ثبات و تغییر در رتبه بندی ابعادی اختلالات شخصیتی در بیماران اعتیاد به مواد مخدر در طول درمان

Abstract The objective of the study was to determine the extent of change in dimensional scores of personality disorders during treatment of drug abuse patients. The drug abuse patients were monitored prospectively during treatment. Over a period of 6 years, at 3-monthly intervals, all residents in a therapeutic community for drug abuse patients were administered the Millon Clinical Multiaxial Inventory-II (MCMI-II), a questionnaire developed to measure Diagnostic and Statistical Manual of Mental Disorders (3rd ed., revised) personality disorders. Subjects who completed the MCMI-II at least at four different times were selected for this study (n = 72). Results of the study showed that treatment had resulted in significant changes in the dimensional scores of some personality disorders, whereas other dimensional scores did not change at all. Implications for treatment are discussed.

Introduction According to the official nomenclature in the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV) of the American Psychiatric Association (1994), a personality disorder (PD) is 'an enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual culture, is pervasive and inflexible, has an onset in adolescence or early adulthood, is stable over time, and leads to distress or impairment'. This definition stresses that the diagnosis of PD is dependent upon a pattern of clinical features manifested consistently over a prolonged period of time and that PDs remain present once a PD is diagnosed. Over the last decade there has been a great deal of interest in the stability of PDs once they have been assessed Ravndal & Vaglum, 1991a, McDavid & Pilkonis, 1996, McMahon & Richards, 1996 and Grilo & McGlashan, 1999. A number of longitudinal studies evaluating the treatment outcome of normals with PD features (Lenzenweger, 1999) and of psychiatric patients with PDs Perry, 1993, Stone, 1993 and Bateman & Fonagy, 2000 have been published. Overall they conclude that a substantial percentage of the psychiatric patients with one or more lifetime PDs had significant symptoms and impairment in social functioning at follow-up. Psychiatric and drug abuse comorbidity is among the highest in health care; axis I disorders are often mentioned (Brooner, King, Kidorf, Schmidt, & Bigelow, 1997) as well as axis II disorders, especially anti-social and borderline PDs Hendriks, 1990, Nace et al., 1991, Verheul, 1997 and Kokkevi et al., 1998. Further, a large proportion of the patients referred for treatment of personality disorders were also diagnosed with comorbid substance abuse disorders (Skodol, Oldham, & Gallaher, 1999). In other studies the prevalence of PDs in drug abuse patients has varied considerably. In their systematic review of the literature on personality disorders in drug abuse patients, Verheul, Van den Brink, and Hartgers (1995) found that across 25 studies of drug abuse patients the median prevalence of any PD was 61%. Most of the substance abusers who seek treatment fit the criteria for more than one PD (Barber et al. 1996). The presence of PDs among drug abuse patients has been associated with longer and heavier drug use histories, with poorer functioning with respect to employment, and with family and social problems (Rutherford, Cacciola, & Alterman, 1994). Rehabilitation of these drug abuse patients is more complicated; treatment times are longer and treatment outcomes poorer Reich & Vasile, 1993, Verheul et al., 1995, Kokkevi et al., 1998 and Ravndal & Vaglum, 1998. Drug abuse patients tend to be more resistant to change and often discontinue treatment prematurely Reich & Vasile, 1993 and Ravndal & Vaglum, 1998. One of the objectives of treatment of drug abuse patients is to instill changes in thinking and behavior, most specifically to have patients change their attitude toward various aspects of life's challenges (Kooijman, 1992). Focusing the treatment of drug abuse patients on PD symptoms sometimes alleviates the specific symptoms. As a result it would be understandable that the percentage of drug abuse patients with a PD may change accordingly. There are several studies on the long-term effects of psychotherapeutic treatment on PDs. However, only few studies have measured change in PDs during treatment, and there is a strong need for studies with repeated measurements (Perry, 1993). Such studies would provide data on changes during the therapeutic process. The Millon Clinical Multiaxial Inventory (MCMI), an instrument to measure PDs, is frequently used in research on drug abuse patients Craig & Weinberg, 1992 and Flynn & McMahon, 1997. McMahon, Flynn and Davidson (1985) administered the MCMI to drug abuse patients at intake and again after 1 and 2 months of treatment. After 1 month of treatment significant changes were recorded on almost all the PD scales of the MCMI. Ravndal and Vaglum (1991a, 1991b) have presented data on substance abusers who followed a long-term program; after treatment most of the dimensional scores on the MCMI-II, the successor of the MCMI, changed significantly. The treatment periods for cocaine abusers in the study of McMahon and Richards (1996) were relatively short, but nonetheless changes on several of the PD scales of the MCMI-II were recorded. Schinka and collaborators (1999), using the MCMI-II, also found significant changes in different dimensional scores before and after treatment of female drug abuse patients. To our knowledge no studies have been published in which the number of (or the dimensional scores of) PDs has been measured more than three times during treatment. In our study we wanted to examine the change in the different PDs during treatment by measuring the severity of pathology every 3 months. Many therapeutic communities consider successful treatment for substance abuse to take at least 12 months Ravndal & Vaglum, 1991a and Kooijman, 1992, thus allowing program-completers to be screened four times. Because there are gender differences in the prevalence of PDs (Verheul, 1997), gender was included as an independent variable.

