Posts Tagged curve fitting
China has run the gauntlet with the novel Coronavirus. They do not report any new cases now and the population is cautiously and gradually returning to normal activity. China has reported on the number of cases and deaths despite very difficult times. One problem they have had as will other countries is how to report deaths that did not happen in a hospital or without testing to verify the presence of COVID-19.
The main question currently in the United states is when will the population be free from restrictions to stay at home? As mentioned in the previous post the experience from china gives some clues. The following are two graphs; the first shows an overview of number of new cases over time (left axis) and the number of deaths (right axis). The data becomes erratic at the peak of the curve when the hospitals and healthcare workers were overwhelmed. The bell-shaped red is a fairly good fit to the number of new cases. The yellow curve is from a similar effort at fitting a curve (see below) to the numbers of deaths per day.
The disease started sometime at the end of December 2019 and spread. Millions of people were confined to their residences finally leading to no new cases about 2 months later. The peak of cases happened around February 8 while the peak of deaths was about 1 week later. The deaths subsided about 2 weeks after the peak, marking continued hospital overload and a reservoir of virus that could escape and threaten the population again.
China has been criticized for the reporting of deaths, and the irregularities are easy to see around the peak of deaths probably related to a saturation of the health care system. Patients likely died at home since admission to hospitals was not possible and furthermore an exact diagnosis based on PCR testing was lacking. Nevertheless, the picture is sufficiently clear to be a guide for other countries regarding what to expect.
In a region such a one of the States in the US several milestones should be reached before social distancing is eased:
- No new cases in a State
- Borders to the State remain closed if adjacent states still have active cases (people could leave states but would likely need 2 weeks of quarantine if they return).
- Hospitals only have a few remaining cases of COVID-19 patients and have the staffing to receive emergencies if needed (this could take about 2 weeks from the last new case).
- The State Health department has the capacity to rapidly trace any new cases and institute strict quarantine.
- Adequate and rapid viral testing is available to the health department and physicians.
Social distancing allowed many persons to avoid infection, associated illness and death. A resurgence of COVID-19 is easily possible in the populations of uninfected individuals.
Healthcare workers and any other workers who have survived the illness could return to work at any time, even now. They are at very low risk due to the immunity developed to clear the virus from their bodies. At some point, a test for immune status based on antibodies present in the blood would be helpful because a large number of persons were probably infected and not diagnosed at the time with PCR testing (the mouth or nose swab). Persons with adequate antibodies also could safely return to work now.
So what does the crystal ball suggest: many people will be returning to work toward the end of April, 2020. There will be states that lag because the virus started in those states later — that’s going to be a difficult pill to swallow. Predictions about what will happen are subject to lots of unknowns. With time, the end-game will be more clear.