In 2009, the National Academy of Sciences was worried hospitals weren’t ready for a crisis. At the time, the subject seemed urgent. It was just after the spring of H1N1 influenza, and experts expected the virus to return again in fall. The organization published a report that year, and several others over the years to follow. The key takeaway: Good crisis care makes decisions based on the health of the community, not just the health of the individual. The U.S. health care system was not used to that kind of thinking. Making that shift would be difficult, but possible — provided there were clear guidelines in place, shared by hospitals across the country, and visible to the public.
But more than a decade later, despite a lot of progress, experts I spoke to said there are still no nationwide crisis regulations or standards in place. State and local plans vary in scope and quality. That means there are no uniform guidelines that decide who gets access to a scarce resource — like a test for novel coronavirus — and who doesn’t. There’s no consistent appeals process for those rejected. And we still have a health care system where nearly every resource, from basics like beds to life-saving tools like ventilators, could very quickly become scarce.
That’s the dilemma the country faces: While the individual risk might be low, the collective risk is almost unfathomable.
In movies, in novels, in the back of our own lizard brains, we’re used to thinking about risk as individuals. Viral pandemics bring up a simple question: “Is this thing going to kill me?”
In that context, the statistics of COVID-19 don’t necessarily sound horrific. You can go look at charts that clearly show that the majority of infections are mild, the bulk of people recover, and people 60 and over are most at risk. Unfortunately, while that information is accurate, it’s also misleading. That’s because the real risk of COVID-19 isn’t about what it does to one person, experts told me, it’s about the community. It’s not about “Will I die?” it’s about “How much will this overwhelm our health care infrastructure?”
And COVID-19 could certainly do that. The devil is in the details. Yes, a study of 44,000 confirmed cases in mainland China showed that 81 percent of infected people had mild cases of the disease. But that leaves 19 percent who suffered from severe pneumonia or worse — situations that are likely to require hospitalization and breathing assistance. That could equate to hundreds of thousands of people. Maybe millions, depending on what the rate of infection turns out to be.
And we don’t have that kind of capacity in our hospitals, said Brendan Carr, chair of emergency medicine at Mount Sinai Health System in New York. Across the country, hospitals have very little money to spare — the median operating margin was 2.1 percent for nonprofit hospitals in 2018, and it’s been going down for years. Between 2015 and 2017 alone, margins fell by 34 percent.
And that has forced hospitals to focus heavily on efficiency. You want every bed filled, every nurse busy, every drug used up quickly and not sitting around for months in a storage cupboard, Carr said. And that system makes sense … until it doesn’t. If demand suddenly and unexpectedly goes up, there’s not a lot of excess room to operate within. “You can’t have extra capacity and efficiency at the same time,” he said. “In fact, for years, we’ve called that waste.”
It’s hard to get numbers for how many hospitals don’t have enough extra capacity for emergencies because there’s no uniform standard for what “good capacity” should look like, said John Hick, medical director for emergency preparedness at Hennepin Healthcare in Minneapolis. He’s also part of that National Academy of Sciences committee that’s been working on crisis care guidelines since the H1N1 flu outbreak. In some parts of the country, it’s normal for emergency patients to wind up on gurneys in hallways because there aren’t enough available rooms — Carr said it happens somewhat regularly in New York. “In other places that would be out of the question,” Hick said.
But even though it’s hard to get numbers for potential shortages nationwide, Hick was able to give me some numbers from his own health system to offer some context for just how efficiently — or dangerously close to the line, depending on your perspective — hospitals operate. “All winter, we’ve been running with less than 5 percent of our [intensive care unit] beds available,” he told me. “Last Tuesday, across 32 hospitals, we had 21 beds available. And that’s with no COVID at all.”
Let’s do some fuzzy math. The population of Hennepin County is 1.25 million. Experts have said somewhere between 20 to 60 percent of Americans are likely to contract COVID-19. That won’t be evenly distributed across the country. But for the purposes of this thought experiment, let’s say 20 percent of Hennepin County residents get COVID-19 and, as in the Chinese study, 19 percent are severe or critical cases and would require hospitalization. That’s 47,000 people. On a given day, Hennepin Healthcare’s hospital network might have 21 empty beds. The Hennepin Healthcare network doesn’t include all the hospitals in Hennepin County, but it’s unlikely that other facilities have way more extra beds sitting around.
Suddenly, the risks of all those cases coming at once becomes starkly clear. When public health types talk about “flattening the curve,” they’re talking about both shrinking the number of infections and spreading those infections out over a longer period of time, so hospitals aren’t left drinking from the firehose. Doctors want you to stay home while you’re sick — even if you’ve just got a mild illness — because that helps slow the spread of the virus. Ditto measures like increased handwashing, canceling big events and bumping elbows instead of shaking hands. The risk is collective, not individual.
