It was mid-March when Dr. Jeffrey Barnard, the chief medical examiner of Dallas County, started to worry. As the number of deaths due to COVID-19 mounted nationwide day-by-day, Barnard felt he wasn’t doing enough in the fight against the disease.

Nearly everyone in the county who is killed or dies of natural causes outside the care of a doctor or hospital ends up in his morgue—an average of fifteen decedents daily. Barnard’s team had begun some screening for the novel coronavirus as a potential cause of death, but a dearth of resources, including swabs, limited them to only about one test per day. They therefore had to be extremely selective, swabbing only those for whom there was evidence of potential COVID-19 symptoms prior to death.

“How many people are being missed because they’re asymptomatic?” Barnard remembers thinking. “Even if they died of something else, like an accident, the public and the health department need to know because they were still potentially spreading the virus.”

Each case confirmed by postmortem testing of the recently deceased enables contact tracing—allowing county health staff to seek out and warn decedents’ friends, family, and coworkers about their possible exposure to the virus. It also might let police and EMTs know that they too may have been exposed by responding to the scene of a fatality involving one of these cases.

On May 11, after swabs and other necessary equipment became more readily available, Barnard dedicated a portion of his in-house DNA lab to COVID-19 testing. Not having to send out samples to other labs, where results took much longer because of backlogs, allowed his office to test all of its decedents, regardless of any other noted cause of death. Among the six Texas counties with the greatest number of reported COVID-19 cases—Harris, Tarrant, Bexar, Dallas, El Paso, and Travis—Dallas is deploying by far the most robust postmortem testing.

Since it began its widespread testing, Barnard’s team has swabbed 95.5 percent of the total 556 deceased it has examined, as of June 18. (Only 353 tests have been completed, however, after a setback in the supply chain created a temporary backlog). Eleven of the completed tests found COVID-19 (a 3.1 percent positivity rate). About half of those confirmed cases came from decedents who arrived with no indication of the presence of the coronavirus—cases that would not have been discovered without Barnard’s more aggressive testing.

It’s difficult to know how widespread postmortem testing has been nationwide since major institutions like the Centers for Disease Control and Johns Hopkins University aren’t tracking it as a separate statistic from overall testing numbers. Spokespeople for the Harris, Tarrant, Bexar, and Travis county medical examiner offices (each of which, like Dallas, also serve some smaller surrounding counties) told Texas Monthly that they’ve limited their testing to decedents who displayed potential symptoms of the disease prior to death, which is as much as the postmortem testing guidelines from the Centers for Disease Control suggest. (El Paso County didn’t respond to multiple inquiries about its testing.)

During roughly the same time in which Dallas has aimed to test all decedents, Tarrant has tested only 6.3 percent of its 380 decedents, with 3 of them positive for COVID-19 (12.5 percent positivity rate). Travis reports testing 12.2 percent of 288 decedents, with 2 of those positive for the coronavirus (5.7 percent positivity). And Harris County has swabbed 12.1 percent of its 596 decedents. Twenty-six of those were positive for COVID-19, a markedly higher positivity rate (36 percent) than in Dallas or elsewhere. The public information manager for the Harris medical examiner didn’t know why this was the case, but possibilities that might explain the disparity among the counties’ positivity rates could be differences in the sensitivity or accuracy of the testing kits each is using.

While the other counties that Texas Monthly surveyed have conducted at least some COVID-19 swabbing since mid-March, Bexar County could not provide a date for when its postmortem testing began. A spokesperson also said the Bexar medical examiner hasn’t tracked the number of tests that it has done. Its records show six deaths attributed to COVID-19, the earliest of which occurred April 30.

Postmortem testing holds the potential to turn up evidence that COVID-19 may have been present in Texas much earlier this year—useful, perhaps, in better understanding whether we missed warning signs. The methods for doing that, however, are trickier than the nasopharyngeal swab used on the recently deceased. Blood stored from past decedents (Dallas, for example, keeps those samples for a year) could be tested for antibodies that indicate the immune system was fighting the coronavirus, but the track record of antibody tests hasn’t shown them to be reliable yet. There’s also the possibility of reexamining tissue samples taken from the lungs during an autopsy, looking for alveolar damage indicative of the disease. Only Harris County said that it’s planning to reconsider some old cases dating as far back as the beginning of 2020.

In Dallas, Barnard said taking another look at old samples is a possibility, but for now, “We’re just trying to keep our head above water.” As the county, along with much of the state, has seen its COVID-19 cases, hospitalizations, and deaths rise, that trend has likewise been reflected in the medical examiner’s morgue.

That’s why Barnard has no plans to stop testing for the coronavirus until a vaccine or some highly effective treatment has been found. “I don’t see this going away anytime soon,” he says. “No matter what the people think who say it’s a hoax. They’re wrong.”