Unit Editing and Data Validation Concepts. Objectives

Transcription

1 Unit Editing and Data Validation Concepts Presented by: Bruce Hallman and Hal Dooley 1 Objectives Become familiar with the types of Unit Edits/Validation Tests and how to research and resolve them 2 1

2 Agenda Unit Data Editing Unit Edit Process Examples Unit Data Validation Unit Validation Process Examples 3 Unit Data Editing 4 2

3 Unit Data Editing Topics Resources for reporting requirements Data grades Types of Edits Field, Logical, Relational, and Net Editing path in Data Collection Edit matrices Edit creation process 5 Resources for Reporting Requirements Statistical Plan Filed rules Unit Statistical Reporting Guidebook Reporting examples Edit and Validation Test Matrices WCIO Data Specifications Manual WCSTAT Online Tools DCA Access Online Unit Pre-Edit Service/Tool Data Transfer via the Internet 6 3

4 Editing Path in Data Collection Pre-Edit Submit Data for Processing Front-End Edits Data Accepts/Rejects Net Edits Submission Results/Reports Corrections/Replacements Update source system if using DCA Access Online 7 Pre-Edit Options Unit Pre Edit Service Submit electronic files to see errors before submitting to production Unit Pre Edit Tool Allows you to update your units online and submit 8 4

5 Unit Edit Data Grades Error Class (Data Grade Description) Data Grade (Level of Severity) Error-Free Informational 1 Suspect 2 Default 3 Priority Error 4 Priority/Critical Error 5 Reject 9 9 Edit Types Field: Checks field values, e.g., numeric, alphanumeric, and valid codes Logical: Checks that the data reported in a field is valid in relation to: One or more fields on the record or Within other records on the unit report 1 5

6 Edit Types Relational: Compares data reported in a field: With other data in the same submission With data on the database Net (Logical and Relational): Uses the current header and latest exposure and loss information at each report level to: Identify error conditions Clear errors on corresponding unit reports 11 Unit Submission Results Report 12 6

7 Unit Edit Matrix 13 Unit Edit Matrix 14 7

8 Unit Edit Matrix 15 Future Enhancement Unit Edit Matrix 16 8

9 Edit Creation Process New Edit Idea (data quality concern or rule requirement) Internal Review External Edit Task Group Future Edit Enhancement Matrix Production Edit Matrix 17 Unit Data Editing Match Game Questions Identifies the Severity of an Error Checks for Numeric/Alpha Values First Step in Editing Path Error Free Data Location of Edit Matrix Answers Pre Edit Unit Statistical Reporting Guidebook Data Grades Field Edit Data Grade 18 9

10 Unit Data Edits Examples 19 Before We Begin, a Few Notes The focus should be on reporting accurate data rather than clearing errors There can be multiple ways to clear an error depending on what data is in error Update source system if using DCA Access Online 2 1

11 Unit Data Edits Edit Description LOSSES WITHOUT EXPOSURE (Front-End) LOSSES WITHOUT EXPOSURE (Net) 75-3 EXPERIENCE MODIFICATION FACTOR INCONSISTENT INCONSISTENT CLASS CODE ACROSS REPORT LEVELS OPEN LOSSES NOT UPDATED ON SUBSEQUENT LOSS MATCHING LOSS MATCHING 21 es Without Exposure Relational Edit : Data Grade: 5 (9 for MN and NC) Element: Classification (on Record) Edit Description: The loss class code on the incoming unit must have a corresponding exposure class code with a nonzero exposure amount

12 es Without Exposure Concept: We don t expect a loss to be paid for a worker that wasn t covered under the policy. Example: A loss is reported for a roofing injury under Classification 5551, but no roofers are covered on the policy i.e., there is no exposure record for es Without Exposure Set Data Issue Up Exposure Class Class Report # and Seq. # Exposure Amount 1- R $1, R 1 $25 Data Grade 2- P 1 $25 Matching R 1 $3 exposure 2-1 class code R $5 5 is missing

