Claude

IntuitionLabs is now a member of the Claude Partner Network – AI training and upskilling with Claude for pharma and biotech. Book a call.

IntuitionLabs
Back to Articles
IntuitionLabs

fda human factors submission requirements · human factors 510(k)

FDA Human Factors Submission Requirements: 2026 eSTAR Map

November 21, 2026
30 min read

FDA human factors submission requirements for 2026: a source-linked map from the May 29 final guidance to eSTAR, with Category 1, 2 and 3 decision logic, an evidence crosswalk and worked examples.

FDA Human Factors Submission Requirements: 2026 eSTAR Map
Summary
  1. 01FDA issued its final human factors content guidance on May 29, 2026, and incorporated it into the eSTAR templates on June 1, 2026. The guidance’s implementation date was August 1, 2026.
  2. 02FDA recommends one human factors category per submission. Category 1 calls for a conclusion and high-level summary, Category 2 calls for a rationale and report sections 1 to 4, and Category 3 calls for a full usability engineering report with validation.
  3. 03Decision Point D weighs interface history, complexity and the adequacy of existing risk controls. For impacted critical tasks with unchanged controls, a rationale needs objective evidence that those controls remain effective.
  4. 04The guidance covers 510(k), De Novo, PMA and HDE submissions. The eSTAR program requires the template for 510(k) and De Novo submissions unless exempted, lists PMA use as voluntary and leaves HDE status unspecified.
  5. 05In eSTAR, use Guide Me or select the category directly, then use the fillable boxes to tell reviewers where to find the supporting evidence.
01

Executive Summary

The U.S. Food and Drug Administration (FDA) issued the final guidance "Content of Human Factors Information in Medical Device Marketing Submissions" on May 29, 2026 ([1]), replacing a revised draft dated December 9, 2022. It sets a risk-based framework for the human factors (HF) content of 510(k), De Novo, premarket approval (PMA) and humanitarian device exemption (HDE) submissions ([2]). The document is nonbinding: FDA describes it as "not binding on FDA or the public" ([3]), and "should" means suggested or recommended, not required. FDA's June 1, 2026 eSTAR update carries the guidance content into the templates ([4]), and FDA did not generally expect the newly recommended information in submissions received before August 1, 2026 ([5]).

The operational core is one category per submission ([6]). Category 1 applies when the user interface, users, uses, environments, training and labeling are unchanged. Category 2 applies when no critical tasks exist, or none are new or impacted, or when a rationale replaces validation data ([7]). Category 3 requires a comprehensive HFE/UE (human factors engineering / usability engineering) report with validation testing ([8]). Decision Point D, the fourth decision point, asks whether validation data should be submitted at all ([9]). FDA staff said this is where "there is the most discretion" ([10]).

For eSTAR, FDA says the template offers a "Guide Me" route or direct category selection ([11]), and that fillable boxes should tell reviewers where evidence sits ([12]). eSTAR use is mandatory for 510(k) and De Novo submissions and voluntary for PMAs ([13]) ([14]). FDA's page lists nIVD and IVD eSTAR Version 7.1 as current. Because FDA verifies dropdown answers and may place a submission on technical screening hold for up to 180 days ([15]), category selection is a quality-controlled step, not a formality.

The practical decision for regulatory and human-factors teams is whether an existing usability-engineering file is submission-ready. This report maps each category to report sections, eSTAR prompts and internal artifacts, separates recommendations from legal duties, and provides a planning calculator (an illustration, not an FDA benchmark). Device-specific questions belong in a Pre-Submission. Several items could not be verified and are listed in the Conclusion.

60 days

Minimum transition time FDA said it and industry might need to put the policies into practice

15 participants

Recommended minimum per user population in human factors validation testing

180 days

Maximum time an eSTAR submission may be on technical screening hold

1GB

Size limit per eSTAR attachment

02

Introduction and Background

Human factors content has long been a common topic of premarket dialogue, but the question "what do we actually put in the submission?" lacked a published structure. FDA's final guidance fills that gap by describing the level of documentation, not the development process: the guidance states it is not intended to inform manufacturers how to perform a human factors evaluation ([16]), and it complements the existing guidance "Applying Human Factors and Usability Engineering to Medical Devices" ([17]). In the town hall, FDA put it directly: the category determines the level of documentation, "not what human factors processes should be applied" ([18]).

