11 Replies Latest reply: May 15, 2014 9:52 AM by Grant Perkins RSS

    Extract Data From Multiple Lines

    jane888 _

      Hi, new to Monarch. using Monarch Pro v7 on WinXP, SP2.  here is the report I'm reading in Monarch:

       

      Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

      **********************************************

      SVCDT      UNIT  CHARGES

      10/1/XXXX    1    XXXX.XX

      10/2/XXXX    1-   XXXX.XX-

      -


        -


        -


      REVCODE  REVTYPE  REVUNIT    REVCHARGES

      ABCD     EFG          1         XXXX.XX

      XYZH     TPT          1        XXXXX.XX

      TWSJ     GUT          1-       XXXXX.XX-

       

      There are over 30 reports like this. The svcdt-unit-charges lines are details and the revcode- revtype-revunit-revcharges lines are continuous of the records but wrapped around in the report.  First, I created a detail report with the svcdt-unit-charges lines.  It worked like a charm.  However, when I tried to create the append template with the revcode- revtype-revunit-revcharges lines, Monarch returned incorrect data. I guess Monarch treated both details in many-to-many relationship.  How should I solve this problem to have data extraction like this:

       

      Three records with columns patient, SSI, DOB, svcdate, unit, charges, revcode, revtype, revunit, revcharges: each record has the save patient, SSI, DOB;

      record 1 has svcdate "10/1/xxxx" line and revcode "ABCD" line;

      record 2 has svcdate "10/2/xxxx" line and revcode "XYZH" line;

      record 3 has blank svcdate line and revcode "TWSJ" line;

       

      pretty confusing? pls let me know if yes.  any help is much appreciated.  :confused:

        • Extract Data From Multiple Lines
          jane888 _

          Hi, new to Monarch. using Monarch Pro v7 on WinXP, SP2.  here is the report I'm reading in Monarch:

           

          Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

          **********************************************

          SVCDT      UNIT  CHARGES

          10/1/XXXX    1    XXXX.XX

          10/2/XXXX    1-   XXXX.XX-

          -


            -


            -


          REVCODE  REVTYPE  REVUNIT    REVCHARGES

          ABCD     EFG          1         XXXX.XX

          XYZH     TPT          1        XXXXX.XX

          TWSJ     GUT          1-       XXXXX.XX-

           

          There are over 30 reports like this. The svcdt-unit-charges lines are details and the revcode- revtype-revunit-revcharges lines are continuous of the records but wrapped around in the report.  First, I created a detail report with the svcdt-unit-charges lines.  It worked like a charm.  However, when I tried to create the append template with the revcode- revtype-revunit-revcharges lines, Monarch returned incorrect data. I guess Monarch treated both details in many-to-many relationship.  How should I solve this problem to have data extraction like this:

           

          Three records with columns patient, SSI, DOB, svcdate, unit, charges, revcode, revtype, revunit, revcharges: each record has the save patient, SSI, DOB;

          record 1 has svcdate "10/1/xxxx" line and revcode "ABCD" line;

          record 2 has svcdate "10/2/xxxx" line and revcode "XYZH" line;

          record 3 has blank svcdate line and revcode "TWSJ" line;

           

          pretty confusing? pls let me know if yes.  any help is much appreciated.  :confused:

          • Extract Data From Multiple Lines
            Grant Perkins

            Originally posted by jane888:

            [font="courier"]Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

            **********************************************

            SVCDT      UNIT  CHARGES

            10/1/XXXX    1    XXXX.XX

            10/2/XXXX    1-   XXXX.XX-

            -


              -


              -


            REVCODE  REVTYPE  REVUNIT    REVCHARGES

            ABCD     EFG          1         XXXX.XX

            XYZH     TPT          1        XXXXX.XX

            TWSJ     GUT          1-       XXXXX.XX-[/font][/quote][/b][/quote]Just to be sure I understood the problem correctly am I right to understand that, in efect, this report really represents a series of long lines that have been "block wrapped" rather than line wrapped for presentation?

