The following study documents adjustment and health-related correlates of a major train derailment with an associated toxic chemical spill. The primary stressor, the effects of which were the focus of this investigation, involved a major train derailment in 1982 in the town of Livingston, a small rural community in Louisiana. As a result of the derailment, multiple toxic chemicals were spilled, exploded, and burned for a period of several days. Some 2,700 community residents, who lived in close proximity to the derailment site, had to be evacuated and were subsequently denied access to their homes, jobs, and schools for a period of approximately two weeks. Upon return to their homes, many residents found dead pets and agricultural animals, stained and discolored buildings, foul odors and structures damaged by explosion. During the following months, residents were subjected to additional stressors as they sought to adjust to altered social and economic conditions, performed clean-up activities and were exposed to possibly hazardous residues in the air, water, and soil.

      Employing a sample of over 100 individuals who had experienced the derailment, this study attempted to determine the proximity of subjects homes to the derailment site to independently assess the degree of stress resulting from this event and to study the relationship between these variables and indices of health and adjustment similar to those employed in prior disaster research. Secondarily, subjects in the derailment group were compared to a smaller sample of individuals from a nearby community in an attempt to further document stress-related correlates.

      Consistent with prior investigations, the present study documented a variety of correlates of living in close proximity to technological disaster. Approximately nine months after a major train derailment involving evacuation from their homes, instances of personal injury, property damage, and toxic contamination, individuals experiencing this event were still found to display significant stress-related difficulties. Specifically, compared to those living in a nearby town, who had not experienced the derailment firsthand, those experiencing the disaster were found to display significantly higher levels of psychiatric symptomatology. In obtaining measures assessing the stressfulness of the derailment, individuals who had experienced the derailment predictably evidenced significantly higher stress scores than those who had not. Indeed, participants living within a half mile of the derailment site had stress scores as great as those previously report for outpatients with post-traumatic stress disorders (Horowitz et al., 1971).

      The aforementioned findings, obtained by comparing individuals in the derailment and control groups, were supported by analyses employing only those individuals involved in the train derailment group. Measures of event-related stress (Impact of Event Scale), as well as measures of actual proximity to the derailment site, were found to be significantly correlated with measures of both physical (Wahler Physical Symptoms Inventory) and psychiatric (Symptom Checklist-90, Revised Edition) symptomatology. Regarding psychological symptoms, significant correlations were found not only for the more general distress indices but also for a range of primary symptom dimensions as well (e.g., somatization, obsessive-compulsive features, depression, anxiety). These findings, like those obtained from group comparisons, suggest not only a link between experiencing the derailment and stress-related symptomatology, but also that the magnitude of distress experienced was significantly related to the closeness of the persons home to the derailment site.

      These findings provide strong support for the notion that the derailment was experienced as a major stressor by those experiencing this event and its aftermath and that living in close physical proximity to the derailment site was associated with symptoms of both a physical and psychological natures. Given that the dependent measures used in this study were collected some nine months post derailment suggests that the stress of the derailment resulted in relatively long-term chronic effects on the health and adjustment of those participating in the study. As such, these results were consistent with earlier studies of the Three Mile Island incident in which stress effects experienced by those living close to the disaster site were found long after the initial occurrence of the disaster (Davidson et al., 1987).

      Three Mile Island was a greater disaster by national standards, by not necessarily local standards. While radiation is well understood as toxic, it may be somewhat ambiguous to a poorly educated rural community. Harrisburg, Pennsylvania, was a relatively progressive and well-educated urban center. Residents had a reason to doubt that their property would lose massive value for long periods of time. Thus, given their overall educational level, they could move to white/blue collar jobs in other places if they desired. Livingston residents, however, were generally under-educated, rural, under-skilled, and ingrown. They experienced a chemical exposure, which was more immediately discernible than radiation. Chemicals smell, burn, turn the stomach, effect water and vegetation, etc. Thus, while the derailment incident investigated in the present study could in some ways be viewed as a stressor smaller magnitude than Three Mile Island, the incident appeared to have been a more tangible and personal disaster in many respects.

      Another point of interest is the finding of a reasonably narrow range of effect that appeared to drop off steeply as the distance from the disaster site increased. Specifically, individuals living within a half mile radius of the derailment site were the ones who displayed the highest level of distress. Thus, it is plausible that there is a critical limit or distance between the home and disaster site beyond which there is relatively little response to the disaster. An alternate view is that there may be a summation of factors that are most likely to be experienced close to the disaster site and which decrease in a linear manner with distance from the disaster locus. In this case, the observed effects of the disaster may not relate solely to living close to the actual disaster site but also to the extent to which one experiences a range of disaster-related consequences (e.g., physical injury, other health related effects or risks, property damage, financial losses). Thus, the increased likelihood of experiencing these consequences close to the disaster site accounts for a decrease in emotional impact as one lives further and further from the site of the accident.

      Regarding directions for further research, it would be desirable that future studies seek indices of health and adjustment problems that are more objective than the self-report measures utilized in this study. While self-report measures employed here and in other published studies provide useful information, it is desirable whenever possible to supplement this type of data with other more unobtrusive measures, such as data regarding changes in physicians, hospital and emergency room visits, visits to mental health facilities by residents during pre- and post-disaster periods. While costly and difficulty to obtain, psychophysiological measures, biochemical indices, and measures of neuropsychological impairment would also yield considerable information.