We are on Misys and our Vitek's are interfaced (thank goodness). I would
hate to have to hand-enter all that data. My condolences to your micro
department.

My husband is a systems analyst for another large local hospital. He manages
interfaces between the hospital's different systems. For example, they have
a registration system that has to send out its demographics to radiology,
lab, etc. They ditched their mainframe quite a number of years ago. These
interfaces allow for the transfer of data between departments. He uses a
commercial software program for this but it has to be fiddled with in order
to get it to work with the various systems. Nothing is easy. Everything in
computers is institution dependent on what else is running. He deals with
many many interfaces on a daily basis.

As far as an open-source LIS: our place has invested mega-bucks getting our
hospital and all the hospitals in our region on one system. I doubt that,
after these investments, they would go to something that is open source.
Also, there are big time security issues involved with any LIS (read that
HIPAA). You have to be able to prevent hackers from getting in. I'm sure
"open source" would sound too scary to most administrators. I would think
your market would be smaller independent labs. These, however, are rapidly
being gobbled up into large lab systems.

All our specimens are assigned an accession number that prints out bar coded
labels. When we put blood culture bottles onto our blood culture
instruments, for example, we scan the bar code for the type of bottle it is
(aerobic or anaerobic) and then scan the accession number bar code into the
instrument, then place the bottle into an available slot (chosen by the
instrument or the tech based on the available slots lighting up at the
time). When we accept a transport list for these blood culture, this
information is relayed to the BC instrument via an interface. This way we
can look up the bottle by accession number of name on the BC instrument's
data screen. Since we have the capability of nearly 500 bottle slots, we
could not possibly look for bottles by hand when a phone call comes in to
the effect of "do you have blood cultures on May Smith yet?" We get their
med record number and search for them that way, ideally, or by name if they
don't have a MR number.

Chemistry, Hematology, and virtually ever other lab department relies
heavily on bar coding as it prevents errors of hand entering numbers and
saves immense amounts of time. Bar coding in labs has been available since
the late 80's, I believe. I know that at my old lab they were bar coding
hematology specimens back then. I don't think bar coding for chemistry came
until a bit later, however. Printers have much improved, making the bar
codes much easier for the instruments to read. In the olden days they
smudged with handling quite often, making scanning a big problem. Better
labels and inks have helped this problem a lot.

I would guess that most of the information that Tim would like to get his
hands on is proprietary and held very close to the vest by the companies
that created it. This is BIG BIG business worth millions of dollars. There
would be no reason to release any of this closely guarded information out to
anyone, except in the parts people like John at their respective labs have
to deal with. I know when these big systems do upgrades, everything has to
go through the vendor, as my husband has dealt with numerous outages and
downtimes throughout his work career. I wouldn't begin to know how all this
stuff works, though :-). I know that some of the fixes we'd like in our
current system "have to wait until the next upgrade."

Judy Dilworth, M.T. (ASCP)
Microbiology