How to Conduct a Known Problems Analysis (KPA) for NMPA Compliance

By Yvonne Limpens and Carolyn David 

Introduction 

Why a known problems analysis is crucial for NMPA compliance 

A known problems analysis (KPA) is a human factors engineering (HFE) activity that aims to identify use issues and user interface flaws with devices currently on the market. Regulators require medical device manufacturers to conduct a KPA to investigate use problems associated with devices that are similar to a device under development (i.e., that have similar user interface elements, user interactions, and/or user groups).  

Many manufacturers will be familiar conducting this process to align with the U.S. Food and Drug Administration (FDA) or IEC62336-1 requirements and will be well versed in identifying appropriate databases, identifying an appropriate time frame, and scoping the KPA according to the countries in which they intend to market the product.  

Notably, in March 2024, China’s National Medical Products Administration (NMPA) Center for Medical Device Evaluation (CMDE) released a finalized version of its guidance pertaining to the application of HFE / usability engineering (UE) to medical device development. The NMPA guidance differs from IEC62366-1 and FDA guidance in a few areas, one of these being the expected scope of a KPA. Note that the NMPA refers to this activity as an “analysis of the post-marketing use of similar medical devices.” In this article, we discuss how the NMPA’s expectations for a KPA differ from the FDA and IEC-62366-1, and we highlight how to most efficiently conduct an NMPA-compliant KPA.  

Key NMPA requirements for conducting a successful KPA 

Below we list the most notable expectations for conducting a KPA that meets NMPA’s expectations per their 2024 finalized HFE guidance: 

• Global reach. NMPA’s guidance specifically calls for a global search covering the world’s major medical device adverse events and recall databases in addition to domestic databases (i.e., Chinese Adverse Event database). Unlike IEC62366-2 and the US FDA’s final HFE guidance, the guidance does not provide a list of suggested databases.  

• Search span. NMPA sets a default search span of 5 years. Alternative search spans might be appropriate but require justification.  

• Recalls.The KPA is not limited to just adverse events because NMPA also expects manufacturers to include recalls of similar medical devices. The report should reflect national differences in recall classification, as needed. 

• Report outline. NMPA reflects which report sections it expects. Specifically, the report should reflect a description of the device, the source range, search span, search dates, search terms / inclusion selection criteria, staff involved, search results, and analysis of the identified use events and recalls.   

• HFE professionals. NMPA expects that staff involved in analysis activities like these have relevant knowledge and work experience in usability engineering and recommends prioritizing HFE professionals over other staff.   

Best practices for conducting an NMPA-Compliant KPA 

Conducting a global KPA can be a challenging and time-consuming effort. Below, we list a few tips based on our experience conducting global KPAs per NMPA’s expectations.  

• Leverage existing KPAs. If you conducted a (recent) KPA per FDA’s expectations or per the requirements set forth in IEC62366-1, then we recommend leveraging that KPA first before expanding to other regions (e.g., China). Note that additional effort will likely still be warranted based on how recent the original KPA is and whether it meets the additional criteria reflected above.  

• Utilize a diverse range of data sources. Consider incorporating data from clinical trials, post-marketing surveillance, literature reviews, and regulatory databases (including but not limited to, FDA’s MAUDE and MEDSUN databases, as well as those listed in IEC62366-2).  

• Register for China’s adverse event database. China’s adverse event database is not accessible to the public. Manufacturers, however, have access to this valuable resource, which should be leveraged to enhance the depth and breadth of your KPA.  

• Define and stick to your search terms. It’s easy to lose track when conducting a KPA, so clearly define your search terms upfront. It is crucial to use these terms consistently (and their accurate translations in other languages, as appropriate) throughout your analysis to ensure accurate and reliable results.  

• Categorize similar problems. We recommend analyzing known problems to identify recurring themes or patterns. Group similar issues together based on common characteristics, such as the action taken or affected product component. This streamlined approach will enhance the efficiency of your analysis.  

• Seek translation support. When conducting a global KPA, you will likely be confronted with adverse event databases in languages that you do not speak. Seek translation support as needed to accurately translate search terms and/or interpret use events. Also, NMPA documentation eventually needs to be submitted in Chinese. As such, engage individuals who can translate documentation as needed.  

• Compare recall classifications. It may be challenging to compare recall classifications from across the globe noting regulators may adhere to different classification levels. As such, it might be easiest to focus on comparing and classifying the types of issues and the resulting corrective measures instead. 

• Provide additional post-market context. If the device being developed already has a direct predecessor on the Chinese market, your internally collected post-market data quality and quantity might be greater than publicly available post-market data for competitor devices (noting that adverse events are underreported). Additionally, the predecessor device might account for a larger market share compared to competitor devices. Consequently, it may appear that the direct predecessor device has higher occurrence of use-related events compared to its competitors. In such cases, it can be helpful to explain this in KPA documentation.  

Navigating NMPA’s HFE guidance with a strategic KPA approach 

While NMPA KPAs share similarities with those required by other regulatory agencies, they also present unique challenges. Key differences include the requirement for a global perspective, specific search span duration, inclusion of recall events, distinct report outlines and the involvement of HFE professionals. By understanding these nuances and adhering to NMPA guidelines, manufacturers can effectively navigate the regulatory landscape in the Chinese market.  

Contact our team for help meeting the NMPA’s HFE requirements, including how we can leverage our exclusive partnership with CMTC, a Shanghai-based state-owned testing institute for medical devices, to conduct robust usability testing in China. Also consider signing up for a complimentary account with OPUS, our team’s software platform that provides HFE training, tools, and templates.  

Yvonne Limpens is a Managing Human Factors Specialist, and Carolyn David is a Human Factors Associate at Emergo by UL.