Journal of Personality Disorders, 29(1), 1-14, 2015 © 2015 The Guilford Press HENGARTNER ET AL. ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY

ASSOCIATION BETWEEN CHILDHOOD MALTREATMENT AND NORMAL ADULT PERSONALITY TRAITS: EXPLORATION OF AN UNDERSTUDIED FIELD Michael P. Hengartner, PhD, Lisa J. Cohen, PhD, Stephanie Rodgers, MS, Mario Müller, PhD, Wulf Rössler, MD, and Vladeta Ajdacic-Gross, PhD

We assessed normal personality traits and childhood trauma in approximately 1170 subjects from a general population–based community sample. In bivariate analyses emotional abuse was most pervasively related to personality, showing significant detrimental associations with neuroticism, extraversion, openness, conscientiousness, and agreeableness. Neuroticism was significantly related to emotional abuse and neglect, physical abuse and neglect, and sexual abuse. Emotional abuse was related to neuroticism in men more profoundly than in women (b = 0.095). Adjusting for the covariance between childhood maltreatment variables, neuroticism was mainly related to emotional abuse (b = 0.193), extraversion to emotional neglect (b = –0.259), openness to emotional abuse (b = 0.175), conscientiousness to emotional abuse (b = –0.110), and agreeableness to emotional neglect (b = –0.153). The proportion of variance explained was highest in neuroticism (5.6%) and lowest in openness (1.9%) and conscientiousness (1.8%). These findings help to understand the complex association between childhood maltreatment and both normal and pathological personality.

Childhood maltreatment is a major public health and social-welfare problem, causing extreme personal distress and societal burden (Gilbert et al., 2009). In the literature the association between childhood maltreatment and personality disorders is well documented and has in various studies been repeatedly replicated as an important risk factor (Afifi et al., 2011; Battle et al., 2004; Hengartner, Müller, Rodgers, Rössler, & Ajdacic-Gross, 2013; Johnson, Cohen, Brown, Smailes, & Bernstein, 1999; Saleptsi et al., 2004). Taking into account all these studies focusing on personality disorders, it is This article was accepted under the editorship of Robert F. Krueger and John Livesley. From Department of Psychiatry, Psychotherapy and Psychosomatics, University of Zurich, Zurich, Switzerland (M.P.H., S.R., M.M., W.R., V A.-G.); Beth Israel Medical Center, New York, New York, USA (L.J.C.); and Collegium Helveticum, a joint Research Institute of the University of Zurich and the Swiss Federal Institute of Technology, Zurich, Switzerland (W.R.). ZInEP was supported by a private donation. We would like to thank the donor cordially. Address correspondence to Michael P. Hengartner, PhD, Department of Psychiatry, Psychotherapy and Psychosomatics, University of Zurich, PO Box 1930, CH-8021 Zurich, Switzerland; E-mail: michael. [email protected]

1

2

HENGARTNER ET AL.

rather surprising that childhood maltreatment has hardly been examined in association with adaptive adult personality traits. Especially in view of the widely accepted conception of personality disorders as extreme variants of normal personality traits (Clark, 2007; Trull & Durrett, 2005; Widiger & Mullins-Sweatt, 2009), it is clear that research should remedy the paucity of knowledge. Furthermore, although some individuals with a history of childhood maltreatment develop severe personality disorders as adults, many do not. Therefore it is important to focus additionally on normal personality traits to detect subtle, less pathological effects of childhood adversity on personality functioning. The need for such an analysis arises from a consistent body of evidence showing that normal personality traits (in particular neuroticism) may detrimentally impact public (mental) health and psychosocial functioning (Krueger, Caspi, Moffitt, Silva, & McGee, 1996; Lahey, 2009). Only a few studies have analyzed adaptive adult personality traits in association with childhood maltreatment in the general population. Our literature search on PubMed and PsycInfo yielded only two recent studies specifically addressing this issue. Moran et al. (2011) found that childhood sexual abuse was mainly related to lower agreeableness and higher neuroticism. Unfortunately, the authors focused only on childhood sexual abuse. Since multiple studies have found sexual abuse to be a predictor of psychopathology in bivariate analyses but not in multivariate analyses when adjusted for other forms of abuse (Cohen et al., 2013; Hengartner, Ajdacic-Gross, Rodgers, Müller, & Rössler, 2013; for reviews see Rind & Tromovitch, 1997; Rind, Tromovitch, & Bauserman, 1998), it is crucial to consider other forms of childhood maltreatment as well. In the other recent article, Allen and Lauterbach (2007) reported that subjects with a history of childhood abuse scored higher on neuroticism and openness. However, they did not report associations for specific forms of abuse and did not include all domains of the Big Five traits of personality (agreeableness and conscientiousness were not assessed). Since there is evidence for an interaction with sex (Moran et al., 2011) and because personality is differentially distributed across the sexes (Schmitt, Realo, Voracek, & Allik, 2008), more attention should also be paid to this potential modifier. Thus, to the best of our knowledge, this is the first study that examines the relationship between various forms of childhood maltreatment and all Big Five personality traits in a large general population–based community sample of male and female adults. METHODS STUDY DESIGN AND SAMPLING

This study was conducted as part of the Epidemiology Survey of the Zurich Programme for Sustainable Development of Mental Health Services (ZInEP; in German: Zürcher Impulsprogramm zur nachhaltigen Entwicklung der Psychiatrie), a research and health care program involving several psychiatric research divisions and mental health services of the canton of Zurich,

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 3

FIGURE 1. Sampling design.

