بازنگری مرزهای مبهی از اسکیزوفرنی: اختلالات طیف در اسکیزوتایپی روان شناسایی شده
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|30205||2015||6 صفحه PDF||سفارش دهید||5084 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Psychiatry Research, Volume 225, Issue 3, 28 February 2015, Pages 335–340
Certain Personality Disorders (PDs) have been found to be present in the prodromal phase of schizophrenia at a higher rate than other personality disorders. Although schizotypal, paranoid, and schizoid PDs are traditionally viewed as spectra for schizophrenia, research suggests that avoidant PD should be included in this group (e.g., Fogelson et al., 2007). The present study examines whether a sample of psychometrically identified schizotypes (SZT) have higher incidence of schizophrenia-spectrum PDs, as well as more symptoms of these PDs, in general, than does a matched comparison (MC) sample. Eighty-five SZT and 78 MC participants were administered the Personality DisorderInterview for DSM-IV (PDI-IV) to assess PD symptoms and diagnoses. Results indicate that the SZT group evidenced significantly more symptoms of avoidant, schizoid, paranoid, and schizotypal PDs than did the MC group. Further, there were significant differences in the incidence of these PDs between the groups.
A long history of research into premorbid personality indicators has supported the notion that there are differences in individuals with increased liability to schizophrenia related illnesses (SRIs) compared to those without increased liability. For example, Bleuler, (1911/1950) noted that individuals who develop SRIs demonstrated oddities in personality from childhood and were likely to be withdrawn from others. Hoch (1910), noting a relationship between a detached personality type and schizophrenia development, referred to a “shut-in” personality. Likewise, Niemi et al. (2005) found that the presence of emotional problems and social inhibition in children predicted later psychotic symptoms. As far back as the time of Kraepelin (1909/1971), it has been noted that symptoms of SRIs appear to aggregate within families, as relatives of individuals with the disorder exhibit a number of anomalies, including eccentric personality. Further, family studies of schizophrenia have indicated a relationship between schizophrenia and personality disorders such as schizotypal personality disorder (Kendler et al., 1993 and Asarnow et al., 2001;Hans et al., 2004). Certain personality disorders have also been found to be present in the prodromal phase of schizophrenia. Indeed, the “Cluster A” disorders (Schizotypal, Paranoid, and Schizoid) are viewed as being related to schizophrenia (Braff et al., 2007). For example, data from the New York High-Risk Project demonstrated that as many as 16–20% of schizophrenia offspring may develop “Cluster A” personality disorders (Erlenmeyer-Kimling et al., 1995). Research, however, has suggested that avoidant personality disorder be included in this group of schizophrenia-related personality disorders. For example, Solano and De Chávez (2000) found that 85% of their sample of patients with schizophrenia had premorbid personality disorders, with avoidant (32.5%), schizoid (27.5%), paranoid (20%), dependent (20%), and schizotypal (12.5%) were the most common; they noted, however, that the generalizability of their findings may be limited by their relatively small (N=40) sample. Likewise, Keshavan et al., 2005 found that “Cluster C” dimensional scores on a semi-structured personality interview schedule, particularly avoidant personality scores, were higher for patients with schizophrenia than for patients with non-schizophrenia psychoses or healthy participants. Such findings have been extended to individuals deemed to be at risk for schizophrenia, as Fogelson et al. (2007) have demonstrated that a relations exists between avoidant personality disorder and liability to schizophrenia even after statistically accounting for paranoid and schizotypal personality disorders, a finding that was supported by Gooding et al. (2007). Bolinskey and Gottesman (2010) found higher rates of reported avoidant personality disorder symptoms among individuals classified as hypothetically psychosis prone compared to a matched control sample, although their study relied on self-report of symptoms. Fogelson et al. (2010) have extended these findings into the neurocognitive realm by demonstrating that avoidant personality disorder symptoms can predict performance on neurocognitive measures associated with schizophrenia liability even after accounting for symptoms of other spectrum disorders. This link between avoidant personality and the schizophrenia spectrum is not surprising given the similar patterns of social withdrawal witnessed among the disorders; indeed ( Millon, 1990 and Millon et al., 2004) have conceptualized schizoid, schizotypal, and avoidant personalities as falling in the detached interpersonal spectrum, with schizoid personality reflecting an entirely passive adaptation style, avoidant personality reflecting an active adaptation style, and schizotypal reflecting a mixed adaptation style.