13 es Without Exposure Correcting Method 1 Exposure Class Class Report # and Seq. # Exposure Amount 1- R $1, R 1 $25 Data Grade 2- P 1 R 1 $25 $3 2-1 R $ P 2 R $5 $5 Based on research, it is determined that the class code on the loss is incorrect. 25 es Without Exposure Correcting Method

14 es Without Exposure Correcting Method 1 For additional priority errors, you can check here. The current data grade for this claim is es Without Exposure Correcting Method

15 es without Exposure Correcting Method 1 The report level data grade is now zero. 29 es without Exposure Correcting Method 1 For further confirmation, you can check here. The report level data grade is now zero. The current data grade for this claim is now zero. 3 15

16 es Without Exposure Correcting Method 2 Exposure Class Class Report # and Seq. # Exposure Amount 1- R $1, R 1 $25 Data Grade 2- P 1 R 1 $25 $3 2-1 R $5 5 Based on research, it is determined that the class code on the loss is correct. 31 es Without Exposure Correcting Method 2 Exposure Class Class Report # and Seq. # Exposure Amount 1- R $1, R 1 $ R 5551 $1,5 2- P 1 R 1 $25 $3 2-1 R $5 5 Data Grade Based on research, it is determined that the class code on the loss is correct

17 Net Edits Compare reported values on incoming units with data on NCCI s database Apply to the newly reported or previously submitted data Corrections must be submitted in sequential report level order, starting with the earliest report level that the error occurred 33 Front-End vs. Net Aspect Front-End Net Aspect Front-End Net Field Yes No Field Yes No Logical Yes Yes Logical Yes Yes Relational Yes Yes Relational Yes Yes Unit Pre-Edit Yes No Tool/Service Unit Pre-Edit Yes No Tool/Service Rejected Units Yes No Rejected Units Yes No Edit Operation Edits operate on All edits operate when submitted records any unit is received Applying Edits Incoming Unit Incoming Unit and unit(s) on database Clearing Errors Yes, by replacing the record Yes, even to records previously submitted 34 17

18 es Without Exposure Net Net Edit : Data Grade: 5 Element: Classification (on Record) Edit Description: The loss class code on a previously reported unit must have a corresponding exposure class code with a non-zero exposure amount. 35 es Without Exposure Net Data Net Set Issue Up Report # and Seq. # Exposure Class 1- R R 8742 Exposure $1, $75 Class Amount R 3 $25 Data Grade 5 2- P 3 R 3 $25 $ P $1, Net

19 Report # and Seq. # es Without Exposure Net Correcting Method 1 Exposure Class 1- R R 8742 Exposure $1, $75 Class Amount R 3 $25 5 Data Grade 2- P 3 R 3 $25 $ P $1, Based on research, it is determined that the class code on 3 is correct and the removal of exposure Class was incorrect. 37 Report # and Seq. # es Without Exposure Net Correcting Method 1 Exposure Class 1- R R 8742 Exposure $1, $75 Class Amount R 3 $25 5 Data Grade 2- P 3 R 3 $25 $ P $1, 1-2 R $1,

20 Report # and Seq. # Exposure Class 1- R R 8742 es Without Exposure Net Correcting Method 2 Exposure $1, $75 Class Amount R 3 $25 5 Data Grade 2- P 3 R 3 $25 $ P $1, Based on research, it is determined that the class code on the claims is incorrect and the removal of exposure Class was correct. 39 Report # and Seq. # Exposure Class 1- R R 8742 es Without Exposure Net Correcting Method 2 Exposure $1, $75 Class Amount R 3 $25 5 Data Grade 2- P 3 R 3 $25 $ P $1, 1-2 P 3 R P 3 R $25 $25 $3 $3 4 2