Switzerland. The Epidemiology Survey is one of six ZInEP subprojects and consists of four components: (1) a short telephone screening, (2) a comprehensive semistructured face-to-face interview followed by self-report questionnaires, (3) tests in the socio-physiological laboratory, and (4) a longitudinal survey. For the present study only data from component (2) was used (see also Figure 1). The telephone screening and semistructured interviews started in August 2010, the tests at the socio-physiological laboratory in February 2011, and the longitudinal survey in April 2011. The screening ended in May 2012 and all other components in September 2012. First, a total of 9829 Swiss males and females aged 20–41 years at the onset of the survey and considered representative of the general population in that age group in the canton of Zurich, Switzerland, were screened by a computer-assisted telephone interview (CATI), using the Symptom Checklist 27 (SCL-27) (Hardt, Egle, Kappis, Hessel, & Brähler, 2004). All participants were randomly chosen through the resident registration offices of all municipalities of the canton of Zurich. Residents without Swiss nationality were excluded from the study. The age range was chosen to align the data to the Zurich cohort study (Angst et al., 2005), a broad psychiatric survey also conducted in the canton of Zurich. The CATI was conducted by GfK (Growth for Knowledge), a major market and field research institute, in accordance with instructions from the research team. The overall response rate was 53.6%. Reasons for non-response were no answer, only telephone responder, incorrect telephone number, communication impossible, unavailability during the

4

HENGARTNER ET AL.

study period, or refusal by a third person or the target person. In cases where potential subjects were available by telephone, the response rate was 73.9%. Second, 1500 subjects (750 males and females each) were randomly selected from the initial screening sample for subsequent face-to-face interviews. We applied a stratified sampling procedure including 60% high scorers (scoring above the 75th percentile of the global severity index of the SCL-27) and 40% low scorers (scoring below the 75th percentile of the global severity index). The basic sampling design was adapted from the longitudinal Zurich cohort study (Angst et al., 2005) and was chosen to enrich the sample with subjects at high risk of mental disorders. Such a two-phase procedure with initial screening and subsequent interview with a stratified subsample is fairly common in epidemiological surveys (Dunn, Pickles, Tansella, & Vazquez-Barquero, 1999). Face-to-face interviews were conducted by experienced and extensively trained clinical psychologists. The interviews took place either at the participants’ homes or at the University Hospital of Psychiatry in Zurich. All participants who completed the semistructured interview (N = 1500) were required to complete additional questionnaires, but because of missing values and uncompleted questionnaires the final sample included in the analysis was approximately N = 1170. The exact numbers for each questionnaire are indicated in Table 1. Participants received a 20 CHF (about $22USD) coupon for completing the semistructured interview. The ZInEP Epidemiology Survey was approved by the ethics committee of the canton of Zurich as fulfilling all legal and data privacy protection requirements and is in strict accordance with the Declaration of Helsinki of the World Medical Association. All participants gave their written informed consent. INSTRUMENT AND MEASURES

The Big Five Inventory short form (BFI-S) (Schupp & Gerlitz, 2008) is a German adaptation of the popular Big Five Inventory by John, Donahue, and Kentle (1991). The questionnaire consists of 15 items divided into the five broad domains neuroticism, extraversion, openness, agreeableness, and conscientiousness. The items are rated on a seven-point Likert scale ranging from “don’t agree at all” to “completely agree.” Neuroticism assesses being frequently worried, tense, and fearsome. Extraversion measures being talkative, outgoing, and sociable. Openness assesses being inventive, imaginative, and sensation seeking. Agreeableness captures being gentle, forgiving, and cordial. Finally, conscientiousness measures being thorough, diligent, and efficient. The BFI-S has shown good reliability and validity (Schupp & Gerlitz, 2008). In the present study internal consistency ranged from α = 0.6 for agreeableness to α = 0.8 for extraversion (see also Table 1). Childhood maltreatment was assessed with the short-form of the Childhood Trauma Questionnaire (CTQ; Bernstein et al., 2003). The CTQ is a popular retrospective measure of child abuse and neglect and consists of 28 items divided into a control scale named denial and the five domains emotional abuse, emotional neglect, sexual abuse, physical abuse, and physical

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 5 TABLE 1. Descriptive Statistics Cronbach’s α

N

Min.

Max.

