On February 3, 2023, a 149-rail-car train in East Palestine, Ohio, USA derailed, and 50 rail cars ignited, 11 of which contained hazardous substances. 1 This event raised concerns about hazardous substances released into neighboring communities including Beaver County, Pennsylvania. Immediately, first responders worked to control the fires and ensure community safety. Responders worked on or near the derailment site, where rail car contents of concern included: vinyl chloride, ethylene glycol, butyl acrylate, isobutylene, and ethylhexyl acrylate.Reference Goodman 2 , 3On February 3, evacuation orders were issued for residents within a 1-mile radius of the train derailment. 4 On February 5, vinyl chloride contents of 5 rail cars became unstable. In response, the evacuation area was extended and a controlled vent and burnFootnote a event was conducted (full timeline of events pertaining to the incident were described at the June 22, 2023, National Transportation Safety Board East Palestine, OH Rail Public Hearing1), causing a chemical plume over the area.
On February 17, the Ohio Department of Health and the Pennsylvania Department of Health requested assistance from the US Centers for Disease Control and Prevention (CDC) and the Agency for Toxic Substances and Disease Registry (ATSDR) to implement an Assessment of Chemical Exposure (ACE) survey for first responders to systematically collect data to inform further public health action. 6 An electronic self-administered responder ACE survey was developed in REDCapReference Harris, Taylor and Thielke 7 , Reference Harris, Taylor and Minor 8with considerable input from partners including Ohio and Pennsylvania health partners, first responder survey pilot participants, and subject matter experts at CDC, ATSDR, and the National Institute for Occupational Safety and Health. The survey began March 5 and ended March 31, 2023. Eligible responders were those who responded to the East Palestine train derailment and were ≥18 years old. Responders residing in any state were eligible to participate. The survey collected information on demographics, occupation, self-reported health outcomes, personal protective equipment (PPE) use, decontamination procedures, prior training, and other topics. The survey also collected responders’ contact information and consent to be contacted in the future. This activity was reviewed by CDC, deemed not research, and was conducted consistent with applicable federal law and CDC policy.Footnote b
Results
Three hundred thirty-nine completed surveys were included in the analysis. Over half of responders reported their role as firefighter (61%), of whom 79% were volunteer. Most reported being non-Hispanic White (96%) and male (88%). The median age was 39 (IQR: 30-50) years. Most responders (86%) reported working on response activities within a 1-mile radius of the derailment site for at least 1 day between February 3-8, the period during which fires were active at the site. Most responders (79%) reported inhaling, touching, or swallowing potentially harmful substances. Of those who reported contact with at least 1 substance, smoke (84%) was the most frequently reported exposure, while 77% reported being exposed to an unknown/uncertain chemical. Most responders (75%) selected “Never” using a face mask or respirator during the incident. Approximately 48% reported at least 1 new or worsening physical symptom since responding to the incident. Of responders who reported at least 1 physical symptom, the most reported symptoms were related to the ears, nose, or throat (82%), heart and lungs (32%), eyes (23%), and skin (12%) (Table 1).
Demographics, occupational role, mask use, and self-reported exposures and health symptoms of individuals who participated in the Responder Assessment of Chemical Exposures Survey (N=339) — East Palestine, Ohio, February 2023-March 2023
a Race and ethnicity variables were collapsed. Percentages are out of those who self-reported their race and ethnicity (n=336). Individuals who identified as Black or African American non-Hispanic, Asian non-Hispanic or Multiracial non-Hispanic were collapsed into 1 category due to small cell sizes.
b Remaining answer options included prefer not to answer, or other.
c Answers are not mutually exclusive. Responders could select more than 1 answer option.
d Denominators for firefighter type are of the total number of individuals reporting being a firefighter.
e Denominator are those who reported working at least 1 day from February 3-8 (n=293).
f Denominator includes individuals who answered the question (n=327).
g Denominator includes responders reporting at least 1 direct contact (touching, inhaling, or swallowing) to potentially harmful chemicals/substances. Remaining responses included those who did not report any type of contact, prefer not to answer or unsure.
h Remaining responses include none of the above.
i For symptoms, responders were asked: “Since the incident, have you experienced a new onset or worsening of symptoms in any of the following (check all that apply): eyes, ears, nose, throat, central nervous system, heart and lungs, gastrointestinal tract, skin, behavioral/mental health, I have not experienced any of these symptoms or unsure.” The total excludes those who reported only a behavioral/mental health symptom, those who did not experience any symptoms, or those who reported unsure.
