Erin Schaff / The New York Times
A medical team tends to a COVID-19 patient at Houston Methodist Hospital in Houston, July 5, 2020.
Friday, Oct. 30, 2020 | 2 a.m.
The coronavirus struck the United States earlier this year with devastating force. In April, it killed more than 10,000 people in New York City. By early May, nearly 50,000 nursing home residents and their caregivers across the country had died.
But as the virus continued its rampage over the summer and fall, infecting nearly 8.5 million Americans, survival rates, even of seriously ill patients, appeared to be improving. At one New York hospital system where 30% of coronavirus patients died in March, the death rate had dropped to 3% by the end of June.
Doctors in England observed a similar trend.
“In late March, 4 in 10 people in intensive care were dying. By the end of June, survival was over 80%,” said John M. Dennis, a University of Exeter Medical School researcher who is first author of a paper about improved survival rates in Britain, accepted for publication in the journal Critical Care Medicine. “It was really quite dramatic.”
Although the virus has been changing slowly as it spreads, and some have speculated that it has become more easily transmissible, most scientists say there is no solid evidence that it has become either less virulent, or more virulent. As elderly people sheltered inside and took precautions to avoid infection, however, more of the hospitalized patients were younger adults, who were generally healthier and more resilient. By the end of August, the average patient was younger than 40.
Were the lower death rates simply a function of the demographic changes, or a reflection of real progress and medical advances in treatment that blunted the impact of the new pathogen?
Researchers at NYU Langone Health who zeroed in on this question, analyzing the outcomes of more than 5,000 patients hospitalized at the system's three hospitals from March through August, concluded that the improvement was real, not just a function of changing demographics. Even when they controlled for differences in the patients’ age, sex, race, underlying health problems and severity of COVID symptoms — like blood oxygen levels at admission — they found that death rates had dropped significantly, to 7.6% in August, down from 25.6% in March.
“This is still a high death rate, much higher than we see for flu or other respiratory diseases,” said Dr. Leora Horwitz, director of NYU Langone’s Center for Healthcare Innovation & Delivery Science and senior author of the paper in Journal of Hospital Medicine. “I don’t want to pretend this is benign. But it definitely is something that has given me hope.”
Other doctors agreed.
“The mortality rates are way lower now,” said Dr. Robert A. Phillips, chief physician executive at Houston Methodist and author of a research letter in JAMA that compared the first and second surges of COVID-19 patients in Houston. But he emphasized that the disease remains “not only deadly — 10 times more deadly probably than a bad influenza — but it also has long-term complications. You don’t have that from the flu.”
While the studies evaluated the death rate, they did not assess the burden of what Phillips called “post-COVID syndrome,” which leaves many patients with lasting respiratory and neurologic problems, cardiac complications, and other lingering issues.
“It’s relatively easy to measure death, but that doesn’t capture all the other health issues,” said Dr. Preeti Malani, an infectious disease expert at the University of Michigan.
Many hospital patients face grueling and protracted recoveries and may require long-term care, while even those who had mild bouts of disease are often left with continuing health problems, like headaches, chronic fatigue or cognitive problems.
“It will take a long time to understand the full clinical spectrum of this disease,” Malani said.
And even as the rates of death decline, the raw numbers of deaths are expected to rise, driven by the increasing cases across the country. According to combined modeling forecasts cited by the Centers for Disease Control and Prevention, the number of weekly deaths in the United States could surpass 6,000 by Nov. 7, and cumulative deaths could reach 250,000 by Nov. 21.
The study from England analyzed the outcomes of 14,958 critical care and intensive care patients hospitalized throughout England from March 1 to May 30. Even after adjusting for differences in age, sex, ethnicity and underlying health conditions, the authors concluded that survival improved by about 10% each week after the end of March for patients in critical care and high-intensity care units (the English study did not adjust for severity of COVID-19 illness at admission).
A combination of factors contributed to the improved outcomes of hospital patients, the authors of the two studies and other experts said. As clinicians learned how to treat the disease, incorporating the use of steroid drugs and non-drug interventions, they were better able to manage it.
The researchers also credited heightened community awareness, and patients seeking care earlier in the course of their illness. Outcomes may also have improved as the load on hospitals lightened and there was less pressure on the medical staff, both of which had been overwhelmed by a surge of patients in the spring.
“We don’t have a magic bullet cure, but we have a lot, a lot of little things, that add up,” Horwitz said. “We understand better when people need to be on ventilators and when they don’t, and what complications to watch for, like blood clots and kidney failure. We understand how to watch for oxygen levels even before patients are in the hospital, so we can bring them in earlier. And of course, we understand that steroids are helpful, and possibly some other medications.”
