New York City's seven-day average test positivity for COVID-19 has declined to 2.5%, alleviating some of the anxiety over the imminent closure of public schools.

In a revision, the city Health Department is now reporting that the daily share of New Yorkers testing positive for COVID-19 has been declining since Sunday. The latest daily test positivity rate was 0.32%, a stunning drop considering that the level had soared above 4% one week ago.

Because the data collection process involves lags and errors, health officials routinely go back and revise prior published test reports. What that means is that all of the latest indicators are subject to change in the future.

Bill Neidhardt, a spokesperson for Mayor Bill de Blasio, said that the city expected the positivity rate to increase "as more tests results are entered into the system, but do not expect that increase to push today’s 7-day average to 3%."

The absolute number of people testing positive is continuing to rise in New York City. The seven-day average of new daily cases is now above 900, having been at roughly 500 two weeks ago.

Still, Saturday's report is expected to bring a small measure of relief for public school parents. De Blasio has consistently said that should the average positivity rate go above 3% he would close city schools, a move that would disrupt the roughly 300,000 students receiving in-person instruction. On Friday, when the positivity rate was believed to be steadily climbing, he told parents to prepare for a closure as soon as Monday.

Dr. Denis Nash, an epidemiologist at CUNY, said the sudden drop could be due to more (negative) New Yorkers showing up to get tested following the election and the ensuing celebrations over the weekend. Across the five boroughs, there have been reports of long lines outside CityMD clinics.

"It's really tricky to interpret the data right now because it could change but there does seem to be a clear trend," he said.

Another possibility is that ongoing test delays may be affecting the sample of results. Nash noted that the practice of pooled testing, in which laboratories combine tests from small groups of people and analyze them together, means that positive tests get reported later. The number of people tested fluctuates from day to day, from between around 15,000 to nearly 60,000, according to city health data.

The current median turnaround time for COVID tests in New York City is two days. But demand is expected to surge as the holidays approach and some residents try to get tested before meeting up with family members. New Yorkers were also recently required by Governor Andrew Cuomo to get tested before and after traveling out of the state.

In an illustration of the way how much revisions can affect test data, the city's report on Friday showed the daily positivity rate rising.

The graph on the NYC Health Department site on November 13, 2020

But on Saturday, the trajectory is now moving in the opposite direction.

The graph on the NYC Health Department site on November 14, 2020

Nash emphasized that such updates to the data are normal. Economic data, such as weekly U.S. employment numbers, is similarly revised and backfilled.

Given that, however, he advised New Yorkers not to place too much stock on the latest readings.

"Let's wait for a few days to see how these indicators shake out," he said.

Asked about testing delays, Michael Lanza, a spokesperson for the Department of Health said, "As we’ve said for the last few months, some facilities are slower to report than others. This is not new."

Lanza did not respond to the question of whether pooled testing was contributing to the delays.

In general, public health experts say test positivity is a complicated measure that depends on who exactly is getting tested and it should not be used as the sole measure of virus spread. For example, health officials will often talk about a "day-of-week effect" with test data, in which more positive results tend to occur on Sundays because typically only the sickest people will seek to get tested on a day when many clinics are closed.

Because of that, Nash said he prefers to keep a close eye on hospitalizations and deaths. While there have not been any sharp increases, that measure is gradually increasing in New York, similar to the rest of the country.

Statewide, the number of people hospitalized for coronavirus has grown to more than 1,600, after having been around 400 in September. That represents a 300% increase in a span of two months.

Although Saturday's indicators amounted to a reprieve for some public school parents, some experts continued to challenge the city's criteria for closing schools given the low level of infections in the school system.

"The 3% trigger is an arbitrary threshold," said Dr. Wafaa El-Sadr, an epidemiology professor at Columbia University, who cited the low level of reported infections—less than a quarter of a percent—in the school system and experience from other countries that have reopened classrooms.

"We should focus on other behaviors where we know the virus are transmitted, and try to keep schools open based on known value of in-person schooling for children, particularly young children," she said.