Telcor Integrations – anyone doing?

[Jim Kosloskey]

Anyone doing integrations to/from TelCor?

Are you willing to share your insights?

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

Topic

[Glenn Friedenreich]

Hi Jim – We installed Telcor here at Christiana Care about 3 years ago.  Interface development was uneventful, and the application’s been running fine.  We have Telecor interfaced to/from the Cerner Classic Lab system.

The LIS app analyst that used to support Telcor is now on our Cloverleaf team;  Give me  a call if you’d like to chat with him.

– Glenn

[Nate Kruse]

We have Telcor (QuickLink) SureStep Glucometer results flowing through Cloverleaf into Epic.  I created a Tcl script to modify the ORU^R30 message.  It wasn’t much modification to make, but for instance I had to add a PV1 segment with the Acct Number, set segment counter for OBR, OBX, and NTE, and set ORC18 with the Device value found in the NTE segment.  The R30 message does not contain the patient’s name or MR#.  The main reference is the patient’s acct number…which makes searching a little trickier as we are usually given patient name and MR#.

I was just involved with the Cloverleaf and Epic part, not the entire process.

Nate.

[Kevin Crist]

We just set this up a few months ago. Wasnt bad at all. We are sending the orders and results through cloverleaf to McKesson Clinicals. Had to make minor changes but nothing drastic..

[Steve Holder]

Jim, we’ve done a results and charge interface to telcor.  Richard Engelbart is an excellent resource with telcor.  Very knowledgeable.

[Jim Kosloskey]

Thanks to all.

We will only be doing and ADT integration to Telcor.

It sounds like we should not have any serious issues.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Robert Denny]

Jim,

did you end up getting the telcor interface running?

We are moving to EPIC and I am tasked with setting up some logic for the results for unsolicited orders.

First time setting up a LAB Results feed, and I first wanted to create a variant for an ORU R30 if possible, but it does not appears to have the choice in CL.

Currently I need to do some logic based on the type of ORDER in OBR-4.

I am wondering if I am going about it incorrectly, but attempting to create the variant.

I’ll keep in touch, as I think we will run into the Patient Name issue that is mentioned by Nate.

[Jim Kosloskey]

Robert,

We are only doing the ADT via Cloverleaf.

However, you can define a variant for ORU R30 (or any Message/Event Type for that matter) even though you do not see one in the distributed set.

You will need the vendor’s Message Structure but typically the ORU or ORU^R01 can be used to start.

What I do is to copy all of the segments in an existing Message (in the GUI), create a new message with the name I want, paste the segments I copied previously then modify to meet the vendor’s structure.

Another thing I normally do is determine if the vendor’s structure can be expressed within the standard HL/7 message structure rather than define a structure that exactly matches the vendor’s description.

Most times the vendor’s description of their structure omit groups, etc which are optional and not used giving the impression the groups cannot exist in the definition.

By keeping those standard definitions (if they work) I eliminate the potential that a future release from the vendor will exploit those groups, etc. and significantly impact my Xlates.

Sounds more complicated than it is because the explanation gets lost in the written word. This concept is more easliy understood with conversation and example.

IMHO with any integration getting the message structure properly defined is one of the most important steps along with proper specification.

Welcome to the world of orders/results. I have often quipped that one has not really experienced the ‘joys’ of Health care integration if all one has done is ADT. Orders/Results really show the power of the Cloverleaf Xlate tool and also challenge the Integrators.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Robert Denny]

Here is an example of the test messages they have sent so far.

MSH|^~&|QML|LAB|EPIC|GSRMC|20120720153305||ORU^R30|2|P|2.4|2012072000001||AL|NE

PID|||100000000002^^^GSRMC|||||||||||||||18602

ORC|NW|||||||||||^^^||||||XP1047A0351164^^GSRMC^ICU^^|

OBR||||POC10^^^POC10|R|20120720124138|20120720124138||||O||||arterial/capillary^^^|^^^|||||||||||||||||||

OBX||NM|POCT GLUCOSE^^^POCT GLUCOSE||148||||||F|||||||XP1047A0351164^^GSRMC^ICU^^|20120720124138

If I understand correctly. I can create an ORU variant, and modify it so that I only have these segments in the variant? I will get started on this and await your reply.

Thanks for your help Jim.

[Jim Kosloskey]

Robert,

I insist on a definition of the vendor’s message structure. Using sample messages is a faulty way to determine the structure a vendor is intneding to use.

However, if this is all you have then from what you show here is what I would do:

1. create a new variant giving it whatever name you feel is descriptive for this vendor.

2. In the GUI HL/7 configurator, copy the segments in the

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

[Robert Denny]

Thanks Jim.

I have followed those steps and it is working. I will get the specs from the vendor this week.

Thanks for your help.

[Robert Denny]

Jim,

Now that I have new messages coming from the POCT, I have a few issues.

First I was getting the IR mismatch error. I clear them by creating variables rather than doing some of the =literal value -> to field.

Now I am only getting the MSH segment of the message.

MSH|^~&|||||||ORU^R30||P|2.4

PID

ORC

OBR

OBX

MSH|^~&|||||||ORU^R30||P|2.4

PID

ORC

OBR

OBX

MSH|^~&|||||||ORU^R30||P|2.4

PID

ORC

OBR

OBX

MSH|^~&|||||||ORU^R30||P|2.4

PID

ORC

OBR

OBX

MSH|^~&|||||||ORU^R30||P|2.4

PID

ORC

OBR

OBX

[Robert Denny]

I created the new variants by copying the ORU-01 and renaming it for an inbound and outbound variant.

Still getting the same results on only the MSH segment.

[bill bearden]

Robert,

It looks like you might be getting just the required segments.

If you are running through an xlate, are you doing a bulk copy to get everything moved over?

Is that the output from the xlate tester?

b

[Robert Denny]

Jim,

it ended up being a problem with my client.

I just received a new pc and we had the PC Tech department install Cloverleaf. There was a step that was missed after installing the client from the CD. We need to install updates. Normally we just copy the updates from a client that already has them installed.

I went out and grabbed those files we keep on the network and now my client is updated.

Now my variants are working and I can get all the segments from the inbound message.

You were a great help in diagnosing this problem.

[Nate Kruse]

Sorry Robert, just saw this thread. Let me know if you have any questions on the results. We now have an ADT interface implemented too. For the most part, it is a pretty hands-off interface with one sticking point….someone having the patient’s chart locked in Epic and the incoming result cant’ file.. So we had our Epic TS create a process where failed results, due to locked charts, get resubmitted every 5 minutes for up to 3 times. Then the result goes to our error work queue.

Nate

[Author]

Replies