Psychosomatics 42:35-40, February 2001
© 2001 The Academy of Psychosomatic Medicine
Linguistic Analysis to Assess Medically Related Posttraumatic Stress Symptoms
Richard J. Shaw, M.B., B.S.,
John E. Harvey, M.D., F.R.C.P,
Kristin L. Nelson, Ph.D.,
Rachel Gunary, M.A., Ms.C. Clin Psychol,
Helen Kruk, R.G.N., and
Hans Steiner, M.D.
Received April 11, 2000; revised July 11, 2000; accepted September 14, 2000. From Stanford University School of Medicine, Stanford, California and Health Services NHS Trust, Bristol, England. Address correspondence and reprint requests to Dr. Steiner, Division of Child Psychiatry, Stanford University School of Medicine, 401 Quarry Road, Stanford, CA 94305–5719.
|
ABSTRACT
|
The authors examined the presence of posttraumatic stress symptoms (PTSS) in 20 patients requiring ventilation after acute respiratory distress. The subjects completed a semistructured interview about their ventilation experience that was subject to content and linguistic analysis. Subjects also completed two self-report measures to assess PTSS and socioemotional adjustment. Subjects who endorsed PTSS were more likely to use a narrative style suggesting emotional involvement in their recall of the stressful event. The authors indicate that the presence of PTSS is a common consequence of traumatic medical experiences and that denial of distress may be an adaptive short-term coping strategy.
Key Words: Posttraumatic Stress Disorder • Linguistic Analysis
|
INTRODUCTION
|
Developments in medical technology have led to greatly improved rates of survival in many medical conditions, although often at the cost of patients being subjected to invasive medical procedures. There is a growing interest in the potential psychological sequelae of these procedures and recognition that some patients may develop symptoms of posttraumatic stress disorder (PTSD), where the traumatic event is the experience of the medical intervention. In this study, we examine the presence of posttraumatic stress symptoms (PTSS) in a group of patients who have undergone emergency intubation as a consequence of acute respiratory failure. We also examine the relationship between PTSS and subjects' self-report on their habitual tendency to use denial of distress as a coping strategy.
Medical Illness as a PTSD Stressor Event
There is substantial literature documenting that PTSD can develop as a consequence of physical trauma and medical illness.1 Of patients surviving burn injuries, 22%–45% report symptoms of PTSD and rates increase in the year after the initial injury.2,3 There has been additional research in cancer survivors where studies have found significant rates of PTSD symptoms in both patients and family members.4–6 In a study by Stuber et al.,7 children with leukemia reported that the traumatic memories that lingered the longest were those of the treatment rather than the illness itself.
To date, there has been little research on the phenomenon of PTSD in patients with acute respiratory distress syndrome (ARDS), specifically, patients who require emergency intubation and ventilation as a consequence of acute respiratory failure. Schelling et al.,8 in the only published study of this kind, studied 80 patients who survived an episode of acute pulmonary failure. Stressors named in this study included potentially traumatizing episodes of respiratory distress, anxiety, or pain during their hospital treatment. Of patients who had PTSD symptoms, 27.5% were able to recall two or more memories of adverse medical events.
Adaptive Coping Mechanisms in PTSD
The finding that the incidence of clinical diagnoses of PTSD in survivors of various traumatic events averages 30% has led to interest in the factors that promote both resilience and vulnerability to trauma symptoms.9 Relevant to this discussion is the literature related to coping style and PTSD.10,11 Problem-focused coping is described as an approach in which individuals take on an active and assertive role in confronting the stressor. There is evidence to suggest that this may be an adaptive mechanism in specific situations (e.g., combat situations).12 By contrast, emotion-focused coping is characterized as a passive approach in which individuals manage their anxiety in response to an acute traumatic event by using coping mechanisms such as denial of distress, which we examine in our study, and dissociation. Emotion-focused coping is thought to be more adaptive in certain situations, such as child sexual abuse, where assertive behavior may provoke retaliation by the perpetrator.12
Linguistic Analysis
In an effort to better understand the finding of PTSS following medical trauma and how subjects try to manage these symptoms, we apply a method of linguistic analysis to subjects' recollections of their ventilation experience. Linguistic analysis of speech transcripts may be used to infer variations in the individuals' internal state of mind as well as to provide information about their defensive operations .34
The study of spontaneous narration of previous experiences shows that individuals employ distinctive structural elements that have been found to generalize across both cultures and languages.13,14 Narrative is described as having a dual or "layered" action.15 It is simultaneously a negotiation of an interpersonal situation, in which the narrator explains the relevance of the story to listeners, referred to as "speaker-now," as well as a "joint imagining," in which the narrator guides listeners through the text of the story and recreates the fictional world in which they first had the experience, referred to as "story-now." Clark15 uses a stage metaphor to clarify this distinction. The real people, narrator and listeners, during speaker-now are like the theatrical audience. They are expected to suspend their involvement in this actual world and instead imagine that the events taking place on stage, during the story-now section, are actually really happening. The following vignette illustrates the shift from speaker-now to story-now.
