Illustration by Sarah Ferone
Illustration by Sarah Ferone

The battle against COVID-19

Hospitalists from New York, New Orleans, and Phoenix share their pandemic stories.


The World Health Organization (WHO) declared COVID-19 a pandemic on March 11, a day that marked the beginning of a series of firsts.

“This is the first pandemic caused by a coronavirus. And we have never before seen a pandemic that can be controlled, at the same time,” said WHO Director-General Tedros Adhanom Ghebreyesus, PhD.

That same day, Montefiore Medical Center in the Bronx in New York City had its first patient test positive for the virus. Also on that day, the city confirmed its first COVID-19-associated death.

Dr Southern photo courtesy of Montefiore Medical Center
Dr. Southern, photo courtesy of Montefiore Medical Center

The uncertainty was humbling. “At that time, of course, we didn't know anything,” said William N. Southern, MD, MS, chief of the division of hospital medicine at Montefiore. “We didn't really know the most appropriate infection control procedures. We certainly didn't have any proven treatments. We didn't know how many patients we were going to get.”

After the first COVID-19 cases and deaths came the first statewide stay-at-home orders and business closures. Meanwhile, hospitalists were having their own kind of firsts: the first surge plans, the first treatment protocols, followed eventually by the first decline in new infections and, in Dr. Southern's case, the first day off in a month.

This is the story of how three hospital medicine programs rose to meet an unprecedented challenge.

***

In the two weeks that followed the first COVID-19 case at Montefiore, hospital medicine leaders looked across the world for guidance. During the epidemic in China, cases had doubled about every six days, so Dr. Southern expected to see a similar exponential curve.

Then a disturbing pattern emerged. “We were doubling approximately every day and a half, just under two days, and that was a huge surprise,” he said. “It's very hard to get your mind around that. . . . I've come to believe we're linear beings, it's hard to think exponentially; we think whatever happened yesterday is going to happen tomorrow. But it's not like that.”

To make matters worse, there were a large number of sick calls among the physicians, physician assistants, housestaff, and nurses, Dr. Southern said, “So we were losing staff while we were gaining size.”

Montefiore's hospital medicine division, which has 102 hospitalists and about 150 physician assistants, staffs the medical service at three core hospitals in the Bronx. Under normal circumstances, the service is “really, really full” when it has 700 to 750 patients, he said. But the projected demand for beds exceeded capacity by two-thirds.

“We were charged with the expansion of our service by about 500 beds, so we would somehow have to figure out both the space and the staffing to get to 1,200,” said Dr. Southern, who is also professor of medicine at the Albert Einstein College of Medicine in the Bronx.

As the hospitalists quickly became overloaded with patient care responsibilities, he had to come up with a plan largely independent of them—and fast. If service was going to expand by 500 beds, staffing would need to drastically increase.

To do this, Dr. Southern worked with the training program leadership to restructure the teaching service, creating three-person teams, each with an attending, a medical resident, and a nonmedical resident. The attendings were recruited from outpatient subspecialties that had gone quiet due to the pandemic, such as rheumatology and endocrinology. They generally had little or no inpatient experience, so they needed some direction.

“The linchpin of those teams was each one had to have at least one medical resident because they know everything,” said Dr. Southern. While this may sound like a joke—what resident knows everything?—he was completely serious. “The most important people in the hospital were the residents. . . . That group had extraordinary responsibility to ensure that the other members of the team knew how to log on to Epic and that they could manage the logistics of patient care.”

After creating a large number of these teams and giving them a census of between 10 and 12 patients each, Dr. Southern rapidly sent them into battle. For example, if a hospital executive called and said a gymnasium with 30 beds would open starting tomorrow at 7 a.m., he would activate a system dispatching three teams to care for those patients.

The additional medicine beds came from everywhere. COVID-19 care took over every inpatient service possible, such as the surgery and gynecological oncology wards, as well as a freestanding surgical center and many nonclinical spaces, from a large event hall to rehabilitation gyms to several conference rooms. Hospitalists, who served as “medical directors” of the new clinical spaces, quickly became the most sought-after hospital staff because of their inpatient care expertise, Dr. Southern said.

Meanwhile, admissions continued to balloon, causing concern about the unfathomable: demand for more than 1,200 beds. “There was a moment when . . . we were, by my calculations, about eight days away from having to put a sign on the emergency department saying, ‘Closed for business,’” he said, “because 1,200 was really hard to conceive of, but beyond 1,200, I couldn't even figure out what that would look like.”

