Witness Testimony of David E. Hunter, Ph.D., Cost Analysis and Research Division, Research Staff Member, Institute for Defense Analyses
Institute for Defense Analyses Study on Analysis of Differences in Disability Compensation in the Department of Veterans Affairs
Mr. Chairman and Members of the Subcommittee, I am pleased to come before you today to discuss IDA’s work on disability compensation conducted for the Department of Veterans Affairs (VA). Let me begin with some background on the study and then I will summarize findings and recommendations.
A total of 2.6 million veterans were receiving disability compensation as of September 2005. The average yearly award for the entire United States was $8,890, and the average varied across states from more than $12,000 in New Mexico to less than $8,000 in Ohio.
In addition, the percentage of veterans receiving compensation differed from state to state. Nationwide, 10.8 percent of veterans were receiving compensation, and this varied from nearly 18 percent in Alaska to about 7 percent in Illinois.
In May 2005, the VA asked the Institute for Defense Analyses (IDA) to conduct a study of the major sources of the observed variation across states in:
- The average payments to veterans receiving disability compensation; and
- The percentage of veterans receiving disability compensation.
My testimony today will be based on the results of that study, which have been documented in IDA Paper P-4175.
There are two potential reasons for the observed state-to-state variations in average awards. First, there may be systematic differences across states in the claim adjudication process. Second, the variation may reflect differences across states in the characteristics of the veteran populations.
Our study quantified the amount of variation attributable to states having veteran populations with different characteristics. To do this, we identified and collected relevant data on disability compensation recipients and the veteran population and used these data to test a wide variety of hypotheses. We used data as of September 2005 as the baseline for our analysis. To identify historical trends, we also examined available historical data.
- Impact of Maximum Awards
Payments to veterans are based on overall disability level, from 0 percent to 100 percent in increments of 10 percent. In addition, veterans may receive an award of Individual Unemployability (IU), which pays them the equivalent of 100 percent disability.
We found that the percentage of recipients receiving a maximum award (100 percent or IU) explains the vast majority of the observed state-to-state variation in average compensation. We calculated that 94 percent of the variation was explained solely by differences across states in the percentage of compensation recipients receiving a maximum award.
This result reflects two underlying facts. First, although veterans receiving maximum awards make up a small percentage (17 percent) of all compensation recipients, they receive the majority (58 percent) of the total compensation dollars. Second, there is variability across states in the percentage of compensation recipients receiving maximum awards, ranging from a low of 10 percent in Alaska to a high of 30 percent in New Mexico.
For the maximum awards, we found the IU awards exhibited the greatest variability across states and alone accounted for 75 percent of the observed variation in average awards. The percentage of compensation recipients receiving IU per state ranges from a low of 3 percent in Maryland to a high of nearly 20 percent in New Mexico.
Given these findings, the key issue our study had to address was: To what extent do the state-by-state variations in maximum awards reflect different treatment of similar veterans and to what extent can they be explained by differences across states in the veteran populations?
- Demographic and Claim-Specific Factors
We tested a wide variety of demographic and claim-specific factors to identify those that influence the award outcomes. We identified three major factors that contribute to the observed variation across states in average disability compensation awards.
- Post-Traumatic Stress Disorder (PTSD). We found that all states have high average awards for veterans with PTSD. However, there are large differences across states in the proportion of compensation recipients with a PTSD award. This difference in the percentage of recipients with a PTSD award accounts for 40 percent of the observed variation in average awards across states.
- Power of Attorney (POA) representation. Nationwide, veterans with POA representation receive an average annual award of over twice that of veterans with no POA representation. We found that differences across states in the percentage of claims with POA account for 16 percent of the variation in average award across states.
- Period of service. The average award for Vietnam veterans is $11,670 – the highest for any period of service. As a single predictive factor, differences across states in the period of service of recipients accounts for 8 percent of the observed variation in average awards.
We calculated the combined effect of the three main factors that we identified: PTSD, power of attorney, and period of service. Note that these factors are correlated, and we could not simply add the percentage of variation explained by each single factor to calculate their combined explanatory power. Taking account of the correlations, we found that 50 percent of the variation across states is explained by these three factors.
Using a more detail model that included several demographic factors related to the veteran’s county of residence, which proved to correlate with average awards, we found that as much as 70 percent of the variation across states is due to differences in the recipient populations. While these observed correlations are of interest, it is important to be careful in interpreting them; they almost certainly do not reflect direct causal relationships.
- Variation in the Percentage of Veterans Receiving Compensation
Our second area of study was the sources of differences in the percentage of veterans receiving compensation.
Two top-level factors influence the percentage of veterans receiving compensation. These factors are application rates and adjudication results. Of these two factors, we found application rates to be more important than adjudication results in explaining variation across states. Using available data over the past 10 years, we calculated that differences in application rates explained over 70 percent of the variation in the percentage of veterans receiving compensation.
We also tested a wide variety of demographic factors to identify those that influence the percentage of veterans receiving compensation. We found that military retirees are over four times as likely to receive compensation as non-retirees. This alone accounts for over 40 percent of the variation across states. The percentage of veterans receiving compensation also varies by period of service. We calculated that differences in state veteran populations by period of service account for 12 percent of the variation across states. Unfortunately, available veteran population data and demographic information on all applicants are insufficient to quantify the total variation accounted for by the combination of these demographic factors.
- The Adjudication Process
As noted above, we found that state-to-state differences in compensation recipients explain 50 percent to 70 percent of the variation in average awards. This implies that as much as 30 percent to 50 percent of the variation in average awards could be due to differences across states in the adjudication process. We examined the VA’s adjudication process and found that most rating decisions are made locally and often call for subjective judgments. We also found that initial and ongoing rater training varies by regional office and has changed over time. On-the-job training and mentoring, an important source of rater education, promotes uniformity within a regional office. The current national quality review program (STAR) focuses on accuracy of individual claims and does not attempt to promote consistency. There is no program to monitor trends in ratings across regional offices aimed at improving understanding of regional differences. For these reasons, the current adjudication process has the potential for allowing regional differences to develop and persist.
Based on our findings and observations, the IDA report presented six recommendations for consideration by the VA.
- Standardize initial and ongoing training for rating specialists.
The VA should consider preparing a set of test cases as part of ongoing training procedures.
- Standardize the medical evaluation reporting process.
Many raters identified variation in quality of medical reports as a possible cause of variation in awards and stated that poor quality reporting hinders their ability to make an accurate rating decision.
- Increase oversight and review of rating decisions.
The VA could strategically select a more significant fraction of rating decisions for review. This selection process should target claims with high leverage and evaluate each on service connection, degree of disability, and IU status determination.
4. Consolidate rating activities to a central location.
Consolidation would remove many of the underlying differences across regional offices that contribute to potential inconsistencies in decisions. Realizing that this may not be feasible, we note that consolidation to fewer regional offices or having regional offices specialize for certain claim types would also improve consistency.
- Develop and implement metrics to monitor consistency in adjudication results.
These metrics would target the key factors that impact the variations in average awards and the percentage of veterans receiving compensation.
- Improve and expand data collection and retention.
The ability to monitor variances is currently limited by lack of available data. Most notably, the VA has not historically tracked data on denied claims. Such data are needed to further understand the underlying reasons for differences across states in the composition of claim recipients. For instance, data do not exist to show how much the denied claims contribute to differences across states in the mix of compensation recipients.
Mr. Chairman and Members of the Subcommittee, that concludes my statement, and I am available for questions.