Journal of Psychiatric Research xxx (2015) 1e7
Contents lists available at ScienceDirect
Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires
What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits Roberto Sudbrack a, Pedro H. Manfro b, Isadora M. Kuhn b, Hudson W. de Carvalho c, Diogo R. Lara a, b, * ~o Lucas da Pontifícia Universidade Cato lica do Rio Grande do Sul, Porto Alegre, Brazil Serviço de Psiquiatria, Hospital Sa lica do Rio Grande do Sul, Porto Alegre, Brazil Faculdade de Bioci^ encias, Pontifícia Universidade Cato c Departmento de Psicologia, Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 September 2014 Received in revised form 27 November 2014 Accepted 8 January 2015
Childhood trauma arises from damaging experiences and the absence of a secure and reliable environment for child development. Despite many studies on the risk for developing psychiatric disorders, much less is known on the relationship between traumatic experiences and personality traits. A total of 10,800 participants (mean age 27.7 ± 7.8 years old, 69.8% women) anonymously answered the Childhood Trauma Questionnaire (CTQ) and the Affective and Emotional Composite Temperament Scale (AFECTS) in the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP). The results showed that total trauma score was associated with lower levels of positive traits (volition, control, coping, and stability), higher scores of negative traits (sensitivity, anger, anxiety) and a higher frequency of depressive and cyclothymic temperaments. Linear regression analysis showed similar findings for emotional abuse and neglect, whereas physical abuse and neglect were positively associated with more volition and coping as well as less fear, sensitivity and anxiety traits, especially in men. This study has a cross-sectional design and the sole use of self-reporting as the main limitations. In conclusion, dysfunctional temperament profiles were associated mainly with emotional trauma and sexual abuse. Some adaptive traits were associated with physical abuse and neglect, especially in men. These results reinforce the negative impact of emotional trauma and people's resilience to physical trauma, which may even translate into adaptive trait expression in males particularly. We propose that the “contextdependent nature” of stress should be further studied to break down the influence of specific types and contexts of adversity on psychological and psychiatric outcomes. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Childhood trauma Abuse Neglect Temperament Traits Personality
1. Introduction Childhood trauma arises from harmful experiences and from the absence of a supporting and reliable environment that fosters child development. Early traumatic events have been widely associated with a diversity of negative outcomes related to psychological maladjustment and psychiatric conditions, such as mood and personality disorders (Afifi et al., 2011; Allen and Lauterbach, 2007; Flory et al., 2009; Heins et al., 2011; McGloin and Widom, 2001; Sansone et al., 2006; Tucci et al., 2010; Widom et al., 2007). Nevertheless, these evidence are mainly based on case-control
^ncias e PUCRS, Av. Ipiranga, 6681 e * Corresponding author. Faculdade de Biocie Pd12A, Porto Alegre, RS 90619-900, Brazil. Fax: þ55 51 33203612. E-mail address: [email protected] (D.R. Lara).
study designs, which are known to be subjective to selection biases (Gordis, 2010), as clinical/case groups tend to be composed of treatment-seeking individuals, which are not representative of all cases within the trauma exposed population and controls that either healthy or have been exposed to very few traumatic events. Also, case-control methodologies are more suitable for rare conditions, which is not the case for abuse and negligence exposure in childhood (Bassani et al., 2009; Briere and Elliott, 2003; Costa et al., 2007). Another possible limitation regarding abuse and neglect research is unreliable reporting. Most studies rely on classic paperand-pencil self-report, telephone or face-to-face interviews and, although scientific work guarantees anonymity in publications, participants may feel vulnerable and exposed when dealing directly with research staff. In this regard, computer-based data collection has important advantages over the aforementioned
http://dx.doi.org/10.1016/j.jpsychires.2015.01.001 0022-3956/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
2
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
methodologies even when both methods are applied anonymously (Gosling et al., 2004; Turner et al., 1998; Couper, 2000). Moreover, volunteers voluntarily and anonymously accessing their private computers probably feel even less concerned about how they appear to others. This is particularly relevant when answering questions on sensitive issues, such as problems of intimate relationships, illicit drug use, sexual experiences, and traumatic events. In addition, the Internet allows the collection of large research samples at a low cost and with little data loss. Thus, internet assisted methods can enhance response reliability and the acquisition of a diverse sample, although selection biases also apply. Despite the common understanding that traumatic life events and psychological maladjustment are intimately connected, few studies have addressed the relationship between childhood traumatic experiences and personality traits. This gap result from the emphasis on categorical diagnosis in the last few decades and from the view that personality is a construct designed mostly to index normal and social behavioral liabilities (Bornstein et al., 2010; Mathews et al., 2003). Nevertheless, using the Five Factor Model of personality, Allen and Lauterbach (2007) reported that people who have gone through early life trauma present higher scores on neuroticism and openness to new experience than controls. Furthermore, individuals who reported a single traumatic experience in early life showed lower scores than controls on interpersonal dependency, while individuals who reported repeated and prolonged traumatic involvement showed higher scores on interpersonal dependency than controls. Rademaker et al. (2008), using Cloninger Psychobiological Model of Personality (Cloninger et al., 1993), showed significant negative associations between traumatic events and character dimensions of self-directedness and cooperativeness in a military sample, which emphasized the deleterious relationship between trauma and psychological traits. Lastly, Li et al. (2014) found significant positive correlations between emotional and sexual abuse scores in the Childhood Trauma Checklist with neuroticism and psychoticism from Eysenck's personality model. Taken together, these findings suggest that traumatic events not only influence psychopathologies, but also have deleterious effects on traits. We have recently developed the Affective and Emotional Composite Temperament (AFECT) model (Lara et al., 2012a), an integrative temperament model that combines specific dimensional traits and global affective temperaments. It is the evolution of the previous “fear and anger” model (Lara and Akiskal, 2006; Lara et al., 2006). The AFECT model comprises twelve affective temperaments (depressive, anxious, apathetic, cyclothymic, dysphoric, volatile, obsessive, euthymic, hyperthymic, irritable, disinhibited and euphoric), which represent specific configurations of the ten emotional traits (volition, desire, anger, fear, caution, emotional sensitivity, coping, control, and the two recently added traits of anxiety and stability). Using the AFECT scale (AFECTS; Lara et al., 2012a; De Carvalho et al., 2013), we recently showed a global impairment of these traits proportionally to time of exposure to bullying. Bullying victims showed lower Volition, Coping and Control scores, and more Emotional Sensitivity, Anger and Fear (Frizzo et al., 2013). Regarding affective temperaments, being bullied was associated with a lower frequency of the healthy euthymic and hyperthymic types, and a higher proportion of depressive and cyclothymic types. The aim of the current study is to advance in the understanding of how childhood trauma relates to emotional traits and affective temperaments using a different methodology from other studies. More specifically, we evaluated the relationship between the AFECTS parameters with indicators of childhood trauma
(emotional, physical and sexual) assessed retrospectively in a crosssectional web survey. Our initial hypothesis was that all types of childhood trauma would be associated with maladaptive scores in emotional traits and types of affective temperaments in a similar fashion as observed with being bullied.
