Best Risk Analysis Approach (ISO 14971) - Class IIa Medical Devices

[tantan]

Hi,

I just wanted to get some feedback from any of you there familiar with the application of risk management (and more specifically risk analysis). We are a specifications developer for Class IIa medical devices. I am new to the company and have noticed that they have only previously done an analysis (FMECA) but with no real management process. Ive read in some articles that FMEA is better for Class III or safety critical parts of a device, and that different analysis methods should be used at different stages of design control. Any feedback on what Ive read or what others are doing would be muchly appreciated!

Cheers

Tan :-)

[Coury Ferguson]

Has anyone performed a Risk Analysis on the Class II Medical device? Any comments/suggestions?

[MIREGMGR]

We always focus more on whether we have input from the right mix of medical procedural experts and device-type-technology historical- and theoretical-failure-potential experts, than we do the formal failure analysis technique and format.

[gholland]

Here are the basics of what we do as a Class II/IIa medical device company. As context we are recovering from a 483 letter and our risk analysis procedure is now <very> thorough. It may be overkill but I know it has the blessing of FDA:

- Develop your Top Level Hazards (TLH). These are the risks to the user, patient, and caregiver. Develop your severity rankings for these TLHs here. The severities do not change and are carried through the entire analysis.

- From your TLHs you develop your basic fault events (BFEs). Basically 'how did my TLH happen?'. Develop your Likelihood of failure here. The multiplication of the severity and the likelihood gives your Risk level. See ISO 14971 for examples.

- Develop a fault tree analysis from this information. Your failure modes leading to your BFEs leading to your TLHs. You can then develop your Risk Control Measures (RCMs) from this information. These RCMs are what you would track and Verify during testing. Basically do your RCMs work as intended to reduce the risk of the BFE? Your RCMs are what you use to reduce the Likelihood number in your Risk analysis. If your RCMs are not effective you are not justified in reducing your Likelihood number, but if you can prove they are effective then you're OK.

- This completes your fault tree. This is a top down approach, i.e. what is my failure and how did I get there? From here you would then go the other way for dFMEA and pFMEA. For your dFMEA you would take each component and ask yourself "if this fails does it lead to a TLH?". If the answer is no then move on to the next component. For you pFMEA you ask yourself "if this process fails can it lead to a TLH?". If no then move on. You would develop more risk control measures from your FMEA if you have missed anything in the FTA.

An example of a TLH would be 'Patient receives a non-sterile infusion' or something similar. A BFE would be 'Packaging failure due to xxx causes sterile barrier breach'. A RCM would be 'Packaging tested to ISTA 2A requirements'. You would then be able to establish traceability from your packaging test to your risk analysis.

[Deepoz]

Hi
We manufacture custom-made sterile Procedure packs, some of which are class IIb due to components used within. I have developed a risk management checklist as per the standards which lists hazards and control measures implemented. We generate similar checklist for all products individually. Since there are no design issues involved there are only a few hazards identified but I have attached a copy of one of our checklists if it helps. It is an excel sheet with links to our component database which makes life easy for generating any new product checklist. It has been audited and accepted by Australian regulator (TGA).
Cheers

[tantan]

Thanks for the attachment! That looks more like the direction I want our risk analysis to head in

[Ajit Basrur]

In Reply to Parent Post by Deepoz

Hi
We manufacture custom-made sterile Procedure packs, some of which are class IIb due to components used within. I have developed a risk management checklist as per the standards which lists hazards and control measures implemented. We generate similar checklist for all products individually. Since there are no design issues involved there are only a few hazards identified but I have attached a copy of one of our checklists if it helps. It is an excel sheet with links to our component database which makes life easy for generating any new product checklist. It has been audited and accepted by Australian regulator (TGA).
Cheers

Great attachment and welcome to the Cove

Hope to see your active participation

[Deepoz]

no worries