A meta-analysis by Fell and Scherer concludes that lowering the legal blood-alcohol-concentration (BAC) limit for driving from 0.08 to 0.05 grams per deciliter (g/dl) in all U.S. states would save 1,790 lives per year. This conclusion is based on a combination of results from 37 studies evaluating a variety of BAC-related policy changes across various locations in different time periods. All of these studies use observational (i.e. non-randomized) data, which leads to concerns about disentangling correlation from causality that are not convincingly addressed. Moreover, while these studies examine some sort of BAC policy, very few actually investigate a reduction from 0.08 g/dl to 0.05, and the few that do examine policies implemented more than two decades ago in countries besides the U.S. Given the lack of high-quality, directly relevant evidence, the findings of Fell and Scherer should be interpreted with considerable caution.