Avoidant personality disorder is a separable schizophrenia-spectrum personality disorder even when controlling for the presence of paranoid and schizotypal personality disorders: The UCLA family study

https://doi.org/10.1016/j.schres.2006.12.023Get rights and content

Abstract

It is unresolved whether avoidant personality disorder (APD) is an independent schizophrenia (Sz)-spectrum personality disorder (PD). Some studies find APD and social anxiety symptoms (Sxs) to be separable dimensions of psychopathology in relatives (Rels) of schizophrenics while other studies find avoidant Sxs to be correlated with schizotypal and paranoid Sxs.

Rates of APD among first-degree Rels of Sz probands, attention-deficit/hyperactivity disorder (ADHD) probands, and community control (CC) probands were examined. Further analyses examined rates when controlling for the presence of schizotypal (SPD) and paranoid (PPD) personality disorders, differences in APD Sxs between relative groups, and whether APD in Rels of Szs reflects a near miss for another Sz-spectrum PD.

Three hundred sixty-two first-degree Rels of Sz probands, 201 relatives of ADHD probands, and 245 Rels of CC probands were interviewed for the presence of DSM-III-R Axis I and II disorders. Diagnoses, integrating family history, interview information, and medical records, were determined.

APD occurred more frequently in Rels of Sz probands compared to CC probands (p < 0.001) and also when controlling for SPD and PPD (p < 0.005). Two Sxs of APD were most characteristic of the Rels of Sz probands: “avoids social or occupational activities…” and “exaggerates the potential difficulties…” 65% of the Rels of Sz probands who had diagnoses of APD were more than one criterion short of a DSM-III-R diagnosis of either SPD or PPD. This indicates that APD is a separate Sz-spectrum disorder, and not merely a sub-clinical form of SPD or PPD.

Introduction

The detection of relatives of schizophrenia (Sz) probands who are affected by a schizophrenia-spectrum diagnosis allows delineation of an extended phenotype that may be helpful for establishing patterns of familial transmission. Accepted schizophrenia-spectrum disorders include: Sz, schizoaffective disorder, atypical psychosis, schizotypal personality disorder and paranoid personality disorder. Some family studies have also found a familial relationship of Sz with avoidant personality disorder which suggests that avoidant personality disorder is a schizophrenia-spectrum disorder (Asarnow, 2001, Baron et al., 1985, Kendler, 1993).

Meehl (1962) hypothesized that social anxiety is part of the schizotypy core, predisposing to Sz. We review below several other lines of research which support the association between symptoms of social anxiety and a liability to Sz: social anxiety symptoms in the relatives of schizophrenics, social anxiety in patients with Sz, examination of “high-risk” children for early predictors of psychosis, and factor analyses demonstrating that social anxiety is separable from other types of schizophrenia-spectrum related psychopathology. These studies offer mixed support for the hypothesis that symptoms of social anxiety may represent an extended phenotype that may be helpful in the delineation of the familial transmission of Sz.

Family studies have found an increased prevalence of social anxiety personality traits in the family members of Sz probands (Torgersen, 1994). Studies of psychosis proneness in relatives (Lyons, 1995, Bailey et al., 1993) report that avoidant personality disorder in the relatives of patients with Sz explains a modest amount of variance in the perceptual aberration and social anhedonia scales completed by these relatives (Eckblad et al., 1982, Chapman et al., 1978). A study of “social closeness” and emotional blunting in first-degree relatives of probands with Sz shows that the relatives score poorly on a measure of social closeness which most likely reflects anhedonia, social anxiety, and odd behaviors (Berenbaum et al., 1994). These studies do not address whether social anxiety in the relatives is separable from other measures of psychosis proneness.

