Background: Misinformation on interactive Knowledge Exchange Social Websites (KESWs) is concerning since it can influence Internet users’ health behaviors, especially during an infectious disease outbreak. Objective: The present study seeks to examine the accuracy and characteristics of health information posted to a Knowledge Exchange Social Website (KESW). Methods: A sample of 204 answers to Ebola questions were extracted and rated for accuracy. Multiple logistic regression modeling was used to examine whether answer characteristics (best answer, professional background, statistical information, source disclosed, link, and word count) predicted accuracy. Results: Overall, only 27.0% of the posted answers were rated as “accurate”. Accuracy varied across question topics with between 11.8% - 45.5% of answers being rated as accurate. When Yahoo Answers’ “best answers” were examined, the overall accuracy was substantially higher, with 80.0% of “best answers” being rated as accurate compared to 16.0% of all other answers. Conclusion: There is need for tools to help Internet users navigate health information posted on these dynamic user-generated knowledge exchange social websites.