~- Taylor&Francis • healthsciences

a Turkish OCD Patterns of Axis-11 comorbidity in sample 2. 1-t»fJ

of C. , s J 5, 1.. f r.

AYSE AYCICEGI 1, WAYNE M. DINN 2

AND CATHERINE l. HARRIS2

Department of Psychology, Istanbul University, Istanbul, Turkey; 2Department of Psychology, Bostan University, Bostan, MA, USA

Correspondence Address Wayne M. Dinn, 42 Washington Terrace, Whitman, MA 02382, USA Tel: +(1)-781-447-6058 E-mail: dinn@bu.edu

Received 26 July 2003; accepted for publication 19 December 2003

INTRODUCTION

OBJECTIVE Obsessive-compulsive disorder (OCD) patients frequently present with Axis-II disorders, particularly Cluster C (anxiety spectrum) and Cluster A (schizophrenic spectrum) personalitj disorders. The present study examined pattems of Axis-II comorbidity in a Turkish OCD sample. In addition, we explored the impact of personality disorder symptoms on OCD-symptom severity and symptom profile.

METHOD: Structured psychiatric interviews and self-report measures of OCD symptoms and Axis-II disorders were administered to patients with OCD and control subjects.

RESULTS : Patients with OCD obtained significantly higher scores on measures of Cluster A and Cluster C personality disorders. Patients with OCD also achieved significantly higher scores on the BPD Subscale; however, they did not obtain significantly higher scores on other Cluster B subscales. Group differences on measures of Cluster A and C disorders were marked.

CONCLUSIONS: Findings are consistent with prior work demonstrating an increased incidence of Axis-II disorders among patients with OCD. (Int J Psych Clin Pract 2004; 8: 85-89)

Keywotds Obsessive-con,:pulsive disorder (OCp) Personality

Axis-II disorders

O bsessive-compulsive disorder (OCD) patients fre-quently present with Axis-Il disorders, particularly

Cluster C (anxiety spectrum) and Cluster A (schizophrenic spectrum) personality disorders_'l-ö Matsunaga et al 4

reported that the majority (53%) ofa group of OCD patients fulfilled criteria for a personality disorder, with Cluster C diagnoses being the most common. Bejerot et al 2 also found that a significant number of OCD patients present with a

meet criteria for Cluster A disorders. Mavissakalian and colleagues5 found that 16% of an OCD patient group met criteria for schizotypal personality disorder (SPD), while Baer and Minichiello1 reported that 35% of a series of patients with OCD seeking treatment at an OCD clinic presented with comorbid SPD.

Black et al 7 concluded that Axis-ll disorders are "highly prevalent among patients wi th obsessive-compulsive dis-order". However, other investigators have found that OCD patients do not exhibit high rates of Axis-Il disorders. For example, Crino and Andrews8 reported that patients with OCD did not demonstrate a high rate of Axis-II diagnoses relative to a clinical concrol group (i.e., patients with anxiety disorders) . As noted previously, Bejerot et al 2 found that approximately one-third of an OCD patient group fulfilled criteria for OCPD. Samuels et al 9 reported that OCD patients exhibited an increased incidence of OCPD relative to control subjects. Diaferia et al 3 also found that OCD patients demonstrated a higher prevalence of OCPD in comparison to psychiatric controls (i.e., panic disorder and major

~· coexistent personality disorder. They reported that 75% of an OCD sample met criteria for an Axis-Il disorder and that more than one-third fulfilled criteria for obsessive-compul-

- sive personality disorder (OCPD). Rodrigues-Torres and Del Porto6 found that 70% of an OCD patient group (n = 40) met criteria for an Axis-Il disorder, while onl y 15 % of matched controls fulfilled criteria for a personality disorder. Again, Cluster C disorders were significantly more prevalent among OCD patients (particularly avoidant and dependent person-ality disorders). A significant subset of OCD patients also r Ü. . .. L . .._ ,'. ..

