If you want to be able to process and transmit incoming messages that are not in HL7 format, you first need to analyze your incoming messages to determine what fields they contain. If you are writing information to a database table, you then need to determine which of these fields contain data that you want to store in the table.

For example, suppose that you have a file containing the following patient data:

[11/13/2009 10:26]        |1|0001533008|4071917902|DOE|JOHN||LESB #3 W/C-ARM|ANESTHESIA|Jones, Frank G|002782|
[11/13/2009 10:26]        |1|0000451253|4071468583|ROE|MARY|L|MED CHECK/POSS PROCEDURE|ANESTHESIA|Jones, Frank G|002782|
[11/13/2009 10:26]        |1|0001472549|4046188712|SMITH|TOMMY|J|PUMP REFILL|ANESTHESIA|Jones, Frank G|002782|

Since these messages are not in the standard HL7 format, you will need to find out what information is supposed to be contained in each of the fields in these messages. For this tutorial, the messages above are assumed to contain the following information:

• The date and time of the message (enclosed in square brackets);
• The sending facility;
• The MRN (medical record number);
• The patient ID;
• The patient's last name, first name, and middle name or initial;
• The procedure performed;
• The procedure service;
• The surgeon name;
• The surgeon code.

As you can see, these messages resemble HL7 messages. Information is separated by | characters, as in standard HL7. However, there are no message segment names, and the data in the message appears in a non-standard order.

The following steps show how to use custom segment definitions and the Global Inbound Script to transform this message into a format that can be understood by Chameleon and Iguana. After that, you can create a message definition to map fields from the message into fields in your database table.