STATEMENT OF
BELINDA J. FINN
ASSISTANT INSPECTOR GENERAL FOR AUDITING
OFFICE OF INSPECTOR GENERAL
DEPARTMENT OF VETERANS AFFAIRS BEFORE
THE SUBCOMMITTEE ON DISABILITY ASSISTANCE AND MEMORIAL AFFAIRS
AND THE SUBCOMMITTEE ON OVERSIGHT AND INVESTIGATIONS
COMMITTEE ON VETERANS' AFFAIRS
UNITED STATES HOUSE OF REPRESENTATIVES
HEARING ON DOCUMENT TAMPERING AND MISHANDLING AT THE
U.S. DEPARTMENT OF VETERANS AFFAIRS

March 3, 2009

INTRODUCTION

Mr. Chairmen and Members of the Subcommittees, thank you for the opportunity to address important issues associated with the integrity and timeliness of the Veterans Benefits Administration’s (VBA’s) handling and processing of compensation claims.  Accompanying me is Mr. James J. O’Neill, Assistant Inspector General for Investigations.  We will discuss the vulnerability of veterans’ claim records and documentation to improper destruction and the accuracy of VBA regional office (VARO) compensation and pension (C&P) benefit claim dates.  In both of these areas, VBA lacked adequate management controls.  The absence of controls to protect claim documentation resulted in unnecessary delays in providing some veterans’ benefits.  Our review of control deficiencies associated with the accuracy of claim dates did not identify any veterans or their beneficiaries who received incorrect or delayed benefit payments.  However, the deficiencies could cause VAROs to report inaccurate claim-processing times to stakeholders such as veterans and Congress.

Improper Shredding of Claim-Related Documents

During our ongoing audit of VBA claim-related mail processing, the Office of Inspector General (OIG) discovered 132 claim-related documents needed to support and facilitate the processing of claims that personnel inappropriately discarded in 37 shred bins located at VAROs in Detroit, MI; Waco, TX; St. Louis, MO; and St. Petersburg, FL.  After we briefed Veterans Affairs (VA) and VBA senior officials regarding this discovery, VBA searched shred bins in all of its regional offices and the St. Louis, MO Records Management Center (RMC) and found 474 additional claim-related documents in 41 of the 58 locations. 

Document shredding services are generally part of routine records management functions performed in VAROs.  Shred bins were located in different work areas throughout the four VAROs visited by our auditors, and allowed employees to deposit documents no longer considered necessary or redundant but containing confidential information that prevented the employees from depositing them in open trash collection bins.  Three of the four VAROs (Detroit, Waco, and St. Petersburg) used locked shred bins, while one (St. Louis) used unlocked shred bins.  Contractors performed shredding either weekly or bi-weekly onsite at three of the VAROs (Detroit, Waco, and St. Petersburg) and offsite at the remaining VARO (St. Louis).

Claim-Related Documents Found in VARO Shred Bins.  This issue came to our attention at VARO Detroit in September 2008 when an employee told our auditors that claim-related documents might have been inappropriately discarded in shred bins.  On hearing this, the OIG audit team reviewed the entire contents of 18 shred bins at that VARO.  We identified 80 documents that were inappropriately discarded—5 documents that could affect claimants’ benefits and 75 documents that would not have affected benefits but should have been retained in claim files. 

After finding the claim-related documents at VARO Detroit, we expanded our review of shred bins and found an additional 49 documents at 3 other VAROs.  These included 31 documents that could affect claimants’ benefits and 18 documents that would not have affected benefits but should have been in claim files. 

The OIG reported preliminary findings to Congress in separate white papers issued in October and December 2008.  At that time, we reported identifying 129 claim-related documents inappropriately discarded in bins for shredding, of which 36 documents could have affected the benefits of 35 claimants.  While working with VARO Detroit officials since that time, we identified an additional nine documents that could have affected claimant benefits.  At 4 VAROs, we recovered a total of 132 documents from examined shred bins, of which 45 documents could have affected claim benefits and 87 other documents, such as death certificates, correspondence from the veterans, and award documents that would not have affected claims but should have been retained in claims files.  (See Table 1.)

