قومیت و چهار اختلال شخصیتی

Abstract The current study examined the relationship between ethnicity and DSM-IV personality disorders. The distribution of four personality disorders—borderline (BPD), schizotypal (STPD), avoidant (AVPD), and obsessive-compulsive (OCPD)—along with their criteria sets, were compared across three ethnic groups (Caucasians, African Americans, and Hispanics) using both a clinician-administered diagnostic interview and a self-report instrument. Participants were 554 patients drawn from the Collaborative Longitudinal Personality Disorders Study (CLPS) who comprised these three ethnic groups and met personality disorder criteria based on reliably administered semistructured interviews. Chi-square analyses revealed disproportionately higher rates of BPD in Hispanic than in Caucasian and African American participants and higher rates of STPD among African Americans when compared to Caucasians. Self-report data reflected similar patterns. The findings suggest that in treatment-seeking samples, Caucasians, Hispanics, and African Americans may present with different patterns of personality pathology. The factors contributing to these differences warrant further investigation.

Results Primary aim: distribution of ethnicity across four personality disorder diagnoses using a semistructured interview (DIPD) Four separate 2 × 3 chi-square tests were conducted. The omnibus chi-square analysis examining the distribution of BPD across ethnic groups was significant [χ2 (2, n = 554) = 12.64, P = .002] (Table 1). There was also a significant difference across groups for STPD [χ2 (2, n = 554) = 7.06, P = .03]. No significant differences were found for OCPD [χ2 (2, n = 554) = .89, P = .64] or AVPD [χ2 (2, n = 554) = .27, P = .87]. Table 1. Chi Square Tests: Distribution of BPD, STPD, OCPD, and AVPD Across Three Ethnic Groups Using DIPD Structured Interview and SNAP Self-Report Caucasian (n = 433) n (%) African American (n = 65) n (%) Hispanic (n = 56) n (%) χ2 df P BPD DIPD 171 (40)a 26 (40)a 36 (64)b 12.64 2 .002 SNAP 121 (28)b 23 (35)b 24 (43)a 5.97 2 .05 STPD DIPD 70 (16)b 18 (28)a 6 (11)b 7.06 2 .03 SNAP 106 (25)b 28 (43)a 17 (30)ab 9.99 2 .007 OCPD DIPD 197 (45) 29 (45) 29 (52) .89 2 .64 SNAP 219 (51) 36 (55) 34 (61) 2.22 2 .33 AVPD DIPD 248 (57) 35 (54) 32 (57) .27 2 .87 SNAP 247 (57) 45 (69) 31 (55) 3.56 2 .17 NOTE. Proportions that share the same superscript are not significantly different from each other. Total percentage exceeds 100 because of multiple personality disorder diagnoses for each subject. N = 433 for Caucasians who completed the DIPD and N = 431 for Caucasians who completed the SNAP. Table options Post hoc 2 × 2 chi-square analyses were performed contrasting the distribution of BPD and STPD across pairs of ethnic groups. Hispanics had disproportionately more BPD diagnoses than Caucasians [χ2 (1, n = 489) = 11.49, P < .001] and African Americans [χ2 (1, n = 121) = 6.16, P < .01] (Table 1). For STPD, African Americans had disproportionately more diagnoses than both Caucasians [χ2 (1, n = 498) = 4.40, P < .05] and Hispanics [χ2 (1, n = 121) = 4.45, P < .05]. Convergent support: distribution of ethnicity across four personality disorder diagnoses using patient self-report (SNAP) Chi-square tests were conducted to verify that similar diagnostic patterns were present using patient’s self-report (the SNAP). Findings were in the same direction as the DIPD results for both BPD and STPD (Table 1). Paralleling DIPD findings, Hispanics had disproportionately more BPD diagnoses than Caucasians [χ2 (1, n = 487) = 4.50, P < .05] and African Americans had disproportionately more STPD diagnoses than Caucasians [χ2 (1, n = 487) = 8.87, P < .005]. Also, consistent with the DIPD interview, no significant differences were found for OCPD [χ2 (2, N = 552) = 2.22, P = .33] or AVPD [χ2 (2, N = 552) = 3.56, P = .17]. Although DIPD and SNAP findings were consistent, it is important to point out that these assessments were not entirely independent given our selection procedures. Secondary analyses Distribution of ethnicity across BPD and STPD personality disorder criteria The presence or absence of a specific criterion was assessed across the three ethnic groups (i.e., Caucasians, African Americans, and Hispanics). As shown in Table 2, significant chi square tests were found for DIPD-BPD criterion 1 (intense anger), DIPD-BPD criterion 2 (affective instability), and DIPD-BPD criterion 9 (unstable