Doctors' and Nurses' Social Media Ads Reduced Holiday Travel and COVID-19 Infections: A Cluster Randomized Controlled Trial

thank Nikhil Shankar and Minjeong Joyce Kim for excellent research assistance. We are particularly grateful to all of the members of the “COVID-19 messaging working group” with whom developed and tested the original messages that led to this at-scale study. The research was supported by the National Science FoundationWe thank the health team at Facebook for their in-kind financial support that allowed us to run the campaign, and for their logistical help. Facebook provided the ad credits used to show the ads and connected the research team with AdGlow, the marketing partner. The ad content went through the usual internal policy review at Facebook for compliance with policies. Facebook had no other role in the design or conduct of the trial, and no role in the interpretation of the data or preparation of the manuscript. In particular, we thank Nisha Deolalikar. We also thank advisors Drew Bernard and Sarah Francis. We thank the team at AdGlow, in particular Camille Orellano and Lauren Novak, for running the campaign. We thank Alex Pompe from Facebook Data for Good for helping us to understand the Facebook mobility data. We thank the team at Damage Control, in particular Pradip Saha, for their tireless work in editing the videos. We (award 2029880), the Physician/Scientist Development Award (PSDA) granted by the Executive Committee on Research (ECOR) at Massachusetts General Hospital, and the National Institutes of Health (awards P30 DK040561, L30 DK118710). Primary outcomes were registered on ClinicalTrials.gov (NCT04644328) and the AEA RCT registry (AEARCTR-0006821). The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.

• supplementary appendix