Premorbid Personality in Psychoses

Transcription

1 Premorbid Personality in Psychoses by Manuel J. Cuesta, Victor Peralta, and Francisco Caro Abstract This study aimed to establish correlates of the dimensions of schizophrenia in the premorbid personality traits of patients. A sample of 112 patients of relatively recent illness onset who were admitted for a psychotic episode were assessed with a semistructured interview for schizophrenia. Positive and negative symptoms were evaluated with the Scale for the Assessment of Positive Symptoms and the Scale for the Assessment of Negative Symptoms at the time of hospital discharge; positive, negative, and disorganization scores were obtained from these scales. Premorbid personality was assessed blindly through a partially modified version of the Personality Assessment Schedule using interviews with the parents or a close relative. traits were significantly associated with negative and positive dimensions. traits were related to the disorganization dimension. Trends toward significance were obtained between passive-dependent traits and the negative and disorganization dimensions, and between the schizotypal dimension and the positive dimension. Partial correlational analyses were used to control for the effect of the remaining personality dimensions on the above relationships. premorbid traits were still significantly related to the negative dimension but to a lesser degree to the disorganization and positive dimensions. The association between sociopathic premorbid traits and the disorganization dimension remained significant. These results suggest the preexistence of a three-dimensional structure predisposing to psychoses within the premorbid personality; this structure is more evident in patients with short illness duration. Key words: Schizophrenia, psychosis, premorbid personality, schizoid, schizotypy, phenomenology, personality traits, psychopathological dimensions, positive dimension, negative dimension, disorganization dimension. Schizophrenia Bulletin, 25(4): ,1999. Premorbid constitution and premorbid personality have long been considered risk factors for the development of functional psychoses. Authors have debated, however, some proposing a "continuity" between premorbid personality and schizophrenia (Bleuler 1924; Kretschmer 1934) and some holding that premorbid personality and schizophrenia are independent (Jaspers 1956; Schneider 1958). Furthermore, in examining relatives of patients with schizophrenia, Bleuler (1972) appreciated milder manifestations of schizophrenia without prominent psychotic symptoms; he called these manifestations "schizoid." Since the days of those early studies of the relationships between premorbid personality and psychoses, several lines of research have developed. Many studies have sought genetic associations between schizophrenia disorders and other mental or personality disorders. These studies gave rise to two terms: the "schizophrenic spectrum" concept, which describes group disorders that could be inherited together and are putatively related to schizophrenia, and the "schizotypy" concept, which describes a characteristic profile of personality that was often found in genetic studies with relatives of schizophrenia patients. The spectrum concept includes schizophrenia and schizoaffective disorders, as well as schizotypal, schizoid, and paranoid personality disorders (Siever et al. 1995). Another strategy used to trace premorbid personality traits of psychoses was to follow populations at high risk for schizophrenia and other psychoses. Two comprehensive studies on schizophrenia risk with a followup period of more than 20 years have been conducted: the Copenhagen Schizophrenia High-Risk Project (Parnas et al. 1982; Parnas and Jorgensen 1989; Cannon et al. 1990; Jorgensen and Parnas 1990), a prospective investigation of children of mothers with schizophrenia, and the New York High-Risk project (Erlenmeyer-Kimling et al. 1995), involving subjects at risk of schizophrenia, subjects at risk Reprint requests should be sent to Dr. M. Cuesta, Psychiatric Unit, Virgen del Camino Hospital, c/ Irunlarrea s/n, Pamplona, Spain. 801

2 Schizophrenia Bulletin, Vol. 25, No. 4, 1999 MJ. Cuesta et al. of affective psychoses, and a population without psychiatric risk followed over a period of 23 years. The latter study examined the cluster A personality disorder from DSM-HI (including paranoid, schizoid, and schizotypal subtypes), discarding subjects who developed schizophrenia or schizoaffective disorder. A significant increase in cluster A as a whole was found for subjects at risk of schizophrenia compared with those at no risk of psychiatric disorders. No differences were found between subjects at risk of schizophrenia and affective psychoses, nor were there differences in schizotypal scores between groups. Studies designed to identify premorbid features from the obstetric period through late adolescence have looked for failures or delays in achievement in neuromotor, social, academic, and functioning development. These studies overlap partially with studies of premorbid personality. Premorbid asociality in puberty, poor school performance during childhood (Mukherjee et al. 1991), and maladjustment at age 7 for male preschizophrenia patients (Done et al. 1994) have been reported as precursors of a predominance of negative symptoms when the schizophrenia subsequently emerges. Several instruments have been developed for the early detection of risk of schizophrenia, such as Chapman and Chapman's scales (Chapman and