In my last newsletter, I wrote about “circuit breakers.” Subsequently, National Public Radio and the New York Times covered a memorandum prepared by Dr. William Hanage (an epidemiologist and a colleague of mine at Harvard) and me and which we sent to the Centers for Disease Control and Prevention (CDC). Here’s the idea, as NPR succinctly put it:
The next question was: “How can we know which counties and states need a circuit breaker?”
The answer is real-time data; real-time Health and Human Services (HHS) and CDC data that we have access to, and which, used correctly, can provide powerful and actionable information to local and state officials.
Knowing that we had the data we needed to identify places at risk of needing a circuit breaker, we created a simple module that pulls in federal and state data on current hospital capacity in 2,237 US counties with reporting hospitals, daily case counts, and recent hospitalization rates (that is, the percent of newly reported cases that ended up being hospitalized in that region recently). It's a prototype, but we think it's worth sharing. The result is a county-level and state-level map that indicates exactly which areas have already exceeded hospital capacity, and identifies ones that are poised overflow if they remain above the thresholds that we calculate for the next 5-10 days. In other words, for every county and state, we can tell officials how many new coronavirus cases can be “tolerated” without being at high risk of hospitals overflowing 1-10 days later.
The circuit breaker threshold dashboard (created by Benjy Renton, Alexander Chen, and me) is now live and can be used by officials to guide local decisions. Yesterday, Dr. Hanage and I sent a new memorandum, this time to Dr. Anthony Fauci and Jeffrey Zients (the White House Covid-19 Response coordinator), outlining all of this (that memo is at the bottom of this newsletter).
As of today. December 22nd, 114 out of 2,237 US counties that report data are either already at or are over 100% of their hospital capacity or are poised to reach that point soon. Of those, 62 counties are already known to be at or above hospital capacity, and another 52 have not yet reached capacity but will soon unless local case counts decrease, or capacity can be increased. These counties are spread across 22 states. So far no state as a whole has reached the circuit breaker threshold.
The dashboard comes with some key caveats and therefore should be used to trigger an on-the-ground assessment by officials. First, there are clearly some discrepancies between federal and local data with respect to hospital capacity. In some areas, it looks like HHS over-estimates the number of open hospital beds (because they seem to count some beds that could not actually be staffed to take care of Covid-19 or other medical patients). But there are likely other areas where HHS under-estimates capacity. In addition, staffing levels may change, especially if hospital workers are sidelined with new coronavirus infections. Also, as testing goes up, the usual rate of hospitalization may decrease. But our dashboard automatically pulls that data in, using an averages from the last 10 days or so in every jurisdiction.
Circuit breakers have many advantages:
• Circuit breakers are designed to achieve a narrow and achievable goal: identifying when the combination of local case rates and real-time hospital capacity indicates that hospitals have or will soon exceed their capacity, and therefore ability to provide even routine life-saving care.
•Circuit breakers shouldn’t take long to work (days, not weeks).
•Circuit breakers are not shutdowns. They are limitations in high-risk exposures, like indoor dining and other large indoor gatherings.
• Circuit breakers are designed to direct local action. Just because one county needs a circuit breaker does not need others do. So, they are far nimbler than our past responses.
• Circuit breakers do not merely rely on case counts but on hospital capacity. That means that circuit breakers do not need to happen in every place where case counts are rising, as long as the region has adequate hospital capacity to keep up.
•With our module/dashboard, we can anticipate whether hospitals can keep up and thus whether circuit breakers are likely to be needed.
As many have correctly said, this is not March 2020. We can track this virus, and the population has a lot more protection than it did, due to vaccines and prior infection. But that does not mean there is not extraordinary danger, due to the speed of Omicron, and the number of people who remain unvaccinated.
So the good news is that 87.5% of US residents ages 65 and up are fully vaccinated, and over half of them have been boosted. That’s important, because around 75% of all deaths have occurred in people 65 and older.
But the bad news is that 12.5% of US residents ages 65 and up are not fully vaccinated, which is over 7 million people.
Omicron will find those people because Omicron will find everybody.
If Omicron were to infect every single person ages 65 and over—not an unrealistic possibility—almost all of the critical Covid-19 hospitalizations and deaths would occur among the 7 million unvaccinated people. (I emphasize critical Covid-19, because the higher the percentage of the population that is vaccinated, the more people hospitalized for other conditions will happen to also have clinically less important or “incidental” coronavirus infections at the same time; we can’t let that fool us).
The big question we face, is what do we need to do now (other than the obvious usual things vaccination, boosters for those who need them, wearing good masks, using tests to limit contagiousness)?
Circuit breaker thresholds can help. Circuit breakers that rely on data-driven thresholds can allow local jurisdictions to anticipate hospital overload.
This is a scary moment but also one in which we are empowered in new ways. Never in human history have we been able to leverage real-time data in this way. Doing so will save lives and only temporarily inconvenience people in the places where that is absolutely necessary.
Remember, the vaccines work. If they didn’t, things would be far worse. But even so, Omicron may be a threat to community safety. That means that case counts might skyrocket, and a circuit breaker may not be needed in some areas, while in others even modest case count increases could spell trouble.
❓💡🗣️ What are your questions? Comments? Join the conversation below!
📬 Subscribe to Inside Medicine here and get updates from the frontline at least twice per week.
Our memorandum sent to Dr. Fauci and Jeffrey Zients (White House Covid-19 Response coordinator):