Table 1. Claim-Related Documents Found in Shred Bins

The 45 documents fell into the following categories:

·        Original claims for benefits (6).  We identified six claims, including two from Global War on Terrorism veterans, which were required to have received priority processing.  We found no record of receipt and VARO managers confirmed that the claims were never established.

·        Reopened claims for benefits (13).  These claims represented requests for increased compensation, new entitlement to pension benefits, or entitlement to service connection for additional disabilities.  For 10 of the 13 claims, the VARO had no record of receipt or action to establish a claim.  Although the VARO had established claims in the remaining three cases, the private medical evidence submitted by the claimant was discarded in two of them, and no copies were in the official records.  In the third case, the VARO had established a claim with a later date based on a telephone inquiry.

·        Claims for burial benefits (4).  The VARO had no record documenting receipt or processing of these claims for reimbursement of burial expenses.

·        Dependency claims (3).  These claims included two requests for benefits on behalf of claimants’ children and one application to add a dependent to an existing claim.  The VARO had no record of receipt or action on the claim.

·        Claim for death benefits (1).  The VARO had no record of receipt or action upon this claim for death benefits by a surviving spouse.

·        Informal claims (4).  An informal claim preserves the claimant’s date of eligibility for benefits for the period of 1 year, indicating a claimant’s intent to apply for one or more benefits.  The discarded informal claims identified the date the VARO received the document, which is necessary to ensure the correct date of eligibility.  However, the VAROs had no record of receipt and control of these documents.  For example, a veteran did not receive his full entitlement of disability benefits because a VARO had not taken any action on an informal claim received prior to discarding the claim in a VARO shred bin.  The VARO received the veteran’s informal claim on January 31, 2008.  In February 2008, the VARO received the veteran’s formal claim and established the claim in the SHARE system, an automated computer system used by VBA to establish and manage pending issue claim data, with a February 12, 2008, date of claim.  The VARO awarded benefits to the veteran based on the February 12, 2008, date of claim with a payment date of March 1, 2008.  The VARO was required to have awarded benefits based on a January 31, 2008, date of claim with a payment date of February 1, 2008.  As a result, the veteran did not receive the entitled disability benefit payment of $117 for the month of February 2008.

·        Various documents (14).  These documents included write-outs and returned mail that the VAROs had not processed and could affect claimants’ benefits.  A write-out is notification to VA that a discrepancy exists between a veteran’s computer record and the amount of monthly benefits the veteran is receiving.  VARO employee action is required to reconcile the discrepancy.

When we completed our reviews, we worked with VARO officials and confirmed that the 45 claim-related documents found in the bins could have affected benefits and copies of the documents were not maintained in claim files.  We returned the identified documents to the VAROs so that appropriate actions could be taken on these claims and the documents could be placed in official records.

OIG Criminal Investigation

Special Agents assigned to VA OIG conducted investigations of improperly shredded claims-related documents at the following seven VAROs:  New York, NY; Cleveland, OH; St. Louis, MO; Pittsburgh, PA; Columbia, SC; Detroit, MI; and St. Petersburg, FL.  To avoid disclosing information protected by the Privacy Act, we will not associate our findings with specific VAROs because the employees subjected to administrative actions as a result of their involvement in shredding would be identifiable to their co-workers. 

At one VARO, an employee was observed on two occasions by two different employees placing claim-related documents in a shred bin.  These documents were recovered and determined to be inappropriate for shredding.  When interviewed, this employee did not dispute the accuracy of the witnesses’ observations but claimed that prescribed medication rendered him extremely disoriented.  He said he was unaware of placing claim-related documents in shred bins on either occasion.  After confirming that the employee is prescribed medication, we presented the facts of this case to an Assistant U.S. Attorney (AUSA), who declined prosecution for violations of 18 USC 2071 (Concealment, Removal, or Mutilation Generally) because of the difficulty in establishing criminal intent.  VA subsequently removed this employee from Government service.