Chapman 1980; Chapman et al. 1994) and Moldin et al.'s index derived from the Minnesota Multiphasic Personality Inventory (Lenzeweger 1994). Finally, many studies have confirmed the preexistence of DSM-HI cluster A personality disorder traits in subjects with schizophrenia (Parnas and Jorgensen 1989; Dalkin et al. 1994; McCreadie et al. 1994), but few studies have focused on the associations between premorbid personality traits and the phenomenological characteristics of schizophrenia or psychotic disorders (Cannon et al. 1990; Jorgensen and Parnas 1990; Peralta et al. 1991). Personality is better understood from a dimensional perspective than from a typological one, given that pure types are rarely found. The dimensional approach has also proved to be superior to typological or categorical models in the study of schizophrenia symptomatology because it incorporates a measure of quantity that can be related to neurobiological correlates (Green and Walker 1985). The dimensional model presumes or hypothesizes that the expression of intensity is related to the etiological agent or dysfunction, even though the agent or dysfunction has not yet been found. Three dimensions have been repeatedly observed in schizophrenia patients in many studies: positive or psychotic, negative, and disorganization (Bilder et al. 1985; Liddle 1987; Ardnt et al. 1991; Peralta et al. 1992). These dimensions, however, have been recently reported to be consistent and widely distributed in all types of psychoses: schizophrenia, schizoaffective disorder, schizophreniform disorder and other nonschizophrenic psychoses (Peralta et al. 1997). This study aimed to find correlates of the clinical dimensions of schizophrenia disorders and other psychoses in the premorbid personality traits of patients. We hypothesized that certain dimensions of premorbid personality may be earlier expressions of the three-dimensional structure of schizophrenia symptomatology in psychosis. Methods The sample was composed of 112 consecutive patients admitted because of a psychotic episode of illness of relatively recent onset. Patients were assessed with a semistructured and multidiagnostic interview for schizophrenia that provides a thorough assessment of epidemiological and clinical variables (Landmark 1982). This interview has demonstrated good interrater reliability (Helmes et al. 1983; Peralta et al. 1991) and is well-suited to the study of the concordance between diagnostic criteria (Landmark et al. 1986; Peralta and Cuesta 1992). In this study, the interview was partially modified to aggregate recent diagnostic criteria with their corresponding items and to include nonschizophrenia patients, as other authors have done (McGorry et al. 1990). DSM-III-R (American Psychiatric Association 1987) diagnoses were 23 schizophreniform disorder (20.5%), 64 schizophrenia disorder (57.2%), 13 schizoaffective disorder (11.6%), and 12 affective disorder (10.7%) (table 1). Positive and negative symptoms were evaluated through the Scale for the Assessment of Positive Symptoms (SAPS; Andreasen 1984a) and the Scale for the Assessment of Negative Symptoms (SA; Andreasen 1984fo) by M.J.C. and V.P. Good interrater reliability for these scales was obtained in an earlier report in a subsample of 30 schizophrenia patients rated together. The intraclass correlation coefficient for the SAPS and SA subscales and items ranged from 0.74 for inappropriate affect to 0.94 for anhedonia-asociality (Peralta et al. 1992). Positive, negative, and disorganization dimensions were derived from the scores of the SAPS and SA scales. Briefly, delusions and hallucinations composed the positive dimension; affective flattening, alogia, abulia, and anhedonia-asociability the negative dimension; and formal thought disorder, bizarre behavior, attention, and inappropriate affect the disorganization dimension. Premorbid personality was assessed through the Personality Assessment Schedule (PAS; Tyrer and Alexander 1988), which assesses 24 personality variables throughout the subject's life. The ratings of each variable are based on a set of specific initial questions and further questions formulated by the interviewer in response to the 802

3 Premorbid Personality in Psychoses Schizophrenia Bulletin, Vol. 25, No. 4, 1999 Table 1. Sociodemographic characteristics of sample (n = 112) Characteristics Mean ± SD Range Age, yr Age at onset, yr Duration of illness Education (yr) GAFP Hospitalizations (n) Chlorpromazine doses Biperidene (mg) Global rating of reliability of information obtained by the rater Gender Male Female Marital status Single Married Divorced Widow Diagnoses (DSM-III-R) Schizophreniform Schizophrenia Schizoaffective Affective Source of personality information Both parents Mother only Father only Older brother or sister Others 1 Note. GAFP = Global Assessment Functioning (past year) ± ± ± ± ± ± ± ± ± Others = Spouses (4%) or second-degree relatives (5%), such as uncles or aunts (65.2) 39 (34.8) 97 (86.6) 8 (7.1) 6 (5.4) 1 (0.9) 23 (20.5) 64 (57.2) 13(11.6) 12(10.7) 34 (30.0) 46(41.0) 8 (7.0) 15(13.0) 9 (9.0) informant's replies. The personality variables are scored on a 9-point scale of intensity: Absent (0), mild expression without repercussion in functioning (ratings 1-3), problems in social adjustment related to the personality traits (ratings 4-6), and severe dysfunction in functioning (ratings 7-8). The advantage of the PAS over other instruments is that it is designed to formalize the assessment of personality disorder, so it can be used with subjects of all kinds, regardless