2. Methods 2.1. Participants The BRAINSTEP web survey (Lara et al., 2012b) was broadcasted in National media (TV news, major newspapers) for recruiting volunteers (18 yrs or older) to answer several psychiatric and psychological scales and questionnaires. The participation and honesty of volunteers was stimulated by the incentive of receiving a brief report on their temperament profile (i.e., traits levels of volition, fear, anger, control and sensitivity, as well as a description of their affective temperament), and the chance of having a psychiatric disorder based on their answers to screening instruments for 19 disorders. Before responding to the questionnaires included in the site (www.temperamento.com.br), respondents gave their electronically informed consent. Their participation was voluntary and they could cancel it at any moment, with no justification necessary. This form fulfills the requirements of the National Health Council of Brazil (Resolution 196/1996) and the Code of Ethics of the World Medical Association (Declaration of Helsinki). The Institutional ~o Lucas from Pontifícia UniReview Board of the Hospital Sa lica do Rio Grande do Sul approved the study's versidade Cato protocol. Age limit was set at 50 years. Reliability of the data was ensured by attention checking questions inserted within the instrument, and throughout the system. Furthermore, volunteers had to deal with specific questions on the degree of attention and sincerity presented while completing the instrument. The initial sample was 16,027 volunteers, only 10,800 passed all of these validity checks.
2.2. Instruments 2.2.1. Assessment of childhood trauma Childhood trauma was evaluated through the CTQ version translated and validated to Brazilian Portuguese (Bernstein et al., 1994; Grassi-Oliveira et al., 2006). This scale is a self-report inventory that can be administrated to adults and adolescents in order to identify traumatic experiences in infancy. Its 25 questions cover 5 types of trauma: physical abuse, sexual abuse, emotional abuse, physical neglect and emotional neglect. Each of these trauma types is dimensionally evaluated through a 5-point Likert scale, ranging from “never” to “frequently”. Thus, these subscales generate scores ranging from 5 (no trauma) to 25 (extreme trauma), and a total scale score varies from 25 (no trauma) to 125 (extreme trauma in all subscales). For the statistical analysis, the total trauma score was divided in quintiles, with the following cut points: 33, 39, 46 and 57. The severity of trauma was categorized according to Tucci et al. (2010): emotional abuse: 8 ¼ minimal, from 9 to 12 ¼ low, 13 to 15 ¼ moderate, >15 ¼ severe; emotional neglect: 9 ¼ minimal, 10 to 14 ¼ low, 15 to 17 ¼ moderate, >17 ¼ severe; physical abuse: 7 ¼ minimal, 8 to 9 ¼ low, 10 to 12 ¼ moderate, >12 ¼ severe; physical neglect: 7 ¼ minimal, 8 to 9 ¼ low, 10 to 12 ¼ moderate, >12 ¼ severe; sexual abuse: 5 ¼ minimal, 6 to 7 ¼ low, 8 to 12 ¼ moderate, >12 ¼ severe.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
2.2.2. Affective and Emotional Composite Temperament Scale (AFECTS) The AFECTS (see Lara et al., 2012a for the complete scale) consists on the following sections: 1) Emotional section: 60 multiple choice questions, with five dimensions composed of 8 items (Volition, Anger, Sensitivity, Coping, Control) and five traits with 4 items (Desire, Fear, Caution, Anxiety and Sensitivity). The questions are scored from 1 to 7 and the total score of each dimension is the sum of the scores of their respective questions. The Global Adaptation Score (GAS) is calculated by adding the adaptive traits according to previous analysis (Lara et al., 2012a) and subtracting the scores of maladaptive traits: Volition þ Coping þ Control þ Caution þ Stability Desire Anxiety Anger Sensitivity Fear. The mean GAS score in this sample was 8.20 ± 60.9, ranging from 156 to 186, with a normal distribution. 2) Affective section: from 12 short descriptions of affective temperaments (depressive, anxious, apathetic, cyclothymic, dysphoric, volatile, obsessive, euthymic, hyperthymic, irritable, disinhibited and euphoric) the subject selects which is most suitable to represent his/her temperament. 2.3. Statistical analysis The statistics of the demographic variables are presented as averages and standard deviation. The scores on the AFECTS0 emotional dimensions for trauma quintiles were analyzed using MANCOVA with Bonferroni's confidence interval, considering sex and age as covariates. The results were presented as averages and 95% confidence interval. In order to analyze how each trauma category relates with specific emotional traits, we conducted a linear regression analysis to calculate beta-values (non-standardized coefficients) and the R2 for men and women separately. Differences in the proportion of affective temperaments according to trauma quintiles were analyzed using chi-square tests for men and women separately. The data was analyzed using SPSS (Statistical Package for the Social Sciences), version 17.0 for Windows. 3. Results The final sample consisted of 10,800 participants (mean age 27.7 ± 7.8 years old, 69.8% women), mostly caucasians with at least a high school degree (Table 1). The most common types of trauma were emotional abuse and negligence, followed by physical abuse
Table 1 Demographic characteristics of the sample (n ¼ 10,800). Males (n ¼ 3257)
Females (n ¼ 7543)
Age 27.7 ± 7.8 28.0 ± 7.7 Ethnicity 66.4/24.8 67.6/24.3 (% caucasian/mixed) Marital status 43.0/24.9 32.6/26.0 (% single/married) Employed/Students (%) 55.6/30.5 50.0/30.3 High school/University 41.3/33.5 35.7/39.6 degree (%) Christian religion/no 49.9/7.3 54.9/6.9 religion (%) Trauma severity (% minimal/low/moderate/severe) Emotional abuse 43/32/12/13 31/30/17/22 Emotional neglect 32/34/16/18 31/33/16/20 Physical abuse 62/17/12/8 64/14/11/11 Physical neglect 65/18/12/5 64/17/12/7 Sexual abuse 74/12/11/3 67/13/13/7
Statistics T ¼ 1.57; p ¼ 0.12 c2 ¼ 7.52; p ¼ 0.11
c2 ¼ 127.18; p < 0.01 c2 ¼ 154.79; p < 0.01 c2 ¼ 51.65; p < 0.01 c2 ¼ 157.48; p < 0.01 t t t t t
¼ ¼ ¼ ¼ ¼
15.36; p < 0.01 2.77; p < 0.05 1.35; p ¼ 0.18 2.50; p < 0.05 9.49; p < 0.01
3
and negligence, and sexual abuse was the least common. Except for physical abuse, all types of trauma were significantly more frequent in women, especially emotional and sexual abuse (Table 1). As shown in Fig. 1, the degree of trauma measured as the total CTQ score was inversely associated with the traits of Volition (F ¼ 359,25, p < 0.001 for all), Stability (F ¼ 327,55), Coping (F ¼ 240,40), Control (F ¼ 214,39) and, to a lesser degree, Caution (F ¼ 79,83). In contrast, more exposure to trauma was related to higher scores on Sensitivity (F ¼ 288,31), Anxiety (F ¼ 271,13), Anger (F ¼ 205,96) and, to a lesser degree, Desire (F ¼ 49,21) and Fear (F ¼ 42,40) traits. Linear regression analysis showed that emotional abuse and neglect were strongly associated with lower Volition, Coping, Control and Stability and higher Sensitivity, Anxiety and Anger in both sexes, although slightly more pronounced in men (Table 2). The associations with sexual abuse were generally in the same direction, being moderate in men and weaker in women. Surprisingly, in the opposite direction of emotional traumas, physical abuse and neglect were positively associated with Volition and Coping scores, while negatively related to Sensitivity, Fear and Anxiety, particularly in men. The GAS score, which represents the adaptive combination of all traits, confirmed that emotional and sexual trauma were more strongly associated with dysfunction in men, but there was also a favorable association with physical trauma. The negative associations with emotional trauma and positive associations with physical trauma were more pronounced in males. We further analyzed the associations of physical abuse with Volition, Coping and Fear in men by dividing total emotional trauma (abuse and neglect) in quintiles and physical abuse in tertiles, given their respective distributions (Fig. 2A). This analysis evidenced the pattern of increment in Volition and Coping and decline in Fear with increasing physical abuse in men (Fig. 2B). It is noteworthy that the differences in these traits in relation to emotional trauma are very similar to the results of total trauma shown in Fig. 1. Regarding affective temperaments, lower frequencies of euthymic and hyperthymic types in both sexes and obsessive type in females were found with higher trauma scores. This was compensated by a higher proportion of the depressive and cyclothymic types in both sexes. We also observed a lower proportion of the apathetic type in the 1sttrauma quintile and a higher frequency of euphoric in the 5th quintile in both sexes (Fig. 3). 4. Discussion The results from this Brazilian community sample evidenced that childhood trauma is associated with a global maladaptive profile. In general, adaptive cognitive and emotional dimensional traits (e.g. volition, coping) and stable affective temperaments (e.g. euthymic and hyperthymic) were negatively associated with total trauma, whereas dysfunctional traits (e.g. sensitivity and anger) and affective types (e.g. cyclothymic and depressive) were proportional to traumatic experiences. However, different types of trauma yielded complex results, with emotional abuse and neglect being associated with less adaptive profile. In contrast, physical abuse and neglect were linked to a few favorable traits, particularly in men. The study of the impact of childhood trauma on human personality is methodologically challenging. Long term prospective studies are considered the best design to establish causeconsequence relationships, but the reliable assessment of traumatic events is hampered by the incomplete disclosure of sensitive issues, since data cannot be collected anonymously. Also, studies rarely used computer based assessments, which have been shown to provide around 5 times more information on sensitive issues
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
4
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Fig. 1. Dimensional traits according to childhood trauma quintiles. Results are shown as mean ± 95%CI of the sum of all five trauma subscales of the CTQ.