At a second VARO, a VBA review of all documents thought to be staged for shredding revealed that approximately 90 percent of claim-related documents were discovered in a shred box under the desk of one employee.  When interviewed, the employee denied intentionally placing the above documents in his shred box.  We terminated the interview when the employee exercised his right to legal representation.  An AUSA declined prosecution for violations of 18 USC 2071 in lieu of administrative remedies.  VA subsequently removed this employee from Government service.

At a third VARO, management learned in October 2008 that an employee had been discovered inappropriately shredding documents approximately 4 months earlier.  At that time, he told prior management that a supervisor had instructed him to shred these documents.  This employee and the supervisor subsequently received verbal counseling.  VBA’s investigation in October identified another employee who said this supervisor, prior to the incident described above, also had instructed him to shred documents that he thought should not be shredded.  We re-interviewed all of the parties involved in this issue.  The supervisor, who was on administrative leave as a result of involvement in another incident, denied instructing either of the above employees to shred documents that should not have been shredded.  An AUSA declined prosecution for violations of 18 USC 2071 in lieu of administrative remedies.  VA subsequently removed this employee from government service.

At a fourth VARO, in response to discoveries made by our audit staff during an unannounced inspection, we opened a criminal investigation into inappropriate shredding of claim-related documents which revealed that the majority of documents discovered by the auditors had been assigned to two employees for processing.  Both employees denied placing these documents in a shred bin and offered to take polygraphs.  Since there was handwritten material on one of the documents, we obtained exemplars from both employees.  Our document analyst concluded that one of these employees had probably written the questioned material.  The paucity of this material was the only reason the analyst had to qualify her opinion; there were no differences identified during the forensic examination between the known writing of the employee and the questioned material.  When confronted with this forensic evidence, the employee continued to deny being the author of the questioned material and reasserted non-involvement in inappropriate shredding.  The second employee also claimed no knowledge of how the documents found their way into the shred bin, but claimed that the other employee processed most of the paperwork in their unit.  After we arranged for another law enforcement agency to administer the polygraphs, both employees declined to be polygraphed.  An AUSA declined prosecution for violations of 18 USC 2071 and 18 USC 1001 (False Statements) in lieu of administrative remedies.  On
February 17, 2009, we provided VBA a report of our investigation so a decision can be made if administrative action will be taken against one or both of these employees.

The investigations conducted at the other three VAROs did not identify anyone who willfully and knowingly shredded claim-related documents inappropriately.

Response by VA and VBA.  On October 14, 2008, we briefed the former Secretary of Veterans Affairs, the Under Secretary for Benefits, and other senior VA and VBA officials concerning the documents found in shred bins.  Immediately following this briefing, the Under Secretary for Benefits directed every VARO to suspend all document shredding.  In addition, the Under Secretary for Benefits instructed every VARO Director to review and inventory all contents in shred bins, report all claim-related mail or original supporting documents found in shred bins, and certify that documentation contained in their shred bins was not inappropriately destroyed.

VBA reported to the OIG that their inventories of VARO shred bins nationwide located 474 documents affecting benefit entitlements and 8 documents that were required to have been returned to claimants at 41 locations nationwide.  Of the 57 VAROs, 40 (71 percent) reported that shred bins included documents affecting benefit entitlements.  VBA’s RMC in St. Louis also reported that six documents were inappropriately discarded.  Of the 474 documents, 242 (52 percent) were at 3 VAROs—Columbia, SC (95 documents); St. Louis, MO (94 documents); and Cleveland, OH (53 documents).  The remaining 232 documents were at 37 other VAROs and the RMC, with the number of documents in shred bins ranging from 1 to 95.

In November 2008, VBA issued a new and sweeping policy regarding procedures for the maintenance, review, and appropriate destruction of veterans’ paper records located in all VBA facilities and worksites.  We have not yet reviewed the implementation or effectiveness of this policy, but we plan to review compliance with the policy during future audits and reviews.

Extent of Inappropriate Claim-Related Shredding Cannot Be Determined.  We cannot determine how long this problem may have been occurring at VAROs or how many documents were potentially shredded.  The bins our auditors reviewed contained 14 or fewer days of material.  Further, we cannot project our findings or know whether the findings represent the typical contents for each shred bin.

