Journal of Affective Disorders 54 (1999) 277–282

Research report

Avoidant personality in social phobia and panic–agoraphobicdisorder: a comparison

a , a a a a*Giulio Perugi , Stefano Nassini , Cristina Socci , Michele Lenzi , Cristina Toni ,a bElisa Simonini , Hagop S. Akiskal

aInstitute of Psychiatry, University of Pisa, Via Roma 67, 56100 Pisa, ItalybInternational Mood Disorder Center, Department of Psychiatry, University of California at San Diego, La Jolla, USA

Received 18 November 1997; received in revised form 10 February 1999; accepted 22 March 1999

Abstract

Background: Avoidant personality disorder (APD) is generally believed to be related to social phobia (SP), especially togeneralized subtype. However, it has also been reported to be prevalent in panic disorder–agoraphobia (PDA). In the presentinvestigation, we wished to explore whether APD in each of these disorders has discriminatory features. Method: We studied71 SP and 119 PDA patients with state-of-the-art clinical instruments based on DSM-III-R. Results: The pattern of socialavoidance in SP was more pervasive: it was characterized by a higher level of interpersonal sensitivity and greater severity,associated with psychopathology as well as a higher rate of Axis I comorbidity. By contrast, avoidance of non-routinesituations characterized APD occurring in the setting of PDA. Limitations: Differences in inclusion criteria and comorbidityrates, as well as overlap between different operational disorders, may have influenced our findings. Conclusion: ADP isoperationally broad, and ‘avoidant’ as a specifier of a personality type is insufficiently precise. ADP captures avoidant traits— which appear secondary to a core dimension such as interpersonal sensitivity — but is basically a heterogeneouscondition influenced by the nature of comorbid Axis I disorders. 1999 Elsevier Science B.V. All rights reserved.

Keywords: Avoidant personality; Interpersonal sensitivity; Social phobia; Panic disorder; Agoraphobia

1. Introduction 1989; Klass et al., 1989; Schneier et al., 1991;Herbert et al., 1992, Holt et al., 1992; Turner et al.,

Avoidant personality disorder (APD) has been 1992) as well as with panic disorder irrespective ofreported to frequently co-occur with social phobia the presence of agoraphobia (PDA) (Renneberg et(SP) (Alnaes and Torgersen, 1988; Brooks et al., al., 1992; Jansen et al., 1994; Noyes et al., 1995).

The overlap is particularly extensive between APDand SP of the generalized subtype (Schneier et al.,*Corresponding author. Tel.: 1 39-50-592-479; fax: 1 39-50-

21-581. 1991; Turner et al., 1991; Herbert et al., 1992, Holt

0165-0327/99/$ – see front matter 1999 Elsevier Science B.V. All rights reserved.PI I : S0165-0327( 99 )00062-2

278 G. Perugi et al. / Journal of Affective Disorders 54 (1999) 277 –282

et al., 1992); although APD is more severe than SP R-Upjohn version (SCID-UP, Spitzer and Williams,in terms of social anxiety and social functioning 1988). With the exception of simple phobia and(Widiger, 1992), few qualitative differences separate generalized anxiety disorder, subjects with currentthese two conditions (Herbert et al., 1992; Holt et al., (past 6 months) comorbidity with mood or other1992; Turner et al., 1992). Moreover, both pharma- anxiety disorders were not included. However, life-cological (Deltito and Stam, 1989; Liebowitz et al., time comorbidity with mood or other anxiety dis-1991) and behavioural (Brown et al., 1995) treat- orders was not excluded. Finally we excluded sub-ments have been shown equally effective in patients jects who had a history of psychosis, organic mentalwith SP with and without APD. These findings disorder or mental retardation, uncontrolled physicalsuggest that the two conditions might represent disease or abnormal laboratory values.‘different points on a continuum of severity’ (Her- Patients were assessed for DSM-III-R APD usingbert et al., 1992, p. 338) and some authors (Schneier the APD section of the Structured Clinical Interviewet al., 1991; Turner et al., 1992) have questioned for DSM-III-R Personality Disorder (SCID-II,whether SP and APD are separate entities or the Spitzer et al., 1990). A specially constructed semi-same condition described from different perspec- structured interview was also administered to investi-tives. It is unclear whether the high overlap between gate demographic features and longitudinal aspectsAPD and SP would persist if recruitment for these of the anxiety disorder (Perugi et al., 1995; Benedettitwo disorders were conducted independently. et al., 1997). The interview lasted | 1.5 h and is