21 Match Game Questions Two Types of Errors Highest Current Data Grade Pre Edit Applies To View Data Grade 4 and 5 Errors Possible Corrective Actions Answers Front End Edit Update Exposure or Classification Front End and Net Report Level Outstanding Priority Errors 41 Experience Modification Factor Inconsistent Relational Edit : 75-3 Net Edit : 75-7 Relational Edit Data Grade: 4 (9 for MN and NC) Net Edit Data Grade: 4 Element: Experience Modification Factor Edit Description: The Experience Modification Factor on the incoming unit must be the same for all class codes with the same Experience Modification Effective Date

22 Experience Modification Factor Inconsistent Concept: We expect the Experience Modification Factor to be the same for all class codes subject to experience rating that share the same Experience Modification Effective Date. Example: Class 8742 with a 1/1/9 Experience Modification Effective Date has an Experience Modification Factor of.79. Class has.82 for the same Experience Modification Effective Date. 43 Experience Modification Factor Inconsistent Report # and Seq. # Exposure Exp Mod Experience Class Eff Date Exposure Mod 1- R /1/9 $25,.79 Data Grade 1-1 P 8742 R 8742 R 1/1/9 1/1/9 1/1/9 $25, $275, $475, Inconsistent Modification Factors

23 Experience Modification Factor Inconsistent Report # and Seq. # Exposure Exp Mod Experience Class Eff Date Exposure Mod 1- R /1/9 $25,.79 Data Grade 1-1 P 8742 R 8742 R 1-2 P 8742 R /1/9 1/1/9 1/1/9 1/1/9 1/1/9 $25, $275, $475, $275, $275, Errors are cleared Based on research, it is determined that the Experience Mod should not have changed. 45 Experience Modification Factor Inconsistent 46 23

24 Experience Modification Factor Inconsistent 47 Experience Modification Factor Inconsistent 48 24

25 Experience Modification Factor Inconsistent 49 Inconsistent Class Across Report Levels Relational Edit : Net Edit : Data Grade: 9 (Relational)/5 (Net) Element: Classification (on Record) Edit Description: The class code must be reported consistently across all report levels for this loss. 5 25

26 Inconsistent Class Across Report Levels Concept: The reported class code of the injured worker is not expected to change over the life of the loss. Example: A claim is reported under Classification (office and clerical) on the first report. Then on the second report, the class is reported as 8742 (Salesperson). 51 Inconsistent Class Across Report Levels Data Issue Report # and Seq. # Class Amount Data Grade 1- R 4 $25 2- P 4 R 4 3- P 4 R 4 Unit Rejected 8742 $25 $3 $3 $

27 Inconsistent Class Across Report Levels Correcting Method 1 Report # and Seq. # Class Amount Data Grade 1- R 4 $25 2- P 4 R 4 3- P 4 R 4 Unit Rejected 8742 $25 $3 $3 $35 9 Based on research, it is determined that the class code on the 3- is incorrect. 53 Inconsistent Class Across Report Levels Correcting Method 1 Report # and Seq. # Class Amount Data Grade 1- R 4 $25 2- P 4 R 4 $25 $3 3- P 4 R 4 Unit Correct Data and Rejected Resubmit Unit 8742 $3 $35 9 Based on research, it is determined that the class code on the 3- is incorrect

28 Inconsistent Class Across Report Levels Correcting Method 2 Report # and Seq. # Class Amount Data Grade 1- R 4 $25 2- P 4 R 4 3- P 4 R 4 Unit Rejected 8742 $25 $3 $3 $35 9 Based on research, it is determined that the class code on the 3- is correct. 55 Inconsistent Class Across Report Levels Correcting Method 2 Report # and Seq. # Class Amount Data Grade R P 4 $25 R $ P 4 R 4 3- P 4 R 4 Unit Rejected 8742 $25 $3 5 $3 $35 9 Based on research, it is determined that the class code on the 3- is correct