Mean

SD

Age

1500

20.00

41.00

29.17

6.84

Neuroticism

1173

3.00

21.00

11.69

3.79

0.70

Extraversion

1171

3.00

21.00

14.33

3.77

0.78

Openness

1169

4.00

21.00

14.92

3.54

0.66

Conscientiousness

1170

3.00

21.00

16.23

3.00

0.61

Agreeableness

1169

4.00

21.00

16.11

2.97

0.57

Emotional abuse

1162

1.00

5.00

1.56

0.72

0.83

Emotional neglect

1160

1.00

5.00

1.88

0.85

0.89

Physical abuse

1161

1.00

4.80

1.14

0.38

0.73

Physical neglect

1165

1.00

3.80

1.28

0.41

0.39

Sexual abuse

1165

1.00

5.00

1.14

0.48

0.92

neglect. The items are rated on a five-point Likert scale ranging from “never true” to “very often true.” For the present study we applied a German adaptation (Wingenfeld et al., 2010). Internal consistency of the different domains is high and test-retest reliability is also good. The CTQ showed good convergent and discriminant validity and can be considered to be a sensitive and valid screening questionnaire for childhood maltreatment (Bernstein et al., 2003; Klinitzke, Romppel, Häuser, Brähler, & Glaesmer, 2012; Wingenfeld et al., 2010). In the present study internal consistency ranged from α = 0.4 for physical neglect to α = 0.9 for sexual abuse (see also Table 1). Internal consistency was unacceptably low for physical neglect, suggesting that the items tapping physical neglect were only slightly correlated. This limitation was previously noted by Wingenfeld et al. (2010). STATISTICAL ANALYSIS

The Big Five personality traits were standardized using z-transformation (mean = 0.0; SD = 1.0). All childhood adversity variables were grand mean centered to reduce the effects of multicollinearity (i.e., each individual score was subtracted from the sample’s mean value). Then we conducted a series of hierarchical multiple linear regression analyses with the Big Five personality traits as the dependent variables. Age was entered in the first block because it is associated with the Big Five traits and may be related to the recall of adverse childhood experiences, followed by sex in the second block. Each form of childhood maltreatment was entered separately in the third block. Finally, the interaction term between childhood maltreatment and sex was fitted in the fourth block. This hierarchical modeling allowed us to estimate the proportion of partial variance explained by each form of childhood maltreatment separately while adjusting for sex and age. A final multivariate model for each Big Five trait was computed by entering sex and age in the first block and all forms of childhood maltreatment together in the second block. All regression analyses were performed with SPSS version 20 for Apple Ma-

6

HENGARTNER ET AL. TABLE 2. Correlation Matrix of Personality Traits and Childhood Maltreatment Variables N

E

O

C

A

EA

EN

PA

PN

Neuroticism (N) Extraversion (E) Openness (O)

–0.183** 0.040

0.255**

Conscientiousness (C)

–0.133** 0.204**

Agreeableness (A)

–0.115** 0.093** 0.104** 0.253**

0.051

Emotional abuse (EA)

0.247** –0.088** 0.114**

–0.054 –0.128**

Emotional neglect (EN)

0.198** –0.177**

0.034

–0.055 –0.171** 0.711**

Physical abuse (PA)

0.104**

–0.035

0.073*

0.019

Physical neglect (PN)

0.138**

–0.052

0.042

–0.019 –0.095** 0.461** 0.592** 0.384**

–0.071* 0.530** 0.432**

Sexual abuse (SA)

0.101**

–0.006

0.055

0.052

–0.007

0.390** 0.310** 0.442** 0.265**

Note. *p < 0.05. **p < 0.01 (two-tailed significance).

cintosh. Post-hoc power analysis was carried out with G*Power version 3 (Faul, Erdfelder, Lang, & Buchner, 2007). RESULTS

The descriptive statistics (raw scores) are reported in Table 1 and the correlation matrix in Table 2. No violation of the assumptions of multiple linear regression analysis was observed. Normality of the residuals, autocorrelation, homoscedasticity, and multicollinearity were acceptable in all models (data not shown). The associations between Big Five personality traits and childhood maltreatment are reported in Table 3. Male sex was significantly and negatively related to neuroticism (b = –0.168), extraversion (b = –0.105), conscientiousness (b = –0.138), and agreeableness (b = –0.078). Emotional abuse was significantly associated with higher neuroticism (b = 0.235), lower extraversion (b = –0.099), higher openness (b = 0.120), lower conscientiousness (b = –0.097), and lower agreeableness (b = –0.134). The interaction between emotional abuse and sex was statistically significant only for neuroticism (b = 0.095), indicating that scores on neuroticism were higher in emotionally abused men than in women. The corresponding separate main effects of sex were b = 0.289 for men and b = 0.208 for women. However, note that the interaction effect was rather small, accounting for less than 1% of the increase in variance explained. Emotional neglect was significantly related to higher neuroticism (b = 0.203), lower extraversion (b = –0.177), lower conscientiousness (b = –0.094), and lower agreeableness (b = –0.166). No significant interaction emerged between emotional neglect and sex. Physical abuse was significantly related to higher neuroticism (b = 0.111), higher openness (b = 0.074), and lower agreeableness (b = –0.064). There was no significant interaction between physical abuse and sex. Physical neglect was significantly associated with higher neuroticism (b = 0.139) and lower agreeableness (b = –0.091), and there was no significant interaction with

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 7 TABLE 3. Associations Between Personality Traits, Sex, and Childhood Maltreatment Male sex

Emotional abuse

Emotional abuse × male sex

Emotional neglect

Physical neglect × male sex

Sexual abuse

Sexual abuse × male sex

Open.

