Hospital Capacity amid the COVID-19 Pandemic

Have hospitals across the country had the capacity to handle the number of hospitalizations? Figure 2 maps the largest two-week change in estimated hospitalizations relative the number of available beds in each county.

This reflects our estimate of the peak strain on hospital capacity through June 30. The estimate requires the two aforementioned assumptions about the share of cases requiring hospitalization and the share of beds used for non-COVID-19 reasons. We also assume that these shares are constant across counties and time. It is important to note that this estimate does not account for the shortage of hospital staff, protective equipment or other critical equipment, such as ventilators. Thus, the measure of available beds only captures a hospital’s capacity to house patients, not the effectiveness of treatment.

With these caveats, our analysis indicates that hospital resources have likely reached capacity in a few areas of the country, particularly in counties within Arizona, Louisiana and Texas. This is consistent with anecdotal reports of cities constructing temporary facilities. However, despite the spike in new cases at the end of June, 96% of counties stayed at or below hospital bed capacity with 91% of all counties at less than 75% capacity.

Capacity and Another COVID-19 Surge

While social distancing measures and government stay-at-home orders have mitigated the strain on hospital systems, there continue to be regional spikes in new cases. Recent increases in Texas, Florida and Arizona have resulted in officials temporarily halting reopening plans. As economies begin to reopen, some counties are operating with less room to handle an increase in new cases.

To study the strain to the hospital system due to regional spikes we modeled a severe case in which hospitalizations per capita rise to 9 per 10,000 people. This represents the 95th percentile of cases per capita at the county level.The magnitude of our assumed two-week spike is based on the current U.S. experience, not an epidemiological model. We also assumed that the number of available hospital beds remained constant, as adding infrastructure to hospitals is costly and cannot be done on short notice.Many regions have developed temporary sites providing more beds to support their patients. Our analysis does not account for this potential expansion. Figure 3 maps this scenario.

This map paints a very different picture of hospital capacity across the U.S. Only 77% of counties would have sufficient bed capacity to handle all new COVID-19 cases under this scenario and remain below 100% capacity on beds. Half of counties would be above 75% capacity.

We should note that another assumption underlying our analysis is that all counties provide similar access to testing. While testing data at the county level is not readily available to verify this assumption, some evidence shows that rural areas might be facing distinct challenges in accessing testing kits.

For instance, testing data from the Covid Tracking Project at the state level show that Alabama (which has many rural counties) has conducted 406,652 tests through the end of June, or 82 tests per 1,000 people. On the other hand, New York (with many urban areas) has conducted more than five times as many tests per capita, at 466 tests per 1,000 people. Assuming asymptomatic people are less likely to be tested when testing is constrained, more widespread testing could reduce our assumption that 12% of cases require hospitalization.

A Closer Look at the Eighth Federal Reserve District

The Eighth Federal Reserve DistrictHeadquartered in St. Louis, the Eighth Federal Reserve District includes all of Arkansas and parts of Illinois, Indiana, Kentucky, Mississippi, Missouri and Tennessee. had 72,217 COVID-19 cases as of June 30 and relative hospital bed capacity averages around 3 per 1,000 people. The average occupancy rate for hospitals in the Eighth District is around 44%, which is in line with the national trend.

For 98% of counties, the peak in new cases was well below hospital bed capacity. In 92% of counties, the peak was less than 75% of capacity. Under the scenario where hospitalizations rise to 9 in 10,000 people, 57% of counties in the Eighth District would have sufficient bed capacity to handle all new COVID-19 cases and remain below 75% capacity on beds, while around 78% of counties would remain at or below 100% capacity.

Conclusion

Many reports note that the health care system is facing severe shortages in terms of equipment and hospital beds as the U.S. manages the COVID-19 outbreak. This article attempts to capture these effects by examining the number of COVID-19 cases per hospital beds to study the magnitude of supply shortages.

We found that hospitals in 4% of counties—most of which are urban—may have been above capacity when new cases reached a peak. While this number may suggest sufficient capacity in many counties in the U.S., it is primarily due to lower infection rates in other areas. We found that 23% of counties would be above capacity with a surge in new cases similar to that of the worst hit counties in the U.S.

States across the country are paying increasing attention to hospital capacity as they reopen. For instance, New York’s plan to reopen includes monitoring the bed vacancy rate, with the desirable vacancy rate being greater than 30% of available beds (or 70% capacity by our metric). While our measure does not provide the exact hospital bed vacancy at any given moment in time, it is still a useful benchmark since it reflects the maximum strain that hospital systems across the country are capable of handling.

The article also raises broader questions about the optimal level of investment into health care capacity in an unfortunate situation of a surge in COVID-19 cases or another health crisis. Society may find it optimal to invest in excess capacity as a way to mitigate the economic impact of health crises.