A call to revise CDC guidelines for reopening

U.S. hospitals are projected to lose $200 billion in revenue by the end of this month. Hundreds face bankruptcy. At a time when we need hospitals to be functioning at peak performance, CDC and state government guidelines for reopening could jeopardize their ability to do so. 

According to the guidance, 30 percent of hospital intensive care unit (ICU) beds should be available before communities can safely reopen. This arbitrary figure will needlessly hamper areas that are ready right now —and could drive hospitals out of business — while also opening the floodgates in areas that are not yet prepared to handle the incoming demand. Instead, the CDC should urgently revise its guidelines to be rooted in evidence and advise state governments to defer relevant decisions about the safety of reopening to local health authorities. 

There is no widely accepted evidence that 70 percent of ICU occupancy will be necessary to handle a future COVID-19 surge. While average occupancy of ICUs was 68 percent before the pandemic, this figure varies substantially in hospitals and cities across America. A 2016 study by Prof. Chan of 15 Northern California Kaiser Permanente hospitals found average ICU occupancy was 80 percent. It is impractical to limit hospitals to 70 percent when they needed more of those beds for patients before COVID-19 struck. 

These restrictions run a real risk of tipping hospitals into insolvency as many face an existential crisis. Complex surgeries are the financial lifeline of many hospitals and require a post-operative ICU stay. If they must idle 30 percent of ICU beds before they can restore normal surgical operations, it’s only a matter of time before they run out of money. 

While it can be comforting to have concrete benchmarks to strive towards, simply capping ICU occupancy at 70 percent fails to capture the nuances of critical care capacity management. The inherent uncertainty of the COVID-19 epidemic means that some days’ demand for ICU beds may be many-fold higher than it is on the “average” day. Patients with critical illness from COVID-19 often require prolonged ICU stays. A new surge that brings only three new ICU patients daily can snowball into 20 patients requiring an ICU bed in under a week.

Hospital size matters substantially, too. In a small community hospital with 10 ICU beds, 70 percent occupancy means there are only three beds available in the event of a “second wave”. In contrast, a large academic medical center with 100 ICU beds will be able to react much more efficiently and effectively to a second wave with 30 empty beds for new patients. 

The new guidelines are ambiguous as to how hospitals or policymakers should define ICU capacity. As the onslaught of COVID-19 patients hit New York City, many hospitals quickly canceled surgeries and created makeshift ICU beds in newly-idle operating rooms and general hospital wards. In doing so, the city’s “ICU capacity” more than doubled. As the number of COVID-19 cases subsides in New York, it is not clear whether and how this “created capacity” — much of which has been returned to its normal function — should be considered. 

We should also consider the trade-offs this massive and rapid expansion of ICU capacity has produced. While it is true that New York City did not run out of ICU beds — he Javitz Center, Comfort, and other temporary hospitals were ultimately used in a very limited capacity — whether these additional ICUs function optimally is a lingering question. Critical care training is a complex, multi-year training process. In the sudden increase in demand, many ICUs had to be staffed by doctors and nurses who don’t usually work in these environments. It will be some time before we understand how these workforce reassignments — coupled with local shortages of essential medications, dialysis machines, ventilators, and personal protective equipment –have impacted patient outcomes and clinician wellbeing. 

As the country moves out of lockdown, criteria need to be tailored for hospitals, cities and states based on geography and patient demographics. In the hardest-hit areas, such as New York City in which an estimated 21 percent of the population has coronavirus antibodies (indicating prior infection), a second wave will look very different than in places like Los Angeles where antibody prevalence is only 4.34 percent. Presuming that infection confers some level of immunity (even if weak), the disease will spread more rapidly in places where fewer people were infected in the first wave. 

Additionally, as COVID19 disproportionately affects the elderly, people with comorbidities (such as diabetes), African-Americans and Hispanics, reopening plans should account for local community risk based on regional population demographics. Hospitals serving at-risk communities likely will need more surge capacity.

With mass protests continuing across the nation and restrictions slowly being relaxed, we can’t risk hobbling our healthcare system with arbitrary requirements for reopening. The CDC’s one-size-fits-all guidelines could hurt our recovery effort and must be revised. State governments should base decisions about reopening on the data and advice of local hospitals and health authorities. 

Carri W. Chan is an associate professor in the Decision, Risk, and Operations Division at Columbia Business School and an expert on hospital operations management. Hayley B. Gershengorn is an associate professor and a Critical Care Physician at the University of Miami Miller School of Medicine.