Conclusion
Survey results show responders reported low rates of respiratory protection use combined with high rates of inhalation exposure to unknown substances. Additional findings will be described in more detail in a separate manuscript. A similar train derailment incident involving vinyl chloride occurred in Paulsboro, New Jersey in November 2012 and an ACE investigation was conducted among emergency responders-the investigation found almost a quarter (23%) of emergency responders did not wear PPE during the response activities, as only 21 of 92 emergency responders self-reported donning self-contained breathing apparatus during the response.Reference Brinker, Lumia and Markiewicz 9
The survey design has several limitations. First, the survey aimed to maximize the detail responders may be able to recall but remains subject to recall bias given post-event administration. Secondly, information on specific timing and frequency of PPE use or contact with substances was not captured in this survey, preventing more exact estimates of exposures. Finally, given the cross-sectional nature of the survey, we were unable to examine causal relationships between variables over time.
A hierarchy of controls framework focusing on elimination of hazards, substitution for less hazardous substances, engineering controls, administrative controls, and PPE usage prioritizes other actions over relying on an individual’s behavioral protective practices. 10 This framework should be used to reduce the likelihood and severity of health effects among responders. It is worth noting, however, that for some events, employing the use of a hierarchy of controls framework may not be feasible. For example, during an emergency response, first responders may not be able to eliminate or substitute a hazard and may only be able to rely on their training and PPE to minimize potential exposure to hazards. Further, volunteer fire departments are often under-resourced and have specific challenges (e.g., inadequate equipment), which may impact their ability to respond effectively to disasters and other emergencies. 11
Increasing awareness and timely communications between rail operators and first response agencies about the hazardous contents of rail cars can improve response efforts. In addition, increasing responders’ timely access to information on substances in their environment might increase their ability to employ engineering controls and other appropriate protective measures. An existing approach that details pre-, during, and post- deployment activities that can be used in emergency events to protect worker safety and health is the Emergency Responder Health Monitoring Surveillance (ERHMS) framework. It includes recommendations and tools to protect emergency responders, including volunteers, during all phases of a response, as well as protocols for rapid identification and sharing of chemical information. 12 Local and state health agencies and responder groups can continue to support the health of responders to this incident by encouraging routine annual preventive care visits with primary healthcare providers, and training in hazardous substances response and appropriate use of PPE.
Acknowledgments
We would like to thank the Columbiana County Health Department, the Ohio Department of Health, and the Pennsylvania Department of Health for their collaboration. We appreciated the input on the responder survey from subject matter experts at the National Institute for Occupational Safety and Health. We also want to acknowledge the ACE investigation field team members for their contribution towards this investigation. Most importantly, we want to thank the first responders to the train derailment for their service.
Author contribution
Survey concept and design: JS, LG, AO, EF, CDew, MO, SC, DS, VP, CC, VM-G, LHM; Acquisition, analysis, or interpretation of the data: AO, CC, VP, EF, MO, SC, DS, VM-G, BG; Drafting of the manuscript: CC, VP, AO, MO; Critical revision of the manuscript for important intellectual content: LG, AO, EF, KB, CDowell, CDew, MO, SC, DS, VP, CC, BG, VM-G, MD, SW, LHM; Subject matter expertise: JS, CDowell, CDew, VP, CC, LHM; Administrative, technical, or material support: MO, JS, CC, VP, BG, CDowell; Supervision: MO. Co-senior authors: Carolyn A. Crisp PhD and Vidisha D. Parasram DrPH.
Funding statement
Support for the ACE investigation was provided by the Centers for Disease Control and Prevention and the Agency for Toxic Substances and Disease Registry.
Competing interest
There are no competing interests to disclose.
Disclaimer
The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry.
