Effects on local emergency departments of large-scale urban chemical fire with hazardous materials spill.

Introduction: On 18 July 2001, a train hauling hazardous materials, including hydrochloric acid, hydrofluoric acid, and acetic acid, derailed in the city of Baltimore, Maryland, resulting in a fire that burned under a downtown street for five days. Firefighters were stymied in their efforts to extinguish the fire, and the city was subjected to thick smoke for several days.

Objectives: To determine whether an urban chemical fire with a hazardous materials spill resulted in a detectable public health impact, and to demographically describe the at-risk population for potential smoke and chemical exposure.

Methods: The United States Centers for Disease Control and Prevention (CDC) was consulted about possible side effects from chemical exposure. Total numbers of emergency department (ED) patients and admissions from 15:00 hours (h), 15 July 2001 to 15:00 h, 21 July 2001 were collected from five local hospitals. Patient encounters citing specified chief complaints from 15:00 h, 15 July to 15:00 h, 18 July (pre-accident) were compared with the period from 15:00 h, July 18 to 15:00 h, 21 July (post-accident). Data were analyzed using Fisher's exact test. The United States Census Bureau's Topologically Integrated Geographic Encoding and Referencing (TIGER) digital database of geographic features and ArcView Geographic Information Systems (GIS) were used to create maps of Baltimore and to identify populations at-risk using attribute census data.

Results: There were 62,808 people residing in the immediate, affected area. The mean of the values for age was 33.7 +/- 3.2 years (standard deviation; range = 16 yrs) with 49% (30,927) males and 51% (31,881) females. A total of 2,922 ED patient encounters were screened. Chief complaints included shortness of breath, pre-event = 109 vs. post-event = 148; chest complaints = 90 vs. 113; burns and/or skin irritation = 45 vs. 42; eye irritation 26 vs. 34; throat irritation = 33 vs. 27; and smoke exposure = 0 vs. 15. There was a statistically significant increase (p < 0.05) for shortness of breath and smoke exposure-related complaints. No statistically significant increase in numbers of admitted patients with these complaints was found.

Conclusions: In the setting of a large-scale urban chemical fire, local EDs can expect a significant increase in the number of patients presenting to EDs with shortness of breath and/or smoke inhalation. Most do not require in-patient hospitalization. Careful assessment of impact on local EDs should be considered in future city-accident planning. Some official warnings were widely misinterpreted or ignored. Public education on potential hazards and disaster preparedness targeted to populations at-risk should receive a high priority. Geographic information systems (GIS) may serve as useful tools for identifying demographics of populations at-risk for disaster planning and responses.