Abstract
Associations between symptoms and interpersonal problems, assessed with the Symptom Checklist–90 (SCL–90–R) and the Inventory of Interpersonal Problems–64-item version (IIP–64), are examined in a large psychiatric outpatient sample. On the basis of the IIP–64 scores, the sample was divided into 8 subgroups, made up of different types of predominant interpersonal problems. These octant groups were used as independent variables in analyses testing hypothesized associations with symptom subscales of the SCL–90–R. In general, strong associations between symptoms and interpersonal problems were found. In addition, hostile and paranoid ideation symptoms displayed significant differences among octant groups, and were associated with interpersonal problems of the vindictive/self-centered kind. Phobic anxiety was associated with interpersonal problems of the socially inhibited kind. Assessing specific combinations of symptoms and interpersonal problems might be useful in treatment planning and evaluation.
Acknowledgments
We would like to thank Svein Friis for his support in completing this article. The article forms part of a doctoral thesis to be submitted to the Department of Psychology, University of Oslo, Norway.
Notes
Several of the personality disorders enlisted in the Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM–IV]; American Psychiatric Association, Citation1994), which have been shown to correlate highly with specific parts of the IPC, have proposed for deletion from DSM–V. Six specific personality disorder types are proposed (antisocial, avoidant, borderline, narcissistic, obsessive–compulsive, and schizotypal), and the remaining (paranoid, schizoid, histrionic, and dependent) are allocated to Personality Disorder Trait Specified (PDTS), and measured by the Levels of Personality Functioning Scale. The levels of personality functioning are based on the severity of disturbances of the self and interpersonal functioning. In this perspective, instruments like the IIP–64 will probably be even more relevant than they are today (Widiger, Citation2010).
The seemingly higher scores in the in the PA octant (cf. ) diminish in the t tests due to different standard errors of differences. For instance, the score difference between PA and HI is 0.954, which, divided with SE diff = 0.116 results in a t value = 8.224, whereas the score difference between BC and HI is 0.895, which, divided with SE diff = 0.088, gives a t value of 10.170. This also applies to seemingly larger mean scores on other symptom subscales, for example, the NO and PA scores on the PAR subscale.