شیوه های مقابله ای در رابطه با اختلالات شخصیتی

Abstract Dysfunctional coping strategies can be expected to be among the criteria distinguishing personality disorders from normal personality functioning. In the present study the relationship between the basic coping modes problem-solving, social support seeking and avoidance was investigated in a sample of 137 psychiatric in-patients, using both a dimensional and a categorical approach. The general pattern of association found was that of a lack of social support seeking, together with an excess of avoidant coping.

1. Introduction The study of coping has its roots in the recognition that there are individual differences in reactions to and outcomes of stress. Some people become distressed or perform poorly, whereas others remain resilient. Coping theorists assume that stress outcomes result from people's coping responses, i.e. the cognitive and behavioral efforts to manage specific external and/or internal demands that are appraised as taxing or exceeding one's resources (Lazarus & Folkman, 1984). The purpose of coping research is twofold: (a) to understand why people differ so greatly in how they cope with stress and (b) to understand how differing responses relate to well-being (Aldwin, 1994). Trying to account for interindividual differences in responses to stress, researchers generally agree that coping is influenced by characteristics of both the person and the situation, but they diverge greatly with respect to the emphasis they give to personal, relative to situational factors, some of them being more person- or trait-oriented, others more situation- or state-oriented (Compas, Worsham, & Ey, 1992). As to the trait-oriented viewpoint, there is a substantial literature on the relationship between coping and personality (for a review, see e.g. O'Brien & DeLongis, 1996; Suls, David, & Harvey, 1996). A growing body of research demonstrates that there is some consistency in coping responses and that coping styles are related to the five basic factors of personality (Costa, Somerfield, & McCrae, 1996; Hewitt & Flett, 1996). Moreover, it is widely recognized that all aspects of coping (initial appraisal of the stressor; emotional response; ability to monitor, identify and regulate stress) are influenced by personal characteristics the individual brings to the situation (Summerfield & Endler, 1996). Another important issue in coping research is the relationship between coping and psychopathology. How a person adjusts to life stress is a major component of his ability to regulate well-being and to maintain mental health. The link between coping and DSM Axis I disorders (Diagnostic and Statistical Manual of Mental Disorders; American Psychiatric Association, 1987 and American Psychiatric Association, 1995) has been studied extensively. Problem-focused coping shows negative associations with a lot of psychopathological symptoms, whereas emotion-focused coping has been found to be positively related to anxious and depressive disorders (Aldwin & Revenson, 1987; Bolger, 1990). Dysfunctional coping strategies can also be expected to be among the criteria distinguishing personality disorders from normal personality functioning (Millon & Davis, 1996). However, empirical research on the relation between coping and DSM Axis II disorders is very scarce. The results of the few available studies indicate that the general pattern of coping associated with personality disorders is that of deficits in active, problem-focused coping and social support seeking, together with an excess of behavioral passivity, mental disengagement and uncontrolled discharge of emotions (Kruedelbach, McCormick, Schulz, & Grueneich, 1993; Vitaliano et al., 1990; Vollrath, Alnaes, & Torgersen, 1994). The assessment of personality disorders is not an easy task. Widely used self-report instruments, such as the Personality Diagnostic Questionnaire (PDQ-R; Hyler & Rieder, 1987), show a tendency towards overdiagnosis of personality disorders, because their items only tap the typicality of dysfunctional personality traits for the person under investigation. According to DSM-III, however, “... it is only when personality traits are inflexible and maladaptive and