Op-Ed: Without context, COVID tallies are misleading
Since the start of the pandemic, public health authorities have been fastidiously counting the number of people infected with the coronavirus. For both the medical profession and the media, these rising figures have been the principal way of framing the pandemic in the U.S.: “124,000 new cases a day,” “802,000 COVID deaths since February 2020.” But this information offers an incomplete picture of the crisis, potentially warping the public’s understanding in ways that could prolong the pandemic and even add to its toll.
What’s missing from the day-to-day conversation is the number of uninfected people and the number of infected people who survive COVID-19. That provides a denominator to put the other figures in context. If there were 124,000 new infections per day, how many people were exposed? If 802,000 people died from COVID, how many were infected but didn’t die?
Indeed, such information is the most underreported story of the pandemic. But it has long been an important piece of public health information. It advances our understanding of the nature of the disease; it hints at the power of precautions such as masks and vaccines; and it can allay fears and trauma that people are experiencing about the seemingly never-ending nature of the pandemic.
Our reliance on numbers to understand epidemics can be traced to the development of epidemiology — when medical and scientific authorities had not yet uncovered how microbes caused the spread of infectious disease. Between 1755 and 1866, when epidemiology emerged, medical practitioners believed that environmental factors caused disease. Based on this inaccurate view, they had few effective metrics to understand the origins of epidemics. As such, they counted the number of uninfected and infected patients; the number who contracted a disease and the number who died; they examined those who were hospitalized and those released.
Counting was a way to rationalize infectious disease and to create a narrative about it. For example, during the Crimean War in the 1850s, the nurse and statistician Florence Nightingale witnessed that more British soldiers died once they were admitted to the hospital, but she couldn’t see the germs that were infecting them. What she could see, she counted: the number of healthy and the number of sick soldiers, inside and outside hospitals. By creating a clear analytical assessment, she then observed how the unsanitary conditions within hospitals correlated with alarming mortality rates. According to Nightingale, a “complete system of sanitary statistics in the army” was necessary “to administer the laws of health with that certainty.”
Statistics, and exploring the behaviors behind them, became a key component in epidemiological analysis because that’s all that health experts had — and it helped them craft treatment strategies.
In response to a cholera outbreak in Calcutta, known today as Kolkata, William Twining, a British military doctor there, published an influential comprehensive volume on diseases in 1832. The treatise provided copious detail of hospital attendants who came into close contact with cholera patients and soiled linens but did not become ill. Had the text focused solely on people who became sick, a reader might have been misled about the risk of the disease, or led to look for its causes in the wrong place. With context about the unafflicted, the study offered key evidence that cholera was not transmitted through direct contact.
It was another set of counterexamples two decades later that helped the young science of epidemiology to zero in on the culprit. John Snow, a physician in London, famously found the common denominator among cholera cases in an 1854 outbreak: Those who became sick seemed to all have drunk water from a pump in the center of a poor neighborhood. Cementing his conclusion was the fact that employees at a nearby brewery, which had its own pump, did not contract cholera.
Learning about the daily lives of these brewery workers led Snow to theorize that cholera was transmitted through contaminated drinking water. To understand how a disease spread, he was equally invested in the infected and the uninfected.
As epidemiology evolved as a field, medical authorities continued to consider the uninfected by developing a new statistic: incidence rate or attack rate, which is still used today. This refers to the number of new infected cases within a specific period measured against the population. While epidemiologists tabulate this rate, the media does not typically broadcast it. Instead, we are inundated with the crude morbidity and mortality (infection and death) rates.
In short, reporting the number of infected offers a numerator but we are missing the denominator. We need a clearer empirical accounting.
A recent example shows why the missing denominator is important: This past summer, the media jumped on one of the first major outbreaks of breakthrough cases in Provincetown, Mass. This provided epidemiologists with valuable evidence of how the Delta variant infected many vaccinated people — but no one actually counted the number of people who were exposed but not infected. (To be fair, documenting exposure among uninfected people is more challenging than counting sick people, as is finding infection among asymptomatic people.)
By focusing on the vaccinated who became infected, the media inadvertently gave the impression that the Delta variant had superpowers. If it is super, it also has a weakness: the vaccines. That’s the picture that emerges if one counts the uninfected and looks at vaccination rates. A narrower focus risks overplaying the danger of the variant and underplaying the value of the vaccines. Epidemiology needs to remember its roots and school the public.
The first generation of epidemiologists were first and foremost storytellers. Without complicated modeling, or much by way of accurate aggregate data, narrating epidemics was at the center of the field, as historian Jacob Steere-Williams explains. Reclaiming this tradition and telling a more complete and nuanced narrative of COVID-19 — using modern data science as well — can help us better understand the virus and make better choices, such as getting vaccinated.
By focusing only on a rising tide of infections and deaths, we veil more of the pandemic than we reveal.
Jim Downs, Gilder Lehrman-NEH professor of history at Gettysburg College, is the author of “Maladies of Empire: How Colonialism, Slavery and War Transformed Medicine.”
A cure for the common opinion
Get thought-provoking perspectives with our weekly newsletter.
You may occasionally receive promotional content from the Los Angeles Times.