I hate it when journalists/reporters do not link to the actual study. It looks like the related publication is https://doi.org/10.1016/j.trc.2019.02.020