This report is written for regulatory affairs leads, human-factors engineers, product owners and software teams preparing a 510(k), De Novo, PMA or HDE after the August 1, 2026 implementation point. As of November 21, 2026, the key dates are: guidance issued May 29, 2026 (Federal Register notice 91 FR 32061 ([19])); eSTAR update June 1, 2026 ([20]); town hall July 22, 2026 ([21]); and the companion 2016 guidance reissued with a date of August 3, 2026 ([22]).

The report's distinctive contribution is a traceability map from guidance sections to eSTAR prompts and internal evidence. It is deliberately not a primer on usability-engineering principles. Throughout, statements are tagged as FDA recommendation, legal requirement, or this report's planning assumption. Nothing below is regulatory advice. Teams should consult qualified regulatory counsel and their FDA review division for device-specific decisions.

F.01
MDUFA V review decision targets
03

Key Changes: From the May 29 Guidance to the June 1 eSTAR

Change 1: A risk-based framework with a fourth decision point

The final guidance states its purpose as providing a risk-based framework ([23]). FDA's Federal Register notice says it added risk-based factors for determining the HF Submission Category, new examples and appendices, and scope clarifications after comments ([24]) ([25]). A law-firm commentary identifies Decision Point D as the notable substantive change ([26]), and notes the draft had appeared to expand the number of submissions requiring HF data ([27]). A practitioner summary says that, after the change, not all devices with new or impacted critical tasks automatically require added validation testing ([28]), a view echoed by another certification body's commentary ([29]). The same law-firm post cautions that in its author's experience CDRH review teams expect to review HF study results ([30]), so the framework should not be read as a general relaxation.

Change 2: Examples and report outlines

Practitioners report the examples now take three times the page count of the draft's ([31]). Trade press describes three appendices, one per category ([32]), and scenarios including a duodenoscope and a stereotaxic navigation system ([33]). In the guidance itself, Appendices A, B and C give report outlines for the three categories (for example Table A-1 for Category 1 ([34])).

Change 3: Transition treatment

FDA recognized that the Agency and industry might need at least 60 days to operationalize the policies ([35]), and said it did not generally expect the newly recommended information in pending submissions or those received before August 1, 2026, while still intending to review such information if submitted at any time ([36]). The Federal Register notice carries the same language ([37]). This differs from the 2022 draft's notice, which referred to up to 60 days after publication of the final guidance ([38]). At the town hall, FDA confirmed a 60-day transition with implementation on August 1 ([39]), and added that earlier submissions are not expected to use the new eSTAR version ([40]). One consultancy summarizes the result as FDA having implemented the new recommendations on August 1, 2026 ([41]). Trade press also dated FDA's publication as May 28, while FDA's own documents say May 29; this report uses FDA's date ([42]).

Change 4: eSTAR incorporation

FDA's eSTAR page carries an update dated June 1, 2026, stating the nIVD and IVD eSTARs now include the human factors content guidance ([4]). FDA describes a major version update as significant template revisions ([43]). The page currently lists nIVD and IVD eSTAR Version 7.1 ([44]) ([45]). A trade-press report, based on the template at release, describes Version 7.0 released June 1, 2026 with a transition date of August 3, 2026 ([46]) ([47]). The two dates describe different events: August 1, 2026 is when the content guidance takes effect, while the trade report gives August 3, 2026 as the eSTAR template transition date. The sources also differ on version number (7.0 versus 7.1); FDA's page carries no release notes for 7.1, so teams should open the current template and read its version history before filling it in. The same trade report says the template adds a Human Factors and Usability section within Performance Testing ([48]).

Change 5: Companion guidance reissued

The 2016 guidance file now reads "Document issued on August 3, 2026" and "originally issued on February 3, 2016" ([49]). Its revised text directs readers to the content guidance for what belongs in a submission ([50]). FDA staff also said they intended minor updates to the older guidance for consistency ([51]). The 2026 reissue is not compared here against the 2016 original, so the participant-number figures cited below are from the current text only.

F.02
FDA human factors guidance and eSTAR milestones
  1. May 29Final guidance issued

    FDA issued the final guidance on human factors information in medical device marketing submissions.

  2. June 1eSTAR templates updated

    The nIVD and IVD eSTAR templates incorporated the human factors content guidance.

