Two Year Later: The Prolonged Traumatic Impact of a Disaster

1The author expresses gratitude to Christian Arrington of the Oakland Unified School System for his assistance with the statistical analyses of the data.

2The author is in private practice and a Clinical Associate Professor with the Department of Psychiatry and Behavioral Sciences at Stanford University School of Medicine

Abstract

The Pebble Beach fire of 1987 destroyed 30 homes. Eighteen months after the fire, ten adults, whose homes and belongings were destroyed, were administered the Rorschach Test and the SCL-90-R Test. Comparison with the means of the normative groups for each test yielded significant results above the p < .01 level. The long term traumatic impact on the personalities was clearly reflected by interferences in thought processes, perceptual distortions and depressive reactions, inter alia.

The Traumatic Impact of a Fire Disaster as Reflected in the Rorschach and SCL-90-R Tests

INTRODUCTION

There was a disastrous forest fire in the Pebble Beach area of California on May 31, 1987 that destroyed a community of 31 homes, severely damaged six other homes and compelled the evacuation of over 200 people. The people had little warning about the severity of the fire and little time in which to leave their homes. In some instances they had only time to run to their cars, escaping while watching their home and all their possessions explode with fire. The fire had begun as an illegal campfire set by local teenagers; it spread quickly and was fanned by winds rising to 50 miles per hour and the temperature reached 500 degrees Fahrenheit. The conflagration swept up a hill, leaping from roof to roof and from tree top to tree top, burning homes as it raced. No one was killed in the fire but the damage was immense in this upper class community.

A law suit was initiated, based on the allegation that the disaster and its scope were predictable and that the community organization had failed to provide adequate protection from such a danger. The author was invited to determine whether any mental health problems of the thirteen fire victims that were related to the traumata of the fire.

The present report summarizes the psychological evaluation of ten of these individuals. Each was given an intensive three to four hour interview, the Rorschach Psychodiagnostic Test, and the Symptom Checklist (SCL-90R). The evaluation occurred 18 to 20 months following the fire.

Knowledge about the effects of traumata and disasters on individuals and on social groups has slowly accumulated through studies in the last fifteen years. Notable among the earliest discussants of the effects of disasters is Martha Wolfenstein (1957). Wolfenstein's analyses of the impact of disasters upon the lives of victims suggested that there were several phases of psychological reactions which included the shock, followed by an initial adjustment to the occurrence. Then came reactions of disillusionment and depression; next were efforts to internalize the facts of the disaster and finally, a reorganization of the personality in the recovery process. When the internalization fails, psychopathological symptoms emerge and fixate. The fixation results in the post-traumatic stress disorder defined in the Diagnostic and Statistical Manual III-R. Summarizing the syndrome therein described, the symptoms include a persistent re-experience of the event as in recollections, dreams, or a sense of reliving the event; persistent avoidance of stimuli associated with the traumatic situation as efforts to avoid thoughts or feelings associated with the event, feeling of detachment, restricted affect, etc.; and, persistent symptoms of increased arousal as difficulty sleeping, irritability, difficulty concentrating, etc.

Factors that contribute to or interfere with re-adaptation include the suddenness and duration of the event (Barton, 1969; Berren, Beigel and Ghertner, 1980), the seriousness of the threat to life, bereavement, the duration of suffering (Berren et al., 1980; Gleser, Green and Winget, 1981), and the scope of the disaster (Barton, 1969; Gleser et al, 1981). In addition, the quality of family and community support (Green, 1982), the source of the disaster--i.e., human error or natural disaster--prior experience with the particular stress, or the victim's vulnerability, and the wealth and availability of resources (Appley and Trumbull, 1986), all assist or impede recovery.

Follow-up studies of natural disasters have consistently shown an increase of symptoms of psychological distress, including symptoms of stress, anxiety, helplessness and depression. The effects on bodily health were evidenced by a significantly heightened frequency of physical ailments, and by visits to physicians and to hospitals. Furthermore, the mortality rate increased for victims in the following year. Social contacts were significantly lessened. People were less satisfied with the quality of their lives and reported less time spent in leisure. These studies have been conducted from ten months to ten years following a disaster. The percentages of persons for whom psychic distress was reported vary from a low incidence of 33% to a much higher 100% (Chamberlain, 1980; MacFarlane 1986; MacFarlane, 1988).

