Welcome to Health Care POV | sign in | join
Stepwise Success

Fast Bar Codes

Published July 27, 2009 7:20 AM by Scott Warner

Let's suppose your information system test database prints at least one barcode specimen label for each test. This label contains the patient name, account number, test order number, a description of the test, and the barcode itself. Now suppose a new barcode label type is available that adds the date of birth and room location.

That would be nice. But you look at your test database:

  • 1000 orderable items
  • 6-digit item code number
  • 3 mouse clicks to get to the barcode field
  • 1 keystroke to change and hit ESC twice to return

At twenty seconds each test, that's thousands of keystrokes and clicks adding up to almost 5 ½ hours of payroll time. Mistakes and distractions add time. Printing a database report can pinpoint errors, adding more time. And don't forget the list of item numbers to read and check off. Figure eight hours to be safe.

Wow. Still, you really want that date of birth on those specimen barcode labels.

There are three basic strategies:

  • brute force – hire someone to sit for a day and pound it out
  • discrete – work your way down, starting with the most commonly ordered test
  • divide and conquer – divvy up the task between many techs over many days to lessen the blow

But wait! Computers are designed to perform mindless tasks. A simple script – a list of plain text instructions – can send keystrokes and mouse clicks to the information system and do the job in minutes, even "reading" and checking off a list. Without mistakes.

If the request is common, your IS vendor may already be able to do this at no extra cost. That's worth a telephone call. If not, there are many scripting languages available, some of which are used by your IT department. There are also freeware languages out there that can easily be programmed (as in plain English syntax) in Notepad. You'll be surprised how easy it is, with a little computer savvy, to make the beast burden your load. After all, eight hours is a day off. Maybe yours.

4 comments

So, printing seems to trigger all tests to be put into a consolidation queue.  Sounds maddening.  Meditech probably has online forums or list servers that may provide a solution.  Good luck!

Scott Warner July 30, 2009 7:33 AM

There's some weird "bug" with our Meditech here in Danville which somehow straightens itself out after you PRINT another set of labels for that accession #.  Even if you don't put the new label on the tube, just putting the tube back on the analyzer after reprinting the labels will lead to all tests getting run on the 123456 bar code (say BMP and CK, if the added-on CK does not run), for example.

Stephanie Mathis, MLS(ASCP), Generalist - Medical Laboratory Scientist, Danville Regional Medical Center July 29, 2009 7:17 PM
Danville VA

As I recall with Meditech, the barcodes are consolidated per instrument and reflex tests require the new barcode to be attached to the tube.  That sounds like what you've described.  Do I have that right?

Scott Warner July 29, 2009 5:57 PM

Our lab has experienced aggravating bar code problems since we implemented our Beckman Coulter Automate system (which pretty much all of us believe is more suited to much larger labs than that of a 300-bed hospital)!

You see, one patient's chemistry lab tests are on the same Meditech accession number (example 0729:C999), but could be distributed among three different 6-digit bar code numbers underneath the same accession number:

example - 123456 BMP, Mg, TSH, CK

               123457 CKMB, Troponin

                123458 Hemoglobin A1C

As if trying to keep up with which DxC 800 and DxI/Access tests were run on each bar code was not fun enough, if the Mg or CK was added on in the lab while the specimen was received, the DxC will not recognize that test as being assigned to bar code 123456 when the bar code label is read - and will not RUN it as a result!

The technologist who works nights with me most of the time had one bar code label on a tube that she loaded into the DxC (example 199999).  For some bizarre reason, the analyzer kept reading it as bar code # 200000 and calling it an "unprogrammed" specimen.  In that case, 200000 had other tests ordered on the same patient associated with it, BUT my colleague and I were extremely freaked out about the possibility of a different patient's tests being run off an incorrect specimen if that did not turn out to be the case the next time this mess happened.

Stephanie Mathis, MLS(ASCP), Generalist - Medical Laboratory Scientist, Danville Regional Medical Center July 29, 2009 4:42 PM
Danville VA

leave a comment



To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the image, reload the page to generate a new one.

Captcha
Enter the security code below: