$TXT Created by COX,KEITH A at DMA.FO-ALBANY.MED.VA.GOV (KIDS) on Wednesday, 05/14/03 at 22:48 ============================================================================= Run Date: MAY 23, 2003 Designation: DG*5.3*244 Package : DG - REGISTRATION Priority: Mandatory Version : 5.3 SEQ #445 Status: Released Compliance Date: JUN 23, 2003 ============================================================================= Associated patches: (v)XU*8*211 <<= must be installed BEFORE `DG*5.3*244' (v)XU*8*240 <<= must be installed BEFORE `DG*5.3*244' (v)DI*22*116 <<= must be installed BEFORE `DG*5.3*244' (v)DG*5.3*489 <<= must be installed BEFORE `DG*5.3*244' (v)DG*5.3*494 <<= must be installed BEFORE `DG*5.3*244' (v)DG*5.3*506 <<= must be installed BEFORE `DG*5.3*244' (v)XU*8*301 <<= must be installed BEFORE `DG*5.3*244' Subject: PATIENT NAME STANDARDIZATION Category: - Routine - Data Dictionary - Other Description: ============ An effort is underway within VHA to establish a greater uniformity to the manner in which person names are stored in computer databases. The benefits of achieving this include greater reliability in matching names from different systems, more powerful person lookup and selection mechanisms and the ability to accurately reformat names for correspondence salutations, etc. This patch implements name standardization for eleven fields in the PATIENT (#2) file, including the patient NAME (#.01) field. While the standardization of names attempts to reduce the input format variations that may occur, for reasons of patient safety, respect and ethnic sensitivity it is important for the approach that is implemented to retain the name value in a format which accurately represents the name as provided by the patient. Name Standardization Rules: =========================== Once standardized, name fields will utilize the format FAMILY_NAME GIVEN_NAME(S)MIDDLE_NAME(S)SUFFIX(ES). Name values will be uppercase alpha characters and contain no other characters except spaces, hyphens, apostrophes and a single comma. Specific cleanup and formatting rules are as follows: o All lowercase alpha characters transformed to uppercase o Embedded accent grave characters ( ` ) will be transformed to apostrophes o Parenthetical text will be removed (text captioned by parenthesis, bracket or brace characters) o All characters other than uppercase alpha, comma, hyphen, apostrophe, numeric and space transformed to a space o All double spaces transformed to a single space o All double hyphens transformed to a single hyphen o All double apostrophes transformed to a single apostrophe o All comma_(space, hyphen or apostrophe) and (space, hyphen, or apostrophe)_comma combinations transformed to a comma o All space_(hyphen or apostrophe) and (hyphen or apostrophe)_space combinations transformed to a space o All hyphen_apostrophe or apostrophe_hyphen combinations transformed to a hyphen o All numeric characters will be removed except for the combinations "1ST", "2ND", "3RD" and "4TH" through "10TH" which must have a space, comma or null value preceding and following to be considered valid o Numeric birth position identifiers ("1ST", "2ND", "3RD", etc.) will be transformed to corresponding roman numeral values, I, II, III, etc. o All characters other than uppercase alpha will be removed from the beginning and end of the family name o All characters other than uppercase alpha will be removed from the beginning and end of the given_middle_suffix name value o Suffixes moved from the immediate left of the comma to the end of the name o "DR" moved to the end of the name if it is found at the immediate right of the comma o Any suffixes between commas following the family name will be moved to the end of the name o "NMI" or "NMN" will be removed if either is used as the middle name If the length of the name value reconstructed from name components is greater than the length allowed for the name field, the following actions will be taken sequentially until the resulting value falls within length restrictions: o Truncate the middle name from the right-most position until only the initial character remains o Drop the suffix o Truncate the given name from the right-most position until only the initial character remains o Truncate the family name from the right-most position Additional rules relative to name formatting: o If there are multiple family, given, middle or suffix values they will be separated by spaces o The user's input value must contain at least one comma in order to add or edit a name value [except for the MOTHER'S MAIDEN NAME (#.2403) field] o At a minimum, a patient name value must consist of a family name followed by a comma (given, middle and suffix may be omitted but this method of entry should only be used for patients whose legal name is a single value such as "MADONNA") Data Dictionary Changes: ======================== The following fields have had their input transforms, short description and long description updated. Each field has a new index type cross reference added which