Table 2 Linear regression of traits versus specific childhood traumas by gender.
Males Volition Control Coping Sensitivity Anger Stabilitya Cautiona Feara Anxietya Desirea GAS Females Volition Control Coping Sensitivity Anger Stabilitya Cautiona Feara Anxietya Desirea GAS
EmAb
EmNe
SeAb
PhAb
PhNe
R2
¡.590 ¡.586 ¡.538 .731 .598 ¡.776 .208 .314 .862 .348 4.297
¡.798 .382 ¡.576 .395 .273 .398 .098 .306 .480 .008 3.061
.247 .224 .345 .288 .125 .240 .132 .130 .156 .172 1.644
.543 .147 .447 .356 .072 .004 .262 ¡.576 .284 .236 1.599
.381 .404 .355 .260 .143 .278 .070 .106 .362 .054 1.923
.16 .10 .11 .13 .10 .14 .03 .05 .14 .04 .20
¡.536 ¡.542 .493 .606 .469 ¡.674 .342 .104 .594 .398 3.701
.676 .265 ¡.545 .282 .254 .312 .014 .158 .388 .132 2.443
.042 .143 .086 .054 .026 .076 .198 .068 .032 .164 .552
.279 .136 .281 .171 .100 .004 .118 .206 .104 .146 .789
.207 .175 .281 .179 .156 .028 .092 .050 .224 .116 1.253
.15 .09 .11 .12 .08 .12 .04 .02 .10 .04 .18
Bold: >0.5 or 0.05), gender and age as covariates. EmAb ¼ emotional abuse, EmNe ¼ emotional neglect, PhAb ¼ physical abuse, PhNe ¼ physical neglect, SeNe ¼ sexual neglect, GAS ¼ global adaptation score. a Multiplied by 2 (traits with 4 items).
than the pen-and-paper format (Turner et al., 1998). Thus, it is difficult to combine the ideal method of data collection on traumatic events (anonymous computer-based self-report) with prospective studies starting during childhood. Another important issue is that, since the different types of trauma tend to co-occur, it is difficult to obtain samples with only one type of trauma, especially only physical abuse with absence of emotional abuse (in this sample, R ¼ 0.49, Spearman test), as well as physical without emotional neglect (R ¼ 0.56). Finally, animal models fall short in validity for emotional abuse, which was the factor most strongly associated with a dysfunctional profile in our study and in previous reports (Spertus et al., 2003; Zhang et al., 2012). The results on the negative impact of emotional and sexual trauma on mental health are in line with several previous studies and with the “cumulative” stress hypothesis, in which adverse experiences early in life predispose individuals to be more vulnerable to adversity later in life (Nederhof and Schmidt, 2012). Unfortunately, the design of this study does not include data on recent stressful events, which would allow a further understanding of the adaptive value of childhood trauma during adulthood. We also have observed that the detrimental associations with emotional abuse and neglect were stronger in men than in women, and the opposite was found for favorable correlations with physical abuse and neglect. Samplin et al. (2013) showed an association between emotional abuse and gender, demonstrating that a history of traumatic emotional experiences was related to a reduced hippocampal volume in males, but not in females. Also, being female was a predictor of resilience in adolescence and young adulthood
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
5
Fig. 2. Specific contributions of emotional trauma and physical abuse regarding volition, coping and fear scores. A. data distribution of emotional trauma scores (emotional abuse þ emotional neglect). B. data distribution of physical abuse. Relationship of emotional trauma quintiles and physical abuse with Volition (C), Coping (D) and Fear traits (E). Results are shown as mean ± 95%CI.
(DuMont et al., 2007) and females were shown to have higher rates of resilience than men for all types of childhood trauma (McGloin and Widom, 2001). On the other hand, Haatainen et al. (2003) showed that adverse childhood experiences are more strongly related with hopelessness among adult women than men. The most surprising result was the adaptively favorable association between physical abuse and neglect during childhood with temperament traits. One possibility is that studies which only address psychopathology are less likely to uncover putative “beneficial” effects of physical punishment since they do not address positive attributes and fail to test for the interactions between types of traumatic events. Importantly, Li et al. (2014), also using a linear regression analysis, found that physical abuse was associated with lower neuroticism, whereas emotional maltreatment was associated with higher neuroticism and psychoticism. They attributed their results to cultural concepts that physical discipline is regarded by many Chinese (including children) as necessary, in agreement with a famous Chinese proverb “no beating, no success”. This possibility seems less likely given the apparent differences between Chinese and Brazilian cultures and the similar results of both studies. Moreover, a volunteer does not answer a questionnaire on personality traits taking into account one's answers on a questionnaire on traumatic experiences or vice versa. However, the present explanatory models do not seem to contemplate the “positive” effects of physical abuse and neglect during childhood. For example, the stress inoculation hypothesis suggests that relatively mild early-life adversity may promote resilience and prepare for similar challenges in later-life (Daskalakis et al., 2013), but our results showed stronger associations with higher physical adversity scores. In contrast, the mismatch
hypothesis states that high levels of psychosocial stress early in life may program for dealing with high psychosocial stress in later life, i.e., the costs or benefits of a past experience depend on how it matches with the present context (Nederhof and Schmidt, 2012). The lack of data on recent events prevents a specific evaluation of this model, but it is nonetheless difficult to reconcile the opposite associations of emotional and physical maltreatment with temperament traits. A third perspective is the differential susceptibility hypothesis, which posits that some individuals are more susceptible than others to both negative (risk-promoting) and positive (development-enhancing) environmental conditions (Ellis et al., 2011). Again, this model does not account for the positive effects of adverse physical experiences. Thus, a new proposal seems necessary. The most well known source on this subject is Nietzsche's quote “From life's school of war: what does not kill (or destroy) me makes me stronger” from his book Twilight of the Idols or How to Philosophize with a Hammer, written in 1888. Our data supports this principle only regarding physical maltreatment, whereas the effects of emotional and sexual traumas are in line with cumulative stress hypothesis, i.e., what does not kill me makes me weaker. Therefore, we suggest that the “context-dependent nature” of stress may play a significant role in the subject of stress and adaptation. From an evolutionary perspective, physical adversity has exerted selective pressure over several million years, i.e., the body is “used to” harshness and is able to recover relatively quickly. In contrast, emotional abuse was only possible after the development of language many thousands of years ago and the meaning of a harsh affirmation may remain for much longer, producing more impact on self-esteem at the trait level. A second reason may derive
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
6
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Fig. 3. Distribution of affective temperaments according to childhood trauma quintiles. Data is shown as percentage of subjects in each category for males (top) and females (bottom) separately. * ¼ a higher proportion and # ¼ a lower proportion within each affective type (Chi-square test, p < 0.05).