of their psychiatric status (Tyrer 1988). It also incorporates a global rating to score the rater's opinion of the reliability of the information obtained (ratings from 0-8). lnterrater and temporal reliability in the assessment of premorbid personality has been reported to be satisfactory (Cicchetti and Tyrer 1988). The PAS can be used to classify patients according to ICD 10 or DSM-FV personality disorders or to assess the personality of patients dimensionally. The dimensional approach is based on Tyrer and Alexander's (1988) study of a cluster analysis of 130 patients, 65 with a clinical personality disorder and 65 with a psychiatric disorder without personality disorder. They found four personality clusters: the schizoid, sociopathic, passive-dependent, and anancastic dimensions. They then presented a formula to calculate the scores of the four clusters from the sum of the scores of the personality variables composing each cluster divided by the number of items included. The items in the schizoid dimension were introspection, shyness, aloofness, suspiciousness, and eccentricity; the items in the sociopathic dimension were irritability, impulsiveness, aggression, callousness, and irresponsibility; the items in the passive-dependent dimension were anxiety, vulnerability, childishness, resourcelessness, and dependence; and the items in the anancastic dimension were introspection, sensitivity, conscientiousness, rigidity, and hypochondria. The PAS dimensions were partially modified to avoid overlapping of items and to develop the schizotypy dimension. The only item that overlapped in two person- 803

4 Schizophrenia Bulletin, Vol. 25, No. 4, 1999 M.J. Cuesta et al. ality dimensions was introspection; it was removed from the anancastic dimension and left in the schizoid. The schizotypy dimension was extracted from two items in the schizoid dimension (suspiciousness and eccentricity) because schizotypy is not included in the four main clusters of the PAS. The rationale for differentiating between these two personality disorders is that the schizotypy dimension has been proposed as a putative schizophrenia spectrum disorder (Kendler et al. 1995a, 1995b) and is therefore presumably more genetically related to schizophrenia than to the schizoid dimension (Torgersen 1994; Siever et al. 1995). The PAS has two versions: one for the subject and the other for parents or close relatives. A rater (EC.) who was blind to the phenomenology and diagnoses of patients administered the PAS to the parents or close relatives, because the information that schizophrenia patients report about themselves is often misleading. Schizophrenia dimensions and personality traits were assessed at the time of hospital discharge when the symptomatology had stabilized. Statistical analyses consisted of Pearson correlation coefficients between personality and schizophrenia dimensions and partial correlations to allow for the influence of overlapping dependent variables. The Bonferroni correction (Grove and Andreasen 1982) was used to exclude spurious significant correlations due to type I errors. Internal consistency of the personality dimensions was assessed through Cronbach's alpha analyses. Regression analyses were carried out through the stepwise method. The SPSS package was used to perform the statistical procedures. Results No differences were found in age, duration of illness, and years of education when patients were grouped by diagnosis, which was probably because onset of illness was recent in all patients. A high global rating of the reliability of information collected from the informants was found by the rater who assessed patients' premorbid personality. This high score may be due to the fact that informants and patients were biologically related in more than 90 percent of cases; in only 9 cases were the informants spouses or secondary relatives, such as uncles or aunts (table 1). The internal consistency of personality dimensions evaluated by Cronbach alpha (Cronbach 1951) was as follows: sociopathic (a = 0.78), passive-dependent (a = 0.77), anancastic (a = 0.57), schizoid (a = 0.86), and schizotypy (a = 0.40). No significant differences in personality or schizophrenia dimensions were found between sexes. key scores were significantly correlated with the positive dimension {r = 0.26, p < 0.003), the negative dimension (r = 0.35, p < 0.000), and the disorganization dimension (r = 0.19, p < 0.023). premorbid traits were significantly related to the disorganization dimension (r = 0.26, p < 0.003). Schizotypal traits were significantly associated with the positive dimension (r = 0.22, p < 0.011). There was also a significant relationship between the passive-dependent dimension and the disorganization dimension (r = 0.20, p < 0.017). No significant relationships were found between schizophrenia dimensions and the anancastic personality dimension. After Bonferroni correction (p 0.003), associations between schizoid traits and the positive and negative dimensions, and between sociopathic traits and the disorganization dimension remained significant (table 2). When only patients with schizophrenia (n = 64) were studied, the same profile of significant relationships appeared, although at a lower level of significance, probably because of the effect of the sample size on statistical power. To avoid the potential effect of illness itself on the relatives' description of premorbid personality, we selected patients with a maximum of 3 years of illness duration. The sample totaled 53 patients (11 schizophreniform, 32 with