  3. July 22FDA town hall

    FDA held a town hall on the final human factors content guidance.

  4. August 1Content guidance implemented

    The transition covered pending submissions and those received before this date. FDA would review the new information if submitted.

  5. August 3Companion guidance reissued

    The companion human factors and usability engineering guidance carries a new issue date of August 3, 2026.

“

even a "small" workflow edit can change the answer at Decision Point C.

05

Choosing Human Factors Submission Category 1, 2 or 3

The decision flow

FDA recommends one category per marketing submission ([6]), and, where a device has several modifications, that submitters consider them collectively ([79]). The flowchart has four decision points. Table 1 below restates them with the evidence each outcome implies.

T.01
Decision pointQuestion (paraphrased)OutcomeEvidence implied
AIs this a modification of an existing authorized device (510(k), PMA, HDE or De Novo)? A new device can answer yes if it leverages the same or a similar user interface of the submitter's own marketed device ([80]).Does not itself yield a category ([81]).Establishes the comparison baseline for later analyses.
BHave the user interface, intended users, intended uses, use environments, training or labeling changed ([82])?"No" gives Category 1 ([83]).Conclusion and high-level summary, with a statement that modifications do not affect HF considerations ([84]).
CDoes the use-related risk analysis (URRA) show critical tasks (new device) or new or impacted critical tasks (modified device)?"No" gives Category 2 ([7]).Rationale for why the answer is no ([85]); report sections 1 to 4 ([86]).
DGiven user interface history of use ([87]), complexity ([88]) and adequacy of existing risk controls ([89]), should validation data be submitted ([9])?"No" gives Category 2 with a rationale; "Yes" gives Category 3 ([90]).Detailed rationale ([91]) or comprehensive HFE/UE report with validation ([8]).

Table 1 shows why category selection is evidence-driven. Decision Point A is a framing step, not a verdict. Decision Point B is a change screen: any change to the six listed elements moves the file past Category 1. Because FDA counts a change in the number of steps in a critical task as potentially impacting that task ([92]), even a "small" workflow edit can change the answer at Decision Point C. FDA also recommends running the URRA on the final finished device, not only on the change, so upstream and downstream tasks are not missed ([93]).

Where discretion sits

Decision Point D is where FDA said sponsors have the most discretion ([10]). FDA declined to define a complex interface, calling complexity resistant to binary categorization ([94]), and described the considerations as a totality-of-evidence assessment ([95]). Validation data are likely needed for complex interfaces or device types historically associated with use errors, with infusion pumps given as an example ([96]). FDA may also require data if there is a significant difference from similar devices ([97]) or an increase in severity of potential harm from use error ([98]). For impacted critical tasks with unchanged controls, a rationale needs objective evidence that the controls remain effective ([99]). One consultancy reads the final text as meaning a "no" at Decision Point D no longer necessarily points toward Category 3 ([100]). FDA told submitters unsure whether a rationale will suffice to seek Pre-Submission feedback ([101]).

06

Evidence Inventory and eSTAR Crosswalk

The four building blocks

  • Use-related risk analysis (URRA): systematic use of information to identify use-related hazards and estimate use-related risk ([102]). FDA's presenters said it should include all user tasks ([103]) and the guidance calls it a living document ([104]). The URRA document itself belongs in the Category 3 report ([105]).
  • Critical tasks: tasks that, if performed incorrectly or not at all, would or could cause serious harm to the patient or user ([106]). The report should categorize them by severity of potential harm ([107]).
  • Comparative use-related risk analysis: for modified devices, a comparison of the modified and existing URRAs ([108]), covering at minimum all tasks in the use scenarios relevant to the modifications ([109]). FDA said it identifies change in risk and can support categorization, particularly Category 3 ([110]).
  • HF validation testing and the HFE/UE report: validation is one portion of design validation ([111]); FDA notes "summative testing" is sometimes used for it, with differing definitions ([112]). HFE and UE are treated as interchangeable terms ([113]).

Report sections by category

Table 2 below crosswalks the report sections to the categories and to the eSTAR prompts that FDA's slides describe. The eSTAR column is built from FDA's slides, transcript and a trade report; the template PDF itself was not reviewed, so on-screen field names should be confirmed in the current template.

