Medical Devices Group

  • Community
  • Webinars
  • Jobs
  • Events
  • Contact
  • Go Premium
« Back to Previous Page
like 35 comments  share
Joe Hage
🔥 Find me at MedicalDevicesGroup.net 🔥
December 2016
ISO 14971:2016 and Labeling
4 min reading time

As originally asked by Rich Meader.

How Do You Handle Labeling Under ISO 14971:2012?

ISO 14971:2012 disallows the use of labeling as a mitigation to risk. This seems to present a problem if your initial assignment of a risk probability does not take into consideration design controls as prescribed by some risk analysis gurus (though not everyone takes this approach). So if your probability of a hazard is high before design controls then this seems to imply that after you carefully design your labeling and subject it to usability assessment, etc. you can not use these efforts to say that your probability has been reduced.

For example, let’s say that you have a medical device that uses ultrasonic energy for a diagnostic or therapeutic indication. It may be important that the device not be used on patients that are implanted with certain devices (e.g. an ICD or pacemaker). How do you mitigate the risk of this hazardous situation other than through labeling? Might one answer be to elevate the labeling to include a “WARNING” statement? Has anyone found warning symbols that might be appropriate for the proposed hazardous situation?

Am I missing something here? How do others out there see this issue?


Erin Seiler, MBA
Account Manager, Color & Additives North America at PolyOne
A huge trend in labeling we are seeing on the manufacturing end, is a change from stickers or other forms of identification to laser marking technology. Laser marking allows companies to etch the surface of plastic and metal taking away the need for labels. This might be a direction that becomes more popular as the medical device industry continues to become more regulated.

Rick McCullar
Senior Quality Engineer, ASQ-CQE, Invacare
Seems like annoy of confusion still remains on this issue in the industry.

Beluh Mabasa Ginting
Section Head of Standardisation at Directorate of Supervision of Medical Devices and Household Health Ministry of Health
Base on of ISO 15223-1: 2012,
Label
“written, printed or graphic information provided upon the medical device itself”
Labelling
” information supplied by the manufacturer that is provided for, associated with, or affixed to, a medical device or any of its containers or wrappers”. So labelling can’t be regard or used as a risk mitigation measure and automatically is not part of the accompanying document (document accompanying ME equipment, an ME system, equipment or an accessory and containing information for the responsible organization or operator particularly regarding basic safety and essential performance: IEC 60601-1:2005)

Daniel John
Passionate/Expertise in Medical Device Strategic Regulatory Affairs & Quality Assurance and R&D
Core of ISO 14971:2012 is with regards to Section 7.4. Risk-Benefit Analysis, which has been made compulsory and if addressed fully then the rest of the sections of ISO 14971 will be adequately covered. So even if the third and final risk control measure, Information for Safety, can not be used as a risk mitigation measure and your RPN remains in the AFAP or Intolerable Region of your Risk Analysis Matrix, it does not matter. We have to conclude that the Residual Risk of subject device is same as already CE marked device/s with objective evidence and that’s what the NB reviewer want to look at. Do we have any other alternative ? No risk can be entirely eliminated but residual risk acceptability rationale is the key. Even if the whole risk thing might not have real-life applicability, it at least ensures that all/most risks are being considered/documented/mitigated to AFAP and the devices are currently, the state of the art.

David Amor
Digital health + combination product Regulatory & Quality executive
Marie – yes NBRG or Team NB-MED. They go through so many acronyms (NBOG was notified body operations group).

Kriss Anderson
Director of Regulatory Affairs, Consumable Medical Devices at ICU Medical
We removed instructions to the user as a risk control from our risk management file to comply with EN ISO 14971:2012. We were then told by the notified body that we had not reduced the risk as far as possible, because we had not used all possible risk control measures. We were forced to put instructions to the user back into the risk management file as a risk control, but then attribute no risk reduction to that risk control. This is one option at address the original question by the poster.

The whole risk thing is mired in bureaucratic nonsense with no real-life applicability. Does anyone really believe that when we provide surgeon training with cadavers (yes, that’s instruction to the user), it does nothing to reduce risk??? The EU has broken risk management and it no longer has any value. No risk can be entirely eliminated, so all risks have residual risk. Does anyone really take every single risk (no matter how negligible) and warn against it in the IFU? The MDD is broken.

James (Jim) Dent, LSSBB, DTM
Sr. Mfg Project Manager at Avalign-Thortex
Mitigating Risk should be accomplished by:
‘– designing the device to be safe, and
‘– implementing protective measures within the device itself

Warning information on labels is only supplemental information and does actually reduce the chance of a device being used incorrectly.

