No, Dr. Oz Didn't Say Only 2–3% of Children Would Die of Coronavirus If Schools Reopened

Dr. Oz deserves criticism, but he was clumsily referencing a real—and actually encouraging—scientific study.


During his many years as a celebrity doctor, Mehmet Oz has promoted a variety of medically unsound products and causes. When he offers advice on how the government should handle the coronavirus pandemic, it is reasonable to be skeptical.

So I was initially inclined to side with the commentators currently criticizing Dr. Oz for remarks he made recently on Fox News. Unfortunately, his critics have misquoted him, wrongly making it sound like he was recommending schools be re-opened despite the possibility of mass death among students. Oz should have chosen his words more carefully, but those who are dunking on him for this are—intentionally or not—promoting misinformation, because it turns out that there is some evidence that schools could be re-opened even as COVID-19 rages.

On his Wednesday night show, Sean Hannity asked what should be done to get the economy going again. Oz replied: "Schools are a very appetizing opportunity. I just saw a nice piece in The Lancet arguing the opening of schools may only cost us 2–3 percent in terms of total mortality. Any life is a life lost but to get every child back into a school where they are safely being educated, being fed, and making the most of their lives, with the theoretical risk on the back side, might be tradeoff some folks would consider."

All your favorite #Resistance blue checkmarks leapt to denounce Dr. Oz, and Fox by extension, for suggesting that the deaths of millions of children could ever be considered an acceptable tradeoff. (The phrase "only 2–3%" was trending on Twitter.)

But Dr. Oz was not describing a death toll in the millions. He said the cost could be "2–3 percent in terms of total mortality," not among all school-aged children or the population at large.

That Lancet article argues that school closures may not be a particularly effective social distancing measure. It cites modeling from Imperial College London that estimates the U.K.'s school closures will reduce overall deaths by about 2–4 percent. In other words, if there are 100,000 COVID-19 deaths despite the schools being closed, having had the schools open anyway would have yielded 2,000–4,000 additional deaths. That's thousands, not millions.

Meanwhile, the article notes that many children depend on school not just for education but for supplemental nutrition. "Following school closures amidst the west African Ebola epidemic, rates of child labour, neglect, sexual abuse, and adolescent pregnancies spiked, and many children never returned to school," the article's authors write. It is possible, they thus conclude, that keeping schools closed would do more harm on net.

The issue is not settled. The Lancet writers are relying on a model, and models can be wrong. But it's a helpful piece of information, and Dr. Oz was not countenancing the deaths of millions of children when he referenced it.