Mr. Chairman and Members of the Subcommittee:
Thank you for this opportunity to present The American Legion’s views on disability claims ratings and benefits disparities within the Department of Veterans Affairs (VA) Veterans Benefits Administration (VBA). The American Legion commends the Subcommittee for holding a hearing to discuss this important and timely issue.
May 2005 VA Office of the Inspector General Report
In response to a December 2004 Chicago Sun-Times article revealing disparities in VA disability compensation payments on a state-by-state basis, the Secretary of VA ordered the VA Office of the Inspector General (VAOIG) to investigate the matter. On May 19, 2005, the VAOIG issued a report addressing the reasons for differences in average monthly VA disability compensation made to veterans living in different states.
The VAOIG noted that for fiscal year (FY) 2004, average annual payments by state ranged from $6,961 to $12,000, a difference of over $5,000. According to the VAOIG the highest paying states were: New Mexico (the highest), Maine, Arkansas, West Virginia, Oklahoma, and Oregon. The lowest paying states were: Indiana, Michigan, Connecticut, Ohio, New Jersey, and Illinois. The VAOIG concluded that no single variable factor was responsible for the discrepancies in compensation payments.
The VAOIG found that there were sixteen possible factors that could cause compensation payment disparities. In its analysis, the VAOIG concluded that there were ten factors that the VA could not control and there were six factors over which the VA could exert some control.
According to the VAOIG, the factors that the VA cannot control are: power of attorney representation, enlisted versus officer, military retirees versus non-military retirees, participation of veterans receiving benefits, period of service, branch of service, dependents, special monthly compensation, age, and the average number of disabilities. The six factors that the VAOIG indicated the VA has some control over are: pending claims, brokered claims, appeal rates, transferred cases, grant rates, and rater experience.
Finally, the VAOIG stated that some disabilities are inherently more susceptible to variations in rating determinations. The VAOIG indicated that the Rating Schedule (38 C.F.R. Part 4), because it is a 60-year-old model, may also cause some inconsistencies. The VAOIG identified post-traumatic stress disorder (PTSD) evaluations, total disability based on PTSD (including individual unemployability or IU), and all veterans rated with IU as rating decisions susceptible to variations.
The VAOIG focused on mental disabilities because of several reasons: mental disabilities have a high variable rate (compared to the other parts of the body systems evaluated by the Rating Schedule); mental disabilities have the highest average evaluation (58 percent); and PTSD, which is a mental disability, is one of the fastest growing service-connected disabilities.
The VAOIG reviewed 2,100 PTSD cases at seven regional offices (RO). They found that the ROs approach stressor verification requirements differently from state to state. In particular, there were differences in how the ROs verified veterans’ allegations about traumatic events in service. The VAOIG also found that, in general, once veterans with PTSD obtain a 100 percent evaluation their receipt of mental health treatment declined.
The VAOIG noted that there were several instances of benefits fraud in the past few years. It was stressed that based on an income match, 8,486 veterans in receipt of IU benefits reported earned income to the Internal Revenue Service (IRS). The VAOIG indicated that some or all of the 8,486 veterans in receipt of IU benefits and in receipt of earned income, may not be entitled to IU benefits.
The VAOIG also surveyed 1,992 rating specialists and Decision Review Officers (DROs) and 1,349 responded. The relevant results indicate:
- 65 percent stated they did not have enough time to provide timely and quality service;
- 57 percent indicated that they had difficulty meeting production standards if they took time to adequately develop claims and thoroughly reviewed the evidence before making a decision;
- 41 percent declared that 30 percent or more of the claims they decided were not ready to rate when presented for rating;
- 20 percent estimated that of the claims not ready to rate more than 10 percent were actually rated without all the needed information; and
- 52 percent responded that they could assign two or more different ratings for the same medical condition.
The May 2005 VAOIG report contained the following recommendations:
- Conduct a study to detect and correct unacceptable payment patterns.
