Differences in PTSD Prevalence and Associated Risk Factors Among World Trade Center Disaster Rescue and Recovery Workers

The mode of World Trade Center Health Registry enrollment was significantly associated with PTSD among sanitation workers and volunteer organizations and thus was left in the model. For all other occupational groups, mode of enrollment was not significantly associated with PTSD.

This is one of the first studies to compare workers from different occupations responding to the same disaster and the first, to our knowledge, to analyze this for the World Trade Center attacks, the largest manmade disaster in U.S. history. The prevalence of PTSD and associated risk factors differed substantially, nearly fourfold across occupations. Workers in occupations less prepared for the type of disaster work required at the World Trade Center site, such as firefighting and search and rescue, were more likely to develop PTSD; the highest rates were documented among volunteers, construction/engineering workers, and sanitation workers. Furthermore, workers who engaged in tasks outside of their training were also at increased risk. When examined by task, the rates of PTSD were highest among emergency medical services/medical/disaster personnel who engaged in firefighting and sanitation workers who engaged in search and rescue. In contrast, few work experiences at the World Trade Center were associated with risk among groups for which prior training and preparedness is most likely to be heterogeneous, and the strongest risk factors for these groups were within-disaster experiences.

Our finding that the prevalence of PTSD varied by occupation is consistent with the previous literature on disaster responders (3 – 7) and suggests that prior training or experience may protect against the psychological distress associated with disaster work, as suggested by others. Potential explanations for why prior training protects against PTSD include an increased sense of self-efficacy or internal locus of control and the satisfaction of applying previous training to successfully fulfill duties (23, 25, 26) .

Among occupations most likely to have had prior training, emergency medical services/medical/disaster personnel and firefighters were twice as likely to have current probable PTSD as police. Other studies of police involved in rescue/recovery work have also reported a lower prevalence of psychological distress (23, 27) and probable PTSD (24) . Police screening procedures may result in selection of a more psychologically resilient workforce. It is also possible that police officers are more likely to underreport symptoms of psychological distress for fear that they will judged as unable to perform their job responsibilities. Firefighters and emergency medical services personnel may also underreport symptoms; however, it is more likely among police officers because compromised psychological health is associated with graver consequences in an environment where carrying a firearm is a daily requirement.

For the World Trade Center disaster, bereavement (28) and self-identification with the victims (29) may have also contributed to the difference in PTSD prevalence. Firefighters lost six times as many comrades as the police (30), and Fire Department of New York funerals continued through September 2003. An assessment 7–13 weeks after Hurricane Katrina, however, documented no disparity in the prevalence of PTSD among police and firefighters (19% and 22%, respectively) (31) . These findings may be explained by other experiences known to increase risk for psychological distress, such as civil disobedience, hostility, and aggression toward police (32), all widely depicted in post-Katrina media coverage.

Lack of access to mental health services may also account for increased risk of PTSD among sanitation workers, construction/engineering workers, and unaffiliated volunteers. Although our survey did not directly assess this, sanitation workers, medical personnel, and other volunteers were excluded from the World Trade Center injury and illness surveillance programs established in local hospitals after the attack to assist rescue/recovery workers injured while working at the site because it was believed that they were not directly involved in rescue/recovery operations (33) . Furthermore, World Trade Center firefighters and police had additional access to mental health support and critical incident stress management services through other organizations. To our knowledge, no comparable services were established for construction/engineering and sanitation workers.

It is also possible that lack of recognition is partially responsible for an increase in psychological distress. Construction and sanitation World Trade Center workers from another study reported feeling that their psychological distress was prolonged because unlike other World Trade Center workers, they did not receive recognition for their efforts, preventing them from obtaining closure (13) .

Among the range of traumatizing experiences during the disaster, being injured on September 11th significantly increased the risk of PTSD for all workers, whereas witnessing trauma was a significant risk factor only for construction/engineering workers, volunteer organizations, unaffiliated volunteers, and other government agencies. One potential explanation for this finding is that previous exposure to trauma during the daily job responsibilities required of police, firefighters, and emergency medical services personnel may have desensitized them to the deleterious effects of witnessing trauma. In contrast, sustaining an injury is a measure of personal life threat, which is less likely to be modified by prior experiences.

