Your article, Experts Debate Using Structure to Resist Fire Loads on a Structure, includes a table of statistics on relative risk levels under fire and other events, which is attributed to me (ENR 10/13 p. 14). My intent in presenting these and other statistics at the NIST/SFPE workshop was to provide some sense of the relative safety implied by current structural design practice, recognizing that the specific limit states and their consequences are not equal across all the hazards. The main point, made correctly in the article, is that compared to other hazards, the risk of structural collapse for modern office buildings is quite low, due to the low likelihood of fully developed fires and to the effectiveness of well maintained active and passive fire protection measures.
That said, the nuclear power facilities entry on the table was misinterpreted and requires clarification. First, the reported risk statistics for nuclear power facilities should refer to the risk of reactor core damage, a less serious risk than meltdown. Furthermore, the mean annual probability (MAP) of 10-4 refers to the target probability of damage to the core from all possible causes, and the MAP of 10-5 corresponds to the core damage risk from earthquakes (ref. ASCE-4 Seismic Analysis of Safety-Related Nuclear Structures and Commentary and related standards).