Acute stress response patterns to accidental injuries

Acute stress response patterns to accidental injuries

Journal of Psychosomatic Research, Vol. 45, No. 5, pp. 419–424, 1998 Copyright  1998 Elsevier Science Inc. All rights reserved. 0022-3999/98 $–see fr...

58KB Sizes 1 Downloads 91 Views

Journal of Psychosomatic Research, Vol. 45, No. 5, pp. 419–424, 1998 Copyright  1998 Elsevier Science Inc. All rights reserved. 0022-3999/98 $–see front matter

S0022-3999(98)00008-7

ACUTE STRESS RESPONSE PATTERNS TO ACCIDENTAL INJURIES ULRICH SCHNYDER and ULRIK FREDRIK MALT (Received 19 August 1997; accepted 28 November 1997) Abstract—We examined, by means of clinical interviews and several self-report measures (IES, STAI), 110 accident victims’ primary appraisal of injury during five different timepoints before, during, and immediately after the accident. The appraisals were combined with the corresponding most dominant emotion in a matrix, revealing six different acute stress response patterns: ordinary (O: 64 subjects—58%); emotional (E: 19 subjects—17%); controlled (C: 9 subjects—8%); derealization (D: 15 subjects—14%); and denial–elation (DE: 3 subjects—3%). The E response pattern was associated with being female (RR 3.31). Psychopathology at the time of the injury or risk of death during the accident increased the risk for a D or E response (RR 1.61 and RR 1.92, respectively). The presence of psychophysiological symptoms or reduced appetite during the hospital stay was associated with E, C, or D response patterns compared with an O pattern (RR 1.76 and RR 2.18, respectively). A DE response was associated with severe brain injury partly undetected by the surgeons. We conclude that the identification of different clinical response patterns may be a meaningful approach to better tailor response-specific interventions for trauma victims.  1998 Elsevier Science Inc. Keywords:

Accidents; Acute stress responses; Injury; Posttraumatic stress disorder; Trauma

INTRODUCTION

Accidental injuries may cause short- and long-term psychological impairment and distress [1–6]. Despite these findings, the results of randomized psychological treatment studies are far from encouraging [7–9]. The lack of effective therapy may partly be due to insufficient knowledge of the phenomenology of the acute psychological responses to the physical trauma in accident victims [10–12]. Thus, clinically valid response patterns to accidental injuries can only be identified by bedside studies of a random sample of patients. The aim of the present article is to analyze data from a previous, carefully conducted clinical phenomenological study to identify clinically meaningful response patterns. METHOD Sample One hundred ten patients, 15–69 years of age, were admitted to a surgical department for accidental injuries and included consecutively from an equal number of each weekday to avoid sampling biases associated with the fact that week-end victims are psychologically different from work-day victims. The Psychiatrische Poliklinik, Universita¨tsspital, Zu¨rich, Switzerland. Address correspondence to: Dr. Ulrich Schnyder, Psychiatrische Poliklinik, Universita¨tsspital, Ra¨mistrasse 100, 8091 Zu¨rich, Switzerland. Phone: 41-1-255 52 80; Fax: 41-1-255 43 83; E-mail: uschnyd@ psyp.unizh.ch

419

420

U. SCHNYDER and U. F. MALT

mean Injury Severity Score [13, 14] was 8.6 (sd 4.9, range 1–35). Forty-six percent were injured in motor vehicle accidents. The majority of patients were males (77.3%) and about half the patients were less than 30 years of age. Twenty-seven patients had been unconscious for several minutes up to 24 hours and 18 patients had possibly been unconscious for a few seconds. Twenty-three patients had a definite retrograde amnesia and 13 patients a probable retrograde amnesia. The mean time from the injury to the interview in the whole group was 84 hours (sd 19.2 hours). Twenty-five percent of the patients were seen within 24 hours following the accident and 78.5% within 96 hours. Details about the sampling procedure, background of the patients, and the injuries sustained have been presented elsewhere [1, 15, 16].