نتیجه گیری انگلیسی
Correlations among the symptom levels for each personality disorder are shown in Table 1. Symptom levels were correlated, but not so highly correlated as to indicate redundancy among the variables. The results of the multivariate test revealed a significant difference between groups, F (4, 158)=6.638, p<0.001. Wilks׳ lambda for the analysis was.856, which suggests that approximately 14% of the variance in the linear combination of symptom levels can be accounted for by group membership. Table 1. Correlations among number of symptoms of schizophrenia-related personality disorders. Disorder Avoidant Paranoid Schizoid Schizotypal Avoidant – Paranoid 0.392⁎⁎ – Schizoid 0.243⁎ 0.423⁎⁎ – Schizotypal 0.350⁎⁎ 0.582⁎⁎ 0.466⁎⁎ – Note: ⁎⁎ p<0.001 ⁎ p<0.01 Table options Since the multivariate test was significant, univariate tests were performed for mean symptom level by group. Mean symptom levels for each group are displayed in Table 2, along with appropriate effect sizes. The results of each comparison were significant at p<0.01. Medium effect sizes were observed for schizoid (d=0.66) and avoidant (d=0.58) personality disorder symptoms, whereas small effects were observed for schizotypal (d=0.45) and paranoid (d=0.44) personality disorder symptoms. Table 2. Means and standard deviations for selected Personality Disorder Interview for DSM-IV scores by group membership, with associated F values and effect sizes. SZT (N =85) MC (N=78) Scale M (S.D.) M (S.D.) F d Level of Avoidant PD sx 2.89 (2.33) 1.65 (1.86) 13.92⁎⁎ 0.58 Level of Paranoid PD sx 1.53 (1.78) 0.86 (1.20) 7.77⁎ 0.44 Level of Schizoid PD sx 1.36 (1.49) 0.58 (0.78) 17.46⁎⁎ 0.66 Level of Schizotypal PD sx 2.14 (1.81) 1.40 (1.44) 8.37⁎ 0.45 Note: SZT=psychometrically-identified schizotype group; MC=matched comparison group; d=Cohen׳s d. ⁎⁎ p<0.001. ⁎ p<0.01. Table options 3.2. Meeting criteria for personality disorders We then examined whether there were differences in the number of individuals meeting diagnostic criteria, as determined by the PDI-IV, between the SZT and MC groups. Results of Barnard׳s and Fisher׳s exact tests are displayed in Table 3. Barnard׳s tests were significant for each disorder. Fisher׳s exact tests were significant for avoidant, schizotypal, and schizoid personality disorders, but fell slightly short of significance for paranoid personality disorder, as the P value fell right at 0.05. Table 3. Number of individuals meeting, or not meeting, criteria for selected personality disorders by group, along with results of Barnard׳s exact tests and odds ratio estimates. Criteria Met Diagnosis Group Yes No Barnard׳s test Fisher׳s test Odds ratio (95% CI) Avoidant SZT 36 49 p=0.001 p=0.002 3.09 (1.52–6.27) MC 15 63 Paranoid SZT 11 74 p=0.049 p=0.050 3.72 (1.00–13.87) MC 3 76 Schizoid SZT 8 77 p=0.006 p=0.007 ⁎ MC 0 78 Schizotypal SZT 11 74 p=0.019 p=0.019 5.65 (0.52–6.27) MC 2 76 Note: SZT=psychometrically-identified schizotypes. MC=matched control. ⁎ An odds ratio could not be calculated for schizoid personality disorder, as none of the MC participants met diagnostic criteria. Table options The odds ratios indicate that individuals in the SZT group were 3.09 times more likely to meet criteria for avoidant personality disorder than were the individuals in the MC group (95% CI=0.52–6.27) and 5.65 times more likely to meet criteria for schizotypal personality disorder (95% CI=0.52–6.27). SZT participants were 3.72 times more likely to meet criteria for paranoid personality disorder than were the individuals in the MC group, although the confidence interval includes 1 (95% CI=1.00–13.87). Odds ratios could not be calculated for schizoid personality disorder, as no members of the MC group met diagnostic criteria. 4. Discussion Consistent with the results of previous research, the present study provides evidence that symptoms of paranoid, schizoid, schizotypal, and avoidant personality disorders are more prominent among individuals determined to be at increased risk for developing schizophrenia through psychometric means than among individuals in a matched comparison group. This finding is important as it demonstrates that these symptoms are present in individuals with subthreshold expression of symptoms of schizotypy, but who are otherwise relatively high functioning. These symptom differences exist even in the absence of identifiable personality disorders, although there was also a difference in the number of individuals meeting diagnostic criteria for spectrum personality disorders between groups. Our first hypothesis was that members of the SZT group would evidence a greater number of symptoms of each of the spectrum disorders than would members of the MC group, without regard to the presence of a diagnosable disorder. Since our participants are drawn from a relatively high-functioning (i.e., college student) sample, we expected that they would not necessarily present with diagnosable disorders, but would evidence attenuated forms of the disorders. This hypothesis was fully supported for each of the disorders (avoidant, schizotypal, and schizoid personality disorders), with medium effect sizes observed for each between-groups comparison. These findings join a growing body of the literature that supports a relationship between schizophrenia spectrum personality characteristics and increased liability to schizophrenia (Erlenmeyer-Kimling et al., 1995, Solano and De Chávez, 2000 and Braff et al., 2007). Our second hypothesis was that SZT participant would have a higher incidence of meeting diagnostic criteria for spectrum personality disorders than would the MC participants. This hypothesis was