Thanks for your reply!
..so do you mean to to decide if the risk is acceptable in terms of the medical benefit outweighing the residual risk? Sorry if this is a stupid question!
It's not a stupid question, it comes from the historical "bad" use of the risk matrix.
And not, it's not related to the benefit.
The problem is how you define your criteria for acceptability. The thing is, risk acceptability is not related only to severity and probability of harm, which is the definition of risk. It has to take into consideration other things (the current ISO 14971 mentions "criteria are based upon applicable national or regional regulations and relevant International Standards, and take into account available information such as the generally accepted state of the art and known stakeholder concerns" and we are expanding this in the revision).
One generic example of criteria (this is a step-by-step I created to show as an example during the revision of ISO 14971, but as it won't be used in the text, I think it would interesting to have it here as an example. Also, the example is not quite correct in some parts, but I'm trying to show the concept here):
1 - Determine the need to establish risk criteria. What risk decisions will be aided by the criteria?
Risk criteria will be used to show that risk related to medical device X are acceptable. The criteria will aid in the decision of:
- When identified and reasonably foreseeable individual risks related to the medical device are to be considered acceptable
- When the aggregate risks related to the medical device are to be considered acceptable
2 - Determine the risks to be addressed (Risk to what ?)
Risk of harm to patients, user and, where applicable, other persons
3 - Determine/classify the populations to be addressed
Individual patients, user, or other persons. The device is not expected to cause harm to groups of person at a time.
4 - Determine which risk criteria to develop.
Individual, societal, or other type of risk?
Will there be a criterion defining de minimis risk?
Individual risk - the risk to a person in the vicinity of a hazard. In particular, the individual risk type to be used will be the maximum individual risk (the individual risk to the person(s) exposed to the highest risk in an exposed population), for patient……because blahblahblah.
There won´t be a de minimis criteria.
5 - Determine philosophy for continuing risk reduction (e.g., ALARP, ALARA, AFAP)
ALARP will be used as philosophy for risk reduction.
6 - Develop individual risk criteria (based on policy)
Example Develop individual risk criteria (based on policy)
1 - Identify possible criteria or basis of criteria in applicable regulatory requirements
Applicable regulations require that risks are acceptable when weighed against the benefits to the patient and are compatible with a high level of protection of health and safety. There´s no different or contradicting expectations in the applicable regulations.
2- Identify possible criteria or basis of criteria in applicable international standards
Device has applicable international safety product standard that can be used as basis (ISO XXX or IECXXX). Other applicable standards exist that can be used as basis for criteria for different aspects of device risks (example, biocompatibility)
3 - Identify state-of-the-art regarding intended use of device, including medical alternatives
There´s several similar devices in the market. Alternative treatment (example drug, manual, etc.) is considered safer but slow in recovery time when compared to devices (and for XXX reason it´s important to have a quicker recovery time)
4 - Identify known stakeholder concerns, including public perception of risk
Patients tend to think that risks related to the device are commonplace (the same risk as being punctured by a needle), and also are more willing to tolerate risks if treatment has quicker recovery time than current treatments
Users generally think that device do pose risks, but are more willing to accept it if protective equipment is required to be used with device so as to diminish involuntary risk.
5 - Identify probability /severity criteria
The following criteria is to be used as a basis, and shall take into consideration the additional criteria mentioned below
See attached image.
Maximum individual risk to patients (fatality /year): 10-3
6 - Identify additional criteria
Risks shall be reduced wherever practicable. This includes cases in which, even if the risk is already deemed acceptable by the criteria, the cost to include the additional risk control is considered so low as to be essentially free.
Further development should not pose any incremental risk.
No single failures/errors should lead to an accident.
Device shall comply with requirements from international standards X, Y, Z. Unless an evaluation shows otherwise, compliance with those standards shall be used as argument to consider the risk as reduced to an acceptable level (ALARP good practice argument).
Risks to patients than can be considered ALARP if:
- there´s a good practice argument which demonstrate that risk control measures comply with relevant good practice and similar solutions in similar devices. This situation can be accepted by the authority of design engineers.
- there´s a qualitative first principles argument based on common sense or professional judgment that weighs possible risk reduction against the gain in recovery time. This situation can be accepted only by the authority of the device design project leader and risk management leader.
- there´s a quantitative first principles arguments based on a Cost Benefit Analysis (CBA) that weighs possible risk reduction against the gain in recovery time (this case would only be required if individual risk is more than 10-4). This situation can be accepted only by the authority of the device design project leader and risk management leader, and top management.
A risk to the user may be acceptable if it can be justified that the use of a protective equipment which is required to be present will reduce the individual risk to less than 10-3
Risks more than 10-4 can only be accepted if it a risk-benefit analysis shows that the total benefits of the device outweighs the aggregate risk profile. This situation can only be accepted by top management.