Man-induced disasters are considered more harmful in their psychological effects than "natural" disasters. The following examples illustrate some incidents in which human error was proven or suspected and summarize briefly the results of the follow-up studies on the victims. The knowledge that the disaster could have been avoided seems to release a rage and anger that are not observable in those affected by natural disasters. Victims experience heightened distrust and suspicion of others and their motives. Their unresolved grief brings about personality changes that involve guilt, rage, demoralization, and a diminished elan vital. The Coconut Grove Fire of 1943, assumed to have been due to negligence, produced tremendous horror and terror in those present at the time as 400 persons were burned to death or otherwise killed in their efforts to escape. One year later, fifty per cent of the survivors still manifested symptoms of sleep disturbances, increased nervousness and anxiety, guilt over survival and fears related to the events of the fire (Adler, 1943; Cobb and Lindemann, 1943).

A four-year follow-up study of a 1963 ship collision revealed that 75% of the survivors had severe work-related problems and persistent psychological distress symptoms (anxiousness); mood disorders actually increased over time and psychosomatic disorders became more frequent (Leopold and Dillon, 1965). The effects on the victims of another shipwreck were similar. One and two years following the disaster, all but one had manifest psychiatric disturbances and none had returned to work on ships (Henderson and Bostock, 1979).

A ten year follow-up study of men who were buried alive for several days in a mine disaster in Langede, Germany in 1963 documented that there were no continuing friendships among the survivors (Ploeger and Andreas, 1974). Most survivors reported significant personality changes characterized by heightened irritability, phobias, and flashbacks to the event that were hallucinatory at times. The shared experiences of a disaster did not result in an intimacy or enduring friendship as might have been anticipated.

The Buffalo Creek dam break and flood of 1972 was a highly publicized and intensively investigated disaster whose victims were interviewed in two-year and five-year follow-up studies (Gleser, Green and Winget, 1978; Gleser et al., 1981; Green and Gleser, 1983; Lifton and Olson, 1976; Rangell, 1976; Titchener and Kapp, 1976). When interviewed two years after the flood, eighty per cent of the disaster victims had disabling symptoms and problems in adjustment. Despair, apathy, aimlessness, depression, hypertension, sleeping problems and anxiousness were common. The use of alcohol and nicotine increased. Psychosomatic and health problems increased significantly and measurably so, even five years after the disastrous flood.

In view of the above, psychological evaluation of the Pebble Beach fire victims can be expected to show the long-term effects of the fire on their personality. This, too, was not a natural catastrophe, but was considered preventable not only because of its accidental inception but also because of the careless management of the fire after it had begun. In order to investigate the degree of internalization of the trauma and the quality of adaptation to the impact of the fire, the Rorschach psychodiagnostic test was administered. This test was expected to tap those effects of the trauma of which the persons might not have been as consciously aware. The SCL-90-R test was used as a self-report measure of distress, which information was also discussed in the interviews.

The difficulties involved in assessing the effects of a trauma on a person 18 months after the event are numerous. The pre-fire personality of each is, by definition, unknown and is not directly measurable; the immediate impact of the fire on the person's behavior and the effects of the loss of home and possessions may only be inferred through self-reports, self-observations or observations by others. In addition, there was much instability, strain, insecurity and anxiety due to the post factum turmoil still present at the time of the evaluations. Some victims were required to move three times in this period while a new home or new possessions were acquired; additional stress came with the process of completing claims for insurance or governmental assistance and waiting an unpredictable time for responses; and, the rigors of the legal process itself, of filing a suit, giving depositions, etc., heightened the tension and re-enlivened the traumata. There is, therefore, sufficient reason to expect that the evaluations would show cumulative effects of the disaster and its aftermath on the personality. These effects are best evaluated through the utilization of the Rorschach Test which includes scores that estimate the degree of stress-related anxiousness.

It was assumed that each fire victim would demonstrate significant symptoms of post-traumatic stress because of the suddenness of the event, the inability to make any preparations for the situation, the total loss of home and possessions, the paucity of resources available to the people, the human error involved in the onset of the disaster, and, the compounding stress of litigation. Further, the post-traumatic effects should be reflected in the results of the SCL-90R and the Rorschach test and thereby, provide evidence that many disaster victims fixate in Phase Four (Wolfenstein, 1957).

METHOD

TIME

The psychological evaluations occurred eighteen to twenty months after the fire. Each person was interviewed and administered the SCL-90-R and the Rorschach test in this period. Therefore, it was assumed that the time factor was a constant.

Three procedures were used, as described below:

(1) A semi-structured interview. The interview lasted three to five hours for each person The contents of the interview included (a) a pre-fire life history, (b) the fire experience, (c) actions taken, and reactions to the fire, in the following week, (d) long term effects of the fire on the victims, and (e) the individual differences in recovery from the fire and its aftermath.

