Although not a mandatory requirement of ISO 14971, it is a common practice in the medical device industry to use the P1, P2 approach for estimating the probability of harm (POH). In many cases, there is a lot of confusion and inaccuracies when using this approach, especially when the post-market data is not complete and/or is of poor quality. In this episode of our weekly live discussion, we are focusing on this topic and discussing some of the common challenges.

Highlights from Episode 4: Challenges of Using P1, P2 Approach 

 Check out the video below which opens our discussion with a brief slide presentation on this topic. ISO 14971:2019 requires that risk of each potential harm associated with the use of a medical device be estimated through a combination of the probability of harm (POH) and severity(S). P1, P2 approach is one of the ways we can estimate the POH. However, it is not a mandatory to use this approach in every situation. 

As defined in ISO 14971:2019, P1 is the probability of a hazardous situation occurring, and P2 is the probability of hazardous situation leading to harm. It may be appropriate, and highly beneficial, to use the P1, P2 method when sufficient amount of high quality data is available. In practice, however, the required level of detail is missing from available data and it is generally not possible to accurately link device or use related events to hazardous situations and harms. As an example, it is not uncommon to find no device or use related issue in a majority of medical device reportable (MDR) injuries in FDA's MAUDE database. 

In these situations, using the P1, P2 approach under the impression that it is mandated by ISO 14971:2019, and expected by notified body auditors, creates a lot of confusion, inconsistencies and inaccuracies in risk estimation. 

In the introductory slides, we highlight two key practical challenges in estimating P1 and P2:

1. Device failure modes do not (always) equal to hazardous situations and exact sequence of events is generally not known to accurately estimate P1 and P2.
2. There is generally no direct link between a specific hazardous situation and a specific harm. 

Following the opening presentation, we discussed the following key points and comments/questions from the attendees:

• Incorrect use of postmarket data for estimating P2 
• Limitations of postmarket data for estimating P1, P2
• How to properly estimate denominator for probability calculations
• Understanding relationship between postmarket data and P1, P2
• How to handle the scenario when no postmarket data is available
• Tips for handling hazards, hazardous situations for estimating P1
• Review of key points

Watch the video above to learn more about these key points and common challenges.

Join our weekly live discussions!

In these weekly discussions, we focus on specific questions or topics that practitioners of risk management find challenging from a practical, day-to-day point of view. Our focus is not on regulatory compliance, but on sharing common problems and best practices. 

These sessions generally last for about 30 minutes. We open with a short slide presentation to introduce the topic or a specific question of interest. You can participate by sharing your question or comment via live chat at any time. However, there are a few key points to note:

• This is an open forum; therefore, do not share any proprietary or business sensitive information
• Our discussion is meant to share information and knowledge purely for the purpose of education

 How you can participate

These discussions will occur via live streaming on LinkedIn and YouTube. If you already follow us on LinkedIn or YouTube, you will get a notification as soon as we go live. Otherwise, connect with us here on the Achieve LinkedIn Page or subscribe to our YouTube channel. During the live meeting, you can use the chat/comments section to ask a question or share comments and best practices. 

Special Offer 

Visit our ACHIEVE website for valuable free offers!