Assessments Assessments included clinical interviews and the Comprehensive Psychopathological Rating Scale (CPRS) [17]. To cover all symptoms needed for the diagnosis of anxiety disorders and stress response syndromes, several items were added to the CPRS interview: startle reactions; intrusive thoughts; bad dreams; rituals related to the injury or the place of the injury; isolation; phobia related to the accident; pessimistic thoughts about the injury; survivor guilt; and vigilance. The patients also filled out the General Health Questionnaire (GHQ) regarding the last fortnight prior to the injury [18]. For valid assessment of preaccident psychopathology and distress, we considered information from several different sources, partly patient-independent (e.g., health registers, prior medical records), when preaccident psychopathology was assessed [1, 16]. If all this information gave any indication of a mental disorder in the past, or problems related to social or occupational functioning, including poor object relations, items from the Schedule of Interviewing Borderlines [19] covering DSM-III personality disorders were applied. Extensive case histories were written for each patient and reviewed by a second psychiatrist [1], to have a quality control of the diagnostic process.

Assessment of clinical response patterns The assessment of the clinical response patterns was based on Lazarus and coworkers’ theory of coping and appraisal and Horowitz’s studies of stress response patterns. The interview assessed the patients’ reactions at five different timepoints (1—the period from when the patient became aware of the threat until the moment of the injury; 2—the immediate response to the injury; 3—the period between the immediate reaction and admission to hospital; 4—the period when the patient was admitted to the surgical ward; and 5—the time of the interview). For each timepoint, the interview independently assessed the patient’s appraisal of the situation, the focus of the cognitive activities, the concomitant emotional experience (e.g., anxiety, sadness, anger), and the behavior (e.g., no action, shouting, running around). The appraisal responses at each point in time were classified as death, disability, threat/harm, no warning/harm, or amnesia. The primary focus of the cognitive activities was mostly the injury, but there was also some focus on the implications of the injury, responsibility issues, and non-personal related damage, particularly at time 2 [15]. The concomitant emotions were classified as anxiety, severe anxiety close to panic, dissociative responses (derealization/depersonalization), irritability, depression or sadness, guilt, relief, elated mood, or no particular dominating emotion. Dissociative responses were assessed by item 27 (derealization) and item 28 (depersonalization) on the CPRS. The type and severity of the injury was assessed by two experienced trauma surgeons using the Abbreviated Injury Scale (AIS) and Injury Severity Score (ISS) [1, 20, 21]. The ISS is a very well accepted index for severity of physical trauma. An ISS score ⬎10 usually defines patients with multiple injuries or a severe single injury. A 12-item version of the Spielberger State Anxiety Inventory (STAI) was used to assess anxiety [22] and the Impact of Event Scale (IES) to assess intrusion and avoidance [23]. A score of 0–8 was considered to reflect a low level of distress, a score of 9–19 a medium level, and a score of ⭓20 a high level of distress [11]. The interrater reliabilities of all variables were assessed in a random sample of 18 interviews. These analyses demonstrated acceptable to very good interrater reliability for all assessments [15, 16].

Identifying and validating stress response patterns First, the reports about appraisals, focus of cognitive activities, prominent emotions, and behavior at the five different timepoints were combined for each of the 110 patients to identify the unique pattern of response of each patient over time. It turned out that the combination of appraisal and emotion best differentiated the responses and gave the most meaningful clinical response patterns. Second, these response patterns were compared with sociodemographic and preaccident health variables and patient scores on self-reported stress measures.

Statistical procedure Chi-square tests were used to analyze contingency tables. A value of p⬍0.05 was considered to represent statistical significance. Fisher’s exact test was used when an expected cell value was ⬍5. Relative

Acute stress response patterns

421

risk (RR) of showing a particular response pattern and Taylor series 95% confidence limits for RR were calculated with EPI Info, version 5 (Dean J, et al., Center for Disease Control, Epidemiology Program Office, Atlanta, GA, 1990).

