If you're trying to figure out why some places have been hit especially hard by the COVID-19 pandemic while others so far seem to be largely unscathed, there is no shortage of hypotheses. But for each seemingly plausible explanation, there are counterexamples that complicate the story.
We know that the true number of infections in any given place is far larger than the number of confirmed cases, although exactly how much larger is a matter of much dispute. We also know that deaths are undercounted, especially if they happen at home and involve people with other serious medical conditions who are never tested for the virus. While the first gap is apt to be much bigger than the second gap, the size of both is likely to vary from one country to another, depending on testing rates and reporting practices.
Still, those differences by themselves cannot account for the striking international differences in COVID-19 deaths. "Even in places with abysmal record-keeping and broken health systems," The New York Times notes, "mass burials or hospitals turning away sick people by the thousands would be hard to miss, and a number of places are just not seeing them—at least not yet."
Stage of the Epidemic
Fewer than 50,000 COVID-19 cases and fewer than 2,000 deaths have been reported in the entire continent of Africa, which has a population of 1.3 billion. That amounts to fewer than two deaths per million people, compared to nearly 200 per million in Europe and a bit more than that in the United States—a huge difference, even allowing for underreporting.
Judging from the first confirmed cases, COVID-19 spread to Europe about three weeks before it hit Africa.* Although the death rate in Africa is bound to rise as the epidemic progresses there, the difference in timing cannot fully account for the enormous difference in fatalities per capita. And Japan, where the first COVID-19 case was reported on January 16, a week before the first confirmed cases in Europe, has a far lower COVID-19 death rate, although not as low as Africa's. South Korea, which reported its first COVID-19 case on January 20, likewise has a remarkably low death rate.
Since COVID-19 fatality rates are dramatically higher among the elderly, one obvious explanation is age demographics. The population of Africa, where the median age is about 20, is much younger than the population of Europe, where the median age is about 43.
As the Times notes, however, some countries with young populations, such as Iran (median age: 32) and Ecuador (median age: 28) are seeing relatively high numbers of COVID-19 deaths, although not as many per capita as Europe. And Japan, which has an even higher median age (48) than Europe, again provides a puzzling counterexample.
"In Thailand and India, where virus numbers are relatively low, people greet each other at a distance, with palms joined together as in prayer," the Times notes. "In Japan and South Korea, people bow, and long before the coronavirus arrived, they tended to wear face masks when feeling unwell."
Then again, the Times says, "there are notable exceptions to the cultural distancing theory. In many parts of the Middle East, such as Iraq and the Persian Gulf countries, men often embrace or shake hands on meeting, yet most are not getting sick." Iraq, which has a population half as big as neighboring Iran's but a substantially lower median age (21 vs. 32), has reported less than 2 percent as many COVID-19 deaths.
Since the COVID-19 virus does not seem to like heat and light, the Times notes, it makes sense that it has made relatively little progress in tropical countries such as Chad and Guyana but is more pervasive in places with more temperate climates, such as Italy and the United States. Yet "some of the worst outbreaks in the developing world have been in places like the Amazonas region of Brazil, as tropical a place as any."
It's no surprise that New York City, which has the highest population density by far of any city in the United States, has had many more COVID-19 deaths per capita than places where people live farther apart. Antibody tests conducted by the state health department in April suggested that more than a fifth of the city's population had been infected. Yet densely populated cities such as Hong Kong, Seoul, Tokyo, Bangkok, New Delhi, and Lagos have not seen anything like the cases and deaths reported in New York.
"Lockdowns, with bans on religious conclaves and spectator sporting events, clearly work," the Times declares, citing the World Health Organization. "More than a month after closing national borders, schools and most businesses, countries from Thailand to Jordan have seen new infections drop." Yet "counter-intuitively, some countries where authorities reacted late and with spotty enforcement of lockdowns appear to have been spared. Cambodia and Laos both had brief spates of infections when few social distancing measures were in place but neither has recorded a new case in about three weeks."
South Korea's response to COVID-19—focused on early and wide testing, coupled with aggressive contact tracing and targeted quarantines—has been notably different from the American response, which was crippled by a government-engineered test shortage and has featured broad business closures and stay-at-home orders. Government-mandated social distancing in South Korea has been significantly less strict and less sweeping.
Likewise in Japan, where schools were closed but there were no American-style lockdowns, although the government recommended precautions such as avoiding unnecessary outings. In Sweden, another country that has eschewed a broad lockdown, the COVID-19 death rate is higher than in other Scandinavian countries but lower than in such countries as Italy, Spain, France, Belgium, the Netherlands, and the U.K.
Some places may be seeing worse epidemics because they happened to have more "super-spreaders": carriers who infected an unusually large number of people at particular gatherings. The Times cites several examples, including the Diamond Princess cruise ship, a funeral in Albany, Georgia, and a church service in Daegu, South Korea.
New York City's epidemic seems to have been seeded by many international travelers, mainly from Europe. Other things being equal, places with fewer visitors can be expected to have fewer chains of transmission. That might be part of the explanation for the striking differences between New York and California, where the virus seems to have been spreading by mid-January (judging from a COVID-19 death, apparently via local transmission, on February 6 in Santa Clara County).
"Far-flung nations, such as some in the South Pacific and parts of sub-Saharan Africa, have not been as inundated with visitors bringing the virus with them," the Times notes. "Health experts in Africa cite limited travel from abroad as perhaps the main reason for the continent's relatively low infection rate."
*Update: A French doctor recently reported that a sample drawn from a hospital patient with respiratory symptoms on December 27 had tested positive for the COVID-19 virus. The former patient said he could not imagine how he was exposed, which suggests the virus was spreading in Europe earlier than public health officials originally thought.