cause either significant impairment in social or occupational functioning or subjective distress [italics added] that they constitute personality disorders” ( American Psychiatric Association, 1980, p. 305). The ADP-IV, a recently developed and very promising instrument (De Doncker, Schotte, Vertommen, & Vankerckhoven, 1997; Schotte, De Doncker, Vankerckhoven, Vertommen, & Cosyns, in press) combines ratings of the typicality of personality traits with ratings of the amount of distress or impairment these traits cause. It was because of this combination that we decided to use the instrument in our study. For the development of instruments to measure coping, traditionally, two methodologies have been used, the one based on deduction (e.g. Beckham & Adams, 1984), the other based on induction (e.g. Salisbury, 1985). Amirkhan (1990)has combined the best of both methodologies: in the inductive tradition, he allowed naturally occurring clusters of responses to emerge from the data and, in line with deductive priorities, he isolated only those clusters common to a wide spectrum of people and events. Over the course of studies, a short self-report questionnaire (Coping Strategy Indicator, CSI) evolved that indicates the extent to which each of three coping modes (Problem-Solving, Social Support Seeking and Avoidance) has been employed in response to a recent stressful event. This trinity, although not exhaustive of coping possibilities, is appealing because it seems to correspond to the most basic human reactions to threat (Amirkhan, 1990). Problem-Solving, a strategy of direct assault, could be considered as derivative of primitive “fight” inclinations, while Avoidance, consisting of escape responses, seems derivative of ancient “flight” tendencies. Finally, Social Support Seeking taps the basic need for human contact in times of duress. The aim of the study was to explore the relationship between personality disorders and the three basic coping modes of Problem-Solving, Avoidance and Social Support Seeking in a sample of psychiatric in-patients. The following questions guided our analyses: Do psychiatric in-patients differ from the normative community sample with respect to the coping modes they engage in? Which coping modes are particularly associated with dysfunctional personality traits (dimensional approach)? Do patients who are diagnosed with a specific personality disorder (categorical approach) differ from patients who do not meet the DSM-IV criteria with respect to the coping modes they use?

3. Results 3.1. Mean ADP-IV scores Mean ADP-IV trait scores for all personality disorder subscales are displayed in Table 1. We have compared these scores to the scores of a Flemish community sample, gathered in a pilot-study on 723 subjects. Compared to this community sample, the patient sample has higher scores on all personality disorder scales of the ADP-IV. Table 1. ADP-IV means and standard deviations of the patient sample compared to those of a normative community sample by means of z-tests Scale Range Patient (n=137) Nor.(n=723) z-value p diff. PAR 7–49 23.46 (8.52) 17.06 (6.51) 11.51 0.000 SZD 7–49 22.40 (8.00) 16.24 (6.07) 11.88 0.000 STP 9–63 28.80 (9.54) 19.61 (7.85) 13.71 0.000 ASP 8–56 17.10 (7.64) 12.58 (5.06) 10.46 0.000 BDL 10–70 40.45 (12.62) 24.30 (8.65) 21.85 0.000 HIS 8–56 25.02 (8.19) 18.43 (7.04) 10.96 0.000 NAR 9–63 23.73 (8.56) 19.08 (6.98) 7.80 0.000 AVO 7–49 27.55 (10.52) 18.24 (8.08) 13.49 0.000 DPD 8–56 29.17 (9.41) 18.24 (7.21) 17.74 0.000 COM 8–56 31.30 (8.94) 23.68 (7.57) 11.78 0.000 DEP 7–49 28.88 (10.06) 17.00 (7.64) 18.20 0.000 PAS 7–49 20.86 (7.31) 14.76 (5.54) 12.89 0.000 PAR means paranoid, SZD schizoid, STP schizotypal, ASP antisocial, BDL borderline, HIS histrionic, NAR narcissistic, AVO avoidant, DPD dependent, COM compulsive, DEP depressive and PAS passive–aggressive. Table options We found gender differences in the patients' scores of four ADP-IV subscales. Men are scoring higher on the antisocial personality disorder scale than women do (t=2.77, p=0.0066), while women show a higher score on the depressive personality disorder scale (t=−3.11, p=0.002), the borderline personality disorder scale (t=−3.61, p=0.0004) and the dependent personality disorder scale (t=−2.30, p=0.0228). 