3. Results 3.1. Stability coefficients Table 1 shows the stability coefficients (correlation coefficients) for the different waves. Lower stability coefficients were observed for the longer treatment time spans. Comparison of waves 2 and 3 gives good stability coefficients. Stability coefficients for the narcissistic and the antisocial personality disorders are among the highest. Table 1. Stability coefficients (Pearson product moment correlations) assessed with the MCMI-II across four waves (n = 72) Disorder Wave 1/ Wave 2 Wave 1/ Wave 3 Wave 1/ Wave 4 Wave 2/ Wave 3 Wave 2/ Wave 4 Wave 3/ Wave 4 Schizoid .64 .48 .30 .71 .52 .66 Avoidant .58 .59 .47 .70 .53 .66 Dependent .75 .63 .54 .79 .69 .66 Histrionic .74 .72 .62 .76 .75 .78 Narcissistic .77 .71 .65 .80 .78 .81 Antisocial .79 .74 .69 .74 .71 .81 Compulsive .67 .65 .64 .71 .73 .69 Passive/aggressive .63 .66 .54 .73 .69 .64 Schizotypal .53 .52 .52 .71 .59 .67 Borderline .55 .56 .41 .61 .60 .55 Paranoid .71 .73 .67 .76 .77 .82 Table options 3.2. Change in personality disorder dimensions Analysis of variance between subjects effects indicated that females exhibited significant lower scores on the personality disorder scales histrionic (F(1, 70) = 14.3, p < .001), narcissistic (F(1, 70) = 13.4, p < .001) and paranoid (F(1, 70) = 8.1, p < .01). There was a significant main effect of assessment wave on six of the 11 MCMI-II subscales. The following scale-scores decreased during treatment: schizoid F(3, 68) = 7.64, p < .001, avoidant F(3, 68) = 17.17, p < .001, dependent F(3, 68) = 6.34, p < .001, passive-aggressive F(3, 68) = 4.73, p < .005, schizotypal F(3, 68) = 13.68, p < .001, and borderline F(3, 68) = 9.83, p < .001. No significant increase was observed on any of the MCMI-II-scales during treatment. In Table 2 the mean value of each scale on each wave is displayed. Significant post-hoc differences between the waves are indicated for each scale. No significant wave × sex interaction was observed. Table 2. Mean (raw) scores on the MCMI-II with post-hoc comparisons between baseline and follow-up for each measurement (n = 72) Disorder Wave 1 Wave 2 Wave 3 Wave 4 Differences Schizoid 22.6 20.0 18.7 16.5 4 < 1; 4 < 3; 3 < 1 Avoidant 28.2 25.5 22.6 18.7 4 <3 < 2 < 1 Dependent 25.5 24.4 23.4 22.3 4 < 1; 4 < 2; 4 < 3 Histrionic 33.9 33.1 33.6 34.2 ns Narcissistic 40.0 40.4 40.5 41.1 ns Antisocial 42.9 41.5 41.8 41.6 ns Compulsive 32.8 34.7 34.2 34.8 ns Passive/aggressive 40.8 38.4 37.0 34.1 4 < 1; 4 < 3 Schizotypal 25.8 23.6 20.8 17.9 4 < 1; 4 < 2; 3 < 1; 4 < 3 Borderline 46.3 41.4 38.7 34.4 4 < 1; 4 < 2; 4 < 3; 3 < 1 Paranoid 32.0 31.4 31.0 30.1 ns Table options Base-rate scores indicate whether the number of personality traits or PDs has changed during treatment. In Table 3 lists these figures for our study. Table 3. Number of drug abuse patients with personality traits (BR = >75, <85) and personality disorders (BR ≥ 85) according to profiles of the MCMI-II in the beginning of treatment (wave 1) and after treatment (wave 4) (n =72) Disorder Wave 1 Wave 4 ≥ 75, <85 ≥85 ç 75, <85 ≥85 Schizoid 8 (11%) 2 (3%) 1 (1%) 1 (1%) Avoidant 15 (21%) 26 (36%) 8 (11%) 9 (13%) Dependent 4 (6%) 10 (14%) 6 (8%) 2 (3%) Histrionic 18 (25%) 6 (8%) 16 (22%) 11 (15%) Narcissistic 15 (21%) 22 (31%) 18 (25%) 28 (39%) Antisocial 16 (22%) 38 (53%) 12 (17%) 43 (59%) Compulsive 7 (10%) 2 (3%) 8 (11%) 4 (6%) Passive/agressive 7 (10%) 52 (72%) 9 (13%) 32 (44%) Schizotypal 1 (1%) 5 (7%) 0 (0%) 1 (1%) Borderline 8 (11%) 25 (35%) 8 (11%) 7 (10%) Paranoid 6 (8%) 5 (7%) 8 (11%) 3 (4%) Table options 3.3. Predictors of time in program Regression analysis, with baseline MCMI-II-scale scores as independent variables and days in program as the dependent variable, showed that the schizotypal personality scale predicts time in program (β = −1.75, t =2.5, p < .05): a high score on the schizotypal scale is significantly associated with early dropout.