from a “corrective” intention behind physical punishment from an authority figure, which is not as evident in insults from a parent towards his or her child. Finally, one should be cautious not to extrapolate this data to physical defeat among peers, which refers to a different context. Our study has significant limitations to be acknowledged. The current sample was comprised primarily of highly educated persons in the Brazilian population and other selection biases may have contributed to the volunteers' motivation to participate. The generalization of the current findings to other cultures and contexts of adversity will depend on future research. A second limitation was the reliance on a single method of data collection (i.e., selfreport questionnaires). Generally, a multi-modal approach to assessment is desirable in assessing trauma history and current emotional and physical functioning. Furthermore, assessments of child and adult trauma exposure were retrospective and crosssectional, preventing any firm conclusions on the direction of the effect (personality traits influencing memory of childhood trauma or childhood trauma affecting personality traits) or the weight of each direction. Prospective studies would address more directly the possibility of causal link and allow for more detailed exploration of mediators and moderators of these relationships. Regarding the association between physical abuse/neglect and more positive personality traits, it is possible that it has been overestimated due to a selection bias. Community members who have “survived” childhood physical abuse/neglect experiences and made it in life might be more willing to participate in such a study than unexposed healthy individuals or individuals with childhood trauma but less beneficial outcomes in adulthood. Other unhidden factors associated with participation and non-participation in this online study (and not physical abuse/neglect alone) may also be
responsible for the positive personality outcomes. Further studies should also address different contexts of physical punishment (e.g. if related to a misbehavior or not). The major strengths were the large sample and the anonymous assessment. Also, despite being cross-sectional, which is prone to memory bias, 18 of the 25 questions of CTQ are about objective situations (e.g. being called names or spanked), not their subjective impact. Finally, the BRAINSTEP has many scales and subjects are unaware of study hypothesis; they voluntarily participate as a means to gain knowledge on their personal profile. In conclusion, total trauma was associated with dysfunctional emotional traits and affective temperaments, in agreement with studies using other designs and methods to evaluate trauma and personality traits. The adverse temperament profiles were related mainly to emotional trauma and, in a lesser degree, to sexual trauma. Positive attributes such as volition, coping and low fear traits were associated with physical abuse and neglect, particularly in men. This suggests that physical trauma could even strengthen individuals and that emotional trauma would be mostly linked to psychiatric and personality disorders. As a society, by focusing on physical trauma against children, which can be assessed objectively, we may be ignoring the real harm of emotional abuse and neglect, which is often undetected due to their subjective nature. Other studies using different methods are necessary to deepen the knowledge on this subject. Role of funding source The funding agency had no influence on study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Contributors DRL designed the study and worked on the implementation of the system. RS, HWC and DRL undertook the statistical analysis, PM and IK managed the literature searches, and all authors contributed in writing the manuscript. All authors have approved the final manuscript. Conflict of interest The authors declare no conflicts of interest. Acknowledgments This research was funded by PRONEX-FAPERGS (10/0055-0). DRL is a CNPq research fellow, PM and IK received a graduate student fellowship from FAPERGS and CNPq. References Afifi TO, Mather A, Boman J, Fleisher W, Enns MW, MacMillan H, et al. Childhood adversity and personality disorders: results from a nationally representative population-based study. J Psychiatr Res 2011;45:814e22. Allen B, Lauterbach D. Personality characteristics of adult survivors of childhood trauma. J Trauma Stress 2007;20:587e95. ria JU, Gigante LP, Figueiredo ACL, Aerts DRGC, et al. Child Bassini DG, Palazzo LS, Be sexual abuse in southern Brazilian and associated factors: a population-based study. BMC Public Health 2009;9:133. Bernstein DP, Fink L, Handelsman L, Foote J, Lovejoy M, Wenzel K, et al. Initial reliability and validity of a new retrospective measure of child abuse and neglect. Am J Psychiatry 1994;151:1132e6. Bornstein MH, Lamb ME. Developmental science: an advanced text book. 6th ed. New York: Psychological Press; 2010. Briere J, Elliott DM. Prevalence and psychological sequelae of self-reported childhood physical and sexual abuse in a general population of men and women. Child Abus Negl 2003;27:1205e22. Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model of temperament and character. Arch Gen Psychiatry 1993;50:975. rbara JFRS, Santos CAST, Gomes WA, de Souza HL. Costa MCO, de Carvalho RC, Ba ^ncia contra crianças e adolescentes, segundo registro de conO perfil da viole ~es de viole ^ncia. Cie ^ncia Saúde selhos tutelares: vítimas, agressores e manifestaço Coletiva 2007;12:1129e41. Couper MP. Review: web surveys: a review of issues and approaches. Public Opin Q 2000:464e94. Daskalakis NP, Bagot RC, Parker KJ, Vinkers CH, De Kloet E. The three-hit concept of vulnerability and resilience: toward understanding adaptation to early-life adversity outcome. Psychoneuroendocrinology 2013;38:1858e73. De Carvalho HW, Bisol LW, Ottoni LG, Lara DR. The affective and emotional temperament composite scale: psychometric analysis including anxiety and instability subscales. In: Abstract book of the third international conference on neurobiology, psychopharmacology and treatment guidance; 2013. p. 176. DuMont KA, Widom CS, Czaja SJ. Predictors of resilience in abused and neglected children grown-up: the role of individual and neighborhood characteristics. Child Abus Negl 2007;31:255e74. Ellis BJ, Boyce WT, Belsky J, Bakermans-Kranenburg MJ, van Ijzendoorn MH. Differential susceptibility to the environment: an evolutionaryneurodevelopmental theory. Dev Psychopathol 2011;23:7e28.