Interviewer: First of all, I'd like to ask you if there is anything you remember about the intensive care unit.
Patient: (Speaker-now) Well, this really was one of the most awful experiences I have ever been through. In fact, I can't remember ever being so afraid or out of control at any other time in my life. I still get nightmares and really find it quite difficult to talk about, even here. (Story-now) When I first woke up, I wasn't even sure that I was still alive. I remember clearly the feeling of the tube in my throat. At one point, I tried to pull the tube out, and one of the nurses told me not to, and the next thing, they were tying my hands down to stop me. I couldn't move, and worst of all, I couldn't talk. I was convinced I was going to die. (Speaker-now) Believe me, you never want to go through something like that.
One identifying feature of story-now discourse is a temporal sequencing of the narrator's spoken clauses so that they match the plot sequence of the story told (i.e., the narrator tells the story by recounting a sequence of events—A happened, then B happened, then C happened, etc.) that occur in exactly the same order in the narrative as they occurred in the original experience. In the above example, the patient recounts during the speaker-now section the following sequence of events: I woke up, and then I felt the tube, and then I tried to pull the tube out, and then the nurse told me not to, and then they tied my hands down, etc. A way of measuring this temporal ordering is the temporal juncture. Labov and Waletzky16 defined this as the separation between two temporally ordered clauses or the equivalent of the temporal conjunction then.
In our study, we hypothesize that during the story-now domain (i.e., themselves during a past traumatic experience), subjects are in a state closer to reexperiencing the past episode than during the speaker-now domain of merely describing it. There are several reasons for this assumption. First, it is only within the story-now domain that the speaker's reference point becomes the past (e.g., "I" and "now" no longer refer to the subject in the interview room, but to the individual in the past, going through a traumatic experience).15,17 Speaker and listener are thus jointly imagining that it is the past. Discourse in the story-now domain is dramatic in that it recapitulates the uncertainty and lack of control that occurs between events in real life. Finally, computer text analysis of large natural speech corpora shows that Referential Activity, defined by Mergenthaler and Bucci18 as "connecting nonverbal experience, including emotional experiences, with language," is higher in narrative language than in nonnarrative texts. Because temporal junctures occur only within the story-now domain, we use the count of the number of temporal junctures in our study as a measure of immersion in the story-now, and thus, a measure of the degree of emotional involvement of the subjects in the recollection of their intubation experience.
Hypotheses
In our study, we ask individuals who have been exposed to stressful and potentially traumatic medical experiences to recall and describe their recollection of the medical events. Our primary hypothesis is that individuals who have intrusion PTSS will relate their experiences in a vivid manner, using a linguistic style (story-now) that suggests they are emotionally aroused and close to their original experience. Our secondary hypothesis is that individuals who endorse the habitual use of denial of distress as a coping strategy will be more likely to relate their experience using a linguistic style that reflects emotional distancing from their original experience.
Our specific hypotheses include the following. 1) Subjects who have a high recall of traumatic memories and stressful experiences will likely endorse intrusion PTSS as rated using the Impact of Event Scale (IES).
2) Subjects who endorse PTSS will more likely describe their traumatic experience in a detailed descriptive manner (i.e., high number of stories) and use a narrative style that suggests the subjects are emotionally involved in the memories being recounted (i.e., high number of temporal junctures).
