Twenty months ago, in July 2022, I wrote a long essay sketching what I called the “pretty brutal” endemic future for Covid: probably about 100,000 deaths annually, at least for the next few years. The number was just a ballpark estimate, drawn from modeling by the epidemiologist Trevor Bedford. But as it is turning out, it looks to have been almost exactly right. The official count is that about 170,000 Americans died of Covid over those 20 months, at a pace that suggests at the end of year two, the total may have been just past 200,000.
Two hundred thousand deaths still counts as “pretty brutal,” and one of the many things the pandemic has taught us again and again is our capacity for normalization. But the burden of Covid is also obviously subsiding, and the shadow of the first-year emergency retreating even further. And as the fourth anniversary of the pandemic brings a new flurry of retrospectives, I find myself marveling not just about the many narratives we’re still getting wrong, but also about how many seemingly contradictory stories can be justified by the facts.
You can see this phenomenon most clearly when examining what was at the start of the pandemic its most immediate and perhaps its most heated question: How many people was it killing, and where? We often used that data to advance arguments about mitigation policy, but it also represented a more fundamental kind of information, around which anxiety and perspective might be calibrated.
In 2020, we often assessed that toll pretty crudely, using raw death counts, which invariably made the United States look like the world’s biggest pandemic failure. This produced one of the dominant morality tales of the pandemic’s first years: that the countries that should have expected to do best were in fact faring worst and that the United States under Donald Trump was the world’s most conspicuous example of pandemic mismanagement.
Occasionally, we would adjust those figures for population size, which yielded per capita measures that showed the country doing only somewhat worse than most rich nations considered American peers. By this measure, Britain became the poster child for rich-world failure — “Plague Island” it was often called in the British press.
But these statistics, though valuable records of the sheer scale of suffering and tragedy within nations, were also biased in two big ways when it came to international comparisons. First, they depended on how much Covid testing was being done: Countries with better disease surveillance tended to register more official Covid deaths while less aggressive places registered far fewer deaths from it. Second, the death counts were driven in part by the age structure of a country’s population, because Covid was so much deadlier for the old and especially very old than for the young and middle-aged. (In an immunologically naïve population facing the original pandemic strains, the infection fatality rate for nonagenarians was perhaps a thousand times higher than for their grandchildren.)
