europa.eu/eli/reg/2017/745/oj
- IEC 62304:2006+AMD1:2015, Medical device software — Software life cycle processes. webstore.iec.ch/en/publication/22790
- 21 CFR Part 807, including § 807.81(a)(3) (when a premarket notification is required for a change), and the limitations of exemption in 21 CFR Parts 862–892. www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-807
- U.S. FDA, Deciding When to Submit a 510(k) for a Change to an Existing Device, guidance for industry and FDA staff. www.fda.gov/regulatory-information/search-fda-guidance-documents/deciding-when-submit-510k-change-existing-device
- U.S. FDA, CLIA Categorizations. www.fda.gov/medical-devices/ivd-regulatory-assistance/clia-categorizations
- U.S. FDA, Deciding When to Submit a 510(k) for a Software Change to an Existing Device, guidance for industry and FDA staff. www.fda.gov/regulatory-information/search-fda-guidance-documents/deciding-when-submit-510k-software-change-existing-device
- U.S. FDA, Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions, final guidance, December 2024, updated August 2025 (FD&C Act section 515C). www.fda.gov/regulatory-information/search-fda-guidance-documents/marketing-submission-recommendations-predetermined-change-control-plan-artificial-intelligence
- MDCG 2020-3 Rev.1, Guidance on significant changes regarding the transitional provision under Article 120 of the MDR with regard to devices covered by certificates according to MDD or AIMDD, May 2023. health.ec.europa.eu/system/files/2023-09/mdcg_2020-3_en_1.pdf
- 21 CFR Part 814, Premarket approval of medical devices, including § 814.39 (PMA supplements) and § 814.84 (reports). www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-814
- U.S. FDA, Reprocessing Medical Devices in Health Care Settings: Validation Methods and Labeling, guidance for industry and FDA staff. www.fda.gov/regulatory-information/search-fda-guidance-documents/reprocessing-medical-devices-health-care-settings-validation-methods-and-labeling
- U.S. FDA, Testing and Labeling Medical Devices for Safety in the Magnetic Resonance (MR) Environment, guidance for industry and FDA staff. www.fda.gov/regulatory-information/search-fda-guidance-documents/testing-and-labeling-medical-devices-safety-magnetic-resonance-mr-environment
- IEC 60601-1:2005+AMD1:2012+AMD2:2020, Medical electrical equipment — Part 1: General requirements for basic safety and essential performance, and IEC 60601-1-2:2014+AMD1:2020 (electromagnetic disturbances). webstore.iec.ch/en/publication/67497
- Regulation (EU) 2017/746 on in vitro diagnostic medical devices (EU IVDR), including Articles 10, 17, 26, 29, 47–48, 56, 84 and 110, and Annexes VI and IX–XI. eur-lex.europa.eu/eli/reg/2017/746/oj
- U.S. FDA, Cybersecurity in Medical Devices: Quality Management System Considerations and Content of Premarket Submissions, final guidance, February 2026 (section 524B of the FD&C Act, cyber devices). www.fda.gov/regulatory-information/search-fda-guidance-documents/cybersecurity-medical-devices-quality-management-system-considerations-and-content-premarket
- 21 CFR Part 806, Medical devices; reports of corrections and removals. www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-806
This checklist is general guidance, not legal or regulatory advice. Adapt it to your own procedures, device type and markets. Requirements are paraphrased; check the current text of each standard, regulation and guidance before relying on it.