3) Subjects who endorse the use "Denial of Distress" as a habitual coping strategy, as rated using the Weinberger Adjustment Inventory (WAI), will less likely endorse intrusion PTSS as rated using the IES.
4) Subjects who use Denial of Distress, assessed using the WAI, will describe their traumatic experience in a less descriptive manner (i.e., low number of stories) manner and use a narrative style that suggests the subjects are emotionally distanced from the memories being recounted (i.e., low number of temporal junctures).
|
METHODS
|
Subjects
We studied 20 patients with diagnoses of acute respiratory distress syndrome who were ventilated for a minimum period of 48 hours in the intensive care unit (ICU) as a consequence of their respiratory illness. Subjects completed a semistructured recorded interview about their ventilation experience and completed two self-report measures, the IES and the WAI. Informed consent was obtained from all subjects for the study.
Measures
Transcript Content Analysis. Transcripts of interviews were rated on a Likert scale of 1 to 5 on the following dimensions: Level of Distress Experienced by Subject; Degree of Anxiety; Traumatic Memories Concerning Illness or Treatment; Themes of Powerlessness or Lack of Control; and Themes of Death.
Narrative Analysis. In the structural linguistic analysis, interviews were rated on the following dimensions: Number of Stories, defined as the number of different tales recounted in a temporal order (i.e., with at least one temporal juncture); and Number of Temporal Junctures/Narrative. To perform the narrative analysis, transcripts are first segmented into independent (main) clauses, identified by bracketed lower-case letters [a], [b], [c], etc. Clauses related to the actual interpersonal speech situation between the interviewer and subject are eliminated. These include summary introductory and concluding statements usually made in the speaker-now domain. The remaining text is examined to determine which clauses are temporally ordered, that is clause [x] cannot be placed after clause [y] without changing the meaning of the narrative. The number of temporal junctures occurring within the narrative is counted where the term temporal juncture indicates the separation between two such temporally ordered clauses. If the transcript contains at least two clauses judged as a chronological sequence of events, containing one temporal juncture, it is designated as a story.
IES. The IES19 is a 15-item self-report measure that identifies symptomatic response to specific traumatic stressors, and in its original version, it taps the B and C criteria used to diagnose PTSD.20 Intrusion symptoms comprise nightmares, intrusive thoughts, or images, while avoidance symptoms comprise attempts to dampen or to avoid experiences associated with the traumatic event, and its associated numbing of responsiveness. The IES has sound psychometric properties, correlates well with other PTSD measures, and has good internal consistency and test-retest reliability.20 Although the IES cannot be used to diagnose PTSD, because it does not assess the full spectrum of symptoms, cut-off scores have been used to identify subjects likely to meet the criteria for PTSD (recommended cut-off scores of 30 for the total score, and of 20 for the Intrusion and Avoidance subscales).21
The WAI. The WAI is an 84-item self-report questionnaire that is used to assess long-term socioemotional adjustment.22,23 The WAI assesses the individual's habitual mode of coping and, as such, is a trait measure, but a measure that is likely to predict how the individual responds during periods of acute stress. The WAI consists of two primary dimensions of distress (with subscales of Anxiety, Depression, Low Self-Esteem, and Low Well-Being) and Self-Restraint (with subscales of Impulse Control, Suppression of Aggression, Consideration of Others, and Responsibility). In addition, there are two Defensiveness scales measuring different aspects of defensive functioning. The Denial of Distress subscale, which refers to defensiveness about normative experiences of distress and is derived from 11 items, measures the refusal to acknowledge the negative emotional significance of events or their subjective experience. In this study, we use this as a measure of the defense mechanism of denial.24 The Repressive Defensiveness subscale, which refers to claims of nearly absolute restraint and is also derived from 11 items, assesses the tendency to extreme self-restraint, the suppression of egoistic desires, and the extreme control of negative emotional responses. Psychometrics of the WAI are excellent and normative data on an independent sample of subjects are available for different age groups.22,23
|
RESULTS
|
Subjects' mean±standard deviation (SD) IES total score was 22.0±15.4. Seven of the 20 subjects (35%) had scores of >30 on the IES, the recommended cut-off for subjects likely to meet the clinical criteria for PTSD. Because our interest is in intrusion PTSS, the Intrusion subscale is used for the subsequent correlational analyses.
