Looking for information on Historical Bloodbank data archive solutions

[spascoe]

Hi, I want to start by saying, I am a software developer, with personal interest in bloodbank data archive systems, but unfortunately, I don't have access to many connections with BloodBankers. If my post is inappropriate, I apologize as I have no intent to offend...

I'm trying to identify ways in which you have solved the challenges of historical transfusion/unit data for systems that your organization may have migrated off of, or where the data was brought into your organization by acquisition (or some such method).

I have a client that needed to move away from an old custom written application. In the process of helping them convert to a new historical archive, I became interested in trying to identify how other organizations have solved these challenges.

Specifically, I'm interested in the following:

Do you use a shrink-wrapped application, a custom built-in-house application, a tool provided by your current vendor, or have no historical archive?

If you have a historical archive:

* Did you use screen-scraping, or discrete database accesses to pull the information out of the old system?

* How many different bloodbank systems do you have aggregated together?

* What type of lookup abilities do you need in a historical archive?

* What do you like about your current system?

* What would you like your current system to do, but it doesn't do it?

[carolyn swickard]

Don't really know what half of this means - but for our system, when we changed from Cerner Classic to Meditech we kept acces to Cerner data for Pathology and Blood Bank on the original alpha drive. It is completely separate and has to be accessed independently, there was no crossover to the new system. Any history from the old system to the new system has to be moved manually. Our archive has recently been moved to a more modern storage system but is still completely separate. We access it as needed.

Would have loved to have been able to transfer all of our historical data to the new system, but that did not seem to be an option. I'm sure most Blood Bank systems would love to be able to incorporate all original data into a new system, but the hospitals would have to be able to afford it too. They would also have to know the option exists and the big Lab software vendors would have to work with it on their individual platforms.

[marvy1]

When working at an old hospital, we changed from Sunquest to Cerner and our IT guy wrote a script to transfer all historical data. This took him a week to figure out and 2 days to do.

In the current hospital I work at, when transferring from 1 older version of Meditech to a newer version, it took ~ 1 year + to make it happen. This included multiple software fixes along the way to accomplish the task. I assume we were one of the first Meditech hospitals to try this (since there were so many software fixes along the way). Meditech tries to make it so hard to transfer data and discourage hospitals from trying. I assume they must feel historical data is not worth transferring, otherwise they would figure out how to make it happen. I could understand if a software vendor had difficulty pulling data out of some old obscure databank, but we wanted to pull data out of their own software.

[ChrisH]

The last two places I have worked, we were able to export select data and imported the data into a access database. I just had to use that database last week to do a lookup.

What ever you do make sure it is easy to use. Not everyone is computer literate.

[Mabel Adams]

I have worked with data that was converted from an old BBIS to a new one in two places. At my current workplace we have both converted data and an Access database from the last system. There is an even older pre-1999 database as well. All archives have to be able to retrieve the final disposition of each unit of blood product, so it has to have the Unit no., product code, and when it was transfused to whom (or that it was transferred out or discarded and when). It has to be searchable by unit number because that is how blood suppliers send recalls and lookbacks. The database must also be searchable by patient identifiers (Name and probably medical record number--whatever you want to be able to use) so you can look up patient history. For some places it would be nice to be able to print out a patient report--although there are rules about lab reports having to print on them the name of the facility that did the testing so that may be more trouble than it is worth. The patient data on file will need to include the interpretations of blood typing and antibody screen tests and probably antibody identifications if there is not a paper archive for that. Some systems may pull only what the patient's blood type is, without the detail of when it was tested or how many times. Results of crossmatch (compatibility) tests of units against that patient might also be necessary. That's a start.

[ChrisH]

I agree with Mabel. The place I work out now has it searchable by name and unit number. If the old system has results it would be most helpful to see those also. On the patient side, if the old system had any special needs, comments, antigen these should also be listed.