Cons.

Agree. –0.078**

β

–0.168**

–0.105**

0.004

–0.138**

0.028

0.011

0.000

0.019

0.006

β

0.235**

–0.099**

0.120**

–0.097**

–0.134**

∆R2

0.053

0.009

0.014

0.009

0.017

β

0.095**

0.047

0.050

0.005

0.035

∆R2

0.006

0.001

0.002

0.000

0.001

β

0.203**

–0.177**

0.037

–0.094**

–0.166**

∆R2

0.040

0.030

0.001

0.009

0.027

0.047

0.029

–0.042

0.026

0.002

∆R2

0.001

0.001

0.001

0.000

0.000

β

0.111**

–0.029

0.074*

0.010

–0.064*

∆R2

0.012

0.001

0.005

0.000

0.004

0.058

0.035

–0.026

0.046

0.037

∆R2

0.002

0.001

0.000

0.001

0.001

β

0.139**

–0.050

0.042

–0.027

–0.091**

∆R2

0.019

0.002

0.002

0.001

0.008

β

0.036

0.049

–0.027

0.024

0.024

∆R2

0.001

0.001

0.000

0.000

0.000

–0.014

0.057

0.020

–0.009

Physical abuse × male sex

Physical neglect

Extra.

∆R2

Emotional neglect × male sex

Physical abuse

Neuro.

β

0.085**

∆R2

0.007

0.000

0.003

0.000

0.000

β

0.064

–0.001

0.048

–0.008

–0.008

∆R2

0.003

0.000

0.002

0.000

0.000

Note. *p < 0.05. **p < 0.01 (two-tailed significance).

sex. Finally, sexual abuse was significantly related only to neuroticism (b = 0.085). No significant interaction between sexual abuse and sex emerged. The multivariate analyses where all forms of childhood maltreatment were adjusted for each other are indicated in Table 4. In neuroticism, childhood maltreatment accounted for 5.6% of total variance explained, and the strongest predictor was emotional abuse (b = 0.193). In extraversion the proportion of total variance explained was 3.7%, and the strongest predictor was emotional neglect (b = –0.259). In openness the proportion of total variance explained was 1.9%, and the main predictor was emotional abuse (b = 0.175). In conscientiousness childhood maltreatment explained 1.8% of total variance, and emotional abuse was again the predominant predictor (b = –0.110). Finally, in agreeableness the proportion of total variance explained was 3.0% and the strongest predictor was emotional neglect (b = –0.153). For conscientiousness and openness the impact of childhood maltreatment on personality traits represented multivariately small effects, for extraversion and agreeableness a small-to-medium effect size, and for neuroticism a medium effect size. We also ran separate multivariate analyses for each form of childhood maltreatment as the dependent variable and Big Five traits as the independent variables (again, adjusted for sex and age). Those results showed that the five personality traits accounted for the following proportion of variance

8

HENGARTNER ET AL.

TABLE 4. Multivariate Associations Between Personality Traits and Childhood Maltreatment, Adjusted for Sex and Age Neuro.

Extra.

Open.

Cons.

Agree.

Emotional abuse

β

0.193**

0.026

0.175**

–0.110*

–0.049

Emotional neglect

β

0.062

–0.259**

–0.107*

–0.080

–0.153**

Physical abuse

β

–0.027

0.032

0.015

0.071

0.001

Physical neglect

β

0.026

0.070

0.016

0.031

0.009

Sexual abuse

β

0.000

0.021

0.015

0.045

0.051

∆R2

0.056

0.037

0.019

0.018

0.030

Variance explained

Note. *p < 0.05 .**p < 0.01 (two-tailed significance).

explained: 8.6% in emotional abuse, 8.2% in emotional neglect, 2.4% in physical abuse, 2.9% in physical neglect, and 1.1% in sexual abuse. The strongest predictor of emotional abuse was neuroticism (b = 0.193), whereas emotional neglect was best predicted by both neuroticism (b = 0.148) and extraversion (b = –0.149). Finally, physical abuse (b = 0.107) and neglect (b = 0.124) as well as sexual abuse (b = 0.083) were again best predicted by neuroticism. Due to the large sample size, the statistical power (1-b) achieved for a partial R2 of 0.02 (which corresponds to a small effect size) set for N = 1110 and α = 0.05 was as impressively high as 0.98. Consequently, the probability of any type II error (b; i.e., the probability that a truly existent small effect could not be detected at a level of statistical significance α = 0.05) was only 2%. Thus, the statistical power achieved in our analyses was very high. DISCUSSION

This is the first study to examine the associations of emotional, physical, and sexual abuse as well as emotional and physical neglect with all personality traits in a large general population–based community sample. A particular strength of the present study is its impressively high statistical power and thus its ability to detect even small effects at the level of statistical significance. In the bivariate analyses emotional abuse was significantly associated with all Big Five traits, but in particular with increased neuroticism and openness as well as low agreeableness. Emotional neglect was mainly related to high neuroticism, low extraversion, and low agreeableness. Physical abuse and neglect primarily predicted neuroticism as well. Finally, sexual abuse was uniquely associated with neuroticism but with a very small effect size. Thus, in conclusion, all forms of childhood maltreatment were bivariately related to increased neuroticism, whereas emotional abuse was associated with all Big Five traits. In the multivariate analysis childhood maltreatment accounted for 5.6% of total variance explained in neuroticism (medium effect size). When mutually adjusted for each other the only forms of childhood maltreatment that were substantially related to Big Five traits were emotional abuse and neglect.