             

            In other words it really should look something like this:

             

            [font="courier"]Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

            **********************************************

            SVCDT      UNIT  CHARGES     REVCODE  REVTYPE  REVUNIT    REVCHARGES

            10/1/XXXX    1    XXXX.XX      ABCD     EFG          1         XXXX.XX

             

            10/2/XXXX    1-   XXXX.XX-     XYZH     TPT          1        XXXXX.XX

             

            -


              -


              -


                TWSJ     GUT          1-       XXXXX.XX-[/font][/quote]Secondly is the report always 3 lines (in effect) or can it vary in the number of lines presented? Can any of the other reports you mention have a variable number of lines? If more than 3 lines are possible what is the maximum number? Is there a maximum number?

             

            I think you have worked out why the append template part of your model did not produce the results you expected. Plus an Append Template needs to be located in the report BEFORE the data to which it is being appended, although there are some possibilities to trap before the detail but capture data which overlaps or follows the detail. Much care required though.

             

            The alternative would be to use a FOOTER template. If you do that you will still get the contents of the footer appended to every detail line.

             

            Ideally what we need to achieve from the extraction is to rebuild the report into the single unsplit detail lines it should have. If we can do this either 'physically' (i.e. reconstitute the report and export it as a new report file, then run a final model on the new version) or virtually (i.e.  deal with the re-grouping of the split lines using calcualted fields in the table) may depend on your answers to the questions above.

             

            If we know the records in the report always have 3 logical data rows split as "row 1 +row 4", "row 2 + row 5" and "row3 + row6" the task should not be too difficult. But if the number of logical detail rows is variable it is not so easy!

             

            Either way is likely to make use of Monarch possibilities that, as a new user, you are unlikely to have come across up to now. So when I say 'not too difficult' I mean 'not too difficult once the approach has been outlined by someone'.

             

            Looking forward to your answers.

             

            Grant

            • Extract Data From Multiple Lines
              Grant Perkins

              Originally posted by jane888:

              [font="courier"]Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

              **********************************************

              SVCDT      UNIT  CHARGES

              10/1/XXXX    1    XXXX.XX

              10/2/XXXX    1-   XXXX.XX-

              -


                -


                -


              REVCODE  REVTYPE  REVUNIT    REVCHARGES

              ABCD     EFG          1         XXXX.XX

              XYZH     TPT          1        XXXXX.XX

              TWSJ     GUT          1-       XXXXX.XX-[/font][/quote][/b][/quote]Just to be sure I understood the problem correctly am I right to understand that, in efect, this report really represents a series of long lines that have been "block wrapped" rather than line wrapped for presentation?

               

              In other words it really should look something like this:

               

              [font="courier"]Patient: ABC SSI: XXX-XX-XXXX DOB: XX/XX/XXXX

              **********************************************

              SVCDT      UNIT  CHARGES     REVCODE  REVTYPE  REVUNIT    REVCHARGES

              10/1/XXXX    1    XXXX.XX      ABCD     EFG          1         XXXX.XX

               

              10/2/XXXX    1-   XXXX.XX-     XYZH     TPT          1        XXXXX.XX

               

              -


                -


                -


                  TWSJ     GUT          1-       XXXXX.XX-[/font][/quote]Secondly is the report always 3 lines (in effect) or can it vary in the number of lines presented? Can any of the other reports you mention have a variable number of lines? If more than 3 lines are possible what is the maximum number? Is there a maximum number?

               

              I think you have worked out why the append template part of your model did not produce the results you expected. Plus an Append Template needs to be located in the report BEFORE the data to which it is being appended, although there are some possibilities to trap before the detail but capture data which overlaps or follows the detail. Much care required though.

               

              The alternative would be to use a FOOTER template. If you do that you will still get the contents of the footer appended to every detail line.

               

              Ideally what we need to achieve from the extraction is to rebuild the report into the single unsplit detail lines it should have. If we can do this either 'physically' (i.e. reconstitute the report and export it as a new report file, then run a final model on the new version) or virtually (i.e.  deal with the re-grouping of the split lines using calcualted fields in the table) may depend on your answers to the questions above.

               

              If we know the records in the report always have 3 logical data rows split as "row 1 +row 4", "row 2 + row 5" and "row3 + row6" the task should not be too difficult. But if the number of logical detail rows is variable it is not so easy!

