1 table omitted
On January 6, 2005, two freight trains collided in Graniteville, South
Carolina (approximately 10 miles northeast of Augusta, Georgia), releasing
an estimated 11,500 gallons of chlorine gas, which caused nine deaths and
sent at least 529 persons seeking medical treatment for possible chlorine
exposure1,2; South Carolina Department of Health and Environmental
Control [SCDHEC], unpublished data, 2005). The incident prompted the Agency
for Toxic Substances and Disease Registry (ATSDR) to review data from its
Hazardous Substances Emergency Events Surveillance (HSEES) system and update
an analysis of 1993-1998 railroad events.3 The HSEES system is
used to collect and analyze data concerning the public health consequences
(e.g., morbidity, mortality, and evacuations) associated with hazardous-substance–release
events* that occur in facilities or during transportation. This
report describes the event in South Carolina, which is not part of the HSEES
system, and two others from HSEES, and summarizes all rail events reported
to HSEES from 16 state health departments† during 1999-2004.‡
Local government agencies, employers, and first responders can help reduce
morbidity and mortality from transit-associated hazardous-substance releases
by examining historical spill data for planning purposes, developing emergency
response plans, undergoing proper hazardous materials (HazMat) training, and
reviewing epidemiologic investigation data.
South Carolina. At approximately 2:40 a.m.
on January 6, in Graniteville, South Carolina, a freight train with three
chlorine tanker cars and one sodium hydroxide tanker car collided with a train
parked on an industrial rail spur. The collision caused a breach in one chlorine
car, which resulted in the immediate release of an estimated 11,500 gallons
of chlorine gas. As a result, nine persons died, and at least 529 persons
sought medical care. Because exposure to high levels of chlorine can result
in corrosive damage to the eyes, skin, and respiratory tissues and lead to
pulmonary edema and, in extreme cases, death,5 local emergency
management officials initially issued a shelter-in-place order for a 1-mile
radius around the site until 4:30 p.m. At noon, South Carolina declared a
state of emergency, giving local authorities responsibility for issuing a
mandatory evacuation for the 5,453 residents within the 1-mile radius. Area
schools and businesses were closed. Four days later, an operation to patch
the leaking chlorine tank car succeeded by applying a temporary repair.2 Federal responders from ATSDR, the U.S. Environmental Protection Agency
(EPA), and the U.S. Coast Guard arrived to assist SCDHEC in sampling air in
factories, homes, and schools within the 1-mile radius.
A rapid epidemiologic assessment determined that, of the 511 persons
examined in emergency departments after exposure to chlorine gas, 69 were
hospitalized in seven area hospitals. An additional 18 persons were treated
at area physician offices. An ongoing assessment is examining the public health
impact associated with exposure to chlorine gas. Those exposed are being interviewed
about their symptoms, the location and duration of the exposures, and demographic
information necessary for monitoring any long-term health effects and psychosocial
Texas. In June 2004, a moving train struck
a stationary train at a rail substation, causing a derailment. One tanker
car was punctured, releasing approximately 90,000 pounds of chlorine gas.
At least 60,000 pounds of chlorine reacted with sodium hydroxide to form sodium
hypochlorite. Also released were approximately 78,000 gallons of urea fertilizer
and 7,000 gallons of diesel fuel. Forty-four persons were injured, including
three who died. The train conductor died from trauma sustained during impact,
and two elderly residents near the site died from chlorine inhalation. Of
the remaining 41 injured, 22 were members of the general public, 13 were employees,
and six were first responders. The most frequent injuries were respiratory
and eye irritation. The majority of those injured (22 [54%]) were treated
at a hospital and released, 12 (29%) were treated on the scene, and seven
(17%) were treated at a hospital and admitted. Nearby residents initially
were ordered to shelter-in-place while a site assessment was conducted. Later,
evacuation of 45 residents for 13 days was ordered when the company prepared
to unload the chlorine car. Responding to the event were a certified HazMat
team; railroad response team; EPA response team; teams from the National Transportation
Safety Board and Federal Railroad Administration; and local health, environmental,
fire, law enforcement, and emergency medical services (EMS) personnel. Twenty
railroad employees and 80 first responders were decontaminated after responding
to the event. The cause of the derailment was determined to be human error
(i.e., failure to stop).
Missouri. In August 2002, approximately 16,900
pounds of chlorine gas were released from a railroad tanker car when a flex
hose ruptured during unloading at a chemical plant. An automatic shut-off
valve on the car and an emergency shut-off system at the plant failed to work
as back-up prevention measures. Sixty-seven persons were injured: 61 members
of the general public and six employees. The most common injury was respiratory
irritation. Sixty-five (97%) of the injured were treated at a hospital and
released; two (3%) were admitted. Approximately 400 nearby residents were
evacuated for 7.5 hours; the release was stopped and contained through the
efforts of a certified HazMat team; company response team; EPA response team;
and law enforcement, fire, EMS, and local environmental personnel.
Of the 49,450 events reported to HSEES during 1999-2004, a total of
12,845 (30%) were transportation related; of these, 1,165 (9%) were rail events.
