Overall residual risk according to ISO 14971:2019

[KrisD]

Hi everyone,

I have few questions regarding the new requirements of ISO 14971:2019. Any comments or suggestions would be greatly appreciated.

1. Anyone has any ideas about how to set up the criteria of acceptability for overall residual risk according to ISO 14971:2019.
2. Is it possible to use the risk matrix like the one used for individual residual risk evaluation to evaluate overall residual risk?

Thanks.

 

[yodon]

They published a guidance document on application of 14971, ISO/TR 29471:2020. It has some good guidance on those topics. It's a worthy investment.

 

[KrisD]

They published a guidance document on application of 14971, ISO/TR 29471:2020. It has some good guidance on those topics. It's a worthy investment.

Thanks for your reply. I already read the technical report, but there still some points regarding overall residual risk evaluation has not been fully understood. Actually, we had a acceptability criteria like "Overall residual risk is judged as available if there is no individual risk that judged as unacceptable". However, I feel like that is not really the appropriate way to define the criteria.

 

[yodon]

Indeed, that's probably a bit on the short side as they want the system considered as a whole. We take several approaches. First, we do a 'heat map' showing where all the final RPNs lie in the acceptability range. If you see clustering towards unacceptable, you may want to either see if you can further reduce those or provide additional justification as to why the benefits outweigh the risks. The second is tied directly to the overall Benefit-Risk Analysis. We abstract the benefits and abstract the risks and then profile each. An expert on the clinical side then weighs in on the acceptability. This FDA Guidance gives an overview and a similar approach is described in the EU MEDDEV 2.7/2 rev. 2 (look at A.7.2).

Unfortunately, it's not a clear-cut path and so you establish what rationale you can and hope for the best.

 

[Tagin]

Thanks for your reply. I already read the technical report, but there still some points regarding overall residual risk evaluation has not been fully understood. Actually, we had a acceptability criteria like "Overall residual risk is judged as available if there is no individual risk that judged as unacceptable". However, I feel like that is not really the appropriate way to define the criteria.

I would agree that it is not appropriate, because it does not take into account even the spirit of TR24971 clause 8, which is to evaluate the ways in which individual residual risks could combine, or occur in sequences, which might result in a larger risk.

 

[KrisD]

Thanks for your comment. I might look into ISO/TR 24971 again and review our overall residual risk acceptability criteria.