RESULTS

A cross-classification between the appraisals reported during the five different timepoints with the dominant emotions experienced at the same time periods revealed five major response patterns among the 110 subjects: the ordinary response pattern (n⫽64; 58%); the emotional response pattern (n⫽19; 17%); the controlled fear response pattern (n⫽9; 8%); and the derealization response pattern (n⫽15; 14%). Three patients (3%) showed a denial–elation response pattern. The latter group did not appraise the accident or injury as very harmful or report any specific emotions. These responses occurred despite the fact that they had sustained serious physical injuries involving the head (ISS ⬎10) with death as a possible outcome. Two of these patients even showed elation. None of them had ever suffered from a psychiatric disorder. One person with denial response developed posttraumatic seizures later. In another patient the severity of the injury was underestimated clinically and first acknowledged at follow-up. Although clinically significant, due to the low number of denial–elation responses, this response was removed from the comparison analyses. The ordinary response pattern was characterized by a combination of no specific emotions or moderate degree of worry or anxiousness during the five different time periods. The corresponding appraisals were no danger or no harm (initially and/or at the interview) or moderate danger or harm. All patients who reported feelings of irritability or anger at time 1 or 2 were included in this group. The risk of not showing an ordinary response pattern was associated with absence of minimal danger during the accident (RR 2.14 [95% CI 1.23–3.73]). The presence of psychophysiological stress symptoms and reduced appetite during the observation period were associated with an increased risk of showing a nonordinary stress response (psychophysiology: RR 1.76 [1.16–2.67]; reduced appetite: RR 2.18 [1.21–3.93]). The patients showing an emotional response pattern either reported strong anxiety during the observation period (n⫽9; 8%) or marked dysphoric–depressed emotions (n⫽10; 9%). Two subtypes of anxiety were seen. Four initially experienced overwhelming anxiety (the overwhelming anxiety subtype) and five patients showed increasing anxiety in the course from the warning period to hospital admittance (increasing anxiety subtype). Despite the responses, none of these patients were injured in accidents with impending death threat, but death could have been a possible, although not probable, outcome in five cases. Only two of these patients had sustained a severe injury (ISS ⬎10). All patients with an anxiety response subtype characterized by overwhelming anxiety qualified for a psychiatric diagnosis according to ICD-9 [24] or DSM-III [25] at the time of the accident. The early appraisal of the accident or injury made by the dysphoric–depressed patients varied. They either suffered from a psychiatric disorder when injured (four patients), were in a state of bereavement (one patient), suffered from a serious illness when injured (one patient), or experienced the accident or the injury as a threat to their selfimage (four patients). The risk of showing an emotional response pattern compared with the other re-

422

U. SCHNYDER and U. F. MALT

sponse patterns was associated with being female (RR 3.31 [1.74–6.29]). This response pattern was also more often seen when the patients had marked inner tension (RR 2.65 [1.30–5.40]); worries about the injury (RR 2.55 [1.48–4.39]); compulsive thoughts about the injury (RR 2.98 [1.52–5.85]); dreams about the accident/ injury (RR 4.63 [1.27–16.95]); and pessimistic thoughts (RR 3.86 [1.31–11.35]). Although not statistically significant, the ten patients demonstrating dysphoric– depressed response tended to show higher mean intrusion scores on the IES and higher mean STAI X-1 scores than those who reported mostly anxiety (IES intrusion 14.1 [sd 5.5] vs. 10.6 [sd 7.2] and STAI X-1 28.0 [sd 4.8] vs. 22.4 [sd 5.0], respectively). The IES avoidance score demonstrated an opposite relationship, however (anxiety responders mean score 18.4 [sd 9.4] vs. 11.9 [sd 5.6] for depressive responders). Compared with the other response patterns, however, the emotional response pattern was associated with significantly more medium and high scores on the IES intrusion (RR 5.10 [2.83–9.18]) and avoidance (RR 2.33 [1.65–3.29]) subscales. The controlled fear response patients appraised the accident or injury as deadly or disabling. However, none of these patients were injured in accidents likely to cause death according to medical evaluation. Despite the serious appraisal of the accident or injury, they did not experience any strong emotions. The controlled fear response pattern was similar to the ordinary response pattern on most background, accident, and clinical variables except for the more frequent absence of minimal danger during the accident. The controlled fear response patients also were more similar to the other nonordinary response patterns with regard to the frequency of psychophysiological symptoms during the observation period, pain reports, and lack of appetite. All 15 patients classified as derealization response subjects experienced depersonalization at least once during the five different timepoints. Their appraisals varied from appraising the injury as harmful to disabling or deadly. As a group they were similar to the emotional response pattern with regard to the frequency of bad childhood, psychopathology at the time of the injury, and possibility of death during the accident. In fact, the presence of psychopathology or death being possible during the accident significantly increased the risk of showing a derealization or emotional response (RR 1.61 [1.02–2.54] and RR 1.92 [1.15–3.2], respectively). Following the experience of derealization these patients reported only moderate worry or anxiousness. Twelve of the 15 patients were under 30 years of age. Two of the patients experienced the derealization when they regained consciousness following a cerebral concussion. Patients who sustained a cerebral concussion but showed an ordinary response did not report derealization when they regained consciousness.

