The spark that started the US coronavirus epidemic arrived during a three-week window from mid-January to early February, before the nation halted travel from China, according to the most comprehensive federal study to date of when the virus began spreading.
That means anyone in the US who thought they had the virus in December or early January probably had the flu, public health researchers said.
Some people have claimed Americans were getting sick from the coronavirus as early as November and that infections were spreading in the US before any case was identified, said Dr Robert Redfield, the head of the US Centres for Disease Control and Prevention.
"[This] puts data into the discussion. Prior to this we had discussion without a lot of data," he said.
In the just-released study, CDC researchers collaborated with health officials in six states as well as genetics researchers and disease modellers in the Seattle area.
They drew on four kinds of data. One was reported illnesses by hospital emergency departments across the country. Another was a look back at about 11,000 respiratory specimens collected in January and February. A third was a genetic analysis of viruses taken from patients in California, Washington and the Northeast. Finally, autopsy findings from California also fit the theory.
The new coronavirus emerged in Wuhan, China, late last year. The first US infection to be identified was a traveller — a Washington state man who returned from Wuhan on January 15 and sought help at a clinic on January 19.
The White House announced a ban on travellers from China on January 31, which was implemented from February 3. Before that, some travellers were screened for symptoms at some airports. Only later did health officials realise the virus could spread before symptoms show rendering symptom-based screening imperfect.
White House officials in February declared the virus was contained and not a current risk to the American public. Until late February, coronavirus infections were too rarely diagnosed by emergency departments to be identified as a growing epidemic, the study found.
But limited spread in some communities was occurring in late January and early February, the study found.
Early instances of infection were found in the 11,000 airway samples collected from six states. The earliest was in a sample collected on February 21 in the Seattle area.
Genetic analysis from early cases suggest a single lineage of virus from China began spreading in the United States between January 18 and February 9.
One of the report's authors, Trevor Bedford, of Seattle's Fred Hutchinson Cancer Research Centre, has been tracking the pandemic using the virus's genetic code. He said there could have been a few undetected cases of the coronavirus in the US in December or January, but flu season was at its height.
"Based on just symptoms in January, it's almost certainly flu or another respiratory infection," Bedford said.
The earliest infections in Washington and California were from viruses originating in China, a genetic analysis found. Several introductions of the virus from Europe followed in February and March.
A separate study, released today by the journal Science, focused on the virus's beginning in New York City, which was hit hard by the virus.
It looked at the genetic information of coronaviruses gathered from 84 patients treated at the Mount Sinai Health System between February 29 and March 18. The information provides clear evidence for multiple introductions of infections — largely from Europe — into New York during the first weeks of March, the authors concluded.
The Trump administration announced a travel ban for most of Europe on March 11, and it went into effect March 13. The United Kingdom and Ireland were initially not included.
Ana Gonzalez-Reiche and her fellow authors wrote that the travel restrictions did little to prevent spread because community-driven transmission was already occurring.
The No. 2 official at the CDC previously acknowledged US officials were slow to understand how much the virus was spreading from Europe.
So, did the White House travel bans come too late to stop outbreaks from beginning?
The CDC study was not designed to assess that, said the agency's Dr Jay Butler.
"It's important to recognise the travel bans were intended to slow introduction of the virus," he said. "We knew it would be fairly unlikely that it would be completely kept out of the United States."
There have been more than 1.7 million confirmed Covid-19 cases in the US and more than 101,000 deaths.