Medical Device Risk Management P1 and P2 Explained

This article presents a strategy that medical device manufacturers can use to assess, analyze, and control risks related to the use of medical devices, compliant to the latest revisions of ISO14971. Copyright 2022, Saegert Solutions Inc.

The ISO14971 standard for the Application of Risk Management to Medical Devices defines risk as the “combination of the probability of occurrence of harm, and the severity of that harm.” In Annex C of the standard (Fundamental Risk Concepts), the probability of occurrence of harm can be expressed as a single probability (P) or as the mathematical product of two separate probabilities, P1 and P2. This article describes opportunities associated with adopting the two-probability approach to risk analysis.

In the two-probability approach to risk, the first probability term (P1), describes the probability that the presence of a hazard will result in a hazardous situation. The second probability term, (P2) describes the probability that a hazardous situation will result in harm.

The alternate approach is typical of that typically used for FMEA, where the probability of occurrence focuses on the occurrence of the hazardous situation, (i.e.: P1), while the severity ratings reflect most severe effects (to be conservative), assuming that if the failure mode occurs, then the stated effect will also occur, with 100% certainty (i.e.: P2 = 100%).

A previous article described how this approach is inefficient, leading to unrealistic expectations when conservative risk profiles demand corrective actions for relatively common product or process failures. These investigations can be difficult to justify, particularly when the observed effects appear much less severe than those predicted by the analysis.       

In reality, a single hazardous situation can result in more than one type of harm, or in harms of similar type, but with differing levels of severity. A hazardous situation may result in bleeding, but the Bleeding Academic Research Consortium (BARC) defines five types of bleeding, each with different levels of severity. In this situation, different values of P2 can be assigned to each type or severity level of bleeding that may be anticipated. The table below lists generic harm types and severity levels as an example:

Harm Type        Description          Severity          P2

Type 1               Negligible                      1                      0.60

Type 2               Actionable                    2                    0.20

Type 3               Serious                           3                    0.15

Type 4               Severe                            4                     0.04

Type 5               Fatal                               5                      0.01

Note that in this example, all values of P2 associated with the hazardous situation sum to 1.00. This indicates that when the hazardous situation occurs, at least one type of harm will occur. If there is a probability that ‘No Harm’ occurs, this too should be assigned its own P2 probability value relative to the other known or anticipated types of harm.

When quantifying risk, the probability of a hazardous situation (P1) and the probability of a resulting harm (P2) produce a specific probability of occurrence (Pocc = P1 x P2).

Summarizing Risk

Just as a single hazardous situation can result in different types and severities of harm, a harm of a specific type and severity may result from more that one hazardous situation. A rule of thumb in risk analysis is that all hazardous situations occur independently of one another; the probabilities of occurrence (Pocc = P1 x P2) for each type of harm can be added together to summarize the overall probability of occurrence for each type of harm, resulting from all anticipated hazardous situations.   

This summary of harm type, severity and overall probability of occurrence can be useful for communicating risk profiles, benchmarking risk profiles between competing devices, or comparing risk profiles between a new device and an incumbent procedure.  

Example:

Consider a manufacturer developing a transcatheter procedure. Intended to replace an ‘open-heart’ surgical procedure, the same harm types and severities have the potential to apply in both cases. However, the overall occurrence rates of each type of harm will likely be different, based on the differences between the procedures. If new procedure can demonstrate that the occurrence rates of each type of harms is lower than the incumbent, then the benefit is of the new procedure is quantified.    

Reducing Risk by Reducing Severity   

Typical approaches to risk reduction usually focus on designing and controlling the failure rates for events that lead to hazardous situations (controlling P1). An FMEA specifies a failure mode and occurrence rate, and controls are defined, verified, and implemented, to reduce that occurrence rate. While efforts to reduce P1 offer the greatest benefit, reductions in the severity of the risk profile can also be quantified if the values of P2 for high severity harms are reduced compared to P2 values associated with lower severities.     

If a hazardous situation results in a high severity harm with a P2 of 80%, and a low severity harm with a P2 of 20%, experience may demonstrate that P2 for the high severity harm is reduced to 40%, while P2 for the low severity harm increases to 60%. Even if P1 remains unchanged, the occurrence rate (Pocc = P1 x P2) of the high severity harm has been demonstrably reduced, lowering the overall risk profile.    

Matching Harms to Hazardous Situations

When preparing a risk analysis, once hazards and hazardous situations have been identified, engage medical advisors and practitioners to determine the types and severities of harms that can result from each hazardous situation. Consult clinical records for similar devices and procedures to gather information about harm types and severities associated with the device or procedure. These form the benchmarks to drive risk reduction efforts.  

For medical devices in development, the list of types and severities of harms associated with the use of the device should be part of the risk file. To ensure accuracy and consistency when validating the device, the same list of harms used to prepare the risk assessment should be used to conduct the clinical evaluation.

Prepare for Post-Market Surveillance

In post-market surveillance, the pre-established list of harms can be used to match reported symptoms with known hazardous situations to help investigate root and contributing causes, and to guide corrective actions. Unexpected or unobserved harms can also be added and re-assessed during periodic risk reviews, as necessary.

This approach can be applied to new products in development, or to existing risk files retroactively, as I have done in the past. Often, the lists of harms and their relative proportions are known, and the organizations risk files can be updated relatively quickly. What's important is a clear and consistent approach, with well-developed methods and techniques. Based on my prior experiences, the results can be transformative.

About the author:

Alex Saegert is founder and principal consultant of Saegert Solutions. He is an ASQ certified reliability engineer (CRE) and supplier quality professional (CSQP), with a specialized ASQ credential in risk management. He has over 20 years experience engineering quality, reliability and safety into products from such diverse fields as medical devices, hydrogen fuel cells, alternative energy powertrains for cars, trucks and locomotives, and in the nuclear power industry. He is licensed as a professional engineer in the provinces of Alberta and British Columbia, Canada.