510(k) Raw Data from Studies - 21 CFR Part 11 Compliance

invitro_spain

Involved In Discussions
Hi Folks,

I need help for this easy or not so easy question. During the premarket submission for a device Class II-Moderate risk, you have to include a summary of your bench studies and raw data (for example, spreadsheets with the data from your studies).

In this case, Should these spreadsheets comply with the 21 CFR Part 11?

Could you advise me about the most suitable way to comply with this regulation?

Thank you in advance
 

yodon

Leader
Super Moderator
If the spreadsheet is maintained as an electronic record or if you applied a signature electronically as required by predicate rule then yes!

Further, a spreadsheet application should also be validated for its intended use. If the spreadsheet is not validated then the data collected COULD be called into question.

A lot will depend on your inspector. The "enforcement discretion" can be anywhere in the spectrum. One option is to make a snapshot of the spreadsheet (PDF) and control that as the record. (There's still potentially a Part 11 concern but it's greatly reduced since alteration would be much more difficult. And the validation concern still exists, too.)
 

invitro_spain

Involved In Discussions
One option is to make a snapshot of the spreadsheet (PDF) and control that as the record. (There's still potentially a Part 11 concern but it's greatly reduced since alteration would be much more difficult. And the validation concern still exists, too.)

Thank you yodon. Sorry for the delay in replying. During the submission, you have to submit a hard copy and a electronic copy of your 510(k) using "esubmitter". I suppose if you upload a spreadsheet to this application you may obtain a pdf controlled version of your initial spreadsheet. Anyway, you can include a hard copy version signed in paper with your 510(k). Would be this methodology correct?
 

Mark Meer

Trusted Information Resource
...I suppose if you upload a spreadsheet to this application you may obtain a pdf controlled version of your initial spreadsheet. Anyway, you can include a hard copy version signed in paper with your 510(k). Would be this methodology correct?

We use this method - spreadsheets, once finalized, are exported to PDF, and then signed. Alternatively, the exported data is attached to a test-report as an appendix, and it is the report that is ultimately signed and controlled.

This should be done at the time the data is done (record closed). The original spreadsheet can be maintained/archived, but as far as your records go, the signed PDF of the data is the "official" record.

Signatures are intended to attest to some activity (e.g. review, approval..) at the time of the activity. So, exporting a PDF and signing it long after the spreadsheet was originally finalized, just for the purposes of Part 11, and a 510(k), may be called into question.

Alternatively what you can do is to collect required signatures elsewhere (e.g. via a separate approvals form). In such case, all you have to do is:
1. Make sure that the dates on the signatures form are after the last modification date of the spreadsheet.
2. The spreadsheet is protected somehow from modification (or, at least, a way to identify modification - e.g. via timestamps, version control), after the necessary signatures have been collected.
 
Top Bottom