               

              Either way is likely to make use of Monarch possibilities that, as a new user, you are unlikely to have come across up to now. So when I say 'not too difficult' I mean 'not too difficult once the approach has been outlined by someone'.

               

              Looking forward to your answers.

               

              Grant

              • Extract Data From Multiple Lines
                jane888 _

                Hi Grant, thanks much for your detailed response.  Here the answers to your questions:

                 

                1) It's the "block wrap", not "line wrap" for presentation.

                 

                Bingo! You're absolutely right.  The report is like the way you mentioned in you reply.

                 

                2) Is the report always 3 lines (in effect) or can it vary in the number of lines presented? Can any of the other reports you mention have a variable number of lines? If more than 3 lines are possible what is the maximum number? Is there a maximum number?

                 

                No. All reports can have various lines and no max from my understanding at this moment.

                 

                3) Create Detail Template before Append Template.

                 

                Yes, I did create a detail template with the svcdt-unit-charges lines already. When I tried to capture the "block wrap" (revcode- revtype-revunit-revcharges lines) by creating an append template, I got the incorrect result, which makes sense because of the many-to-many relationship.

                 

                4) Ideally what we need to achieve from the extraction is to rebuild the report into the single unsplit detail lines it should have.

                 

                Unfortunately, it's not an option because this is the only format I can get from vendor. I've tried to get Executives involved, but still no luck to resolve.

                 

                Knowing all these contraints, I guess this may belong to "Not too easy" group.  For a new greenie like me, mission impossible 4!

                 

                  :eek:

                • Extract Data From Multiple Lines
                  jane888 _

                  Hi Grant, thanks much for your detailed response.  Here the answers to your questions:

                   

                  1) It's the "block wrap", not "line wrap" for presentation.

                   

                  Bingo! You're absolutely right.  The report is like the way you mentioned in you reply.

                   

                  2) Is the report always 3 lines (in effect) or can it vary in the number of lines presented? Can any of the other reports you mention have a variable number of lines? If more than 3 lines are possible what is the maximum number? Is there a maximum number?

                   

                  No. All reports can have various lines and no max from my understanding at this moment.

                   

                  3) Create Detail Template before Append Template.

                   

                  Yes, I did create a detail template with the svcdt-unit-charges lines already. When I tried to capture the "block wrap" (revcode- revtype-revunit-revcharges lines) by creating an append template, I got the incorrect result, which makes sense because of the many-to-many relationship.

                   

                  4) Ideally what we need to achieve from the extraction is to rebuild the report into the single unsplit detail lines it should have.

                   

                  Unfortunately, it's not an option because this is the only format I can get from vendor. I've tried to get Executives involved, but still no luck to resolve.

                   

                  Knowing all these contraints, I guess this may belong to "Not too easy" group.  For a new greenie like me, mission impossible 4!

                   

                    :eek:

                  • Extract Data From Multiple Lines
                    Grant Perkins

                    Hi Jane,

                     

                    Briefly to the points list;

                     

                    1 - Good, we know what is required.

                     

                    2 - A small, fixed number fixed number of lines might have offered some interesting possibilities - but we can ignore those for now!

                     

                    3 - Appends must start before detail BUT don't apply here in that way anyway because we really have a split detail line not a detail an append.

                     

                    4 - OK, we'll do it with Monarch!

                     

                     

                    Are you ready for this? This is not really newbie stuff but it won't be too difficult.

                     

                    I'm going to suggest 3 steps to this process so that we can sort of recreate the report (as much as we need to) to recombine the split lines and then work with the resulting table.

                     

                    We will be using a couple of versions of a simple model and then a final model using the Monarch Pro Database and External Lookup functions to create the final records.

                     

                    If you have access to the Monarch Learning Guide some of the sections could be useful pre-reading. Failing that read the Help sections that relate to the topics. Exports, Project exports, Databases and External LLokups would all be useful reading.

                     

                     

                    The second block of lines may be easier to use as a 'PRIME' detail record since they are easy to trap whereas the first block may not be so easy because we need to keep the line with dashes. So the easy traps are not helped by that. We have to have the SAME NUMBER OF ROWS in each part of the  extraction for each patient to get the match to work. If the report does not always deliver this things may get more complicated!