T.03
HFE/UE report sectionCat. 1Cat. 2Cat. 3Typical internal source documenteSTAR note
1. Conclusion and high-level summary, including residual risk discussion ([114]) ([115])Yes ([116])YesYesDesign validation summary; risk management reportCategory dropdown or "Guide Me" ([11]); Category 1 is conclusion and summary only ([117])
2. Intended users, uses, environments, training ([118])Summary level onlyYesYesIntended-use and user-profile documentsCross-reference allowed where already in the device description ([119])
3. Device user interface description, with comparison to the existing device for modified devices ([120])Summary level onlyYesYesUI specification, labeling, screenshotsFillable boxes should point reviewers to where information sits ([12])
4. Known use problems ([121])Summary level onlyYesYesPost-market surveillance review; predicate and prior-model literatureState explicitly if none are known
5. Preliminary HFE/UE analyses and evaluations ([122]) ([123])NoNoYesFormative evaluation reportsDocumented locations in eSTAR
6. URRA, with comparative URRA for modified devices ([124])NoUnderlying analysis supports the Decision Point C answerYesURRA / hazard analysis traceability matrixLiving document
7. Critical tasks ([125])NoBasis for "no critical tasks" claimYesCritical task list with severity categoriesSeverity scale described in the report
8. Validation testing of the final design ([126])NoNo (rationale instead)YesValidation protocol, scripts, forms, resultsFull protocol and sample scripts appended ([127])

Table 2 should be read as a minimum map, not a checklist of everything FDA might ask. For Category 2, the content is a rationale rather than a report, and FDA recommends a "detailed rationale for why human factors validation should not be" submitted. For Category 1, only a conclusion and high-level summary are recommended. Previously reviewed HFE/UE evaluations should not be resubmitted; a cross-reference may help ([128]). The report summarizes evaluations and does not typically include all raw data from validation testing ([129]).

Validation testing expectations

The final content guidance defers testing method to the companion guidance. Section 8 of the report asks for a rationale for test type (simulated use, actual use or clinical study) ([130]). The current text of the companion guidance states that validation is generally conducted under simulated use ([131]), that participants represent intended actual users ([132]), that all critical tasks are performed ([133]), and that the testing should include at least 15 participants from each user population ([134]). Pediatric and adult users of one device are treated as distinct populations ([135]). "Think aloud" is acceptable in formative evaluation but not in validation testing ([136]). Manufacturer employees should generally not serve as participants ([137]), participants for U.S. users should reside in the U.S. ([138]), training given should approximate real-world training ([139]), and testing should not follow training immediately because retention decays ([140]). Critical task performance is measured by observation ([141]), while knowledge tasks can be assessed by questioning participants ([142]).

For modified devices, FDA says the validation test may be limited to aspects of users' interactions affected by the modification ([143]), consistent with the companion guidance that testing can focus on modified aspects of use ([144]). The need for additional validation depends on the risk analysis of the modification ([145]).

F.03
Category 2 and Category 3 submission evidence
Category 2Rationale
  • Use Category 2 when there are no critical tasks, none are new or impacted, or Decision Point D supports submitting a rationale in place of validation data.
  • Include a rationale with report sections 1 to 4.
  • For impacted critical tasks with unchanged controls, support the rationale with objective evidence that the controls remain effective.
Category 3Validation data
  • Submit validation data when critical tasks are present or impacted and Decision Point D leads to yes.
  • Provide a comprehensive human factors engineering / usability engineering (HFE/UE) report with validation testing.
  • Include a use-related risk analysis, a critical task list and a validation report.
07

Implementation Considerations and Process Changes

Gap-assessment workflow and document controls

A practical workflow for an existing usability-engineering file has six steps. Each step produces a controlled record; the numbering below is this report's suggestion, not an FDA sequence.

  • Step 1, baseline the device: identify the authorized device, or the same-submitter device whose user interface is leveraged. Record the submission number and report section for any earlier HFE/UE evaluation you intend to cross-reference.
  • Step 2, run the change screen: list every change to the user interface, users, uses, environments, training and labeling. Consider changes collectively.
  • Step 3, refresh the URRA on the finished device: update the URRA across all tasks and re-categorize critical tasks by severity. Count step additions or removals as potential impact.
  • Step 4, assess Decision Point D: document history of use, complexity and control adequacy, and the objective evidence for any rationale.
  • Step 5, assemble by category: populate report sections per Table 2 and decide where each lives in the submission; FDA wants locations documented in eSTAR.
  • Step 6, release gate: a regulatory lead signs that the eSTAR category selection, attached documents and report conclusion agree. This matters because FDA intends to verify the accuracy of responses ([146]) and an eSTAR can be placed on technical screening hold for up to 180 days ([15]).