- Work with the Veterans’ Disability Benefits Commission to clarify and revise the rating schedule.
- Conduct Review of rating practices for certain disabilities such as PTSD and IU.
- Expand national VA quality review to include review of PTSD evaluations for consistency, and to determine if the stressor was fully documented.
- Coordinate with the Veterans Health Administration to improve the quality of medical examinations.
- Ensure that VA regional offices are adequately staffed and equipped.
- Consider establishing a lump-sum payment option in lieu of recurring monthly payments for veterans with disability evaluations of 20 percent or less.
- Analyze differences in claim submission patterns to determine if certain veteran sub-populations, such as World War II veterans or veterans living in certain areas, have been underserved and perform outreach based on the results of the analysis.
For years, The American Legion and other veterans service organizations (VSOs) have stated that the driving force behind most VA adjudications is the need for VA to process as many claims as possible in the fastest possible time. This emphasis on quantity and speed of adjudication results in premature adjudications, improper denials of benefits, and of course, inconsistent decisions.
The VAOIG report confirms much of what we have been saying about the VA claims adjudication process. Essentially, the VAOIG acknowledges that because the VA often does not take the time to obtain all relevant evidence and information, there is a good chance that these claims are not properly adjudicated. The VAOIG, to its credit, quoted raters and DROs who indicated that VA management is much more concerned with quantity than quality. Some VA adjudicators stated that awards and bonuses are centered around production. The report, however, did not mention that in most claims where the VA does not obtain all relevant information, the claim is denied or under evaluated.
The overall tone of the VAOIG report was disappointing. It implied that where the VA fails to develop claims properly, there are only improper grants of benefits. The VAOIG ignored the fact that many deserving veterans have their claims denied or under evaluated because the VA, in a rush to claim work credit, failed to, or refused to, comply with the duties to assist and notify. Although the VAOIG conceded that VA often makes errors, it failed to consider or discuss whether these errors could result in the unlawful denial of benefits or the under evaluation of service-connected disabilities.
This negative tone exists throughout the VAOIG report. For example, when discussing the differences between adjudications in New Mexico and Illinois, the VAOIG noted that New Mexico had the highest average monthly VA disability compensation payments at $11,206. The VAOIG indicated that the high New Mexico payments “may be a cause for concern.” The VAOIG, however, did not express any concern about the low paying ROs. Apparently, the possibility that some veterans may be underpaid or unfairly denied did not alarm the VAOIG.
The VAOIG also attacked the current rating schedule as “a 1945 model that does not reflect modern concepts of disability” even though most of the major body systems have been updated in the last 20 years. Also, it did not define the term “modern concepts of disability” and did not explain why the current rating schedule would cause inconsistent payments.
According to the VAOIG, whether a veteran was represented by a VSO was the single most important factor in determining the amount of compensation payments made to that veteran. The VAOIG reported that on the average, veterans who are represented by a VSO, receive $6,225 more per year than those veterans without representatives. This is a telling statistic. VA operates a disability benefits program that is required to be non-adversarial and ex parte. (See 38 C.F.R. § 3.103(a).) The huge disparity between non-represented veterans and represented veterans supports the conclusion that VA’s claims adjudication system is more adversarial than VA cares to admit.
Additionally, the VAOIG report appears to assume that the states with high levels of compensation payments are doing something wrong. The VAOIG apparently did not consider that the states paying a high level of benefits are making correct legal decisions--doing a better job than the states with low levels of payments. The American Legion asserts that it is quite possible that some, if not all, ROs are incorrectly denying a considerable number of claims for compensation and under-evaluating some service-connected conditions. We believe there are more veterans being unfairly denied benefits and underpaid benefits than there are veterans who are being unfairly granted benefits and/or overpaid benefits.
This conclusion is based on the following fact. In the past few years The American Legion has jointly reviewed the quality of adjudications in approximately 40 ROs. Our quality review team has found errors in all of the VA offices reviewed, including the regional office in New Mexico. For example, the review of the VA regional office in New Mexico generated the following comments.