(2) The Symptom Checklist-90-Revised (SCL-90-R). This is a frequently used self-report test that requires the subject to make self observations about anxiety-related symptoms. The test is composed of ninety items to which the subject responds by evaluating the level of discomfort experienced on a five-point scale varying from not at all to extremely. The subjects were asked to evaluate the discomfort of each symptom they experienced since the time of the fire. The test is factored into scores of somatization, obsessive-compulsive tendencies, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation and psychoticism. There are two derived scores - the Global Severity Index and the Positive Symptoms Index.

The scores of the fire victims were compared with two samples originally reported in the SCL-90-R administration manual (Derogatis, 1977). These are a psychiatric outpatient group (N = 1002) and a non-patient normative group (N = 974). Because of the small sample size of the present study (N = 10), a t-test for independent samples was used to determine possible significant differences between the victims and the two groups presented by Derogatis. The level of confidence chosen was p01.

(3) The Rorschach Psychodiagnostic Test. This test was administered, scored and interpreted using the Exner method. The scores were then grouped to provide a comparison between the research fire group and the normal non-patient adult group reported by Exner (Exner 1989).

A pooled estimate t-test for two independent samples (Hayes, 1963, p.320) was used to determine whether significant differences existed between the means of the scores of the Fire Victim Group and the selected normative groups.

SAMPLE

The sample consisted of ten persons whose homes had been destroyed in the fire. There were four females and six males; their ages ranged from 32 to 65 years. All were of the upper middle socioeconomic class. All had a college education. One person did not take the SCL-90-R test due to time limitations.

RESULTS

The purpose of the study was to determine whether victims of a community disaster (i.e., a fire) manifested symptoms of post-traumatic stress as measured by the SCL-90-R and the Rorschach Psychodiagnostic Test.

Hypothesis 1. There will be no significant differences in the means of each SCL-90-R dimension between the Fire Victim Group and the Normative Psychiatric Outpatient Group or the normative non-patient normals comparison group.

Insert Table 1 about here

Table 1 presents the means, standard deviations, and the t-test values for the Fire Victim Group and the Normative Psychiatric Outpatient Group as reported by Derogatis (1977) in the administration manual. There were significant differences between the fire group and the outpatient group on the scales of Interpersonal Sensitivity (t = 3.422 p < .0005), Depression (t = 2.859, p < .005), Anxiety (t = 2.574, p < .01), Phobic Anxiety (t= 7.205, p < .0005), Paranoid Ideation (t = 4.874, p

There were three symptom dimensions in which the Fire Victim Group did not differ from the Normative Psychiatric Outpatient Group. These were Somatization, Obsessive-Compulsive, and Hostility symptoms. These results suggest that the fire victims were as severely disturbed as the psychiatric outpatient group on these dimensions of personality. Therefore, there are some significant differences between the Fire Victim Group and the Normative Outpatient Psychiatric Group and the null hypothesis is rejected for those specific symptoms. However, on the GSI (General Stress Index) the Fire Victim Group was significantly lower, i.e., reported fewer symptoms, than the Normative Psychiatric Outpatient Group and the null hypothesis is rejected for the GSI.

Insert Table 2 about here

It was assumed that the scores of the Fire Victim Group would differ from those of the Normative Outpatient Psychiatric Group because of the lack of previous, acknowledged, psychiatric difficulties. It was also assumed that their traumatization symptoms would differ from those of the normed group for the SCL-90-R. Table 2 presents the means, standard deviations and t-values for the Fire Victim Group and the Normative Non-Patient Adult Group. There are significant differences in the t-value between the groups on all symptom dimensions except that of Phobic Anxiety symptoms. The Fire Victim Group also scored significantly higher than the normative group on the Global Severity Index of symptoms (M = -.89, t = -4.181, p < .005), indicating that the fire victims suffered a greater level of general distress than the normative group.

Hypothesis 2. There will be no significant differences between the means of the Fire Victim' Group and the means of the Normative Non-patient Adult Group (Exner, 1986) an the scores of the Rorschach Psychodiagnostic Test.

Insert Table 3 about here

Table 3 presents the means and standard deviations for the two groups on the primary variables in the scoring of the Rorschach Test. The DO+ score was significantly different at the p < .01 level indicating that the normative group gave significantly more synthesized responses.

The variables that were significantly different at the p < .005 level of confidence were: (1) the D score which indicates that the Fire Victim Group gave significantly fewer large detail responses; (2) the paired responses (2) were given significantly less frequently by the Fire Victim Group; (3) Pure F responses (M = 5.3, sd= 3.71) were given less frequently by the Fire Victim Group; and (4) the Experience Actual sum (EA) was significantly greater for the Fire Victim Group.

