already been the subject of significant study and its recognition that, despite that study, the system is still plagued by backlog and delay and the frustration that accompany them. Making major changes will not be easy-there will be opposition from those feeling threatened by new ways of doing business, but I believe that, if all the stakeholders are permitted to participate actively in crafting these changes, they can happen.

I have three major recommendations, the first of which would affect both the administrative and judicial processes, the second of which would affect primarily the former, and the third of which would affect primarily the latter.

1. Despite the controversy and resources expended over the former requirement of a claimant having to present a well-grounded claim before being entitled to the VA-provided duty to assist, and despite the controversy and resources already expended in the purported fixing of the problem, it appears that little has changed regarding what is, in most cases, the major need for such a duty: That is, to provide a thorough medical opinion as to the causative relationship between present disability and military service. Indeed, it appears that 38 U.S.C. §5103A(d)(2)(B), in essence, reimposes a well-grounding requirement to obtain such an opinion. To avoid much future litigation and future cycles of both BVA and Court ordered remands, this provision should be clarified.

Where causation is at issue, I believe that obtaining a medical opinion on this issue at the earliest possible time in the claims process would likewise result in much earlier finalization. One approach could be that once there is evidence of both present disability and a possible event in service, a claimant would be entitled to such an opinion.

Needless to say, this would put a heavy burden on VA, but I would be hopeful that both cost and emotional savings from early resolutions, would offset the expenditures and reorganization that such a change would require. One possible quick fix would be for every VA physician treating a new condition to fill out a standard form addressing causation including possible comment on the need for the consideration of additional documentation prior to rendering an opinion.

2. I have seen too many claims that have remained in the VA administrative system despite the passage of more than a decade. Many of these are caught in a cycle of remands between a VA regional office (RO) and the BVA and the confusion and bureaucracy that is created by the back and forth transmission of documents. And many times the claimants are themselves part of the problem by continuously sending in more and more papers that in turn result in delay and frequently new adjudications.

In my view, in keeping with the theme expressed in my preceding recommendation, the adjudicative objective should be to finalize as many claims as possible at the RO level. In order better to achieve this objective, I also recommend that Administrative Law Judges (ALJs) or, at a minimum, Veterans Law Judges (VLJs), working at the final stage of RO adjudication where there is claimant disagreement, should insure that all necessary development has taken place and that, in the event of such a judge's continuing denial, should prepare, in lieu of a Statement of the Case, a decision as thorough as one now prepared by the BVA. In essence, what is being suggested is to decentralize high level administrative decisionmaking. (I would also note that the VA itself has taken initial steps in this direction by implementing a voluntary Decision Review Officer program staffed by more experienced adjudicators.)

Once such a decision was rendered, only formal motions that specified and articulated errors in the decision or made offers of proof would be accepted by the ALJ or VLJ. In such event, claimants would be permitted to hire counsel, if they chose to do so, to file such motions. Under present law, counsel may only be retained after an adverse BVA decision. Some will oppose such a change as upsetting the non-adversarial agency process, which in my mind is illusory once you have said "no" to a claimant. Permitting such a motion prior to an appeal to the BVA would allow for additional building of the evidentiary record often critical to success. Only after the prerequisite motion had been filed and a response from the ALJ/VLJ resulted in continuing denial would an appeal to the BVA be allowed. At this point, the record would be closed and no further evidentiary submissions could be made.

The approach suggested here will likely require many more ALJs or VLJs_than the number of VLJs presently at the Board. Some will come from the present Board but others will have to be hired along with staff. Thus, as with the additional expenditures and reorganization of the medical side of the house likely needed for the implementation of my first recommendation, this recommendation also will carry with it additional expenditures and the need for reorganization of the adjudicative side of the house. Nevertheless, it is possible that here, too, overall savings will re

sult by removing vast amounts of paper and vast numbers of the adjudicative hours required by the present system.

3. I am a big believer in the success of independent judicial review. It has caused VA decisionmaking to be light years ahead of where it was before such review by requiring that decisions be based on the real evidence and hard analysis, often previously missing. The bottom line is that judicial review has done much to bring about accurate decisions and helped insure fairness to our nation's veterans.