Social anxiety is highly prevalent in outpatients with Sz and is unrelated to clinical psychotic symptoms (Pallanti et al., 2004). Social anxiety comorbid with Sz is associated with higher risk for suicide attempts, greater lethality of suicide attempts, more substance abuse, lower social adjustment, and lower overall quality of life (Bayle et al., 2001, Blanchard et al., 1998, Pallanti et al., 2004, Taiminen, 2001). The rate of social anxiety ranges from 13% to 39% in patients with Sz (Bermanzohn, 2000, Cassano et al., 1999, Cosoff and Hafner, 1998, Kendler et al., 1995a). “Shyness” as a personality trait is greater in schizophrenics than in controls (Flanagan, 1992). Social anxiety and shyness have been studied in the interactions of schizophrenic patients with others (Fingeret et al., 1985, Heinssen and Glass, 1990, Monti, 1984, Morrison and Bellack, 1987, Penn et al., 1994, Pilkonis et al., 1980). These studies were not designed to determine whether social anxiety was an independent factor from positive or negative symptoms. However, one study that examined this question directly found social anxiety to be unrelated to clinical psychotic symptoms (Pallanti et al., 2004). More recently, studies have shown that avoidant personality scores were highly intercorrelated with all DSM cluster A personality disorder (Spitzer et al., 1990) dimensional scores and that avoidant personality disorder is highly prevalent in outpatients with schizotypal personality disorder (Keshavan, 2005, Battaglia et al., 1995). These latter studies are at odds with the finding that social anxiety symptoms are unrelated to other dimensions of psychosis (Pallanti et al., 2004).

Studies of children at high risk for developing Sz have examined the prevalence of social anxiety related symptoms. These subjects score poorly on measures of peer relations, adaptation to school, hobbies, interests, and social–sexual adjustment (Dworkin et al., 1993), which may reflect high levels of social anxiety. Two studies with similar findings (Auerbach et al., 1993, Hans et al., 2000) support the conclusion that adolescents at risk for Sz have social deficits that are independent of early-onset schizotypal personality disorder and paranoid personality disorder and may reflect a vulnerability indicator to Sz. A prospective study found that premorbid social phobia was associated with more than a 3.5 times increased odds of developing Sz (Tien and Eaton, 1992). Prospective studies found that during childhood, early adolescence and adolescence, poor social competence is more characteristic of children at risk for Sz than children at risk for affective disorder (Dworkin, 1991, Dworkin et al., 1994). These studies are unable to sort out whether social anxiety in the relatives is separable from other measures of psychosis proneness.

Factor analytic studies have found that avoidant personality disorder and avoidant/social anxiety symptoms are a separable dimension of psychopathology found in normals, individuals who are at heightened risk for developing schizophrenia-spectrum disorders, individuals with schizotypal personality disorder, and the first-degree relatives of probands with Sz (Bentall et al., 1989, Fogelson, 1999, Kendler et al., 1995b, Raine, 1994, Tyrka, 1995, Vollema and van den Bosch, 1995).

These studies have left unresolved whether avoidant personality disorder and social anxiety in first-degree relatives of Sz probands overlap with other Sz-spectrum disorders. Deciphering the relationship between avoidant personality disorder and other schizophrenia-spectrum disorders requires blind evaluation of relatives of patients with Sz compared to community control relatives. (Keshavan et al., 2005) To address these concerns we examined if avoidant personality disorder shows heightened rates among first-degree relatives of Sz probands compared to relatives of control groups and shows heightened rates even when controlling for the presence of schizotypal and paranoid personality disorders. We also examined whether some individual avoidant personality disorder symptoms are more characteristic of relatives of schizophrenics compared to relatives of community controls and if avoidant personality disorder when present in the first-degree relatives of Sz probands mainly reflects a near miss for a diagnosis of another schizophrenia-spectrum personality disorder. We examined the specificity of these familial relationships by comparing these rates and characteristics in the relatives of the Sz probands to the relatives of CCs as well as to the relatives of ADHD probands, another disorder with prominent cognitive features.

Section snippets

Diagnosis of first-degree relatives

Three hundred sixty-two first-degree relatives, age 18 and older, of probands with adult-onset schizophrenia (AOSz, n = 111 probands) and with childhood-onset schizophrenia (COSz, n = 51 probands), 201 relatives of probands with attention-deficit/hyperactivity disorder (ADHD, n = 113 probands), and 245 relatives of adult and child community control (CC) probands (n = 48 adult probands; n = 71 child probands) were blindly and directly interviewed for the presence of DSM-III-R axis I and II disorders. All

Statistical methods

Because members of the same family cannot be treated as independent observations due to shared genetics and shared social environment, family membership was included in all statistical models as a random effect. Analyses were conducted using SAS 9.1 PROC MIXED for linear models with random effects, or PROC GLIMMIX for logistic regression with random effects. Denominator degrees of freedom were calculated using the “containment method” as specified in SAS, and thus the df reported below may vary

Results

There were 34 (9.4%) cases of avoidant personality disorder in the Sz relatives, 5 (2.0%) in the CC relatives, and 11 (5.5%) in the ADHD relatives (see Table 2). The differences between the Sz group and the CC group were statistically significant (F = 11.28, df = 1,325, p < 0.001), whereas those between the Sz and ADHD groups were not (F = 2.65, df = 1,288, p = 0.11). The odds ratio is 5.0 for the Sz and CC relatives and 1.8 for the Sz and ADHD relatives.