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 9

These results are mostly in line with the literature. For instance, Allen and Lauterbach (2007) as well as Moran and colleagues (2011) found significant associations of neuroticism with childhood trauma and sexual abuse, respectively. In line with the present study Allen and Lauterbach (2007) also reported a significant association with higher openness, which is related to psychoticism (Widiger & Mullins-Sweatt, 2009). In a small sample of incarcerated male sexual offenders, Becerra-Garcia, Garcia-Leon, and Egan (2012) found higher scores in neuroticism and openness as well as lower scores in conscientiousness and agreeableness in offenders with a history of childhood maltreatment. The major limitation of the latter study is that a sample of sexual offenders is hardly representative of the general population. In contrast to Moran et al. (2011), we did not find that childhood sexual abuse was related to lower agreeableness. Using a longitudinal design with a sample of low-income children, Rogosch and Cicchetti (2004) found that at age 6 maltreated children already showed altered personality traits, in particular lower agreeableness, conscientiousness, and openness as well as higher neuroticism. Moreover, those alterations in personality functioning were stable from ages 6 to 9, suggesting that those patterns are consistent and enduring. The consistent association between neuroticism and all forms of childhood maltreatment is highly meaningful with respect to personality disorders. Neuroticism is the most important and predominant trait underlying personality disorder symptomatology (Hengartner, Ajdacic-Gross, Rodgers, Müller, & Rössler, 2014; Saulsman & Page, 2004). A growing body of evidence suggests that personality disorders are best conceived as five continuous higher order domains (Krueger, Derringer, Markon, Watson, & Skodol, 2012). These five domains, labeled negative affect, detachment, antagonism, disinhibition, and psychoticism, align closely with the personality traits neuroticism, introversion, agreeableness (–), conscientiousness (–), and openness (Thomas et al., 2013). Thus, the findings of our study provide further evidence for an association between childhood adversity and personality pathology. Our data also demonstrate that childhood maltreatment relates not only to maladaptive personality disorder traits or extreme variants along continuous personality dimensions (Hengartner, Ajdacic-Gross, et al., 2013), but also to the whole “normal” range of personality functioning. Although due to the cross-sectional design of this study we cannot draw strict causal conclusions, we suggest that adverse childhood experiences interfere with and influence personality development significantly in the long term. The impact of childhood maltreatment on personality functioning may range from mainly subtle increases in neuroticism, rendering someone more anxious and insecure, up to radical dysfunctional personality pathology, causing, for instance, severe temper tantrums and self-mutilation. We posit that a gene–environment interaction (Caspi et al., 2002) may describe the causal pathway most appropriately. Neuroticism has been related to functional impairment (Judge & Ilies, 2002; Michon et al., 2008) and health care utilization (Goodwin, Hoven,

10

HENGARTNER ET AL.

Lyons, & Stein, 2002; ten Have, Oldehinkel, Vollebergh, & Ormel, 2005). It has also been shown that personality traits predict treatment utilization independently of PD symptomatology (Miller, Pilkonis, & Mulvey, 2006), thus these normal personality traits possess important incremental validity, which supports their assessment in addition to personality disorders. The association between neuroticism and childhood maltreatment has therefore important implications for public mental health policies as well as clinical practice. These diverse associations of neuroticism may in part also explain why general personality disorder traits are so pervasively related to functional impairment (Hengartner, Müller, Rodgers, Rössler, & Ajdacic-Gross, 2014a, 2014b). However, the clinical relevance of neuroticism and its significance for public health go far beyond personality functioning, as this personality trait strongly relates to various other mental disorders as well as physical health and global quality of life (Lahey, 2009). Therefore, childhood maltreatment may have extensive and numerous detrimental consequences apart from and along with personality disorder development. High openness and low agreeableness are indicative of aggressiveness and impulsive sensation seeking, which in turn are facets of a superordinate factor called externalization, comprising substance abuse and other risky or antisocial behaviors (Markon, Krueger, & Watson, 2005). Therefore, the associations of those personality traits with childhood maltreatment may help to understand why maltreated subjects show increased risk of substance abuse as well as promiscuous sexual and criminal behavior (Gilbert et al., 2009). Furthermore, it may in addition partially explain the phenomenon of revictimization, which is very common in abused subjects (Widom, Czaja, & Dutton, 2008), because childhood maltreatment predicts externalizing problems and the latter predict revictimization (Fargo, 2009). Of note, the data support the differential associations of different types of maltreatment. Emotional abuse and neglect had the strongest effect sizes, and physical and sexual abuse the smallest. This is consistent with other multivariate studies of the relative impact of different types of maltreatment (Cohen et al., 2013; Grilo & Masheb, 2002; Johnson et al., 1999; Joyce et al., 2003). These findings suggest that forms of childhood maltreatment that most directly impact the attachment between parent and child may have a more pernicious effect on personality development than those maltreatment types characterized by episodic abusive behavior. Such an interpretation is consistent with the proposed changes to personality disorder diagnoses in DSM-5 (alternative personality disorder model from section III), such that impaired interpersonal functioning is listed as a gateway criterion to any personality disorder diagnosis (Skodol, 2012). It is also possible that physical and sexual abuse occur more episodically and thus with less chronicity than emotional abuse and neglect and therefore have a less pervasive effect. Unfortunately, all those studies cited above, including our own, have in common that they cannot provide data about the personality structure before maltreatment of a child occurred. As a consequence we do not know if the traumatic event itself causes disruptions in a child’s personality development.