7
Flory JD, Yehuda R, Grossman R, New AS, Mitropoulou V, Siever LJ. Childhood trauma and basal cortisol in people with personality disorders. Compr Psychiatry 2009;50:34e7. Frizzo MN, Bisol LW, Lara DR. Bullying victimization is associated with dysfunctional emotional traits and affective temperaments. J Affect Disord 2013;148: 48e52. Gosling SD, Vazire S, Srivastava S, John OP. Should we trust web-based studies? A comparative analysis of six preconceptions about internet questionnaires. Am Psychol 2004;59:93. Grassi-Oliveira R, Stein LM, Pezzi JC. Translation and content validation of the Childhood Trauma Questionnaire into Portuguese language. Rev Saude Publica 2006;40:249e55. € J, Honkalampi K, Koivumaa-Honkanen H, Hintikka J, Haatainen K, Tanskanen A, Kylma et al. Gender differences in the association of adult hopelessness with adverse childhood experiences. Soc Psychiatry Psychiatr Epidemiol 2003;38:12e7. Heins M, Simons C, Lataster T, Pfeifer S, Versmissen D, Lardinois M, et al. Childhood trauma and psychosis: a case-control and case-sibling comparison across different levels of genetic liability, psychopathology, and type of trauma. Am J Psychiatry 2011;168:1286e94. Lara DR, Akiskal HS. Toward an integrative model of the spectrum of mood, behavioral and personality disorders based on fear and anger traits: II. Implications for neurobiology, genetics and psychopharmacological treatment. J Affect Disord 2006;94:89e103. Lara DR, Bisol LW, Brunstein MG, Reppold CT, de Carvalho HW, Ottoni GL. The Affective and Emotional Composite Temperament (AFECT) model and scale: a system-based integrative approach. J Affect Disord 2012a;140:14e37. Lara DR, Ottoni GL, Brunstein MG, Frozi J, de Carvalho HW, Bisol LW. Development and validity data of the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP). J Affect Disord 2012b;141:390e8. Lara DR, Pinto O, Akiskal K, Akiskal HS. Toward an integrative model of the spectrum of mood, behavioral and personality disorders based on fear and anger traits: I. Clinical implications. J Affect Disord 2006;94:67e87. Li X, Wang Z, Hou Y, Wang Y, Liu J, Wang C. Effects of childhood trauma on personality in a sample of Chinese adolescents. Child Abus Negl 2014;38:788e96. Mathews G, Deary IJ, Whiteman MC. Personality traits. 2nd ed. Cambridge: Cambridge University Press; 2003. McGloin J, Widom CS. Resilience among abused and neglected children grown up. Dev Psychopathol 2001;13:1021e38. Nederhof E, Schmidt MV. Mismatch or cumulative stress: toward an integrated hypothesis of programming effects. Physiol Behav 2012;106:691e700. Rademaker AR, Vermetten E, Geuze E, Muilwijk A, Kleber RJ. Self-reported early trauma as a predictor of adult personality: a study in a military sample. J Clin Psychol 2008;64:863e75. Samplin E, Ikuta T, Malhotra AK, Szeszko PR, DeRosse P. Sex differences in resilience to childhood maltreatment: effects of trauma history on hippocampal volume, general cognition and subclinical psychosis in healthy adults. J Psychiatr Res 2013;47:1174e9. Sansone RA, Pole M, Dakroub H, Butler M. Childhood trauma, borderline personality symptomatology, and psychophysiological and pain disorders in adulthood. Psychosomatics 2006;47:158e62. Spertus IL, Yehuda R, Wong CM, Halligan S, Seremetis SV. Childhood emotional abuse and neglect as predictors of psychological and physical symptoms in women presenting to a primary care practice. Child Abuse Negl 2003;27(11): 1247e58. ^a F, Souza-Formigoni MLO. Childhood trauma in substance use Tucci AM, Kerr-Corre disorder and depression: an analysis by gender among a Brazilian clinical sample. Child Abus Negl 2010;34:95e104. Turner CF, Ku L, Rogers SM, Lindberg LD, Pleck JH, Sonenstein FL. Adolescent sexual behavior, drug use, and violence: increased reporting with computer survey technology. Science 1998;280:867e73. Widom CS, DuMont K, Czaja SJ. A prospective investigation of major depressive disorder and comorbidity in abused and neglected children grown up. Arch Gen Psychiatry 2007;64:49e56. Zhang T, Chow A, Wang L, Dai Y, Xiao Z. Role of childhood traumatic experience in personality disorders in China. Compr Psychiatry 2012;53(6):829e36.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
Contents lists available at ScienceDirect
Journal of Psychiatric Research journal homepage: www.elsevier.com/locate/psychires
What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits Roberto Sudbrack a, Pedro H. Manfro b, Isadora M. Kuhn b, Hudson W. de Carvalho c, Diogo R. Lara a, b, * ~o Lucas da Pontifícia Universidade Cato lica do Rio Grande do Sul, Porto Alegre, Brazil Serviço de Psiquiatria, Hospital Sa lica do Rio Grande do Sul, Porto Alegre, Brazil Faculdade de Bioci^ encias, Pontifícia Universidade Cato c Departmento de Psicologia, Universidade Federal de Pelotas (UFPel), Pelotas, RS, Brazil a
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 September 2014 Received in revised form 27 November 2014 Accepted 8 January 2015
Childhood trauma arises from damaging experiences and the absence of a secure and reliable environment for child development. Despite many studies on the risk for developing psychiatric disorders, much less is known on the relationship between traumatic experiences and personality traits. A total of 10,800 participants (mean age 27.7 ± 7.8 years old, 69.8% women) anonymously answered the Childhood Trauma Questionnaire (CTQ) and the Affective and Emotional Composite Temperament Scale (AFECTS) in the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP). The results showed that total trauma score was associated with lower levels of positive traits (volition, control, coping, and stability), higher scores of negative traits (sensitivity, anger, anxiety) and a higher frequency of depressive and cyclothymic temperaments. Linear regression analysis showed similar findings for emotional abuse and neglect, whereas physical abuse and neglect were positively associated with more volition and coping as well as less fear, sensitivity and anxiety traits, especially in men. This study has a cross-sectional design and the sole use of self-reporting as the main limitations. In conclusion, dysfunctional temperament profiles were associated mainly with emotional trauma and sexual abuse. Some adaptive traits were associated with physical abuse and neglect, especially in men. These results reinforce the negative impact of emotional trauma and people's resilience to physical trauma, which may even translate into adaptive trait expression in males particularly. We propose that the “contextdependent nature” of stress should be further studied to break down the influence of specific types and contexts of adversity on psychological and psychiatric outcomes. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Childhood trauma Abuse Neglect Temperament Traits Personality
1. Introduction Childhood trauma arises from harmful experiences and from the absence of a supporting and reliable environment that fosters child development. Early traumatic events have been widely associated with a diversity of negative outcomes related to psychological maladjustment and psychiatric conditions, such as mood and personality disorders (Afifi et al., 2011; Allen and Lauterbach, 2007; Flory et al., 2009; Heins et al., 2011; McGloin and Widom, 2001; Sansone et al., 2006; Tucci et al., 2010; Widom et al., 2007). Nevertheless, these evidence are mainly based on case-control
^ncias e PUCRS, Av. Ipiranga, 6681 e * Corresponding author. Faculdade de Biocie Pd12A, Porto Alegre, RS 90619-900, Brazil. Fax: þ55 51 33203612. E-mail address: [email protected] (D.R. Lara).