3.2. Mean CSI scores Mean CSI subscale scores are given in Table 2. We have compared these scores to the scores of a Flemish community sample, gathered in a study on 298 subjects (Bijttebier & Vertommen, 1997). Compared to this sample, the patient sample reports an equal amount of social support seeking, but a smaller amount of problem-solving and a larger amount of avoidance. Table 2. CSI means and standard deviations of the patient sample compared to those of a normative community sample by means of z-tests Scale Range Patient (n=137) Nor. (n=289) z-value p diff SS 11–33 22.97 (5.90) 23.72 (5.60) −1.57 ns PS 11–33 23.59 (5.65) 25.40 (5.04) −4.20 0.000 AV 11–33 21.73 (3.96) 17.67 (3.96) 12.00 0.000 SS means Social Support Seeking, PS Problem-Solving and AV Avoidance. Table options We found no gender differences in the amount of social support seeking nor in the amount of problem-solving reported by the patients. With respect to avoidance, women report a higher use of this coping mode then men do (t=−2.36, p=0.0200). 3.3. ADP-IV personality disorder diagnoses The ADP-IV scores were used to obtain DSM-IV personality disorder diagnoses. To accomplish this, all ratings were dichotomized according to the following algorithm: a criterion is not fulfilled, unless the trait score is higher than 4 and the distress/impairment score is higher than 1. These data transformations make it possible to obtain personality disorder diagnoses according to the DSM-IV criteria. A subject is diagnosed with e.g. a paranoid personality disorder in case at least four of the seven criteria for this disorder are fulfilled. For each personality disorder diagnosis the proportion of subjects in the sample diagnosed with this personality disorder was calculated: Compulsive PD 43.8%, Borderline PD 42.3%, Avoidant PD 39.4%, Dependent PD 24.8%, Paranoid PD 24.8%, Histrionic PD, 22.6% Schizoid PD 13.1%, Schizotypal PD 12.4%, Narcissistic PD 5.1% and Antisocial PD 2.9%. Using these criteria, 15.3% of the sample received no personality disorder diagnosis. As can be derived from these percentages, there remains a considerable overlap between the personality disorder diagnoses, even when using the combined trait-distress score of the ADP-IV. 3.4. Dimensional approach: correlations between CSI and ADP-IV dimensional scores Table 3 displays Pearson correlations between the coping modes and the “personality trait” scores for all subscales, clusters and the total score. Table 3. Correlations between ADP-IV and CSI Cluster Scale SS PS AV A −0.37∗∗∗ 0.06 0.33∗∗∗ PAR −0.34∗∗∗∗ −0.03 0.27∗∗ SZD −0.29∗∗∗ 0.06 0.24∗∗ STP −0.36∗∗∗∗ −0.08 0.24∗∗ B −0.17 −0.00 0.22∗ ASP −0.27∗∗ −0.09 0.08 BDL −0.18∗ −0.09 0.30∗∗∗ HIS 0.00 −0.08 0.16 NAR −0.08 0.22∗ 0.12 C −0.17 0.02 0.25∗∗ AVO −0.29∗∗∗ 0.01 0.29∗∗∗ DPD −0.07 −0.15 0.23∗∗ COM −0.03 0.15 0.07 DEP −0.24∗∗ −0.06 0.23∗∗ PAS −0.29∗∗∗ −0.04 0.20∗ Total Score −0.21∗ 0.06 0.33∗∗∗ ∗∗∗∗p<0.0001, ∗∗∗p<0.001, ∗∗p<0.01, ∗p<0.05. PAR means paranoid, SZD schizoid, STP schizotypal, ASP antisocial, BDL borderline, HIS histrionic, NAR narcissistic, AVO avoidant, DPD dependent, COM compulsive, DEP depressive and PAS passive–aggressive. Table options As can seen from the table, all significant correlations between Social Support Seeking and the dimensional personality disorder scores are negative. Especially the cluster A personality disorder scales and the avoidant and the passive–aggressive personality disorder scale show a negative association with Social Support Seeking. This is — to a lesser extent — also the case for the antisocial, the depressive and the borderline personality disorder scale. With respect to Avoidance, all significant correlations with the personality disorder scales are positive. This is especially the case for the borderline and the avoidant personality disorder scales. Lower, but still significant correlations are found between Avoidance and the cluster A scales, the dependent and the depressive personality disorder scales. There is only one significant correlation with Problem-Solving, revealing a positive association with the narcissistic personality disorder subscale. The most frequently occurring pattern of correlations