keeps the field in synch with the corresponding NAME COMPONENTS (#20) file record. Additionally, the NAME (#.01) field has a new index type cross reference added which maintains the 'NOP' special lookup cross reference. Field Name Field # ----------------------------------- ------- NAME .01 K-NAME OF PRIMARY NOK .211 K2-NAME OF SECONDARY NOK .2191 FATHER'S NAME .2401 MOTHER'S NAME .2402 MOTHER'S MAIDEN NAME .2403 E-NAME .331 E2-NAME OF SECONDARY CONTACT .3311 D-NAME OF DESIGNEE .341 DISPOSITION LOG-IN DATE/TIME (multiple) 1000 ATTORNEY'S NAME 30 ALIAS (multiple) 1 ALIAS .01 The following index type cross references have been added: Index name File or sub-file Field ------------ ---------------- -------- NOP 2 .01 ANAM01 2 .01 ANAM1001 2.101 30 ANAM201 2.01 .01 ANAM211 2 .211 ANAM2191 2 .2191 ANAM2401 2 .2401 ANAM2402 2 .2402 ANAM2403 2 .2403 ANAM331 2 .331 ANAM3311 2 .3311 ANAM341 2 .341 Eleven new fields have been added. These fields contain pointers to the NAME COMPONENTS file and represent the relationship between the existing PATIENT file fields listed above and their corresponding NAME COMPONENTS record. Field Name Field # ----------------------------------- ------- NAME COMPONENTS 1.01 K-NAME COMPONENTS 1.02 K2-NAME COMPONENTS 1.03 FATHER'S NAME COMPONENTS 1.04 MOTHER'S NAME COMPONENTS 1.05 MOTHERS MAIDEN NAME COMPONENTS 1.06 E-NAME COMPONENTS 1.07 E2-NAME COMPONENTS 1.08 D-NAME COMPONENTS 1.09 ALIAS (multiple) 1 ALIAS COMPONENTS 100.03 DISPOSITION LOG-IN DATE/TIME (multiple) 1000 ATTORNEY'S NAME COMPONENTS 100.21 Changes to Options: =================== The name components of the patient NAME (#.01) field can be viewed with the 'View Registration Data' [DG REGISTRATION VIEW] option. Name components for the NAME field may be edited through screen 1 of the demographic edit functionality contained in the 'Register a Patient' [DG REGISTER PATIENT] and 'Load/Edit Patient Data' [DG LOAD PATIENT DATA] options. The following is an example of what this might look like: PATIENT DEMOGRAPHIC DATA, SCREEN <1> BELL,DONALD MD; 000-00-3456 NSC VETERAN ============================================================================== [1] Name: BELL,DONALD MD SS: 000-00-3456 DOB: MAY 12,1949 Family: BELL [2] Alias: BELL,DAVIS G 000-00-3434 Given: DONALD SMITH,DUCKY Middle: DRAKE,D 000-00-4321 Prefix: MR JONES,DOE SMITH 000-00-5567 Suffix: MD BELL,ORMSBY R Degree: PHD < More alias entries on file > [3] Remarks: SEE SOCIAL WORKER NOTES [4] Permanent Address: [5] Temporary Address: APARTMENT 43 LONESOME ONE MOTEL BUILDING G ROOM 999 93 FAIRMONT PLAZA 53935 WEST HIGHWAY 50 PITTSBURGH,PA 15201 MOO MESA,NV 89898 County: ALLEGHENY (003) County: ESMERALDA (021) Phone: 555-444-3333 Phone: 777-890-7654 Office: 555-432-1234 From/To: DEC 31,2001-APR 30,2002 to CONTINUE, 1-5 or ALL to EDIT, ^N for screen N or '^' to QUIT: 1 NAME: BELL,DONALD MD// FAMILY (LAST) NAME: BELL// BELL GIVEN (FIRST) NAME: DONALD// DONALD MIDDLE NAME: JOE PREFIX: MR// MR SUFFIX: MD// MD DEGREE: PHD// PHD Ok to file 'BELL,DONALD JOE MD' and its name components? Yes// (Yes) SOCIAL SECURITY NUMBER: 000003456// DATE OF BIRTH: 05/12/1949// If the patient's name components have been edited, the user is prompted to confirm that they wish to file the newly reconstructed name and its components. If the user responds "no" the name component edits and newly reconstructed name value will not be filed. Changes to PATIENT file lookups: ================================ In cases where standardized representation (stripped of spaces, punctuation, etc.) of a patient name is different than the value stored in the NAME (#.01) field, the standardized value is stored in the 'NOP' cross reference described above. This facilitates lookup matches to be identified for common name input variations based on a single user input value. For example, the entry of OBRIEN will return OBRIEN, O'BRIEN and O BRIEN variations. Previously, lookups on the Social Security Number were limited to the first character of the last name and the last four digits of the SSN; just the last four digits of the SSN; or the complete nine or ten digit SSN value. The SSN lookup logic has been modified allow two or more numeric characters to be applied as a lookup value. Programmer API: =============== A new application programming interface (API), LIST^DPTLK1 has been created which facilitates the non-interactive return of patient record match lists based on a specified lookup value. Access to this API is available by a controlled subscription Integration Agreement (IA). Interactions with the MPI: ========================== When the patient name conversion utility distributed in this patch is performed, any change to Patient Name or Mother's Maiden Name (other than just a comma added to Mother's Maiden Name) will generate an A31 update message to the MPI for any patient that has a National ICN. Test Sites: =========== Battle Creek, MI Bay Pines, FL Charleston, SC Clarksburg, WV Dublin, GA Grand Junction, CO Lexington, KY Miami, FL New Orleans, LA Northern California