June 24, 2020: Covid spiked?

A recent Accuweather interview of Penn State professor, using JHU data, report that 117,000 Covid deaths moves the pandemic to second place behind Spanish flu epidemic, passing the 1968-70 flu. According to professor, Spanish flu killed about 685,000 Americans, infected 500 million people (a third of the world population) and killed 50 million people globally. He said technology difference today is better, more rapid communication.

But there is much more to the story than these data points. You must have perspective.

· USA and global population one hundred years ago was ¼ to 1/5 what it is today. So for a useful comparison, reduce the JHU count by 80%, to about 25,000, to compare it to the Spanish flu death rate.

· The number of USA elderly population (over 65) was less than ten percent of what it is today, and Covid is especially lethal to the obese, sick elderly, few of whom existed in 1918; life expectancy was about 50 years. Nursing homes as we know them today did not exist. The annual mortality rate for those over seventy in the United States is about 14%, higher than the mortality rate for those over seventy who contract the virus.

· Need to question deaths assigned as Covid when often there are co-existing conditions and no positive test for the virus, pre or post-mortem. States, governors and hospitals have a strong economic incentive to classify a death as Covid related, especially as the federal government pays the hospital a large fee for a reported Covid death.

· China did not communicate the problem, making it worse and gainsaying the tech communications advantage in prevention. The USA was very well prepared to cope, with a strong internet, great cell phone network, nascent work from home practices, and growing online education. The problem was that China kept the important information about the virus a secret, while allowing millions of people to flee Wuhan, many of them heading for what came to be the immediate flashpoint, namely the USA and Italy.

· No vaccines for the viruses in 1918-1920. And no lockdown in 1968-69 despite comparable number of deaths among a population half the size of the US population today.

· Science and knowledge of virus has increased, how to treat it, transmit it, stop its transmission, plus alternate work from home strategies using technology flattens the curve. Ventilators in New York were stored but had fallen into such disrepair as to be unusable. An auto plant converted to making ventilators, then doctors learned that ventilators were not very effective anyway. Meanwhile, New York screamed for aid, only to not even use the 2,500 beds in the Javits Center or the 1,00 beds on the US Navy hospital ship dispatched to the New York harbor. Worse, the NY governor instead sent sick, hospitalized nursing home patients back to their nursing home, close to a death sentence for many of the elderly residents. I have one, very reliable first hand account of a nursing home outside New York City where the staff refused to work, leaving care of 86 elderly to four remaining employees. Thirty-eight of the residents died. Their remains required a refrigerator truck to handle the bodies.

Bottom line is that one needs to see hard data of clear measures analyzed using solid statistical methods. The consumer has to understand mean, median and mode, she must understand the difference between correlation and causality. A forecast without clear explanation of the standard error in the forecast is close to useless.

One must also consider the source and the platform for the message. Journalists know little or nothing about statistics and they often rely on a single 'expert' to verify the journalist's personal bias or theory. Their reports often lack historical or even concurrent perspective. They use terms such as 'spike' or 'record' (as in 'hits a new record') way too casually. What exactly does a spike mean? Has an already high baseline increased by 200%, or might it be a very low baseline increasing by 50%, when the baseline number was two. 'Spike' is a meaningless yet highly charged term of art among lazy journalists, editors and headline writers. 'Spiked' used to mean that the story was not even published.