Some of the New Mexico rating decisions reviewed by The American Legion team exhibited lack of knowledge or carelessness. For example:
- In some instances the RO incorrectly denied service connection for a congenital disease because the RO misinterpreted 38 C.F.R. § 4.9.
- In some instances the Global Assessment of Functioning (GAF) score was ignored.
- The effective dates assigned for individual unemployability (IU) created problems. According to an RO official, the RO assigned an effective date from the receipt of the VAF 21-8940 – instead of the date of the informal claim for IU. The official stated this was a recurrent problem in this RO.
- Some VA examinations were inadequate.
- Some ratings concerning claims for increase should have, but did not, consider 38 C.F.R. § 3.400(o)(2).
- Some inferred issues were either missed or ignored.
- The rules concerning new and material evidence were not correctly applied.
- In some instances, special monthly pension (SMP) was not correctly considered or improperly rejected.
- In some cases, the RO issued confusing and misleading development and notice letters.
- In some instances the RO failed to clarify the appellate process to veterans who clearly were confused.
Many of the types of errors identified in New Mexico were similar to the errors that we found in low paying ROs like Chicago. If the New Mexico RO, the highest paying office according to the VAOIG, exhibited these underpayment and improper denial problems, it is possible that all VA ROs under-compensate some claimants to various degrees. The VAOIG never considered this possibility. In fact, all ROs reviewed by The American Legion’s quality review team exhibited patterns of improper denial and underpayment. Of course, some ROs exhibited much better quality than other ROs.
Also, in FY 2007 the Board of Veterans’ Appeals (BVA or Board) remanded or reversed 56 percent of the appeals it reviewed. It is very unlikely that any of those remands or reversals involved overpayments of benefits or the improper grant of service connection. The BVA reversal/remand rate reveals that ROs commit many errors adverse to veterans.
In spite of the inescapable fact that there is a serious quality problem within the ROs that unfairly deprives many deserving veterans of VA benefits, the VAOIG did not mention or even allude to this situation. This omission is a disservice to veterans and casts doubt on most of the VAOIG conclusions.
Institute for Defense Analyses (IDA) Report
In response to the VAOIG’s recommendation, VA contracted IDA to conduct a study in order to gain a better understanding of the potential causes of the variances in disability payments.
The IDA offered six recommendations for improving the consistency of VBA’s claims adjudication process:
- Standardize initial and on-going training for rating specialists.
- Standardize the hospital evaluation reporting process.
- Increase oversight and review of rating decisions.
- Consider consolidating all or selected parts of the rating process into one location.
- Develop and implement metrics to monitor consistency in adjudication results.
- Improve and expand data capture and retention.
The American Legion agrees with IDA’s recommendation to increase VBA oversight and review of RO rating decisions. We also note that this recommendation specifically stated that denied claims should also be reviewed, something the VAOIG did not consider in its investigation and subsequent report.
Regarding its training recommendation, IDA noted that although VBA provides centralized training modules for training purposes, many regional offices supplement this training with material developed locally. IDA also noted that many rating specialists interviewed stated that they received “on-the-job” training from senior raters and identified these individuals as the biggest influence on their rating styles. IDA suggested that a “stronger mechanism” would reduce the potential for persistent differences among regional offices in ratings and ensure that raters VA wide are receiving the same training. IDA further recommended that raters be given standardized test cases, reflecting the most likely areas of variation, as part of an ongoing training process.
The American Legion is appreciative of the importance the Under Secretary for Benefits has placed on training of VBA personnel. We are also aware of the centralized training program that has been implemented; however, a national training standard/requirement, in addition to the centralized training conducted by Compensation and Pension Service (C&P), for regional office personnel is also needed. Consistent and standardized training at each regional office must take place for all personnel—experienced and new hires alike. The American Legion believes it is crucial that such a program be implemented and closely monitored for compliance by the Under Secretary for Benefits. Management in stations not in compliance with such training requirements must be held accountable; otherwise any national or centralized training effort will not be successful.