The numbers fell short of those calculations . . . barely. Montefiore's capacity peaked at 1,194 beds, filled mostly by patients with the novel coronavirus. “We had almost no patients on the medical service who were not COVID positive. . . . It was—no exaggeration—virtually 100%,” Dr. Southern said.

***

A few weeks earlier and 1,000 miles south, there was a big party.

On Feb. 25, the longstanding New Orleans celebration of Mardi Gras went on as usual. Two weeks later, Ochsner Health announced its first case of COVID-19, said Steven B. Deitelzweig, MD, MMM, FACP, system chairman of hospital medicine for the health system, which has hospitals in Louisiana and Mississippi.

Dr Deitelzweig photo courtesy of Ochsner Health
Dr. Deitelzweig, photo courtesy of Ochsner Health

The city was the region's epicenter of COVID-19, which then spread to other areas, he said. “The timing of the initial case peak in New Orleans does reflect the Mardi Gras season, which is a two-week season. People didn't know to really have the social distancing and masks or anything else, so I think that's part of the reason, if not the main reason, why we were hit so hard so early in the pandemic.”

Case counts at the health system rose rapidly before peaking on April 7 with 973 COVID-positive patients and persons under investigation for the virus, said Dr. Deitelzweig, who is also professor of medicine at University of Queensland (Brisbane) and Ochsner Clinical School in New Orleans. Much like Montefiore, Ochsner Health stepped up its staffing to meet the high demand. To do this effectively, it used two military-inspired techniques: boot camps and daily situational reports.

The health system's staffing response included not just its 75 hospitalist physicians and 35 advanced practitioners, but other specialists who weren't seeing patients, such as podiatrists. To get these redeployed physicians up to speed with inpatient care, there was a “boot camp” of an online learning module and in-person lectures, said Dr. Deitelzweig, adding that in some cases, a hospitalist was partnered with a nonhospitalist due to the high patient volume.

With rapid changes on policies such as visitor restrictions, the biggest challenge was ensuring that information quickly reached all of these clinicians, he said. “We used military tactics to get communications out. . . . I have an administrative dyad for the hospital medicine service line, Kevin Green, who actually was a hurricane hunter who was in the Air Force for a good number of years,” he said. “So he shared during our huddles that the way the military would get things out across 20 campuses is [through] a daily situational report.”

The reports were sent up the organizational chain to leadership at higher levels. Each included a situational statement on expected staffing (both physicians and advanced practitioners), capacity, and the overall morale of the team. In addition, the reports outlined clinical and safety discussion points, equipment supply, and whether help was needed, said Dr. Deitelzweig. “We did this every day, and we would send it out to each of our campuses.”

The reports were kept in a book at each campus, both electronic and hard copy. Each day, the hospitalists would come in, check for any updates, and be able to hit the ground running, even if they were just starting their first week of service after seven days off. “I think that was very useful, it was creative, and we did it fast,” he said. “Usually we have to plan some of these things. There was no planning; we didn't have a hell of a lot of time. We thought through things, and then we got it launched.”

Yet another challenge was, and still is, family and patient communication. Since the medical students were not allowed to see COVID-19 patients but still wanted to be engaged, they served as family communication champions, Dr. Deitelzweig said. “We had them connect with the hospitalists and then connect with the patients and families . . . and scripted a couple of core messages to communicate. While we could use technology to do that with telemedicine, we also had that more personal touch.”

Finally, to increase capacity, Ochsner Health completed a dramatic expansion in June. The Federal Emergency Management Agency and the state of Louisiana funded the construction of three additional stories to the main campus, Ochsner Medical Center, a 767-bed hospital located just outside New Orleans. “Literally three floors, right on top,” Dr. Deitelzweig said, adding that a prior expansion of the hospital's critical care tower facilitated the growth. “That was done in three months. That's 100 beds. . . . It's miraculous that we use those beds now.”

On April 20, the health system discharged its 1,500th COVID-19 patient.

***

Meanwhile, Mayo Clinic Hospital in Phoenix had been waiting for a sudden spike of COVID-19 patients since the end of January, said hospitalist and ACP Member Benjamin Dangerfield, DO.

Dr Dangerfield photo courtesy of Mayo Clinic
Dr. Dangerfield, photo courtesy of Mayo Clinic

“We developed plans where we could manage [an additional] 25%, upwards of 50% of our licensed bed capacity if need arose,” he said. With outpatient areas closed, leaders of the nearly 300-bed hospital decided that if volume got high enough, a variety of outpatient physicians would help manage inpatients alongside the hospitalists.