Document-control points: keep one controlled copy of the HF information regardless of whether it is submitted; version the URRA as a living document; retain the evidence behind each Category 2 rationale; and keep attachment names within the 124-character limit ([147]), with attachments under 1GB each and 4GB per eSTAR ([148]). FDA does not expect a properly prepared eSTAR to go through Refuse to Accept review, but technical screening still applies ([149]). If a team prepared a submission on a prior eSTAR major version, FDA says previous versions may continue where major changes do not affect the device ([150]); given that the human factors section is new, check this against the specific template.

An evidence-gap tracker needs only six columns: evidence item, category it supports, guidance section, owner (role), status (not started, drafted, reviewed, filed) and eSTAR location. Typical gaps found when a legacy file is mapped to Table 2 are a URRA that covers only the changed function, a Section 4 known-use-problem summary that was never written, and a rationale with no objective evidence.

Worked examples

FDA's own examples anchor the three scenarios requested by software teams, device engineers and new-device sponsors.

  • Software user interface change. In FDA's anesthesia gas machine example, a font size increase from 12 to 14 with a larger screen leaves layout and format unchanged and no critical task impacted, so it is Category 2 with a rationale that may rest on formative data ([151]). An algorithm-only change to a lesion-detection device with no user interface change is Category 1 ([152]). Contrast an augmented reality interface added to a stereotaxic navigation system, which FDA treats as more complex than existing 2D interfaces, so validation data are needed ([153]).
  • Physical control change. In the same gas machine, replacing a physical knob with a digital slider changed the interaction ([154]); FDA judged the interface complex enough that validation data should be submitted, making it Category 3 ([155]). Adding a new user group (for example pediatric users) triggers Category 3 even with no interface change ([156]).
  • New device. A blood lancet with critical tasks but low-complexity, long-established interface components is Category 2 ([157]). A guided ultrasound whose intended users include professionals unfamiliar with the technique is Category 3 ([158]). A new stent delivered by an unchanged balloon catheter system is Category 1 because the delivery system is the same as approved ([159]).

(Hypothetical Example) A team plans a mobile app update that reorders the screens of a dose-confirmation workflow, adds a pediatric caregiver user group, and leaves hardware unchanged. Decision Point B is "yes" (interface and user group). The URRA likely shows impacted critical tasks (dose confirmation), and a new user group is a pattern FDA illustrates as triggering validation, so the planning assumption would be Category 3 with a validation cohort for each distinct user population. This illustration is the report's, not FDA's.

Review gates and questions for FDA

FDA says device-specific questions go through the Pre-Submission pathway ([160]). A Pre-Submission is not a pre-review of a submission or of data ([161]). The Q-Submission guidance (current copy dated May 29, 2025 ([162])) includes sample HF questions: whether a validation protocol is adequate ([163]), whether the participant recruitment plan is adequate ([164]), and whether the URRA plan is adequate ([165]). Written feedback is due 70 calendar days after receipt in the MDUFA V commitment ([166]), or 70 days or 5 days before the meeting, whichever is sooner ([167]). A consultancy recommends using the program when there is uncertainty about category selection or testing expectations ([168]). Good questions to pose: whether FDA agrees with the proposed category, whether a Category 2 rationale is sufficient, which tasks FDA considers critical, and which participant populations should be separate.

How practitioners read the framework

Independent commentary is largely consistent with FDA's text, with differences of emphasis. These are secondary sources and should not replace the guidance.