relationships). Follow-up post hoc 2 × 2 chi-square analyses (alpha = .05) revealed that Hispanics had disproportionately higher endorsement rates than Caucasians on criterion 1 (intense anger) [χ2 (1, n = 489) = 8.99, P < .001], criterion 2 (affective instability) [χ2 (1, n = 489) = 11.18, P < .001], and criterion 9 (unstable relationships) [χ2 (1, n = 489) = 12.51, P < .001]. African Americans also had disproportionately higher rates than Caucasians on criterion 9 [χ2 (1, n = 498) = 4.75, P < .05]. Table 2. BPD Criteria Across Ethnic Groups Using the DIPD (structured interview) Caucasian n (%) African American n (%) Hispanic n (%) χ2 df P DIPD-BPD 1 205 (47)b 39 (60)ab 39 (70)a 12.46 2 .002 Intense anger DIPD-BPD 2 234 (54)b 42 (65)ab 44 (79)a 14.50 2 .001 Affective instability DIPD-BPD 3 189 (44) 33 (51) 31 (55) 3.50 2 .17 Emptiness DIPD-BPD 4 153 (35) 22 (34) 25 (45) 1.98 2 .37 Confusion over identity DIPD-BPD 5 161 (37) 20 (31) 25 (45) 2.48 2 .29 Paranoia DIPD-BPD 6 131 (30) 23 (35) 25 (45) 4.82 2 .09 Fears of abandonment DIPD-BPD 7 127 (29) 14 (22) 18 (32) 2.13 2 .35 Self-injury, suicide DIPD-BPD 8 201 (46) 36 (55) 33 (59) 4.41 2 .11 Impulsivity DIPD-BPD 9 174 (40)b 36 (55)ab 37 (66)a 16.93 2 .001 Unstable relationships NOTE. DIPD (N = 554). Proportions that share the same superscript are not significantly different from each other. Table options The STPD criteria that were differentially distributed across ethnic groups were DIPD-STPD criteria 1 (excessive social anxiety), DIPD-STPD criterion 2 (no friends), DIPD-STPD criterion 3 (odd beliefs), and DIPD-STPD criterion 5 (paranoia) (Table 3). African Americans had disproportionately higher rates of social anxiety [χ2 (1, n = 498) = 5.28, P < .05], no close friends [χ2 (1, n = 498) = 7.45, P < .01], odd beliefs [χ2 (1, n = 498) = 5.59, P < .05], and paranoia [χ2 (1, n = 498) = 9.05, P < .01] than Caucasians. African Americans also had disproportionately higher rates of social anxiety and odd beliefs compared to Hispanics: [χ2 (1, n = 121) = 6.03, P < .01] and [X2 (1, n = 121) = 5.96, P < .05], respectively. Table 3. STPD Criteria Across Ethnic Groups Using the DIPD (structured interview) Caucasian n (%) African American n (%) Hispanic n (%) χ2 df P DIPD-STPD 1 77 (18)b 20 (31)a 6 (11)b 8.85 2 .01 Social anxiety DIPD-STPD 2 112 (26)b 28 (43)a 18 (32)ab 8.61 2 .01 No friends DIPD-STPD 3 92 (21)b 23 (35)a 8 (14)b 8.8 2 .01 Odd beliefs DIPD-STPD 4 114 (26) 25 (39) 15 (27) 4.17 2 .12 Unusual perceptions DIPD-STPD 5 152 (35)b 36 (55)a 22 (39)ab 9.92 2 .01 Paranoia DIPD-STPD 6 135 (31) 29 (45) 20 (36) 4.78 2 .09 Ideas of reference DIPD-STPD 7 37 (9) 2 (3) 1 (2) 5.27 2 .07 Odd behavior DIPD-STPD 8 47 (11) 7 (11) 3 (5) 1.64 2 .44 Odd thinking DIPD-STPD 9 38 (9) 8 (12) 1 (2) 4.51 2 .11 Constricted affect NOTE. DIPD (N = 554). Proportions that share the same superscript are not significantly different from each other. Table options Gender: simple main and moderating effects To analyze potential gender effects, separate chi-square analyses were performed for men and women. As shown in Table 4, significant differences were found for BPD among men and STPD among women. Post hoc comparisons revealed that Hispanic men had disproportionately more BPD diagnoses than Caucasian men [χ2 (1, n = 186) = 11.59, P < .001]. African American women had more STPD diagnoses than both Hispanic women [χ2 (1, n = 88) = 3.99, P < .05] and Caucasian women [χ2 (1, n = 313) = 11.28, P < .001]. No gender effects were found for AVPD or OCPD. Logistic regression analyses were conducted to test for an interaction between ethnicity and gender predicting BPD and STPD, with gender, ethnicity, and the interaction term entered simultaneously. Contrast coding was used to examine the effect of each group relevant to the average effects of the other groups. All of the gender by ethnicity interactions were not significant. Table 4. Ethnic Distribution of Personality Disorders by Gender Caucasian n (%) African American n (%) Hispanic n (%) χ2 df P Men n = 169 n = 16 n = 17 BPD 46 (27)b 5 (31)ab 12 (71)a 13.54 2 .001 STPD 45 (27) 5 (31) 3 (18) .87 2 .65 OCPD 88 (52) 5 (31) 5 (29) 5.25 2 .07 AVPD 94 (56) 7 (44) 10 (59) .94 2 .62 Women n = 264 n = 49 n = 39 BPD 125 (47) 21 (43) 24 (62) 3.41 2 .18 STPD 25 (10)b 13 (27)a 3 (8)b 12.36 2 .002 OCPD 108 (41) 24 (49) 24 (62) 6.36 2 .04 AVPD 154 (58) 28 (57) 22 (56) .07 2 .97 NOTE. Proportions that share the same superscript are not significantly different from each other.