Weaponizing statistics

When death becomes an aggregate abstraction, the truth either matters, or it doesn’t

Dave Kirby's father passed away on Feb. 9 from "acute respiratory distress" in Marin County, California. There's no way to know if he was one of the first COVID-19-related deaths in the United States.
Courtesy of Dave Kirby

“When one man dies, it’s a tragedy; when 100,000 men die, it’s a statistic.”

This quote is variously attributed to Joseph Stalin, Adolf Hitler or Napoleon Bonaparte. It bears more than a little relevance these days.

We live in a time when we are besieged by statistics. Number of infections, number of hospitalizations, number of ventilators or masks. Number of unemployed. Number of recovered. Number of deaths. 

Number of deaths.

Unaccustomed as we, and most of the rest of the world, are to grasping the scope and context of a global pandemic, we can extend ourselves and our need for understanding not much further than numbers, as the culprit is essentially invisible, thrives and moves according to its own whim and flourishes remorselessly, as life itself is wont to do. The numbers are the virus’ footprints in the snow, and to a large extent, all we can do is follow its tracks, estimate its path, anticipate its exhaustion, coax it somehow into a no-exit cul de sac and prod it toward its own extinction. 

In February, my elderly father passed away in Marin County, California. A few weeks after his 98th birthday, he took a fall at home while his wife was at work, and his daytime caregiver, responding to California law, called an ambulance over my father’s fruitless objections. He wasn’t hurt, not too badly anyway except for a slightly bruised arm, but they took him in, checked him over, kept him for a few days and then released him to a rehab center. They wanted to see him gain some strength in his legs, in the hopes he could avert, or at least significantly delay, another fall. 

While at the rehab center, he caught what was described as a fast-moving respiratory virus, was quickly transferred back to a hospital ICU, and was gone in a day. Acute respiratory distress. It was Feb. 9. 

He was 98 years old, a relatively healthy 98 at that; seldom sick, a little wobbly and inconsistently in command of his emotions and his memory, but to my own recollection, never seriously ill. Gave up smoking in 1952, drinking in 1955. A passionate golfer, until he had to give it up in his early 80s due to creeping immobility of his shoulders. He was a World War II veteran, a flight instructor (he taught the baseball great Ted Williams how to fly) and, by the time he left the military in late 1945, he was a captain in the Marines, at age 24. 

Doddering and feeble he was not. He was indifferent to the term “the Greatest Generation,” shrugged it off as a meaningless media creation, but he was one of its number. He couldn’t operate a home computer. He didn’t know who Mick Jagger was. An imperfect but decent and erudite man, full of humor and insight, a devourer of history, and a Roosevelt Democrat to his core. When they killed John Kennedy, he wept on and off for three weeks. 

By early March, the term “coronavirus” had ingrained itself into the news cycle, into our daily conversation, into our heads. Names and epidemiological terms we were unaccustomed to became commonplace. “Distance” became a verb. “Flattening the curve” sounded like a call to arms from a Richard Simmons video, but it had to do with the ICU. 

By early April, locked down in front of our televisions and laptop screens, working from home or not working at all, we settled into the routine of watching the rolling toll that the virus exacted on the population, on the economy, on living itself. All of it in numbers.

And then something happened. Health officials started digging into the numbers, prompted by a need to understand and potentially thwart the virus via its pathology and tracked infectiousness. Because so few people were being tested for the virus, but deaths were being recorded anyway as a matter of course, public health statisticians and researchers came to the realization that more people were dying from this thing, or dying from its proximate complications, than were originally recorded. 

In the aggregate, we know about how many people die every day, from disease or misadventure or lifestyle neglect or influenza or old age. Medicare, Social Security, insurance carriers access these numbers all the time. Hospitals supply themselves based on these numbers. Pharma companies determine much of their product to manufacture. It’s actuarial science. A blanket set of numbers that, more or less, provides a statistical fabric across which our collective mortality is tallied. Grim predictability. Some people spend their entire professional lives studying these things.   

And not just here, but in other countries ravaged by this spiky little ball, there were far more people dying than actuarial tables, appended by test-verified COVID-19 mortality, could explain. A lot more. The inescapable conclusion was that far more people were getting this thing, and far more were dying from it, than “official” tallies could narrate. Dying alone or dying in fear of (or lacking the means to access) hospitals. Dying because they couldn’t get tested, and thus treated. The mortality figures climbed. 

At this writing, in the United States, it’s about 83,000. The official number. The real number is higher — everyone who has been tracking this event and knows anything about actuarial science and epidemiology now knows this. 

Across the world, the statistics bear this undercount out. For the month between March 24 and April 24, the New York Times reported that the U.K.’s death count was more than 14,000 higher than normal plus confirmed COVID-19 mortalities. In France, from mid-March to mid-April, the difference was more than 6,000. In New York City alone, the difference in the period March 11 to May 4 was over 4,000 

But this is an election year. At the top of the ticket, we have an incumbent whose primary campaign platform was to plant his flag atop the “greatest economy in the history of the world,” now fending off criticism of inaction, executive dithering, cheap finger-pointing and unpersuasive magical thinking, in the face of a life-and-death, slowly fused health crisis. Plenty has been and will be written on President Trump’s behavior and performance during this event, and the point is not to litigate it here. 