Many scores attained a level of significance at the p < .001 level. (1) The Fire Victim Group gave significantly fewer popular responses (P) even though there was no significant difference in the total number of responses given; (2) the mean of the Whole (W) responses was significantly greater for the Fire Victim Group (M = 14.5) as compared with the normative group (M = 8.58); the Lambda (L) ratio was significantly lower for the Fire Victim Group and suggests difficulty in utilizing their inner strengths effectively; (3) the Form Quality (~FQ) scores showed significant differences between the groups. Surprisingly, the Fire Victim Group gave significantly more FO+ responses1 representing superior, over-elaborated responses that are "unique by the manner in which details are defined." (Exner, 1986, p. 148). This group also gave significantly more FQu and FQ- responses, demonstrating that their perceptions were uncommon and showed a "distorted, arbitrary and unrealistic use of form in creating a response" (Exner, 1986, p. 148). The Fire Victim Group provided significantly fewer responses of ordinary form quality (FO) The lack of appropriate form quality of the responses given by the Fire Victim Group was also substantiated by the percentages of total responses that reflect perceptual accuracy. The X+% (conventional form) ratios and the F+% (conventional pure form) were significantly lower for the Fire Victim Group whereas the X-% (distorted form) ratios were significantly greater than expected; (4) With regard to the individual determinants of the Rorschach responses, the Fire Victim Group produced significantly more blend responses, i.e. used more determinants in their responses than the Normative Non-patient Adult Group, which suggests again that their responses were elaborated and complex. While there was no significant difference between the means of the groups on human and animal movement (M, FM) responses, the Fire Victim Group gave significantly more Ma (active human movement) and a (total active movement) responses. The significance of these ratios as they relate to the ability to cope with stress will be discussed. More importantly, the Fire Victim Group gave significantly more inanimate movement (m) responses, substantiating an assumption of increased inner distress.

The color responses (FC, CF, and C) relate to the modulation of affect. The expectation would be that traumatized individuals might show lessened ability to control or to modulate their emotional expressions. The responses of the Fire Victim Group do show fewer FC responses but not at a significant level, and a greater number of CF responses, again not at the selected level of significance. However, the Fire Victim Group did give significantly more pure color (C) responses (M = 1.0) than the Normative Non-patient Adult Group (M = .12) at the p

Other determinants that were given at the significance level of p < .01 by the Fire Victim Group include the achromatic color (C’) responses, the texture (T) responses and the reflection (r) responses. These scores suggest that the Fire Victim Group were more tense and self-oriented than would be anticipated. This seems substantiated by the es (Experienced Stimulation) sum which would indicate that the Fire Victim Group experienced many more demands on their coping abilities than the normative population.

Those scores and ratios that were significantly different at the p < .001 level of confidence were as follows: (1) The mean of the whole (W) responses was significantly greater at the p < .001 level for the Fire Victim Group (M= 14.5) compared to the normative group (M = 8.58); (2) the Suicide Constellation (S Con) ratio showed that the Fire Victim Group had much greater suicidal ideation (M = 5.2 for the Fire Victim Group, M = 2.07 for the normative group), as well as significantly greater morbid (MOR) preoccupations; (3) the Depression Index (DPI) showed that the Fire Victim Group (M = 2.3) had significantly more depression responses than those of the normative adult group (M = .40); and, (4) the Schizophrenia Index (SCZI) represents a cluster of variables that are "related to problems in thinking and perceptual adequacy" (Exner, 1986, p. 182); here, the Fire Victim Group manifested a significantly greater SCZI (M = 2.3, s.d. = 1.73) than the normative adult group (M = .95, s.d. = 1.08) at the p < .001 level of confidence but it is important to note that the SCZI scores do not attain clinical significance.

Another level of significance proposed by Exner (1986) posits that any variables which deviate from the non-patient group by more than one standard deviation are considered to be clinically significant. It seems important to observe that the S-Con, the DEPI, and the SCZI scores did not attain clinical significance, thus the Fire Victim Group cannot be characterized as severely psychopathological.

The comparison of the means of the Fire Victim Group with those of a Normative Psychiatric Outpatient Group and the comparison with the means of the Normative Non-Patient Adult Group for the SCL-90-R and the comparison of the Rorschach responses with the normative adult sample give clear evidence that the fire victims were distinctly different from the comparison groups.

DISCUSSION

The purpose of this study was to determine whether victims of a fire that had destroyed their homes some eighteen months earlier manifested measurable signs of stress on the SCL-90-R and the Rorschach Psychodiagnostic Test. Ten of the fire victims were administered the Rorschach test; nine additionally took the SCL-90-R.