schizophrenia, 5 schizoaffective, and 5 affective). The distribution of significant relationships between personality and schizophrenia dimensions differed slightly from that found in the total sample. The negative dimension was strongly associated with the schizoid dimension, and a strong relationship also appeared between the positive dimension and schizotypy. The relationships between disorganization and sociopathic dimension and between the negative dimension and passive-dependent traits approached significance after the Bonferroni correction (table 2). The personality dimensions correlated strongly (average r = 0.35, range = ). The only association that was not significant after Bonferroni correction was between schizoid and sociopathic dimensions (r = 0.12, ). The high overlap between personality dimensions required the use of partial correlational analyses in order to control its effect on the relationships between schizophrenia and personality dimensions. premorbid traits were significantly related to the three schizophrenia dimensions: the association was strong with the negative dimension, intermediate with the disorganization dimension, and weak with the positive dimension. traits were significantly associated with the disorganization dimension (table 3). The anancastic dimension tended to be negatively associated with positive and negative symptomatology. The schizotypal dimension showed only a modest relationship with positive symptomatology after allowing for the influence of the remaining personality dimensions. The passive-dependent dimension was not 804

5 Premorbid Personality in Psychoses Schizophrenia Bulletin, Vol. 25, No. 4, 1999 Table 2. Pearson correlation coefficients between dimensions of personality and SAPS-SA schizophrenia dimensions Passive-dependent Anancastic Schizotypy Sample with < 3 yrs Passive-dependent Anancastic Schizotypy r Positive illness duration (n = 53) P r Negative P Disorganization Note. Boldface indicates associations that remained significant at p s 0.05 after Bonferroni correction (p = 0.003). SAPS = Scale for the Assessment of Positive Symptoms; SA = Scale for the Assessment of Negative Symptoms; = not significant. r P significantly related with schizophrenia symptomatology (table 3). To better understand the relationships between personality dimensions and symptomatology, we carried out a separate set of regression analyses using positive, negative, and disorganization dimensions as dependent variables and personality dimensions as independent variables. These results are represented graphically in figure 1 for the total sample and in figure 2 for the sample with a maximum of 3 years of illness duration. The relationships between the three-dimensional model of schizophrenia symptoms and premorbid traits of personality structure were more evident in patients with short illness duration. Discussion In this relatively young mixed sample of subjects with psychoses, there were significant relationships between schizoid premorbid personality and the negative and positive schizophrenia dimensions, and between sociopathic premorbid personality and the disorganization dimension. The study also showed weaker associations between schizoid premorbid traits and the disorganization dimension, and a slight tendency of the schizotypal dimension to be related with positive symptomatology. Furthermore, these results were more evident in patients with a maximum of 3 years' illness duration, who were analyzed to collect more reliable information. The schizoid dimension was significantly associated with the negative schizophrenia dimension, and schizotypy with the positive dimension. traits and the disorganization dimension showed a strong tendency toward significance with the Bonferroni correction. The relationship between passivedependent traits and the negative dimension was also near statistical significance. Taken together, these results provide support for a continuity between premorbid personality traits and schizophrenia dimensions. Table 3. Partial correlations between personality and SAPS-SA schizophrenia dimensions after allowing for the influence of the remaining personality dimensions Passive-dependent Anancastic Schizotypy r Positive P r Negative P Disorganization Note. Boldface indicates associations that remained significant at p s 0.05 after Bonferroni correction (p = 0.003). SAPS = Scale for the Assessment of Positive Symptoms; SA = Scale for the Assessment of Negative Symptoms; = not significant. r P

6 Schizophrenia Bulletin, Vol. 25, No. 4, 1999 M.J. Cuesta et al. Figure 1. Relationships between premorbid personality dimensions and schizophrenia dimensions dimension dimension Schizophrenia Dimensions Results of separate multiple regression analyses through stepwise procedure of personality dimensions with schizophrenia dimensions as dependent variable (only parameters of personality variables entering in the equation are shown). Variables Passive-dependent Anancastic Schizotypy B 0.12 Positive Beta T B 0.22 Beta 0.35 Negative T B 0.15 Disorganization Beta 0.26 T MR 0.26 R 2 F df , MR 0.35 R 2 F df 1,109 sig MR 0.26 R 2 F df 1, Note. MR = multiple R;R 2 = R square; F = F de Snedecor; df= degrees of freedom; = significant p. No previous studies of the relationships between premorbid personality and schizophrenia symptomatology have been performed from a three-dimensional perspective of psychotic disorders. Our results, however, corroborate in part a number of studies that report a greater frequency of schizoid and schizotypal traits in