Gather More Info As COVID-19 Testing Expands

In an opinion piece in Stat, two former Obama administration officials make a case that data collection and analysis needs to go hand-in-hand with expanded testing for COVID-19.

Farzad Mostashari and Ezekiel Emanuel say, for example, that laboratories should be required to submit information on everyone tested not just those who test positive and that there should be serological surveys of multiple communities. They also mention serological surveys of randomly selected people would be the way to assess the “real percentage” of those who are positive for COVID-19, not just the “individuals who are bringing themselves in for testing.”

Their piece was published in Stat yesterday.

Mostashari, MD, was the National Coordinator of Health Information Technology in the Obama administration for two and a half years. After leaving the federal government, he founded Aledade, a company that operates ACOs for primary care physicians. Emanuel was a senior health adviser during the Obama administration and is commonly credited with being one of the architects of the ACA. He is a professor at the University of Pennsylvania.

At a White House briefing today, Deborah Birx, MD, said that 370,000 tests had been done in this country, 220,000 of them in the last eight days.

Mostashari and Emanuel said there should be four types of information gathering and analysis as COVID-19 testing is expanded.

• As condition of approval of tests, private labs should be required to submit age, county of residence, and other basic information on every person tested not just those who test positive. Mostashari and Emanuel say having knowing the ratio of positive tests to those tested-the numerator and the denominator-is important to knowing whether the COVID-19 pandemic is getting worse or better.

• Serological surveys- aka blood tests - should be conducted from randomly selected people in defined populations, Mostashari and Emanuel argue. They say in the Stat opinion piece that the information collected could be a “key input to models trying to predict when herd immunity can begin to blunt the outbreak.”

• Sentinel testing of samples of patients that should be conducted that would compile data on clinical aspects of the disease and patient exposures.

• Analysis of information currently in hand could provide insight into, for exampole,  emergency department use and how much the increase in patients in recent days is truly related to COVID-19.