THE SYMPTOM CHECK LIST

When the results of the SCL-90-R test taken by the Fire Victim Group were compared with the sample of psychiatric outpatients, the scores on the scales of somatization, obsessive-compulsive traits and hostility did not differ. Moreover, the fire victims, as a group, shared these traits to a similar degree with persons in outpatient psychotherapy. Thus, they may be assumed to have developed at least moderately severe psychosomatic and physical symptoms, to be wrestling with compulsive urges to undo the trauma, and to experience unresolved anger and irritability provoked by the disaster and its consequences. Further, in comparison with the normative standardization group on the SCL-90-R, they were significantly more depressed, experienced more anxiety, had heightened interpersonal sensitivity, and suffered more paranoid ideation and psychoticism (confused thinking) than the normative group.

Another important feature is that the general severity of the fire victims' symptoms (Global Severity Index) was significantly less than that of psychiatric outpatients but significantly greater than that of the Normative Non-Patient Adult Group. This provides some basis for the assumption that these individuals were not psychiatrically ill prior to the fire and that, even after the disaster, they were not that emotionally disturbed. They did, however, report significantly more distress than the normative group, suggesting that the fire and its aftermath, remained highly stressful even eighteen months after the disaster.

THE RORSCHACH TEST

The scores and score ratios on the Rorschach obtained from the Fire Victim Group were compared with the Non-patient Adult Group reported by Exner (1986, pp. 257-258). It became quite apparent that the Fire Victim Group demonstrated many significant differences from the Non-Patient Adult Group which suggested that they experienced intrapsychic difficulties to the extent that these emotional problems interfered with their everyday functioning both in their orientation to events, in their thought processes, and in the expression of affect. The comparison utilized Exner's (1986) criteria for clinical significance as well as the confidence level of .01 or greater. The following discussion focuses on understanding the results of the Rorschach test results of the Fire Victim Group.. Validity of the test results

The protocols of the Fire Victim Group were of sufficient length to warrant careful consideration, that is, they gave an average of 23.6 responses; Exner states that 17-27 responses are normal. The Lambda (L) score is significantly below the mean for the Non-patient Group (p < .001); this suggests that the fire victims' apprehensions interfere with concentration or logical reasoning, that they became over-involved with stimuli and have difficulty perceiving economical solutions to problems. The protocols may be assumed to be valid and representative of the psychological state of the group.

Stress tolerance and control

Exner (1986, p. 315) avers that "a cluster of six variables provides the first data set" from which to interpret the test results in order to evaluate the ability to tolerate stress and control responses. These variables are: the EB (Erlebnistypus), EA (Experience Actual), eb (Experience Base), es (Experienced Stimulation), the D score and the adjusted D Score. Two of these variables, the EB and the eb ratios, are meaningful only for the interpretation of individual protocols and are not considered in this report. The D score of the fire victims falls within the normal limits, although at the lower end of expectability. The score indicates that under most circumstances the fire victims had sufficient resources to be able to direct their behavior in a deliberate and meaningful way without loss of the ability to control their actions. However, when the D score and the adjusted D score are in the minus range, the implication is that the persons are more vulnerable to being overwhelmed by situational demands. "People who fall into the D-2, D-3 or lower categories are in an almost continuous state of overload. They are upset with more experienced demands for responses than they can handle easily." The mean frequency of m responses is also significantly greater than expected (M = 3.0, p < .001); accordingly, this suggests a fear of the disintegration of controls over their behavior due to situational distress and reflects an inner sense of possible disruption in their lives. In fact, the fire victims experience much more anxiety along with a sense of doubt about their ability to cope with the demands of life. (Exner 1986 pp. 317-318).

The EB ratio and the eb ratio could not be computed meaningfully for a group because they are meaningful for individual protocols only. However the EA (Experience Actual) sum and the es (Experienced Stimulation) sum were computed,. both attained clinical significance and both were significantly greater than those of the normative group. The suggestion from the EA sum is that the fire victims did not have ready access to their inner resources because of the stress under which they were functioning. They could not easily summon their coping strengths to resolve everyday situations. That they do have sufficient inner strengths is indicated by their production of the expected level of M responses. The fire victims differed significantly in the number of chromatic color responses that they gave on the Rorschach Test. They produced fewer FC responses than the normative sample, indicating a lowered ability to modulate or control affective expression. This characteristic becomes more obvious when considering the CF and C responses which were given significantly more frequently by the fire victims. These scores emphasize that the fire victims experience some lack of control over emotional expression, have become more susceptible to stress, and are less able to modulate their emotional life during stressful events.