patients with schizophrenia (Hogg et al. 1990) and in their relatives (Kendler 1985). Moreover, in a previous study in a sample of 115 patients with schizophrenia, we found higher negative symptoms (affective flattening and alogia) in patients with previous schizoid or schizotypal personalities (Peralta et al. 1991). Cannon et al. (1990) followed a group of 138 high-risk individuals and found that those individuals regarded by their teachers as passive, socially isolated, and unresponsive to praise in late childhood and early adolescence displayed a predominantly negative schizophrenia disorder. In contrast, schizophrenia patients with predominantly positive manifestations were described by their teachers during the same period as overactive, irritable, distractible, and aggressive. Finally, McCreadie et al. (1994) found more current negative symptoms in schizophrenia patients with schizoid or schizotypal premorbid traits. To avoid the overlapping of asociality in personality and schizophrenia dimensions, we removed anhedoniaasociality from the negative dimension and repeated the 806

7 Premorbid Personality in Psychoses Schizophrenia Bulletin, Vol. 25, No. 4, 1999 Figure 2. Sample with a maximum of 3 years illness duration (n = 53) Schizotypal dimension dimension R 2 =0.17 dimension R 2 =0.08 Schizophrenia Dimensions Results of separate multiple regression analyses through stepwise procedure of personality dimensions with schizophrenia dimensions as dependent variable (only parameters of personality variables entering in the equation are shown). Variables Passive-dependent Anancastic Schizotypy B 0.27 Positive Beta T B 0.24 Beta 0.41 Negative T B 0.17 Disorganization Beta T MR 0.41 R 2 F df 1, MR 0.41 R 2 F df 1, MR 0.30 R 2 F df , Note. MR = multiple R;R 2 = R square; F = F de Snedecor; df= degrees of freedom; = significant p. same correlational analyses between this dimension and the five personality dimensions. nificant relationships were similar to those of the original negative dimension, suggesting that the association of negative symptoms and personality dimensions was independent of the overlapping of asociality in both domains. The relationship between sociopathic personality and the disorganization dimension has not been reported previously. Dalkin et al. (1994), however, found explosive and paranoid traits to be more common in patients with schizophrenia than in patients with other nonorganic psychoses. Jackson et al. (1991) also found higher proportions of schizotypal and antisocial subtypes of personality in a sample of recent-onset inpatients. Explosive and antisocial traits can be included in the domain of sociopathic personality. Berembaum and Fujita (1994) report the same results in a review of the literature focusing on the differentiability of personality traits of patients with schizo- 807

8 Schizophrenia Bulletin, Vol. 25, No. 4, 1999 M.J. Cuesta et al. phrenia and those who do not have schizophrenia. Several results reported in the Copenhagen high-risk studies suggest that "class-disturbing behavior" had a higher predictive power for schizophrenia than for schizotypy in followup (Parnas and Jorgensen 1989). And formal thought disorder, included in the disorganization dimension, was predictive of later development of a schizophrenia spectrum disorder (Parnas et al. 1982). A three-dimensional structure is therefore demonstrated in schizophrenia disorders (Bilder et al. 1985; Liddle 1987; Ardnt et al. 1991; Peralta et al. 1992); across all psychoses (Peralta et al. 1997); in schizotypal traits of nonpsychotic populations, such as normal students (Muntaner et al. 1988; Bentall et al. 1989; Raine et al. 1994); in patients with personality (Bergman et al. 1996) or psychiatric disorders (Vollema and van den Bosch 1995); and in the premorbid personality of patients with psychoses included with other dimensions. This threedimensional psychotic structure from normality to schizophrenia spectrum disorders appears to indicate that schizophrenia dimensions and their related premorbid personality dimensions share underlying neurobiological abnormalities. In a related study, Siever (1994) presented arguments based on genetic, neuropsychological, neuroimaging, neurofunctional, and biochemical abnormalities to suggest strong similarities between the deficit-like symptoms of schizotypal disorder, which are close to those in the schizoid dimension in our study, and negative or deficit schizophrenia, and between the psychotic-like schizotypy and the psychotic dimension of schizophrenia. Maier et al. (1994), like Siever, found that negative symptoms of schizotypal personality were "the main source of familial aggregation, but psychotic-like personality features are also contributing factors" in schizophrenia spectrum disorders. Moreover, schizotypal traits in the normal population were associated with neuropsychological and structural disturbances similar to those of schizophrenia patients (Raine et al. 1992); schizotypal patients with a history of schizophrenia displayed more neurocognitive disturbances than those without a family background of schizophrenia, although the clinical profile did not differ between the groups (Condray and Steinhauer 1992); and, as a group, patients with schizotypal personality disorder presented more cognitive disturbances than normal subjects or those with any other personality disorders (Trestman et al. 1995). Support for the concept of continuity between personality and symptomatology of psychosis is found in the work of classic psychiatrists, who proposed