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

    capturing inconsistent data

    BaWahoo _

      I am running Monarch 8.02 and having some difficulty with capturing inconsistent data.  I thought of using the Floating Trap option, but it does not work. 

       

      The data below is what I am trying to capture, but sometimes certain data are blank.  Take for example the Edit Number in the last line or the Date Created is blank. When there are blank data the whole line shifts to the left.  I tried using the Floating Trap with “/”, but that did not work.  Any suggestion is greatly appreciated.

       

      XYZ COMPANY                                                                               

      Prebill XX-11                                                                               

      EDIT SUMMARY REPORT                                                                               

      DATE:12/07/2006                        PAGE: 145                                                                               

      TIME:05:42:02 PM                                                                               

      FILE: NEICNEICST      D.DBF                                                                               

      Edit    Date    Date                                                                               

      Number  Created Modified        Disabled        Description                                                                               

      -


        -


            -


            -


              -


           

      720    /  /    10/20/2006      No      BILLED     Set claims to a billed status                     

      227    10/14/2004      1/14/2006       No      NOPAY      Change carrier to NOPAY                   

      893    8/23/2006       8/23/2006       No      |RESUM BIL Resummed Total Charges                    

      721    /  /    /  /    No      SKIP EDITS Skip the remaining edits                                  

      953    2/5/2003        2/5/2003        No      SPOT       Remove carrier if equal to SPOT           

      950    2/5/2003        2/5/2003        No      SPOT       Check for prompt pay indicator   

      951    2/5/2003        2/5/2003        No      SPOT       Strip SPOT from cert ID                   

             /  /    /  /    No      CARRIER    Carrier assigned to primary claim for Accubill

        • capturing inconsistent data
          BaWahoo _

          I am running Monarch 8.02 and having some difficulty with capturing inconsistent data.  I thought of using the Floating Trap option, but it does not work. 

           

          The data below is what I am trying to capture, but sometimes certain data are blank.  Take for example the Edit Number in the last line or the Date Created is blank. When there are blank data the whole line shifts to the left.  I tried using the Floating Trap with “/”, but that did not work.  Any suggestion is greatly appreciated.

           

          XYZ COMPANY                                                                               

          Prebill XX-11                                                                               

          EDIT SUMMARY REPORT                                                                               

          DATE:12/07/2006                        PAGE: 145                                                                               

          TIME:05:42:02 PM                                                                               

          FILE: NEICNEICST      D.DBF                                                                               

          Edit    Date    Date                                                                               

          Number  Created Modified        Disabled        Description                                                                               

          -


            -


                -


                -


                  -


               

          720    /  /    10/20/2006      No      BILLED     Set claims to a billed status                     

          227    10/14/2004      1/14/2006       No      NOPAY      Change carrier to NOPAY                   

          893    8/23/2006       8/23/2006       No      |RESUM BIL Resummed Total Charges                    

          721    /  /    /  /    No      SKIP EDITS Skip the remaining edits                                  

          953    2/5/2003        2/5/2003        No      SPOT       Remove carrier if equal to SPOT           

          950    2/5/2003        2/5/2003        No      SPOT       Check for prompt pay indicator   

          951    2/5/2003        2/5/2003        No      SPOT       Strip SPOT from cert ID                   

                 /  /    /  /    No      CARRIER    Carrier assigned to primary claim for Accubill

          • capturing inconsistent data
            Nick Osdale-Popa

            Sorry, my original suggestion did not work.... please disregard this message.

            • capturing inconsistent data
              Nick Osdale-Popa

              Sorry, my original suggestion did not work.... please disregard this message.

              • capturing inconsistent data
                Nigel Winton

                Hi There

                Sorry I do not have time to have a good look at this, but this may get you started.

                Trap the whole line and make this one field. Then using calculated fields and the Instr function to locate to "No " to get a datum point you shoould be able to use other functions like Len and Instr for the first "/" to find specific parts of the data. Ie Right of the field using the length less the start point of "No " (note the space to clarify which No) will give you the text. Similar tricks should give the dates etc. If your "/" is found at position 1 then you have no record number. Substr function will also come in useful.

                I will try to have a better look some time.

                 

                Regards

                 

                Nigel

                • capturing inconsistent data
                  Nigel Winton

                  Hi There

                  Sorry I do not have time to have a good look at this, but this may get you started.

                  Trap the whole line and make this one field. Then using calculated fields and the Instr function to locate to "No " to get a datum point you shoould be able to use other functions like Len and Instr for the first "/" to find specific parts of the data. Ie Right of the field using the length less the start point of "No " (note the space to clarify which No) will give you the text. Similar tricks should give the dates etc. If your "/" is found at position 1 then you have no record number. Substr function will also come in useful.

                  I will try to have a better look some time.

