Pentagon COVID-19 Case Registry Is Riddled with Errors, Watchdog Finds

Army specialist performs COVID-19 testing at the mobile lab.
U.S. Army Spc. Crystal Gonzales, assigned to the 44th Medical Brigade, conducts COVID-19 testing at the mobile Analytical Laboratory System unit outside the Javits Center in New York City, May 18, 2020. (U.S. Army Photo by Cpl. Rachel Thicklin)

A Defense Department database of military health system patients diagnosed with COVID-19 is incomplete and inaccurate, shortcomings that make it ineffective for steering medical treatment or public health decisions on the illness, the Pentagon's top watchdog has found.

A DoD inspector general report released this month found that the department's COVID-19 registry did not contain roughly 7,200 patients who qualified for inclusion and there were errors in 24 of 25 records reviewed by inspectors.

The issues raise questions about the database's overall accuracy, which was supposed to be right for at least 90% of the records, according to its contract, said Carol Gorman, the assistant DoD inspector general for audit, cyberspace operations and acquisition, contracting and sustainment.

Read Next: A Mysterious Osprey Clutch Issue Caused the Deaths of 5 Marines. But V-22s Remain Flying.

Since last year, lawmakers have raised concerns over the accuracy of DoD medical databases and registries, including the Defense Medical Epidemiology Database, or DMED, a data set that includes current and historical data on patient health conditions, hospitalizations and medical events of service members.

Last week, Sen. Ron Johnson, R-Wis., wrote to Defense Secretary Lloyd Austin over discrepancies in the numbers reported to his office for 15 illnesses or conditions. Johnson said a whistleblower has questioned the accuracy of DoD data, and the department failed to disclose that it provided incomplete data to the senator when asked.

For example, Johnson noted, the DoD provided his office erroneous information that diagnoses of hypertension rose by 2,181% in 2021, when compared with the years 2016 to 2020. After correcting the data that led to that conclusion, the DoD said the number of cases rose by 2%.

A whistleblower told Johnson, however, that cases rose by 12%.

"I am grateful to the whistleblowers who continue to come forward to provide my office with information you and other DoD officials are unwilling to produce," Johnson wrote in the letter.

The Defense Department's public COVID-19 dashboard, which differs from the registry, reported that the DoD had 740,942 cases from 2020 to 2022, with 690 deaths, including 96 service members and 36 dependent family members. The remainder of the deaths were among civilian DoD employees and contractors.

Overall, there have been 103 million confirmed cases of COVID-19 in the U.S. and 1.1 million deaths, according to the World Health Organization.

The Defense Health Agency established the registry in July 2020 to track all COVID-19 cases and events in the military health system. To create the database, the U.S. Army Medical Research Acquisition Activity awarded a contract worth up to $4.75 million to an unnamed company to review patient records, confirm the records and enter the information into the registry.

The contract was renewed in 2022 for continued support services for $2.3 million.

But the DoD inspector general audit found that the database identified 182,159 COVID-19 cases among DoD patients in the health system, while the DoD IG identified 189,372 patients -- 7,213 more than the previous tally.

The auditors also examined 25 patient registry records and found 24 errors in their demographics information, symptoms or pre-existing condition data -- information that physicians could use to assess and improve COVID-19 medical care and track the illness.

"A properly designed and executed registry can provide the DoD with real-time information on patient outcomes and can contribute to the research and development of treatments," the auditors wrote.

Given that the contract called for a high rate of accuracy in building the database, the IG questioned whether the DoD has recourse to recoup any of the $6.2 million it spent on the program.

"We recommend that the [Joint Trauma System] Chief conduct an analysis to determine whether the contractor complied with the terms of the contracts," they wrote.

To qualify for the registry, military hospital or clinic patients must have had a confirmed positive test, been hospitalized for COVID-19 or told to self-isolate or quarantine. As part of the process, the database was loaded with data for 187 categories in the system, including demographics, health data and more.

Since the DoD's electronic medical records didn't automatically glean the information for the registry, the department contracted out the job to a company that hired reviewers and nurses who were supposed to ensure that the registry was complete and accurate.

But the DoD didn't follow guidelines published by the Department of Health and Human Services for patient registries, and it also lacked its own policies for developing and building patient registries, the IG noted.

The watchdog made a dozen recommendations to improve the process in the future, recommending that the DoD build a policy for developing and populating patient registries; review all existing registries for accuracy and fix them if necessary; and determine whether the department should receive some of its money back.

The DoD pushed back on 75% of the recommendations, however, arguing against the IG's methodology for determining errors, its characterization of the contracting process and payment concerns.

DoD officials said they would work with the Agency for Healthcare Research and Quality and other federal agencies to develop policies for overseeing and developing future patient registries.

Meanwhile, Johnson has been seeking data from the department as part of a concerted campaign against the federal response to the COVID-19 pandemic, vaccine mandates and government shutdowns.

He ran on a platform that included a belief that the COVID-19 vaccine caused harm to patients and that such incidents are not rare occurrences. He has been seeking information from the department to determine the scope of any vaccine-related injuries among service members.

Side effects of COVID-19 vaccines are mainly mild and disappear within days of receiving the vaccine, including fatigue, pain at the injection site, joint pain, headache and malaise.

According to the Centers for Disease Control and Prevention, serious but rare side effects include anaphylaxis, in 5 cases per one million doses; Guillain-Barreé Syndrome, an immune response that can cause weakness and paralysis that occurred among recipients of the Johnson & Johnson vaccine at higher rates than the other immunizations; heart conditions such as myocarditis and pericarditis, particularly in young men in their teens and 20s, following the second dose of the vaccine; and severe blood clotting, which occurred in 4 cases per one million doses.

The CDC noted, however, that millions of Americans have safely received the COVID-19 vaccine and the benefits continue to outweigh any potential risks.

-- Patricia Kime can be reached at

Related: Nearly 100 Deaths, Half a Million Cases: The Toll from 3 Years of the Coronavirus Pandemic on the Military

Story Continues