“We got everyone in the outpatient environment fit-tested for N95s, and we got them through training in the inpatient Epic environment, and we had them on call schedules and ready to come in and respond,” Dr. Dangerfield said. “I think it gave the inpatient practitioners reassurance that there was support for them, that they wouldn't be left alone to manage huge volumes of patients.”

But as of March and April, volumes at the hospital were far from huge.

While the first COVID-19 case (rather, three cases on the same day) came around the third week of March, the patient census was unusually low, said Dr. Dangerfield, vice chair of the hospital practice subcommittee, which represents the medical leadership for the hospital.

“When we first started taking COVID-19 patients in the hospital, our numbers actually drastically fell,” he said. “There was a [state] stay-at-home order, we stopped taking procedural elective and other types of elective admissions to the hospital, and the overall hospital census dropped to nearly 50% of our normal census.”

For a while, it stayed that way. Due to very low numbers on the hospital internal medicine service, “We actually stopped bringing in as many providers as we typically have on the rounding services. . . . For whatever reason, there was less people coming into the hospital and needing our care, so for a good time, we were a little overstaffed,” said Dr. Dangerfield.

It would still be a couple of months before Arizona made the news as part of a second surge of COVID-19 cases.

***

Back in March, hospitalists knew very little about treatment for COVID-19. “Almost everything that had been published was either anecdotal or observational, with major flaws,” said Dr. Southern. “So we started with having our infectious disease division create a treatment protocol, which we all understood was based on flawed evidence because there wasn't anything else.”

The hospitalist group quickly realized that those on the ground simply couldn't keep up with the emerging evidence. “So we created a centralized portal and published protocols and added to them as the evidence accumulated,” he said. “And even sometimes when the evidence didn't accumulate. For instance, steroids was one of those.”

The steroid protocol was initiated by a hospitalist “who had been working a lot of days in a row and was incredibly clinically astute,” Dr. Southern said. “In a last-ditch effort, he treated a few patients who he knew were going to get intubated with steroids.”

After seeing improvement in all five of the initial patients, the hospitalist went to Dr. Southern's office to report his findings. “He's a really humble guy, but he said, ‘You know, I really believe it.’ So we published in early April a protocol for administering steroids to the sickest patients,” he said. “If you think about it, that's a little bold, right? There was no evidence, and, in fact, most of the agencies who were writing recommendations at that time were specifically recommending against giving patients steroids.”

Therapeutic anticoagulation was another early protocol. Around April 1, the chief of hematology approached Dr. Southern and said that the coagulation parameters she was seeing suggested that COVID-19 produced an extraordinarily prothrombotic state and hypothesized that some patients might have been succumbing to thrombotic events rather than acute respiratory distress syndrome and inflammation.

The chief advocated for therapeutic-dose anticoagulation as prophylaxis for patients with D-dimers greater than 3.0 µg/mL, regardless of symptoms or enrollment in a study, Dr. Southern said. “At that time, the best evidence that we had was that there were a few anecdotes of autopsies from China that suggested there were microthrombi in virtually everyone's lungs. So she was really bold, and I have to say, for about 24 hours, I was resistant.”

Despite Dr. Southern's initial reservations, the hospital's first steroid and anticoagulation protocols were published on April 5 and 10, respectively, on a web-based portal and app for easy access. Because of the evolving evidence, each one was revised on a near-daily basis and had dozens of versions.

“I believe that those two things—the steroids and the anticoagulation—saved many people . . . although we had to adjust who were the right people to treat,” he said. “Also, it behooved us to create a system where we could publish and disseminate that information to all the various literally hundreds of providers who were offering care on the inpatient side who didn't participate in the normal inpatient channels of communication.”

***

Ochsner Health also had a systemwide approach to protocols and treatments. Dr. Deitelzweig even used his own clinical and research expertise in anticoagulation, as well as his position on the board of directors of the Anticoagulation Forum, a nonprofit organization of physicians, nurses, and pharmacists that provides information on best practices, to develop protocols for the health system.

“I was moved to stay in touch with colleagues around the globe about what they were doing, so with anticoagulation, I adopted some of those practices,” he said, adding that his hospitalist group even had a secret weapon when it came to building the order sets. “We are an Epic EMR shop, but in my department there is an Epic writer: The assistant chief medical information officer is a hospitalist here, and he was able to build the Epic order sets in real time with us.”