The shading variables (C’, T, V, and Y) are all related to "impinging or irritating affects" (Exner, 1986, p. 337), that is, they signal the presence of distress in the individual. The significantly greater T scores among the fire victims (M = 2.2, p < .001) strongly suggest that they have experienced an emotional loss; that is, these scores reflect the traumatic effect of the loss of house and belongings on the fire victims as a group. According to Exner( 1986, p.339), persons whose T scores are elevated, experience stronger than usual needs to be dependent on others. The Y scores and the V scores did not differ significantly from the normative group and provide evidence that the fire victims did not suffer from feelings of helplessness. Also the fire victims did not differ significantly from the normative group in their ability to view problem situations with an appropriate perspective (V).

The achromatic responses (C’) were also significantly more frequent in the Fire Victim Group (M = 2.2, p < .001). These responses indicate that the fire victims experienced more depressive affect; that is, they placed an internal constraint over their emotions and such constraint causes a sense of discomfort and uneasiness.

The results of these scores on the Rorschach point to a lessening of coping capacity among the fire victims; they have increased anxiety, experience fears of losing control over their behavior, feel somewhat constrained and depressed, and have lessened ability to organize their resources to react well to stressful situations.

Cognitive Qualities

The level of cognitive operations appears quite high for the fire victims. As a group they expended greater effort to organize their responses (Zf = 16.7, p < .001) than the adults of the normative sample, i.e., they appeared to have chastened by their experience and thus they showed a need to deal with events in a particularly careful and thorough manner. However, the efficiency of their organization attempts is within that of the normative sample. They also produced significantly more whole responses (W) which suggests a need to deal with a stimulus situation in its entirety; this is supported by the fewer detail responses and fewer unusual detail responses (D and Dd). Their cognitive functioning has been affected deleteriously. The Contamination (CONTAM) score occurred significantly more frequently among the fire victims' responses; yet the WSum6 score showed that the Fire Victim Group gave significantly fewer signs of disturbed thought processes than the normative group. The CONTAM score suggests that the fire victims' responses may have been compromised by the intrusion of symbols of their traumatic experiences with the fire and its aftermath. It appears that the fire victims suffered a loss of practicality and efficiency in responding to situations because, motivated by alarm, they are compelled at once to take account of the whole stimulus field.

Perceptual Orientation

The orientation to reality is an important process for the individual as it directs the cognitive, affective and social reactions to situations. It was assumed that the fire victims would suffer a distortion of their view of everyday contingencies because of their experience with the fire, its unpredictability and their unpreparedness for such a disaster; this would sharpen their sensitivity to possible dangers around them. The results of the Rorschach clearly support the assumptions of a disruption of normal perceptual processes during a status emergens. The popular responses (P) were clinically lower than those of the normative group; the X+% (Conventional Form percentage) was significantly lower (p < .001); the F+% (Conventional Pure Form percentage) was clinically lower; and the X-% (Distorted Form percentage) was significantly greater than expected (p < .001). All these percentages were also clinically at variance with those of the normative group. It thus appears that the fire, which was a disaster of overpowering suddenness, had a severely disruptive impact upon its victims' ability to process reality experiences and this corroborates the hypothesis that they now experience difficulties in making appropriate responses, especially under stress.

Special Scores: S-Con, DEPI, SCZI

Scrutiny of the indices of psychopathology in the Rorschach makes it evident that the fire disaster and its aftermath have caused these victims severe emotional injury. The S-Con (Suicidal Constellation) score is significantly greater than for the normative group (p < .001), but it does not meet Exner's (1986, p. 414) criteria for subjects at high risk for self-destruction. Nevertheless, since the catastrophe presented the fire victims as a group with the peril of losing their lives, it must have undermined their usual sense of invulnerability or indestructibility. The results also confirm some preoccupation with morbid, self-annihilative thoughts, for the fire victims, as a group, had significantly more Morbid (Mor) responses (M = 2.0, p < .001) than the normative group, which substantiates their preoccupation with destructive ideas or dysphoric feelings. Some individuals who gave these responses more frequently could conceivably be at risk for suicidal attempts.

The Depression Index (DEPI) was also significantly greater for the Fire Victim Group (p < .001). While Exner's criteria (Exner, 1986, p.425) state that three or more variables of the Depression Index are required before a severe depressive reaction may be presumed to be present, the indications are that the fire victims as a group do experience frequent episodes of depression and it seems quite likely that one or more of them may have been experiencing severe depressive reactions following the fire.

It is assumed that some individuals within the victim group suffer from depressive experiences, in turn elevating the Depression Index, and that the group, as a whole, suffers more depressive moods than would be anticipated when compared with a Normative Non-patient Adult Group.

The Schizophrenic Index (SCZI) for the fire victims does not meet the essential criteria of five positive variables necessary for the diagnosis of schizophrenia. That is, the fire victims are not psychopathologically compromised, nor do they show signs of severe psychiatric disturbances.