HCS Reno, NV Tennessee Valley HCS Routine Summary: ================ The following is a list of routines included in this patch. The second line of each of these routines will look like: ;;5.3;Registration;**[patch list]**;Aug 13, 1993 CHECK^XTSUMBLD results Routine name Before Patch After Patch Patch List ============ ============ =========== ========== DG53244P N/A 5765684 244 DG53244Q N/A 20392109 244 DG53244R N/A 31006811 244 DG53244S N/A 8984289 244 DG53244T N/A 25590231 244 DG53244U N/A 13284605 244 DG53244V N/A 13356281 244 DGBUL 3509242 4065107 31,244 DGDDC 4841232 5005727 489,244 DGLOCK2 11684785 12302800 18,244 DGNFUNC 901162 244095 149,244 DGRP1 14523092 14821010 109,161,506,244 DGRPDD1 18965250 19626105 72,136,244 DGRPE 24923895 26258242 32,114,139,169, 175,247,190,343, 397,342,454,415, 489,506,244 DPTLK 17131072 16844003 32,72,93,73,136, 157,197,232,265, 277,223,327,244 DPTLK1 11676550 12893756 32,50,197,249, 317,391,244 DPTLK2 16365827 14251761 32,197,214,244 DPTNAME N/A 7767521 244 DPTNAME1 N/A 7789567 244 Installation Instructions: ========================== This patch can be installed with users on the system, however it should be installed during off hours to minimize disruption to users. An installation post-init will build the 'NOP' cross reference. The amount of time to accomplish this will vary depending on the size of the PATIENT (#2) file but will take at least several minutes. On a quiet VMS/Cache system it took about 7 minutes to build the 'NOP' cross reference for 100,000 patients. If the installation is not queued the KIDS progress bar will be updated to reflect the completion rate of this activity. 1. Review your mapped set. If the routines listed in the ROUTINE SUMMARY section are mapped, they should be removed from the mapped set at this time. 2. Use the INSTALL/CHECK MESSAGE option on the PackMan menu. 3. From the Kernel Installation and Distribution System Menu, select the Installation menu. 4. From this menu, you may elect to use the following options (when prompted for the INSTALL NAME, enter DG*5.3*244). a. Backup a Transport Global - this option will create a backup message of any routines exported with the patch. It will NOT backup any of the other changes. b. Compare Transport Global to Current System - this option will allow you to view all changes that will be made when the patch is installed. It compares all components of the patch (routines and options). c. Verify Checksums in Transport Global - this option will allow you to ensure the integrity of the routines that are in the transport global. 5. Use the Install Package(s) option and select the package DG*5.3*244. 6. When prompted, 'Want KIDS to INHIBIT LOGONs during the install? YES//' answer NO. 7. When prompted 'Want to DISABLE Scheduled Options, Menu Options, and Protocols? YES//', respond YES. When prompted to select the options you would like to place out of order, enter the following: DG REGISTRATION VIEW View Registration Data DG LOAD PATIENT DATA Load/Edit Patient Data DG REGISTER PATIENT Register a Patient 8. If any routines were unmapped as a part of step 1, they should be returned to the mapped set once the installation has run to completion. 9. Upon successful completion of the KIDS install, you may delete the routines DG53244P, DG53244Q, DG53244R and DG53244S. 10. Use PRINT^DG53244T to print a report of name value exceptions which will be encountered when the name conversion is performed. This report should be reviewed to determine if there is data that needs to be cleaned up prior to the name conversion. This report provides for the printing of statistics and detailed name exception lists. The detailed format should be used to review name exceptions prior to doing the conversion. Individual fields or all fields may be selected for review. It is recommended that all fields be reviewed. This report groups name exceptions by the following categories: 1 Name value contains no comma 2 Parenthetical text is removed from name 3 Name value cannot be converted 4 Characters are removed from the name Of these types, the first three are considered important to review. Names with no comma may not be parsed correctly by Kernel name standardization utilities. Names with parenthetical text, e.g. "(INELIGIBLE)" may require clean up prior to conversion. Names that cannot be converted should be corrected or deleted. The fourth category will apply to the majority of name changes. These are instances where characters such as periods, have been removed from the name and probably do not require review. You may review all four categories but limiting the report to the first three will produce a more relevant and manageable list. 11. Use CONVERT^DG53244T to perform the name conversion. This process will establish a NAME COMPONENTS (#20) file record for each data value associated with the 11 fields to be converted. Data values will be transformed to match the rules listed in the patch description above as indicated. Due to the volume of activity, the patient name change bulletin