Around the city's case peak on April 17, the hospital's patient-flow center played a crucial role. PILOTs, or physicians in lead of transfers, were responsible for managing bed availability, staffing availability, and the overall census, Dr. Deitelzweig said. “We have four [full-time equivalent] hospitalists assigned to that role, and these people are basically traffic cops. . . . We were moving people around the whole health system like I've never seen before, because the virus was moving. So if it was heavy in New Orleans, maybe some of the New Orleans [patients] would go to Slidell, which is about 30 or 40 miles away. And then when it got really bad in Slidell, those people would come down here.”

The hospitalists were up for the challenge on the clinical side as well, he said. “The team has been doing heroic work for a long time now. There were some folks that were very hesitant to really engage with the COVID patients, but they did some other things like telemedicine. But really, most everyone stepped up and continues to step up.”

Dr. Deitelzweig said he continues to emphasize to them that the battle against COVID-19 is a marathon, not a sprint. “But they've demonstrated incredible resilience, and the morale has been high,” with survey results in June showing greater morale on the medicine service than in other parts of the organization.

He attributes these successes to leadership visibility and rounds involving the executive team, as well as the front line, across the health system. “We've made every effort to go to every place where we have programs multiple times during the last few months. We would do it before, but not like this,” he said, adding that his own flights to Shreveport, La., were replaced by 10-hour round-trip car rides.

Asked whether he's exhausted, Dr. Deitelzweig replied, “You know how a president looks not four years older, but more like 15 years older? That's how I feel, but I'm not exhausted.”

***

Montefiore reached its high watermark of COVID-19 cases on April 11, exactly one month after its first case. “You can imagine, we went from 0 to 1,200 in 31 days,” said Dr. Southern. “I slept, but I bet I didn't take a day off, and neither did any of the people in my division.”

In New York City Montefiore Medical Centers resident teaching teams spent many long days fully garbed in their PPE Pictured here from left are Kevin Zheng MD ACP Member Shitij Arora MD and A
In New York City, Montefiore Medical Center's resident teaching teams spent many long days fully garbed in their PPE. Pictured here, from left, are Kevin Zheng, MD, ACP Member, Shitij Arora, MD, and Amber Khan, MD. Photo courtesy of Montefiore Medical Center

He told the hospitalists that everyone could start taking days off as soon as the hospital's census was lower than the day before, which happened to be April 12. “One of the other things I said to everyone is that ‘If April 11 was our peak, then April 12 will be our second worst day,’” he said with a laugh. “But I do have to say that getting off the exponential part of the curve was an extraordinary relief because I believed that we were going to fail in some way.”

In addition to relief, Dr. Southern has other strong feelings about the whole experience. “Early on, I recognized that all of the people in my division and all of the people outside of my division I would call on to help had to step into something that was really, really scary. . . . It felt very, very much to me like I'd be sending people into battle and that I knew that not everyone was going to come home,” he said.

Fortunately, while many clinicians got sick, no one in the division died or has permeant sequelae, Dr. Southern reported. “But in March, I didn't believe that to be true. I thought that, from a numbers standpoint, if we sent literally hundreds of people into the hospital to take care of COVID patients, that someone was going to not make it.”

Nonetheless, he needed to be calm and reassure clinicians that they weren't on their own. “I think the thing that was most important is that people had to understand and realize that we were behind them, we were supporting them, and the seemingly infinite number of issues that came up that had to be dealt with would all be dealt with,” said Dr. Southern, who has led the division since 2011.

Part of that reassurance came in the form of team calls at 9 p.m., which were attended by 150 to 200 people each night. These updates covered practical concerns, such as treatment protocols and the supply of personal protective equipment, but they also served a more emotional function.

“So many people were self-quarantining from their families or, if they were living alone, they weren't having any outside contact. . . . So what many people said about those calls was they're sort of like the fireside chats [given by President Franklin D. Roosevelt], that they just wanted to hear someone's voice every night,” Dr. Southern said.

As he walked the hospital floors, he thanked people for their service and said he saw an outpouring of energy and bravery. “A really large number of people said, ‘This is what I'm supposed to be doing. This is why I became a doctor.’ Some people even said, ‘I've been waiting for this. I had been wondering whether there was going to be a moment.’”

On the other hand, some physicians were distressed by the incessant overhead codes, the pressure of dealing with end-of-life issues, and flashbacks to traumas from their own residency training, Dr. Southern said. But everyone played a role in the response.

“No one ever left their post. No one ever didn't step in. No one ever said, ‘No, we can't do that.’ No one,” he said. “Everyone stepped right in.”

***

As case counts continued their gentle declines in New York City and New Orleans, they suddenly shot up in Arizona.