When examining for the presence of avoidant

Discussion

We find that first-degree relatives of probands with COSz or AOSz are at increased risk for avoidant personality disorder compared to first-degree relatives of CC probands, but not compared to first-degree relatives of ADHD probands. This finding holds for avoidant personality disorder that is present in the absence of other schizophrenia-spectrum personality disorders. Even when controlling for the presence of paranoid and schizotypal personality disorders, the analysis demonstrated that

Acknowledgements

The Della Martin Foundation Grant support: MH-41953, Drs. Kendler, Neale, and Jacobson MH-49716 and MH-45112, Drs. Fogelson, Nuechterlein, Asarnow, Subotnik, and Payne MH-37705 and MH-66286, Dr. Nuechterlein.

The authors wish to thank Jim Mintz for his assistance with the statistical analysis.

References (59)

  • R.F. Asarnow

    Schizophrenia and schizophrenia-spectrum personality disorders in the first-degree relatives of children with schizophrenia

    Archives of General Psychiatry

    (2001)
  • J.G. Auerbach et al.

    Neurobehavioral functioning and social behavior of children at risk for schizophrenia

    Israel Journal of Psychiatry and Related Sciences Science

    (1993)
  • B. Bailey et al.

    The convergent and discriminant validity of the Chapman scales

    Journal of Personality Assessment

    (1993)
  • M. Baron et al.

    Familial transmission of schizotypal and borderline personality disorders

    American Journal of Psychiatry

    (1985)
  • M. Battaglia et al.

    A family study of schizotypal disorder

    Schizophrenia Bulletin

    (1995)
  • R.P. Bentall et al.

    The multidimensional structure of schizotypal traits: a factor analytic study with normal subjects

    British Journal of Clinical Psychology

    (1989)
  • S.A. Berenbaum et al.

    Family study of schizophrenia and personality

    Journal of Abnormal Psychology

    (1994)
  • P.C. Bermanzohn

    Hierarchical diagnosis in chronic schizophrenia: a clinical study of co-occurring syndromes

    Schizophrenia Bulletin

    (2000)
  • J.J. Blanchard et al.

    Anhedonia, positive and negative affect, and social functioning in schizophrenia

    Schizophrenia Bulletin

    (1998)
  • E. Bleuler

    Dementia Praecox or the Group of Schizophrenias

    (1950)
  • G.B. Cassano et al.

    Multiple anxiety disorder comorbidity in patients with mood spectrum disorders with psychotic features

    American Journal of Psychiatry

    (1999)
  • L.J. Chapman et al.

    Schizotypy and thought disorder as a high risk approach to schizophrenia

  • S.J. Cosoff et al.

    The prevalence of comorbid anxiety, schizophrenia, schizoaffective disorder, and bipolar disorder

    Australian and New Zealand Journal of Psychiatry

    (1998)
  • R.H. Dworkin

    Social competence and positive and negative symptoms: a longitudinal study of children and adolescents at risk for schizophrenia and affective disorders

    American Journal of Psychiatry

    (1991)
  • R.H. Dworkin

    Childhood precursors of affective vs. social deficits in adolescents at risk for schizophrenia

    Schizophrenia Bulletin

    (1993)
  • R.H. Dworkin et al.

    Social competence deficits in adolescents at risk for schizophrenia

    Journal of Nervous and Mental Disease

    (1994)
  • M.L. Eckblad et al.

    The Revised SAS

    (1982)
  • A.L. Fingeret et al.

    Social perception, social performance, & self-perception: a study with psychiatric and nonpsychiatric groups

    Behavior Modification

    (1985)
  • R. Flanagan

    Shyness, egocentricity, and psychopathology: their relationships among nonhospitalized individuals and mental hospital patients

    Psychological Reports

    (1992)
  • Cited by (0)

    The work described in this article has been carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans.

    View full text