Lack of Controls Is a Contributing Factor.  At the time of our audit, VBA had no requirement for any supervisor or other official to review documents placed in the shred bins.  Therefore, an employee could easily dispose of documents, either unintentionally or purposely.

VBA officials also said that some VAROs held “mail amnesty” periods to encourage employees to turn in unprocessed mail and other documents without penalty or repercussions.  During an amnesty period in July 2007 at VARO Detroit, VARO employees turned in almost 16,000 pieces of unprocessed mail including 700 claims and 2,700 medical records and/or pieces of medical information.  The VARO determined that none of these claims or documents were in VBA information systems or associated claim files.  VBA management told us of similar amnesties at other VAROs, such as an amnesty at VARO New York in December 2008 that recovered 717 documents from VARO employees.

We consider any loss of claim-related documents to be unacceptable.  These actions result in an inaccurate or incomplete record of activities needed to process a veteran’s claim, delay claims processing, and potentially result in denial, under, or overpayment of benefits.  Further, this situation increases the distrust that some veterans and beneficiaries already have in VA’s ability to adequately protect documents and provide timely benefits. 

We believe that, taken together, this information indicates that claim-related mail processing and the protection of records submitted by veterans in support of claims are a high-risk area for VBA.  Had we not discovered this situation, some veterans claims may have languished with no action or been inappropriately denied.  VBA took immediate actions to address the shredding problem.  We will monitor these controls at the VAROs to ensure that veterans’ claims and records are promptly processed and appropriately protected.

Inaccurate Compensation and Pension Benefit Claim Receipt Dates

In October 2008, at the request of the Ranking Member of the HVAC, Steve Buyer, we initiated a review to evaluate the accuracy of VARO C&P benefit claim receipt dates.  VARO staffs are required to document claim receipt dates in claim folders and an automated computer system named SHARE.  When VAROs use inaccurate claim receipt dates as the effective dates of awards, payments to veterans or their beneficiaries can potentially be delayed.  Also, since VBA uses claim receipt and completion dates to measure claim processing timeliness, inaccurate receipt dates can cause reported claim-processing times to be incorrect.

Congressman Buyer requested our review after VBA reported the results of its August 2008 Administrative Investigation Board (AIB) investigation of claim receipt dates at VARO New York.  During the AIB, staff obtained sworn testimony from 34 employees and C&P Service staff reviewed receipt dates for 390 claims.  The investigation concluded that VARO management instructed staff to intentionally establish erroneous claim receipt dates and staff did so for 220 (56 percent) of 390 claims reviewed and had been establishing erroneous dates for a number of years.  However, VBA said that the errors did not cause any veterans or their beneficiaries to receive incorrect or delayed benefit payments.  Subsequently, VBA has held several VARO New York managers accountable for intentionally establishing erroneous claim receipt dates.

We evaluated a statistical sample of 1,515 total claims at VAROs Albuquerque, Boston, San Diego, and Winston-Salem to determine if inaccurate claim receipt dates caused veterans or their beneficiaries to receive incorrect benefit payments and if inaccurate claim receipt dates caused VBA to report incorrect claim-processing times to stakeholders.  We found that the claim receipt dates for the majority of the 94,920 claims completed at the 4 VAROs during fiscal year (FY) 2008 were accurate.  We projected that 88,639 (93.4 percent) claim receipt dates were accurate, and 4,520 (4.7 percent) were inaccurate.  In addition, we also projected that the 4 VAROs had not documented a receipt date for the remaining 1,761 (1.9 percent) claims and could not support the dates recorded in SHARE with receipt dates in claims folders.

While we found some inaccurate claim receipt dates at all four VAROs, none of the VAROs’ inaccuracy rates approached the 56 percent rate VBA reported for VARO New York.  VARO Boston had the highest inaccuracy rate of 10 percent.  Inaccuracy rates were 5 percent for VARO Albuquerque, 4 percent for VARO Winston-Salem, and 3 percent for VARO San Diego.  The calculations of inaccurate dates do not include the 1,761 undocumented dates.  Therefore, the actual rates of inaccurate claim dates could be higher.

