Labelling can be use as risk control..?

Hi,

I have a question, and I am looking for standards (ISO13485, ISO14971, FDA or MDR) that say labeling can be used as a risk control measure.
Basically, what I am understanding is that we can try first to define design change, and control environment, and if it does not work, then we can use labelling as a risk control measure.
But there is no clear information I can find.

If someone has knowledge or experience with this, please help me.
Thank you in advance.

Read A Beginner's Guide to Product Safety Standards.

The risk mitigation hierarchy is to start with the top strategy and only go lower if a higher level strategy is not feasible.

1. Design the hazard out of the product.
2. Design the product to fail in a safe mode.
3. Guard against the hazard.
4. Warn about the hazard.

Warning such as applying labels is the least effective approach as people have become desensitized to warning labels due to the proliferation of silly warnings found on consumer products. These were not made up. I have actually seen these.

And let's not forget the required warning that hot coffee is hot!

The EU has stated that if labeling alone is the control then the risk should not be lowered. I personally agree with this. If (outside the EU) you do believe the risk is lowered based on labeling alone, you'd want really good data to show the labeling was effective in reducing the risk.

That said, labeling is very much a valid control, just not very effective (if at all) for the reason @Miner noted.

Miner said:

The risk mitigation hierarchy is to start with the top strategy and only go lower if a higher level strategy is not feasible.

1. Design the hazard out of the product.
2. Design the product to fail in a safe mode.
3. Guard against the hazard.
4. Warn about the hazard.

Warning such as applying labels is the least effective approach as people have become desensitized to warning labels due to the proliferation of silly warnings found on consumer products. These were not made up. I have actually seen these.
View attachment 30024 View attachment 30025
View attachment 30026

Click to expand...

Thanks for clear information

yodon said:

And let's not forget the required warning that hot coffee is hot!

The EU has stated that if labeling alone is the control then the risk should not be lowered. I personally agree with this. If (outside the EU) you do believe the risk is lowered based on labeling alone, you'd want really good data to show the labeling was effective in reducing the risk.

That said, labeling is very much a valid control, just not very effective (if at all) for the reason @Miner noted.

Click to expand...

Hi, so if we want to use it as a risk control in EU, what should we prepare for proof?

evan_kim90 said:

Hi, so if we want to use it as a risk control in EU, what should we prepare for proof?

Click to expand...

If you use it (alone) to reduce the risk; i.e., lower the probability.

Generally speaking, we would normally include a review of labeling for higher severity risks in a usability (summative) study. We still wouldn't lower the risk, but it does provide insight that the information is understandable and the user understands the correct actions to take.

A couple more examples:
Some labels can be useful and heeded if they are common and directly useful: “Refrigerate after opening” is one that is quite useful and followed.
Some labels are punitive: the warning on cigarette packages for example. No one really believed that people would heed the warning but legally it served as a public statement that cigarette manufacturers knew their product was harmful.

IF you can prove that a label is helpful AND you can demonstrate how much the occurrence of mis-use is reduced, THEN you can reduce the occurrence rating. But don’t just guess, or use engineering judgment. These non-data based decisions are usually just wishful thinking that overstate the reduction in occurrence.

However a label is never a bad idea and can provide some legal shelter from mis-use claims…

Yes. We had to apply a laser warning for a printer that used a laser for an alignment of the film before printing. The person really needed to stick their face in the printer for the risk to manifest but we did it anyways.

Ed Panek said:

Yes. We had to apply a laser warning for a printer that used a laser for an alignment of the film before printing. The person really needed to stick their face in the printer for the risk to manifest but we did it anyways.

Click to expand...

That is government mandated though.

If there is no other option to reduce risk than labelling is the last choice to do so and risk reduction through information is seen as residual risk. If labelling is state of the art für such a device, then this is acceptable, e.g. the flammability risk of an alcohol based disinfectant couldn't be reduced and labelling with a hazardous symbol is state of the art (Remember risk can be seen as acceptable, if the state of the art is achieved). But if there are other options to reduce the risk of your product than labelling, which are state of the art, then labelling as risk reduction measurement wouldn't be acceptable. Thus I would look for the state of the art, which also includes other regulatory requirments than MDR related ones.