To answer the question, turn on "Prohibit modification during data entry" and you will get a new creation timestamp. I know counter intutive, but welcome to FM.
You should have two Timestamp fields, creation and modification, each set as above, as even though this is a duplication it is a new record.
And if you don't already, you should also have a serial number field, with full validation, can't modify, always validate, numeric only and unique.
In filemaker you can "Auto Enter" a timestamp on the creation of a field or on the modification of a field.
When I use this I usually make two fields:
TimestampCreation - Auto enter Timestamp on Creation
TimestampModification - Auto enter Timestamp on Modification
If you have a multi user system you can also add the accountname that created or modified a record.
But I feel you are trying to create unique ID's using large calculations. It's not really neccesary.
1. TimeStamping creation of a new record I understand.
2. However, if I've prohibited modification during data entry, which TimeStamp will react when I duplicate a record in order to enter some new data for a recalculation? ie. I can't see when TimeStamp on modification can work if I've prohibited modfication on data entry.
3. What's the difference between copy and duplicate?
Copy will copy the selected contents of a field, duplicate will recreate the entire record, save those fields mentioned below, when the prohibit mod is turned on.
So let's start over. You need to track when a record was created and when a record is modified. It doesn't hurt to track who created the record and who, if it ever gets modified, made the modification. Plus every record in every table even if you don't need it right now needs a unique serial number for that record.
The first four are often called developer fields, but in the case of medical records, they can be important for tracking purposes.
So in a test file create the above five fields and a text field to watch the mod Ts change. Put the prohibit on like suggested and then create new records and then duplicate some of these records. I believe you will see the results you initally requested.
You haven't mentioned your schema, but if you do not already have a patient table, you might consider doing so as this will allow you to put the name, address, Dx, etc. in one record that if anything ever needs updating you only have to go to that one patient record and it is changed for all related records, such as your caculation records and it also make it possible to view the calc records in a portal and you can spot trends, etc.
when assuming Patient ID has been obtained from 'Lastname&Firstname&DOB&sex'
I do not recommend using this method for patientID's. While duplicates are unlikely, you can't absolutely guarantee that they will be unique. Two John Smith's born on the same day are far from impossible.
Also, data entry errors not immediately discovered and corrected will play havoc with your data. Imagine entering several related records for our "John Smith" and then discovering that his name is really John Schmidt.... To correct the name, you have to track down and update all the records where this data is used to link records in your database.
You can avoid both issues if you use an internally generated serial number or (if using FileMaker 12) an auto-enter field that uses the new Get ( UUID ) function to generate a unique ID for each Patient in your Patients table.