                     

                    Open the report and using one of the lines in the second block of lines

                     

                    [font="courier"]REVCODE  REVTYPE  REVUNIT    REVCHARGES

                    ABCD     EFG          1         XXXX.XX

                    XYZH     TPT          1        XXXXX.XX

                    TWSJ     GUT          1-       XXXXX.XX-  /font[/quote]generate a trap (A numeric in the column under REVUNIT and the . in the REVCHARGES column look good for that) then  paint in your fields making sure they are wide enough for the max possible field filling. That's your detail at this point.

                     

                    Save that template and add another template for the APPEND of the Patient Details. I would suggest trap on "Patient" and then grab the NAME, SSI code and DOB. We will need those later.

                     

                    Save the APPEND template. Save the model somewhere nice and safe and obvious and then save the PROJECT. Give the model and project names that are easily remembered as being the FIRST PHASE of this process.

                     

                    Go to the TABLE screen. Under Data add a calculated field to the table. It will be a numeric field. The formula will be ROWNO(). This it the ROW NUMBER of the CURRENT DATA TABLE so it is a variable and is not directly fixed to the record so be aware of its limitations.

                     

                    Save the model again.

                     

                    Go to Project Exports, add a new export and create an export of the table to a known file name in a known place of your choice. The file can be an Excel file, or a fixed width text file or a delimited text file. Any should do the job - I tend to favour delimited text files. Have a look in the Menu>Options>Export and Clipboard and then the "Text Files" section to what is set up that defines how the files are exported. It doesn't matter what it is but it is nice to know what to expect!

                     

                    Make sure the export options GENERAL tab under "Field Names" specifies that field names are to be included in the first row of the output.

                     

                     

                    You may want to play around with some exports for experience if you are not familiar with them already.

                     

                    Finally export the file. (The file should ALWAYS be overwritten - set that in the appropriate screen of the output definition when prompted.)

                     

                    Save the model and the project once more.

                     

                    Now go back to you REPORT screen. Save the model again WITH A NEW NAME indicating a different step in the process. Open the DETAIL TEMPLATE and clear the fields. Pick one of the rows from the first block of rows and promote that to the sample area. (Button to do that looks like a page of text with a red line and an up arrow).

                     

                    Define a trap that will pick up all the lines in that block - including the dashes line but excluding the column names of course!

                     

                    Go to the table view and you should now see a revised version of your first table with the Patient details and ROW Number still in place but the other fields changed. The ROW NUMBERS IN EACH TABLE MUST GIVE YOU A MATCH ON THE LINES WE NEED TO JOIN. That is very important as I am sure you will appreciate.

                     

                    Repeat the process for defining the export and saving the model and projects as before but with new identifiable names for the second step in the process. (If you see any anomaly exports or whatever left over from the first model don't worry about them, they should do no harm.)

                     

                    Now, all that reads like a lot of work but in fact it is not much at all once you get into it - especially if you are already familiar with exports and project exports.

                     

                    Final step.

                     

                    Close everything ("Close All" from the FILE menu.)

                     

                    Then FILE>Open Database.

                     

                    Use the BROWSE function to navigate to wherever you saved the exported files. Open one of them - I would suggest the second block version first but you may find it does not matter. The system should read the file and ask a few question or offer some option settings which you don't need to worry about

                    other than select ALL fields.

                     

                    You will see a table.

                     

                    Now go to the MENU>DATA>EXTERNAL LOOKUPS

                     

                    Click on NEW (if this is the first time through, otherwise if you have allocated a name you can just choose an existing lookup) and BROWSE to find the other exported file. Follow the steps and when asked retain the suggested PATIENT to PATIENT field link (it may have a different name depending on what you called the field in your template) and also add the links for SSI to SSI, DOB to DOB and ROW NUMBER to ROW NUMBER (Again these names will be whatever you used in the models.)

                     

                    You should now see all the data you require in a single table in single record rows!

                     

                    Save the latest model and project files with a new name to indicate that it is the last step in the process.

                     

                    If you want to make some changes to the field order, add some more calcualted fields, add some summaries, define further exports or whatever, do that in this model an project and save the results as before.