study designs, which are known to be subjective to selection biases (Gordis, 2010), as clinical/case groups tend to be composed of treatment-seeking individuals, which are not representative of all cases within the trauma exposed population and controls that either healthy or have been exposed to very few traumatic events. Also, case-control methodologies are more suitable for rare conditions, which is not the case for abuse and negligence exposure in childhood (Bassani et al., 2009; Briere and Elliott, 2003; Costa et al., 2007). Another possible limitation regarding abuse and neglect research is unreliable reporting. Most studies rely on classic paperand-pencil self-report, telephone or face-to-face interviews and, although scientific work guarantees anonymity in publications, participants may feel vulnerable and exposed when dealing directly with research staff. In this regard, computer-based data collection has important advantages over the aforementioned
http://dx.doi.org/10.1016/j.jpsychires.2015.01.001 0022-3956/© 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
2
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
methodologies even when both methods are applied anonymously (Gosling et al., 2004; Turner et al., 1998; Couper, 2000). Moreover, volunteers voluntarily and anonymously accessing their private computers probably feel even less concerned about how they appear to others. This is particularly relevant when answering questions on sensitive issues, such as problems of intimate relationships, illicit drug use, sexual experiences, and traumatic events. In addition, the Internet allows the collection of large research samples at a low cost and with little data loss. Thus, internet assisted methods can enhance response reliability and the acquisition of a diverse sample, although selection biases also apply. Despite the common understanding that traumatic life events and psychological maladjustment are intimately connected, few studies have addressed the relationship between childhood traumatic experiences and personality traits. This gap result from the emphasis on categorical diagnosis in the last few decades and from the view that personality is a construct designed mostly to index normal and social behavioral liabilities (Bornstein et al., 2010; Mathews et al., 2003). Nevertheless, using the Five Factor Model of personality, Allen and Lauterbach (2007) reported that people who have gone through early life trauma present higher scores on neuroticism and openness to new experience than controls. Furthermore, individuals who reported a single traumatic experience in early life showed lower scores than controls on interpersonal dependency, while individuals who reported repeated and prolonged traumatic involvement showed higher scores on interpersonal dependency than controls. Rademaker et al. (2008), using Cloninger Psychobiological Model of Personality (Cloninger et al., 1993), showed significant negative associations between traumatic events and character dimensions of self-directedness and cooperativeness in a military sample, which emphasized the deleterious relationship between trauma and psychological traits. Lastly, Li et al. (2014) found significant positive correlations between emotional and sexual abuse scores in the Childhood Trauma Checklist with neuroticism and psychoticism from Eysenck's personality model. Taken together, these findings suggest that traumatic events not only influence psychopathologies, but also have deleterious effects on traits. We have recently developed the Affective and Emotional Composite Temperament (AFECT) model (Lara et al., 2012a), an integrative temperament model that combines specific dimensional traits and global affective temperaments. It is the evolution of the previous “fear and anger” model (Lara and Akiskal, 2006; Lara et al., 2006). The AFECT model comprises twelve affective temperaments (depressive, anxious, apathetic, cyclothymic, dysphoric, volatile, obsessive, euthymic, hyperthymic, irritable, disinhibited and euphoric), which represent specific configurations of the ten emotional traits (volition, desire, anger, fear, caution, emotional sensitivity, coping, control, and the two recently added traits of anxiety and stability). Using the AFECT scale (AFECTS; Lara et al., 2012a; De Carvalho et al., 2013), we recently showed a global impairment of these traits proportionally to time of exposure to bullying. Bullying victims showed lower Volition, Coping and Control scores, and more Emotional Sensitivity, Anger and Fear (Frizzo et al., 2013). Regarding affective temperaments, being bullied was associated with a lower frequency of the healthy euthymic and hyperthymic types, and a higher proportion of depressive and cyclothymic types. The aim of the current study is to advance in the understanding of how childhood trauma relates to emotional traits and affective temperaments using a different methodology from other studies. More specifically, we evaluated the relationship between the AFECTS parameters with indicators of childhood trauma
(emotional, physical and sexual) assessed retrospectively in a crosssectional web survey. Our initial hypothesis was that all types of childhood trauma would be associated with maladaptive scores in emotional traits and types of affective temperaments in a similar fashion as observed with being bullied.
2. Methods 2.1. Participants The BRAINSTEP web survey (Lara et al., 2012b) was broadcasted in National media (TV news, major newspapers) for recruiting volunteers (18 yrs or older) to answer several psychiatric and psychological scales and questionnaires. The participation and honesty of volunteers was stimulated by the incentive of receiving a brief report on their temperament profile (i.e., traits levels of volition, fear, anger, control and sensitivity, as well as a description of their affective temperament), and the chance of having a psychiatric disorder based on their answers to screening instruments for 19 disorders. Before responding to the questionnaires included in the site (www.temperamento.com.br), respondents gave their electronically informed consent. Their participation was voluntary and they could cancel it at any moment, with no justification necessary. This form fulfills the requirements of the National Health Council of Brazil (Resolution 196/1996) and the Code of Ethics of the World Medical Association (Declaration of Helsinki). The Institutional ~o Lucas from Pontifícia UniReview Board of the Hospital Sa lica do Rio Grande do Sul approved the study's versidade Cato protocol. Age limit was set at 50 years. Reliability of the data was ensured by attention checking questions inserted within the instrument, and throughout the system. Furthermore, volunteers had to deal with specific questions on the degree of attention and sincerity presented while completing the instrument. The initial sample was 16,027 volunteers, only 10,800 passed all of these validity checks.
2.2. Instruments 2.2.1. Assessment of childhood trauma Childhood trauma was evaluated through the CTQ version translated and validated to Brazilian Portuguese (Bernstein et al., 1994; Grassi-Oliveira et al., 2006). This scale is a self-report inventory that can be administrated to adults and adolescents in order to identify traumatic experiences in infancy. Its 25 questions cover 5 types of trauma: physical abuse, sexual abuse, emotional abuse, physical neglect and emotional neglect. Each of these trauma types is dimensionally evaluated through a 5-point Likert scale, ranging from “never” to “frequently”. Thus, these subscales generate scores ranging from 5 (no trauma) to 25 (extreme trauma), and a total scale score varies from 25 (no trauma) to 125 (extreme trauma in all subscales). For the statistical analysis, the total trauma score was divided in quintiles, with the following cut points: 33, 39, 46 and 57. The severity of trauma was categorized according to Tucci et al. (2010): emotional abuse: 8 ¼ minimal, from 9 to 12 ¼ low, 13 to 15 ¼ moderate, >15 ¼ severe; emotional neglect: 9 ¼ minimal, 10 to 14 ¼ low, 15 to 17 ¼ moderate, >17 ¼ severe; physical abuse: 7 ¼ minimal, 8 to 9 ¼ low, 10 to 12 ¼ moderate, >12 ¼ severe; physical neglect: 7 ¼ minimal, 8 to 9 ¼ low, 10 to 12 ¼ moderate, >12 ¼ severe; sexual abuse: 5 ¼ minimal, 6 to 7 ¼ low, 8 to 12 ¼ moderate, >12 ¼ severe.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
