This post was originally published on this site
New York City and New York state pull testing data from the same source, but process and analyze it slightly differently.
AFP via Getty Images
New York City Mayor Bill de Blasio said Friday that parents should prepare for the possibility that schools will be closed on Monday if COVID-19 infections surpassed a critical metric over the weekend—something he warned the city was “quite close to.”
But how close? The city’s Department of Health and Mental Hygiene reported Friday the positivity rate on a seven-day rolling average had climbed to 2.83%—indeed, just shy of the critical 3% threshold that would trigger schools to close. The state’s Department of Health, however, reported that same metric for the five boroughs at 2.3%, significantly less when even the slightest uptick could be decisive.
Any data nerds (ahem, like us) might be confused as to how, when so much is at stake, the city and state figure could diverge by more than half a percentage point.
Coronavirus update: U.S. again shatters daily case record as experts worry holidays will be ‘superspreader’ event
The answer is more than a little technical, but it all comes down to where the data is coming from and how it’s processed.
Both the city and state pull infection data from a state repository known as the Electronic Clinical Laboratory Reporting System, or ECLRS (pronounced like the French pastries), said Corinne Thompson, an epidemiologist with the city’s Department of Health and Mental Hygiene.
“If you’re a New York state resident, you’ll go, you’ll get a test, the swab will be sent to a laboratory,” she said. “Then the laboratory is legally required to report to that system, which is ECLRS.”
Once a day, the state sends the city testing data it received for patients in the five boroughs, according to Thompson and a spokesman for the state Health Department.
The next step is largely where the discrepancy arises. New York state classifies a case based on when the test result came in, whereas New York City classifies each case based on the date a person was tested in the doctor’s office. Same raw data, slightly different analyses.
“The reason we do that is because we think it’s most epidemiologically relevant because it’s most close in time to when someone actually started to feel sick,” Thompson said.
New York City also does a little extra data culling that could cause the city and state figures for the five boroughs to differ slightly. For instance, the city sometimes comes across a patient who actually lives in New Jersey but tested in New York City. That fix isn’t reported back to the state, she said.
There are also hiccups like twins—same birthday, same household, often similar names—which can trip the algorithmic data processing and require manual clean up. But that fine-tuning is largely insignificant when it comes to the final metrics.
So whose number is more important?
When it comes to school closings, sadly for anxious public school parents, it’s the seven-day rolling positivity rate as reported by the city, i.e. 2.83%.
Were the city’s dedicated data collectors to tabulate that figure at 3% or higher this weekend, all public schools will be closed on Monday, excluding community-based pre- and 3-K providers, the mayor told radio talk show host Brian Lehrer on Friday.
“This is not something any parent, you know, wants to have to deal with, but we should get ready and parents should have a plan for the rest of the month of November,” he said.
Gov. Andrew Cuomo, who has overruled de Blasio on school decisions in the past, affirmed in his own public briefing that the mayor would be within his power to shutter schools — apparently, even if the state’s metric remains below 3%.
“That is the agreement that New York City announced,” Cuomo said on Friday. “I would hope that the mayor and the teachers and the parents work together to reopen the schools as quickly as possible.”
