Regulatory Open Forum

 View Only
  • 1.  Usability engineering file vs. FDA Appendix A HFE/UE Report

    This message was posted by a user wishing to remain anonymous
    Posted 15-Apr-2020 13:52
    This message was posted by a user wishing to remain anonymous

    Hi all, 
    I am going through IEC 62366-1:2015 (Ed.1) and the FDA guidance on human factors and I am trying to figure out differences and similarities between what IEC calls "usability engineering file" and FDA "HFE/UE Reports". I see differences both in terms of content and outline when I compare the section heading of the standard vs. the structure of the documentation requested by the FDA.
    Does any of you have a checklist or a good template to use for such documentation to comply with both at the same time and make happy both the NBs in Europe and the FDA?
    Thanks in advance for your help.


  • 2.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    Posted 16-Apr-2020 17:40

    First and foremost, HF/UE studies, as part of risk assessment, are within a risk management process. We need to familiarize ourselves with ISO 14971 and other FDA-recognized standards on Human Factors: IEC 62366, IEC 60601, and AAMI/ANSI HE75. Once you have RM process set in place, a common approach to satisfy both US and EU Submissions HFE/UE Report would contain the following elements:

    1. Understand users and use characteristics to define use specification - patients, user, use-environment, etc
    2. Assess user interface characteristics related to safety and evaluate known use-related problems – g., Use FEMA
    3. Formative Evaluation(s) - expert review, physician feedback, usability test, etc., may require more than one formative evaluations depending on the type of the device
    4. Summative Evaluation - validation testing with representative users, use-environment and final design
    5. HFE/UE Report - summarize all HF activities, concluding the product is safe and effective, and potential use errors and failures have been eliminated or limited to the extent possible through the application of the HF methods.

    There are situations where full-scale human factors or usability engineering process may not be required (e.g., legacy devices or low-risk products), every project is different, the process should be tailored accordingly. Good luck.



    ------------------------------
    Jo Huang RAC
    Assoc Manager - Regulatory Affairs
    Irvine CA
    United States
    ------------------------------



  • 3.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    Posted 17-Apr-2020 09:28
    Thanks, Jo for mentioning the risk management standard, ISO 14971:2019, is the current edition.  The point you make is that these are all connected processes.  IEC 62366-1 in Annex A, I think, has a diagram that illustrates the connections between the usability processes and the risk management processes.  It is essential to a complete overall residual risk evaluation (Clause 8 in 14971) that you have all the risks, including those identified in usability engineering, as illustrated in the diagram in 62366-1, sent to the risk management process. These and other process standards operate as part of a complete quality management system under ISO 13485:2016. It is vital that we take a systems view of our processes, they are not disconnected, rather all interconnected!

    ------------------------------
    Edwin Bills MEd, CQA, RAC, BSc, CQE, ASQ
    Principal Consultant
    Overland Park KS
    United States
    elb@edwinbillsconsultant.com
    ------------------------------



  • 4.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    This message was posted by a user wishing to remain anonymous
    Posted 17-Apr-2020 10:48
    This message was posted by a user wishing to remain anonymous

    Thank you both for your very helpful insights.
    In terms of final documentation, what is the best strategy to avoid double work and format the usability engineering file in a (smart) way that it is compliant with 62366-1 and can be directly used for the 510k submission? 
    Please let me know your thoughts on this point. 
    Thanks


  • 5.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    Posted 17-Apr-2020 13:15
    Although FDA has a recommended structure, there is no standardized/harmonized format for the HFE/UE report itself. As long as you cover the above mentioned elements and reference your RM files as Edwin pointed out, including portions of risk analyses focusing on user interactions with the device and specific risk analysis processes, it should satisfy both parties. The level of detail of documentation submitted should be sufficient to describe your identification, evaluation, and final assessment of all use-related hazards for the device. 



    ------------------------------
    Jo Huang RAC
    Assoc Manager - Regulatory Affairs
    Irvine CA
    United States
    ------------------------------



  • 6.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    Posted 18-Apr-2020 17:04

    The IEC 62366-1:2015 Usability Engineering File and the FDA HFE/UE Report are different because they serve different purposes.

    The Usability Engineering File collects all the documents related to the usability engineering process. You should make your own checklist for your usability engineering process by searching IEC 62366-1:2015 for the phrase "Compliance is checked by". You will get everything that needs to be in file. Then determine how you implemented the usability engineering process. For example, you might have tailored it or you might have a legacy interface with UOUP. Adjust the checklist. Then identify the specific documents required by your implementation. Be mindful of EU-MDR Annex II(4)(d) which requires, "the precise identity of the controlled documents offering evidence of conformity with each harmonized standard, CS, or other method applied to demonstrate conformity with the general safety and performance requirements". The contents of the Usability Engineering File fall under this provision. See the EU-MDR Annex I(5), for example, since it includes use-error.

