Considerations When Using Postmarket Data in Risk Management

By Roberta Goode, Julie Cabezas
September 8, 2015


• Categories: 04-Product Design & Development, 05-Quality/Regulatory, RC - Medtech Regulatory Resource Center

The following article is Part II of “The Integration of Complaint Handling and Risk Management”.

The key to integrating complaint handling and risk management is to ensure feeder processes are in place to bring data from the post-market surveillance group to the risk management tools, in the form of:

• Adding new or previously unforeseeable hazards, hazardous situations, and harms into the risk management tools for analysis, evaluation and control
• Evaluating the new and existing risks against acceptance criteria on intervals commensurate with product risk
• Determining if the risk is maintained within acceptable levels, considering state-of-the-art and known stakeholder concerns

REGISTER HERE for MedTech Intelligence’s Integrated Complaint Management Conference! September 17-18, 2015 | Washington, DC. Attend in-person or VIRTUALLY. The challenges with this have to do with designing methodologies that assess the updated risk information in a way that accurately and appropriately indicates where risk mitigation action is required. Questions may arise, such as:

• How often do we update our frequencies of occurrence?
• How do we use complaint data that is not substantiated by a product investigation, if the product is not returned?
• Should we have separate evaluations in vulnerable patient populations like the sick, elderly and pediatrics? If so, are there different state-of-the-art criteria for different patient populations?
• What are the risks of multiplying serial fractions in quantified frequencies of occurrence?  If we use qualitative occurrence scales, how do we use them correctly?
• How do we account for patient perceptions of current and future risk and current and future benefit? For example, people generally prefer present benefits over future benefits and future harms over present harms, so time-preference may need to be captured when risks and benefits occur at different times.¹

VIDEO: Roberta Goode discusses the responsibility of suppliers in a medical device recall at MedTech Intelligence’s conference on health hazard evaluations, risk assessment and recalls. VIEW VIDEOIn our consulting practice, we have noted that risk evaluation methodologies using post-market data should be designed to address concerns such as:

• Low-frequency, high-risk outcomes, such that they are not underreported or underestimated by manufacturers multiplying serial fractions for frequency of occurrence
• Low-frequency, low-risk outcomes, such that they are not under-mitigated, considering vulnerable populations individually, where applicable. (Note: FDA, for example, has a low tolerance for low-frequency, less-threatening events in vulnerable populations such as the elderly and children.)
• The identification of systemic trends, even at low frequencies, rather than solely focusing on quantitative frequencies of individual hazards or failure modes at a given point in time
• Risk benefit analyses and decision-making processes related to recalls. For example, not making a correction in the field, despite an unfavorable risk benefit analysis, due to the belief that product would be too scarce, impacting public benefit, is most often neither accurate nor acceptable. (For example, the way FDA looks at this situation is to ask not what percent of the market share the company has, but rather, if you had to recall, could your competitors ramp up their production to fill that gap? This is a far different question than most manufacturers are asking.)

Final Thoughts

Navigating the waters of complaint handling amidst changing risk management standards and best practices can be challenging. By focusing on mitigating risk as far as possible, considering generally accepted state-of-the-art, and using post-market data to monitor risk, it is possible to strike a balance that prioritizes patient safety and minimizes organizational burden. Through our work with the technical committee and our clients, we’ve established best practices for how to balance these apparently competing elements.

Just like you, we anxiously await updates to the ANSI/AAMI/ISO 14971(2007)/(R)2010 standard. Until then, let’s carefully watch the notified bodies and FDA to see how they interpret existing Annexes Z of EN ISO 14971:2012 and implement best practices as noted herein. If you’re struggling to implement a compliant risk management solution, we can help.

Reference

1. Rothstein, Z. A. (May 20, 2015). “Re: White Paper on Risk Principles: A Post-Market Perspective,” Official commentary by AdvaMed to AAMI.