The advent of the novel coronavirus (COVID-19) has renewed the debate about the credibility of health information shared across social network sites, (SNS), such as Facebook. Despite these concerns, empirical studies into the reliability and accuracy of Facebook posts are largely limited. The objective of this study is to judge the accuracy and reliability of COVID-19-related information shared on Facebook using a mixed methodology. The authors extracted Facebook posts (text and embedded video links) in English from Buzzsumo between March 11, 2020, to April 31, 2020, using predetermined keywords. The first 50 posts pertaining to each keyword were extracted and analyzed using a one-way ANOVA, Tukey test for post hoc analysis, and modified DISCERN score. A qualitative content analysis was then conducted to contextualize the data. 196 posts were included in the analysis, with 24,084,564 shares and 3,436,275 comments. Useful posts had higher interactions compared to misleading and irrelevant posts (p < 0.05). The mean mDISCERN score for reliability was 3.2/5 for useful videos. The majority of the posts (56%) comprised positive appeals. Facebook is a widely utilized SNS with the potential of communicating accurate and reliable public health information if utilized efficiently by academic or research institutions and government health agencies.