Okay, you did ask nicely.
It's the FDA guidelines as to what they expect/prohibit/obiligate when you mix computers and medicine.
Mostly they worry about industry. Where I work, boss wanted to take us wireless as to stocking received material. It was promptly shot down, for the time being.
If it looks like a database, smells like a database, and you have critical (clinical tests, maybe?) information on it, then you need to reconsider.
If you search on that phrase, you will find CD-ROMs on validation protocols, seminars on validation protocols, consultants, etc.
Another buzzword is IVT, instrument validation and testing.
FDA is worried about chains of custody, who entered the data, etc. If the program simply prints out, and doesn't save anything (and you can prove it's save mechanism is disabled) perhaps you are okay.
The main grief is who entered the data, and how to prove/disprove.
I am basically ignorant of this stuff. As time goes by, and in-house Regulatory visits the netherworld of Receiving, I will know more.
The nightmare scenario currently making the rounds is Baxter. They 'blew off' the FDA, and finally, the FDA 'shut them down'. All I know is our company is now a bit phobic about new applications that could be interpreted as 'database'.
Another nightmare scenario was the radiation therapy machine that interpreted a technician
s typos without giving an error. I think ten times the expected dosage was given. Yes, people died.
Just be careful when you mix Microsoft Office and anything pharmaceutical.