Ever since Covid-19 first began to spread, the question of how many people are infected with the disease has remained difficult to answer. Official case counts, it is widely acknowledged, represent only the tip of the iceberg—a gap in our knowledge even more felt today, now that the virus is ten times as transmissible as earlier strains.
Just how widespread is infection actually? In Spain, a broad survey of the country’s population found that about 5 percent—that is, 2.3 million people—were infected. Prevalence varied according to occupation (health workers, 10 percent) and age (children, 3 to 4 percent). The actual rate of infection, the study suggests, was about ten times the roughly 250,000 cases picked up by standard genome tests.
Two JAMA articles, both published this month, report seroprevalence data that tells a similar story—only this time, the setting is 10 different locations in the United States. One article is a scientific-medical manuscript, the other a commentary. While the rate of infection varied widely from site to site, on average it was found to be about 10 percent. Since the population of the United States is about 330 million, a 10 percent infection rate would translate to a whopping 33 million infected people.
As of today, the number of people who have tested positive in the United States is just shy of 4 million. This means that the actual rate of infection, as was the case in Spain, may be 10 times higher than originally reported. If that many more people are infected—and cable of infecting others—than previously thought, the implications are vast. One being that people who are infected but who don’t develop symptoms, no matter their age, are spreading the virus further and faster than we think.
People of all ages who are asymptomatic, from young children to those 80 and older, are well capable of transmitting the virus to others, who may fall seriously ill or die. This observation reinforces the advice I often give to others: that you should assume everyone outside your personal “bubble” may be infected and act accordingly. Certainly it justifies the continuance of broad social distancing measures and restrictions on social and work activity.
An infection rate of 10 percent would also present a challenge to those who are still holding out hope for so-called herd immunity, which requires prevalence of 60 to 70 percent to be achievable. As the authors of the JAMA commentary point out, the rates of 5 to 10 percent observed in Spain and the United States fall significantly short of this mark.
The articles also tell us that infection rates vary not just between cities, but within cities as well. Perhaps unsurprisingly, given all the data we have on connections between social inequality and vulnerability to Covid-19, intra-city variation correlates with income and occupation. The lower the income level, the higher the infection rate. Those with lower incomes often have no choice but to continue working service jobs, exposing them to risks that those with higher incomes may avoid.
As long as infection persists in any community, SARS-CoV-2 will continue to pose risks to the entire population. In Singapore, where case counts were so low for so long, disease outbreaks in neglected guest worker dorms did much to bring back the dangers of Covid-19 to all.
To put a stop to this pandemic, we need to intensify our focus on reducing infections in marginalized and low-income populations. Whatever the number of undetected Covid-19 cases, these are the people the disease will hit hardest.