Soon after Gov. Doug Ducey announced that the state's stay-at-home order would expire on May 15, patient volumes picked back up to normal before exceeding typical volumes at Mayo Clinic Hospital, said Dr. Dangerfield, who is also medical director of the hospital's operations command center.

ACP Member Benjamin Dangerfield DO describes the command center at Mayo Clinic Hospital in Phoenix as a smaller version of NASAs mission control Photo courtesy of Mayo Clinic
ACP Member Benjamin Dangerfield, DO, describes the command center at Mayo Clinic Hospital in Phoenix as a smaller version of NASA's mission control. Photo courtesy of Mayo Clinic

The command center is staffed by a multidisciplinary group, which manages hospital operations in one room with three big screens on the wall that provide real-time data, “kind of like a smaller version of NASA mission control,” he joked. “We can monitor hospital operations and try to keep an eye on where there are bottlenecks in patient movement and then try to make improvements.”

As the hospital became busier in June and cases peaked in July, it was advantageous that the command center had just launched on March 21 and was ready to manage capacity issues, Dr. Dangerfield said. “We started planning it and building it in October of last year, and it was just coincidental that it came online right as the pandemic of the century appeared.”

The center was heavily involved in managing day-to-day changes in hospital capacity and determining where to place COVID-19 patients in the hospital. “It was just an ever-shifting landscape that required real-time evaluation and intervention,” said Dr. Dangerfield.

The hospital never ended up having to pull outpatient doctors into inpatient practice. With some help from volume triggers, the hospital internal medicine service's hospitalist team, which includes 46 physicians and 23 advanced practitioners, handled the load.

Certain patient volumes would trigger additional support from residents and hospitalists who were off service or in nonclinical administrative roles at the time, Dr. Dangerfield said. “Our typical schedule is seven-on, seven-off, but that wasn't always the norm. . . . There was a little bit less continuity in the care of the patients than seven days straight,” he said. “We may need to experiment more with that moving forward just because that seven-on, seven-off schedule is a challenge.”

The percentage of COVID-19 patients at the hospital peaked at about 50%, with the hospital straining capacity and the closed ICU being most heavily hit due to long lengths of stay, Dr. Dangerfield said. Initially, the protocol was to send any COVID-19 patient needing more than 6 L of oxygen directly to the ICU, but that became unsustainable.

To support their critical care colleagues, the hospitalists comanaged COVID-19 patients who were on high-flow nasal cannula in the progressive care step-down unit.

“We would round twice a day with an intensivist and keep an eye on these patients,” Dr. Dangerfield said. “Actually, I spent some time in there myself [providing care]. . . . We were proning [patients] and giving them maximal supportive care. It was a bit of a challenge, but for me, it was not an unwelcome one. I enjoyed it.”

Still, he and his colleagues felt ongoing anxiety. “There was always this fear that suddenly there'd be an even bigger surge and we'd have to be managing patients in the unlicensed areas of the hospital,” he said.

Ultimately, those concerns never manifested. “It was a little bit odd [that] we had all the surge plans in place, and then we didn't use them for an extended period,” Dr. Dangerfield said. “Then we got into June and July, and things got a bit more hectic. But still, I would say our numbers never went above a fairly manageable level. And we are very grateful for that.”

***

As of September, all three hospitalists reported declines in the number of COVID-19 patients they were seeing.

At Montefiore, “We have very, very few patients with COVID, really just a handful. We haven't—I'm knocking wood—yet seen any sign of an uptick at any one of our hospitals, so our COVID census is low and, for the most part, both our medical and surgical services are operating at something like between 60% and 80% capacity, and we're usually at 100% capacity,” said Dr. Southern.

At Ochsner Health, “It's been a steady number of COVID patients. We're at a lower level. Of course, [in] the health system, maybe 150 to 200 COVID patients on any given day are hospitalized, and we've been pretty much at that level for six weeks,” Dr. Deitelzweig said.

At Mayo Clinic Hospital, “We've seen a gentle decline since late July . . . and now it's gone back down towards our normal volumes again. Still present, but at lower levels,” Dr. Dangerfield said.

In this period of relative calm, the hospitalists stressed the importance of being able to re-engage their plans and of keeping all hands on deck if the pandemic worsens. As Dr. Deitelzweig put it, “There is no off deck.”

There's a famous quote attributed to President and General Ulysses S. Grant to the effect of, “In every battle there comes a time when both sides consider themselves beaten. Then he who continues the attack wins.” By all accounts, the battle against the novel coronavirus is not over, and no one knows what's in store for the first combined COVID-19 and influenza season.

But one thing is clear: Hospitalists are ready to continue the attack.