SUMMARY

The fire victim group showed signs of internal distress even eighteen months following the experience with a fire that destroyed their homes and endangered their lives. The results of the SCL-90-R and the Rorschach Test support the extrinsic evidence that such a disaster leaves severe personal anxieties in its wake. The victims were angry, hostile, and resentful at the loss of their homes, had obsessive preoccupations about the fire and the events that followed, and suffered an increase in somatic complaints on a level reported by persons in outpatient psychotherapy. The orientation to other events was disturbed, and paranoid-like projections increased their anxiety and their fears of unexpected harmful experiences. The characteristics of their thought processes showed that they were an intelligent group who struggled to take account of each and every stimulus in a situation, even at the risk of thereby incurring disadvantages and acting without prudence. Their coping skills have been damaged and, as a group, they seem less able to organize their resources to meet stimulus demands effectively.

The Rorschach test results suggest that, on the whole, the reactions to the destructive fire persist. The anxieties have been internalized and the emotional harm remains unrelieved. The fire victims carry the scars of the damage to their ability to cope. Only one victim chose to struggle with these anxieties through psychotherapy and two had sought crisis counseling that was, by their report, highly ineffective.

Wolfenstein's (1957) analysis that individuals progress through five identifiable phases as they attempt to adapt to the impact of disasters suggests that, as a group, the fire victims became fixated in the fourth phase. They were still involved with struggles to accept the effects of the disaster and to regain mastery over their lives. To the extent that the symptoms remain fixated, each fire victim is at risk for maintaining a less than adaptive, depressive personality style that represents a characterological ossification of symptom formations. Guilt, irritability, mourning, psychosomatic ailments, heightened sensitivities, perceptual distortions and compulsions were in evidence as the victims had reconstructed their lives some eighteen months after the fire. As a group, then, they well fit the diagnostic category of the Post-Traumatic Stress Disorder described in the DSM-III-R (1989).

REFERENCES

Adler, A. (1943). Neuropsychiatric complications in victims of Boston's Coconut Grove disaster. Journal of the American Medical Association, 123, 1098-1101.

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, D.C.: Author.

Barton, A. H. (1969). Communities in disaster: a socio1ogica1 analysis of collective stress situations. New York: Doubleday.

Berren, M., Beigel, A., & Ghertner, S. (1980). A topology for the classification of disasters. Community Mental Health Journal, 17, 103-111.

Burgess, A. W. & Holstrom, L. L. (1974). Rape trauma syndrome. Archives of General Psychiatry, 13, 981-986

Chamberlin, B. C. (1980). Mayo seminars in psychiatry: The psychological aftermath of disaster. Journal of Clinical Psychiatry, 41, 238-244.

Cobb, S., & Lindemann, E. (1943). Neuropsychiatric observations during the Coconut Grove fire. Annals of Surgery, 112, 814-824.

Derogatis, L. R. (1977). SCL-90-R Baltimore, MD.: Johns Hopkins University Press.

Derogatis, L. R., Lipman, R. S., & Covi, L. (1973). SCL-90:An outpatient psychiatric rating scale - preliminary report. Psychopharmacology Bulletin, 9, 13-28.

Exner, J. E., Jr. (1986). The Rorschach: A comprehensive system. Volume 1.(2nd ed.). New York: John Wiley & Sons

Exner, J. E., Jr. (1974). The Rorschach: A comprehensive system Volume 1. New York: John Wiley & Sons.

Gleser, G. C., Green, B. L., & Winget, C. N. (1981) Prolonged psychosocial effects of disaster: A study at Buffalo Creek. New York: Academic Press.

Gleser, G. C., Green, B. L., & Winget, C. N. (1978). Quantifying interview data on psychic impairment of disaster survivors. Journal of Nervous and Mental Disease, 166, 209-216.

Goldberg, J., True, W., Eisen, S., & Henderson, W. (1990). A twin study of the effects of the Vietnam war on posttraumatic stress disorder. The Journal of the American Medical Association, 263, 1227-1232.

Green, B. L. (1982). Assessing levels of psychological impairment following disaster: Consideration of actual and methodological dimensions. Journal of Nervous and Mental Disease, 170 544-552.

Green, B. L., & Gleser, G. C. (1983). Stress and long-term psychopathology in survivors of the Buffalo Creek disaster. In D. Ricks & B. S. Dohrenwend (eds.), Origins of Psychopathology (pp. 73-90). New York: Cambridge University Press.

Green, B. L., Grace, M. C., & Gleser, G. C. (1985).Identifying survivors at risk: Long-term impairment following the Beverly Hills Supper Club Fire. Journal of Consulting and Clinical Psychology, 53, 672-678.

