Thread by @DataDrivenMD, /NEW: #COVID19 Hospitalizations down to the *hospital-level*(I had a few hours to [...]

/NEW: #COVID19 Hospitalizations down to the *hospital-level*
(I had a few hours to spare)

It’s based on the newly-released HHS dataset. The data itself is directionally solid (H/T to the data team at HHS).

Explore by visiting the @CodersCovid site:
https://reports.codersagainstcovid.org/hospitals

You can look at the national picture, or focus on one (or more states) just by adjusting the state filter at the top.

The other filters are dynamic: changing one automatically adjusts all of the others so that you’re only working with the data you want.

Here’s Texas:

The drop-down filter for “County” is searchable, so you can quickly find information about where you live.

Side note: you’ll be surprised to see how many counties in the U.S. have 0 hospitals

I included a searchable filter for hospital names, too. This is helpful if you’re looking for historical trends.

Notes:

Reporting requirements have evolved, and you can see hiccups in the data. Don’t worry: the data is *directionally* accurate — focus on the trend

I included a searchable filter for hospital names, too. This is helpful if you’re looking for historical trends. Notes: Reporting requirements have evolved, and you can see hiccups in the data. Don’t worry: the data is *directionally* accurate — focus on the trend

Note on hospital names (cont.)

There are several field hospitals in the dataset. These don’t have a name, hence the “null” values here. I’ll make some time later this week to clean up the dataset a little more and make this context self-evident.

Note on hospital names (cont.) There are several field hospitals in the dataset. These don’t have a name, hence the “null” values here. I’ll make some time later this week to clean up the dataset a little more and make this context self-evident.

There’s also a filter for the *kind of hospital*

A few notes:

“Short Term” is what we all think of when we say “hospital”

“Long term” hospitals focus on rehabilitation. A #COVID19 patient might end up in one of these if they need ventilator support for a long time….

There’s also a filter for the *kind of hospital* A few notes: “Short Term” is what we all think of when we say “hospital” “Long term” hospitals focus on rehabilitation. A #COVID19 patient might end up in one of these if they need ventilator support for a long time….

“Critical Access Hospitals” are found throughout the country, and tend to be smaller and limited in their ability to manage very sick patients. As you can see, there are several throughout the country that are being pushed to their limits, and dozens that are trending that way

“Critical Access Hospitals” (cont.): CAHs provide a “safety net” for rural communities, but looking at them on a map is misleading.

There are 1300 of them, they're everywhere, right?…

Don’t be fooled by the map: all of those Critical Access Hospitals are *tiiiiiiny* — all told, we’re talking about 35,000 beds.

For context, the state of Texas has over 62,000 beds.

Childrens Hospitals are in there, but I would take that data with a grain of salt.

Pediatric medical data is *exceedingly* difficult to de-identify. I won’t get into the details, but suffice it to say that a domino effect results in most of the peds data getting censored.

“What parts of the country are running low on ICU capacity?”

This is mission-critical information that will help us mobilize resources, so I added a dedicated filter for that, too.

And we have another eye-opening revelation:

1 out of every 3 *hospitals* don’t have an ICU…

1 out of every 3 hospitals in the U.S. don’t even have the physical space to deliver ICU-level care.

So, where are these “ICU deserts”? Everywhere. Many are Critical Access Hospitals, all but a few are in rural America.

Health disparities take many forms, and they affect people of all races and ethnicities: “ICU deserts” illustrate how geography can be drive poor health outcomes. *This* is why data is critical: we can’t fix what we can’t see.

Now, try filtering for ICU capacity < 15%…

This is what keeps me up at night: metro hubs are running out of ICU beds.

That’s a *huge* problem b/c US healthcare delivery is predicated on the notion that small, rural communities always have the ability to transfer patients to “higher levels of care”

Remember the thousands of critical access hospitals?

Recall that 1 out of 3 hospitals in America don’t even have the *physical space* for an ICU (let alone the staff)?

Where are these #COVID19 patients going to go if there’s no room in Sacramento, SLC, San Francisco, Las Vegas?

What about all of these #COVID19 patients that are waiting in the wings?

Or all of these #COVID19 patients that found out that they contracted a deadly virus within the last couple of weeks?

We’re already running out of ICU capacity throughout the country.

These are all the hospitals that have less than 15% of their ICU capacity remaining.

It’s not just California, It’s also Arizona as @Cleavon_MD has called out several times per week for several weeks.

It’s also El Paso, every metro hub in TX + the Panhandle, OK, Missouri, Mississippi, Louisiana, Alabama, ATL, Florida…the northeast, the midwest, the northwest…

cc @OliviaTroye @morgfair @PeterHotez @ashtonpittma @drmt @NicoleCarrWSB @DWUhlfelderLaw @IHME_UW @juliettekayyem

And to round everything out: there’s *also* a racial and ethnic disparity in access to hospital care: there are fewer hospitals and fewer hospital beds per 100K persons in counties where >55% of the population are persons of color.

cc @DrNunezSmith

These communities are in the same parts of the country where ICU capacity is dwindling.

So, taken together, these data show us that:

we’re running out of hospital capacity at the local level *and* regional levels

geographic *and* racial/ethnic disparities are each playing a role in creating a bottle neck in ICU-level care
(cont.)

there are 2+ waves of newly-infected persons that will push the system beyond its limit — that’s a fact — and that’s not taking Thanksgiving entirely into account yet.
(cont.)

Final words:

The vaccine is coming, but it won’t be here in time to save us from the next 2 waves, which will hit us in the next 1-2 weeks.

We can *and must* do more to protect ourselves: #WearAMask

and #HunkerDown

Plan for virtual holiday gatherings for the rest of year

Above all: don’t lose sight of the fact that we will get through this.

There are better days ahead. I want you to be there to enjoy them. Your friends and family do, too.

So long as we focus on what we *have* to do now, we’ll get back to what we *want* to do in the future./

Latest Threads Unrolled: