Comparing bad apples to orange soda: Flaws and Errors in an Estimation of Years of Life Lost Associated With School Closures and COVID-19 deaths by Christakis, Van Cleve, and Zimmerman

We are writing this openly-published letter to express deep concerns regarding the paper recently published in JAMA Network Open: Estimation of US Children’s Educational Attainment and Years of Life Lost Associated With Primary School Closures During the Coronavirus Disease 2019 Pandemic https://doi.org/10.1001/jamanetworkopen.2020.28786 The paper by Christakis, Van Cleve, and Zimmerman (2020, abbrev. CVZ) is built upon multiple critically flawed assumptions, obvious misuse of the standard analytical tools, and clear mistakes in study design. Additionally, the analysis presented contains crucial mathematical and statistical errors that completely revert the main results, sufficient that if the estimates had been calculated according to the declared methodology, the results would completely contradict the stated conclusions and policy recommendations. These are not idle criticisms. This study has received enormous public attention, and its results immediately appeared in discussions of public health policies around schools worldwide. The central question is resolving an evidence base for the inevitable trade-off between (a) the very real harms of missed education provoked by policies that decrease viral spread vs. (b) the resumption of education as a social good which increases viral spread. This is an incredibly important public health question, and it demands careful cost-benefit analysis. To that end, this paper adds no usable evidence whatsoever.