29 Inconsistent Class Across Report Levels Correcting Method 2 Report # and Seq. # Class Amount Data Grade R P 4 $25 R $ P 4 R $3 $25 $3 3- P $3 R $35 9 Unit Resubmit Rejected Rejected 3- Unit Based on research, it is determined that the class code on the 3- is correct. 57 Open es Not Updated on Subsequent Relational Edit : Data Grade: 5 (NC and WI are not included) Element: Miscellaneous edit (on Header Record) Edit Description: The incoming unit does not contain updates for all open claims that exist at the previous report level

30 Open es Not Updated on Subsequent Concept: All open losses must be reported on subsequent units. Example: If a loss is open at the 1st report, the loss must be updated and reported on the 2nd report. 59 Open s Not Updated on a Subsequent Report Level Data Issue Report # and Seq. # 1- Header 2- Header R - 6 R - 7 P - 7 R - 7 Status O - Open O - Open O - Open C - Closed Class Amount $25 $125 $125 $25 $25 $3 Data Grade Since 6 was open at the 1st report level and not updated at the 2nd report, the header record for the 2- receives a DG 5 error. Then the header record of the 1- receives a Net DG 5 error. 5 5 Edit Writes the Error Edit

31 Open es Not Updated on a Subsequent Report Level Correction Method 1 Report # and Seq. # 1- Header 2- Header R - 6 R - 7 P - 7 R - 7 Status O - Open O - Open O - Open C - Closed Class Amount $125 $25 $25 $3 Data Grade 5 5 After carrier research, it is determined that 6 was still open at the 1st report and requires an update. 61 Open es Not Updated on a Subsequent Report Level Correction Method 1 Report # and Seq. # 1- Header 2- Header 2-1 Header R - 6 R - 7 P - 7 R - 7 P - 6 R - 6 Status O - Open O - Open O - Open C - Closed O - Open C - Closed Class Amount $25 $125 $125 $25 $25 $3 $125 $175 Data Grade Clears the Error No Error These edits write to the header record. When the 2-1 arrived, it became the latest header at the 2 nd report and it has no errors. The DG 5 on the 2- is just historical

32 Open s Not Updated on a Subsequent Report Level Correction Method 2 Report # and Seq. # 1- Header 2- Header R - 6 R - 7 P - 7 R - 7 Status O - Open O - Open O - Open C - Closed Class Amount $125 $25 $25 $3 Data Grade 5 5 After carrier research, it is determined that 6 should have been closed at the 1st report. 63 Open s Not Updated on a Subsequent Report Level Correction Method 2 Report # and Seq. # 1- Header 2- Header 1-1 Header R - 6 R - 7 P - 7 R - 7 P - 6 R - 6 Status O - Open O - Open O - Open C - Closed O - Open C - Closed Class Amount $25 $125 $125 $25 $25 $3 $125 $125 Data Grade Submitting a 1-1 loss correction that closes 6 will clear the 1st report error by replacing the current header

33 Open s Not Updated on a Subsequent Report Level Correction Method 2 Report # and Seq. # 1- Header 2- Header 1-1 Header R - 6 R - 7 P - 7 R - 7 P - 6 R - 6 Status O - Open O - Open O - Open C - Closed O - Open C - Closed Class Amount $25 $125 $125 $25 $25 $3 $125 $125 Data Grade Clears the Error Next, Net edit will operate to physically clear the error on the Matching Relational Edit : Data Grade: 9 Element: Update Type Edit Description: If the reported Update Type is P, C, or D, a previously reported R, A, or C Update Type must be found on the database for that claim. Matching fields used to identify previously reported loss records are: and Accident Date

34 Matching Relational Edit : Data Grade: 9 Element: Update Type Edit Description: If the reported Update Type is A or R without a corresponding D or P on the same unit, the claim must not already exist on the database. Matching fields used to identify previously reported loss records are: and Accident Date. 67 Matching Concept: If a loss is being updated, it must have the same claim number and accident date as the loss on our database. If a new loss is being added, it must not match a claim that already exists on our database. Maintaining the and Accident Date for the life of the loss is very important because these fields uniquely identify the claim