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 11

Furthermore, although the effect sizes of emotional abuse and neglect were clearly different in magnitude to those of physical and sexual abuse, the confidence intervals of the regression coefficients slightly overlapped, which is why we may not certainly assert that they also differ in their statistical significance. One interaction term was statistically significant in the present study, that is, the interaction between emotional abuse and sex in neuroticism, indicating that the association was stronger in males than in females. However, the corresponding effect size was quite small and accounted for less than 1% of total variance. Moran et al. (2011) found increased effects of childhood sexual abuse on extraversion and neuroticism in males. Although we did not find a significant interaction with respect to sexual abuse, both findings are consistent in that they both indicate that men may be more strongly affected in their personality structure by childhood maltreatment than women. This is an important finding that should be considered in clinical practice, especially in respect to the association between childhood abuse and conduct problems, the latter being clearly more prevalent in males (Moffitt, Caspi, Rutter, & Silva, 2001). Finally, main effects of sex were statistically significant for neuroticism, extraversion, conscientiousness, and agreeableness, indicating that women scored slightly higher than men in those four personality traits. The sex differences in broad personality domains reported here are fully consistent with a large survey that replicated these findings across 55 nations (Schmitt et al., 2008). The results of the present study need to be interpreted in the context of the following limitations. First, because of the cross-sectional design the assessment of childhood adversity was necessarily retrospective. Thus, we were not able to determine a clear temporal order and accordingly cannot draw causal conclusions from our data. At this time we still do not know whether childhood maltreatment alters personality functioning and development or whether specific personality traits predispose someone to maltreatment. Moreover, we must also consider that persons with specific personality traits may more readily appraise a given experience as abuse or alternatively that they are more or else less willing to report any experience. Second, a serious limitation of retrospective assessment and self-report instruments may be a recall bias. There is evidence of substantial underreporting of childhood adversity (Hardt & Rutter, 2004), which may have decreased the associations reported in this study and which leads to underestimation of the true effect size. Moreover, the CTQ sexual abuse scale is particularly limited as it does not specify either the perpetrator or nature of sexual abuse, which is of critical import in measuring the impact of sexual abuse (Collishaw et al., 2007; Cutajar et al., 2010). Finally, we used a short form to assess the Big Five personality traits. Although this instrument is valid and reliable, it does not assess all the facets of the complete form. This may have slightly influenced our results. In sum, in the context of these limitations this study adds valuable information to current research on the association between childhood maltreat-

12

HENGARTNER ET AL.

ment and normal personality traits. Our data showed that, within the general population, multiple forms of childhood maltreatment, but in particular emotional abuse and neglect, have a moderate and meaningful relationship to a broad range of normal personality domains. References Afifi, T. O., Mather, A., Boman, J., Fleisher, W., Enns, M. W., Macmillan, H., et al. (2011). Childhood adversity and personality disorders: Results from a nationally representative population-based study. Journal of Psychiatric Research, 45(6), 814–822. Allen, B., & Lauterbach, D. (2007). Personality characteristics of adult survivors of childhood trauma. Journal of Traumatic Stress, 20(4), 587–595. Angst, J., Gamma, A., Neuenschwander, M., Ajdacic-Gross, V., Eich, D., Rössler, W., et al. (2005). Prevalence of mental disorders in the Zurich cohort study: A twenty year prospective study. Epidemiologia e Psichiatria Sociale, 14(2), 68–76. Battle, C. L., Shea, M. T., Johnson, D. M., Yen, S., Zlotnick, C., Zanarini, M. C., et al. (2004). Childhood maltreatment associated with adult personality disorders: Findings from the Collaborative Longitudinal Personality Disorders Study. Journal of Personality Disorders, 18(2), 193–211. Becerra-Garcia, J. A., Garcia-Leon, A., & Egan, V. (2012). Childhood abuse history differentiates personality in sex offenders. Journal of Forensic Psychiatry and Psychology, 23(1), 61–66. Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., et al. (2003). Development and validation of a brief screening version of the Childhood Trauma Questionnaire. Child Abuse and Neglect, 27(2), 169–190. Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., et al. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297(5582), 851– 854. Clark, L. A. (2007). Assessment and diagnosis of personality disorder: Perennial issues and an emerging reconceptualization. Annual Review of Psychology, 58, 227–257. Cohen, L. J., Foster, M., Nesci, C., Tanis, T., Halmi, W., & Galynker, I. (2013). How do different types of childhood maltreatment relate to adult personality pathology? Journal of Nervous and Mental Disease, 201(3), 234–243. Collishaw, S., Pickles, A., Messer, J., Rutter, M., Shearer, C., & Maughan, B. (2007). Resilience to adult psychopathology following