Spearman correlation coefficients were calculated between scores on the IES Intrusion subscale and those for the Transcript Content analysis. There are significant correlations between scores on the IES Intrusion subscale and Transcript Content in the narratives related to themes of Distress, Lack of Control, Death, and Traumatic Memories of the Illness (all P values<0.05). This is consistent with our first hypothesis that subjects who have a high recall of traumatic memories and stressful experiences are more likely to endorse intrusion PTSS.
Table 1 shows the Spearman correlation coefficients between scores on the IES Intrusion subscale and the linguistic analysis. There are significant correlations between scores on the IES Intrusion subscale and number of stories and number of temporal junctures (P values<0.05). This is consistent with our second hypothesis that subjects with intrusive PTSD symptoms will describe their traumatic experience in a detailed and descriptive (high number of stories) manner, with a narrative style that suggests the subject is emotionally involved and close to the memories being recounted (high number of temporal junctures).
Table 1 also shows the Spearman correlation coefficients between scores on the WAI and scores on the IES Intrusion subscale and linguistic analysis. There is a significant negative correlation between WAI Denial of Distress and scores on the IES Intrusion subscale (P<0.05). This is consistent with our third hypothesis that subjects who use Denial of Distress as a coping strategy are likely to have fewer symptoms of psychological distress (as rated on the IES). There are significant negative correlations between WAI Denial of Distress and number of stories and number of temporal junctures (P values<0.05). This is consistent with our hypothesis that subjects who use Denial of Distress as a coping strategy will relate their memory of the traumatic event in a style that suggests emotional distancing from their traumatic memories. Parenthetically, we found a significant positive correlation between IES Intrusion subscale scores and the WAI Distress score, where high scores indicate the habitual tendency to report general states of anxiety, depression, low self-esteem, and low emotional well-being. This correlation supports findings from the PTSD literature that chronic elevation in symptoms of anxiety and depression are associated with the PTSD symptoms, either as a risk factor, or as comorbid psychiatric disorders.25
|
DISCUSSION
|
Our study builds on the existing literature to suggest that PTSS are important and often unrecognized sequelae of traumatic medical experiences. The finding of our study that 35% of subjects had IES scores indicating likely clinically significant PTSS is consistent with previous studies of PTSD in subjects with medical illnesses, including postventilation patients.1,8 Our study is also consistent with findings by Schelling et al.8 in their study of ARDS patients. Schelling at al. found that subjects with traumatic memories of their intubation and treatment experiences were more likely to report PTSS.
Our study is the first, to our knowledge, to use the technique of linguistic analysis of transcripts of a traumatic medical experience to examine the relationship between traumatic stress reactions and narrative style. Our primary measure of narrative style, the temporal juncture, suggests that subjects are reliving their traumatic experience as it is narrated. Our study demonstrates that patients who endorse intrusion PTSS on a self-report measure are more likely, when asked to talk about their trauma, to retell their story in an expressive and emotional manner, indicating involvement and emotional arousal. This suggests that the tendency to experience intrusive recollections of the event is reflected in the individual's narrative style.
Linguistic analysis is useful in clarifying the potential mechanism by which denial functions to moderate the development of PTSS. In our study, subjects who endorse the habitual use of Denial of Distress were less likely to report intrusion PTSS. These subjects were also found to provide narratives that were shorter and less descriptive and that had a narrative style suggesting less emotional involvement in the traumatic memories. One possible explanation for this finding is that Denial of Distress protects individuals from experiencing traumatic symptoms and that this is reflected in their distancing narrative style. However, it is also possible that the construction of the highly constricted narrative in itself, by minimizing the activation of traumatic memories, serves to moderate the development of PTSS.