35 Matching Fields States All States Except MD and VA TX Policies Effective 1/1/15 and Subsequent TX Policies Effective 1/1/8 Through 12/31/14 MD and VA Only Policies Effective 1/1/8 and Subsequent MD, TX, and VA Policies Effective Prior to 1/1/8 Part 5 Unit Statistical Reporting Guidebook Fields Accident Date Accident Date Condition : Act Condition : Type of Accident Date Condition : Type of Accident Date Condition : Act Condition : Type of Condition : Type of Recovery Condition : Type of Vocational Rehabilitation Indicator Jurisdiction State 69 Matching Data Issue Report # and Seq. Accident Date Class Amount Data Grade # 1- R 8 4/1/6 $25 2- P 8 R 8 4/2/6 4/1/6 $25 $3 9 9 Unit Rejected Previous (P) record rejected for (matching) Revised (R) record rejected for (duplicate) 7 35

36 Report # and Seq. # Matching Correcting Accident Date Class Amount 1- R 8 4/1/6 $25 Data Grade 2- P 8 R 8 4/2/6 4/1/6 $25 $3 9 9 Based on research, it is determined that the date on the previous record of the 2- should have been 4/1/6. 71 Matching Correcting Report # and Seq. Accident Date Class Amount Data Grade # 1- R 8 4/1/6 $25 2- P 8 R 8 4/2/6 4/1/6 4/1/6 $25 $3 9 9 Correct Data and Resubmit Unit Now the previous (P) record for the loss on the 2- correctly matches the revised (R) record for the loss on the 1- and the unit loads to the database

37 New Unit Edits Edit Description Ensure that the correct Deductible Reimbursement Amount is being applied based on date. (Colorado only) 95 2 Ensures that the Incurred ant Attorney Fee Amount does not exceed the incurred amount of the claim. (Florida only) Identifies lump sum or closed claims that have an increase in Paid Indemnity that isn t identified as having a settlement Identifies suspect claims that have not been deleted at every report level. 73 Match Game Questions Classification Experience Modification Factor Must Be Consistent Open s Matching Fields Proper Reporting of Deleted s Answers Must Be Reported at Next Report Level New Unit Edits Should Never Change Same Experience Modification Effective Date and Accident Date 74 37

38 Unit Data Validation 75 Unit Data Validation Tests Quality checks after the data is loaded Verify the completeness, validity, accuracy, and reasonability of the data Identify patterns and data anomalies over a specified time Check for reasonability against countrywide or state norms 76 38

39 Unit Data Validation Tests Apply to data effective 2 and subsequent Performed prior to the state rate filings Responses are required within a predetermined time frame Results and responses via Notifications in DCA Access Online 77 Countrywide Tests Identify suspect patterns across states Countrywide Tests Impact ratemaking and experience rating Identify aggregate data issues Review of historical data before implementation of an edit 78 39

40 State-Based Tests Identify suspect patterns within a state State Tests Detect unusual year-to-year changes Have shorter validation periods Note: Similar issues may occur in numerous states. 79 Unit Validation Process NCCI Database Run Validation Tests Outbound Results via DCA Access Online Data Provider Research Detailed Explanation Take Appropriate Action Correction 8 4

41 Unit Validation Tests Matrix Details on All Validation Tests Assists with working notifications Unit Statistical Validation Matrices: All Validation Tests in production Future Validation Test Enhancements Part 12 Unit Statistical Reporting Guidebook 81 Unit Statistical Validation Test Matrix 82 41

42 Unit Statistical Validation Test Matrix 83 Unit Statistical Validation Test Matrix 84 42

43 Unit Validation Match Game Questions Identifies Data Anomalies Countrywide and State Validation Tests View and Respond to Validation Tests Identifies State Reporting Patterns Resource for Working Notifications Answers Notifications via DCA Access Online Validation Test Matrix Validation Tests The Two Types of Validation Tests State Based Tests 85 Validation Tests Validation Type C821 Validation Description Suspect Exposure/s Distribution E133 Duplicate Exposure for Uneven Split Periods 86 43

