If you look at a map of coronavirus cases in the US, it can be hard to figure out any patterns. California reported some of the first coronavirus infections — including the country’s first instance of community transmission — but New York has reported 12 times as many deaths.
Michigan, meanwhile, has the third-highest number of coronavirus deaths in the US, but neighboring Ohio doesn’t even register in the top 10.
Florida was slow to shut down in response to the pandemic, but so far it’s reporting fewer deaths than nine other states, many of which were quicker to act.
Why? What explains why some states become hot spots for Covid-19, the disease caused by the SARS-CoV-2 coronavirus, but not others?
I turned to experts with these questions. While they cautioned that there’s a lot about the coronavirus and how pandemics work in modern society that we don’t know and are still learning about, they gave two major explanations for why some areas become hot spots and others don’t.
First, sometimes it just comes down to chance. Some places just happen to have populations, like the elderly and those with underlying medical conditions, that are more vulnerable to Covid-19. Factors beyond an area’s immediate control, such as its population density and perhaps even its weather, can also contribute to the virus’s spread. Whether any given place experiences a super-spreading event, in which one or more individuals transmit the coronavirus to a disproportionate number of people, can partly come down to chance, too.
Second, early action did appear to prevent coronavirus cases. Even in states that are not suffering a high number of Covid-19 deaths, chances are those figures would be even lower if they — or if the country as a whole — had acted sooner. There’s good evidence for this in some states, but there’s also research of past pandemics to back it up.
“A hot spot is a reflection of the combination of the random nature of things — in terms of who gets hit harder earlier — and then the timing of what we call non-pharmaceutical interventions” such as social distancing, William Hanage, an epidemiologist at Harvard University, told me.
The key here is that states have direct control over one of these two factors. They can’t do much about luck. They have little ability to predict whether someone infected within their borders happens to spread the virus to a lot of people, and little control over their local weather or population density. But they can take swift, aggressive action to mitigate their chances of an outbreak — to limit the risk, for example, that someone becomes a super-spreader or that people are gathering in very dense crowds.
In other words, some of this comes down to uncontrollable chance. But to the extent that it doesn’t, different places need to be ready to take aggressive action and sustain it as long as necessary.
Chance plays a role in Covid-19 outbreaks
Some of the major contributors to a big, deadly outbreak are issues that are, at least for now, largely out of any given place’s control (even if past years of policy contributed to them). “There is a role for luck,” Hanage said.
For example, the age and general health of a population can contribute to just how bad a major outbreak gets, with older and less healthy populations more likely to become seriously ill and die from Covid-19. Italy has suffered one of the deadliest coronavirus outbreaks in the world — with a death rate of more than 10 percent among confirmed cases — and one potential explanation for that is it has the second-oldest population in the world.
Cities, states, and countries could perhaps have taken some steps prior to Covid-19 to shield their more vulnerable residents from infectious disease — by, for example, providing better health care or elderly care services. But once the coronavirus hit, places had to deal with the realities on the ground.
There are other factors outside of any particular place’s immediate control, too.
Timing is a big one. If a place is among the first hit by a new disease outbreak, it’s going to have fewer examples to learn from in figuring out what to do. States now look to New York as an example of things going very wrong, but that was possible only because New York got hit hard before most of them. Whether a place is among the first hit is largely up to chance or, at least, variables largely outside of any government’s control.
Places with higher population density are likely more vulnerable to the spread of the coronavirus. Places that rely on packed public transportation are likely more vulnerable, too. It’s possible that colder weather can make the virus easier to spread, as is true with some other viruses. (New York City, perhaps not coincidentally, checks all of these boxes as the densest city in the US, with the highest rates of public transportation ridership, and relatively cold weather.)
Whether a place gets hit by a super-spreading event can also come down to chance. There are things that governments and members of the public can do to mitigate the chance of such an event, like trying to keep people at home, restricting travel, and stopping large gatherings.