The findings of our study also support previous work suggesting that the use of denial is associated with decreased presence of psychiatric symptoms, in this case PTSS. There is a fairly extensive literature related to the use of denial during physical illness for example in studies of oncology patients where numerous researchers have described both adaptive and maladaptive consequences of denial.26–28
There are several limitations to our study. First, we have a small sample size and lack a control group. Second, we are relying on self-report measures to assess both coping strategies and PTSS, both of which may be concepts that are prone to distortion by subjects.29 Third, we have no long term follow-up at this time to examine the outcome of subjects who appear successfully to be managing their psychological distress by using denial as a coping strategy. We have already noted that previous studies have shown that PTSD related to medical trauma may have a delayed onset. In addition, Coursin and Coursin30 have drawn attention to the fact that PTSD symptoms, particularly the avoidance and hyperarousal criteria, may occur in subjects who do not have conscious recollection of the traumatic events. Thus, focusing on intrusion symptoms may result in missing a number of subjects with the "unconscious manifestations of PTSD that may have the greatest effect of the subsequent health and quality of life of the patient."30
Although the measures used in our study did not provide us with the ability to accurately diagnose PTSD in our subjects, we believe there are several potentially interesting issues to consider when assessing medically related PTSD. First, patients with PTSD may show high levels of depression, suicidality, and substance abuse.31,32 It is not known whether medical related PTSD is also associated with psychiatric comorbidity, but this should be an important area of further research. Second, patients with PTSD may be at particular risk of increased levels of anxiety related to subsequent hospitalizations. The finding that 35% of our sample had PTSS suggests the need to pay attention to measures used to reduce stress related to invasive medical procedures. Third, avoidance, which may be one of the "unconscious manifestations" of PTSD, may have other clinical implications (e.g., in contributing to patterns of nonadherence and avoidance of medical treatment).33
|
ACKNOWLEDGMENTS
|
This study was supported by a grant to Dr. Steiner from the C.B. Wilson Foundation.
|
REFERENCES
|
-
Green BL, Epstein SA, Krupnick JL, et al: Trauma and medical illness: Assessing trauma-related disorders in the medical settings, in Assessing Psychological Trauma and PTSD, edited by Wilson JP, Keane TM. London, The Guilford Press, 1997, pp 160–191
-
Roca RP, Spence RJ, Munster AM: Posttraumatic adaptation and distress among adult burn survivors. Am J Psychiatry 1992; 149:1234–1238
-
Perry S, Difede J, Musngi G, et al: Predictors of posttraumatic stress disorder after burn injury. Am J Psychiatry 1992; 149:931–935[Abstract/Free Full Text]
-
Cella DF, Mahon SM, Donovan M: Cancer recurrence as a traumatic event. Behav Med 1990; 16:15–22[Medline]
-
Alter CL, Pelcovitz D, Axelrod A, et al: The identification of PTSD in cancer survivors. Psychosomatics 1996, 37:137–143
-
Erikson S, Steiner H: Somatization as an indicator of trauma adaptation in long-term pediatric cancer survivors. Clin Child Psychol Psychiatry 1990; 4:415–426
-
Stuber ML, Christakis D, Houskamp B, et al: Posttraumatic symptoms in childhood leukemia survivors and their parents. Psychosomatics 1996; 37:254–261[Abstract/Free Full Text]
-
Schelling G, Stoll C, Haller M, et al: Health-related quality of life and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med 1998; 26:651–659[CrossRef][Medline]
-
Kessler RC, Sonnega A, Bromet E. et al: Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1995; 52:1048–1060
-
Lazarus RS: Psychological Stress and the Coping Process. New York, McGraw-Hill, 1966
-
Solomon Z, Mikulincer M, Avitzur E: Coping, locus of control, social support, and combat-related posttraumatic stress disorder: a prospective study. J Pers Soc Psychol 1988; 55:279–285[CrossRef][Medline]
-
