Monday, June 15, 2020

Parameters involved in Covid19 forecasts

**New York Governor Cuomo said:

So we are the exact opposite. We, since we've reopened, the number has continued to go down, believe it or not. We reopened. It continues to go down because we've been disciplined in our reopening, and that's what we have to continue to do. This is a website where the founders of Instagram now track the rate of transmission of states across the nation. New York State, the lowest rate of transmission, meaning the virus is spreading at the lowest rate in the State of New York, of every state in America, that is incredible.



These are up-to-date values for Rt, a key measure of how fast the virus is growing. It’s the average number of people who become infected by an infectious person. If Rt is above 1.0, the virus will spread quickly. When Rt is below 1.0, the virus will stop spreading.

The metric being tracked here (Rt) represents the effective reproduction rate of the virus calculated for each locale. It lets us estimate how many secondary infections are likely to occur from a single infection in a specific area. Values over 1.0 mean we should expect more cases in that area, values under 1.0 mean we should expect fewer

The new version of this model accounts for variation in serial interval and delay between onset of symptoms and a positive test. Because of these changes, it is also more robust to large changes in reported tests. However, this also means that Rt will be far less variable day-to-day than the previous model. [This sounds like Ro.]

There is an incubation period in which people are likely infectious but not symptomatic. This model assumes infectiousness begins with symptoms. While future versions may correct for this, a simple heuristic is to shift all values of Rt 5 days into the past.
While we make a best effort to accurately describe Rt, nothing can do that perfectly.


One cannot be sure "how qualified" the people at Instagram are to assess Rt, which does sound like Ro. Their Rt is clearly NOT a rate; it is dimensionless. If you believe this stuff, you should be worried, because Hawaii is at 0.99, not like a safe state like NY or NJ, which are the two "best" states (lowest Rt)

As usual, the data relied upon is tests of symptomatic people. Although they make a nod to pre-symptomatic people, they ignore asymptomatic people (who never get symptoms). If one is worried about the kinetics of spreading, this is incomplete, because it ignores: asymptomatic to symptomatic, symptomatic to asymptomatic and asymptomatic to asymptomatic.

At this point in time, this is not a viable approach.

Recent work by Aravinda de Silva of UNC shows how we can fine tune analysis of "asymptomatic" individuals.

From Genetic and Engineering News:

The antibody test is based on the receptor-binding domain (RBD) of the spike protein. The RBD-based antibody test measures the levels of that domain, which the authors found correlate to the levels of neutralizing antibodies related to the paper published in Science Immunology titled “The receptor-binding domain of the viral spike protein is an immunodominant and highly specific target of antibodies in SARS-CoV-2 patients.”

[The importance resides in specificity to Covid19]

The RBD of the spike protein in SARS-CoV-2 is not shared among other known human or animal coronaviruses. Therefore, antibodies against this domain are likely to be highly specific to SARS-CoV-2, and so these antibodies reveal if an individual has been exposed to the virus that can cause COVID-19.

[Of data]

By day 9 after the onset of symptoms, the authors wrote that “the recombinant SARS-CoV-2 RBD antigen was highly sensitive (98%) and specific (100%) for antibodies induced by SARS-CoVs.” They noted that they observed a strong correlation “between levels of RBD binding antibodies and SARS-CoV-2 neutralizing antibodies in patients.”

[Of the relevance to asymptomatic people]

“We don’t see our research as a means to replace commercial tests,” said de Silva. “Commercial tests are critical, especially for making decisions about the clinical management of individual patients. But it’s too early in the pandemic to know if the commercial assays are suitable for identifying people who experienced very mild or no disease after infection or if the assays tell us anything about protective immunity, as researchers are still learning about this virus.”


Post a Comment

<< Home