The errors we reviewed did not cause any veterans or their beneficiaries to receive incorrect or delayed benefit payments.  The payments were correct because VARO personnel used the correct claim receipt dates, which staff documented in claims folders, when establishing the effective dates of benefit awards in the C&P Payment Master Records system instead of relying on dates in SHARE.  Interviews of VARO staff and reviews of claims folders indicated that the inaccurate dates were mostly unintentional errors.  Because the inaccurate receipt dates were both before and after the correct dates, the inaccuracies did not significantly affect most of the four VAROs’ reported FY 2008 average claim-processing times.  The only exception was VARO Boston, where the errors caused the understatement of projected average processing times by 4 days (176 days using recorded dates and 180 days using actual dates).

Our results indicated that the claim date inaccuracies were mostly unintentional errors.  Only 1 of the 1,515 claims folders included evidence of an intentional inaccurate claim date and only one VARO employee stated that an inaccurate date was intentional.  We concluded that the timing of the inaccurate claim dates also indicated that the inaccuracies were unintentional because the inaccurate dates were occurring both before and after the correct dates.  Inaccurate receipt dates that are after actual receipt dates will result in lower claim-processing times and indicate better performance.  Inaccurate receipt dates that are before actual receipt dates will result in higher claim-processing times and indicate poorer performance.  Therefore, if VAROs intentionally manipulated claim-processing times to indicate better performance, they would have consistently recorded SHARE receipt dates that were after actual receipt dates.

Our review indicated that the majority of the inaccurate claim receipt dates were unintentional errors.  For 109 (99.1 percent) of the 110 inaccurate claim receipt dates, the timing of the dates, our claim folder reviews, and interviews of staff indicated the VARO personnel probably entered the inaccurate dates in SHARE unintentionally.  The following case is an example of how the evidence indicated that the error was most likely unintentional.

·        VARO Winston-Salem received a veteran’s statement in support of a claim for an increased disability rating because of a lower back condition.  The claim receipt date documented in the claim folder was September 28, 2007, which was the receipt date the VARO stamped on the document.  However, the September 8, 2007, receipt date reported in SHARE was 20 days before the date documented in the claim folder.  Our review of the claim folder and interviews of VARO staff found no evidence that the VARO intentionally reported the inaccurate date in SHARE.  Therefore, we concluded that the inaccurate date was most likely the result of the Veterans Service Representative (VSR) erroneously entering an “8” instead of a “28” when entering the date in SHARE.  The incorrect claim receipt date did not affect the veteran’s benefits.

VARO Boston Employee Intentionally Recorded Inaccurate Dates.  We found intentional recording of inaccurate dates at one location.  During our visit to the Boston VARO, a Senior Veterans Service Representative (SVSR) told us that he entered an inaccurate receipt date in SHARE.  In this case, VARO Boston received a letter from a veteran claiming service-connection disability benefits for sleep deprivation.  The VARO Triage Team correctly stamped September 24, 2007, as the receipt date on the back of the veteran’s letter.  On the last page of the letter, we found an annotation with no initials that established the receipt date as
January 22, 2008.  The date in SHARE was the incorrect date of January 22, 2008, which was 124 days after the actual receipt date of September 24, 2007.  The incorrect claim receipt date did not affect the veteran’s benefits, but it did contribute to the VARO Boston 4-day understatement of average claim processing time previously discussed.

The SVSR told us that he entered inaccurate receipt dates in SHARE because he had a “general impression” of responsibility to help the Veterans Service Center (VSC) achieve the “goal to make numbers meet.”  He stated that he entered receipt dates in SHARE that were within 30 days of the input date instead of the actual receipt date for claims over 1 year old.  For claims that were between 7 and 30 days old, he entered dates in SHARE that were within 7 days of the input date.  The employee also said he knew of no other supervisors entering inaccurate dates, he was unaware of any other supervisor that instructed staff to do this, and VARO management did not direct him to instruct staff to enter inaccurate claim receipt dates in SHARE.  The SVSR retired the end of January 2009.  The VARO Director is conducting a comprehensive investigation to assess the extent of this SVSR’s actions. 