44 Validation Tests Validation Type L246 L51 L33 Validation Description s With No Recovery Deleted s on Subsequent Report Level Suspect Permanent Total s 87 Validation Tests Validation Type Validation Description L256 (New) Catastrophe Greater Than 1 L741 (New) L742 (New) Increased Paid Indemnity With Lump Sum and No Settlement Increased Paid Indemnity With Closed and No Settlement 88 44

45 Suspect Exposure/s Distribution Validation Type C821 Validation Type SUSPECT EXPOSURE/ CLAIMS DISTRIBUTION Validation Description CLAIM DISTRIBUTION SUSPECT COMPARED TO EXPOSURE Validation Summary Unusually low claim counts compared to high exposure/premium or unusually high claim counts compared to low exposure/premium in a given class. The inaccurate distribution of exposure and claims across states or within a state may result in overstated and/or understated loss costs/rates. 89 Unit Notifications via DCA Access Online 9 45

46 Unit Notifications via DCA Access Online 91 Unit Notifications via DCA Access Online 92 46

47 Unit Notifications via DCA Access Online 93 Suspect Exposure/s Distribution counts are high or low for policy period Payroll increases with decrease in claim count Payroll decreases with increase in claim count 94 47

48 Suspect Exposure/s Distribution Example State Policy Class Exposure Indemnity s Medical Only s 1 WC , WC ,975, WC , Suspect Exposure/s Distribution Verify Class (Exposure and ) Confirm Exposure Amount Submit Correction or Detailed Explanation 96 48

49 Validation Type E133 Duplicate Exposure Uneven Split Validation Type SUSPECT DUPLICATES RECON- CILIATION Validation Description DUPLICATE EXPOSURE IS REPORTED IN UNEVEN EXPERIENCE MODIFICATION AND/OR RATE EFFECTIVE PERIODS Validation Summary The attached grid displays exposure records on the same policy containing identical class code and payroll amount in unevenly divided experience modification and/or rate effective periods. Please verify the exposure amounts or submit corrections to reconcile the exposure with the duration of the split experience periods by the due date. 97 Duplicate Exposure Uneven Split Identifies exposure records on same policy with the same class code and payroll but uneven split periods 98 49

50 Duplicate Exposure Uneven Split Example State Policy Eff. Date Policy Class Exposure Rate Premium Mod Eff. Date Rate Eff. Date Split Period 2 1/1/1 WC , ,315 1/1/1 1/1/1 2 1/1/1 WC , ,315 9/1/1 9/1/ Duplicate Exposure Uneven Split Confirm Exposure Amounts Confirm Split Dates Are Correct Submit Correction or Detailed Explanation 1 5

51 s With No Recovery Validation Type L246 Validation Type CLAIMS WITH NO RECOVERY Validation Description DECREASING PAID LOSSES REPORTED ON CLAIMS WITH NO RECOVERY Validation Summary The attached grid displays claims with a total paid amount less than the total paid amount at the prior report level in absence of a recovery. Please verify the recovery type code and paid amounts at all applicable report levels by the due date. 11 s With No Recovery s total paid amount is less than the total paid amount at the prior report level No Recovery 12 51

52 s With No Recovery Example Rpt. # Policy Eff. Date Policy Inj. Type Paid Indemnity Paid Med 1 1/1/1 WC1111 WC ,553 25, /1/1 WC1111 WC ,648 28, /1/1 WC1111 WC ,159 3, /1/1 WC1111 WC ,526 3,521 1 Type of Recovery 13 s With No Recovery Verify Paid Amounts Confirm if a Recovery Was Received Submit Correction or Detailed Explanation 14 52

