Thread by @JaymeKFraser

Do hospitals have enough beds to care for seriously ill patients if there’s a surge of #COVID19 cases?

That’s the question @mattwynn and I tried to answer. #coronavirus #WhatReportersDo THREAD /1

To be clear, this is a mental exercise, not a hard-and-fast prediction. We just wanted rough numbers to help us ask better questions about the tough decisions that lie ahead. Infectious diseases experts have already warned hospitals could be overwhelmed. /2

In short, no state has enough hospital beds to treat seriously ill #covid19 patients if #coronavirus cases spike like other countries. We found that there could be six patients with severe or critical symptoms for every existing hospital bed. usatoday.com/in-depth/news/… /3

Most hospital beds are already in use, so reality could be far worse: about 17 seriously ill people per open bed. /4

In total, we estimated 23.8 million Americans could contract #COVID19, including 4.7 million severe or critical cases. For comparison, John Hopkins University experts estimated 38 million total cases and up to 9.6 million hospitalized. /5

The picture varied by state and metro area. Oregon, Washington, New Mexico and California might need eight times as many beds. South Dakota and North Dakota have the smallest gap between beds and potential need, but still could draw three patients for every bed. /6

Read the full story here: usatoday.com/in-depth/news/…
Keep reading this thread for a walk through of how we got these numbers. #WhatReportersDo /7

We gathered population and age data from the U.S. Census for states and for Core-Based Statistical Areas (CBSAs), a type of metro-area geography. The American Hospital Association reports the number of hospital beds by state and by CBSA, so we were able to match those up. /8

Then we had to make some assumptions about coronavirus. We assumed an infection rate of 7.4%, which was the lowest figure for the standard flu in the U.S. in the last five years. Experts we talked to said this was a conservative number to use. /9

We also used figures from the WHO that 13.8% of cases are severe and 6.1% are critical. This is the group we assumed would need hospital-level care at some point.

/10

Then we ran the numbers. (Population X Infection Rate X Severity Rate) divided by the number of total hospital beds. We also ran variations that looked only at the population 60+ and that assumed 2/3 of beds were already in use, the national figure reported by AHA. /11

To get a rough sense of how much states would need to spread out the timeline of cases in order to have adequate capacity, we assumed people would be hospitalized 10 days. (Slightly less than the 11 days found by Los Alamos National Laboratory’s review of data from China.) /12

We also assumed a standard 30-day month for ease of calculations. /13

Example: About 44,000 people in Arkansas might need hospitalization in a state with ~3,400 open beds. So the state could manage if the cases were spread out over five months or more. But if the surge lasts four months or less, Arkansas would quickly run out of beds. /14

We also provided some metro-level data to local reporters in our company. They are best positioned to ask questions about actual bed vacancies, surge plans, etc. because they live in those places. I'll add them to thread as I see them. /15

The reporting also was informed by interviews with lots of infectious disease and public health experts. I have 41 pages of typed notes for this story. And we read many of the early studies based on data from other countries to make sure we weren't being overly pessimistic. /16

And. . . that’s it. That’s how @mattwynn and I did this story. Hope it was helpful to walk you through it. Let us know if you have any questions or any ideas for future stories we should do.
usatoday.com/in-depth/news/…
/17