The answer usually given is that we must “flatten the curve.” The pandemic curve, a roughly bell-shaped curve, shows the number of new infections rising and then falling over time. With a highly transmissible virus the curve will have a high peak, with a steep ascent and steep descent on either side. If measures are taken to limit transmission the curve will be flattened—the slope will be more moderate, and the peak will be lower—but the process will play out over a longer period of time and the total number of infections will not necessarily be any different. And, barring a vaccine, once measures are relaxed the number of infections will rise again because herd immunity never had a chance to develop and there are still enough susceptible people to fuel transmission.
The pandemic curve translates into a similar curve of how many people will, after being infected, become seriously ill and require hospitalization and intensive care unit (ICU) care, which could include ventilators. Health care planners were afraid that with a steep curve there would be large numbers of patients needing ICU care and ventilators all at the same time and that the health care system’s capacity would be overwhelmed. That would increase the number of deaths as both COVID-19 and non-COVID patients were unable to obtain care. They prescribed suppression measures including social distancing of the entire population and school closures. Delaying and lowering the utilization peak would give hospitals time to increase capacity and supplies and allow more time to develop treatments.
But the planners were wrong. Outside of a small number of hospitals in a few hotspots, U.S. health care capacity was not overwhelmed. In fact, many hospitals, forced to suspend elective procedures, have plenty of empty beds. The additional bed capacity that New York built went largely unused. This underutilized capacity is not proof that a lockdown is superior to less aggressive mitigation measures. Despite widespread criticism that Britain waited too long to impose suppression measures, its poorly resourced health system was not overwhelmed. Sweden, which has been roundly criticized for not imposing society-wide social distancing and school and business closures, has never come close to having a shortage of ICU beds.
From the start, planners misjudged the height of the hospitalization/ICU curve because they overestimated how deadly infection with the virus is. They were undoubtedly influenced by scenes of overwhelmed Italian hospitals and reports of doctors forced into tragic choices as to who would receive care and who would die. But the Italian experience was not applicable here.
COVID-19 is especially lethal among older people—95 percent of COVID-19 deaths in Europe and 85 percent in New York were in people 60 and older; only 2 percent of New York deaths are in people under 40. Older COVID patients are also far more likely to have severe cases needing hospitalization. COVID-19 hospitalization rates for people over 65 are double the rates for ages 50-64 and six times the rates under 50. Hospitalizations below age 18 are rare. Italy’s population is disproportionately elderly. Twenty-three percent of Italy is 65 or older. The comparable U.S. figure is 16 percent. And the U.S. was better prepared. While the U.S. has roughly the same number of hospital beds per capita as Italy, we have nearly three times as many ICU beds per 100,000 population as Italy has. The U.S. has more than twice as many modern ventilators per capita as Italy and the number of U.S. ventilators could more than double if older, less full featured units are pressed into service. Finally, Italian cases rose several weeks before the U.S., so the U.S. had time to institute mitigation measures at an earlier phase of the pandemic.
Planners were also concerned that COVID-19 could be as severe as the 1918 influenza pandemic, where 50 million people died worldwide including 675,000 in the U.S., because both pandemics involve a new virus for which the population had little or no immunity. But the 1918 pandemic had a unique mortality distribution not seen before or since. While seasonal influenza typically has mortality peaks among the very young and very old, the 1918 pandemic was especially lethal among young adults—nearly half the total deaths were in the 20-40 age group. The absolute risk of death was higher in those younger than 65 than in those who were older. In contrast, COVID-19 deaths are heavily concentrated among the elderly.
Finally, planners were influenced by high initial fatality estimates—on March 3 the World Health Organization (WHO) reported a fatality rate of 3.4 percent. But that estimate was artificially high because it calculated rates based on deaths (the numerator) among known cases (the denominator), which were the more severe cases. Asymptomatic and mild cases went uncounted. The WHO report also erroneously reported that the number of asymptomatic infections was only 1 percent.
Some epidemiologists suggested the true number of infected people in the U.S. is far higher than the number of confirmed cases. One study of excess influenza like illnesses reported by the Centers for Disease Control and Prevention suggests that a high percentage (87 percent) of symptomatic COVID-19 cases were not identified because they were never formally tested. In addition, reports from California, New York, and Europe found that most COVID-19 infections are asymptomatic and go undetected. A report from prisons in four states shows that large numbers of prisoners are testing positive and 96 percent are asymptomatic.
If the true numbers of symptomatic and asymptomatic COVID-19 infections are grossly undercounted, it would lead to two encouraging conclusions: First, the fatality rate is far lower than feared. Second, there is probably a large repository of people who were infected, have recovered, and now have antibodies making them immune to future infection. A study of confirmed COVID-19 cases in New York showed that they reliably develop antibodies, likely making them immune. Another study of random samples in New York City found more than 20 percent of people test positive for antibodies. This indicates we may be closer to the herd immunity that usually signals the end of pandemics than has been appreciated.
Good serology studies are imperative to determine how prevalent antibodies are in the population. The Food and Drug Administration has given emergency approval to new, highly reliable tests that should facilitate the process. If infection rates are high, lockdown measures should be relaxed.
This does not mean ending all measures. Public health measures to combat COVID-19 are a continuum. Applying just three mitigation interventions—home isolation of known cases, voluntary home quarantine of household contacts of cases, and social distancing of those over 70—was predicted to lower peak health care demand by two thirds and deaths by half. A recent MIT study found that targeting mitigation policies to the most vulnerable, elderly population would actually lower fatalities and economic costs, compared to measures applied uniformly to the entire population. Continuing school closures and social distancing of the entire population makes little sense, since younger people have a near-zero risk of serious complications and death from COVID-19. Allowing the young to be in environments where they risk exposure to infection is possible if we continue protection measures for the most vulnerable—the elderly and those with underlying medical conditions.
Testing, targeting our efforts at protecting the most vulnerable, and relaxing public health measures as local circumstances dictate, will be an ongoing process. This course is far preferable to another year of lockdown while we await a vaccine that may or may not come.