The association between SP and APD could be instituted by residents with extensive experience infurther clarified by comparing APD with patients the diagnosis and treatment of anxiety disorders.selected on the basis of another anxiety disorder. Patients’ self-evaluation was recorded on the Hop-Jansen et al. (1995), comparing 32 patients with SP kins Symptoms Checklist (HSCL-90) (Derogatis etand 85 patients with PDA, found similar rates of al., 1973). The reliability of our methods has beenAPD (31.3 vs. 23.5, respectively); SP patients ap- documented elsewhere (Perugi et al., 1998a).peared to be more disturbed on Axis II, and APD Comparison between groups on continuous vari-criterion 6 (fears being embarrassed) discriminated ables were analysed by Student’s t-test, and categori-

2SP the most from PDA on the item level. A major cal variables by x analysis. Because of multiplelimitation here was the fact that Axis I diagnoses comparisons only values of P , 0.01 can be consid-were not based on structured interview. ered significant.

The aim of our study, which attempts to redressthe foregoing methodological shortcoming, wastwofold: (1) to evaluate the distribution of APD 3. Resultsaccording to DSM-III-R (APA, 1987) criteria in twogroups of patients with PD and SP; (2) to clarify the Patients with PDA and SP differed in terms ofquestion of the specificity of the co-presence of mean age (respectively 36.1611.7 vs. 30.669.4, t 5

APD, by means of a direct comparison of the clinical 3.3, df 5 179, P 5 0.001), age at onset (28.7610.6features and the personality profile in the two groups vs. 14.366.5, t 5 10.1, df 5 179, P 5 0.0001) andof patients. gender distribution (males 5 39, 32.8% vs. 43,

260.6%; x 5 14.0, df 5 1, P 5 0.0002). As expected,the rate of patients who received the co-diagnosis of

2. Patients and methods APD was higher in SP (n 5 50, 70.4%) compared2with PDA (n 5 45, 37.8%) (x 5 18.9, df 5 1, P 5

We selected 71 patients with SP and 119 with PD, 0.0001).consecutively admitted to a long term treatment Comparisons between demographic features andprogram for anxiety disorders located with the comorbidity in APD patients with SP and PDAInstitute of Psychiatry of the University of Pisa. showed several significant differences (Table 1).

Information on Axis I diagnoses was obtained by Social phobic APD were younger, with earlier age atusing the Structured Clinical Interview for DSM-III- onset, more often males, single and had a higher

G. Perugi et al. / Journal of Affective Disorders 54 (1999) 277 –282 279

Table 1Demographic characteristics and lifetime comorbidity in 50 SP patients and 45 PDA patients with APD

2SP with APD PDA with APD t or x (df) P(n 5 50) (n 5 45)

Age, mean (S.D.) 29.3 (8.9) 36.7 (10.8) 3.67 (93) 0.0004Age at onset, mean (S.D.) 12.7 (5.2) 29.8 (10.1) 10.34 (93) 0.0001Gender: Male, n (%) 31 (62.0) 12 (26.7) 11.93 (1) 0.0006Marital status, n (%)

Single 40 (80.0) 10 (22.2)Married 8 (16.0) 33 (73.3)Separated or divorced 2 (4.0) 2 (4.4) 33.07 (2) 0.0001

Education, n (%)University graduates 6 (12.0) 2 (4.6)High school 31 (62.0) 12 (27.9)Junior high school 13 (26.0) 17 (39.5)Less than 7 years 0 (0.0) 12 (27.9) 22.53 (3) 0.0001

Lifetime comorbidity, n (%)Major depression 27 (54.0) 15 (33.3) 4.10 (1) nsBipolar disorder II 14 (28.0) 2 (4.4) 9.38 (1) 0.002Obsessive compulsive disorder 20 (40.0) 4 (8.9) 12.14 (1) 0.0005Generalized anxiety disorder 8 (16.0) 12 (26.7) 1.62 (1) nsPanic disorder–agoraphobia 22 (44.0)Social phobia 7 (15.6)Simple phobia 13 (26.0) 9 (20.0) 0.48 (1) nsUse of alcohol and/or sedatives 14 (28.0) 5 (11.1) 4.22 (1) ns

Separation anxiety 21 (42.0) 20 (44.4) 0.06 (1) ns

education level than PDA patients. SP patients by scores compared to PDA in all items, except ‘feelingcontrast, showed a higher rate of comorbid Bipolar II uncomfortable about eating or drinking in public’.and obsessive compulsive disorder. While 44% of SPshowed lifetime comorbidity with PDA, only 15.6%of PDA patients reported lifetime SP. 4. Discussion