DISCUSSION

To our best knowledge, this is the first comprehensive, prospective clinical study of accidentally injured adults, describing clinical reponse patterns as they evolve over a short period of time before, during, and after an accident. This study demonstrates that, when carefully interviewing about the appraisal of the injury and the corresponding emotions experienced from the time of the accident until the admittance to the ward, clinically meaningful response patterns may be isolated. Our results suggest two quite opposite major response patterns, the ordinary re-

Acute stress response patterns

423

sponse pattern and the emotional response pattern. These response patterns are well known from previous clinical descriptions of response to accidental injury. We believe the identification of two response patterns falling in between the two extreme opposite responses, namely the controlled response pattern and the derealization response pattern, are of particular clinical interest. These two response patterns share the physical symptomatology (psychophysiology, appetite) of the emotional group, but not the psychological distress symptoms (inner tension, worry, intrusion, avoidance), and thus point to the presence of a psychosomatic response pattern between the ordinary response and the clear-cut emotional response patterns. Furthermore, more psychopathology and real danger were observed in the derealization group. This indicates that any effort to intervene must be based on individual identification of the most important sources determining the stress response. Accordingly, providing one type of therapy to all victims would most likely fail [9]. The limited number of subjects emphasizes the need for more studies addressing these issues. So far, the present results must be considered preliminary. Based on our extensive clinical experience with the treatment of accident victims, however, we believe the following implications of our results should be considered in clinical care. If a person responds with elation to an accidental injury (denial), the possibility of a severe head injury should be considered. The same response has been described following heart surgery [26], but has been attributed to psychological processes only. None of the patients with a denial–elation response appeared manic from the standpoints of DSM-III-R, DSM-IV, or ICD-10. The possibility that some of these responses could nevertheless reflect a secondary manic response due to brain injury [27], but modified by the treatment of the injury during the hospital stay, cannot be ruled out, however. A response pattern characterized by a severe primary appraisal but no corresponding strong emotions (controlled fear response pattern) is typically seen among males. Clinically, such patients may be prone to somatize their distress. To decrease the risk for somatic complications during the hospital stay, early intervention should be considered if the psychophysiological symptoms are marked. The observation that the majority of those patients have sustained only minor injuries emphasizes the importance of inquiring about the personal meaning of a trauma to assess the psychological impact. The derealization response pattern is typically observed among young individuals involved in a frightening accident. The relatively high percentage of patients with psychopathology and/or bad childhood environment may indicate that this response mirrors immaturity. If derealization symptoms are marked, or do not disappear quickly, psychosocial intervention may be warranted to decrease the risk of longterm problems. Patients with marked anxiety or dysphoric–depressed response patterns should be offered psychological support. When treating these subjects, factors present before the accident or the injury should be considered. These suggestions fit with the findings of other research groups [2, 28–30]. Acknowledgments—This study was supported by the Norwegian Research Council for Sciences and the Humanities, the University of Oslo, the Norske Folks Stress Fund, and the Anders Jahres Foundation and also by legacies administered by the University of Oslo.

