Arizona, California, Washington, Wisconsin appear to have more hospitalized patients than beds. 24 states poised to join them.
Lagging state and federal data paint a far rosier picture than harrowing ground conditions.
“Are there enough staffed hospital beds to take care of the sick?” This is perhaps the essential question for assessing hospital capacity and safety. For the first time in memory—if not ever—the answer in many US counties cannot be said to be certain.
Based on the model that runs the Covid-19 Hospital Capacity Circuit Breaker Dashboard my team developed, today we now project that hospitals in Arizona, California, Washington, and Wisconsin are at capacity or over capacity; there are more hospitalized patients than staffed beds in those states. Hospital care may be compromised in some areas.
In addition, 24 more states are forecasted to reach that point in the next week or so, unless changes are made, changes like increasing nursing-to-patient ratios, cancelling elective procedures, taking risks by sending admitted patients home sooner, and discharging emergency room patients that might normally be hospitalized. Another 17 are at risk of a capacity crunch too. Only 5 states and Washington DC are in the low risk, (“have capacity,”) category. Below, is a timeseries, created by Benjy Renton, dramatically showing how things progressed from November through January 12th.
As the two maps below illustrate, our model paints a far worse picture than supposedly real-time data made available by the US Department of Health and Human Services. Our model generates the mostly red, orange, yellow, and purple map on the left, indicating that hospitals are in trouble. Meanwhile HHS’s data generates a map that suggests everything is hunky-dory in 35 states. In fact, our model says that 89% of US hospital beds are full. HHS says only 79% are. That’s a major difference. We say 4 states have reached or exceeded hospital capacity. HHS says no state has reached that point.
On the left, a map generated by our Covid-19 Hospital Capacity Circuit Breaker dashboard using our nowcast model. We project 4 states are at or above capacity, 24 are poised to reach that point, and all but 4 states and DC are at risk. On the right, a map that uses real-time HHS data, which generates a far less dramatic picture of things.
Why does our dashboard make things look so much more dire than federal reports? Because we “nowcast” to make up for the fact that federal data is clearly lagging several days (or more) behind.
Meanwhile, reports from healthcare workers in hospitals describe an increasingly chaotic picture in many if not most regions of the United States, including the ones our model highlights as areas particularly in danger.
•••
I’ve always been for data over anecdote. Am I abandoning my principles? No. In this case, we have conflicting data and when two sets of data disagree, on-the-ground information carries a lot of water. In this case, the anecdotes match our model's data, not the federal (and most state and local) data that are available to the public. And we know why: most official data have been lagging too far behind, reflecting what was, not what is.
I ask another doctor I used to work with, “are there any beds?” The response is brief. “Um, no haha.”
To learn more—and really to reassure myself that our model and dashboard was not inappropriately alarmist—I asked healthcare workers in these regions and others to tell me what they’re seeing. I assured them all anonymity. I know many of them personally and for the others, I’ve verified they are who they claim to be (i.e. not fakers or bots). Note: Each of the following quotations are from different doctors.
Washington state (HHS says hospital capacity is at 86%; we estimate they are at 102%):
“Even within the system can’t find beds or move people around. All elective surgeries and procedures cancelled. Patients waiting 12-48 hours for a bed and some ultimately getting discharged from ED. No higher level beds available.”
“I can still typically move sick kids out without huge delay, but adults are boarding.” [i.e. getting their care in ERs and sleeping there].
“Not enough beds. But cancelling elective surgeries, opening new units, pulling floor/OR staff to ERs for boarders has kept us functional. Barely.”
Arizona (HHS says hospital capacity is at 83%; we estimate they are at 112%):
“Small community access hospital. Last shift was 1 week ago. At that time had been boarding a gastrointestinal bleed for 4 days. And I caught COVID along with several ER staff, all fully vaxxed.”
One ER doctor told me, “I finally managed to get a patient with sepsis admitted to a hospital an hour from the ER where I was working. Most patients who are admitted board for 20+ hours.” But otherwise, admissions were impossible. This doctor told me that patients with serious skin infections were getting intravenous antibiotics, sent home, and told to come back a few days in a row for more doses.
I ask another doctor I used to work with, “Are there any beds?” The response is brief. “Um, no haha.” A few minutes later I get a follow-up text. “Our hospital is over full.” I wait to hear more. But the texts stop.