Fifteen of the 16 HSEES states reported rail events, with Texas (249 [21%]
events) and Louisiana (175 [15%]) reporting the most. Rail events occurred
most frequently in industrial (47%) and commercial areas (27%). A total of
1,080 (93%) events involved the release of only one chemical. Of the 1,299
total substances released, the most common were sulfuric acid (73 [6%] releases),
sodium hydroxide (60 [5%]), and hydrochloric acid (53 [4%]). Chlorine gas,
the substance released in all three case reports, accounted for 11 (0.8%)
of the releases reported to HSEES in rail events.
Approximately 60% of the known quantities released were measured in
gallons. Of these, quantities ranged from <1 gallon to 400,000 gallons
(median: 7.5 gallons). Of the 1,055 (91%) railroad events for which a primary
cause was identified, 645 (61%) resulted from equipment failure and 258 (24%)
from human error.
Forty-six (4%) of the 1,165 identified rail events resulted in injuries
to 271 persons, including four deaths. The persons most frequently injured
were members of the general public (e.g., nearby residents) (150 [55%]) and
employees (e.g., of railroads and plants) (77 [28%]). Of the 370 total injuries
sustained by the 271 persons, the most frequently reported were respiratory
irritation (147 [40%]), headache (40 [11%]), and eye irritation (36 [10%]).
Of the 271 injured, 205 (76%) were treated at hospitals and released, 29 (11%)
were treated on the scene, 15 (6%) were treated at hospitals and admitted,
and four (1%) died.
Of the 938 (81%) railroad events for which population data were available,
185,801 persons lived within one-quarter mile of the release (range: 0-3,000
persons; median: 38 persons). Seventy-five (6%) railroad events involved ordered
evacuations, of which 61 had a known number of evacuees. A total of 11,497
persons (range: 2-2,500 persons; median: 50 persons) were known to have evacuated.
Durations of evacuation ranged from <1 hour to 13 days (median: 4 hours).
C Henry, Missouri Dept of Health and Senior Svcs. A Belflower, MSPH,
D Drociuk, MSPH, JJ Gibson, MD, Div of Acute Disease Epidemiology, South Carolina
Dept of Health and Environmental Control. R Harris, Texas Dept of Health.
DK Horton, MSPH, S Rossiter, MPH, M Orr, MS, Div of Health Studies; B Safay,
T Forrester, Div of Regional Operations; S Wright, Div of Toxicology; Agency
for Toxic Substances and Disease Registry. MA Wenck, DVM, EIS Officer, CDC.
Approximately 800,000 shipments of hazardous substances travel daily
throughout the United States by ground, rail, air, water, and pipeline; approximately
4,300 shipments of hazardous materials travel each day by rail, including
chemical and petroleum products.6 Although nearly all of these
materials safely reach their destinations,7 many are explosive,
flammable, toxic, and corrosive and can be extremely dangerous when improperly
released. These materials frequently are transported over, through, and under
areas that are densely populated or populated by schools, hospitals, or nursing
homes, where the consequences of an acute release could result in environmental
damage, severe injury, or death.8
Findings from the HSEES system suggest that rail events constitute only
2% of total hazardous-substance releases. Furthermore, most rail events involved
small-scale releases (75% of events involved ≤70 gallons). However, large-scale,
acute releases during rail transit can occur (10% of events involved ≥2,200
gallons) and can cause substantial injury and death, as demonstrated by the
The findings in this report are subject to at least two limitations.
Reporting of any event to HSEES is not mandatory; therefore, participating
state health departments might not be informed about every event. Second,
only 16 state health departments provided data to HSEES during the analysis
period; therefore, the data represent only a proportion of the total hazardous-substance
releases in the United States.
Examining data on locations, types, and times of previous hazardous-substance
releases is crucial to preventing or planning responses to future releases
(see sidebar ). HSEES does not anticipate a new
funding announcement until 2008; however, nonparticipating states can use
the U.S. Department of Transportation Hazardous Materials Information Reporting
System (HMIRS) to acquire data on railroad and other transportation-related
hazardous materials incidents in their area. Although HMIRS does not actively
collect detailed public health consequence data, nonparticipating states can
request such data from HSEES participant states to increase their knowledge
of hazardous-substance releases.
REFERENCES: 9 available
*An HSEES event is the acute release or threatened release of a hazardous
substance(s) into the environment in an amount that requires (or would have
required) removal, cleanup, or neutralization according to federal, state,
or local law.4 A hazardous substance is one that can reasonably
be expected to cause an adverse health effect.
†Alabama, Colorado, Iowa, Louisiana, Minnesota, Mississippi,
Missouri, New Jersey, New York, North Carolina, Oregon, Rhode Island, Texas,
Utah, Washington, and Wisconsin.
‡Data for 2004 are preliminary.
Public Health Consequences From Hazardous Substances Acutely Released During Rail Transit—South Carolina, 2005; Selected States, 1999-2004. JAMA. 2005;293(16):1968-1972. doi:10.1001/jama.293.16.1968