In December, The Department of Health and Human Services (HHS) released a facility-level dataset about hospital utilization, capacity, and COVID-19 patients. This is the most detailed data ever made available on how COVID-19 is affecting American hospitals. This data is much more granular than what The COVID Tracking Project has been able to compile from state data, though in aggregate, HHS’s numbers are comparable to ours. So, we’re presenting it here.
The data presented for each facility on the map include:
- Adult COVID-19 patients currently in hospital: both confirmed and suspected COVID-19 cases (includes ICU patients)
- Percent of adult inpatient beds occupied by COVID-19 patients: Current adult COVID-19 inpatients divided by total staffed adult inpatient beds (includes ICU patients and beds)
- Percent of adult inpatient beds occupied by all patients: Current adult inpatients divided by total staffed adult inpatient beds (includes ICU patients and beds)
- Available adult inpatient beds: Total staffed adult inpatient beds minus currently occupied beds
- Adult COVID-19 patients currently in ICU: both confirmed and suspected COVID-19 cases
- Percent of adult ICU beds occupied by COVID-19 patients: Current adult COVID-19 ICU patients divided by all staffed adult ICU beds
- Percent of adult ICU beds occupied by all patients: Current adult ICU patients divided by total staffed adult ICU beds
- Available adult ICU beds: Total staffed adult ICU beds minus currently occupied beds
- Reporting completeness: Percent of all daily data fields reported by facility during reporting week
- Reporting week (spans Friday to Thursday): Data from facility’s most recently available reporting week
In some cases fields might have the following values:
- Between 0 and 4: the facility reported a low number of currently hospitalized patients (between 0 and 4 on average for the week). In the HHS dataset, these are displayed as
- N/A: this field was unreported or could not be calculated in the case of percent occupancy because the numerator (number of patients) or denominator (number of beds) was a number less than 4 or unreported.
- >100% of beds occupied: in most instances where reported percent bed occupancy is slightly greater than 100%, the facility may be experiencing an overflow of hospitalized patients beyond its total staffed bed capacity (including surge/expansion bed capacity). In rarer instances, usually where a facility’s percent occupancy is much greater than 100%, this is more likely due to erroneous reporting.
The HHS data comes either directly from hospitals or through state public health agencies and hospital associations that aggregate this data. It is submitted directly into the HHS Protect system or via a portal run by the company TeleTracking. To learn more about the facility-level dataset, see our initial analysis. For more granular information, see this community-created Frequently Asked Questions page. Original data files and data definitions are provided at the page hosting this dataset. Feedback and inquiries about the data can be shared with HealthData.gov.
In addition to being granular, the HHS dataset also presents metrics that were not available from states. Most importantly, we can see hospital capacity—or, what share of available beds in a hospital are currently occupied by patients—as well as new admissions of COVID-19 patients and demographic breakdowns of patient counts. Hospital inpatient capacity and a similar metric, intensive care unit (ICU) capacity, can indicate whether a particular hospital or region is becoming overwhelmed. As a hospital approaches 100% capacity, or all available beds are occupied, healthcare workers are increasingly overstretched and may be unable to devote the time and resources needed to best treat every patient.
The COVID Tracking Project has visualized hospital capacity and COVID-19 patients at each reporting facility. Bubble size corresponds to the number of adult COVID-19 patients currently at each facility while bubble color corresponds to the share of staffed inpatient or ICU beds which are occupied by COVID-19 patients. These counts include both confirmed and suspected COVID-19 hospitalizations.
This dataset is updated every Monday and reflects the prior week’s average of data reported daily by every facility.
Facilities represented in this data include Short Term Acute, Long Term Acute, Critical Access, and Children’s hospitals. Not included are Psychiatric, Rehabilitation, Indian Health Service (IHS) facilities, U.S. Department of Veterans Affairs (VA) facilities, Defense Health Agency (DHA) facilities, and religious non-medical facilities. Facilities reporting anomalous data may be minimized and flagged in display. Facilities without ICUs (those reporting zero total ICU beds) are not displayed on the “ICU patients” layer of the map.
In addition, facility locations and names presented here may not match up with physical hospital campus locations and names. When it is available, hospitals report to HHS based on unique identifier codes issued by the Centers for Medicare & Medicaid Services (CMS). One hospital system with multiple sites may report in aggregate for all sites, or each site within a system may report on its own, depending on how each facility is identified and set up for reporting to CMS.
The HHS Open Data team notes with the dataset that “Patients should not be discouraged from seeking hospital care based on their interpretation of the data. Hospitals have protocols in place to keep patients safe from exposure and to ensure all patients are prioritized for care.”