A few commenters are dismissive of Oster as an economist commenting on health issues. Let's set aside the fact that "health across populations" has basically b

A few commenters are dismissive of Oster as an economist commenting on health issues. Let's set aside the fact that "health across populations" has basically been Oster's research focus throughout her career. There's a bigger issue here than the mistaken belief that she's not an expert on this subject. If the drive for credentialism means that we don't allow smart statisticians to review existing literature, then we've failed at whatever we're trying to achieve. I think I understand the drive for expert-only opinions. We've probably all been disappointed by misinformed scientific discussions, probably on some policy issue. Excluding opinions from random people on the street yields more productive discussions. After random people on the street, we might also exclude people in positions of irrelevant authority, like religious or political leaders. We might continue to exclude voices right up until we're just left with those who have actually conducted some research directly on topic. There comes a point, though, where by narrowing the field of voices you listen to, you're actually lowering the signal/noise ratio rather than upping it. In fact, if you can only talk to one person about scientific consensus, absolutely do not ask a scientist in that field. One researcher out of a field full of contentious debates is a terrible sample! If you can only talk to one person, start with a smart statistician who engages in lit reviews or meta-analysis. That way you have one conversation, but sample a much greater percentage of the field, and you get to hear it from someone who has some distance from the interpersonal debates going on in that narrow corner of academia. If you can only talk to one person about a topic, talk to someone like Emily Oster.