The comparison of the DSM-III-R diagnosticcriteria for APD (Table 2) showed that ‘sensitivity to Several limitations of this study need to becriticism or disapproval’, ‘avoidance of activities that recognized. Different inclusion criteria may haveinvolve significant interpersonal contact’, and ‘reti- influenced our comparison: social phobics werecence in social situations because of a fear of saying recruited to have moderately severe symptoms, andsomething inappropriate’, were significantly more panic subjects were required to have had at least onefrequent in the SP group, while avoidance of ‘unusu- panic attack per week in the last month. A relatedal situations’ appeared to be more represented in factor has to do with differences in lifetime comor-PDA patients. bidity between the PDA and SP. A final constraint

Concerning the self-evaluation scores on the generic to all diagnostic research based on post-HSCL-90 (Table 3), the subjects with APD with SP DSM-III-R criteria derives from the unavoidablescored significantly higher values in SCL-90 factors, overlap between different operational disorders. Theexcept for ‘Anxiety’, ‘Phobic Anxiety’, ‘Somatisa- present study represents an attempt to clarify —tion’ and ‘Psychoticism’ that obtained similar mean within these limitations — the relationship betweenscores in the two groups. The greater severity of anxiety and anxious personality disorders.APD with SP is confirmed by the scores of the items Our findings confirm that APD is not restricted toon the ‘Interpersonal Sensitivity’ factor. These pa- SP, but is common among patients with PDA. Hightients obtained statistically significant higher mean frequencies of APD in PDA have been reported by

280 G. Perugi et al. / Journal of Affective Disorders 54 (1999) 277 –282

Table 2DSM-III-R diagnostic criteria for avoidant personality disorder in PD and SP patients with APD

PDA with APD SP with APD(n 5 45) (n 5 50)

2n (%) n (%) x (df 5 1) P

Is easily hurt by criticismor disapproval. 37 (82.2) 50 (100.0) 9.71 0.002Has no close friends orconfidant (or only one)other than first-degree relatives. 29 (64.4) 37 (74.0) 1.02 nsIs unwilling to get involvedwith people unless certainof being liked. 31 (68.9) 41 (82.0) 2.22 nsAvoids social or occupationalactivities that involve significantinterpersonal contact. 24 (53.3) 46 (92.0) 18.26 0.0001Is reticent in social situationsbecause of a fear of sayingsomething inappropriate orfoolish, or of being unable toanswer a question. 32 (71.1) 48 (96.0) 11.03 0.0009Fears being embarrassed byblushing, crying, or showingsigns of anxiety in front ofother people. 35 (77.7) 41 (82.0) 0.26 nsExaggerates the potentialdifficulties, physical dangers,or risks involved in doingsomething ordinary but outsidehis or her usual routine. 33 (73.3) 21 (42.0) 9.48 0.002

Reich et al. (1987) and Alnaes and Torgersen diagnosis of SP. This problem may be partially(1988), and a comparable frequency of APD in PDA clarified by studies on clinical populations selectedpatients (32%) was found by Renneberg et al. on the basis of a principal diagnosis of APD with(1992). different sets of criteria.

The main result of our study concerns the different The association between SP and APD does notdistribution of DSM-III-R criteria for APD diagnosis necessarily mean that APD is always a severe formin the two groups. The pattern of avoidance with SP of SP. APD is also diagnosed, albeit less frequently,comorbidity is more pervasive and generalized, in other anxiety disorders as well as in such con-depending on higher levels of interpersonal sensitivi- ditions as Body Dysmorphic Disorder (Veale et al.,ty and behavioural inhibition, probably as a result of 1996), Atypical Depression (Perugi et al., 1998b),greater severity of the symptomatology and of a Eating Disorders (Oldham et al., 1995). SP mayhigher rate of comorbidity with Axis I disorders. covary with APD, however, although the definitionsAvoidance of unusual situations appears to be corre- of the two disorders are quite similar, the onelated with the presence of APD in PDA patients. emphasizes phobic symptoms and the other avoidant

Criteria for APD have been partially modified in traits and their interpersonal consequences. APD ‘isDSM IV (APA, 1994). The last criterion has been essentially a problem of relating to persons; socialmodified and refocused on ‘embarrassing’ new ac- phobia is largely a problem of performing situations’tivities. In our opinion the DSM IV criteria seem to (Millon, 1991). For the present, there does not seemreflect a definition of APD closer to the profile of to be a simple solution to this conceptual diagnostictraits presented by patients with primary Axis I problem. Here, as elsewhere, the distinction between