That said, judicial review has done little, if anything, to improve timeliness and, indeed, viewed objectively, can be seen as a real part of the problem. In the worst case, which happens more than occasionally, a veteran dies, leaving a case unresolved. Just as the administrative process itself is involved in the ever-revolving ROBVA two-step, judicial review turns that two-step into a four-step, adding on additional years to the process with a cycle of remands between the U.S. Court of Appeals for the Federal Circuit (Federal Circuit) and the Court and between the Court and the BVA. As to the latter, I myself have seen too many cases come back to the Court after two previous Court remands to the BVA and the passage of nearly a decade since the initial appeal to the Court was first brought. As to the former, an appeal to the Federal Circuit from the Court often carries with it 2 more years of the claimant's life; and in the event of a Federal Circuit remand back to the Court, I would estimate that another year can be added on, to say nothing of the additional years that will be involved if the Court in turn remands the case back to the BVA. Under existing law, there are four levels of possible appeal-one administrative appeal to the BVA and three levels of judicial involvement: The Court, the Federal Circuit, and the Supreme Court. Stated simply, this is more "justice" than the system can properly bear. Indeed, justice delayed is justice denied and the timeliness problem cannot be fixed without reforming the judicial process.

There is no compelling reason to have so many layers of judicial review. The only fathomable argument in support is that the party who has lost at the Court will have one more opportunity to demonstrate the rightness of that party's view. Although there is no question that the Court does make mistakes and is not omniscient, the same is true of the Federal Circuit. Indeed, the Court has far greater expertise in veterans' law. This capability is an outgrowth of nothing more complicated than the fact that this subject is the Court's sole business, while it is only a part-time focus of the Federal Circuit. Moreover, the reality of confusion over conflicting judicial decisions is directly proportionate to the number of judicial bodies involved in the process.

In my view, the best fix would be to make the Court, in all respects, the final arbiter of veterans' law, short of the Supreme Court to which appeals still, of course, should be allowed. Moreover, to provide for greater finality and fewer remands to the BVA, I would change the Court's organic law to clarify the Court's power to review BVA benefit-of-the-doubt determinations. Where a fully developed evidentiary record clearly reflects entitlement to a benefit or clearly reflects a claimant's inability to succeed, the Court, in spite of otherwise remandable BVA error, should end the matter, either respectively, with an award of a benefit or an affirmance based on non- prejudicial error.

The other possible approach to eliminating layers of judicial review would be to merge the Court into the Federal Circuit. The Federal Circuit's history itself reflects one of merger and spin-off. Such a merger would give the Federal Circuit a much bigger diet of veterans cases, thereby increasing its expertise. And it would provide for Article III decisionmaking, the very reason that the Federal Circuit was originally put into the process. Despite these considerations, it is my view, with 15 years of history behind the Court, that the preferable course of action would be to eliminate Federal Circuit review. First, I think it is preferable to have judicial review exclusively focused on veterans' cases. Second, even with an added focus on such cases, the Federal Circuit's primary focus will remain with intellectual property matters, the compelling reason for its own creation. At this juncture, I would think that few proponents still remain of the need for Article III review, short of the Supreme Court, of veterans cases. The Court's history shows that the threat of its being captured by its constituents has never materialized.

Lastly, the model of Article I court review being the final stop before review to the Supreme Court has been in place for half a century. Indeed, appeals from the only other Article I appellate court-The U.S. Court of Appeals for the Armed Forces are brought directly to the Supreme Court.

Thank you for the opportunity to be heard. If I can be of further assistance in providing our veterans with the best justice system possible, I stand ready to help in any way that I can.

Chairman CRAIG. Judge Kramer, thank you very much for that testimony.

Now, let's turn to Cynthia Bascetta, Director, Education, Workforce, and Income Security, Government Accountability Office. Cynthia, please proceed.

STATEMENT OF CYNTHIA BASCETTA, DIRECTOR, EDUCATION, WORKFORCE, AND INCOME SECURITY, GOVERNMENT ACCOUNTABILITY OFFICE

Ms. BASCETTA. Thank you, Mr. Chairman, other Committee members. We appreciate the opportunity to be here today to share our views on VA's performance in processing compensation and pension claims.

