Stanford Study shows BVA’s accuracy claims are misleading

A recent paper published by the Stanford Institute for Economic Policy Research confirms what veteran’s advocates have long known to be true – BVA’s claimed 90-95% accuracy rates are, at best, misleading. 

In this study, researchers looked at data on nearly 600,000 BVA appeals over a 15 year period to determine whether BVA’s quality review process reduced appeals or reversals.  BVA randomly selected 5 percent of original appeals and 10 percent of appeals remanded from the Court of Appeals for Veterans Claims (CAVC) for review. Draft decisions were reviewed so that errors could be corrected before the decisions were released. They analysis found “the program had no appreciable effect on reducing appeals or reversals.  Based on internal records, we demonstrate that this inefficacy is likely by design, as meeting the performance measure of ‘accuracy’ was at cross purposes with error correction.” 

In 2018, 4,842 BVA decisions were appealed to CAVC.  Only 382 (7%) of these BVA decisions were affirmed. 80 percent were remanded in whole or in part.  In the same year, BVA claimed to have maintained a 93.5% accuracy rate.  If the definition of ‘accuracy’ is to be immune to reversal or remand, then clearly these numbers don’t add up.   The Stanford Study pointed out that BVA’s standard of review is not consistent with CAVC’s, citing internal BVA documents which show that quality review errors are only called when they are considered undebatable. The result is that there is no statistically significant difference in the rate of remands and reversals for a case that has been through the quality review process and a case that has not.

In addition to concluding that BVA’s quality review process seems to be a waste of time and resources, this study reveals that the accuracy numbers reported by BVA are a misrepresentation. BVA is not calling quality review errors for the failure to adequately explain decisions, despite the fact that this type of error represents 62 percent of the remands from CAVC. Quoting from the Stanford study:

“The VA’s own Office of General Counsel (OGC) has sharply questioned the BVAs reported accuracy rate of 94 percent. OGC noted that in 2009, CAVC alone had reversed or remanded a higher absolute number of cases than would be mechanically possible under a mere 6 percent error rate. Vice Chairman Keller responded by stating that CAVC reversal or remand did not necessarily mean that the BVA decisions contained error. Keller argued that a remand for failure to provide an adequate explanation - textbook administrative law - should not be counted as error because the standard is ‘highly subjective and inconsistently applied.’”

BVA’s accuracy rate is is an important performance metric, which is reported to congress to support budget requests. By manipulating the standard of review to produce a 90-plus percent accuracy rate, BVA creates the illusion of effective and efficient workload management and renders their claimed accuracy rates meaningless.