Creating SPARCS files Special Procedures for Solucient data >>>

This overview is still valid as a guide for doing special SPARCS runs, such as
On-Request yearly or quarterly runs. However, for the regular monthly runs:


Life has been made easier - Monthly SPARCS now automated!

     We create seperate SPARCS files for InPatients and OutPatients because NY State wants it that way. There is a limit to the number of Patient Types brought down in a particular run, so Inpatient data could theoretically require more than one run if there are more than nine Inpatient types.

     NYS also requires seperate files for each physical location, so Fishkill gets its own file. Fishill has only Outpatient data, so a special Patient Type code of 'F' was built for them, with Patient Types 'O' being regular POK Outpatients and 'E' being OutPatients via the Emergency Room. If Fishkill ever has Inpatients, we would need to create a 4th file and would need to create unique Patient Type codes for them, to distinguish between POK and Fishkill Inpatients, just as POK types 'O' & "E" distinguish from Fishkill type 'F'.

     Beyond that, there is a special run for POK Outpatients of type 'Q', subtype 'CTH' (CathLab). Normally, there is one SYSIN card which specifies, patient type. For POK Outpatients, this card specifies 'OE' to get both types. In the case of the Q/CTH patients, there are two cards, the first specifying type 'Q' and the second specifying 'CTH'.

When I build JCL, I follow certain member/job naming conventions. For example:

  FKJUL06 would be for Fishkill July 2006 data (Outpatients)
  OPJUL06 would be for POK July 2006 data (Outpatients types 'OE')
  IPJUL06 would be for POK July 2006 data (Inpatients)
  CTJUL06 would be for POK July 2006 data (Outpatients type 'Q', subtype 'CTH')
  IPQ106 would be for POK Inpatients, 1st quarter 2006
  FKQ205 would be for Fishkill Outpatients, 2nd quarter 2005
  IPYR05 would be for POK Inpatients, all year 2005

     A given file can be set as TEST or PROD when sent to NYS, with PROD being the default.

     While Invision allows for automatic creation of these files, VMBC personnel need to make corrections before we run the jobs. This is an ongoing process, so we never know in advance when a particular batch of data will be ready for transmission. We therefore run the jobs On Demand. We also sometimes do special runs to correct errors, with records being set to 'mythical' dates such as January 1966 or April 2017 and the record dates being reset properly after the SPARCS run.

     A given SPARCS file contains data for a particular time period: Monthly, Quarterly, Semi-annually or Annually. It also has data for specific patient types. The contents of the files are controlled by Sysin data and parms in the JCL, which is in RLS0.SPARCS,JCL. The JCL used is a modification of Siemens' job PADTPA7$, so if this job is redelivered on a SUT, we must propagate any changes or additions to our JCL

     We create new jobs for every run, copying from previous jobs. For example, to create files for month of June 2003, we might create members MIP0603 (InPatient) and MOP0603 (OutPatients). For 1st Quarter 2003, we might create Q1IP2003 and Q1OP2003.

In the JCL:
Parm "DATE=mm/dd/yy" is set to the first of the month FOLLOWING the desired time period of the report.
  For June 2003 reports, this would be "DATE=07/01/03"
  For 1st Quarter 2003, this would be "DATE="04/01/03

For example:
JULY 20003
...
//           DATE=08/01/03,
...
OutPatients at POK
...
//PATINIT.SCHFILE DD  *
1V0PATAPES   111893FN  111793FN   A7
2V0PATAPES   PATAPES 000          A7NYOOM00        3330OE
...

OutPatients at Fishkill
...
//PATINIT.SCHFILE DD  *
1V0PATAPES   111893FN  111793FN   A7
2V0PATAPES   PATAPES 000          A7NYOOM00        3330F
...

InPatients (POK)
...
//PATINIT.SCHFILE DD  *
1V0PATAPES   111893FN  111793FN   A7
2V0PATAPES   PATAPES 000          A7NYIIM00        3330BIJKLNPSV
...

On the 2nd card:
For InPatients, the value of columns 39-40 = "II" for InPatients, "OO" for OutPatients

Column 41 = M (monthly), Q (quarterly), S (Semi-Annual) or A (Annual)

Column 42-43 = "00" (zero - no delay...create report now)

Columns 44-51 blank if Monthly
 if Quarterly, "03060912" because Fiscal Year = Calendar year
 if Semi-annual, "0612    "
 if Annual, "12      "

Columns 56-64
OutPatients:
   "OE" for POK Outpatients and ER Outpatients
   "F" for Fishkill Outpatients
InPatients:
   "BIJKLNPSV"
NOTE:Only 9 possible patient types. If more needed, would have to be seperate job.

Files created:
PDSMS.SMPA.DGA7HCV0.PANYITPn
PDSMS.SMPA.DGA7HCV0.PANYOTPn
 Note: There may be several files. Check creation date for your files

You can edit these files if necessary to change them from PROD to TEST in columns 143-146

Reports:
Reports go to the GQ and to Report-To-Web. A copy is also in the Spool and can be printed.

Downloading files to Server so Medical Records can upload to NY State
Edit member FPTPOK
Go to bottom and change OS filename(s) to what you created
Change downstream filename to something identifiable. For example, for JULY 2003:
put 'PDSMS.SMPA.DGA7HCV0.PANYITP0' ipjuly03.txt
put 'PDSMS.SMPA.DGA7HCV0.PANYOTP0' opjuly03.txt
Save
Submit the job and check OUTPUT Sysout to be sure it downloaded to the server okay.

After the Poughkeepsie OP and IP files have been created and downloaded to the server, create the Fishkill OP file. Edit FTPFK and set the appropriate file name, then run the job.

Contact Medical Records (Ann Sullivan).
Give her the filenames created on the server - in this example, 'ipjuly03.txt', 'opjuly03.txt' and 'fkjuly03.txt'


*** IMPORTANT *** *** IMPORTANT *** IMPORTANT *** IMPORTANT *** IMPORTANT ***
*** IMPORTANT *** *** IMPORTANT *** IMPORTANT *** IMPORTANT *** IMPORTANT ***


Create and DOWNLOAD any Outpatient data (Types OQ, Q/CTH or F) BEFORE creating another OutPatient file, since the new file may overlay the old!

Scenario:
  Create POK IP and OP files
    Download both files to server
  Create POK Q/CTH file
    Download file to server
  Create Fishkill OP file
    Download file to server
  Notify user


PRINT