When the samples were analysed later in each city, most turned out to be flu, but some turned out to be positive for SARS-CoV-2, the virus that causes COVID-19, the researchers said.
In Seattle, they estimate there could have been more than 9,000 cases by March 9, the week schools were closed, although official figures at the time reported 116 cases.
While 6,153 patients have so far died of the virus in the country including 2,307 patients died in Sindh, 2,180 in Punjab, 1,236 in Khyber Pakhtunkhwa, 173 in ICT, 138 in Balochistan, 60 in Gilgit-Baltistan and 59 patients died in AJK after contracting COVID-19.
Like flu, when someone does develop symptoms, they usually begin with fever, cough and fatigue, potentially leading to shortness of breath. According to the latest batch of documents released by SAGE, the experts mentioned that people who are returning from holidays must be tested twice.
"Even before we realized that COVID-19 was spreading, the data imply that there was at least one case of COVID-19 for every two cases of flu", said Lauren Ancel Meyers, a professor at UT Austin.
And about a third of those cases were in children. But Seattle alone may have actually had thousands of infections at that point, according to a new study, because so many illnesses were thought to be the flu instead.
The team found a similar pattern when they ran their model on swabs from Wuhan, China, where coronavirus was first identified last winter.
When Wuhan locked down on January 22 - the first city in the world to do so - 422 cases had been identified there.
Although in both cities the majority of the reports were flu, they both demonstrated a wider spread of the virus than reports from the time suggested, weeks before lockdown measures were implemented in either city.
Dr Meyers notes that this pattern may be repeated now as restrictions to slow the spread of coronavirus are lifted in many parts of the USA and world.
'We can go back and piece together the history of this pandemic using a combination of investigative techniques and modeling, ' she said. "This helps us understand how the pandemic spread so quickly around the globe and provides insight into what we may see in the coming weeks and months".