Twitter is becoming a popular social media platform for syndromic surveillance. We explored the Twitter discussion in the context of the 2016 Zika Outbreak. The Zika virus can have severe long-term (such as microcephaly in newborns) effects and not so serious immediate (such as fever or headache) effects. We explored whether social media discussions effectively capture the severity of these long-term concerns. We performed volumetric and text mining analysis, such as the co-occurrence of words and hierarchical clustering, to explore the different underlying themes in the Twitter discussion regarding the immediate and long-term concerns. Our findings suggest that the concerns related to the long-term consequences are dominant and consistent, but this is not the case for the immediate effects. © 2016 IEEE.