2.2.2. Affective and Emotional Composite Temperament Scale (AFECTS) The AFECTS (see Lara et al., 2012a for the complete scale) consists on the following sections: 1) Emotional section: 60 multiple choice questions, with five dimensions composed of 8 items (Volition, Anger, Sensitivity, Coping, Control) and five traits with 4 items (Desire, Fear, Caution, Anxiety and Sensitivity). The questions are scored from 1 to 7 and the total score of each dimension is the sum of the scores of their respective questions. The Global Adaptation Score (GAS) is calculated by adding the adaptive traits according to previous analysis (Lara et al., 2012a) and subtracting the scores of maladaptive traits: Volition þ Coping þ Control þ Caution þ Stability Desire Anxiety Anger Sensitivity Fear. The mean GAS score in this sample was 8.20 ± 60.9, ranging from 156 to 186, with a normal distribution. 2) Affective section: from 12 short descriptions of affective temperaments (depressive, anxious, apathetic, cyclothymic, dysphoric, volatile, obsessive, euthymic, hyperthymic, irritable, disinhibited and euphoric) the subject selects which is most suitable to represent his/her temperament. 2.3. Statistical analysis The statistics of the demographic variables are presented as averages and standard deviation. The scores on the AFECTS0 emotional dimensions for trauma quintiles were analyzed using MANCOVA with Bonferroni's confidence interval, considering sex and age as covariates. The results were presented as averages and 95% confidence interval. In order to analyze how each trauma category relates with specific emotional traits, we conducted a linear regression analysis to calculate beta-values (non-standardized coefficients) and the R2 for men and women separately. Differences in the proportion of affective temperaments according to trauma quintiles were analyzed using chi-square tests for men and women separately. The data was analyzed using SPSS (Statistical Package for the Social Sciences), version 17.0 for Windows. 3. Results The final sample consisted of 10,800 participants (mean age 27.7 ± 7.8 years old, 69.8% women), mostly caucasians with at least a high school degree (Table 1). The most common types of trauma were emotional abuse and negligence, followed by physical abuse
Table 1 Demographic characteristics of the sample (n ¼ 10,800). Males (n ¼ 3257)
Females (n ¼ 7543)
Age 27.7 ± 7.8 28.0 ± 7.7 Ethnicity 66.4/24.8 67.6/24.3 (% caucasian/mixed) Marital status 43.0/24.9 32.6/26.0 (% single/married) Employed/Students (%) 55.6/30.5 50.0/30.3 High school/University 41.3/33.5 35.7/39.6 degree (%) Christian religion/no 49.9/7.3 54.9/6.9 religion (%) Trauma severity (% minimal/low/moderate/severe) Emotional abuse 43/32/12/13 31/30/17/22 Emotional neglect 32/34/16/18 31/33/16/20 Physical abuse 62/17/12/8 64/14/11/11 Physical neglect 65/18/12/5 64/17/12/7 Sexual abuse 74/12/11/3 67/13/13/7
Statistics T ¼ 1.57; p ¼ 0.12 c2 ¼ 7.52; p ¼ 0.11
c2 ¼ 127.18; p < 0.01 c2 ¼ 154.79; p < 0.01 c2 ¼ 51.65; p < 0.01 c2 ¼ 157.48; p < 0.01 t t t t t
¼ ¼ ¼ ¼ ¼
15.36; p < 0.01 2.77; p < 0.05 1.35; p ¼ 0.18 2.50; p < 0.05 9.49; p < 0.01
3
and negligence, and sexual abuse was the least common. Except for physical abuse, all types of trauma were significantly more frequent in women, especially emotional and sexual abuse (Table 1). As shown in Fig. 1, the degree of trauma measured as the total CTQ score was inversely associated with the traits of Volition (F ¼ 359,25, p < 0.001 for all), Stability (F ¼ 327,55), Coping (F ¼ 240,40), Control (F ¼ 214,39) and, to a lesser degree, Caution (F ¼ 79,83). In contrast, more exposure to trauma was related to higher scores on Sensitivity (F ¼ 288,31), Anxiety (F ¼ 271,13), Anger (F ¼ 205,96) and, to a lesser degree, Desire (F ¼ 49,21) and Fear (F ¼ 42,40) traits. Linear regression analysis showed that emotional abuse and neglect were strongly associated with lower Volition, Coping, Control and Stability and higher Sensitivity, Anxiety and Anger in both sexes, although slightly more pronounced in men (Table 2). The associations with sexual abuse were generally in the same direction, being moderate in men and weaker in women. Surprisingly, in the opposite direction of emotional traumas, physical abuse and neglect were positively associated with Volition and Coping scores, while negatively related to Sensitivity, Fear and Anxiety, particularly in men. The GAS score, which represents the adaptive combination of all traits, confirmed that emotional and sexual trauma were more strongly associated with dysfunction in men, but there was also a favorable association with physical trauma. The negative associations with emotional trauma and positive associations with physical trauma were more pronounced in males. We further analyzed the associations of physical abuse with Volition, Coping and Fear in men by dividing total emotional trauma (abuse and neglect) in quintiles and physical abuse in tertiles, given their respective distributions (Fig. 2A). This analysis evidenced the pattern of increment in Volition and Coping and decline in Fear with increasing physical abuse in men (Fig. 2B). It is noteworthy that the differences in these traits in relation to emotional trauma are very similar to the results of total trauma shown in Fig. 1. Regarding affective temperaments, lower frequencies of euthymic and hyperthymic types in both sexes and obsessive type in females were found with higher trauma scores. This was compensated by a higher proportion of the depressive and cyclothymic types in both sexes. We also observed a lower proportion of the apathetic type in the 1sttrauma quintile and a higher frequency of euphoric in the 5th quintile in both sexes (Fig. 3). 4. Discussion The results from this Brazilian community sample evidenced that childhood trauma is associated with a global maladaptive profile. In general, adaptive cognitive and emotional dimensional traits (e.g. volition, coping) and stable affective temperaments (e.g. euthymic and hyperthymic) were negatively associated with total trauma, whereas dysfunctional traits (e.g. sensitivity and anger) and affective types (e.g. cyclothymic and depressive) were proportional to traumatic experiences. However, different types of trauma yielded complex results, with emotional abuse and neglect being associated with less adaptive profile. In contrast, physical abuse and neglect were linked to a few favorable traits, particularly in men. The study of the impact of childhood trauma on human personality is methodologically challenging. Long term prospective studies are considered the best design to establish causeconsequence relationships, but the reliable assessment of traumatic events is hampered by the incomplete disclosure of sensitive issues, since data cannot be collected anonymously. Also, studies rarely used computer based assessments, which have been shown to provide around 5 times more information on sensitive issues
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
4
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Fig. 1. Dimensional traits according to childhood trauma quintiles. Results are shown as mean ± 95%CI of the sum of all five trauma subscales of the CTQ.
Table 2 Linear regression of traits versus specific childhood traumas by gender.
Males Volition Control Coping Sensitivity Anger Stabilitya Cautiona Feara Anxietya Desirea GAS Females Volition Control Coping Sensitivity Anger Stabilitya Cautiona Feara Anxietya Desirea GAS
EmAb
EmNe
SeAb
PhAb
PhNe
R2
¡.590 ¡.586 ¡.538 .731 .598 ¡.776 .208 .314 .862 .348 4.297
¡.798 .382 ¡.576 .395 .273 .398 .098 .306 .480 .008 3.061
.247 .224 .345 .288 .125 .240 .132 .130 .156 .172 1.644
.543 .147 .447 .356 .072 .004 .262 ¡.576 .284 .236 1.599
.381 .404 .355 .260 .143 .278 .070 .106 .362 .054 1.923
.16 .10 .11 .13 .10 .14 .03 .05 .14 .04 .20
¡.536 ¡.542 .493 .606 .469 ¡.674 .342 .104 .594 .398 3.701
.676 .265 ¡.545 .282 .254 .312 .014 .158 .388 .132 2.443
.042 .143 .086 .054 .026 .076 .198 .068 .032 .164 .552
.279 .136 .281 .171 .100 .004 .118 .206 .104 .146 .789
.207 .175 .281 .179 .156 .028 .092 .050 .224 .116 1.253
.15 .09 .11 .12 .08 .12 .04 .02 .10 .04 .18
Bold: >0.5 or 0.05), gender and age as covariates. EmAb ¼ emotional abuse, EmNe ¼ emotional neglect, PhAb ¼ physical abuse, PhNe ¼ physical neglect, SeNe ¼ sexual neglect, GAS ¼ global adaptation score. a Multiplied by 2 (traits with 4 items).
than the pen-and-paper format (Turner et al., 1998). Thus, it is difficult to combine the ideal method of data collection on traumatic events (anonymous computer-based self-report) with prospective studies starting during childhood. Another important issue is that, since the different types of trauma tend to co-occur, it is difficult to obtain samples with only one type of trauma, especially only physical abuse with absence of emotional abuse (in this sample, R ¼ 0.49, Spearman test), as well as physical without emotional neglect (R ¼ 0.56). Finally, animal models fall short in validity for emotional abuse, which was the factor most strongly associated with a dysfunctional profile in our study and in previous reports (Spertus et al., 2003; Zhang et al., 2012). The results on the negative impact of emotional and sexual trauma on mental health are in line with several previous studies and with the “cumulative” stress hypothesis, in which adverse experiences early in life predispose individuals to be more vulnerable to adversity later in life (Nederhof and Schmidt, 2012). Unfortunately, the design of this study does not include data on recent stressful events, which would allow a further understanding of the adaptive value of childhood trauma during adulthood. We also have observed that the detrimental associations with emotional abuse and neglect were stronger in men than in women, and the opposite was found for favorable correlations with physical abuse and neglect. Samplin et al. (2013) showed an association between emotional abuse and gender, demonstrating that a history of traumatic emotional experiences was related to a reduced hippocampal volume in males, but not in females. Also, being female was a predictor of resilience in adolescence and young adulthood
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
5
Fig. 2. Specific contributions of emotional trauma and physical abuse regarding volition, coping and fear scores. A. data distribution of emotional trauma scores (emotional abuse þ emotional neglect). B. data distribution of physical abuse. Relationship of emotional trauma quintiles and physical abuse with Volition (C), Coping (D) and Fear traits (E). Results are shown as mean ± 95%CI.