But faced with the rising tally of dead Americans, a seemingly non-negotiable fact inherent in its cruelly binary nature (there is the living, and the no longer living), a cornered chief executive seeking re-election may be making the non-negotiable… negotiable. 

Axios reported recently that the death counts are the subject of ongoing internal conversation among Trump’s inner circle, who appear to be reinforcing Trump’s view, predictably based on little or no substantial evidence, that they are overstated. It’s hard to imagine this concern fading away, with the prospect of COVID-19 fatality statistics blaring across presidential election campaign ads. 

What will we start hearing? Blue state governors are padding the numbers to make Trump look bad? Deep state public health officials, embarrassed by shaky modelling early in the crisis, are now running up the score to validate their earlier predictions of mass mortality? How can you trust these obviously politically motivated statisticians? There’s… an agenda at work here. 

The perverse twist is that while a president running for re-election and trying desperately to inoculate himself from the fallout of this thing, to whatever degree the voters hold him accountable, is trying to deflate the statistics coming from suddenly politically motivated institutions like Johns Hopkins or the World Health Organization or the CDC or state health agencies, the real number is probably higher than reported, since many of the virus’ victims are in the ground, unautopsied, cremated or otherwise untested and unavailable for examination. All we’re left with is the numbers. 

Such was the case with the 1918 Spanish Flu pandemic. Historians are certain the death count, here and especially abroad, was likely much higher than the staggering numbers tossed around these days. Actuarial science, in the absence of computers and centralized data collection, was barely in its infancy at the time.

A high school friend with whom I reconnected recently through social media — a practicing, published and highly regarded physician in Massachusetts — has been posting about the progress of this pandemic. He was the first, to my eyes, to identify the benefit of anti-coagulants for certain severely ill COVID-19 patients. He follows and posts frequently on clinical studies. He was an early skeptic of hydroxychloroquine when Trump’s supporters in the media were still trumpeting it as our redeeming elixir. He also tested positive for antibodies, though never felt sick. His wife and his son both tested negative. 

I asked him about my father, especially after hearing the news that the earliest probable fatality from COVID-19 in California was February 6. In northern California…where my father lived. I have no idea how many medical or rehab personnel came in contact with him. Likely a couple of dozen, or more. 

“Better than even chance that was COVID, IMHO,” he replied.

According to KQED, Dr. Scott Morrow, the health officer for San Mateo County, wrote in an email in the last week of April, “I told my colleagues in mid-January that it was likely spreading under our noses.” 

My father’s wife has since tested positive for antibodies, and she was sick with body aches and a raging cough for much of January, only being tested well after the fact. And a week or so after the exchange with my physician friend, I learned that my half-brother had flown out to California from New York City to visit my father less than two weeks before his death. He was sick in early April, although never tested. His toddler son was tested and confirmed positive for antibodies. All have since recovered.    

Did my father get it from the rehab center? From his wife? Did his son bring it from New York? Did he infect my father, or vice versa? Did my half-brother get it on a plane, or from his mother? Or his since deceased father? Or his son?  

We don’t know, and we won’t. The statistics will say that my stepmother and half-brother recovered from it. But thus far, they won’t reflect that my father died from it. 

In the mid-1990s, I had the opportunity to interview James Burke, the renowned science historian and BBC correspondent, who wrote and hosted the original Connections television series (1978) that aired on public TV here in the States. Burke was on a speaking tour, to promote a book or his followup series… I can’t remember which, if not both.

In our chat, and in the speech he gave at Denver’s Paramount Theatre, he talked expansively and energetically about the coming information revolution. The advent of the internet, where, quite literally, every person on Earth would have access to the sum total of the human experience. History, literature, science, art, philosophy, religion. Everything that could be rendered digitally would be available to everybody. It would be a global transformation of incalculable proportion. The world, the entirety of humanity, would become smarter, healthier, more productive. Presumably, more prosperous. Everybody could know everything. 

It kind of worked out like that, and kind of didn’t, of course. After his speech, during his Q&A, I raised my hand. All this sounds terrific, but at some point you’re relying on centralized control — technology, administration… content. What if Goebbels had the internet, free to craft his party’s version of history. Stalin? Mao? How do you keep content unbiased, factual, dispassionate, unmoored from institutional self-interest? From a distance, the question seems quaint and naïve. 

Burke brushed it aside. “I have faith,” he said, “in the crusade of the ‘hacker.’ The guy who will sniff out disinformation and fix it, obscure it, inoculate us from it. There have always been these people.”

Burke’s faith in the benign inevitability of truth was persuasive at the time, but as we have come to see almost three decades later, sadly over-optimistic. Hackers are at least as much malign as benign. Facts on a digitally rendered screen are every bit as fungible as they were before the internet, and only as reliable as the perceived credibility of their source. When you’re dealing with the abstraction of mortality figures, who prints them — or who fiddles with them — matters. 

A lie can travel around the world in the time it takes the truth to pull up its pants. Mark Twain is credited for that remark, but it could have been Goebbels. 

Tell a lie three times, and it becomes the truth. Hitler said that. 

It doesn’t matter how many caskets there are. It only matters who’s counting them. Now, go get that haircut. 

My father’s late March memorial service, which was to be attended by family members from two countries and at least seven states, was canceled, likely not to be rescheduled. 

He’s a statistic now. Or not. 

Does it matter?