is that of a positive association between personality disorder scales and Avoidance, and a negative association between these scales and Social Support Seeking. This correlation pattern was found for all cluster A scales, the avoidant and the depressive personality disorder subscale. Beyond that, we have on one hand the antisocial and the passive–aggressive personality disorder scales, showing the negative association with Social Support Seeking but not the positive association with Avoidance and on the other hand the dependent personality disorder scale, showing the positive association with Avoidance but only a very marginally significant negative association with Social Support Seeking. Furthermore, we have the narcissistic personality disorder scale, which is the only scale associated with Problem-Solving. Finally, neither the compulsive personality disorder scale nor the histrionic personality disorder scale showed an association with any of the coping scales. With respect to the cluster scores, the cluster A score shows the clearest association with coping: we found a positive correlation with Avoidance and a negative correlation with Social Support Seeking. Both cluster B and cluster C scores also seem to be associated in a positive way with Avoidance, but these correlations are lower than the correlation with the cluster A score. As expected from these data, the total personality disorder score is associated in a positive way with Avoidance and in a negative way with Social Support Seeking. 3.5. Categorical approach: DSM-IV personality disorder diagnoses in relation to coping modes We explored differences in coping strategies reported by subjects diagnosed with a specific personality disorder compared to subjects without personality disorder diagnosis by means of multivariate analyses of variance (MANOVAs). Wilk's λ criterion was significant for the paranoid personality disorder [F(3, 51)=3.51, p=0.0184], the schizoid personality disorder [F(3, 48)=3.61, p=0.0197], the schizotypal personality disorder [F(3, 41)=7.63, p=0.0004], the antisocial personality disorder [F(3, 22)=3.24, p=0.0417], the borderline personality disorder [F(3, 80)=3.26, p=0.0258] and the avoidant personality disorder [F(3, 77)=4.24, p=0.0079]. Follow-up univariate ANOVAs were conducted, revealing the following significant main effects: paranoid PD on Social Support Seeking [F(1, 53)=7.87, p=0.0070] and on Avoidance [F(1, 53)=4.53, p=0.0184]; schizoid PD on Social Support Seeking [F(1, 50)=6.82, p=0.0119] and on Avoidance [F(1, 50)=4.32, p=0.0429], Schizotypal PD on Social Support Seeking [F(1, 43)=11.51, p=0.0015] and on Avoidance [F(1, 43)=7.95, p=0.0072]; Antisocial PD on Social Support Seeking [F(1, 24)=7.81, p=0.0101]; Borderline PD on Social Support Seeking [F(1, 82)=4.43, p=0.0384] and on Avoidance [F(1, 82)=5.87, p=0.0176]; Avoidant PD on Social Support Seeking [F(1, 79)=8.53, p=0.0046] and on Avoidance [F(1, 79)=4.40, p=0.0392]. Post-hoct-tests of the significant effects suggest a smaller amount of Social Support Seeking and a larger amount of Avoidance in patients diagnosed with a Paranoid PD, a Schizoid PD, a Schizotypal PD, a Borderline PD or an Avoidant PD. There is also evidence for a smaller amount of Social Support Seeking in patients diagnosed with an Antisocial PD. Adjusted means for all personality disorders with significant effects are presented in Table 4. Table 4. Adjusted means, t-values and significance levels of the post-hoc tests concerning differences in coping strategies between patients diagnosed with a specific personality disorder and patients without personality disorder diagnosis CSI subscale With PD (Madj) Without PD (Madj) t-value p Paranoid PD Social Support Seeking 20.40 24.90 2.80 0.0070 Avoidance 22.74 20.30 −2.13 0.0320 Schizoid PD Social Support Seeking 20.47 24.90 2.61 0.0119 Avoidance 22.66 20.30 −2.08 0.0429 Schizotypal PD Social Support Seeking 19.24 24.90 3.39 0.0015 Avoidance 23.56 20.30 −2.82 0.0072 Antisocial PD Social Support Seeking 17.83 24.90 2.79 0.0101 Borderline PD Social Support Seeking 21.86 24.90 2.10 0.0384 Avoidance 22.77 20.30 −2.42 0.0176 Avoidant PD Social Support Seeking 20.54 24.90 2.92 0.0046 Avoidance 22.51 20.30 −2.10 0.0392