Scientists Warn CDC Testing Data Could Create Misleading Picture of Pandemic
Some epidemiologists are reacting unfavorably to news that the Centers for Disease Control and Prevention (CDC) pools results from two different types of coronavirus tests to track screening levels across the United States. While it says most of the data are derived from polymerase chain reaction (PCR) testing, which detect active infections, CDC acknowledges that it also incorporates data from serology testing, which flags antibodies in the blood as an indicator of previous infection but does not contribute real-time information on the number of new infections. Scientists argue that combining the test results could skew the national take on the pandemic by suggesting that testing of active cases was performed at a greater extent than actually occurred and also by artificially lowering the number of tests that appear to be generating positive results. CDC spokeswoman Kristen Nordlund said the agency, which gets information from multiple sources, hopes to break out the data between PCR and antibody screens over the coming weeks. Officials also are in the process of developing standardized criteria that they hope will minimize confusion over reporting requirements.
CDC Estimates That 35 Percent of Coronavirus Patients Don't Have Symptoms
Based on real-world data, the Centers for Disease Control and Prevention (CDC) estimates that 35 percent of people with COVID-19 are asymptomatic, that 40 percent of infections occur before they feel sick, and that 0.4 percent of patients with symptoms do not survive. The agency emphasizes that the numbers are not projections of the expected impact of the pandemic, nor do they reflect the impact of social distancing or other responses. Rather, it clarifies, the data will be used for planning purposes by mathematical modelers and public health officials viewing the pandemic under different potential conditions. In the best-case scenario, for example, the overall mortality rate for symptomatic coronavirus patients would reach only 0.2 percent. The worst-case scenario, however, raises that share to 1 percent. At least one modeling expert, Carl Bergstrom of the University of Washington, has already challenged the data. "While most of these numbers are reasonable, the mortality rates shade far too low," Bergstrom, a biologist, complained. "As I see it, the 'best estimate' is extremely optimistic, and the 'worst case' scenario is fairly optimistic even as a best estimate."
These Labs Rushed to Test for Coronavirus. They Had Few Takers.
New York Times
In the wake of California's stay-at-home order in March, the Innovative Genomics Institute at the University of California, Berkeley, led by Jennifer Doudna, converted the facility to test solely for COVID-19. The reconfigured lab was offering the service for free with support from philanthropies. However, there was little demand for the tests as many local hospitals and doctors' offices continued sending samples to national laboratory firms. It was difficult for health care providers to switch to a new lab because of bureaucratic hurdles. Fyodor Urnov, who supervised the transformation of the institute into a clinical laboratory, observes: "It's still amazing to me, like, how can that be the case, that there is not a more systematic way to address a central need?" Scott Becker, chief executive of the Association of Public Health Laboratories, notes there is little coordination among public and private health labs. Experts say that doctors who order coronavirus tests also often send out tests of other viruses, such as influenza, and many insurers require those tests to be done by in-network labs. The large companies said they accept the overflow from some public health laboratories. Meanwhile, many smaller laboratories are focusing on broad-based testing of high-risk groups such as health care workers and nursing home residents. Dr. Bob Kocher, a member of California's testing task force, believes the situation will change as the state reopens and more people will need to be tested.