As was found among responders to the Oklahoma City bombing (25), the duration of work at the disaster site was associated with PTSD. For all occupations except police, the probability of PTSD was elevated for those who worked for more than 3 months. Stratified analyses also revealed that the relationship between time worked at the site and the probability of PTSD was strongest for those who started on September 11th, when exposure to trauma and risk of injury was greatest in all groups except for police. It is possible that the lack of association between the start date, duration of work, and PTSD among police may be related to tasks performed because they were responsible for patrolling the perimeter and less likely to have been involved with tasks on the pile for which they had no prior training. Supporting this argument is evidence from our cohort that few police reported operating heavy equipment (1.2%), performing light construction work (6.9%), or firefighting (1.9%) at the World Trade Center site.

A few limitations of this study must be noted. First, among sanitation workers and volunteer organizations, mode of enrollment was significantly associated with PTSD. Individuals within these groups who self-identified into the registry were more likely to be diagnosed with probable PTSD. This raises the possibility of reporting bias and overestimation of the mental health impact of the event; however, we appropriately controlled for this in each of the multivariable analyses across all occupations, and our findings remained the same. Second, individuals from hundreds of different organizations responded to the disaster and initiated work at varying time periods, and ascertaining the representativeness of this cohort is challenging. An assessment of coverage and enrollment rates found that approximately one-third of those eligible enrolled in the World Trade Center Health Registry with coverage varying among organizations: the enrollment rate was 68.9% for the New York City Police Department, 50.4% for the New York City Department of Sanitation, and 23.5% for the Fire Department of New York (34) . Although coverage for some groups was modest, for the five major occupational groups included in this study, we successfully recruited ample numbers from both self-enrollment and employee lists to examine differences in prevalence and risk factors. Third, degree of previous training and professional experience were not directly assessed but rather inferred based on occupation. However, the observation that tasks inconsistent with typical occupational roles were significant risk factors for PTSD provides greater confidence in our assumptions about prior training. Examination of characteristics such as mental health history and social support and mental health care use are also necessary to further elucidate the relationship between disaster work and PTSD. Such assessments are planned for the first follow-up survey of World Trade Center Health Registry registrants. Finally, it is important to note that among some of the groups assessed, such as volunteer workers and responders from other government agencies, individuals most likely self-selected for participation in rescue/recovery work. It is possible that some persons who self-selected to respond were not well screened compared to police and firemen and may have been at increased risk for PTSD.

The World Trade Center Health Registry is the largest disaster registry in U.S. history, and the rescue/recovery worker sample includes the largest number of workers from multiple occupations exposed to the same disaster. Current probable PTSD was identified with a standardized, well-validated assessment, providing confidence in our prevalence estimates and enabling comparisons across studies.

The results from the present analysis suggest that rescue/recovery workers who responded to the September 11th disaster continue to experience substantial psychological distress years later. Our findings demonstrate a need for targeted interventions specific to the diversity of workers who respond to large disasters to reduce the psychological burden associated with participation. Disaster preparedness training and shift rotations to enable a shorter duration of service at the site may reduce PTSD among workers and volunteers in future disasters.

Received Oct. 6, 2006; revisions received Dec. 8, 2006, and Jan. 30 and March 5, 2007; accepted March 19, 2007 (doi: 10.1176/appi.ajp.2007.06101645). From the Cognitive Neurophysiology Laboratory, Nathan S. Kline Institute for Psychiatric Research; the Divisions of Epidemiology and Environmental Health, New York City Department of Health and Mental Hygiene, New York; and the Agency for Toxic Substances and Disease Registry, Atlanta. Address correspondence and reprint requests to Ms. Perrin, the Cognitive Neurophysiology Laboratory, Nathan S. Kline Institute for Psychiatric Research, Bldg. 35, 140 Old Orangeburg Rd., Orangeburg, NY 10962; [email protected] (e-mail).

All authors report no competing interests.

The authors thank the World Trade Center Health Registry baseline interviewers employed by Research Triangle International. The success of the World Trade Center Health Registry baseline enrollment also resulted from a concerted effort of the outreach, communication, and data teams of the New York City Department of Health and Mental Hygiene, Research Triangle International, and the Agency for Toxic Substances and Disease Registry. The authors also thank Thomas Frieden, M.D., Lloyd Sederer, M.D., Jim Cone, M.D., and Scott Kellerman, M.D., for their review of this manuscript, the organizations that provided us with lists of workers who participated in the relief efforts, and all of the participants in the study.