The Finding
In May 2023, GAO published Electronic Health Records: VA Needs to Address Management Challenges with New System (opens in a new tab) (GAO-23-106731). About 79 percent of surveyed users (1,640 of 2,066) disagreed or strongly disagreed that the system enabled quality care. Efficiency scored worse: about 89 percent (1,852 of 2,074) disagreed or strongly disagreed that it made them as efficient as possible.
GAO made ten recommendations, and VA concurred with all of them. Two bear on evidence.
Recommendation 9 says VA should make certain that future system trouble tickets are resolved within established timeliness goals. Recommendation 10 tells VA to reinstitute plans for an independent operational assessment. As of January 2023, VA had assessed system performance at two sites but had not conducted that assessment as originally planned.
An earlier report, GAO-21-224 (opens in a new tab), asked VA to hold deployments until critical and high severity test findings were resolved. GAO has closed both of those recommendations.
Why It Keeps Happening
GAO reports that VA did not ensure trouble tickets were resolved within timeliness goals. VA then worked with the contractor to reduce the number of tickets more than 45 days old, but GAO also reports that the overall number of open tickets has steadily increased since 2020.
The operational assessment has a different problem: nobody has set a date. GAO's status note says VA had not established a target date for completing it.
The deployment calendar adds pressure. VA announced in April 2023 that it would halt deployments, and GAO's September 2026 testimony describes a restart in 2026. The same testimony says VA aims to reach about 170 sites by 2031. A program moving that fast can treat each go-live as the test of the last one, and the evidence GAO asked for gets deferred to the next wave.
Why It Matters to Your Bid
GAO's September 2026 testimony, GAO-26-109393 (opens in a new tab), says VA had not fully implemented 14 of 18 EHR recommendations across five reports as of August 2026, and warns that future deployments risk repeating earlier management problems.
An evaluator reading that testimony may look for a plan to hear from clinicians before a go-live and again after it, with named measures. Tickets matter too: can the offeror tie them to the screens and tasks behind them? Last comes the line between the contractor's own testing and evidence from a second set of eyes.
Recompetes and task order renewals raise the stakes. An incoming team inherits a ticket queue and a satisfaction score it did not create. A baseline taken early gives it a fair starting number.
Technical Approach
The method comes from Avalon's User Testing & Feedback Analysis service. It describes how the work is done, at the methodology level, on one named clinical workflow at a time.
- Plan around one workflow and one user segment. Write the research questions and task scenarios with the product owner from the workflow's top tasks, such as one frequent clinician task, confirmed against ticket and survey data. Draft the moderation guide and consent language, and write a Paperwork Reduction Act pathway memo. Research with federal employees acting within their duties is outside the PRA, so VA clinicians the program supplies keep the plan simple. The PRA decision stays with the agency.
- Run moderated, task-based sessions. Five to eight participants per segment, remote on the agency's authorized conferencing platform or in person, using the think-aloud protocol. Record task success, errors, time on task, and the path taken. Score the System Usability Scale where a benchmark helps. Recordings of internal-system screens are handled as potentially CUI-bearing material, and consent language clears the agency's privacy review before the first session.
- Build a feedback annex from data VA already holds. Take the trouble-ticket and survey exports, then rank ticket categories by volume and cross-check them against what the sessions showed. Where a high-volume ticket category matches an observed task failure, the report says so. Where they disagree, it says that too.
- Rate every finding and hand off a backlog. Affinity-map the observations, give every finding a severity rating on a defined scale, and attach a recommended fix with a rough level of effort. Each finding cites the session and timestamp where it was observed. The issue log arrives formatted for the program's tracker (Jira, GitHub, or Azure DevOps).
- Leave a repeatable scorecard. A one-page metrics summary (task success, errors, time on task, SUS) uses the same tasks each round, so the program can compare a pre-deployment round with a post-deployment one.
The boundaries matter here. This work tests and advises. It does not run the independent operational assessment GAO asked for, accept the system on the agency's behalf, resolve tickets, or perform load, performance, or security testing. It produces task-level user evidence that an assessment, or a program's own readiness decision, could draw on.
What "Fixed" Looks Like
GAO has not published closure evidence for recommendations 9 and 10, so these are targets.
For recommendation 9, tickets resolved within the timeliness goals VA has set, with ticket categories tied to the specific workflows that generate them.
For recommendation 10, an independent operational assessment of the system's suitability and effectiveness, with a target completion date.
GAO's closure of the 2021 recommendations shows what it accepts when evidence is countable. In July 2026, it verified that VA had no critical or high severity test findings at go-live across nine subsequent deployments.
Status Check
As of GAO's December 2025 update, recommendation 9 is open and partially addressed. Recommendation 10 is open, with no target date in GAO's status note. In its September 2026 testimony, GAO counted six of the ten GAO-23-106731 recommendations open and four partially implemented as of August 2026. GAO-21-224's two recommendations are closed.
Teaming Fit
| Field | Detail |
|---|
| NAICS | 541611, 541512 |
| Avalon's sub role | Moderated, task-based usability sessions with the agency's supplied clinician users on one named workflow, a feedback annex built from the agency's ticket and survey exports, and a severity-rated, backlog-ready findings log, under the prime's EHR modernization or digital services task |
| Small-business status | Small business; HUBZone (in progress) |
| Likely vehicles | VA T4NG2 (VA's Enterprise Contract for IT Services); GSA MAS (SIN 54151S IT Professional Services); GSA OASIS+ |
If VA EHR deployment or sustainment work is in your capture plan, Avalon teams as the sub that gives the program severity-rated, session-level clinician evidence and a ticket-to-task annex to put beside GAO's open recommendations.