van der Kolk B, van der Hart O, Marmar CR: Dissociation and information processing in posttraumatic stress disorder, in Traumatic Stress. The Overwhelming Experience on Mind, Body, and Society, edited by van der Kolk BA, McFarlane AC, Weisaeth L. London, The Guilford Press, 1996, pp 303–327
-
Fleischman S: Tense and Narrativity: From Medieval Performance to Modern Fiction. Austin, TX, University of Texas Press, 1990
-
Labov W (eds): The transformation of experience in narrative syntax, in Language in the Inner City: Studies in Black English Vernacular. Philadelphia, PA, University of Pennsylvania Press, 1979, pp 354–396
-
Clark HH: Using Language. Cambridge, UK, Cambridge University Press, 1996
-
Labov W, Waletzky J: Narrative analysis: oral versions of personal experience, in Essays in the Verbal and Visual Arts: Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society, edited by Helm J. Seattle, WA, University of Washington Press, 1967, pp 12–44
-
Zubin DA, Hewitt LE: The deictic center: A theory of deixis in narrative, in Deixis in Narrative: A Cognitive Science Perspective, edited by Duchan JF, Bruder GA, and Hewitt LE. Hillsdale, NJ: Lawrence Erlbaum, 1995, pp 129–155
-
Mergenthaler E, Bucci W: Linking verbal and nonverbal representations: Computer analysis of referential activity. Br J Med Psychol 1999; 72:339–354
-
Horowitz MJ, Wilner N, Alvarez W: Impact of Event Scale: A measure of subjective stress. Psychosom Med 1979, 41:209–218
-
Weiss DS, Marmar CR: The Impact of Event Scale—Revised, in Assessing Psychological Trauma and PTSD, edited by Wilson JP, Keane TM. New York, Guilford Press, 1997, pp 399–411
-
Horowitz MJ, Field NP, Classen C: Stress response syndromes and their treatment, in Handbook of Stress, 2nd edition, edited by Goldberger L, Breznitz S. New York, The Free Press, 1993, pp 757–773
-
Weinberger DA, Schwartz GE: Distress and restraint as superordinate dimensions of self-reported adjustment: a typological perspective. J Pers 1990; 58: 381–416
-
Weinberger DA: Distress and restraint as measures of adjustment across the life span: confirmatory factor analyses in clinical and nonclinical samples. Psychol Assess 1997; 9:132–135[CrossRef]
-
American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition. Washington, DC, American Psychiatric Association 1994, p 755
-
Yehuda R, McFarlane A: Conflict between current knowledge about posttraumatic stress disorder and its original conceptual basis. Am J Psychiatry 1995; 152:1705–1713
-
Connor SR: Denial in terminal illness: To intervene or not to intervene. Hospice Journal 1992; 8:1–15
-
Goldbeck R: Denial in physical illness. J Psychosom Res 1997; 43:575–593[CrossRef][Medline]
-
Muskin PR, Feldhammer T, Gelfand JL, et al: Maladaptive denial of illness: A useful new "diagnosis." Int J Psychiatry Med 1998; 28:463–477[Medline]
-
Epstein RS: Avoidant symptoms cloaking the diagnosis of PTSD in patients with severe accidental injury. J Trauma Stress 1993; 6:451–458[CrossRef]
-
Coursin DR, Coursin DB: Survivors, beware of posttraumatic stress disorder: what shall we tell the men in black? Critical Care Med 1998; 26:634–635
-
Breslau N, Davis GC, Andreski P, et al: Risk factors for PTSD related traumatic events: a prospective analysis. Am J Psychiatry 1991; 152:529–535[Abstract/Free Full Text]
-
Davidson JRT, Hughes D, Blazer DG, et al: Posttraumatic stress disorder in the community: an epidemiological study. Psychol Med 1991; 21:713–721[Medline]
-
Shemesh E, Lurie S, Stuber M, et al: A pilot study of posttraumatic stress and nonadherence in pediatric liver transplant recipients. Pediatrics 2000; 105:E29
-
Nelson KL: Narrative structure in recounted sad memories. Dissertation Abstracts International 1998; 59:4477B
This article has been cited by other articles:
|
|
|
|
 
D. S. Davydow, S. V. Desai, D. M. Needham, and O. J. Bienvenu
Psychiatric Morbidity in Survivors of the Acute Respiratory Distress Syndrome: A Systematic Review
Psychosom Med,
May 1, 2008;
70(4):
512 - 519.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. J. Shaw, T. E. Robinson, and H. Steiner
Acute Stress Disorder Following Ventilation
Psychosomatics,
February 1, 2002;
43(1):
74 - 76.
[Full Text]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 2001
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
