Virus models

The use of models to predict the course of the coronavirus is to some extent an exercise in circular reasoning. The models are never wrong, even though events depart drastically from what the models predict. That can always be explained by saying that the predictions reflect a range of probabilities or that mitigation measures have gone better or worse than expected.

Even though the models are shaky, the scientists behind the models are happy to announce their predictions to the world.

The attraction of being the center of attention is very strong. Predicting a vast number of deaths increases the attention.

The New York City Department of Health publishes a daily data summary of cumulative deaths allegedly from the coronavirus. Of 1,145 deaths to April 2, only 24 patients did not have underlying conditions such as diabetes, lung disease, cancer, immunodeficiency, heart disease, hypertension, asthma, kidney disease and Gi/liver disease. Further, only one patient was younger than 18, 64 were younger than 45 and 68 percent of the patients were older than 65. There are about 55,000 deaths in New York City every year, or 4,500 a month.

Apparently, the city counts every death as a coronavirus death if the deceased tested positive. Certainly this is wrong. If an elderly person dies from a heart attack, it is not a coronavirus death just because he tests positive. But there might be a financial advantage (federal payments) in classifying as many deaths as possible from coronavirus.