VARO Documentation of Claim Receipt Dates Needs Improvement.  We found that VAROs needed to improve the documentation of claim receipt dates in claims folders.  We projected that for 1,761 (1.9 percent) of the 94,920 claims the 4 VAROs completed during FY 2008, claim folders did not include sufficient documentation to determine if SHARE claim receipt dates were accurate.  VARO managers could not explain why staff had not documented SHARE receipt dates in the sampled claims folders.  Because we could not confirm the accuracy of the receipt dates for these claims, we could not determine if veterans or their beneficiaries received correct benefit payments or if there was any effect on the accuracy of reported claim-processing times.

VBA Needs To Take Action.  To help ensure the accuracy of reported claim-processing times, which VBA managers and stakeholders use to measure and monitor VARO performance, VBA needs to better monitor claim receipt dates and documentation.  In November 2008, VBA’s Quality Assurance Systematic Technical Accuracy Review (STAR) began evaluating the accuracy of receipt dates during their regular monthly VARO reviews.

However, we recommended that VBA further improve the reporting reliability of claim-processing times by implementing two actions.  The first action is to establish claim receipt date accuracy goals, which will improve transparency and accountability for VARO performance.  VBA has used goals to help improve VARO performance in other areas such as accuracy rates for pension authorizations and burial claims processed.  The second action is to require VAROs to perform Systematic Analysis of Operations’ (SAOs) of claim receipt date accuracy and documentation.  VAROs routinely use SAOs as a self-audit technique to improve various aspects of operations.  Performing specific SAOs of claim receipt dates and documentation will help ensure VARO and VBA managers have accurate and reliable claim-processing time data for their decision-making purposes.  We also recommended the VARO Boston Director identify any other claims where the SVSR intentionally entered inaccurate receipt dates in SHARE and ensure benefit payments related to these claims are correct.

VBA Response.  VBA reported that claim date accuracy reviews will be a permanent addition to the STAR program to ensure continued monitoring.  VBA will collect and study accuracy data from these reviews and establish a sound goal by June 1, 2009.  By April 1, 2009, C&P Service will revise its policies to add the requirement for validating the accuracy of claim dates to the existing “Quality of Control Actions” SAO.  In addition, VARO Boston will examine claims established by the SVSR who intentionally entered the incorrect date of receipt identified by OIG and correct dates to ensure benefit payments are accurate.  The VARO Director is conducting a comprehensive investigation to assess the extent of this SVSR’s actions. 

Ongoing OIG Oversight Work.  In addition to the two reviews already discussed, the OIG has been aggressively increasing its presence in, and oversight of, VBA’s regional offices.  In addition to our audit of VBA claim-related mail and mailroom operations and our review of claim-date accuracy, we are also auditing two of VBA’s quality assurance programs.  These include VBA’s “Site Visit” program that reviews C&P functions, and the Systematic Technical Accuracy Review program, which measures accuracy of claim processing decisions made in all regional offices.  We also performed a special review of management controls to prevent fraudulent payments for retroactive benefits of $25,000 and above.  This week we are also initiating an audit to evaluate the effectiveness of VBA’s Control of Veterans Record System, which tracks the location of claims folders within VBA offices.   We will provide reports on these reviews upon completion.  In May 2008, the OIG issued a report on the impact of VBA’s hiring initiative on reducing the claims backlog and we are planning to begin another review to examine the effectiveness of VBA’s efforts integrating new staff into their workforce.  OIG teams conducted evaluations onsite at 16 VAROs during 2008 (we visited 3 VAROs twice on 2 different reviews).

In addition, we are staffing the Benefits Inspection Division to provide continuous oversight of VARO operations nationwide.  The inspections will evaluate how well VAROs are accomplishing their mission of providing accurate and timely benefits and services to veterans and their dependents.  The goal of the inspection program is to complete at least 12 inspections each fiscal year, allowing coverage of all 57 VA Regional Offices within a 5-year period. 

Mr. Chairmen, that concludes our remarks and we thank you for the opportunity to discuss these important issues and your continued support.  We would be pleased to answer any questions that you or other members of the Subcommittee may have.