'1'c,r.irb~~ No :/V\ ... ..... "Jo

.. No

DO l l O .1080/1'\6 'i Vi004 00064 11

depressive disorder patients). However, other found that OCPD was not strongly associated with OCD; rather, OCD patients were more likely to present with avoidant,

1. d. d 4-610-12 dependent, or schizotypal persona ers. · The relation between OCD and OCPD merits funher study.

WHY EXAMINE PATTERNS OF CoMORBIDITY IN OCD?

Clearly, a significant number of OCD patients present with Axis-11 disorders. Why is it important to examine patterns of Axis-11 comorbidity in OCD7 Clinicians and researchers may benefit from an understanding of the relation between OCD and Axis-11 disorders . The presence ofa personality disorder among patients with OCD is associated with a poor response to treatment. 13 For example, OCD patients presenting with

. l h 14-16 I schizotypal respond poor y to t erapy. n addition, OCD patients presenting with and without schizo-typal personality features demonstrate distinct neurocogni-tive profiles.17 Patients with primary OCD (without coexistent schizotypal features) displayed performance def-icits on tasks considered sensitive to orbitofrontal dysfunc-tion (e.g., tests of alternation learning and response inhibition). OCD patients presenting with pronounced schizotypal features (OCD/SP) also exhibited performance deficits on orbitofrontal tasks; however, OCD/SP patients also performed poorly on tests of executive functioning and reported a significantly greater number of attentional diffi-culties and executive dysfunction relative to healthy controls and patients with primary OCD (i e., without prominent schizotypal features). 17

PURPOSE OF PRESENT STUDY

The present study examined patterns of Axis-II comorbidity in a Turkish OCD sample. in addition, we explored the impact of personality disorder symptoms on OCD-symptom severity and symptom profile . We conducted structured psychiatric interviews and administered self-report measures of OCD symptoms and Axis-11 disorders to patients with OCD and control subjects. To our knowledge, this is the first study exploring patterns of Axis-11 comorbidity among patients with OCD in Turkey.

METHOD

Twenty-one OCD patients (14 female and seven male) were recruited from a major psychiatric hospital (Bakirkoy Hospital) and from a private psychiatric clinic in lstanbul, Turkey. They ranged in age from 17 to 58 years (M = 35.3, SD = 1 O. 7) and their mean educational level was 9 .4 years (SD = 4.6) Control subjects (n = 21) (14 female and seven male) were recruited from the general population . They ranged in age from 17 to 51 years (M = 34.4, SD = 9 5) and their mean educationallevel was 10.0 years (SD = 4 2). OCD pa tients were attending outpatient clinics and diagnoses were

made by clinicians specializing in the treatment of anxiety disorders. in addition, we used the MiNi OCD Diagnostic Module to confirm diagnoses. The Mini lnternational Neuropsychiatric lnterview (MINI) is a structured psychiatric interview based on DSM-IVand ICD-10 diagnostic criteria 18 The MiNi provides d1agnostic algorithms and is a psycho-metrically sound instrument which compares favourably to older, established measures such as the SCID-P 18 All OCD patients exceeded the syrnptom threshold on the MiNi OCD Diagnostic Module. A Turkish translation of the MiNi was used. 19

Panicipants also completed the Obsessive-Compulsive lnventory (OCl). 20 OCD patients obtained high scores on the OCl (see Table 1) The OCI is a self-report measure which yields a total score and subscale scores reflecting: washing rituals; checking behaviour; obsessive doubting; obsessional ideation; compulsive hoarding; ordering; and mental neu-tralizing. The lead author translated the OCl into Turkish. A second bilingual clinical psychologist independently trans-lated the Turkish version into English and the second author (WD) compared this translation to the original instrument. Discrepancies in meaning with the original English version were noted and the Turkish translation was adjusted.