                     

                    Now when you want to repeat the whole process you can just open the Projects and run the exports in the steps defined and the result will be there. Remember that the projects know a specific file name and location for the original report. If the name changes between analyses then you need to edit the project to pick up the new report. Or set up a generic name for the project internal use and copy or rename the new report to that generic name and location before processing.

                     

                    If this whole process is a very regular task you might want to consider automating the entire process, perhaps using a batch script to make each step run sequentially and creating the output automatically. Setting up the steps as projects makes that quite easy to do. Then you only need to click a desktop icon (or something similar to that idea) and have the results produced for you. If that is of interest we can cover it once you have the other parts of the process operational.

                     

                    Sorry it's long. It reads to sound more complex that it really is but I decided to assume you have not too much knowledge of these areas of functionality at the moment. My apologies if I am wrong on that.

                     

                    I hope this all helps.

                     

                    I think this is the quckest and most reliable way to get the results you need right now. If you qwant to automate with scripts or even using a VB or similar program at some point you can of course do so usning this process as the core of the automation.

                     

                    Good luck. I will be happy to answer any questions. I would prep some Projects for you but the internal references tend to be environment specific so to attempt to do so does not always make good sense.

                     

                    Have fun.

                     

                     

                    Grant

                     

                    [size="1"][ December 09, 2006, 04:42 AM: Message edited by: Grant Perkins ][/size]

                    • Extract Data From Multiple Lines
                      Grant Perkins

                      Hi Jane,

                       

                      Briefly to the points list;

                       

                      1 - Good, we know what is required.

                       

                      2 - A small, fixed number fixed number of lines might have offered some interesting possibilities - but we can ignore those for now!

                       

                      3 - Appends must start before detail BUT don't apply here in that way anyway because we really have a split detail line not a detail an append.

                       

                      4 - OK, we'll do it with Monarch!

                       

                       

                      Are you ready for this? This is not really newbie stuff but it won't be too difficult.

                       

                      I'm going to suggest 3 steps to this process so that we can sort of recreate the report (as much as we need to) to recombine the split lines and then work with the resulting table.

                       

                      We will be using a couple of versions of a simple model and then a final model using the Monarch Pro Database and External Lookup functions to create the final records.

                       

                      If you have access to the Monarch Learning Guide some of the sections could be useful pre-reading. Failing that read the Help sections that relate to the topics. Exports, Project exports, Databases and External LLokups would all be useful reading.

                       

                       

                      The second block of lines may be easier to use as a 'PRIME' detail record since they are easy to trap whereas the first block may not be so easy because we need to keep the line with dashes. So the easy traps are not helped by that. We have to have the SAME NUMBER OF ROWS in each part of the  extraction for each patient to get the match to work. If the report does not always deliver this things may get more complicated!

                       

                      Open the report and using one of the lines in the second block of lines

                       

                      [font="courier"]REVCODE  REVTYPE  REVUNIT    REVCHARGES

                      ABCD     EFG          1         XXXX.XX

                      XYZH     TPT          1        XXXXX.XX

                      TWSJ     GUT          1-       XXXXX.XX-  /font[/quote]generate a trap (A numeric in the column under REVUNIT and the . in the REVCHARGES column look good for that) then  paint in your fields making sure they are wide enough for the max possible field filling. That's your detail at this point.

                       

                      Save that template and add another template for the APPEND of the Patient Details. I would suggest trap on "Patient" and then grab the NAME, SSI code and DOB. We will need those later.

                       

                      Save the APPEND template. Save the model somewhere nice and safe and obvious and then save the PROJECT. Give the model and project names that are easily remembered as being the FIRST PHASE of this process.

                       

                      Go to the TABLE screen. Under Data add a calculated field to the table. It will be a numeric field. The formula will be ROWNO(). This it the ROW NUMBER of the CURRENT DATA TABLE so it is a variable and is not directly fixed to the record so be aware of its limitations.

                       

                      Save the model again.

                       

                      Go to Project Exports, add a new export and create an export of the table to a known file name in a known place of your choice. The file can be an Excel file, or a fixed width text file or a delimited text file. Any should do the job - I tend to favour delimited text files. Have a look in the Menu>Options>Export and Clipboard and then the "Text Files" section to what is set up that defines how the files are exported. It doesn't matter what it is but it is nice to know what to expect!