an etiological relationship between the two dimensions, and in the study by Claridge and Broks (1984), who, following Eysenck's hypothesis, postulated that biological, cognitive, and personality disturbances of patients with schizophrenia are also present in normal personalities. Furthermore, they suggested that these disturbances are the genetic traits that produce the vulnerability to schizophrenia. In contrast, Meehl (1990) suggested that the schizotypy construct is independent of normal personality and is biologically or genetically derived. This continuity model could also be accounted for by one of the most compelling hypotheses of schizophrenia disorders, described by Murray (1994) and based on disturbances in the neurodevelopmental process, since neuromotor abnormalities and childhood behavioral problems have been found to be precursors of adult symptom dimensions in schizophrenia (Foerster et al. 1991; Baum and Walker 1995; Walker et al. 1996). This hypothesis holds that premorbid personality dimensions should be considered as precursors of psychosis; in addition, some of these dimensions have been found as asymptomatic traits in relatives. Kendler et al. (1995a) reported that three schizotypal factors very close to the schizophrenia dimensions could discriminate between relatives of schizophrenia patients and control relatives, suggesting that these factors may constitute the familial vulnerability to schizophrenia and other nonaffective psychoses. These results suggested that personality dimensions, rather than the specific typology of personality, can be considered as putative phenotypes related to neurodevelopmental abnormalities for genetic studies. Therefore, not only is a continuity between personality and schizophrenia dimensions likely, but it could be hypothesized that these three personality dimensions contribute to patients' individual vulnerability to psychoses. The addition of these and other risk factors would determine the vulnerability of each patient (McGlashan and Johannessen 1996). This dimensional approach may also overcome problems in the search for spectrum phenotypes in genetic studies (Faraone et al. 1995). A number of studies of vulnerability seem to corroborate our hypothesis. Vulnerability to delusions over time in first-episode affective patients with delusions has been reported to remain as a trait-like vulnerability (Harrow et al. 1995). Nonetheless, vulnerability not only appears after illness has begun, but a poor premorbid functioning, among other factors, was predictive of poor outcome in unipolar affective disorders (Sands and Harrow 1995; Vocisano et al. 1996). Poor premorbid functioning, as it is usually evaluated, corresponds to a deficiency in establishing significant relationships in late childhood and adolescence, one of the hallmarks of the schizoid personality. Moreover, Dworkin et al. (1994) found that adolescents at risk of schizophrenia demonstrated more deficits in social competence (another function that is disturbed in the schizoid dimension) than those at risk of affective disorder or normal controls, but this was not the case in school 808

9 Premorbid Personality in Psychoses Schizophrenia Bulletin, Vol. 25, No. 4, 1999 adjustment. Finally, thought disorder, which is included within the disorganization dimension, has been proposed as a marker of familial vulnerability to schizophrenia (Hain et al. 1995), and it has also been suggested that thought disorder is distributed as a continuum ranging from normality to schizophrenia symptomatology (Johnston and Holzman 1979). Several limitations of this study should be taken into account. First, the sample was predominantly composed of males (65.2%), and the personality structure may not be the same in both sexes; however, no sex-related differences were found between dimensions of premorbid personality either in our study or in others (Jackson et al. 1991; Dalkin et al. 1994). Second, retrospective assessments after illness onset are not the best way to assess premorbid personality. To account for this limitation we chose the interview with relatives (predominantly biological) as a more reliable method than self-questionnaires or self-description of personality traits. We also reanalyzed the data in a subset of patients with a shorter period of evolution and found quite similar results. Moreover, the PAS interview includes a global rating to score the quality of information received from the relatives, which achieved a high average (table 1). Third, one of our modifications of the PAS interview was the derivation of a schizotypy dimension from the original schizoid PAS dimension, given the special interest of this dimension in schizophrenia disorders. The schizotypy dimension, however, showed a low internal consistency, and the results obtained for this dimension should be treated with caution. Fourth, we studied only patients with psychotic symptoms, and the results cannot be extrapolated to other mental disorders without psychotic symptoms. While the most important conclusion of this study is that three personality dimensions (schizoid, sociopathic, and schizotypy) may increase vulnerability to psychosis, it is also interesting that passive-dependent and anancastic dimensions were not related to the schizophrenia symptomatology and therefore cannot be considered as protective factors against the risk of suffering schizophrenia symptoms. References American Psychiatric Association. DSM-IH-R: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed., revised. Washington, DC: The