A Tur.kish version of the SCID-11 Patient Questionnaire (SCID-11 PQ) was administered to patients and control subjects. 21 The SCID-11 PQ is a self-repon measure based on DSM-lll-R diagnostic criteria for personality disorders. Of course, we are not equating performance on a self-repon measure with a diagnosis established by a clinician experi-enced in the differential diagnosis of Axis-ll disorders. Nevertheless, we will be able to determine if OCD patients demonstrate a greater number of personality disorder (PD) symptoms relative to control subjects and we will examine the impact of PD symptoms on OCD-symptom severity and symptom profile. lnformed consent was obtained from ali participants. Since multiple comparisons were planned, we employed the Bonferroni procedure to reduce the risk of Type-1 error. A conservative Bonferroni-corrected a level was used (P < O 003)

Table 1 OCI subscale scores: mean (SD)

]it' '. :~t~~~ii •. ö

~!!t:t :::i:g~ •p < 0 .001.

OCO ,

21 31;3''4 (17..,3)

33:i:(fa: ij . ,,;; .. •• -. ,,. 24.8 (9.8) 34.2 (21.1)

Con tTol

1_7., l\C p ,_6-) .. · -~~9 ~J.p) : ii\~ o)

· ' S._8 (5.3) ~9.4)

13.Q (8.2) 14.0 (8 9)

RESULTS

GROUP DIFFERENCES

As shown in Table 1, patients achieved significantly higher scores on OCI subscales assessing compulsive checking behavior (t(40) = 4.12, P < 0.001), obsessional doubting (t(40) = 6.79, P < 0.001), obsessional ideation (t(40) = 6.34, P < 0.001), ordering (t(40) = 4.17, P < 0.001),

mental neutralizing (t(40) = 5.95, P < 0.001), and compul-sive washing rituals (t(40) = 4.02, P < 0.001). Group differ-ences on the hoarding subscale were not significant (P > 0.35) . Groups did not differ on age or educational !eve! (P values > 0.67).

As shown in Table 2, OCD patients obtained significantly higher scores, relative to control subjects, on scales assessing Avoidant Personality Disorder (Av1'D) (t(40) = 3.42, p < .001), Obsessive-Compulsive Personality Disorder (OCPD) (t(40) = 3.61, p < .001), Dependent Personality Disorder (DPD) (t(40) = 3.12, p < .003), Borderline Person-ality Disorder (BPD) (t(40) = 3.47, p < 001), Schizotypal Personality Disorder (SPD) (t(40) = 5.24, p < .001), and Schizoid Personality Disorder (SzPD) (t(40) = 3.77, p < .001). Patients with OCD obtained higher scores on the Narcissistic Personality Disorder Subscale; however, group differences were not statistically significant (t ( 40) = 2.51, p = .016). In summary, OCD was strongly associated

with Cluster C and Cluster A personality disorders. Group differences on the remaining SCID-II subscales were not significant (ps > .40) (Paranoid Personality Disorder Sub-scale and the remaining Cluster B disorders) .

Table 2 Personality Disorder Subscale Scores: Mean (5D)

OCD Çontrol

N 21 21 Age 35.3 (1,0.7) 34.4 (9.5) Education 9.43 ( 4.6) 10.0 (4 2)

Cluster A Schizotypal personality (SPD}* 3 .42 (1.74) 1:19 (0.87) Schizoid personality (SzPD)• 2.38 (1.35) 1.09 (0 .76) Paranoid personality (PPD) 4.47 (1.93) 4.00 (1.70)

B Histrionic personality (HPD) 5.71 G.84) 5,09 (2 .42) Narcissistic personality (NPD) 4.42 (1.96) 2 .. 80 (2 . .20) Antisocial personality (A.PD) 1.19 (1,.6Q) (Ul) Borderline personality (BPD)* 7.85 (3. 16) 4.52 (3.Ö4)

Cluster C OC personality (OCPD)* 7.71 (2.45) 5.09 (2.23) Avoidant personality (AvPD)* 4.42 (1 .85) 2.47 (1.83) Dependent personality (DPD)* 3.57 (1.59) 2.14 (1.35)

•p < 0.003.