                       

                      Make sure the export options GENERAL tab under "Field Names" specifies that field names are to be included in the first row of the output.

                       

                       

                      You may want to play around with some exports for experience if you are not familiar with them already.

                       

                      Finally export the file. (The file should ALWAYS be overwritten - set that in the appropriate screen of the output definition when prompted.)

                       

                      Save the model and the project once more.

                       

                      Now go back to you REPORT screen. Save the model again WITH A NEW NAME indicating a different step in the process. Open the DETAIL TEMPLATE and clear the fields. Pick one of the rows from the first block of rows and promote that to the sample area. (Button to do that looks like a page of text with a red line and an up arrow).

                       

                      Define a trap that will pick up all the lines in that block - including the dashes line but excluding the column names of course!

                       

                      Go to the table view and you should now see a revised version of your first table with the Patient details and ROW Number still in place but the other fields changed. The ROW NUMBERS IN EACH TABLE MUST GIVE YOU A MATCH ON THE LINES WE NEED TO JOIN. That is very important as I am sure you will appreciate.

                       

                      Repeat the process for defining the export and saving the model and projects as before but with new identifiable names for the second step in the process. (If you see any anomaly exports or whatever left over from the first model don't worry about them, they should do no harm.)

                       

                      Now, all that reads like a lot of work but in fact it is not much at all once you get into it - especially if you are already familiar with exports and project exports.

                       

                      Final step.

                       

                      Close everything ("Close All" from the FILE menu.)

                       

                      Then FILE>Open Database.

                       

                      Use the BROWSE function to navigate to wherever you saved the exported files. Open one of them - I would suggest the second block version first but you may find it does not matter. The system should read the file and ask a few question or offer some option settings which you don't need to worry about

                      other than select ALL fields.

                       

                      You will see a table.

                       

                      Now go to the MENU>DATA>EXTERNAL LOOKUPS

                       

                      Click on NEW (if this is the first time through, otherwise if you have allocated a name you can just choose an existing lookup) and BROWSE to find the other exported file. Follow the steps and when asked retain the suggested PATIENT to PATIENT field link (it may have a different name depending on what you called the field in your template) and also add the links for SSI to SSI, DOB to DOB and ROW NUMBER to ROW NUMBER (Again these names will be whatever you used in the models.)

                       

                      You should now see all the data you require in a single table in single record rows!

                       

                      Save the latest model and project files with a new name to indicate that it is the last step in the process.

                       

                      If you want to make some changes to the field order, add some more calcualted fields, add some summaries, define further exports or whatever, do that in this model an project and save the results as before.

                       

                      Now when you want to repeat the whole process you can just open the Projects and run the exports in the steps defined and the result will be there. Remember that the projects know a specific file name and location for the original report. If the name changes between analyses then you need to edit the project to pick up the new report. Or set up a generic name for the project internal use and copy or rename the new report to that generic name and location before processing.

                       

                      If this whole process is a very regular task you might want to consider automating the entire process, perhaps using a batch script to make each step run sequentially and creating the output automatically. Setting up the steps as projects makes that quite easy to do. Then you only need to click a desktop icon (or something similar to that idea) and have the results produced for you. If that is of interest we can cover it once you have the other parts of the process operational.

                       

                      Sorry it's long. It reads to sound more complex that it really is but I decided to assume you have not too much knowledge of these areas of functionality at the moment. My apologies if I am wrong on that.

                       

                      I hope this all helps.

                       

                      I think this is the quckest and most reliable way to get the results you need right now. If you qwant to automate with scripts or even using a VB or similar program at some point you can of course do so usning this process as the core of the automation.

                       

                      Good luck. I will be happy to answer any questions. I would prep some Projects for you but the internal references tend to be environment specific so to attempt to do so does not always make good sense.

                       

                      Have fun.

                       

                       

                      Grant

                       

                      [size="1"][ December 09, 2006, 04:42 AM: Message edited by: Grant Perkins ][/size]

                      • Extract Data From Multiple Lines
                        jane888 _

                        Hi, Grant

                         

                        Thank you very much for this useful recommendation!

