EditorialThe relationship of social phobia to avoidant personality disorder: a proposal to reclassify avoidant personality disorder based on clinical empirical findings
Introduction
There has been a long-standing question as to how social fears relate to personality disorders. At one time anxiety was considered the hallmark of a personality disorder, e.g., the concept of neurosis. The development of the avoidant personality disorder (APD) category in DSM-III stimulated interest in the relationship of social fears to personality disorders. By examining the relationship of social phobia (SP) to the personality disorders clinicians and researchers could examine this question empirically in disorders defined by specific criteria.
The conceptualization of APD in DSM-III relied heavily on the work of Millon [22]. Millon felt that although APD would overlap to some extent with schizoid, schizotypal, dependent and paranoid personality disorders, it was clearly a category in its own right. (Many DSM personality disorders overlap with each other to some extent.) For Millon the distinction between APD and SP was clear. As he put it, “Avoidant is essentially a problem of relating to persons; social phobia is largely a problem of performing in situations. The avoidant PD has a feeling of low self-esteem; social phobia implies no such self critical judgment” [22].
SP researchers have examined the relationship between SP and personality disorder. One review is by Johnson and Lydiard. They address the diagnostic question, “…social phobia itself has many features in common with a personality disorder” [18]. They note that the rate of personality disorders appears higher among patients diagnosed as having SP than among other anxiety disorders. The high rate of APD in SP as well as other personality disorders (dependent, borderline, schizotypal and obsessive) is discussed. They make the observation that some empirical studies indicate that the presence of co-morbid depression appears to correlate highly with the presence of APD. APD without co-morbidity with SP may therefore, at times, be an epiphenomena of co-morbidity with another Axis I disorder. This would mean that the APD, in this case, would be only associated with depression and would not appear when the depression is not present. In this example a person who had the criteria for APD when depressed does not have these symptoms when the depression has resolved.
Johnson and Lydiard also examine the overlap between the generalized form of SP (GSP) and APD. Their review of empirical studies shows that while some distinctions can be made in different studies between GSP and APD the most remarkable finding is the similarity between the two disorders. They raise the question as to whether there are different subtypes of APD. One would be an overlapping concept with GSP while the second APD without GSP (which is rare, but diagnosed) would be a separate subtype. Johnson and Lydiard believe that at least four items of DSM-IV APD (criteria 1, 2, 3 and 7) emphasize shame and embarrassment and therefore overlap with SP; the clinical overlap of the two disorders is understandable.
Widiger [42] reviewed three empirical studies examining the overlap between GSP and APD. These studies used DSM-III-R criteria. He notes that while there are cases of GSP without APD, there are very few cases of APD without GSP. He feels that the criteria are written in such a way that there will be very few of the APD without GSP cases found. He concludes that although there are distinctions that can be made between GSP with and without APD, there is no evidence for a demarcation between the two disorders, which would justify the diagnosis of two separate categorical disorders. He concludes that the disorder should be listed on both the anxiety and personality disorder categories, even though it appears to be a single disorder.
Section snippets
Diagnostic issues in using the DSM
The diagnosis of both SP and APD using the DSM system has been a bit of a moving target as the criteria change somewhat with each revision (see table I). The DSM-III criteria for APD emphasized low self-esteem, social withdrawal and sensitivity to rejection. Social phobia emphasized social withdrawal, but also lists fear of being humiliated, which could be seen as very similar to the APD criteria of low self-esteem. Already we have questions about the overlap of the two disorders. If someone is
Studies comparing social phobia to APD
We now turn to empirical literature comparing SP and APD to answer two questions. The first is the co-morbidity of the two disorders. If the disorders were separate we would expect, at best, modest co-morbidity. If they were highly related or identical disorders we would expect a much higher overlap. (We wouldn’t expect 100% overlap due to inherent measurement errors – especially for the personality disorders – and the different wording of the two sets of criteria.) If there was a high overlap
The association of social phobia to other personality disorders
A review of the studies which examine social phobia and other personality disorders 2, 8, 12, 17, 24, 29, 30, 34, 35, 39 shows some mild association with the DSM schizoid personality disorder cluster, but more strongly to the DSM anxious personality disorder cluster. It is quite possible that some of the association with the schizoid personality cluster (especially in the self-report instruments) may be due to measurement artifact. It can be difficult for personality instruments to distinguish
Treatment and outcome studies for social phobia and APD
Although treatment response is not part of the definition for SP or APD, these studies can give us valuable nosologic information. If the same treatments work for different disorders, or symptoms of one disorder get better as a second disorder is treated, this increases the possibility that these disorders are related or may even be the same disorder.