Marie Suetsugu
QA/RA
Sorry, I should perhaps have said NBRG one…

Marie Suetsugu
QA/RA
Dear David, could you please let me know where exactly one can find the ‘NBOG Annex Z guidance’? Or, do you mean the NB-MED one? http://www.team-nb.org//wp-content/uploads/2015/05/nbmeddocuments/NBRG_WG%20RM_Interim_NBmed_Consensus_Version_140812_1_1.pdf

Marcelo Antunes
Regulatory affairs strategy consultant at SQR Consulting
ISO 14971 details two types of information.
1 – Information for safety – the third option of risk control measure, usually indication of the need for an action or no action to prevent or control a hazardous situation.
2- Information (the standard calls it disclosure) about residual risks – not a risk control measure, this is related to risks remaining after controls have been implemented.

Section A.2 of Annex I to Directive 98/79/EC says that users shall be informed about the residual risks (2) and the manufacturers shall not attribute any additional risk reduction to this information . This type of information (about residual risks) is what the deviation focus on. So, any information that discloses residual risks (2) cannot reduce risks.

Any information for safety (1) can reduce risks.

David Amor
Digital health + combination product Regulatory & Quality executive
Hi folks: please recognize that “labeling” is not sufficient a distinction to distinguish between whether it constitutes a “risk control” or not. There are two types of information per Annex J in 14971: disclosure of residual risk (which cannot be considered a source of risk control as explaining what risks may occur from use of a product logically does not mitigate such use) and information for safety (which infers safe use and thus MAY be treated as a risk control). I would suggest reviewing Edwin Bill’s comments on reviewing the NBOG Annex Z guidance for their interpretation of the content deviations. Please defer to the actual statutory requirements and inform yourselves on what Annex ZA/B/C ACTUALLY say.

Hugh F. McCann, Jr
CEO/Co-Owner at Identification Products Corporation
www.idproducts.com

Les Henderson
QA/RA Consultant at NSC, LLC
I agree with Jon, but there is a distinction that can be made that labeling should be the only mitigation considered.

Rich Meader
Principal at Meader Consulting, LLC
Hi Mark, can you please expand a bit on your comment. I am not sure that I am following correctly. Lots of back and forth here and would like to understand the bottom line on this issue. Thank you for weighing in on this.

Mark Swanson
Principal Quality and Regulatory Consultant
This is NOT correct. The information on residual (as required by the regulations) does not mitigate risk is what the content deviation states People need to quit misinforming on this.

Rich Meader
Principal at Meader Consulting, LLC
John, yes I agree regarding the Z annexes. Harmonization with the MDD and Essential Requirements which disallow the use of labeling to mitigate risk?

Michael Wienholt, RAC
Director of Regulatory Affairs at Q² Solutions
Jon makes an important distinction with this comment.

Jon Speer
Founder of Greenlight Guru – Helping medical devices get to market faster with less risk
Great comments. The Z annexes of EN ISO 14971:2012 indicate labeling cannot be used as a risk mitigation. This version of 14971 implies that labeling is mandatory part of a medical device.

But as the comments indicate, labeling is definitely a risk control measure.

Stacy Tseng
Regulatory Compliance Specialist
The other common way to do is to add a warning instruction of use symbol (blue man) which must stay in color of blue in label. Then you can add clear warning statement in IFU.

Andrew Keirns
RWS Sr. Director Business Development
I’ll get Marc Miller to weigh in – he took us through 14971 many years ago

Markus Weber
Think in data and content, not in documents.
The thing is about risk management everyone likes to reduce the risk as low as possible. in my opinion all labeling activities are just to address the thing about liability, but labeling is never really reducing any risks. training with written confirmation could be used as mitigation in my opinion. Finally it’s the question how many people are going to die with this device with foreseeable miss use and how many will survive based on the use of this device (benefits). think about performing a risk assessment about a scalpel or an injection needle. how are you going to reduce the risk of cutting blood vessels?

Marcello Conegliano
Regulatory Affairs Engineer
If alert symbol or label with warning statement haven’t a risk mitigation effect, you should evaluate that the removal of those not lead to an increase of incidents …
So, you can mitigate the risk “as far as possible” by the use of alert symbol or warning states because you haven’t the control on possible mistake of the operator (even if the operator is skilled) .

Jeff Gutmann
Safety and Regulatory Engineer at General Electric
Labeling alone is not adequate. Design, Protection and Labeling are options in that priority. I like the protection idea of the software confirming with the operator that they have checked for an incompatible device such as a pacemaker. Labeling is often not read. Asking the user to perform an ACTION that is mandatory to move ahead should decrease risk.

Ivan Liljegren
Director QA&RA at Breas Medical AB
I think common sense asserts that you need to inform users about critical restrictions in the indications for use (e.g. “Not to be used on patients with pacemakers”). In doing so, you must not rely solely on written statements in the IFU. If your device has a software user interface, you could add a mandatory prompt which asks the user to confirm that the patient is part of the intended population (i.e. Does not have a pacemaker) before commencing treatment.