Wisconsin (HHS says hospital capacity is at 75%; we estimate they are at 101%):
“Always boarding, but beds do turn over eventually. Critical transfers (heart attacks, trauma, strokes needing acute treatment) get out the door. But everyone else waits and waits.”
“Central Wisconsin. I work at the tertiary center, so we don’t generally have to send patients to outside facilities. However, I can tell from experience that the number of transfers we are able to take in are pretty minimal. The beds we do have, we try to save for the critical transfers. So, I can’t speak to the difficulty with transferring out, but I can say that we are seeing patients in halls, waiting rooms, admitting then from the waiting room, etc. The non-critical patients (when we can squeeze them in) have generally been sitting in another ED somewhere for 24+ hours.”
“Southeastern WI. Vaccinated docs are starting to have Covid-19 themselves. Staff (nurses and docs) are stretched so thin. Boarding patients in the ED, but not enough nursing staff to care for them.”
A friend of mine in Wisconsin texts me to say he’s seen our dashboard. I ask if it matches what he’s seeing. “I’m not sure and can’t currently verify this, but it feels like even your data is under-playing how bad things are compared to my own experience in several rural hospitals in Wisconsin in the past 9 days. I just flew a patient to Chicago.” He wants to talk on the phone. It's good to hear his voice. But he sounds pretty beaten down. His voice and cadence, somehow, reminds me of a mentor we both once shared, and who liked to tell war stories, always boisterously smiling his way through hair-raising ER stories. This, I thought, is our war.
California (HHS says hospital capacity is at 82%; we estimate they are at 101%):
“Inland Southern California. No beds, tons of boarding, transfers near impossible. It's a nursing issue and not a physical bed issue. Also, transferring patients is almost impossible.”
“Los Angeles. We have patients in hallways, and are starting to turn clinics into inpatient wards.”
Two other doctors, both in Northern California respond to a post of mine, and both say there are no beds, transfers are not happening, and patients are boarding in the ER for prolonged periods.
I ask a former colleague of mine, now working in Northern California, if there are any beds in his region. “Some, not many.”
Ground conditions indicate that our hospital capacity model depicts reality more closely than the portrait according to HHS. And these regions are not alone. One physician on the east coast DM'd me the following nightmare: “Seriously I spent 2 hours on the phone with health officials last night trying to figure out literal disaster plans bc our trauma centers are doing no surgeries except the most time sensitive, and our largest hospitals’ ICUs are ~50% COVID patients (& >90% capacity) and my emergency department has wait times of 6-12 hours. Many of my staff are ready to give up as are half my physician friends. But whatever, we are exaggerating ☹️.”
“Los Angeles. We have patients in hallways, and are starting to turn clinics into inpatient wards.”
•••
Because I’m a practicing physician, I have the opportunity to get my questions to other doctors answered quickly, and from trusted sources in many cases. I am fortunate that I can simply pull my phone out and text or email dozens of healthcare workers and get a bunch of responses almost immediately. I can post on internet forums where people’s identifications have already been verified.
But what about the general public? For those wondering if local hospitals are full, websites publishing state and federal data supposedly in “real time” have great appeal. Trouble is that almost all of these “dashboards” rely on out-of-date information. In Omicron time, data that is a few days old—let alone a week or more—is obsolete and can’t be relied on. If you were to think there are prominent or accurate warnings to this effect on most (if any) of the major hospital capacity dashboards published by federal or local governments, you’d be incorrect.
Now add in the usual crop of pundits looking yet again to minimize the impact of Covid-19. Apparently, the archetypal tech-bro who trades in his familiarity with objective “data” seems to lack the sophistication required to understand simple things like reporting lags. Fine print isn’t their thing. The amateur stops searching for answers when he finds one that delights him, not one that is irrefutably accurate.
We all tolerate different levels of risk. And for many nowadays, whether they realize it or not, learning that local hospitals are overflowing remains about the only thing that will stop them from going about their normal lives. “Pandemic fatigue” has, in some places, been replaced wholesale by “pandemic pretend-it-doesn’t exist.” Hospital workers do not have that luxury.
•••
❓💡🗣️ What are your questions? Comments? Join the conversation below!
Follow me on Twitter, Instagram, and on Facebook and help me share accurate frontline medical information!
📬 Subscribe to Inside Medicine here and get updates from the frontline at least twice per week.