But there are some things that are harder to control. Perhaps someone spreads a virus widely before she’s symptomatic and before it’s known that the community is having a problem with that virus. Maybe some people are just more infectious for reasons we don’t yet know. Some communities can have infected people travel to them at the wrong time, widely spreading a disease.
It’s kind of like rolling a die. A die can be loaded in all sorts of ways, which vary from place to place and time to time, to prevent a bad roll that leads to a massive outbreak. But there’s always a chance that the roll goes very, very wrong — maybe the one person in an otherwise well-behaved town who disobeys a stay-at-home order turns out to be highly infectious despite showing no symptoms, launching a super-spreading event.
With the coronavirus, there’s also a lot we don’t know, from exactly how and where the virus most often spreads to just how deadly Covid-19 truly is. There could be variables affecting coronavirus outbreaks that we don’t even know exist. That makes outbreaks of the virus less predictable, leaving more of the situation to chance than would otherwise be true.
Early action and vigilance matter
While chance plays a role in any place becoming a hot spot, that doesn’t mean countries, states, and cities are powerless — far from it. Experts say that during a big outbreak, governments should do what can feel like an overreaction: The big goal is to prevent things from getting bad, so once policymakers are reacting to a bad outbreak, they’re already acting too slowly.
“Any time you are dealing with an outbreak, if it appears like you overreacted, then you probably did the right thing,” Krutika Kuppalli, a fellow in the Johns Hopkins Center for Health Security Emerging Leaders in Biosecurity program, told me.
To put it another way, states have to do as much as they can, even if it feels like too much, to load the die so that people aren’t spreading the disease to others.
Consider the stories of California and New York. Despite reporting some of the first Covid-19 cases in the US, California has avoided a massive outbreak. Again, some of that could come down to chance or uncontrollable variables, such as New York City’s higher population density and use of public transit, or California’s generally warmer weather.
But one likely contributor is that California, particularly the San Francisco Bay Area, reacted quicker to the outbreak. The Bay Area issued the first shelter-in-place order in the US on March 16, and California issued a statewide stay-at-home order three days later — while New York didn’t impose its own mandate until March 22. Even before these orders, some parts of California seemed to adopt social distancing early: OpenTable data suggests that seated dining on March 1 was down just 2 percent in New York City, but 18 percent in San Francisco. (It was down by only 3 percent in Los Angeles, though, so not every city in California acted the same.)
The extra days and weeks may not seem like that much time. But when coronavirus cases and deaths can double every few days, that short time span is important. “It’s exponential,” Hanage said. “And if you’re too late with it, you’re making the problem much, much, much worse.”
That’s especially true because the coronavirus can spread without any clear indication. The research increasingly shows that people can transmit the virus before they develop any symptoms, and regardless of whether they develop symptoms at all. Given the continued lack of testing in the US, that can mean the coronavirus is already spreading before people are aware it’s in the community — and it also might mean some places are underestimating just how bad their outbreaks are, potentially hiding hot spots before they become too bad to ignore.
“By the time that you start seeing cases show up in the hospital, there’s already a fair amount of transmission in the community,” Natalie Dean, a biostatistics professor at the University of Florida, told me. “If you wait until that happens and then you clamp down, at some level it’s already going to be out of control at that point.”
To that end, it’s important for states to act not only earlier on the calendar but also earlier in terms of their overall outbreak. For example, Ohio’s and Michigan’s stay-at-home orders took effect with only a two-minute difference, but Ohio’s order came when it was reporting around 100 new cases a day while Michigan’s came when it was reporting 500-plus cases a day. So Ohio’s order was likely more effective at preventing a bigger outbreak — and that could help explain why Ohio has one-fifth as many Covid-19 deaths as Michigan, despite having a slightly larger population.
One potential counterpoint to the need for early action is Florida. The governor there was slow to take action, not issuing a stay-at-home order until April 1. Yet the state has yet to see a huge outbreak, falling outside the top 20 for confirmed coronavirus deaths per 100,000 people.