Stefan Eisenring
Leiter Regulatory Affairs & Qualitätsmanagement
I like Maurizio’s approach. Though, it would be interesting to know what the risk score of this hazard is. If it was AFAP and you could not mitigate it any further other than by labeling, you would have to assess and document whether the medical benefit outweighs the risk. Anyhow, have you considered to let the user confirm in the GUI that patient does not have such a device implanted before start?

Michael Wienholt, RAC
Director of Regulatory Affairs at Q² Solutions
To Julie’s point, the standard does no such thing. See D.5.1 Risk control option analysis, which states: There are several approaches to reducing risk, which can be used alone or in combination with each other.
The designer/engineer has accordingly to explore different options how, in a reasonably practical way, to
reduce risk(s) to acceptable levels. The following is a non-exhaustive list of risk control approaches that are
commonly used.
c) Providing information for safety by:
⎯ placing warnings in the labelling of the medical device,

Michael Wienholt, RAC
Director of Regulatory Affairs at Q² Solutions
“D.5.1(c) Risk control option analysis” explicitly allows labeling as a risk mitigation: “Providing information for safety by:
⎯ placing warnings in the labelling of the medical device”

Katarzyna Zofia C.
Senior Quality Assurance & Regulatory Affairs & Clinical Affairs (BE + MSc + MBA Student)
Risk Controls should be focused on the specific design features first and labeling as a last resort.

Katarzyna Zofia C.
Senior Quality Assurance & Regulatory Affairs & Clinical Affairs (BE + MSc + MBA Student)
Why? Because the information for safety is a requirement of medical devices and can NOT be used to reduce risk.

Katarzyna Zofia C.
Senior Quality Assurance & Regulatory Affairs & Clinical Affairs (BE + MSc + MBA Student)
Julie O. One of the news of the Z annexes of EN ISO 14971:2012 (no ISO 14971:2007) is that the risk reduction has to be more than information provided to the user (e.g. labeling and Ifu)

Michael Eberhard
QA & RA Manager bei RUDOLF Medical GmbH + Co. KG; Head of R&D
Rich,
Maurizio is right, those instruments are used by trained personnel only and they know or remember the meaning of the warning signs.
You can combine a pacemaker warning sign with information about the source of danger – see http://www.safetysign.com/pacemaker-warning-signs

Jayeshkumar Dangruchiya
Digital Devices Expert
I think Rich is referring to Annex ZA of ISO 14971.

Rich Meader
Principal at Meader Consulting, LLC
Thanks Maurizio.

Julie Omohundro
Principal Consultant at Class Three, LLC
Rich, can you provide the section of ISO 14971 that disallows the use of labeling as a mitigation of risk?

Maurizio Colombo
Quality Assurance & Regulatory Affairs at MOSS S.p.A.
If the device is to be used by professional users then risk mitigation might be obtained through a specific and documented training to be supplied during device installation.
Hope this might help…

Marked as spam
Posted by Joe Hage
Asked on December 12, 2016 3:21 am
4059 views
  • Follow
  • Unfollow
  • Report spam
like 35 comments  share

Meet your next client here. Join our medical devices group community.

« Back to Previous Page
Ask a Question
Leave a Comment

We still use LinkedIn to access our site because it’s the only way to “pull in” your LinkedIn photo, name, and hyperlink to your profile page, all vital in building your professional network. When you log in using LinkedIn, you are giving LinkedIn your password, not me. I never see nor store your LinkedIn credentials.

Stay connected with us.

By signing up you are agreeing to our Privacy Policy.

Categories

  • Capital/Investment
    • Business Model
    • Funding
  • Careers
  • Design/Devel
    • Design
    • Development
    • Human Factors
    • Labeling
    • Material Selection
    • R&D
    • Trials and Post-Market
  • Featured
  • Industry
    • Announcements
    • Device Tax
    • Hospital and Health Care
    • Innovation
    • Medtech
  • LinkedIn, etc.
  • Markets
    • Africa
    • Americas
    • Asia
    • Australia
    • Europe
  • Regulating
    • CE Marking
    • EU
    • FDA
    • FDA/EU etc.
    • Notified Bodies
    • Quality
    • Regulatory
  • Selling
    • Distribution
    • Intellectual Property
    • Marketing/Sales
    • Reimbursement
  • Worth bookmarking!
Feature your job here.
logo

Companion to LinkedIn's 350,000 member community

  • Contact
  • Medical Device Marketing
  • In Memoriam
  • Medical Device Conference

The Medical Devices Group   |   Copyright © 2025 Terms, Conditions & Privacy

Medical Devices Group
Powered by  GDPR Cookie Compliance
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.

Strictly Necessary Cookies

Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.

If you disable this cookie, we will not be able to save your preferences. This means that every time you visit this website you will need to enable or disable cookies again.