  • Decision Point D: consultancies call it the most impactful structural change ([169]) and say the rationale should directly address the applicable decision point ([170]). Another describes category assignment as still requiring sponsor judgment ([171]).
  • Category 2 scope: a human-factors consultancy notes some modified devices may fall into Category 2, which does not require validation testing ([172]). A caution from another consultancy: choosing not to submit validation data does not make the underlying work optional ([173]).
  • Comparative analysis: commentators say comparative analyses may help support reliance on prior information ([174]) and that FDA gave added instruction on the comparative URRA for modified devices ([175]). One goes further and calls a Table 3 style comparative URRA the new norm for any modification ([176]); the guidance itself recommends it for modified devices but frames it as a recommendation.
  • The URRA as backbone: several sources treat the URRA as the backbone of the submission ([177]) or the engine of the process ([178]), and say the URRA and critical-task rationale become the load-bearing documents ([179]). FDA's own advice is to assess the URRA on the whole finished device, echoed in practitioner summaries ([180]).
  • Justification and evidence: a regulatory consultancy says a rationale should include objective evidence that risk controls remain effective ([181]), tracking the guidance text, and another says a clear rationale supported by a URRA is required ([182]).
  • Process advice: commentators recommend planning human factors work earlier ([183]), keeping one controlled record early in submission planning ([184]), and recording the decision points for every software release ([185]). The last of these matches the document-control step proposed in this report.
  • eSTAR: one consultancy says the templates now require a category selection ([186]) and another calls the category a mandatory eSTAR field ([187]). FDA's own materials describe a "Guide Me" or direct-selection route; the template should be checked directly to see which fields block submission.
  • Pre-Submission: commentary points submitters toward the Pre-Submission program rather than guessing ([188]), consistent with FDA.

A cautionary note on dates: one practitioner advises planning as though August 1, 2026 is the effective date ([189]), and another describes the transition as covering submissions already under review or received before that date ([190]). Both match the guidance's wording ([5]).

Standards and non-U.S. context

FDA's content guidance aligns its user interface definition with ANSI/AAMI/IEC 62366-1 ([191]), and the FDA recognition database lists the IEC 62366-1 Edition 1.1 consolidated version ([192]). The IEC store describes the standard's process as evaluating the usability of a medical device as it relates to safety ([193]), notes that the consolidated edition combines the 2015 first edition and its 2020 amendment ([194]), and states that it does not assess or mitigate risks associated with abnormal use ([195]). A 2025 edition of the companion AAMI human factors standard is described by ANSI as covering general principles and attribute-specific principles ([196]). Teams should confirm in FDA's database which editions are currently recognized, because the guidance points readers there for current editions.

Regulators elsewhere express related expectations in different forms, which matters for global filers reusing a usability-engineering file. EU Medical Device Regulation Annex I sets duties for reducing risks related to use error ([197]) and refers to design for lay, professional, disabled or other users ([198]). The UK MHRA's human factors guidance states it does not represent a compliance requirement ([199]) and asks summative studies to include representatives from all identified user groups ([200]). Australia's TGA guidance expects human factors testing with a representative sample of intended users where applicable ([201]). None of these replaces the FDA category decision, and no equivalence of category logic across jurisdictions is claimed here.

F.04
Suggested gap assessment for an existing usability-engineering file
01Establish the device baseline

Identify the authorized device or the same submitter’s device with the interface being reused. Record the submission number and report section for earlier evaluations you plan to cite.

02Screen for changes

List changes to the interface, users, uses, environments, training and labeling. Assess the changes together.

03Update the risk analysis

Refresh the URRA across all tasks on the finished device. Reclassify critical tasks by severity and flag steps added or removed as potential impacts.

04Assess Decision Point D

Document the history of use, interface complexity, adequacy of risk controls and objective evidence for any rationale.

05Assemble the submission

Fill in the sections for the selected category and use eSTAR to document where each appears in the submission.

06Confirm agreement before release

Have a regulatory lead sign off that the selected eSTAR category, attachments and report conclusion agree.

08

Data Analysis and Evidence

Scope and eSTAR status by submission type

Table 3 below consolidates the scope facts.

T.02
Submission typeIn guidance scopeeSTAR status (FDA program page)Notes
510(k)YesMandatory since October 1, 2023Applies to modifications of 510(k)-exempt devices too
De NovoYesMandatory since October 1, 2025Must use the CDRH Portal ([202])
PMAYesVoluntarynIVD and IVD eSTARs cover PMA ([203])
HDEYes ([204])Not listed on the program page as mandatory or voluntarySlides state HDEs are in scope
Combination productsNot addressedeSTAR mandatory for 510(k) and De Novo combination productsContact Center and review division

Table 3 highlights a three-way mismatch for HDE and PMA teams: the guidance applies, but the eSTAR mechanics are voluntary or unlisted. Those teams should still follow the framework in the narrative section of their submission and confirm format expectations with their review division.