But there’s another explanation: Florida is much more than its state government, and other parts of the state took relatively early action. For example, some cities, such as Miami and Orlando, had stay-at-home orders in place in the weeks before the state took action. There’s also evidence that Floridians started to embrace social distancing before officials demanded it: OpenTable’s restaurant data shows that seated dining in the state started to fall in March and was collapsing, with drops of 20 percent or more, by the second week of the month.
There’s also evidence supporting the efficacy of early action from the 1918 flu pandemic, which was linked to as many as 100 million deaths globally and about 675,000 deaths in the US. A 2007 study in PNAS found that the places that took quicker action on social distancing — closing schools and banning big public gatherings — saw better outcomes:
[C]ities in which multiple interventions were implemented at an early phase of the epidemic had peak death rates ≈50% lower than those that did not and had less-steep epidemic curves. Cities in which multiple interventions were implemented at an early phase of the epidemic also showed a trend toward lower cumulative excess mortality, but the difference was smaller (≈20%) and less statistically significant than that for peak death rates.
One example cited in the study is the difference between Philadelphia, which was slow to act, and St. Louis, which was faster. As this chart shows, St. Louis did a much better job of flattening the curve and averting excess deaths:
The goal with disease outbreaks is to look less like Philadelphia and more like St. Louis. And the key is early, aggressive action.
States can’t control luck, but they can be vigilant
Experts are clear that, even though some elements may be outside of any city, state, or country’s immediate control, no one should rely on luck to get them by this crisis.
Instead, experts cautioned that everyone should assume coronavirus will eventually come to them. The reality is that in a full-blown pandemic the risk is so widespread that it’s likely a matter of time, not chance, before an infection hits a place not taking the right precautions. “Even if you’re not rolling the dice now,” Hanage said, “you’re almost certainly going to be rolling them before this is over.”
Experts especially emphasize the need for much, much more testing. While America has built greater testing capacity since March, progress appears to have stalled out in recent weeks at around 150,000 tests a day — far from the 500,000-plus or even millions of tests nationwide that some experts argue is necessary. Even that might overstate how well some states are doing; while New York has among the highest testing rates in the country at 32 per 1,000 people, Florida has less than half that at 12 per 1,000 and Texas has even less at 6 per 1,000.
Testing remains crucial to addressing the crisis. It gives officials the means to isolate sick people, track and quarantine the people whom those verified to be sick came into close contact with (a.k.a. “contact tracing”), and deploy community-wide efforts if a new cluster of cases is too large and uncontrolled otherwise. Without it, the only way to deal with the outbreak is more social distancing, which further hurts the economy, or letting the disease run its course — at the cost of potentially hundreds of thousands or millions of lives.
The recognized need for more testing is one of the major reasons much of the US is social distancing right now. “The whole point of this social distancing is to buy us time to build up capacity to do the types of public health interventions we know work,” Dean said. “If we’re not using this time to scale up testing to the level that we need it to be … we don’t have an exit strategy. And then when we lift things, we’re no better equipped than we were before.”
To fix the gaps, experts argue, the federal government needs to relax criteria for testing, invest in new supplies and labs, and better coordinate supply chains to address, among other issues, chokepoints. States, with limited resources and little control of the national supply chain, simply can’t do this all on their own.
Such fixes, however, are easier said than done. Experts warn the next phase of testing will be much more difficult than the initial phase, which largely required getting existing labs to start doing coronavirus testing — the low-hanging fruit.
“We’ve made substantial progress ramping testing this month,” Scott Gottlieb, a former commissioner of the Food and Drug Administration, wrote in a tweet on April 10. “But many gains were made by getting players into fight (clinical labs, academic labs). Now we must expand lab capacity, platforms, throughput, test kits. Getting next million tests/week will be harder than getting first.”
Until that work is done, states will need need to remain cautious. While some of the Covid-19 risk is due to uncontrollable factors, aggressive early actors like California, Germany, and South Korea show the risk can be mitigated. It’s important cities, states, and countries don’t leave this all to chance.