Government briefings and news reports during the COVID-19 pandemic tend to lead with two statistics: the number of confirmed cases and the number of deaths.
Both numbers are important. But it strikes this observer that neither of these data points offers the public a good sense of the pandemic’s severity.
To explain why, it might be useful to contrast these headline numbers with those associated with a more common large-scale emergency.
Unless you’ve just moved to North Carolina, you’ve lived through a hurricane. As in the case of the current COVID-19 pandemic, each major hurricane that reaches this state generates regular briefings from the governor and members of his emergency response team.
Those briefings feature their own statistics. Chief among them tend to be the hurricane’s death toll and the number of power outages. Both numbers convey useful information.
The number of deaths associated with a hurricane in North Carolina rarely climbs above a few dozen. But each death reminds us about potential hazards associated with the storm. Along with rare reports of a person being killed by flying debris or a toppled tree, hurricane-related death tolls often include drownings, electrocutions, and other accidents.
The list typically does not include those who die in hospitals or nursing homes during the course of the storm. It does not list those who happen to lose a battle with cancer or other major disease as the high winds blow through the state.
Once you learn that a hurricane has caused deaths, you understand that it’s dangerous. It differs from a storm that skirts the coastline with no loss of life. The higher the death toll during a hurricane, the greater the sense that public safety might be jeopardized. That heightened sense of danger remains until trees are cleared, flood waters subside, and electricity returns.
Speaking of electricity, regularly updated power outage totals offer an even better indicator of the extent of a hurricane’s damage. The numbers and locations of outages offer valuable clues. A storm that knocks out electricity to 2,000 people in one county presents one kind of challenge. Different problems emerge when a storm knocks 500,000 people offline across a wide region.
Unlike the death toll, the number of power outages eventually goes down. Once the number drops, people start to get a sense that storm damage has peaked. It’s easy then to make a mental shift toward recovery mode.
North Carolinians familiar with these statistical signals cannot draw similar conclusions from the two COVID-19 headline numbers.
The total death count tied to the coronavirus tells us little about danger to the community at large. To do so, that death toll would have to distinguish among children, seniors, and working-age adults. It would have to contrast those with previously clean bills of health from those with underlying medical conditions. The count would have to separate those who died because of COVID-19 from those for whom the disease served as an aggravating, but nonlethal, factor.
Tell me that 200 North Carolinians have died from COVID-19, and I feel sadness and concern. But my approach to processing that news can vary widely.
It’s one thing if those 200 victims represent a random cross-section of people in the virus’s path. In that case, anyone could be threatened. Everyone should be wary.
The reaction changes if those deaths primarily involve a population of older, sicker residents who are more susceptible to infection. It’s still important to take precautions. But those precautions would focus on protecting vulnerable populations while allowing others to proceed with their lives.
The overall number of COVID-19 cases does little more to help us gauge our risks. For one thing, testing in the earliest days of the pandemic focused almost exclusively on people with severe symptoms. Even as testing expanded, most healthy people have avoided the process. It’s clear that the total number of cases has omitted some people who contracted COVID-19 without feeling especially sick. The tally also omits those who felt sick but never bothered to access a test.
Even if the total case number proved more reliable, that number never will decline. In other words, if North Carolina reports 6,000 cases, that number combines those currently fighting COVID-19 with those who have conquered it. The total case number tells us nothing about the number of people grappling with the disease — and trying to avoid infecting others — at this moment.
Unlike the number of hurricane power outages, we can’t look at the number of total COVID-19 cases for a sign that danger has peaked. The best we can do is hope that the rate of increase in total cases slows and eventually plateaus. Even then, the limits of testing will impact the data’s accuracy.
State Senate leader Phil Berger, R-Rockingham, and colleagues have highlighted the need for better data. Citing vocal national experts such as Stanford Medical School’s John Ionnidis, a professor of epidemiology and population health, Berger has led the charge for random testing. He has urged fellow state officials to gather the type of data needed to determine how COVID-19 is affecting the population at large.
The pandemic has produced plenty of data points. It will produce many more. Some will prove more useful than others.
Let’s not get stuck on the two headline numbers. They tell us little about what we need to know to move forward.
Mitch Kokai is senior political analyst for the John Locke Foundation.