VA provided almost $30 billion in cash disability benefits to more than 3.4 million veterans and their survivors in fiscal year 2004. As you know, for years the claims process has been the subject of concern and numerous studies, mostly focused on persistently long waits for decisions, large backlogs, and inaccurate decisions.

We believe these longstanding concerns, coupled with the need to modernize Federal disability programs, support GAO's decision to designate VA disability programs as a high risk area.

My comments today draw from numerous GAO reports and testimonies on this topic. To update our work, we reviewed recent claims processing performance data, VA's fiscal year 2006 budget justification and its 2004 performance and accountability report. After briefly addressing the current state of the disability claims process, I would like to focus on factors that we believe impede VA's ability to improve its performance.

The bottom line, as we have all been discussing, is that VA continues to experience claims processing problems characterized by a large number of pending claims and lengthy processing times. VA did make considerable progress in reducing the size and age of its inventory through fiscal year 2003, but it has recently lost some ground.

Pending claims for example, have increased from 254,000 at the end of fiscal year 2003, to 340,000 by the end of this March. This is about 50,000 cases more than their goal of 290,000 cases for fiscal year 2005. More importantly, claims pending over 6 months, an indicator of a growing backlog, have increased more than 60 percent during the same period.

VA has also reduced the average age of its pending claims from 182 days at the end of fiscal year 2001, to 111 at the end of fiscal year 2003. But the trend is slightly upward, to 119 days at the end of this March. This is far from VA's strategic goal of 78 days by the end of fiscal year 2008.

VA's reported performance on accuracy is better: 87 percent of claims were decided accurately in fiscal year 2004, close to its goal of 90 percent. But despite improvements in accuracy, consistency remains a significant problem.

To ensure that similarly situated veterans who submit claims in different regional offices for similar conditions receive reasonably consistent decisions, we recommended in August 2002, and again in November 2004, that VA undertake systematic review of the consistency of its decisions. Just last Thursday, the IG published

the first study of disparities in average payments between States, which was initiated by the Secretary following the adverse publicity at the end of last year.

Our work also shows that program design and other aspects of the current system may constrain how much VBA can improve its performance, especially in the timeliness dimension. First, as you have heard, the law and court decisions, which have tended to protect veterans' rights and expand their entitlement to benefits, have at the same time adversely affected VBA's workload.

For example, presumptive eligibility for certain benefits has increased the volume of claims, and certain court decisions have added administrative complexity to the decisionmaking process. In addition, veterans are filing claims at a growing rate, as Admiral Cooper testified. And VA reports that the number of disabilities per claims is also increasing, compounding the complexity of the decisions they need to make.

Second, we reported that VBA will need to rely on productivity improvements to achieve its claims processing goals. VA assumes a 16 percent increase in rating related claims decided per FTE this year. However, based on available information, we believe it is unclear whether this is an achievable goal.

Third, program design may limit performance improvements in both timeliness and consistency. For example, timeliness is affected by the overall size of the workload, which consists mainly of claims filed for increases in disability ratings. Most of these claims are for veterans who have less severe disabilities.

We and others believe that consistency could be improved by consolidating regional offices. In fact, in 1995, VA listed more complete claims development and improved accuracy and consistency of decisions among the potential benefits of consolidation.

To sum up, the system we have today has evolved over several decades. Like other Federal disability programs, VA needs to modernize, and faces persistent and perhaps intractable problems improving timeliness and consistency. Tackling these issues will be critical to assuring that VA's disability programs meet the needs of 21st century veterans.

In addition, we believe that more fundamental reform, while a daunting task, presents an opportunity to achieve more than incremental gains in performance of VA's disability programs. That concludes my remarks.

[The prepared statement of Ms. Bascetta follows:]

PREPARED STATEMENT OF CYNTHIA BASCETTA, DIRECTOR, EDUCATION, WORKFORCE, AND INCOME SECURITY, GOVERNMENT ACCOUNTABILITY OFFICE

Mr. Chairman and Members of the Committee:

I am pleased to be here today to discuss claims processing issues in the Department of Veterans Affairs' (VĂ) disability compensation and pension programs. Through these programs, VA provided almost $30 billion in cash disability benefits to more than 3.4 million veterans and their survivors in fiscal year 2004. For years, the claims process has been the subject of concern and attention within VA and by the Congress and veterans service organizations. Many of their concerns have focused on long waits for decisions, large claims backlogs, and inaccurate decisions. Our work and recent media reports of significant discrepancies in average disability payments from State to State has also highlighted concerns over the consistency of decisionmaking within VA. In January 2003, we designated modernizing Federal disability programs as a high-risk area, in part because of VA's continuing challenges to improving the timeliness and consistency of its disability decisions.