                         

                        I was trying the same approach except that I was doing it in Access rather than in Monarch.  I used AutoNo feature (similar to the RowNo) combined with patient demographic fields (SSN and DOB, etc) to join two tables exported from Monarch: one was Patient demographic with SvcDate blocks and one was Patient demographic with Revcode blocks.  When taking this approach, I'm making a big assumption that all the data were extracted in the order that showed in the report.  Hopefully it is.  Due to the crunch time frame I'm having, I'll temporarily continue with this approach.  Once I'm ready to move on, I'll try it again in Monarch using what you recommended and also try it in COBOL program on OS/400 to find the most effective one with minimal manual process.

                         

                        Many thanks for your help!  Have a great weekend! 

                        • Extract Data From Multiple Lines
                          jane888 _

                          Hi, Grant

                           

                          Thank you very much for this useful recommendation!

                           

                          I was trying the same approach except that I was doing it in Access rather than in Monarch.  I used AutoNo feature (similar to the RowNo) combined with patient demographic fields (SSN and DOB, etc) to join two tables exported from Monarch: one was Patient demographic with SvcDate blocks and one was Patient demographic with Revcode blocks.  When taking this approach, I'm making a big assumption that all the data were extracted in the order that showed in the report.  Hopefully it is.  Due to the crunch time frame I'm having, I'll temporarily continue with this approach.  Once I'm ready to move on, I'll try it again in Monarch using what you recommended and also try it in COBOL program on OS/400 to find the most effective one with minimal manual process.

                           

                          Many thanks for your help!  Have a great weekend! 

                          • Extract Data From Multiple Lines
                            Grant Perkins

                            Originally posted by jane888:

                               Due to the crunch time frame I'm having, I'll temporarily continue with this approach.  /b[/quote]Hi Jane,

                             

                            If it helps I have models that will do the job with the sample data. MOditfy those if required (I assume it will be required to some extent) for the real report and then define the exports and save the projects and you would be in good shape. It shoudl only take a few minutes.

                             

                            A few lines in a batch file for automation and you would have something that would work based on running the batch file. Manual input would probably be limited to getting the reports to the right place and with the right name for the process. Maybe a little more work to extend the batch script to include this part of the process as well (but external to Monarch of course).

                             

                            As for checking the credibility of the data (row orders and correct number of printed lines and so on) I am guessing that it may be possible to come up with some forms of checks to run for each line and highlight any lines that fail.

                             

                            That may not be full proof but would at least provide some degree of confidence. It should be easy to do with Monarch and also easy to extend quickly with additional potential checks as they become evident without the risk of compromising stable program code.

                             

                            If you can use the models send me a private message with an email address that will accept mails with attachments and I will provide them. (I would have sent this information via a PM but MPs are disabled in your profile.)

                             

                            HTH

                             

                             

                            Grant

                             

                            PS. Weekend? What weekend? 

                            • Extract Data From Multiple Lines
                              Grant Perkins

                              Originally posted by jane888:

                                 Due to the crunch time frame I'm having, I'll temporarily continue with this approach.  /b[/quote]Hi Jane,

                               

                              If it helps I have models that will do the job with the sample data. MOditfy those if required (I assume it will be required to some extent) for the real report and then define the exports and save the projects and you would be in good shape. It shoudl only take a few minutes.

                               

                              A few lines in a batch file for automation and you would have something that would work based on running the batch file. Manual input would probably be limited to getting the reports to the right place and with the right name for the process. Maybe a little more work to extend the batch script to include this part of the process as well (but external to Monarch of course).

                               

                              As for checking the credibility of the data (row orders and correct number of printed lines and so on) I am guessing that it may be possible to come up with some forms of checks to run for each line and highlight any lines that fail.

                               

                              That may not be full proof but would at least provide some degree of confidence. It should be easy to do with Monarch and also easy to extend quickly with additional potential checks as they become evident without the risk of compromising stable program code.

                               

                              If you can use the models send me a private message with an email address that will accept mails with attachments and I will provide them. (I would have sent this information via a PM but MPs are disabled in your profile.)

                               

                              HTH

                               

                               

                              Grant

                               

                              PS. Weekend? What weekend?