Additionally, The American Legion also believes it is essential to proper training that information (reasons for remand or reversal) from BVA decisions, Court of Appeals for Veteran Claims decisions, DRO decisions and errors noted in the National Systematic Technical Accuracy Review (STAR) be tracked and examined for patterns. This information should then be analyzed by VBA and provided to ROs in mandatory formal training to ensure that common errors and other discrepancies occurring in regional office rating decisions are not repeated. This information should also be used for remedial training purposes when patterns of errors are identified for specific individuals. Although such data is currently being collected and disseminated to the ROs, it appears that consistent utilization of this data in regular formalized and specific training has been lacking. Unless ROs (both managers and individual adjudicators) learn from their mistakes and take corrective action, there will continue to be a high rate of improperly adjudicated claims, resulting in a consistently high appeals rate and subsequent high BVA remand/reversal rate of RO decisions.
In addition to our training-related concerns discussed above, we also have concerns regarding VA’s skill certification testing program to ensure competency and proficiency. C&P conducted an open book (pilot) job skill certification test for veterans service representatives (VSR) several years ago in which the pass rate was extremely low (approximately 23 percent). Even more alarming than the low-test scores was the fact that those who took the test had several years of experience in the position and were considered to be proficient.
C&P subsequently finalized its VSR proficiency test and conducted tests in May and August 2006. Employees participating in the testing underwent 20 hours of training prior to taking the test. Although the pass rate (about 42 percent) for these tests was much higher than the pilot test, it is still very low and can hardly be considered acceptable. C&P did not conduct any tests in FY 2007.
The American Legion applauds the new testing program as a step in the right direction, but we still have concerns. Although successful completion of the test will be required for promotion or assignment to a rating board, it is not mandatory as a condition of employment in that position and is completely optional. C&P is in the process of developing a test for rating veterans service representatives (RVSR) and DROs, but a timeline for completion or implementation has not yet been determined. Unfortunately, like the VSR test, the test for RVSRs and DROs will not be mandatory as a condition of employment.
The ultimate goal of proficiency or competency testing should be to ensure that an individual in any given position is competent, proficient, and otherwise qualified to perform the duties required of that position. This goal will not be achieved if testing is not mandatory, or is not provided for all levels or for all positions, and remedial training or other corrective action is not required for those who do not successfully pass the test. Although this concept may not be embraced by some, the ultimate goal is to have qualified and competent staff who will provide the best service possible for America’s veterans.
Lastly, The American Legion opposes IDA’s recommendation supporting rating consolidation. It is likely that some VA managers also like the idea of consolidation because of the economic advantage to the VA. It is cheaper to have 10 or 16 offices than to pay for 57 regional offices. However, in our experience, many of the bigger VA offices have more quality problems than the smaller ROs. The American Legion quality reviews reveal that the fact that raters and DROs are under the same roof does not mean they will all rate claims consistently. Also, consolidation, especially consolidation in low cost of living rural areas, would hamper access to the VA regional offices for many veterans, especially low income and minority veterans. Obviously, that is not a good thing.
In closing, The American Legion recommends increased oversight by VBA as well as more frequent transferring of RO service center managers in order to create a “national” culture to avoid regional differences and biases. We also recommend the establishment of an independent quality review program with accountability built in for managers and adjudicators. Additionally, until substantive changes are made in the work measurement system, a piecemeal “band-aid” approach will not make a major difference. The creation of a work measurement system that rewards prompt, but fair and complete adjudications would improve consistency and quality. Such changes would be the fastest, least expensive way to make the biggest positive impact on the VA’s claims adjudication system.
Mr. Chairman, that concludes my statement. The American Legion welcomes the opportunity to work closely with you and your colleagues on this and any other issue that concerns this nation’s veterans.