Review clocks and volumes

Available FDA data give context for planning. The MDUFA V commitment letter sets goals of a decision within 90 FDA days for 95% of 510(k)s ([205]), 150 FDA days for 70% of De Novo requests ([206]), and 180 FDA days for 90% of PMAs without advisory committee input ([207]). FDA's fiscal 2025 report lists 4,151 510(k)s received against a five-year average of 3,881 ([208]), and 4,502 Pre-Submissions, 26% above average ([209]). These figures describe FDA workload, not the share of submissions with human factors content; FDA did not publish such a share in the sources reviewed.

Planning calculator (assumptions, not an FDA benchmark)

The only numeric participant figure in the sources is the companion guidance's minimum of 15 per user population. The calculator in Table 4 multiplies it by the number of distinct populations; the 10% recruitment buffer is an assumption of this report, not an FDA recommendation, and the critical-task counts are invented inputs.

T.04
Scenario (Hypothetical Example)User populations (U)Minimum participants = 15 × URecruit with 10% buffer (assumption)Critical tasks (C)Critical-task observations = min. participants × C
Single professional group115178120
Lay user plus caregiver2303310300
Pediatric, adult and clinician3455012540

Table 4 is a sizing aid for budgeting and scheduling. Use environments, training arms, repeat devices and any FDA feedback may change the real number. The statistical rationale in the companion guidance's appendix is that a sample of 15 detected a minimum of 90% and an average of 97% of problems in the cited study ([210]), with 30 users detecting less than 2% more ([211]). Independent literature shows the limits of this logic: a resampling study of 60 users found sets of 5 found between 55% and 99% of problems and sets of 20 reached at least 95% ([212]) ([213]); earlier work reports diminishing returns with sample size ([214]); and a 2022 analysis argues the decision rule used to size usability tests has statistical limitations ([215]), with the expected net benefit of sampling reaching zero after 90 participants in their example ([216]). A peer-reviewed review argues that no a priori cohort size can guarantee a reliable assessment ([217]). A peer-reviewed paper restates that usability testing is mandatory for market access ([218]), though that statement is the authors' framing, not FDA's wording. A published example of a summative test with 15 intended users shows the common practice in the literature ([219]), and the literature describes formative testing as primarily conducted during design and development ([220]).

09

Implications and Future Directions

For teams filing after August 1, the main shift is that evidence packaging is now explicit and tied to a category dropdown. The framework rewards a living URRA and a maintained comparison against the existing device, since both determine Decision Points C and D. It also moves effort from "run a study by default" toward "justify the decision either way", which raises the value of objective evidence for rationales.

Open points to monitor as of November 2026:

  • Version drift: FDA lists eSTAR 7.1 while a trade report describes 7.0 at release. Check release notes before each submission.
  • Newer Q&A: only the town hall transcript and slides were identified, with no separate FAQ document issued afterward; the search was not exhaustive.
  • Companion guidance revisions: FDA said it intended to update the older guidance, and its file now carries an August 2026 date.
  • HDE and PMA mechanics: eSTAR status for these types is voluntary or unlisted.

A practical suggestion is that the category framework is most useful when software and human-factors teams share one traceability record, so that a UI ticket automatically triggers the Decision Point B and C screen. That is an operating suggestion, not an FDA position.

10

Frequently Asked Questions (FAQs)

What human factors information does a 510(k) need? The guidance recommends content scaled to the category: a conclusion and summary (Category 1), a rationale with report sections 1 to 4 (Category 2), or a full HFE/UE report with validation (Category 3). These are recommendations.

Does the guidance apply to De Novo and PMA? Yes, and to HDE. eSTAR is mandatory for De Novo and voluntary for PMA.

When is human factors validation data required? The guidance is nonbinding, but FDA recommends submitting validation data when critical tasks are present or impacted and Decision Point D leads to yes. FDA may still request data in specific cases.

Is "summative evaluation" the same as validation testing? FDA notes "summative testing" is sometimes used for validation testing but can be defined differently.

Is there an HFE report template? Appendices A, B and C give sample outlines for the three categories.

What happened to submissions received before August 1? FDA did not generally expect the new information but would review it if submitted.

