Additional Information (AI) Requests are a part of FDA’s procedure to provide a medical device company with an opportunity to submit additional information missing from the initial submission. This information is deemed necessary by the FDA to be able to approve the medical device for marketing in the US. Here’s our first example of an AI Request:
FDA Additional Information Request – Missing Documentation Across Entire Submission
You provided software documentation in the software section of your eSTAR submission.
However, as noted in several deficiencies in this letter, your software documentation lacks traceability and detail and overall indicates that there is poor design control of your software. None of the software documentation you have submitted adequately describes or tests your software as it is described and tested in other parts of your submission, such as cybersecurity and labeling.
Therefore, your software documentation lacks the traceability and adequacy described in the FDA guidance document “Content of Premarket Submissions for Device Software Functions”. This information is important to assess the adequacy of the implementation of the software requirements in terms of intended use, functionality, safety, and performance, which is important to determine the testing needed to mitigate relevant software-related risks. Adequate testing and risk mitigation are important because software with unmitigated risks could result in adverse health effects for the patient.
Please update the entirety of your software documentation and testing to be complete, detailed, and traceable to the rest of your submission and properly reflect the software and software version used in your device system.
In this specific example, the medical device company has quite a bit of work to do since the FDA found many issues with the company’s submission; enough that the FDA asked them to update the entirety of the firm’s software documentation and testing.
The FDA is looking for a coherent, complete, detailed, and traceable set of documents. Per the above findings, the FDA also observed challenges with configuration management (software version control in particular).