will be disabled during the conversion. The conversion is a lengthy and reasonably intensive process. It should be performed during non-business hours. The name conversion should not be performed concurrently with the patient record merge process. Patient names in records that are being merged by the patient record merge functionality will not be converted. To avoid the system and network messaging impact that would result from the volume of changes that occur during the conversion process, the HL7 messages that would normally be generated by the VAFC ADT-A08 SERVER protocol are disabled for the activity related to the conversion. Local subscribers to this protocol such as VistA Imaging and RAI/MDS will still receive HL7 notification messages through a special protocol created for the conversion process. To assist sites in managing impacts to users, system journaling, etc. limitations may be specified when queuing the conversion process which will stop the conversion after a specified entry number or time of day. This will allow portions of the conversion to be performed on subsequent days. For example, the entry number limitation may be used to process a group of 20,000 patients; the date/time limitation would allow the site to start the process at 10 pm and have it stop at 6 am the next day. CONVERSION PROCESSING IMPACTS: Although eleven fields in the PATIENT file are being converted, not all of them may contain data in a given record. In testing it was observed that approximately five NAME COMPONENTS records were created per PATIENT record. A quiet VMS/Cache system was able to process about 5,000 PATIENT records per hour. SYSTEM JOURNALING: A VMS/Cache system journaling all globals including ^TMP and ^XTMP used about one gigabyte of journal space per 10,000 PATIENT records processed. Approximately two thirds of this activity was ^TMP and ^XTMP. Sites may need to utilize the conversion limiting methods described above to avoid running out of journal space during a conversion processing session. GLOBAL GROWTH: Due to the creation of records in the NAME COMPONENTS (#20) file, the ^VA global can be expected to grow approximately 73 megabytes for every 100,000 PATIENT file records converted. STOPPING THE CONVERSION PROCESS: It is important that each session of the conversion process ends gracefully. This will insure the reactivation of the patient name change bulletin, tasking of HL7 name change messaging and the accurate recording of the conversion progress to that point. System utilities should not be used to restore (kill) the conversion process. In addition to the limitations which may be specified during queuing, the conversion process may be stopped gracefully by using the STOP^DG53244T entry point or by using the 'Stop task' functionality of the 'TaskMan User' [XUTM USER] option on the 'User's Toolbox' [XUSERTOOLS] menu. When the conversion process ends or is stopped with the methods noted above, a mail message is sent to the user that initiated the conversion process. This message contains statistics regarding records converted, exception types, run time, processing rate, etc. If the process is stopped without processing the entire PATIENT (#2) file, it is important that it be subsequently re-tasked to run as needed to complete the conversion. The results of the conversion may be printed with PRINT^DG53244T. RETENTION OF REPORT AND CONVERSION DATA: Data generated for reporting (prior to conversion) will be retained for 90 days after the first report generation. Once the conversion process has begun, conversion data will be retained for 90 days after the conversion was last started. 12. When the conversion has processed to completion and all desired review reports have been obtained, the routines DG53244T, DG53244U and DG53244V may be deleted. Routine Information: ==================== Routine Name: - DG53244P Routine Checksum: Routine Name: - DG53244Q Routine Checksum: Routine Name: - DG53244R Routine Checksum: Routine Name: - DG53244S Routine Checksum: Routine Name: - DG53244T Routine Checksum: Routine Name: - DG53244U Routine Checksum: Routine Name: - DG53244V Routine Checksum: Routine Name: - DGBUL Routine Checksum: Routine Name: - DGLOCK2 Routine Checksum: Routine Name: - DGNFUNC Routine Checksum: Routine Name: - DGRP1 Routine Checksum: Routine Name: - DGRPE Routine Checksum: Routine Name: - DPTLK1 Routine Checksum: Routine Name: - DPTLK2 Routine Checksum: Routine Name: - DPTNAME Routine Checksum: Routine Name: - DGDDC Routine Checksum: Routine Name: - DPTNAME1 Routine Checksum: Routine Name: - DPTLK Routine Checksum: Routine Name: - DGRPDD1 Routine Checksum: ============================================================================= User Information: Entered By : SHER,JOEL Date Entered : AUG 03, 1999 Completed By: FAIN,ZACHARY Date Completed: MAY 23, 2003 Released By : ALEXANDER,DIANNE Date Released : MAY 23, 2003 ============================================================================= Packman Mail Message: ===================== $END TXT