NEW YORK (NYTIMES) - When there were just over a dozen official recorded cases of the new coronavirus in the US, at least 50 people who later tested positive were already feeling ill. By the time 50 cases were officially confirmed, at least 1,200 people had already started showing symptoms of Covid-19, the disease caused by the virus.

This new picture of the US coronavirus outbreak is based on data published by the Centres for Disease Control and Prevention (CDC) that estimates the date on which people who were later tested and confirmed positive for Covid-19 said they first started to experience symptoms.

The data helps show "just how far behind the virus we are," said Justin Bahl, an associate professor of infectious disease and epidemiology at the University of Georgia, "and, potentially, how much more transmission is happening in the community before we can identify and isolate those people."

Looking at the date patients' symptoms started tells epidemiologists more about the true progression of an outbreak than confirmed case numbers alone. That's because there are often lags between the moment someone is infected, when they fall ill and when they receive a positive test result.

"Ideally you would have a curve that shows when everyone got infected, but that's often impossible to know," said Dr Cecile Viboud, an epidemiologist at the National Institutes of Health's Fogarty International Centre.

"People don't know when they got infected; they don't suddenly light up with a sign that says, 'You got infected,'" she said. "So the closest we can get is symptom onset, and since we know the incubation period, we can then estimate when they got infected."

The date a person became ill is usually determined from a patient's medical history or interviews by public health professionals. In some instances, the CDC has estimated onset dates for coronavirus cases.

"There is the whole time people are sick before they seek medical help," Assoc Prof Bahl said. "All the time beforehand, that's potential transmission time."

A separate study that examined the case history of the first Covid-19 patient in the United States offers a real-world example of the gap between when symptoms start and a case is identified. (The CDC did not publish individual case timelines.)

The first reported US case was a 35-year-old man in Washington state who had recently travelled to Wuhan, China, where the virus was first detected. He reported seeking medical help because he had seen an alert from the CDC about the new coronavirus outbreak in China that matched his symptoms. Most new infections in the United States today come from local contacts.

The time between the start of illness and case identification tends to be greater at the beginning of an outbreak, when people are less aware of the symptoms, but it decreases over time. The current average lag from symptom onset to positive test is four days, according to the CDC.

Narrowing the gap is crucial to combating the virus, Assoc Prof Bahl said: "The ultimate goal is to identify and isolate. If we can shorten the time frame, then we can control the epidemic."

Symptom onset data is not available for every reported Covid-19 case in the United States, but even the subset published by the CDC, which includes more than 40,000 cases, shows a substantially higher number of infections in late February and early March compared to confirmed case tallies from the same period.

This data shifts public understanding of the US epidemic curve days to weeks earlier, but it still does not reflect the true scale of Covid-19 cases in the country. The symptom onset data is based on patients who tested positive and misses cases that are treated at home and others that may not receive testing.

By some estimates, the United States has reported only a fraction of the coronavirus cases currently circulating in the country. The CDC also estimates that as many as 25 per cent of people infected may not show symptoms at all.

The data also does not reflect the real timeline of infections - what Dr Viboud called the "ideal" epidemic curve. The virus' incubation period is estimated to be five days on average following exposure but can last up to 14 days. That means people reporting symptoms today may have been infected up to two weeks ago.

Efforts are underway to trace the true beginning of the US outbreak, including genetic sequencing methods that track the virus's spread.