53 Match Game Questions Researching Distribution Test Paid Decreases With No Recovery Alternative to Working Grid in DCA Access Online Confirm Split Dates Always Do This Answers Downloadable Spreadsheet Submit Corrections or Provide Explanations Verify Both Exposure and Values Verify Paid Amounts and if There Was a Recovery Uneven Splits Identified With Duplicate Exposure 15 Deleted s on Subsequent Reports Validation Type L51 Validation Type DELETED CLAIMS ON SUBSEQUENT REPORTS Validation Description CLAIMS HAVE BEEN DELETED IN ERROR ON SUBSEQUENT UNIT REPORTS LEVELS Validation Summary The attached grid displays claims that have been deleted or reduced to zero at one or more subsequent report levels. If claims were deleted or reduced to zero in error, please submit corrections to add the necessary claim data at all applicable report levels by the due date. If the removal of a claim is justified at a subsequent report level, submit corrections to remove the claim at prior report levels and/or add the appropriate loss condition coding

54 Deleted s on Subsequent Reports was deleted or had zero filled Amounts on a subsequent report level 17 Deleted s on Subsequent Reports Example Upt. Type Policy Eff. Date Policy Rpt. # Class Inj. Type Inc. Indem. Inc. Med R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1,258 R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1,

55 Deleted s on Subsequent Reports Example Upt. Type Policy Eff. Date Policy Rpt. # Class Inj. Type Inc. Indem. Inc. Med R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1,258 R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1, Deleted s on Subsequent Reports Example Upt. Type Policy Eff. Date Policy Rpt. # Class Inj. Type Inc. Indem. Inc. Med R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1,258 R 1/1/1 WC WC ,45 1,258 P 1/1/1 WC WC ,45 1,

56 Deleted s on Subsequent Reports Review claim to determine if reported in error on subsequent report Determine appropriate corrective action Submit Correction Report 111 Suspect Permanent Total s Validation Type L33 Validation Type SUSPECT PERMANENT TOTAL CLAIMS Validation Description CLAIMS REPORTED WITH A PERMANENT TOTAL INJURY TYPE BUT HAVE SUSPECT INCURRED INDEMNITY AND/OR MEDICAL AMOUNTS Validation Summary The attached grid identifies claims with a Permanent Total injury type that appear inconsistent with the incurred indemnity and/or medical amount reported. Please verify the claim information or submit corrections to reflect the appropriate injury type and incurred amounts at all report levels by the due date

57 Suspect Permanent Total s Permanent Total injury type inconsistent with incurred indemnity and/or medical amount Injury usually likely to develop and require more medical expenses and lost time 113 Suspect Permanent Total s Example Policy Eff. Date Policy Rpt. # Inj. Type Inc. Indem Inc. Med Part Nature Cause 1/1/1 WC WC ,2 61, /1/1 WC WC ,356 62, /1/1 WC WC ,398 63,

58 Suspect Permanent Total s Verify the Injury Type Confirm the Indemnity and Medical Amounts Submit Correction or Detailed Explanation 115 Catastrophe Greater Than 1 Validation Type Validation Type Validation Description Validation Summary L256 CATASTROPHE NUMBER GREATER THAN 1 CLAIMS ARE REPORTED WITH A CATASTROPHE NUMBER GREATER THAN 1 AND ARE NOT PART OF AN EXTRAORDINARY LOSS EVENT The attached grid displays claims reported with a catastrophe number greater than 1, which do not appear to be extraordinary loss events. Please submit corrections to reflect the appropriate catastrophe coding at all report levels by the due date

59 Catastrophe Greater Than 1 Catastrophe greater than 1 but not an Extraordinary Event 117 Catastrophe Greater Than 1 Example Policy Eff. Date Policy Rpt. # Acc. Date Inc. Indem Inc. Med Cat. # 1/1/12 WC WC1234 1/21/12 58,2 61, /1/12 WC WC4321 1/21/12 65,356 62,