Association, Andreasen, N.C. Scale for the Assessment of Positive Symptoms (SAPS). Iowa City, IA: University of Iowa, 1984a. Andreasen, N.C. Scale for the Assessment of Negative Symptoms (SA). Iowa City, IA: University of Iowa, Ardnt, S.; Alliger, R.F.; and Andreasen, N.C. The distinction of positive and negative symptoms: The failure of a two-dimensional model. British Journal of Psychiatry, 158: , Baum, K.M., and Walker, E. Childhood behavioral patterns of adult symptom dimensions in schizophrenia. Schizophrenia Research, 16: , Bentall, R.P.; Claridge, G.S.; and Slade, P.D. The multidimensional nature of schizotypal traits: A factor analytic study with normal subjects. British Journal of Clinical Psychology, 28: , Berembaum, H., and Fujita, F. Schizophrenia and personality: Exploring the boundaries and connections between vulnerability and outcome. Journal of Abnormal Psychology, 103: ,1994. Bergman, A.J.; Harvey, P.D.; Mitropulou, V.; Aronson, A.; Marder, D.; Silverman, J.; Trestman, R.; and Siever, L.J. The factor structure of schizotypal symptoms in a clinical population. Schizophrenia Bulletin, 22: ,1996. Bilder, R.M.; Mukherjee, S.; Rieder, R.O.; and Pandurangi, A.K. Symptomatic and neuropsychological components of defect states. Schizophrenia Bulletin, 11:409^119, Bleuler, E. Textbook of Psychiatry. MacMillan, New York, NY: Bleuler, M. The schizophrenic disorders: Long-term patients and family studies. New Haven, CT: Yale University Press, Cannon, T.D.; Mednick, S.A.; and Parnas, J. Antecedents of predominantly negative and predominantly positive symptom schizophrenia in a high risk population. Archives of General Psychiatry, 47: , Chapman, L.J., and Chapman, J.P. Scales for rating psychotic and psychotic-like experiences as continua. Schizophrenia Bulletin, 6: , Chapman, L.J.; Chapman, J.P.; Kwapil, T.R.; Eckblad, M.; and Zinser, M.C. Putatively psychosis-prone subjects 10 years later. Journal of Abnormal Psychology, 103: , Cicchetti, D., and Tyrer, P. Reliability and validity of personality assessment. In: Tyrer, P., ed. Personality Disorders: Diagnosis, Management and Course. London: Wright, pp Claridge, G., and Broks, P. Schizotypy and hemisphere function. I. Theoretical considerations and the measurement of schizotypy. Personality and Individual Differences, 5: ,

10 Schizophrenia Bulletin, Vol. 25, No. 4, 1999 M.J. Cuesta et al. Condray, R., and Steinhauer, S.R. Schizotypal personality disorder in individuals with and without schizophrenic relatives: Similarities and contrasts. Schizophrenia Research, 7:33-41, Cronbach, L.J. Coefficient alpha and the internal structure of tests. Psychometrika, 16: , Dalkin, T.; Murphy, P.; Glazebrook, C; Medley, Y.; and Harrison, G. Premorbid personality in first-onset psychosis. British Journal of Psychiatry, 164: , Done, J.; Crow, T.; Johnstone, E.C.; and Sacker, A. Childhood antecedents of schizophrenia and affective illness: Social adjustment at ages 7 and 11. Lancet, 309: , Dworkin, R.H.; Lewis, J.; Cornblatt, B.A.; and Erlenmeyer-Kimling, L. Social competence deficits in adolescents at risk for schizophrenia. Journal of Nervous and Mental Disease, 182: , Erlenmeyer-Kimling, L.; Squires-Wheeler, E.S.; Adamo, U.H.; Basset, A.S.; Cornblatt, B.A.; Kestenbaum, C.J.; Rock, D.; Roberts, S.A.; and Gottesman, I.I. The New York High-Risk Project: Psychoses and cluster A personality disorders in offspring of schizophrenic parents at 23 years of follow-up. Archives of General Psychiatry, 52: , Faraone, S.V.; Kremen, W.S.; Lyons, M.J.; Pepple, J.R.; Seidman, L.J.; and Tsuang, M.T. Diagnostic accuracy and linkage analysis: How useful are schizophrenia spectrum phenotypes? American Journal of Psychiatry, 152: , Foerster, A.; Lewis, S.W.; Owen, M.J.; and Murray, R.M. Premorbid personality in psychosis: Effects of sex and diagnosis. British Journal of Psychiatry, 158: , Green, M., and Walker, E. Neuropsychological performance and positive and negative symptoms in schizophrenia. Journal of Abnormal Psychology, 94: , Grove, W.M., and Andreasen, N.C. Simultaneous tests of many hypotheses in exploratory research. Journal of Nervous and Mental Disease, 170:3-9, Hain, C; Maier, W.; Hoeschst-Janneck, S.; and Franke, P. Subclinical thought disorder in first-degree relatives of schizophrenic patients. Acta Psychiatrica Scandinavica, 92: ,1995. Harrow, M.; MacDonald, A.W., III; Sands, J.R.; and Silverstein, M.L. Vulnerability to delusions over time in schizophrenia and affective disorders. Schizophrenia Bulletin, 21:95-109, Helmes, E.; Landmark, J.; and Kazarian, S.S. Inter-rater reliability of twelve diagnostic systems of schizophrenia. Journal of Nervous and Mental Disease, 171: , Hogg, B.; Jackson, H.J.; Rudd, R.P.; and Edwards, J. Diagnosing personality disorders in recent-onset schizophrenia. Journal of Nervous and Mental Disease, 178: ,1990. Jackson, H.J.; Whiteside, H.L.; Bates, G.W.; Beel, R.; Rudd, R.P.; and Edwards, J. Diagnosing personality disorders in psychiatric inpatients. Acta Psychiatrica Scandinavica, 83: , Jaspers, K. General Psychopathology. Chicago, IL: University of Chicago Press, Johnston, M.H., and Holzman, P.S. Assessing Schizophrenic Thinking. San Francisco, CA: Jossey-Bass, Jorgensen, A., and Parnas, J. The Copenhagen high-risk study: Premorbid and clinical dimensions of maternal schizophrenia. Journal of Nervous and Mental Disease, 178: , Kendler, K.S. Diagnostic approaches to schizotypal personality disorder: A historical perspective. Schizophrenia Bulletin, 11: , Kendler, K.S.; McGuire, M.; Gruenger, A.M.; and Walsh, D. Schizotypal symptoms