                   

                  Regards

                   

                  Nigel

                  • capturing inconsistent data
                    mdk _

                    "Inconsistent Data" is my life, apparently.  Below please find a sample from one "report" I'm working from.  I am having difficulty grabbing the complete time from each and every line.  I have tried using the floating trap, but not successfully here- it recognizes all lines, but doesn't capture the full time data.  A secondary issue is that when I can grab the times, I cannot get the multiple line entry (each of the entries below is one line, except for second from bottom).  Any advice?

                     

                    8:15Introduction P.D. Wagner, MD, La Jolla, CA                          

                    8:20Patient's Perspective Speaker To Be Announced                       

                    8:25Classification Of PAH D.B. Badesch, MD, Denver, CO                  

                    8:50Genetics Of PAH C.G. Elliott, MD, Salt Lake City, UT                

                    9:15Pathogenesis of PAH L.J. Rubin, MD, La Jolla, CA                    

                    9:40Patient's Perspective Speaker To Be Announced                       

                    9:45Current Therapies: Prostanoids, Endothelin Receptor Antagonists     

                    And Phosphodiesterase Inhibitors R.J. Barst, MD, New York, NY           

                    10:10Future Therapies: Antiproliferatives And Others J.H. Newman,       

                    MD, Nashville, TN                                                       

                    10:35Stem Cell Therapy D.J. Stewart, MD, Toronto, ON, Canada

                    • capturing inconsistent data
                      mdk _

                      "Inconsistent Data" is my life, apparently.  Below please find a sample from one "report" I'm working from.  I am having difficulty grabbing the complete time from each and every line.  I have tried using the floating trap, but not successfully here- it recognizes all lines, but doesn't capture the full time data.  A secondary issue is that when I can grab the times, I cannot get the multiple line entry (each of the entries below is one line, except for second from bottom).  Any advice?

                       

                      8:15Introduction P.D. Wagner, MD, La Jolla, CA                          

                      8:20Patient's Perspective Speaker To Be Announced                       

                      8:25Classification Of PAH D.B. Badesch, MD, Denver, CO                  

                      8:50Genetics Of PAH C.G. Elliott, MD, Salt Lake City, UT                

                      9:15Pathogenesis of PAH L.J. Rubin, MD, La Jolla, CA                    

                      9:40Patient's Perspective Speaker To Be Announced                       

                      9:45Current Therapies: Prostanoids, Endothelin Receptor Antagonists     

                      And Phosphodiesterase Inhibitors R.J. Barst, MD, New York, NY           

                      10:10Future Therapies: Antiproliferatives And Others J.H. Newman,       

                      MD, Nashville, TN                                                       

                      10:35Stem Cell Therapy D.J. Stewart, MD, Toronto, ON, Canada

                      • capturing inconsistent data
                        Data Kruncher

                        mdk,

                         

                        Start by highlight your 9:45 row to build your detail template.

                         

                        In order to trap line, I would simply use a single numeric trap in column 1. Now select the entire line as one field - let's call this Data. In the advanced tab, set the End Field On: End of left justification.

                         

                        Now in the Table view, create a calculated field named Time, and give it the following formula:

                        [font="courier"]Left(Data,Instr(":",Data)+2)  /font[/quote]This will give you the time extracted properly, and will pick all of the text on multiple line fields.

                         

                        As to properly extracting ALL of the remaining fields for this particular data set, it's time for some Voodoo! That 9:45 row, with the comma in the presentation name, is particularly troublesome...

                         

                        Let us know how you make out.

                         

                        Kruncher

                        • capturing inconsistent data
                          Data Kruncher

                          mdk,

                           

                          Start by highlight your 9:45 row to build your detail template.

                           

                          In order to trap line, I would simply use a single numeric trap in column 1. Now select the entire line as one field - let's call this Data. In the advanced tab, set the End Field On: End of left justification.

                           

                          Now in the Table view, create a calculated field named Time, and give it the following formula:

                          [font="courier"]Left(Data,Instr(":",Data)+2)  /font[/quote]This will give you the time extracted properly, and will pick all of the text on multiple line fields.

                           

                          As to properly extracting ALL of the remaining fields for this particular data set, it's time for some Voodoo! That 9:45 row, with the comma in the presentation name, is particularly troublesome...

                           

                          Let us know how you make out.