424

U. SCHNYDER and U. F. MALT

REFERENCES 1. Malt U. The long-term psychiatric consequences of accidental injury. A longitudinal study of 107 adults. Br J Psychiatry 1988;153:810–818. 2. Blanchard EB, Hickling EJ, Taylor AE, Loos WR, Gerardi RJ. Psychological morbidity associated with motor vehicle accidents. Behav Res Ther 1994;32:283–290. 3. Mayou R, Bryant B. Duthie R. Psychiatric consequences of road traffic accidents. BMJ 1993; 307:647–651. 4. Blanchard EB, Hickling EJ, Taylor AE, Loos WR. Psychiatric morbidity associated with motor vehicle accidents. J Nerv Ment Dis 1995;183:495–504. 5. Schnyder U, Buddeberg C. Psychosocial aspects of accidental injuries—an overview. Langenbecks Arch Chir 1996;381:125–131. 6. Eitinger L, Strøm A. Mortality and morbidity after excessive stress: a follow-up investigation of Norwegian concentration camp survivors. New York: Humanities Press 1973. 7. Roth S, DeRosa RR, Turner K. Cognitive-behavioural interventions for PTSD. In: Giller EL, Weisæth L, eds. Post-traumatic stress disorder, 1st ed, vol 2. London: Baillie`re 1996:282–296. 8. Marmar CR, Weiss DS, Pynoos RS. Dynamic psychotherapy of PTSD. In: Giller EL, Weisæth L, eds. Post-traumatic stress disorder, 1st ed, vol 2. London: Baillie`re 1996:297–316. 9. McFarlane AC. Individual psychotherapy for post-traumatic stress disorder. Psychiatric Clin N Am 1994;17:393–408. 10. Bryant RA, Harvey AG. Initial posttraumatic stress responses following motor vehicle accidents. J Trauma Stress 1996;9:223–234. 11. Horowitz MJ. Stress response syndromes and their treatment. In: Goldberger L, Breznitz S, eds. Handbook of stress. Theoretical and clinical aspects, 1st ed. New York: Free Press 1982:711–732. 12. Weisæth L. A study of behavioural responses to an industrial disaster. Acta Psychiatrica Scand 1989;355(suppl):13–24. 13. Baker SP, O’Neill B, Haddon W. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma 1974;14:187–196. 14. Baker SP, O’Neill B. The injury severity score: an update. J Trauma 1976;16:882–885. 15. Malt UF, Olafsen OM. Psychological appraisal and emotional response to physical injury: a clinical, phenomenological study of 109 adults. Psychiatric Med 1992;10:117–134. 16. Malt U, Myhrer T, Bilkra G, Høivik B. Psychopathology and accidental injuries. Acta Psychiatrica Scand 1987;76:261–271. ˚ sberg M, Perris C, Schalling D, Sedvall G. The CPRS: development and applications of a psychiat17. A ric rating scale. Acta Psychiatrica Scand 1978;60(suppl 271). 18. Malt UF. The validity of the General Health Questionnaire in a sample of accidentally injured adults. Acta Psychiatrica Scand 1989;355(suppl):103–112. 19. Baron M, Gruen R. Schedule for interviewing borderlines. New York: New York State Psychiatric Institute 1980. 20. Committee on Injury Scaling. The Abbreviated Injury Scale, 1980 revision. Park Ridge, Illinois: American Association of Automotive Medicine 1980. 21. Greenspan L, McLellan BA, Greig H. Abbreviated Injury Severity Score: a scoring chart. J Trauma 1985;25:60–64. 22. Spielberger CD, Barker L, Russel S, Crane SD, Westberry L, Knight J. Preliminary manual of the State–Trait Personality Inventory (STPI). Tampa, Florida: University of South Florida, Department of Psychology 1979. 23. Horowitz MJ, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med 1979;41:209–218. 24. World Health Organization. Mental disorders: glossary and guide to their classification in accordance with the ninth revision of the International Classification of Diseases. Geneva: WHO 1978. 25. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 3rd ed. Washington, DC: APA 1980. 26. Kimball CP. Psychological responses to the experience of open heart surgery. Am J Psychiatry 1969;126:248–359. 27. Jorge RE, Robinson RG, Starkstein SE, Arndt SV, Forrester AW, Geisler FH. Secondary mania following traumatic brain injury. Am J Psychiatry 1993;150:916–921. 28. McFarlane AC, Papay P. Multiple diagnoses in posttraumatic stress disorder in the victims of a natural disaster. J Nerv Ment Dis 1992;180:498–504. 29. McFarlane AC. Abnormal stimulus processing in posttraumatic stress disorder. Biol Psychiatry 1993;34:311–320. 30. Spurrell MT, McFarlane AC. Life-events and psychiatric symptoms in a general psychiatric clinic: the role of intrusion and avoidance. Br J Med Psychol 1995;68:333–340.