childhood maltreatment: Evidence from a community sample. Child Abuse and Neglect, 31(3), 211–229. Cutajar, M. C., Mullen, P. E., Ogloff, J. R., Thomas, S. D., Wells, D. L., & Spataro, J. (2010). Psychopathology in a large cohort of sexually abused children followed up to 43 years. Child Abuse and Neglect, 34(11), 813–822. Dunn, G., Pickles, A., Tansella, M., & VazquezBarquero, J. L. (1999). Two-phase epidemiological surveys in psychiatric research. British Journal of Psychiatry, 174, 95–100. Fargo, J. D. (2009). Pathways to adult sexual revictimization: Direct and indirect behavioral risk factors across the lifespan. Journal of Interpersonal Violence, 24(11), 1771–1791. Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191. Gilbert, R., Widom, C. S., Browne, K., Fergusson, D., Webb, E., & Janson, S. (2009). Burden and consequences of child maltreatment in high-income countries. Lancet, 373(9657), 68–81. Goodwin, R. D., Hoven, C. W., Lyons, J. S., & Stein, M. B. (2002). Mental health service utilization in the United States. The role of personality factors. Social Psychiatry and Psychiatric Epidemiology, 37(12), 561– 566. Grilo, C. M., & Masheb, R. M. (2002). Childhood maltreatment and personality disorders in adult patients with binge eating disorder. Acta Psychiatrica Scandinavica, 106(3), 183–188. Hardt, J., Egle, U. T., Kappis, B., Hessel, A., & Brähler, E. (2004). Symptom Checklist SCL-27: Results of a representative German survey. Psychotherapie Psychosomatik Medizinische Psychologie, 54(5), 214–223. Hardt, J., & Rutter, M. (2004). Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology and Psychiatry, 45(2), 260–273. Hengartner, M. P., Ajdacic-Gross, V., Rodgers, S., Müller, M., & Rössler, W. (2013). Childhood adversity in association with person-

ASSOCIATION BETWEEN CHILD MALTREATMENT AND ADULT PERSONALITY 13 ality disorder dimensions: New findings in an old debate. European Psychiatry, 28(8), 476–482. Hengartner, M. P., Ajdacic-Gross, V., Rodgers, S., Müller, M., & Rössler, W. (2014). The joint structure of normal and pathological personality: Further evidence for a dimensional model. Comprehensive Psychiatry, 55, 667–674. doi:10.1016/j. comppsych.2013.10.011 Hengartner, M. P., Müller, M., Rodgers, S., Rössler, W., & Ajdacic-Gross, V. (2013). Can protective factors moderate the detrimental effects of child maltreatment on personality functioning? Journal of Psychiatric Research, 47(9), 1180–1186. Hengartner, M. P., Müller, M., Rodgers, S., Rössler, W., & Ajdacic-Gross, V. (2014a). Interpersonal functioning deficits in association with DSM-IV personality disorder dimensions. Social Psychiatry and Psychiatric Epidemiology, 49, 317–325. doi:10.1007/ s00127-013-0707-x Hengartner, M. P., Müller, M., Rodgers, S., Rössler, W., & Ajdacic-Gross, V. (2014b). Occupational functioning and work impairment in association with personality disorder traitscores. Social Psychiatry and Psychiatric Epidemiology, 49, 327–335. doi:10.1007/ s00127-013-0739-2 John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory – Versions 4a and 54. Berkeley, CA: University of California, Berkeley, Institute of Personality and Social Research. Johnson, J. G., Cohen, P., Brown, J., Smailes, E. M., & Bernstein, D. P. (1999). Childhood maltreatment increases risk for personality disorders during early adulthood. Archives of General Psychiatry, 56(7), 600–606. Joyce, P. R., McKenzie, J. M., Luty, S. E., Mulder, R. T., Carter, J. D., Sullivan, P. F., et al. (2003). Temperament, childhood environment and psychopathology as risk factors for avoidant and borderline personality disorders. Australian and New Zealand Journal of Psychiatry, 37(6), 756–764. Judge, T. A., & Ilies, R. (2002). Relationship of personality to performance motivation: A meta-analytic review. Journal of Applied Psychology, 87(4), 797–807. Klinitzke, G., Romppel, M., Häuser, W., Brähler, E., & Glaesmer, H. (2012). The German version of the Childhood Trauma Questionnaire (CTQ)—Psychometric characteristics in a representative sample of the general population. Psychotherapie Psychosomatik Medizinische Psychologie, 62(2), 47–51. Krueger, R. F., Caspi, A., Moffitt, T. E., Silva, P. A., & McGee, R. (1996). Personality traits are differentially linked to mental disorders: A multitrait-multidiagnosis study of an ado-