and signs in the Roscommon family study. Archives of General Psychiatry, 52: , 1995a. Kendler, K.S.; Neale, M.C.; and Walsh, D. Evaluating the spectrum concept of schizophrenia in the Roscommon family study. American Journal of Psychiatry, 152: , 1995*. Krestchmer, E. Krestchmer's Textbook of Medical Psychology. London: Oxford University Press, Landmark, J. A manual for the assessment of schizophrenia. Acta Psychiatrica Scandinavica, 65(Suppl 298): 1-80, Landmark, J.; Cernvsky, Z.Z.; Merskey, H.; and Leslie, B. Interrelationships of systems for diagnosing schizophrenia. Comprehensive Psychiatry, 27: , Lenzeweger, M.F. Psychometric high-risk paradigm, perceptual aberrations and schizotypy: An update. Schizophrenia Bulletin, 20: , Liddle, P.F. The symptoms of chronic schizophrenia. British Journal of Psychiatry, 151: , Maier, W.; Lichtermann, D.; Minges, J.; and Heun, R. Personality disorders among the relatives of schizophrenia patients. Schizophrenia Bulletin, 20(3):481^93, McCreadie, R.G.; Connolly, M.A.; Williamson, D.J.; Athawes, R.W.B.; and Tilak-Sihgh, D. The Nithsdale Schizophrenia Surveys. XII. Neurodevelopmental schizo- 810

11 Premorbid Personality in Psychoses Schizophrenia Bulletin, Vol. 25, No. 4, 1999 phrenia: A search for clinical correlates and putative aetiological factors. British Journal of Psychiatry, 165: , McGlashan, T.H., and Johannessen, J.O. Early detection and intervention with schizophrenia: Rationale. Schizophrenia Bulletin, 22: , McGorry, P.D.; Copolov, L.; and Singh, B.S. Royal Park Multidiagnostic Instrument for Psychosis: Part I. Rationale and review. Schizophrenia Bulletin, 16: , Meehl, P.E. Schizotaxia revisited. Archives of General Psychiatry, 46: ,1990. Mukherjee, S.; Reddy, R.; and Schnur, D.B. A developmental model of negative syndromes in schizophrenia. In: Greden, J.F., and Tandon, R., eds. Negative Schizophrenic Symptoms: Pathophysiology and Clinical Implications. Washington, DC: American Psychiatric Press, pp Muntaner, C; Garcia-Sevilla, L.; Fern&ndez, A. and Torrubia, R. Personality dimensions, schizotypal and borderline personality traits and psychoses proneness. Personality and Individual Differences, 9: , Murray, R.M. Neurodevelopmental schizophrenia: The rediscovery of Dementia Praecox. British Journal of Psychiatry, 165(Suppl 25):6-12, Parnas, J., and Jorgensen, A. Pre-morbid psychopathology in schizophrenia spectrum. British Journal of Psychiatry, 155: , Parnas, J.; Schulsinger, F.; Teasdale, T.W.; Feldman, H.; and Mednick, S.A. Perinatal complications and clinical outcome within the schizophrenia spectrum. British Journal of Psychiatry, 140: , Peralta, V.; Cuesta, M.J.; and De Leon, J. Premorbid personality and positive and negative symptoms in schizophrenia. Acta Psychiatrica Scandinavica, 84: , Peralta, V., and Cuesta, M.J. A polydiagnostic approach to self-perceived cognitive disorders in schizophrenia. Psychopathology, 25: , Peralta, V.; Cuesta, M.J.; and Farre, C. Factor structure of symptoms in functional psychoses. Biological Psychiatry, 42: , Peralta, V.; De Leon, J.; and Cuesta, M.J. Are there more than two syndromes in schizophrenia. A critique of the positive-negative dichotomy. British Journal of Psychiatry, 161: , Raine, A.; Reynolds, C; Lencz, T.; Scerbo, A.; Triphon, N.; and Kim, D. Cognitive-perceptual, interpersonal, and disorganized features of schizotypal personality. Schizophrenia Bulletin, 20(l): , Raine, A.; Sheard, C; Reynolds, G.P.; and Lencz, T. Prefrontal structural and functional deficits associated with individual differences in schizotypal personality. Schizophrenia Research, 7: , Sands, J.R., and Harrow, M. Vulnerability to psychosis in unipolar major depression: Is premorbid functioning involved? American Journal of Psychiatry, 152: , Schneider, K. Psychopathic Personalities. London: Cassell, Siever, L.J. Biological factors in schizotypal personality disorders. Acta Psychiatrica Scandinavica, 90(Suppl 34):45-50, Siever, L.J.; Bergman, A.J.; and Keefe, R.S.E. The schizophrenia spectrum personality disorders. In: Hirsch, S.R., and Weinberger, D.R., eds. Schizophrenia. Oxford, England: Blackwell Science, pp Trestman, R.L.; Keefe, R.S.E.; Mitropoulou, V.; Harvey, P.D.; devegvar, M.L.; Less-Roitman, S.; Davidson, M.; Aronson, A.; Silverman, J.; and Siever, L.J. Cognitive function and biological correlates of cognitive performance in schizotypal disorder. Psychiatry Research, 59: , Torgersen, S. Personality deviations within the schizophrenia spectrum. Acta Psychiatrica Scandinavica, 90 (Suppl 34):40-44, Tyrer, P. Personality Disorders: Diagnosis, Management and Course. London: Wright, Tyrer, P., and Alexander, J. Personality Assessment Schedule (PAS). In: Tyrer, P., ed. Personality Disorders: Diagnosis, Management and Course. London: Wright, pp Vocisano, C; Klein, D.N.; Keefe, R.S.; Dienst, E.R.; and Kincaid, M. Demographics, family history, premorbid functioning, developmental characteristics, and course of patients with deteriorated affective disorder. American Journal of Psychiatry, 153: ,1996. Vollema, M.G., and van den Bosch, R.J. The multidimensional! ty of schizotypy. Schizophrenia Bulletin, 21:19-31, Walker, E.; Levine, R.R.J.; and Neuman, C. Childhood behavioral characteristics and adult brain morphology in schizophrenia. Schizophrenia Research, 22:93-101, The Authors Manuel J. Cuesta, M.D., is staff psychiatrist; Victor Peralta is staff psychiatrist, both at Virgen del Camino Hospital in Pamplona, Spain. Francisco Caro is staff psychologist at Instituto de Psiquiatria Pere Mata in Reus, Spain. 811

12