As noted previously, a Turkish version of the SCID-II Patient Questionnaire (SCID-II PQ) was administered to patients and control subjects. The lead author translated the Turkish SCID-II PQ into English The second author compared this translation to DSM-IV diagnostic criteria for Axis-II disorders. We were thus able to determine if participants fulfilled diagnostic criteria for DSM-IV person-ality disorders. A significant number of OCD patients met DSM-IV diagnostic criteria for Cluster C disorders (based on SCID-II Questionnaire scores). More than 80% of the OCD patient group met criteria for .avoidant personality disorder (AvPD) (17/21), while 33.3% of controls fulfilled criteria for AvPD. Similarly, 81 % of the OCD patient group fulfilled criteria for OCPD, while 42 .8% of control subjects met criteria. In addition, 28% of the patient group met criteria for dependent personality disorder (DPD). None of the control subjec ts met criteria for DPD. A-si.gnificant subset of the OCD group also met criteria for Cluster A disorders . Approxi-mately 50% (10/21) of the OCD group fulfilled diagnostic criteria for SPD. None of the control subjects met criteria for SPD. Ten OCD patients (47 6%) fulfilled criteria for SzPD, while only one control subject met criteria for SzPD ( 4. 7%). Six OCD patients met criteria for BPD (28 5%), while two control subjects (9.5%) fulfilled criteria for BPD. Group differences on the remaining SCID-II subscales were not significant.

CORRELATIONAL ANALYSIS

The Pearson product-moment correlation was used to determine the strength of association between OCD-symp-tom subtypes and personality disorder symptoms within the patient group. Correlational analysis revealed that scores on the 10 SCID-II Subscales were not significantly associated \.vith OCD-symptom severity (i.e ., total OCI score) (P values > 0.10). Cluster A scores (PPD, SPD, and SzPD) were not significantly associated with performance on OCI subscales assessing compulsive checking (P values > 0.31), doubting (P values > 0.16), obsessional ideation (P values > 0.61), and mental neutralization (P values > 0.46). Scores

on the Paranoid and Schizotypal Personality Disorder subscales were marginally associated with performance on the Hoarding Subscale, with r = 0.51 , P < 0.02 and r = 0.40, P < 0.071 . Interestingly, analysis revealed an inverse relation between scores on the Paranoid and Schizotypal Personality Disorder subscales and performance on the Washing Subscale, with r = - 0.67, P < .001 and r = - 0.43, P < .054 .

Cluster B scores (APD, BPD, HPD, and NPD) were not significantly associated with OCD symptom severity (to tal OCI score) (P values > 0.12) or performance on OCI subscales assessing compulsive checking (P values > 0.40), doubting (P values > 0.14), mental neutralizing (P values > 0.21), obsessional ideation (P values > 0.2D), ordering

(P values > 0.08) , and washing (P values > 0.30) with the following exception. Performance on the BPD Subscale was inversely related to scores on the ordering and washing

88 A Aycicegi et al

subscales, with r= -0.42, P <0.054 and r= -0.39, P < 0.08.

Cluster C scores (AvPD, DPD, and OCPD) were not significantly associated with OCD-symptom severity (total OCI score) (P values > O 36) or with performance on OCI subscales assessing compulsive checking (P values > 0.15), doubting (P values > 0.33), obsessional ideation (P values > 0.29), ordering (P values > 0.77) , and mental neutraliza-tion (P values > 0.38). Performance on the Hoarding Sub-scale correlated significantly with scores on the Avoidant, Dependent, and Obsessive-Compulsive Personality sub-scales, with r = 0.50, P < 0.021, r = 0.46, P < 0.035, and r = 0.46, P < 0.034, respectively Analysis revealed an in-verse relation between scores on the Dependent and Obsessive-Compulsive Personality Disorder subscales and performance on the Washing Subscale, with r = - 0.38, P < 0.083 and r = - 0.40, P < 0.072.

