3 Replies Latest reply on Dec 2, 2008 10:19 AM by TSGal

    Deleting old repeated records



      Deleting old repeated records


      I would like to merge data at a central level from different clinics' databases that are using the same patient database. (I am using importing from file). The imported records may have an older clone of them in the host databse. I wonder how can I delete those old repeated records and keep only the updated ones during importing or by using a special script to delete them.

        • 1. Re: Deleting old repeated records

          Some very important questions.


          1)  Is there a unique identifier in each of these records, like Social Security Number, that will allow the computer to tell that two records are, indeed, the same?


          2)  Do your records record modification dates so that you know which is older?  (Or is the central file always going to be older?)


          3)  Is it possible for data to have been entered on the same patient in multiple clinics?  (So that you can't just delete one but, instead, have to merge the information?) 

          • 2. Re: Deleting old repeated records

            Thanks for your assistance, let me try to answer your questions.

            1. Yes, there is a unique number that identifies every patient. Each patient may have more than one follow up visits with different visit dates that can determine the old from the updated records.


            2. Yes, there is a modification date fields in the tables where there is no visit date.


            3. There is no concurrent data entry for the same patient in different clinics, but a patient can be transferred out to other clinics for different reasons, where the records for that patient are emaild to the reciepient clinic to import in its file.



            • 3. Re: Deleting old repeated records



              Thank you for your posts.


              The unique number for every patient is important.  However, how do you determine if a record needs to be deleted or a record needs to be merged?  There needs to be some sort of criteria to determine what steps to take.


              Once you import the data, you can run a script to look at the modification dates.  If they are the same, then you know you probably have the same record, and one of those could be deleted.  You may also want to include the Creation date, just in case two people modify the same patient at different places, but the records were created at different times.


              What problems are you running into at this time?  That may help determine a proper answer.



              FileMaker, Inc.