You asked us to discuss the current state of VA's disability claims process and factors that may impede VA's ability to improve performance. My testimony today draws on numerous GAO reports and testimonies on VA's compensation and pension claims-processing operations. (See related GAO products.) To update our work, we reviewed recent claims processing performance data, VA's fiscal year 2006 budget justification, and VA's fiscal year 2004 Performance and Accountability Report. We did not perform independent verification of VA's data. We conducted our work in May 2005 in accordance with generally accepted government auditing standards.

In summary, VA continues to have disability claims processing problems. For example, as of the end of March 2005, rating-related claims 1 were pending an average of 119 days, 8 days more than at the end of fiscal year 2003, and far from its strategic goal of 78 days. During the same period, the rating-related inventory grew by about 86,000 claims to a total of about 340,000 claims. While VA has improved the accuracy of its decisions to 87 percent in fiscal year 2004, it is still below its strategic goal of 96 percent in fiscal year 2008. Further, we have identified concerns about the consistency of decisions across VA's regional offices. VA has begun studying one indicator of inconsistency, the wide variations in average payments per veteran from State to State, in response to adverse media coverage.

We identified factors that may impede VA's ability to improve its disability claims processing performance. The impacts of laws, court decisions, and the filing behavior of veterans can significantly affect VA's ability to decide claims, as well as the volume of claims received. Also, VA's ability to improve the productivity of its claims processing staff may affect its ability to improve performance. More dramatic gains in timeliness and inventory reduction might require fundamental changes in the design and operations of VA's disability programs.

BACKGROUND

VA's disability compensation program pays monthly benefits to veterans with service-connected disabilities (injuries or diseases incurred or aggravated while on active military duty) according to the severity of the disability. Also, VA pays dependency and indemnity compensation to some deceased veterans' spouses, children, and parents and to survivors of service members who died on active duty. The pension program pays monthly benefits based on financial need to wartime veterans who have low incomes, served in a period of war, and are permanently and totally disabled for reasons not service-connected (or are aged 65 or older). VA also pays pensions to surviving spouses and unmarried children of deceased wartime veterans. When a veteran submits a claim to any of VA's 57 regional offices, a veterans' service representative (VSR) is responsible for obtaining the relevant evidence to evaluate the claim. Such evidence includes veterans' military service records, medical examinations and treatment records from VA medical facilities, and treatment records from private medical service providers. Once a claim is developed (i.e., has all the necessary evidence), a rating VSR, also called a rating specialist, evaluates the claim and determines whether the claimant is eligible for benefits. If the veteran is eligible for disability compensation, the rating specialist assigns a percentage rating based on degree of disability. Veterans with multiple service-connected disabilities receive a single composite rating. For veterans claiming pension eligibility, the regional office determines if the veteran served in a period of war, is permanently and totally disabled for reasons not service-connected (or is aged 65 or older), and meets the income thresholds for eligibility. A veteran who disagrees with the regional office's decision for either program can appeal sequentially to VA's Board of Veterans' Appeals (BVA), the U.S. Court of Appeals for Veterans' Claims, and the U.S. Court of Appeals for the Federal Circuit.

In January 2003, we designated modernizing VA's disability programs, along with other Federal disability programs, as high-risk. We did so, in part, because VA still experiences lengthy processing times and lacks a clear understanding of the extent of possible decision inconsistencies. We also designated VA's disability programs as high-risk because our work over the past decade found that VA's disability programs are based on concepts from the past. VA's disability programs have not been updated to reflect the current state of science, medicine, technology, and labor market conditions.

In November 2003, the Congress established the Veterans' Disability Benefits Commission to study the appropriateness of VA disability benefits, including disability criteria and benefit levels. The commission held its first public hearing in May 2005.