The New York Times gets it wrong on COVID-19

David Leonhardt has a Good Morning column in the 4/13/21 New York Times. It is titled “With Covid cases rising, more experts think the U.S. should start delaying second vaccine shots. The column focuses on the possible benefit of delaying second vaccine shots in order to increase the number of people who have been vaccinated at least once. I will not point out the many flaws in DL’s argument about this. Instead, I will focus on his being unable to resist taking a shot at the US record on cases and deaths vs. Britain’s – where second shots have been delayed. Here is what DL says.

"There is real-world evidence — from Britain — showing large benefits from maximizing the number of people who get one shot."

"Britain and the U.S. have now given a roughly similar number of total shots per capita. The difference is that Britain has deliberately delayed second shots, by up to 12 weeks. The results are impressive."

"Despite being the country where the B.1.1.7 variant was first detected, Britain now has the pandemic under better control than the U.S. does. Both cases and deaths have fallen more sharply, highlighting the power of a single vaccine dose. “The levels of antibodies after the first shot are sky-high,” Dr. Robert Wachter of the University of California, San Francisco, told me."

DL then presents the following chart as justification for his conclusion that Britain has done better than the US..

Note that the red line (Britain) crosses the blue line (US) in late February. Measured from then, Britain’s COVID deaths per capita decline faster than the US’s. On the other hand, the US deaths per capita never reached the far higher peak of Britain’s. Measured from that peak, Britain’s rate of decline in deaths per capita relative to the US is truly impressive.

Let me describe what would have been an even more impressive result – according to DL’s method of analysis. Suppose Britain had done nothing about COVID and it had spread rapidly, killing 90 % of the population by, say, September of 2020. That would have guaranteed a really high peak in deaths per capita ended by herd immunity with a subsequent really rapid decline in deaths per capita subsequent to the peak. The comparison between the US and Britain in deaths per capita after that would have favored Britain even more – there would have been hardly any deaths per capital in Britain.

What this example illustrates is the incredible inadequacy of DL’s analysis. Either he knew the flaws in his analysis or he didn’t. In either case, we can conclude that DL is not credible.

What should we make of the chart? What is more important, the death rate per capita at particular times or the total number of deaths per capita? Clearly, the latter. And the chart suggests no clear superiority for Britain based on total deaths per capita. For the period shown in the chart, the difference between the total deaths per capita in Britain and those in the US is the area between the red line and the blue line. Here, the higher peak of the red curve and the higher level of the red curve in the mid October to December 2020 offsets to a large extent – and perhaps completely – the other times when the red curve is moderately below the blue curve. In other words, a more sensible analysis shows no clear advantage for Britain and a possible advantage for the US.