How are combination products handled? They are outside the guidance.

11

Conclusion

The May 29 guidance and June 1 eSTAR update give regulatory and human-factors teams a usable structure: one category per submission, four decision points, report sections 1 to 8 scaled by category, and an eSTAR flow that either guides the choice or accepts it directly. An existing usability-engineering file is submission-ready when it can answer the Decision Point B and C screens against the finished device, supports any Category 2 rationale with objective evidence, and, for Category 3, contains a URRA, critical task list and validation report mapped to Table 2.

Limitations of this report: the eSTAR template itself and its on-screen field names were not read; the QMSR design-validation provisions incorporated from ISO 13485 were not read directly; no FDA statistic on HF deficiencies or eSTAR adoption was found; and the planning calculator rests on stated assumptions. Treat the crosswalk as a starting point and confirm device-specific decisions with FDA through a Pre-Submission and with qualified regulatory counsel.

The publisher

About IntuitionLabs

Build practical AI for pharma and biotech with IntuitionLabs. We help life-science teams turn complex information and workflows into useful software, governed knowledge systems and AI tools.

IntuitionLabs is an AI consulting, custom software development and data engineering firm serving pharmaceutical, biotechnology, medical-device and other life-science organizations. We work with clinical, regulatory, medical-affairs, commercial, quality and IT teams to connect technology decisions with the work people need to accomplish.

AI consulting and adoption

Our AI enablement services cover readiness assessments, use-case selection, governance and policies, team workshops, adoption measurement and ongoing advisory support. We help organizations structure the information layer behind AI: source material, context, permissions and maintained knowledge that make generated answers useful and reviewable. Private LLM inference and hosted AI options support teams evaluating how to operate AI with appropriate control over their data and infrastructure.

Software, data and life-science workflows

IntuitionLabs develops custom software for pharma and biotech, integrates enterprise systems, and builds data engineering and business intelligence solutions. Areas of focus include AI agents, regulatory research, medical writing, medical affairs, CMC information, competitive intelligence and clinical-document workflows. Our eTMF intelligence work includes cross-system reconciliation and inspection-readiness support.

Enterprise platforms and regulated delivery

We provide Veeva services, application support, managed services, integrations and custom applications, alongside enterprise content work involving platforms such as Egnyte. For regulated workflows, our services include GxP enablement, computer-system validation and software development addressing 21 CFR Part 11 requirements. The applicable controls, validation responsibilities and acceptance criteria are defined for each engagement.

Work with IntuitionLabs

Explore AI enablement, pharma and biotech software development, data engineering and BI, and Veeva services. Contact IntuitionLabs to discuss your workflow, information sources and implementation needs.

IntuitionLabs publishes educational research to help life-science teams make informed technology decisions. Coverage of a product or organization does not imply a client relationship, endorsement or partnership.

Sources / 220
Adrien Laurent

Need Expert Guidance on This Topic?

Let's discuss how IntuitionLabs can help you navigate the challenges covered in this article.

I'm Adrien Laurent, Founder & CEO of IntuitionLabs. With 25+ years of experience in enterprise software development, I specialize in creating custom AI solutions for the pharmaceutical and life science industries.

Disclaimer

The information contained in this document is provided for educational and informational purposes only. We make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability, or availability of the information contained herein. Any reliance you place on such information is strictly at your own risk. In no event will IntuitionLabs.ai or its representatives be liable for any loss or damage including without limitation, indirect or consequential loss or damage, or any loss or damage whatsoever arising from the use of information presented in this document. This document may contain content generated with the assistance of artificial intelligence technologies. AI-generated content may contain errors, omissions, or inaccuracies. Readers are advised to independently verify any critical information before acting upon it. All product names, logos, brands, trademarks, and registered trademarks mentioned in this document are the property of their respective owners. All company, product, and service names used in this document are for identification purposes only. Use of these names, logos, trademarks, and brands does not imply endorsement by the respective trademark holders. IntuitionLabs.ai is an AI software development company specializing in helping life-science companies implement and leverage artificial intelligence solutions. Founded in 2023 by Adrien Laurent and based in San Jose, California. This document does not constitute professional or legal advice. For specific guidance related to your business needs, please consult with appropriate qualified professionals.

Related Articles

Need help with AI?

© 2026 IntuitionLabs. All rights reserved.