NCHR Testimony at 2016 FDA Workshop on Role of Hospitals in Modernizing Evidence Generation for Device Evaluation

Thank you very much for the opportunity to speak today.  My name is Dr. Jay Ronquillo and I am speaking on behalf of the National Center for Health Research.  I am a physician who trained at Massachusetts General Hospital, have two engineering degrees from Cornell, a Master of Public Health from Harvard, and a master’s in biomedical informatics from Harvard Medical School.  These are the perspectives I bring with me today.  Our research center analyzes scientific and medical data and provides objective health information to patients, providers, and policy makers.  We do not accept funding from the drug or medical device industry and I have no conflicts of interest.

Hospitals are expected to play a key role in evidence generation and adverse event reporting for the National Evaluation System for Health Technology.  While the collection and analysis of real-world data will be important for post-market surveillance of medical devices, the quality of information being collected and used for decision-making must be focused on making patient safety a priority.

Software is a critical component of many medical devices, from implantable cardiac pacemakers, to drug infusion pumps, to health IT like electronic health records and clinical decision support.  However, software-related errors for medical devices are complex.  They can be hard to identify and clearly separate from other types of problems, such as human error or mechanical failure.  Software vulnerabilities also represent a growing cybersecurity concern that could directly affect individual patients and even populations.  As a result, accurate risk assessment and adverse event reporting will require clear guidance regarding how all types of software malfunctions are measured, monitored, and could impact patient care.  We currently do not have that guidance today.

Similarly, real-world evidence collected from new data sources, such as electronic health records (or EHRs), is expected to be central to the National Evaluation System for Health Technology.  However, EHRs in their current state are not very usable or user-friendly, making it difficult for hospital staff to interact with data in a structured format without hindering patient care.  For example, the FDA medical device database has reported several thousand EHRs in oncology and anesthesiology that were recalled due to software flaws over the last five years.  These defects reportedly caused incorrect drug dosage calculations or displayed medical data for the wrong patient.  Unless there are robust regulatory safeguards for improving the design, development, and implementation of EHRs, poor system usability and interoperability will limit the quality of surveillance data being collected as well as any conclusions or decisions based on that data.

In summary, we recognize the potential value of a National Evaluation System for Health Technology that leverages real-world evidence from hospitals.  However, the complexity of software errors, along with the limitations of current EHRs could limit the quality of evidence being generated and used for post-market surveillance of medical devices.  We recommend clear guidelines on the regulation of health IT, and stronger safeguards to ensure that all medical devices, along with their software and data, remain safe and effective for patients.

Thank you again for the opportunity to speak today and for consideration of our views.