lescent birth cohort. Journal of Abnormal Psychology, 105(3), 299–312. Krueger, R. F., Derringer, J., Markon, K. E., Watson, D., & Skodol, A. E. (2012). Initial construction of a maladaptive personality trait model and inventory for DSM-5. Psychological Medicine, 42(9), 1879–1890. Lahey, B. B. (2009). Public health significance of neuroticism. American Psychologist, 64(4), 241–256. Markon, K. E., Krueger, R. F., & Watson, D. (2005). Delineating the structure of normal and abnormal personality: An integrative hierarchical approach. Journal of Personality and Social Psychology, 88(1), 139–157. Michon, H. W., ten Have, M., Kroon, H., van Weeghel, J., de Graaf, R., & Schene, A. H. (2008). Mental disorders and personality traits as determinants of impaired work functioning. Psychological Medicine, 38(11), 1627–1637. Miller, J. D., Pilkonis, P. A., & Mulvey, E. P. (2006). Treatment utilization and satisfaction: Examining the contributions of Axis II psychopathology and the Five-Factor Model of personality. Journal of Personality Disorders, 20(4), 369–387. Moffitt, T. E., Caspi, A., Rutter, M., & Silva, P. A. (2001). Sex differences in antisocial behaviour: Conduct disorder, delinquency, and violence in the Dunedin Longitudinal Study. Cambridge: Cambridge University Press. Moran, P., Coffey, C., Chanen, A., Mann, A., Carlin, J. B., & Patton, G. C. (2011). Childhood sexual abuse and abnormal personality: A population-based study. Psychological Medicine, 41(6), 1311–1318. Rind, B., & Tromovitch, P. (1997). A meta-analytic review of findings from national samples on psychological correlates of child sexual abuse. Journal of Sex Research, 34(3), 237–255. Rind, B., Tromovitch, P., & Bauserman, R. (1998). A meta-analytic examination of assumed properties of child sexual abuse using college samples. Psychological Bulletin, 124(1), 22–53. Rogosch, F. A., & Cicchetti, D. (2004). Child maltreatment and emergent personality organization: Perspectives from the five-factor model. Journal of Abnormal Child Psychology, 32(2), 123–145. Saleptsi, E., Bichescu, D., Rockstroh, B., Neuner, F., Schauer, M., Studer, K., et al. (2004). Negative and positive childhood experiences across developmental periods in psychiatric patients with different diagnoses—An explorative study. BMC Psychiatry, 4, 40. Saulsman, L. M., & Page, A. C. (2004). The fivefactor model and personality disorder empirical literature: A meta-analytic review.

14 Clinical Psychology Review, 23(8), 1055– 1085. Schmitt, D. P., Realo, A., Voracek, M., & Allik, J. (2008). Why can’t a man be more like a woman? Sex differences in Big Five personality traits across 55 cultures. Journal of Personality and Socical Psychology, 94(1), 168–182. Schupp, J., & Gerlitz, J.-Y. (2008). BFI-S: Big Five Inventory-SOEP. In A. Glöckner-Rist (Ed.), Zusammenstellung sozialwissenschaftlicher Items und Skalen (ZIS Version 12.00) [Compilation of items and scales in the social sciences (ZIS Version 12.00)]. Bonn, Germany: GESIS. Skodol, A. E. (2012). Personality disorders in DSM-5. Annual Review of Clinical Psychology, 8, 317–344. ten Have, M., Oldehinkel, A., Vollebergh, W., & Ormel, J. (2005). Does neuroticism explain variations in care service use for mental health problems in the general population? Results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Social Psychiatry and Psychiatric Epidemiology, 40(6), 425–431.

HENGARTNER ET AL. Thomas, K. M., Yalch, M. M., Krueger, R. F., Wright, A. G., Markon, K. E., & Hopwood, C. J. (2013). The convergent structure of DSM-5 personality trait facets and five-factor model trait domains. Assessment, 20(3), 308–311. Trull, T. J., & Durrett, C. A. (2005). Categorical and dimensional models of personality disorder. Annual Review of Clinical Psychology, 1, 355–380. Widiger, T. A., & Mullins-Sweatt, S. N. (2009). Five-factor model of personality disorder: A proposal for DSM-V. Annual Review of Clinical Psychology, 5, 197–220. Widom, C. S., Czaja, S. J., & Dutton, M. A. (2008). Childhood victimization and lifetime revictimization. Child Abuse and Neglect, 32(8), 785–796. Wingenfeld, K., Spitzer, C., Mensebach, C., Grabe, H. J., Hill, A., Gast, U., et al. (2010). The German version of the Childhood Trauma Questionnaire (CTQ): Preliminary psychometric properties. Psychotherapie Psychosomatik Medizinische Psychologie, 60(11), 442–450.