Green, B. L., Grace, M. C., Lindy, J. D., Titchener, J. L., & Lindy, J. G. (1983). Levels of functional impairment following a civilian disaster: The Beverly Hills Supper Club Fire. Journal of Consulting and Clinical Psychology, 51, 573-580.

Green, B. L., Lindy, J. D., & Grace, M. C. (1985). Posttraumatic stress disorder: Towards DSM-IV. Journal of Nervous and Mental Disease. 173, 406-411.

Hocking, F. (1970). Psychiatric aspects of extreme environmental stress. Diseases of the Nervous System, 31, 542-545.

Horowitz, M. J. (1976). Stress response syndromes New York: Aronson.

Ingham, J. (1981). Neurosis: Disease or distress. In J.K. Wing et al. (eds.), What is a case? The problem of definition in psychiatric community surveys. L. N. Grant McIntyre, London.

Leopold, R. D., & Dillon, H. (1963). Psycho-anatomy of a disaster. A long term study of posttraumatic neuroses in survivors of a marine explosion. American Journal of Psychiatry, 119, 913-921.

Leivesley, S. (1984). Psychological response to disaster, in Epidemiology of Natural Disasters, Volume 5: Contributions to Epidemiology and Biostatistics. Edited by Seman J., Leivesley, S., & Hogg, C. Basel Kargen Publications.

Lifton, R. J. & Olson, E., (1976). The human meaning of total disaster: The Buffalo Creek experience. Psychiatry, 39, 1-18.

Lindy, J. D., Grace, M. C., & Green, B. L. (1981). Survivors: Outreach to a reluctant population. American Journal of orthopsychiatry, 51, 468-478.

Lindy, J. D., Green, B. L., Grace, M. C., & Titchener, J. L. (1983). Psychotherapy with survivors of the Beverly Hills Supper Club Fire. American Journal of Psychotherapy, 37, 593-610.

Logue, J. M., Melick, H. E., & Hansen, H. (1981). Research issues and directions in the epidemiology of health factors of disasters. Epidemiology Review, 3, 140-162.

Madakasira, S., & O'Brien, K. (1987). Acute posttraumatic stress disorder in victims of a natural disaster. Journal of Nervous and Mental Disease, 175, 286-290.

McFarlane, A. C. (1989). The aetiology of post-traumatic morbidity: Predisposing, precipitating factors. British Journal of Psychiatry, 154, 221-228.

McFarlane, A. C. (1988a). The phenomenology of posttraumatic stress disorders following a natural disaster. Journal of Nervous and Mental Disease, 176, 22-29.

McFarlane, A. C. (1988b). The longitudinal course of posttraumatic morbidity: The range of outcomes and their predictors. Journal of Nervous and Mental Disease, 176, 30-39.

McFarlane, A. C. (1988c). The aetiology of post-traumatic stress disorders following a natural disaster. British Journal of Psychiatry, 152, 116-121.

McFarlane, A. C. (1987). Life events and psychiatric disorder: The role of a natural disaster. British Journal of Psychiatry. 151, 362-367.

McFarlane, A. C., (1986a). Long term psychiatric morbidity after a natural disaster: Implications for disaster planners and emergency services. Medical Journal of Australia, 145, 561-563.

McFarlane, A. C. (1986b). Posttraumatic morbidity of a disaster: A study of cases presenting for psychiatric treatment. Journal of Nervous and Mental Disease, 174, 4-14.

McFarlane, A. C. (1985). The effects of stressful life events and disaster: Research and theoretical issues. Australian and New Zealand Journal of Psychiatry, 19, 409-421.

McFarlane, A.C., & Raphael, B. (1984). Ash Wednesday: The effects of a fire. Australian and New Zealand Journal of Psychiatry, 19, 913-921.

Rangell, L. (1976). Discussion of the Buffalo Creek disaster: The course of psychic trauma. American Journal of Psychiatry, 133, 313-316.

Raphael, B. (1984). Psychosocial aspects of disasters: Some Australian studies and the Ash Wednesday bushfires. Medical Journal of Australia, 141, 268-270.

Raphael, B., & Middleton, W. (1987). Mental health responses in a decade of disasters: Australia, 1974-1983. Hospital and Community Psychiatry, 38, 1331-1337.

Tennant, C. (1983). Life events and psychological morbidity: the evidence from prospective studies. Psychological Medicine, 13, 483-486.

Titchener, J. L, & Kapp, F. T. (1976). Family and character change at Buffalo Creek. American Journal of Psychiatry, 13, 295-299.

Wolfenstein, M. (1977). Disaster: A Psychological Essay. New York: Arno Press.