G. Perugi et al. / Journal of Affective Disorders 54 (1999) 277 –282 281

Table 3HSCL-90 factors and items of interpersonal sensitivity in PD and SP patients with APD

PDA with APD SP with APD(n 5 45) (n 5 50)Mean (S.D.) Mean (S.D.) t P

HSCL-90 factorsInterpersonal sensitivity 1.25 (0.81) 2.37 (0.62) 2 7.60 0.0001Anxiety 1.58 (0.80) 1.84 (0.76) 2 1.58 nsObsession–compulsion 1.39 (0.70) 1.97 (0.75) 2 3.81 0.0002Depression 1.63 (0.77) 2.10 (0.73) 2 4.14 0.0001Paranoidism 1.06 (0.86) 1.64 (0.76) 2 3.51 0.0007Phobic anxiety 1.24 (0.84) 1.25 (0.70) 2 0.01 nsSomatization 1.36 (0.81) 1.02 (0.70) 2.19 nsHostility–anger 0.84 (0.75) 1.24 (0.71) 2 2.64 0.01Psychoticism 0.76 (0.71) 1.16 (0.67) 2 2.00 ns

Items of interpersonal sensitivityFeeling critical of others 0.85 (1.04) 1.62 (1.28) 2 3.22 0.002Feeling shy or uneasywith the opposite sex 0.80 (1.04) 2.42 (1.16) 2 7.14 0.0001Your feelings beingeasily hurt 1.58 (1.27) 2.70 (1.01) 2 4.78 0.0001Feeling others do not understandyou or are unsympathetic 1.13 (1.14) 2.30 (1.11) 2 5.05 0.0001Feeling that people areunfriendly or dislike you 1.18 (1.19) 2.06 (1.33) 2 3.39 0.001Feeling inferior to others 1.62 (1.39) 2.86 (1.11) 2 4.83 0.0001Feeling uneasy when people arewatching or talking about you 1.40 (1.27) 2.86 (0.97) 2 6.34 0.0001Feeling very self-consciouswith others 1.51 (1.24) 2.96 (0.88) 2 6.63 0.0001Feeling uncomfortable abouteating or drinking in public 1.22 (1.28) 1.54 (1.34) 2 1.18 ns

Alnaes, R., Torgersen, S., 1988. The relationship between DSMwhat is labelled personality (Axis II) and psychiatricIII symptom disorders (Axis I) and personality disorders (Axissyndrome (Axis I) remains blurred (Akiskal, 1984;II) in an outpatient population. Acta Psychiatr. Scand. 78,

Widiger, 1992). 485–492.In our view, ADP may be considered a heteroge- American Psychiatric Association, 1987. Diagnostic and Statistical

Manual of Mental Disorders, 3rd ed., Revised. APA, Washing-neous condition whose boundaries are probablyton, DC.influenced by the presence of an Axis I comorbid

American Psychiatric Association, 1994. Diagnostic and Statisticaldisorder. As a corollary, we submit that the core Manual of Mental Disorders, 4th ed. APA, Washington, DC.defining trait in APD might be interpersonal sen- Benedetti, A., Perugi, G., Toni, C., Simonetti, B., Mata, B.,sitivity, whereas in SP it is avoidance of social Cassano, G.B., 1997. Hypochondriasis and illness phobia in

panic–agoraphobic patients. Compr. Psychiatry 38, 124–131.situations, and in PDA avoidance of non-routineBrown, E.J., Heimberg, R.G., Juster, H.R., 1995. Social phobiasituations. These considerations further suggest that

subtype and avoidant personality disorder: effect on severity of‘avoidant’ as a qualifier for a personality type is social phobia, impairment, and outcome of cognitive be-insufficiently precise. havioral treatment. Behav. Ther. 26, 467–486.

Brooks, R.B., Baltazar, P.L., Munjack, D.J., 1989. Co-occurrenceof personality disorders with panic disorder, social phobia, andgeneralized anxiety disorder: a review of the literature. J.

References Anxiety Disord. 3, 259–285.Deltito, J., Stam, M., 1989. Psychopharmacological treatment of

Akiskal, H.S., 1984. Characterologic manifestations of affective avoidant personality disorder. Compr. Psychiatry 30, 498–504.disorders. Integrative Psychiatry 2, 83–88. Derogatis, L.R., Lipman, R.S., Rickels, K., 1973. The Hopkins

282 G. Perugi et al. / Journal of Affective Disorders 54 (1999) 277 –282

Symptom Checklist (HSCL): a measure of primary symptom Perugi, G., Akiskal, H.S., Lattanzi, L., Cecconi, D., Mastrocinque,dimensions in psychological measurement. In: Pichot, P. (Ed.), C., Patronelli, A., Vignoli, S., Bemi, E., 1998b. The highModern Problems in Pharmacopsychiatry, Karger, Basel, Swit- prevalence of ‘soft’ bipolar II features in atypical depression.zerland. Compr. Psychiatry 39, 63–71.