DISCUSSION

Patients with OCD obtained significantly higher scores on measures of Cluster A and Cluster C personality disorders. Patient and control group differences were striking.

Of course, we are not equating performance on a self-report screening measure with a diagnosis established by a clinician experienced in the differential diagnosis of Axis-II disorders. Moreover, the high rates of avoidant and OC personality disorders among OCD and control participants suggest that these subscales "overdiagnosed" AvPD and OCPD. Nevenheless, group differences on measures of Cluster A and C personality disorders were marked. Findings are consistent with prior work demonstrating an increased incidence of Axis-ll disorders among patients with OCD. Patients with OCD also achieved significantly higher scores on the BPD Subscale; however, they did not obtain significantly higher scores on other Cluster B subscales. It is interesting to note that BPD is characterized by intense and unstable affect and severa l studies reported that a significant number of bipolar disorder present with OCD. The relation between OCD and affective dysregulation merits funher study

Correlational analysis showed that scores on SCID-Il subscales did not correlate with performance on OCI subscales assessing classic OCD symptoms. Indeed, scores on SCID-II subscales assessing Paranoid , Schizotypal, Bor-derline, Dependent, and Obsessive-Compulsive personality disorders were inversely related to performance on the compulsive washing subscale; however, compulsive hoarding was associated with Paranoid, Schizotypal, Avoidant, Depen-dent , and Obsessive-Compulsive Personality Disorder symptoms. Since personality disorder symptoms were not associated with OCD symptom severity, poor treatment response among OCD patients with a coexistenl Axis-Il

disorder may not be associated with an increase in the intensity or frequency of classic OCD symptoms. Rather, resistance to treatment may stem from specific PD symptoms (e.g., schizotypal symptoms)

Our findings must be interpreted with caution. The sample size was relatively small (total N = 42) and we relied on a self-report measure of personality disorder symptoms based on the SCID-Il . Nevertheless, group differences were striking and findings confirm that a significant number of OCD patients present with coexistent personality disorder symptoms. Clinicians may consider screening OCD patients for Axis-Il disorders since the presence of a personality disorder is associated with a poor response to treatment and targeting comorbid Axis-II disorders (e g., using atypical antipsychotics to treat schizotypal symptoms) may enhance the OCD patient's response to standard biobehavioural interventions.

Researchers attempting to identify the pathophysiological correlates of OCD may also benefit from an understanding of the relation between OCD and Axis-ll disorders. Our prior work 17 suggests that OCD can be subdivided into clinical subtypes (based on the presence or absence of schizotypal personality features), and that distincl prefrontal subsystems may be differentially involved in these subtypes That is, core OCD symptoms may be associated with orbitofrontal dysfunction, while executive function deficits are associated with comorbid schizotypal symptoms (possibly reflecting dorsolateral-prefrontal dysfunction).

KEY POINTS

• Obsessive-compulsive disorder (OCD) patients fre-quently present with Axis-II disorders

• The present study examined patterns of Axis-11 comorbidity in a Turkish OCD sample

• Patients with OCD obtained significantly higher scores on measures of Cluster A and Cluster C personality disorders

• Patients with OCD also achieved significantly higher scores on the BPD Subscale; however, they did not obtain significantly higher scores on other Cluster B subscales

• Group differences on measures of Cluster A and C disorders were marked

ACKNOWLEDGEMENTS

The authors are grateful to Dr. Husnu Erkmen for his assistance with patient recruitment

REFERENCES Baer L, Minichiello W (1986) Behavior therapy for obsessive-compulsive disorder. in M Jenike, L Baer, W Minichiello (eds) Obsessive-compulsive disorders: Theory and management (pp. 45- 75). Chicago, iL: Year Book Medical Publishers.

