Jeong on how to interpret daily covid-19 case counts


TRIB LIVE -  As coronavirus cases mount in Allegheny and surrounding counties, questions often arise about how the data should be interpreted. Viewed without context, the numbers can lead to some misconceptions.

Allegheny County health officials reported 158 new cases Thursday, for instance, along with 12 hospitalizations — but those infections didn’t happen overnight.

“Some people get reported at a later time and there’s a lag,” said Dr. Jong Jeong, professor and interim chair of the Department of Biostatistics at the University of Pittsburgh Graduate School of Public Health. “We don’t know exactly what happened today. Some reports can come in that happened a few weeks ago.”

The positive cases reported by Allegheny County on Thursday, for example, are from tests administered between June 25 and July 7. Because test results can take weeks to return, or as quickly as a few days, it’s difficult for experts to know exactly what’s happening on any specific date. Day by day, the snapshot changes, Jeong said.

Reports of deaths and hospitalizations are even more complicated. Of the 36 deaths reported in Pennsylvania on Thursday, 10 occurred in April, May and June.

Information about the exact dates Thursday’s hospitalizations occurred could not be found, but county health officials have reiterated in the past that hospitalization data often trails behind daily case counts. It can take up to several weeks for an infected individual to get tested, get the results and become sick enough to be hospitalized.

Because of the reporting delays, experts and officials likely won’t know the most accurate record of the virus until it’s already passed, Jeong said.

The number of tests administered each day also fluctuates, which can affect the raw number of positive cases detected. In recent weeks, President Trump has said increased testing capacity explains rapid spike in cases across the United States — a claim health experts have called misleading, as the rate of infections far exceeds the proportionate number of tests.

For more clarity, Jeong said it’s important to look at the percentage of positive cases out of tests administered, to see if the number of infections is disproportionate.

Read full story


Search for an Article