Psychopharmacologic treatment studies
The second half of table III lists pharmacologic treatment studies for APD and SP that include APD or avoidant personality traits. Although the studies vary in many respects, there is preliminary evidence that benzodiazepines, SSRIs and various forms of MAOIs may be effective for APD or avoidant traits associated with SP. Of specific note are the studies of Fahlen 9, 10 Liebowitz et al. [19] and Reich et al. [29]. All of these studies had reasonably good sample sizes and careful measurements of
Psychological treatments
The first part of table III lists psychological treatment studies of APD or SP associated with APD. These studies use cognitive or behavioral treatments or both. Overall it appears that both APD and SP co-morbid with APD do respond to treatment. The SP without GSP tends to be least disabled, followed by GSP, followed by GSP co-morbid with APD. Although all start at different baselines of morbidity, all seem to respond to treatment. Since some subjects are more disabled their final scores don’t
Course of social phobia
There are relatively few studies of the course of SP. Two prospective studies review this literature and also give us valuable information. Reich et al. 27, 28 find very few longitudinal studies of SP, with many of them using a weaker retrospective design and most studies not extending as long as a year. The general findings are that SP tends to be chronic with low remission rates. Using the more rigorous methodology and prospective methods gives rates of complete remission at 65 weeks as about
Discussion
One of the first questions that must be asked is whether SP and APD are the same or different disorders. The empirical evidence we have now on co-morbidity and treatment comes down strongly on the side of there being just one disorder with different subtypes. There is clearly no distinct symptomatic delineation between SP, GSP and APD. Treatment studies do not distinguish between subtypes. The original “performance in situations versus problems in relating to people” theoretical distinction has
Conclusions
The available empirical evidence supports the conclusion that SP and APD are not separate disorders. It is suggested that the disorder be placed on Axis I with an Axis II cross-listing of “A chronic Axis I disorder with significant personality disorder features, secondary to social phobia, with dysfunctional avoidant personality traits.” Since APD would essentially be eliminated, dependent personality disorder (one of its nearest neighbors) should have a subtype of “with avoidant features” as
References (42)
- et al.
Avoidant personality disorder, interpersonal problems as predictors of treatment response
Behav Ther
(1993) - et al.
Social phobia subtype and avoidant personality disorder: effect on severity of social phobia, impairment, and outcome of cognitive behavioral treatment
Behav Ther
(1995) - et al.
A case of social phobia with avoidant personality disorder treated with MAOI
Compr Psychiatry
(1986) - et al.
Psychopharmacological treatment of avoidant personality disorder
Compr Psychiatry
(1989) - et al.
Secondary social phobia in patients with major depression
Psychiatry Res
(1992) - et al.
Social phobia with and without avoidant personality disorder: preliminary behavior therapy outcome findings
J Anxiety Disord
(1995) Relationship between DSM-III avoidant and dependent personality disorders
Psychiatry Res
(1990)- et al.
A prospective follow-along study of the course of social phobia
Psychiatry Res
(1994) - et al.
Intensive behavioral group treatment of avoidant personality disorder
Behav Ther
(1990) - et al.
Prevalence of personality disorders in patients with major depression and dysthymia
Psychiatry Res
(1992)