                           

                          Kruncher

                          • capturing inconsistent data
                            Grant Perkins

                            Originally posted by mdk:

                            [font="courier"]8:15Introduction P.D. Wagner, MD, La Jolla, CA                          

                            8:20Patient's Perspective Speaker To Be Announced                       

                            8:25Classification Of PAH D.B. Badesch, MD, Denver, CO                  

                            8:50Genetics Of PAH C.G. Elliott, MD, Salt Lake City, UT                

                            9:15Pathogenesis of PAH L.J. Rubin, MD, La Jolla, CA                    

                            9:40Patient's Perspective Speaker To Be Announced                       

                            9:45Current Therapies: Prostanoids, Endothelin Receptor Antagonists     

                            And Phosphodiesterase Inhibitors R.J. Barst, MD, New York, NY           

                            10:10Future Therapies: Antiproliferatives And Others J.H. Newman,       

                            MD, Nashville, TN                                                       

                            10:35Stem Cell Therapy D.J. Stewart, MD, Toronto, ON, Canada /font[/quote][/b][/quote]Does that original look anything like the above?

                             

                            I se 2 wrapped lines but you only mention 1 so I wonder if what posted is at all close to what you are dealing with.

                             

                            You can get at the full time using the floating trap if you define the trap using a wide time field - i.e. one of the lines with hh:mm rather than just h:mm - and then set a trap as the alpha character which immediately follows. Bit of a problem if it is not always alpha of course.

                             

                            Define the time field as 5 chars wide. When the trap moves left for the shorter time field the field for 5 chars will be shortened to 4.

                             

                            A byproduct of that is that the wrapped line, assuming is does not begin with a numeric, will also be picked up in the 'text' field that you would define from the alpha trap point. unfortunately it won't be 'as one' with the rest of the text in the field from the line above.

                             

                            You could export the table produced as a new 'report' and then make an easy model for that version of the file.

                             

                            Another entirely different would be to to trap all lines based on the first character being numeric. (No floating here.) Paint one field for the entire width of the text including the time and then define the field as a multiline field using "End field on" = 'None of the above' in the advance properties tab.

                             

                            The line can then be sliced and diced, to some extent, using calculated fields. For example the time can easily be extracted by checking whether the ":" separator is the 2nd or 3rd character in the field and then taking the first 4 or the first 5 characters accordingly for the time field.

                             

                            I'm not sure what you can do with the rest of the line if you need to break it into component parts. There does not seem to be much to work with.

                             

                            What do you need to do with it? Is there anything available to you that would help to identify where to split the line?

                             

                            HTH.

                             

                             

                            Grant

                            • capturing inconsistent data
                              Grant Perkins

                              Originally posted by mdk:

                              [font="courier"]8:15Introduction P.D. Wagner, MD, La Jolla, CA                          

                              8:20Patient's Perspective Speaker To Be Announced                       

                              8:25Classification Of PAH D.B. Badesch, MD, Denver, CO                  

                              8:50Genetics Of PAH C.G. Elliott, MD, Salt Lake City, UT                

                              9:15Pathogenesis of PAH L.J. Rubin, MD, La Jolla, CA                    

                              9:40Patient's Perspective Speaker To Be Announced                       

                              9:45Current Therapies: Prostanoids, Endothelin Receptor Antagonists     

                              And Phosphodiesterase Inhibitors R.J. Barst, MD, New York, NY           

                              10:10Future Therapies: Antiproliferatives And Others J.H. Newman,       

                              MD, Nashville, TN                                                       

                              10:35Stem Cell Therapy D.J. Stewart, MD, Toronto, ON, Canada /font[/quote][/b][/quote]Does that original look anything like the above?

                               

                              I se 2 wrapped lines but you only mention 1 so I wonder if what posted is at all close to what you are dealing with.

                               

                              You can get at the full time using the floating trap if you define the trap using a wide time field - i.e. one of the lines with hh:mm rather than just h:mm - and then set a trap as the alpha character which immediately follows. Bit of a problem if it is not always alpha of course.

                               

                              Define the time field as 5 chars wide. When the trap moves left for the shorter time field the field for 5 chars will be shortened to 4.

                               

                              A byproduct of that is that the wrapped line, assuming is does not begin with a numeric, will also be picked up in the 'text' field that you would define from the alpha trap point. unfortunately it won't be 'as one' with the rest of the text in the field from the line above.

                               

                              You could export the table produced as a new 'report' and then make an easy model for that version of the file.

                               

                              Another entirely different would be to to trap all lines based on the first character being numeric. (No floating here.) Paint one field for the entire width of the text including the time and then define the field as a multiline field using "End field on" = 'None of the above' in the advance properties tab.

                               

                              The line can then be sliced and diced, to some extent, using calculated fields. For example the time can easily be extracted by checking whether the ":" separator is the 2nd or 3rd character in the field and then taking the first 4 or the first 5 characters accordingly for the time field.

                               

                              I'm not sure what you can do with the rest of the line if you need to break it into component parts. There does not seem to be much to work with.

                               

                              What do you need to do with it? Is there anything available to you that would help to identify where to split the line?

                               

                              HTH.

                               

                               

                              Grant