Moving from Cloverleaf 6.0.2 to 6.1.1 – Delimiters changed

[Jim Rawls]

We’re running message comparisons prior to upgrading from Cloverleaf 6.0.2 to 6.1.1, and we’re finding numerous instances where the delimiters are being changed by the new Cloverleaf. Here is an example:

6.0.2: 13532-7^XANTHOCHROMIA CSF QL^LN

6.1.1: 13532-7&XANTHOCHROMIA CSF QL&LN&XANCSF&XANTHOCHROMIA CSF&L&&v1&XANTHOCHROMIA CSF^1

This is field OBR-26. The first component separator ^ is being changed to subcomponent separator &. It also appears that truncation is occurring in the OLD 6.0.2 version that is no longer occurring in 6.1.1. Note the length difference in the 2 samples above.

Example 2, also of OBR-26:

6.0.2: 17284-1^MITOCHONDRIA AB SER QL IF^LN

6.1.1: 17284-1&MITOCHONDRIA AB SER QL IF&LN&AMTOAB&ANTI MITOCHONDRIAL AB&L&&v1&ANTI MITOCHONDRIAL AB^1

How can we prevent this from happening? Nothing has changed in our configurations (Xlt files, Tcls, HL7 variants), just the Cloverleaf release.

(Also using same release of Red Hat Linux 6.7)

Topic

[Jim Kosloskey]

Jim,

Does the input llook like what is the 6.0.2 example shown?

What version of HL/7 for the outbound?

Also what Xlate Actions are used to make the COPY (can you show the properties of the Action)?

Any Tcl involved?

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

[John Mercogliano]

Do you send this field to a xlt proc.

John Mercogliano
Sentara Healthcare
Hampton Roads, VA

[Jim Rawls]

This is an HL7 2.5.1 ORU^R01^ORU_R01 lab result message (both inbound and outbound).  This input matches the 6.1.1 output, as do the delimiters, so it appears that 6.0.2 (and 5.7.2, where this was developed) had been making a substitution of ^ for & in this case.

The copy is straight across for the whole field.

There is a post-xlt Tcl proc that renumbers the OBX and NTE segments.  It splits segments at the field level, updates OBX-1 or NTE-1 and reassembles the segments.

[Jim Kosloskey]

Using COPY at the field level when there are subcomponents is problematic. That field in the HL/7 standard has subcomponents in at least the first component if I recall correctly and it appears the sending system is following the standard.

PATHCOPY at the field level or COPY at the subcomponent level is more reliable.

Of course the situation now may be the receiving system wants everything as components and not components with subcomponents.

In that case it may be necessary to COPY from component and subcomponent to component.

So I think the real issue is with the original approach (COPY @ field level when subcomponents are possible) and now Cloverleaf has exposed that flaw.

Anyway those are my thoughts.

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

[James Cobane]

With regards to the truncation no longer occurring, there is a note in the Release notes:

See “Issue in CIS 6.1 while parsing the subcomponent separator (&)” section in the Release Notes.

Hope this helps.

Jim Cobane

Henry Ford Health

[John Mercogliano]

Jim,

  Based on what you said it sounds like it is a side affect of the change to there ampersand handling.  The release notes have a pretty good description of this change with some examples.  

  I remember having to deal with this myself in 5.7.  I my case, I needed the ampersands and had to use a pre a post proc to convert the ampersands to /T/ and back again.  

  In 6.1.1, I’m now able to remove the procs because they now handle the ampersands correctly.

John Mercogliano
Sentara Healthcare
Hampton Roads, VA

[Jim Rawls]

Thanks guys.  Looks like Infor fixed what we’ve been coding around all of these years…

[Author]

Replies