FDA’s MAUDE (Manufacturer and User Facility Device Experience) database has long been a resource for industry stakeholders to enhance medical device safety and quality. Understanding and analyzing MAUDE data helps us meet compliance obligations and serve a greater purpose: safeguarding patient health and optimizing device performance. In this article, we will discuss best practices for MAUDE data utilization and constraints to keep in mind during analysis.

Acknowledging the Source

The FDA deserves considerable recognition for its role as perhaps the most transparent and informative health agency worldwide, making valuable data, including the MAUDE database, accessible to the public despite the significant effort and cost involved.

Understanding MAUDE Data

Medical Device Reporting (MDR) regulation 21 CFR 803 requires manufacturers, importers, and clinical care centers to report certain device-related adverse events and product issues.[1] According to the FDA:

Manufacturers must report incidents where a device might have caused or contributed to a death or serious injury, as well as any malfunction that could likely recur and lead to such outcomes.
Importers are required to notify the FDA and the manufacturer if a device could have caused or contributed to a death or serious injury. They must report to the manufacturer alone if their devices have malfunctioned in a way that could likely cause or contribute to a death or serious injury if repeated.
Device User Facilities, such as hospitals and outpatient facilities, must report suspected device-related deaths to both the FDA and the manufacturer, and serious injuries to the manufacturer (or the FDA if the manufacturer is unknown).

These reports, as well as voluntary reports that come from the public, comprise the MAUDE database.

Considerations and Limitations

It is important to recognize that due to the nature of the reporting process, MAUDE reports are not verified by FDA, may contain inaccurate or incomplete information, and do not contain information regarding the frequency of device use.[2] In addition, although reports should be filed within specified timelines (5, 10, or 30 days in accordance with § 803.53(b)), delays are common and unavoidable when initial data is incomplete or not properly reported to the manufacturer. Keeping this in mind, MAUDE data remains an important primary data source for quality monitoring, postmarket surveillance, risk management, and management review.

Analyzing Data & Best Practices

A systematic approach to MAUDE monitoring and analysis is required for effective and actionable outputs. Continuous monitoring along with a regular cadence of reviewing, reporting, and disseminating information to the relevant stakeholders is best practice.

When analyzing MAUDE data, it’s important to consider both quantitative and qualitative aspects. Quantitative analysis involves tracking the number and frequency of reported events over time, which can help identify sudden spikes or gradual increases that may indicate emerging issues. For example, a significant increase in reports related to a specific device or failure mode, or reported from a particular geography may signal the need for further investigation and potential corrective actions. Qualitative analysis, on the other hand, focuses on the content of individual reports, such as the description of the event, patient factors, and device-related details. This can provide valuable insight into the nature and severity of the issues reported and inform risk management records and activities.

Different roles within the company should monitor and analyze MAUDE data for different outputs. The three most common roles and the associated use cases include:

Global Compliance monitors organizational and product trends and performs benchmarking. Specifically, Global Compliance monitors the number of MAUDE and types of reports over time benchmarking against competitive organizations and product lines. These individuals also use the data to identify industry-wide product quality issues. The output of these analyses is used in management review and as early warning signals of quality trends. In Redica’s linked data models, inspections linked to postmarket events are used to ensure continuous inspection readiness.
Postmarket Reporting is performed in compliance with quality system procedures (ISO 13485 and ISO/TR 20416) and to regulatory authorities. Product-specific MAUDE reports and trends should be utilized to contextualize types and occurrences of events at the product level. In the Redica database, specialized filtering, search, and visualizations are provided to ensure that appropriate product and procedural benchmarking can occur.
Risk Management is performed for products during design controls and requires the identification of hazards and harms associated with products. The product family, similar products, and similar technologies can be used as filters to identify all potential applicable hazards and harms and to consider the apparent prevalence of the risks. MAUDE trends and clinical risk can be evaluated with specialized visualizations and filtering.

Redica Systems provides data visualization tools such as charts, graphs, and dashboards, to monitor and summarize the large volume of data contained in MAUDE and highlight trends or outliers. We utilize methods such as time-series analysis and correlation analysis to identify relationships between different variables and assess the significance of observed patterns. Real-time data pipelines ensure that data is available upon publication and continuous monitoring can occur.

Figure 1 | Screenshot of Redica Systems – Key Performance Indicators – FDA Recall Trends

In addition, Redica provides detailed and customizable filters to identify the data that matters to your company and products. These include procedure-based filters, text searching, and product, site, and organizational linkages. These tools are crucial for extracting insights rather than noise from the overwhelming amount (more than 17 million records) of MAUDE data.

Challenges and Solutions

When interpreting MAUDE data, it’s crucial to consider the context and limitations of the information provided as well as the queries being made of the data. As mentioned earlier, MAUDE reports are not verified by FDA and may contain incomplete or inaccurate information; context is key for interpreting the results. Additionally, the number of reports for a particular device does not necessarily reflect the actual incidence of adverse events, as reporting rates can be influenced by factors such as device usage, market share, and reporting practices.

With regard to querying the data correctly, ensure the right filters are being used to generate useful outputs. For example, a request for site-level data can give you insights about historical and current reporting volumes and may be indicative of quality issues, but it won’t give you information about a particular product or event. Similarly, MAUDE reports cannot in many circumstances be linked to other events such as recalls or inspections until a proper qualitative analysis is performed. These issues in particular can be difficult to explain to industry outsiders, and should always be explained in context with deep understanding of the reports and possible contributing events.

It’s essential to use tools that allow for identification and visualization of the “correct” (i.e. meaningful and useful) data. Then we must supplement that data with other sources of information, such as internal complaint data, scientific literature, and expert opinions, to gain the most comprehensive understanding of the issues at hand.

Conclusion

MAUDE data is a valuable resource for medical device professionals seeking to enhance product safety, quality, and compliance. By understanding the reporting requirements, analyzing the data effectively, and applying best practices, we can leverage this information to drive meaningful improvements in patient care and device performance.

It is crucial to approach MAUDE data with a critical eye and a deep understanding of its regulatory framework. With the appropriate use of MAUDE data in conjunction with other sources of information we can gain deep new insights into product quality and safety. By proactively monitoring adverse events, identifying trends and patterns, and investigating potential issues, we can take swift action to mitigate risks, improve device design, and enhance patient outcomes. In addition, with a commitment to data-driven quality improvement and regulatory compliance, we can build trust with patients, healthcare providers, and regulatory authorities.

[1] This post does not discuss the Voluntary Malfunction Summary Reporting (VMSR) program which allows manufacturers to report certain malfunctions on a quarterly basis as summaries.

[2] A note to industry outsiders: MAUDE data is often misunderstood by those outside of the medical device industry, and we should be clear that a MAUDE report is not necessarily evidence that a device caused an adverse event. Incomplete and conflicting information is reported to manufacturers from many different sources and companies aim to report this information as best they can pursuant to the regulations. Each medical device manufacturer and facility has procedures specifying criteria for reporting within the definitions provided by FDA. These procedures are routinely inspected by FDA to ensure that companies remain compliant with the regulation. Most companies report events “conservatively” meaning that over-reporting of events is common, particularly in products that are used in critically ill or elderly patients. These patients, regardless of treatment or device, often experience many adverse events due to comorbidities which are reported regardless of device causation or contribution.