If you want to see what happens when collective risk becomes medical reality, all you have to do is look to Italy. There, doctors are deciding, based on patients’ age and pre-existing conditions, who gets access to limited facilities and who gets sent home. Who gets to use a ventilator and who must try to survive without it. And while they care for COVID-19 patients, regular patients get shoved to the side. Where a heart attack is usually processed within minutes, people now may have to wait an hour or longer, Dr. Christian Salaroli, an anesthesiologist at the Pope John XXIII Hospital in Bergamo, Italy, told the Italian paper Corriere Della Sera. “The system cannot bear the ordinary and the extraordinary at the same time,” he said.
And Italy may have had more capacity to begin with than the United States has now. The latest data is from 2017, but at that point, the country had a usage rate of only about 48 percent for its intensive care beds. It’s hard to pin down an exact number for the U.S. — hospital participation in the surveys that count bed occupancy varies a lot from year to year — but some estimates place occupancy of ICU beds in this country at around 68 percent. A 2018 paper found that 35 percent of U.S. hospitals had an ICU bed occupancy rate above 71 percent.
It’s not just the number of available beds that matter in determining collective risk. The entire U.S. health care system has evolved to focus on what can be done to save the individual in front of you, Hick told me. And those are very different priorities — very different practices — from those you use to decide how to save the most people with the least resources. The last time a serious respiratory pandemic swept the country was 1968. Think about the amazing technologies we have available today. Mechanical ventilators that can keep people breathing when their lungs are damaged. Artificial lung machines that can temporarily take over the whole job if a patient’s own lungs can’t function at all. Those are remarkable developments. But the way we’ve distributed and used them is designed to keep a few very sick patients alive longer — not deal with a pandemic.
We don’t know how many ventilators there are in the U.S. right now, said Dan Hanfling, vice president on the technical staff at In-Q-Tel, a nonprofit technology development organization that grew out of the CIA. He’s also a practicing emergency physician at Inova Fairfax Hospital in Virginia and another researcher who has been part of that National Academy of Sciences crisis guidelines committee. The best estimate available comes from a 2010 survey that estimated the number of full-function ventilators in the country at around 62,000. In addition to that, Hanfling said, more machines are held in reserve in the Strategic National Stockpile — a collection of secret warehouses that contain supplies meant to help the country cope with a biological weapon attack, nuclear bombing or other localized large-scale emergency. That number of ventilators in that cache has not been made public, but Hanfling said he believes there are at least 10,000 there, and probably more, because of the efforts of former Centers for Disease Control director Tom Frieden. Hanfling’s best guess: We’ve got about 100,000 ventilators. For the whole country.
There are even fewer external lung machines, known as ECMOs. “We’ve got over 400 ICU beds in the [Minneapolis] metro area, but I can put 26 people on ECMO,” Hick said. “That’s it.”
One of the problems with not having good crisis care plans in place is that no plan means there are no good systems for making sure those few machines go to the people who actually need them. If the rules for who gets access to a machine aren’t the same everywhere, people might commute to hospitals with less-stringent standards — or even just rumors of less-stringent standards — which might flood some hospitals. On top of that, cases of a viral outbreak aren’t distributed evenly across the country. Some cities will be hit harder than others — but might not have access to unused tools sitting around at a neighboring facility. “You’re really at the risk of sowing social chaos,” Hanfling said. “Patients shouldn’t die for lack of resources if they’re available 20 miles up the road.”
Although the nature of collective risk means we’re all facing the dangers, Carr emphasized that the takeaway shouldn’t be doom. People are working behind the scenes to create backup plans, and backup plans to the backup plans. Some states, including Minnesota, have crisis care plans in place that stipulate who is making decisions at the hospital level, how those decisions get made, and how different hospitals work together. Even the efforts over the past few years to add extra ventilators to the strategic stockpile is a win.
But to reduce risk, we do have to start thinking about ourselves as part of a community, experts told me. And that means making choices we wouldn’t otherwise make, that aren’t personally convenient, and that may even be pretty uncomfortable. For example, you should continue self-isolating for illness even after you feel better because you could still be contagious. Or, Hick suggested, you may need to have a conversation with the older adults in your life to find out what they want from doctors, if they do end up in an overcrowded intensive care unit with COVID-19. Do they want to go on a ventilator at all? Would they give their ventilator to a younger person? Under what circumstances?
These aren’t hypotheticals or over-preparedness, he told me. “It’s happening right now in Italy. It happened in Wuhan [China]. It will happen in the U.S. if we don’t do social distancing really, really well,” he said. Once someone has a tube down their throat, it’s too late to know what choices they’d have made. “I’d rather have those hard choices revolve around patients’ choices,” Hick said, “rather than me deciding for you.”
Davide Mancino contributed reporting.