considered secondary because we did not expect that many of the individuals in the SZT group would meet full criteria for a spectrum disorder at this early stage of the study; we expect, however, that the number of individuals meeting criteria will increase over the course of the study as our SZT group moves through the period of greatest risk for developing schizophrenia related illnesses. Our second hypothesis was fully supported for three of the disorders (avoidant, schizotypal, and schizoid personality disorders), whereas the significance of the results for paranoid personality disorder depends on which test is considered. The results for Barnard׳s exact test were significant, although the results for Fisher׳s exact test fell just short of significance at p¬ =0.05. As noted above, Fisher׳s test has been criticized for being overly conservative with small samples; it is likely that with a larger sample, this comparison would have been significant. The results clearly indicate, however, a trend toward greater incidence of paranoid personality disorder within the SZT group. Taken together, these results indicate that there are sub-threshold expressions of symptoms of schizophrenia spectrum personality disorders among those identified through psychometric means to be at increased risk for developing schizophrenia. An additional, and not unimportant, implication is the additional support that these results provide for the psychometric assessment of schizotypy by means of self-report. This is especially noteworthy, as the majority of previous research has focused on individuals deemed “at risk” for schizophrenia via different means, such as genetic predisposition or family history (Kendler et al., 1993 and Asarnow et al., 2001; Hans et al., 2004). Finally, the current results add to the literature (e.g. Solano and De Chávez, 2000, Keshavan et al., 2005, Gooding et al., 2007, Fogelson et al., 2007, Fogelson et al., 2010 and Bolinskey and Gottesman, 2010) demonstrating a relationship between avoidant personality disorder and liability to schizophrenia. Social withdrawal and detachment have long been noted to be a hallmark feature of liability to schizophrenia (cf. Hoch, 1910, Bleuler, 1911/1950, Meehl, 1962 and Meehl, 1990) and both the diagnostic and phenomenological overlap between avoidant and schizoid personality disorders have been represented in commentary and research (e.g., Trull et al., 1987 and Thompson-Pope and Turkat, 1993), as well as models of personality disorders (Millon, 1990 and Millon et al., 2004). Fogelson et al. (2007) suggested that with avoidant personality disorder – specifically, its associated social withdrawal and interpersonal sensitivity – represents a separable indicator of schizophrenia liability, but one consistent with Gottesman and Shields (1982) epigenetic liability model. We concur with that assessment, but suggest that avoidant personality disorder may reflect a somewhat attenuated form of schizophrenia liability than schizoid personality in which this social detachment and sensitivity is not yet associated with ambivalence. Thus, from this perspective, the ambivalence associated with schizoid personality represents increased liability to schizophrenia. A major strength of this study is our use of psychometrically identified schizotypy as a dependent variable; further, we incorporated a matched comparison sample for our criterion group, rather than using convenience samples or relying on sometimes unreliable family history information to determine risk status. Our design allows for a direct analysis of the relationship between spectrum personality disorders and schizotypy without the influence of selection bias or demographic influence. Further, our study incorporated measures of personality disorder symptoms and personality disorder diagnoses that were obtained through semi-structured interviews administered by trained clinicians and subjected to review by at least one other trained administrator, rather than relying on self-report, as in the Bolinskey and Gottesman (2010) study, or chart diagnoses that are often notoriously unreliable (cf., Garb, 1998 and Jensen and Weisz, 2002). A weakness of the present study is the relatively small sample size resulting in a small number of individuals exhibiting fully developed personality disorders. This is particularly notable when considering low base rate characteristics like paranoid or schizoid personality disorders. As noted above, however, it was not expected that we would find a large number of individuals with fully developed personality disorders at baseline, given both the age of the individuals (late teens) and their relatively high functioning at the present time. In addition, it should be considered that the present sample overrepresented females and therefore results should be interpreted with caution as it is unclear to what degree the present results might be influenced by the gender distribution in our sample. We have recently completed collecting baseline data for a 10-year prospective study; we also gathered baseline data on processes such as eye-tracking, working memory, executive functioning, and additional measures of personality. Since the project from which the current study comes is a prospective study we will be able to follow these individuals through most of their highest risk period (approximately 18–35 years old) for developing further symptoms of spectrum disorders or SRP. We expect that, over time, we will see an even greater disparity in the number of personality disorder symptoms and the number of individuals meeting criteria for personality disorder between the two groups.