Risk Evaluation Matrix-Product Portfolio

alimary15

alimary15

Involved In Discussions
Hello,

a quick question :)

Does it make sense to have the same Risk Evaluation Matrix for completely different products?

Shouldn?t different products ( such as Apps, or heavy machines) have different tresholds of acceptable/unacceptable risks?

What is the best practice? Is it better to your experience to have 1 Evaluation Matrix for all, or to define 1 evaluation Matrix for each single product/device?

Thank you very much!
 
M

Marcelo

Inactive Registered Visitor
The evaluation matrix you mentioned is the graphical interpretation of the risk acceptability criteria.

The criteria should be based on the risk management policy, and should be created for each device.

So yes, you should have different ones for different devices. The only rationale for having the same for different devices is if the devices have generally the same risks, for example, if all are lung ventilators. If you manufacture a lung ventilator and an implant, or a lung ventilator and ultra-sound machine, there?s no way the risk acceptability criteria would be the same.

However, most of the time people use only one. This is common practice, not best practice, thought.
 
Last edited:
alimary15

alimary15

Involved In Discussions
Marcelo Antunes said:
The evaluation matrix you mentioned is the graphical interpretation of the risk acceptability criteria.

The criteria should be based on the risk management policy, and should be created for each device.

So yes, you should have different ones for different devices. The only rationale for having the same for different devices is if the devices have generally the same risks, for example, if all are lung ventilators. If you manufacture a lung ventilator and an implant, or a lung ventilator and ultra-sound machine, there?s no way the risk acceptability criteria would be the same.

However, most of the time people use only one. This is common practice, not best practice, thought.
Click to expand...
Thanks for the feedback !

In case the devices are totally different I would go for the option to have different acceptability criterias. However, if having a general and "good-for-all" policy, would it then make any sense to discuss possible residual risks in a risk/benefit analysis?

By having an acceptability criteria that is unique for all, wouldn?t there be a risk of classifyng as acceptable risks that might not be acceptable? I am concerned about and if it is possible to lose any kind of information within the RM process by doing so.

What is your opinion?

Thanks
 
M

Marcelo

Inactive Registered Visitor
However, if having a general and "good-for-all" policy, would it then make any sense to discuss possible residual risks in a risk/benefit analysis?
Click to expand...

The risk management policy is the high-level policy on how the enterprise deal with risk. Policies are direction, but really not directly actionable. That why you need to define the criteria for each device based on the policy. Please see an example of policy discussion in this thread - http://elsmar.com/Forums/showthread.php?t=52843

By having an acceptability criteria that is unique for all, wouldn?t there be a risk of classifyng as acceptable risks that might not be acceptable? I am concerned about and if it is possible to lose any kind of information within the RM process by doing so.
Click to expand...

Yes, that why is makes no sense to have a risk acceptability criteria for all devices, as I mentioned. This problem might even happen if the devices are similar but have modifications that can impact the risks.
 
alimary15

alimary15

Involved In Discussions
Marcelo Antunes said:
The risk management policy is the high-level policy on how the enterprise deal with risk. Policies are direction, but really not directly actionable. That why you need to define the criteria for each device based on the policy. Please see an example of policy discussion in this threadhttp://elsmar.com/Forums/showthread.php?t=52843
Click to expand...

Sorry for the bad use of words- I was always speaking about the risk criteria of acceptability!

Thanks for the help!
 
alimary15

alimary15

Involved In Discussions
This problem might even happen if the devices are similar but have modifications that can impact the risks.
Click to expand...

In this case for example, when upgrading the same device, it is practice to start from the previous RA and then move forward to identify new possibile risks that are coming from the new modification.

My only concern is using the same acceptability criteria for software, hardware, and different kind of devices. I think it would make things way more general.

I will try to discuss this further within my team and try to change things.

Thanks for the confirmation :thanx:
 
Last edited by a moderator:
M

Marcelo

Inactive Registered Visitor
Ah, ok, well, people generally use those concepts interchangeably, but they are very different, which causes confusion in the discussion and understanding.

So, if your question is

However, if having a general and "good-for-all" risk acceptability criteria, would it then make any sense to discuss possible residual risks in a risk/benefit analysis?

Yes, it would, because the residual risk are related to risk control measure, and even if similar devices would have similar risks, it does not mean that they would have the same risk control measures (in fact, newer devices in a device family usually have different risk controls measures to some risks because that?s how technology evolution works.
 
M

Marcelo

Inactive Registered Visitor
In this case for example, when upgrading the same device, it is practice to start from the previous RA and then move forward to identify new possibile risks that are coming from the new modification.
Click to expand...

Yes, and this is acceptable by ISO 14971 IF you perform and evaluation and concludes that you can use the proviso RA.

My only concern is using the same acceptability criteria for software, hardware, and different kind of devices. I think it would make things way more general.
Click to expand...

Exactly. They probably need separate ones, based on the same high level risk management policy.
 
alimary15

alimary15

Involved In Discussions
Ok great thank you!

Then I might have a more complex question:

Let?s consider a scenario where we have different modules that could be or could be not present on the field and that are sold as a complete system to the costumer.

Then what is it the best practice? Make RA for each single module? Shall RA also be made to the complete and overall system to include possible risks arising from interactions between each single module?

What?s the best practice in this case?

Thanks
 
M

Marcelo

Inactive Registered Visitor
Then I might have a more complex question:

Let?s consider a scenario where we have different modules that could be or could be not present on the field and that are sold as a complete system to the costumer.

Then what is it the best practice? Make RA for each single module? Shall RA also be made to the complete and overall system to include possible risks arising from interactions between each single module?

What?s the best practice in this case?

Thanks
Click to expand...

Risk management (as defined in ISO 14971 and as required by most regulations) is applicable to the medical device as a whole.

So yes, you would need to focus on the complete device, and as you mentioned, part of the focus is on interactions between each single module - in fact, think of it as interactions in the different configurations.
 
You must log in or register to reply here.
Top Bottom