By Lynne Peeples

The first case of COVID-19 in the United States was reported on Jan. 19. However, the novel coronavirus may have arrived and begun spreading domestically as early as late December, according to a new retrospective study in Los Angeles.

Researchers compared the number of patients visiting University of California, Los Angeles (UCLA) medical facilities with complaints of a cough this past winter with the same period in previous years. Beginning the week of Dec. 22, 2019, numbers were “consistently elevated every single week in our facilities,” said Joann Elmore, MD, a professor of medicine at the David Geffen School of Medicine at UCLA and lead author on the paper (J Med Internet Res 2020;22[9]:e21562). 

ADVERTISEMENT

“You could say that maybe this was a bad year for influenza. But this would be really, really bad,” Dr. Elmore added. She suggested that neither wildfire smoke nor vaping illness would explain the difference, as last year was not a major fire season and the CDC was already reporting declines of e-cigarette/vaping-associated lung injury by that time.

She and her colleagues searched the electronic health records from more than 9.5 million outpatient visits, nearly 575,000 ER visits and almost 250,000 hospital admissions going back more than five years. To identify signs of possible COVID-19, they pulled out records that included notes of a patient complaining of a cough. Among clinic visits, they counted nearly 3,000 cough patients in the 13 weeks between Dec. 1, 2019, and Feb. 29, 2020, which was about 1,000 more than the average number seen during the same three-month period in the previous five years. 

ADVERTISEMENT

“That was my first, ‘oh my,’” Dr. Elmore said. Her team then found similar patterns in records from ER visits and hospital admissions. “Again, we saw numbers well above what you’d expect by chance,” she added.

Even if just one or two of those excess cases in late December or January represented COVID-19 patients, she and her co-authors argue, that would be supportive of an earlier arrival of the novel coronavirus, SARS CoV-2. Dr. Elmore suggested that the findings, while not definitive, “add to the growing body of literature.” 

Other studies have hinted at an earlier presence of the virus. George Rutherford, MD, the head of the Division of Infectious Disease and Global Epidemiology at the University of California, San Francisco, pointed to a preprint study that evaluated satellite imagery of hospital parking lots and internet searches for COVID-19 symptoms in Wuhan, China. The authors concluded that “both signals precede the documented start of the COVID-19 pandemic in December.” A study of sewage water in Italy found traces of the virus back in December as well.

“That’s all suggestive, as is this new study,” Dr. Rutherford said. “The problem is, last year was a big influenza A year. Differentiating one from the other is going to be problematic.”

ADVERTISEMENT

Dr. Elmore and her team did not have samples to test for the virus itself. 

Still, the narrative that the first spillover of the novel coronavirus from animal to human happened in late December may be incorrect, according to Dr. Rutherford. “The virus had been circulating for at least a few weeks in southern China before it got into that market and started a secondary outbreak. That’s become increasingly clear,” he said. 

“It likely arrived in the western U.S. in late January,” he added. “It could’ve been circulating a little earlier—maybe. There are a lot of flights back and forth to China from California.” 

Another study identified seven patients treated at the Los Angeles County + USC Medical Center in mid-March for what turned out to be COVID-19 (JAMA 2020;323[19]:1966-1967). The researchers suggested the patients had become infected through sustained community transmission; none of them had clear ties to anyone who had traveled to a COVID-19 hot spot. However, Brad Spellberg, MD, the chief medical officer at Los Angeles County + USC Medical Center and the lead author on that paper, suggested the figure would have been “much higher” had the virus been circulating widely in Los Angeles in December. 

“Once COVID-19 starts spreading, in a location with no public health measures in place, it will keep spreading. The numbers of infections will double approximately every five to seven days,” added Ben Cowling, PhD, a public health professor at the University of Hong Kong School of Public Health, who was also not involved in the new study. “If the spike was due to COVID, it would have kept growing and growing.

“I can believe that COVID-19 arrived into LA earlier than we had previously thought,” he said. “But I am not convinced that COVID-19 is responsible for the spikes [in patient visits for cough seen] in this paper.”

Drs. Elmore, Rutherford, Spellberg and Cowling reported no relevant financial relationships.