Herbert, J.D., Hope, D.A., Bellack, A.S., 1992. Validity of the Perugi, G., Nassini, S., Lenzi, M., Simonini, E., Cassano, G.B.,distinction between generalized social phobia and avoidant McNair, D.M., 1995. Treatment of social phobia with fluox-personality disorder. J. Abnorm. Psychol. 101, 332–339. etine. Anxiety 1, 282–286.

Holt, C.S., Heimberg, R.G., Hope, D.A., 1992. Situational do- Reich, J.H., Noyes, R., Troughton, E., 1987. Dependent personali-mains of social phobia. J. Anxiety Disord. 6, 63–77. ty disorder associated with phobic avoidance in patients with

Jansen, M.A., Arntz, A., Merckelbach, H., Mersch, P.P., 1994. panic disorder. Am. J. Psychiatry 144, 323–326.Personality disorders and features in social phobia and panic Renneberg, B., Chambless, D.L., Gracely, E.J., 1992. Prevalencedisorder. J. Abnorm. Psychol. 103, 391–395. of SCID-diagnosed personality disorders in agoraphobic out-

Klass, E.T., DiNardo, P.A., Barlow, D.H., 1989. DSM-III-R patients. J. Anxiety Disord. 6, 111–118.personality diagnosis in anxiety disorder patients. Compr. Schneier, F.R., Spitzer, R.L., Gibbon, M., Fyer, A.J., Liebowitz,Psychiatry 30, 251–258. M.R., 1991. The relationship of social phobia subtypes and

Liebowitz, M.R., Schneier, F.R., Hollander, E., Welkowitz, L.A., avoidant personality disorder. Compr. Psychiatry 32, 496–502.Saoud, J.B., Feerick, J., Campeas, R., Fallon, B.A., Street, L., Spitzer, R.L., Williams, J.B.W., 1988. Structured Clinical Inter-Gitow, A., 1991. Treatment of social phobia with drugs other view For DSM-III-R, Upjohn Version Revised (SCID-UP-R),than benzodiazepines. J. Clin. Psychiatry 52 (11 Suppl.), 10– New York State Psychiatric Institute, New York.15. Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1990.

Millon, T., 1991. Avoidant personality disorder: a brief review of Structured Clinical Interview For DSM-III-R Personality Dis-issues and data. J. Pers. Disord. 5 (4), 353–362. orders (SCID-II), American Psychiatric Press, Washington DC.

Noyes, R., Woodman, C.L., Holt, C.S., Reich, J.H., Zimmerman, Turner, S.M., Beidel, D.C., Borden, J.W., Stanley, M.A., Jacob,M.B., 1995. Avoidant personality traits distinguish social R.G., 1991. Social phobia: Axis I and II correlates. J. Abnorm.phobic and panic disorder subjects. J. Nerv. Ment. Dis. 183, Psychol. 100, 102–106.145–153. Turner, S.M., Beidel, D.C., Townsley, R.M., 1992. Social phobia:

Oldham, J.M., Skodol, A.E., Kellman, H.D., Hyler, S.E., Doidge, a comparison of specific and generalized subtypes and avoidantN., Rosnick, L., Gallaher, P.E., 1995. Comorbidity of Axis I personality disorder. J. Abnorm. Psychol. 101, 326–331.and Axis II disorders. Am. J. Psychiatry 152, 571–578. Veale, D., Boocock, A., Gournay, K., Dryden, W., Shah, F.,

Perugi, G., Toni, C., Benedetti, A., Simonetti, B., Simoncini, M., Willson, R., Walburn, J., 1996. Body dysmorphic disorder: aTorti, C., Musetti, L., Akiskal, H.S., 1998a. Delineating a survey of fifty cases. Br. J. Psychiatry 169, 196–201.putative phobic-anxious temperament in 126 panic– Widiger, T.A., 1992. Generalized social phobia versus avoidantagoraphobic patients: toward a rapprochement of European and personality disorder: a commentary on three studies. J. Ab-US views. J. Affect. Disord. 47, 11–23. norm. Psychol. 101, 340–343.