VA Modernization Effort Failing to Reduce Claims Processing Times
Despite a massive effort to convert the Department of Veterans Affairs to a modernized, Web-based claims processing system, the VA has thus far failed to reduce wait times. In fact, they have actually made the claims process more difficult.
This was the finding of an internal investigation by the Inspector General’s office of the Department of Veterans Affairs. The IG published his report, Review of Transition to a Paperless Claims Processing Environment, earlier this month.
The VA’s goals were laudable: They intended for their massive transformation system to improve their claims processing throughput by 40 to 65 percent, while reducing their error rate. A key part of the transition was the adoption of their new Web-based automated claims processing software, VBMS.
The report faulted the VA for a failure to come up with a detailed plan for modernization or think through the system requirements. This poor planning, in combination with an incremental approach under the Agile system of software development, means that the Veterans Benefits Administration will continue to face significant problems.
Among the Inspector General’s findings:
- The Veterans Benefits Administration’s efforts to scan and digitize veterans’ claims have not been built from a detailed plan and analysis of requirements.
- Users stated that developers did not visit the pilot sites for the first time until August 2012 to understand their business needs and system functionality requirements.
- Users indicated that test scenarios were not realistic because functionality in the test environments did not replicate functionality in the production environment.
- Test cases did not process claims end-to-end within VBMS.
- While it took approximately 4 minutes to establish a claim with multiple contentions—contentions are veterans’ disabilities or health issues—in the legacy systems, it took approximately 18 minutes to establish the same claim in the VBMS pilot system.
- VBMS performance issues caused some documents to take 3 to 4 minutes or longer to open. On numerous occasions, inefficient system use of memory caused the system to crash and users had to reboot after opening multiple documents.
- VBMS-generated Veterans Claims Assistance Act letters contained errors and spacing issues and did not provide capabilities to edit or modify the documents. System users complained these letters often contained the wrong VARO addresses and VBMS did not provide the capability to make the necessary corrections.
- Ratings calculators had been deployed. However, because the calculators were not functioning properly, they were disabled and therefore not used to support disability claims determinations. A Rating Veterans Service Representative disclosed that rating an average claim in VBMS typically took 1hour in the legacy systems, but required 2 or more hours in VBMS.
- VA began scanning and digitizing veterans’ claims before it had a detailed plan and analysis of requirements for automating claims intake.
- VA proceeded with claims scanning and digitizing without a detailed plan outlining what this process would entail.
- Because a methodology was not well planned, VA encountered issues in scanning and digitizing claims folders to support the VBMS pilots. Specifically, the eFolders used to store the scanned images were disorganized and VA did not ensure proper management of hard copy claims folders.
Also, the IG’s office noted that as of September 2012, it was still not possible to complete a claim entirely using the VBMS system ‘end to end.’ This is, in part, due to the Agile approach to software development and project management, which brings the system to completion gradually, in stages.
According to the VA’s own data, average processing times actually increased at the VA claims centers designated as test sites for the VBMS rollout. The Fort Harrison facility reported an increase from 78 days to 125.6 days in the amount of time it took to process a pending disability claim during the test period, from October 2011 (pre-VBMS) to September 2012 (post VBMS). The Wichita facility reported an increase of 159.2 to 172.3 days.
Data for the other two Beta sites, Salt Lake City and Providence, RI, were not yet available for comparison, according to the report.