60 Catastrophe Greater Than 1 Verify ELE Catastrophe s ELE s and Descriptions on ncci.com Submit Correction 119 Increased Paid Indemnity With Lump Sum and No Settlement Validation Type Validation Type Validation Description Validation Summary L741 LUMP SUM FLAG Y AND NO LOSS CONDITION SETTLEMENT CODE () WITH INCREASED PAID INDEMNITY AMOUNTS CLAIMS REPORTED WITH A LOSS CONDITION SETTLEMENT CODE OF, A LUMP SUM FLAG OF Y, AND PAID INDEMNITY AMOUNTS THAT HAVE INCREASED SIGNIFICANTLY s reported with a Lump Sum Flag of Y and a large increase in paid indemnity amount on a subsequent report or is reported for the first time with a large paid indemnity amount. These claims are reported with a Condition Settlement Type of. 12 6

61 Increased Paid Indemnity With Lump Sum and No Settlement Lump Sum Indicator = Yes Condition Type of Settlement = Significant Increase in Paid Indemnity 121 Increased Paid Indemnity With Lump Sum and No Settlement Example Policy Eff. Rpt. Paid Lump Cond. Date Policy # Indem. Sum Settlement 1/1/1 WC WC ,2 N 1/1/1 WC WC ,356 Y

62 Increased Paid Indemnity With Lump Sum and No Settlement Verify Paid Indemnity Amount Check Lump Sum Indicator Submit Correction or Detailed Explanation 123 Increased Paid Indemnity With Closed and No Settlement Validation Type Validation Type Validation Description Validation Summary L742 CLOSED CLAIMS INDICATE NO LOSS CONDITION SETTLEMENT CODE () WITH AN INCREASE OR HIGH PAID INDEMNITY AMOUNT CLAIMS REPORTED WITH A LOSS CONDITION SETTLEMENT CODE OF (), CLOSED CLAIM STATUS, AND PAID INDEMNITY AMOUNTS THAT HAVE INCREASED SIGNIFICANTLY The attached grid displays closed claims reported with a large increase in paid indemnity amount on a subsequent report or is reported for the first time with a large paid indemnity. These claims are reported with a Condition Settlement Type of

63 Increased Paid Indemnity With Closed and No Settlement Status = Closed Condition Type of Settlement = Significant Increase in Paid Indemnity 125 Increased Paid Indemnity With Closed and No Settlement Example Policy Eff. Rpt. Paid Cond. Date Policy # Indem. Status Settlement 1/1/1 WC WC ,2 1/1/1 WC WC ,

64 Increased Paid Indemnity With Closed and No Settlement Verify Paid Indemnity Amount Check Status Submit Correction or Detailed Explanation 127 Editing Process Recap Data provider submits data to NCCI Updated/ corrected data is submitted NCCI edits applied Submission Results report issued

65 Validation Process Recap Data is subject to NCCI validation tests NCCI validates response closes notification or requests more information Verification or correction requested via DCA Access Online Unit Notifications Data provider submits corrections or provides detailed explanation 129 Unit Notifications es Without Exposure Identifies Suspect Patterns Across States Clears Errors on NCCI s Database Data Grade 9 Nonextraodinary Event USRG Hal s Beverage Choice 13 65

66 Thank You!

67 Supplemental Information

68

69 Presenter Biographies Bruce Hallman, AIDM, has worked in NCCI s Data Services Department for 18 years. He is a product support manager, subject matter expert, and the data liaison between NCCI and the independent bureaus. Bruce participates in many projects, including enhancements to Unit and Policy edits. He is actively involved in various industry groups, including the Workers Compensation Insurance Organization, Policy Advisory Task Group, and Advisory Statistical Work Group Committees. Bruce earned the Associate Insurance Data Manager designation from the Insurance Data Management Association (IDMA). Before joining NCCI, he was employed by the Pennsylvania Compensation Rating Bureau. Hal Dooley, AIDM, works in NCCI s Data Quality Department as a lead analyst and unit data subject matter expert. He is involved in all aspects of unit data editing and leads unit data projects. During his 23 years at NCCI, Hal has held positions in Customer Service, Experience Rating, and Data Resources. Hal earned the Associate Insurance Data Manager designation from the Insurance Data Management Association (IDMA). He holds a bachelor of science degree in business management from Nichols College in Massachusetts.