CryptoPointers

Here is another one, in the BCMA MEDICATION LOG File:

53.79,.11 ORDER REFERENCE NUMBER .1;1 FREE TEXT (Required)

             INPUT TRANSFORM:  K:$L(X)>15!($L(X)<1)!'(X?.N1"U"!

(X?.N1"V")!(X?.N1"P")) X

             LAST EDITED:      MAY 9,2001
             HELP-PROMPT:      Answer must be 1-15 characters in

length.

             DESCRIPTION:
                               Contains the IEN to the actual order

in PHARMACY PATIENT (#55) followed by a U for Unit Dose or V for IV

             RECORD INDEXES:   AORDX (#47), AUID (#147)

I call this a "crypto-pointer to the ORDER file, File 100.

I recently

dealt with this by creating a COMPUTED POINTER to File 100, using this field. Details on request.

BTW, a quick search of FOIA VistA finds exactly one computed pointer, field 3.3 of File 798, pointing to File 4.

---------------------

Of course the classic example is in the lab data file where a combination of PARENT FILE (.02) and NAME (,03) are used as a pseudo variable pointer to either the patient file (2) or referral patient file (67).

---

And is another one in the Domain file: 4.21,1.3 PHYSICAL LINK / DEVICE 0;5 FREE TEXT

               INPUT TRANSFORM:K:$L(X)>30!($L(X)<1) X I $D(X) S

%ZIS="NQRS",IO

                               P=X D ^%ZIS K:POP X S:$D(X) X=ION W:$D(X) "

Sto

                               red internally as ",X D ^%ZISC S IOP="HOME"

D ^

                               %ZIS K IOP,%ZIS
               LAST EDITED:    MAR 12, 1993
               HELP-PROMPT:    Answer must be 3-30 characters in length.
               DESCRIPTION:    The physical link is the channel that the
                               transmission will take place on.  This field

is

                               always a free text pointer to the device

file.

                               See field 17 / Physical Link Device for more
                               information.

Just as a historical point about the Lab Data file, that was Lab's solution for a variable pointer before at least 1986. Long before there was such a field type within FM.

Chris Edkins