Information about anthrax released by news media from October 4 to December 3, 2001, was identified, sampled, coded, and compared with information released by CDC during that period using statistical analysis. In addition, communications about two anthrax-related issues were examined in depth. The quantitative analysis showed that, overall, CDC information releases and news coverage tracked fairly closely. When weight was defined as number of mentions, both sources gave the same weight to reports of risk for the population. The news sample gave roughly half the weight as CDC to who was exposed, how people were exposed, and what role antibiotics play in preventing anthrax. The samples were widely divergent (CDC high, news sample low) for public health precautions and other details. The in-depth, qualitative analysis showed that some reporters misinterpreted information provided by CDC, but they responded to requests to clarify the issue. The findings of this study suggest ways to improve future crisis communication efforts and demonstrate how differing methods of analysis can yield substantially different conclusions.