2. Bejerot S, Ekselius L, von Knorring L (1998) Comorbidity between obsessive-compulsive disorder (OCD) and personality disorders. Acta Psychiatr Scand 97: 398-402.

3. Diaferia G, Bianchi !, Bianchi ML et al (1997) Relationship between obsessive-compulsive personality disorder and obsessive-cornpulsive disorder. Compr Psychiatry 38 38-42.

4. Matsunaga H, Kiriike N, Miyara A et al (1998) Personality disorders in patients with obsessive-compulsive disorder in Japan . Psychiatr Scand 98: 128-34.

5. Mavissakalian M, Hamann MS, Jones B (1990) Correlates of DSM-Ill personality disorder in obsessive-compulsive disorder. Compr Psychiatry 31: 481-9.

6. Rodrigues-Torres A, Del Porto JA (1995) Cornorbidity of obsessive-compulsive disorder and personality disorders: A Brazilian controlled study. 28 : 322-9.

7. Black DW, Noyes R Jr, Pfohl Bet al (1993) Personality disorder in obsessive-compulsive volunteers, well comparison subjects, and their first-degree relatives. Am J Psychiatry 150: 1226-32.

8. Crino RD, Andrews G (1996) Personality disorder in obsessive-compulsive disorder: a controlled study J Psychiatr Res 30: 29-38.

9. Samuels J, Nestadt G, Bienvenu OJ et al (2000) Personality disorders and normal personality dimensions in obsessive-compulsive disorder. Br J Psychiatry 177: 457-62.

10. Baer L, Jenike MA, Ricciardi JN et al (1990) Standardized assessmem of personality disorders in obsessive-compulsive disorder. Arch Gen Psychiatry 47 : 826-30.

Pattems of Axis-ll comorbidity in Turkey 89

11. Maina G, Bellino S, Bogetto F. Ravizza L (1993) Personality disorders in obsessive-cornpulsive patients: A study repon. Eu.r J Psychiatry 7: 155-63 .

12 . Thomsen PH, Mikkelsen HU (1993) Development of personality disorders in children and adolescerits with obsessive-cornpulsive disorder: A 6- to 22-year follow-up study. Acta Psychiatr Scand 87: 456-62.

13. AuBuchon PG, Malatesta VJ (1994) Obsessive compulsive patients with comorbid personality disorder: associated problems and response to a comprehensive behavior therapy. J Clin Psychiatr 55: 448-53.

14. Jenike MA, Baer L, Carey RJ (1986) Coexistent obsessive-compulsive disorder and schizotypal personality disorder: A poor prognostic indicator. Arch Gen Psychiatry 43 : 296.

15. Minichiello WE, Baer L, Jenike MA (1987) Schizotypal personality disorder: A poor prognostic indicator for behavior therapy in rhe treatment of obsessive-compulsive disorder. J Anxiety Disord l : 273-6.

16. Ravizza L, Barzega G, Bellino S et al (1995) Predictors of drug treatment response in obsessive-compulsive disorder J Clin Psychiatry 56 368-73.

17. Harris CL, Dinn WM (2003) Subtyping obsessive-compulsive disorder: Neuropsychological correlates. Behav Neurol 14: 75-87.

18. Sheehan D, Janavs J, Baker R et al (1998) Mini International Neuropsychiatric lnterviewJ Clin Psychiatr 59 (Suppl 20) : 34-5 7.

19. Omek T, Keskiner A (1998) in Sheehan ·o, Janavs J, Baker R, et al Mini lntemational Neuropsychiatric lnterview. J Clin Psychiatr 59 (Suppl 20) : 34-57.

20. Foa EB, Kozak MJ, Salkovskis PM et al (1998) The validation ofa new obsessive-compulsive disorder scale: The Obsessive-Compulsive lnventory. Psychol Assess 10: 206-14.

21. Sorias S, Saygili R, Elbi H et al (1990) SCID II Hasta Formu. Ege Üniversitesi, Bornova.