Records - Do's and don't' of record entries (FDA - 21 CFR 820)

We all are familiar with the do's and don't of record entries for cGMP.
Don't scribble out, don't use white out, initial and date changes etc.
The question came up in a discussion as to where specially is this stated in the requirements.
We have searched CFR 820 and can't find anything.
Suggestions?

Many requirements come from good laboratory practices. 21CFR58

Sec. 58.130 Conduct of a nonclinical laboratory study.
(e) All data generated during the conduct of a nonclinical laboratory study, except those that are generated by automated data collection systems, shall be recorded directly, promptly, and legibly in ink. All data entries shall be dated on the date of entry and signed or initialed by the person entering the data. Any change in entries shall be made so as not to obscure the original entry, shall indicate the reason for such change, and shall be dated and signed or identified at the time of the change. In automated data collection systems, the individual responsible for direct data input shall be identified at the time of data input. Any change in automated data entries shall be made so as not to obscure the original entry, shall indicate the reason for change, shall be dated, and the responsible individual shall be identified.

It's all about data integrity and ties in with record controls. If the information is questionable, the FDA may refuse it. Google ALCOA and ALCOA+

Everywhere in 21CFR820 it says "the manufacturer shall establish and maintain...……" procedures, records, etc, I would take the word "maintain" to include the documents must be attributable, legible, etc. (ALCOA+), and follow good documentation practices.

Thanks everyone for the information, we practice the guidelines for documentation but were never aware of where it was stated. As a converter we don't have any involvement with the pharmaceutical community where many of the guidelines initiate.
Thanks again

I have some similar questions about this topic, specifically "Good Documentation Practices". My company has a very similar situation to the one that SMBIZZQA is describing... low risk Class I medical devices, no pharmaceutical involvement. The reference to 21 CFR 58 is extremely helpful and something I will definitely be looking into further. However, while the company I work for certainly performs activities that could be considered "nonclinical laboratory studies", we document data for many functions outside of the laboratory as well (especially production), and I'm not sure if the same regulations are necessary in those situations.

I also agree with your statement, chris1price, but I also understand some of the confusion that SMBIZZQA is experiencing. ALCOA+ clearly has very close ties with pharma practices, but the extent to which it should be implemented in the medical device world is a bit fuzzier. The same is true for good documentation practices... they both seem to ALWAYS be associated with pharma, and sometimes medical devices are mentioned on the side. It's very challenging to find resources on data documentation regulations that relate specifically to medical devices without it being a side-note to pharma practices.

So are ALCOA+ and GDP methods identical for medical devices and pharma? I've struggled with this question for some time and I'm not sure where to find a good answer. Maybe someone here on Elsmar knows...???

You are correct, the Data Integrity and ALCOA principles have come from the pharma arena. I have not seen an FDA warning letter for a device company discussing data integrity and I am not aware of anything FDA has published that is specific to devices and excludes pharma. Perhaps requiring ALCOA for a device company is a step too far. However if you follow the basic principles of ALCOA, your documentation practices will be very good.

I agree with @chris1price - why WOULDN'T you want to ensure the highest degree of data integrity? The risks, to me, are too high to do otherwise. I have seen 483s basically for Good Documentation Practices, just framed around (lack of) control of records.

A large part of why I am asking these questions is to defend our current procedure. Several of our upper management personnel come from industries outside of the medical device world and find our current procedures to be "excessive". I would like to say that our current procedures are somewhat necessary, but it's challenging to defend when there isn't much literature on good documentation practices specific to medical devices.

From the perspective of upper management, if we can take away some procedures that truly are excessive, it could free up a lot of time that production and quality have to spend ensuring that our DHRs are correctly completed.

It should be unlikely that good record keeping is too burdensome so there may be some deeper issues. If, for example, to capture a record, operators have to make a lot of changes, what are the reasons? Are you trying to capture too much info? Is it possibly redundant? Do you have too many checkpoints that require too many signatures? If you made an investment in some software to better automate DHR capture, might the long-term savings outweigh the investment cost?

Are you able to have an "adult conversation" with management? What, specifically, do they believe is "excessive"? Specific areas are always better than generalizations.

Have you talked to the folks that are capturing the records to find out what they feel is excessive?