    There is often a question of the residence of the Usability Engineering File. Usability engineering is one of the specific applications of Risk Management. ISO 14971:2007 is a normative reference, but you should implement ISO 14971:2019. Some people like to include the Usability Engineering File in the ISO 14971:2019 Risk Management File. Others keep it as a separate file. In either case, the ISO 14971:2019 Risk Management Plan applies. You will need, at least, the risk acceptability criteria for the case when you cannot estimate the probability of occurrence of harm. You will not be able to estimate the probability of use-error as you analyze the hazard-related use scenarios.

    There is often a question of the residence of the Risk Management File and the Usability Engineering File. In an FDA QSR implementation, they are in the Design History File. In an ISO 13485:2016 implementation, they are in the Design and Development File. After release to production certain parts of these files move for maintenance in the Production and Post-production phases.

    The FDA HFE/UE Report is just that, a report. It is a document not a file. There are two situations for FDA, whether or not the pre-market submission requires the report. There is a draft guidance document, List of Highest Priorities Devices for Human Factor Review, that answers the question.

    In QSR, usability is part of the 820.30(g) design validation because it addresses patient or user needs. If you need to write the report, follow the format in the guidance document Applying Human Factors and Usability Engineering to Medical Devices, Appendix A. Include the report in the DHF. I recommend including it in the Usability Engineering File.



    ------------------------------
    Dan O'Leary CQA, CQE
    Swanzey NH
    United States
    ------------------------------



  • 7.  RE: Usability engineering file vs. FDA Appendix A HFE/UE Report

    Posted 19-Apr-2020 08:28
    I have to make a comment.  I think auditors, consultants, Registrars, NBs, and regulators expecting a certain document to be in "this file" or "that file"' or expecting tools using thirteen ways from Friday to be used to estimate risk are really turning people off from doing anything. This can lead to all sorts of extra non-value procedures and activities for companies, or they just throw up their hands and do nothing until they have a major finding or lawsuit.

    For example, for even something as "simple" as filing, my simple suggestion - make a "folder" for a Design File in an electronic or paper system .  In that folder, have a sub-file for RMF.  In THAT folder make a sub, sub-folder for UEF in there.  KEEP track of those and all specific documents in the DHF Index , which is similar to an EP or ER checklist.  Keep it updated.


    DONE.

    Everytime I see comments lamenting clients' lack of understanding about risk management (ISO-14971), I want to protest.   Yes, you can use thirteen tools, numerous experts, literature reviews, statisticians for probability estimations.  Does one need all that for lower risk devices? Please consultants do not scare innovators in application of over a voluntary standard.  It just seems to me over time "we" (the industry) are creating a monster that only certain qualified "Risk Management Expert Consultants" charging $300 an hour will be deemed qualified to perform. 

    Having to read and apply the 100 page TIR 24971 to "interpret" ISO 14971:2019  is overwhelming for a start-up.  To then criticize any effort is demoralizing to them. It just seems to me, someone who has seen very bad (lack of) RMFs and over-complicated RMFs even risk management experts in a large company can explain, that "we" have screwed this up.

    That said, I have been involved in RM since implementation of ISO 14971:2007, as a user and now a consultant. I have seen its application evolve through "preliminary [software] hazard analysis leading to hazard analysis.  I have seen and used detailed software and use hazard analysis with event sequence interactions leading to software safety classification and use studies. Other tools used: sFMEA, dFMEA, pFMEA, uFMEA, FTAs of all sorts.  I have seen many different ways to write the final risk management report.

    This said,  I am myself uncertain where "we" are letting requirements of a "voluntary standard" end versus committee and consultant decided "best practices" start to creep in, so I myself  am taking a refresher 2 day course on ISO 14971:2019 at the end of the month.

    After reading the redlined version of ISO 14971:2019, its TIR, the standard IEC 80002-1, and the risk standards on cybersecurity risk management multiple times, it seems to me that it is time for committees to stop birthing standards and for consultants to keep it simple when it can be. 

    Ok, thanks for reading my rant. Peace be with you.




    ------------------------------
    Ginger Cantor, MBA, RAC
    Founder/Principal Consultant
    Centaur Consulting LLC
    River Falls, Wisconsin 54022 USA
    715-307-1850
    centaurconsultingllc@gmail.com
    ------------------------------