(DuMont et al., 2007) and females were shown to have higher rates of resilience than men for all types of childhood trauma (McGloin and Widom, 2001). On the other hand, Haatainen et al. (2003) showed that adverse childhood experiences are more strongly related with hopelessness among adult women than men. The most surprising result was the adaptively favorable association between physical abuse and neglect during childhood with temperament traits. One possibility is that studies which only address psychopathology are less likely to uncover putative “beneficial” effects of physical punishment since they do not address positive attributes and fail to test for the interactions between types of traumatic events. Importantly, Li et al. (2014), also using a linear regression analysis, found that physical abuse was associated with lower neuroticism, whereas emotional maltreatment was associated with higher neuroticism and psychoticism. They attributed their results to cultural concepts that physical discipline is regarded by many Chinese (including children) as necessary, in agreement with a famous Chinese proverb “no beating, no success”. This possibility seems less likely given the apparent differences between Chinese and Brazilian cultures and the similar results of both studies. Moreover, a volunteer does not answer a questionnaire on personality traits taking into account one's answers on a questionnaire on traumatic experiences or vice versa. However, the present explanatory models do not seem to contemplate the “positive” effects of physical abuse and neglect during childhood. For example, the stress inoculation hypothesis suggests that relatively mild early-life adversity may promote resilience and prepare for similar challenges in later-life (Daskalakis et al., 2013), but our results showed stronger associations with higher physical adversity scores. In contrast, the mismatch
hypothesis states that high levels of psychosocial stress early in life may program for dealing with high psychosocial stress in later life, i.e., the costs or benefits of a past experience depend on how it matches with the present context (Nederhof and Schmidt, 2012). The lack of data on recent events prevents a specific evaluation of this model, but it is nonetheless difficult to reconcile the opposite associations of emotional and physical maltreatment with temperament traits. A third perspective is the differential susceptibility hypothesis, which posits that some individuals are more susceptible than others to both negative (risk-promoting) and positive (development-enhancing) environmental conditions (Ellis et al., 2011). Again, this model does not account for the positive effects of adverse physical experiences. Thus, a new proposal seems necessary. The most well known source on this subject is Nietzsche's quote “From life's school of war: what does not kill (or destroy) me makes me stronger” from his book Twilight of the Idols or How to Philosophize with a Hammer, written in 1888. Our data supports this principle only regarding physical maltreatment, whereas the effects of emotional and sexual traumas are in line with cumulative stress hypothesis, i.e., what does not kill me makes me weaker. Therefore, we suggest that the “context-dependent nature” of stress may play a significant role in the subject of stress and adaptation. From an evolutionary perspective, physical adversity has exerted selective pressure over several million years, i.e., the body is “used to” harshness and is able to recover relatively quickly. In contrast, emotional abuse was only possible after the development of language many thousands of years ago and the meaning of a harsh affirmation may remain for much longer, producing more impact on self-esteem at the trait level. A second reason may derive
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
6
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Fig. 3. Distribution of affective temperaments according to childhood trauma quintiles. Data is shown as percentage of subjects in each category for males (top) and females (bottom) separately. * ¼ a higher proportion and # ¼ a lower proportion within each affective type (Chi-square test, p < 0.05).
from a “corrective” intention behind physical punishment from an authority figure, which is not as evident in insults from a parent towards his or her child. Finally, one should be cautious not to extrapolate this data to physical defeat among peers, which refers to a different context. Our study has significant limitations to be acknowledged. The current sample was comprised primarily of highly educated persons in the Brazilian population and other selection biases may have contributed to the volunteers' motivation to participate. The generalization of the current findings to other cultures and contexts of adversity will depend on future research. A second limitation was the reliance on a single method of data collection (i.e., selfreport questionnaires). Generally, a multi-modal approach to assessment is desirable in assessing trauma history and current emotional and physical functioning. Furthermore, assessments of child and adult trauma exposure were retrospective and crosssectional, preventing any firm conclusions on the direction of the effect (personality traits influencing memory of childhood trauma or childhood trauma affecting personality traits) or the weight of each direction. Prospective studies would address more directly the possibility of causal link and allow for more detailed exploration of mediators and moderators of these relationships. Regarding the association between physical abuse/neglect and more positive personality traits, it is possible that it has been overestimated due to a selection bias. Community members who have “survived” childhood physical abuse/neglect experiences and made it in life might be more willing to participate in such a study than unexposed healthy individuals or individuals with childhood trauma but less beneficial outcomes in adulthood. Other unhidden factors associated with participation and non-participation in this online study (and not physical abuse/neglect alone) may also be
responsible for the positive personality outcomes. Further studies should also address different contexts of physical punishment (e.g. if related to a misbehavior or not). The major strengths were the large sample and the anonymous assessment. Also, despite being cross-sectional, which is prone to memory bias, 18 of the 25 questions of CTQ are about objective situations (e.g. being called names or spanked), not their subjective impact. Finally, the BRAINSTEP has many scales and subjects are unaware of study hypothesis; they voluntarily participate as a means to gain knowledge on their personal profile. In conclusion, total trauma was associated with dysfunctional emotional traits and affective temperaments, in agreement with studies using other designs and methods to evaluate trauma and personality traits. The adverse temperament profiles were related mainly to emotional trauma and, in a lesser degree, to sexual trauma. Positive attributes such as volition, coping and low fear traits were associated with physical abuse and neglect, particularly in men. This suggests that physical trauma could even strengthen individuals and that emotional trauma would be mostly linked to psychiatric and personality disorders. As a society, by focusing on physical trauma against children, which can be assessed objectively, we may be ignoring the real harm of emotional abuse and neglect, which is often undetected due to their subjective nature. Other studies using different methods are necessary to deepen the knowledge on this subject. Role of funding source The funding agency had no influence on study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
R. Sudbrack et al. / Journal of Psychiatric Research xxx (2015) 1e7
Contributors DRL designed the study and worked on the implementation of the system. RS, HWC and DRL undertook the statistical analysis, PM and IK managed the literature searches, and all authors contributed in writing the manuscript. All authors have approved the final manuscript. Conflict of interest The authors declare no conflicts of interest. Acknowledgments This research was funded by PRONEX-FAPERGS (10/0055-0). DRL is a CNPq research fellow, PM and IK received a graduate student fellowship from FAPERGS and CNPq. References Afifi TO, Mather A, Boman J, Fleisher W, Enns MW, MacMillan H, et al. Childhood adversity and personality disorders: results from a nationally representative population-based study. J Psychiatr Res 2011;45:814e22. Allen B, Lauterbach D. Personality characteristics of adult survivors of childhood trauma. J Trauma Stress 2007;20:587e95. ria JU, Gigante LP, Figueiredo ACL, Aerts DRGC, et al. Child Bassini DG, Palazzo LS, Be sexual abuse in southern Brazilian and associated factors: a population-based study. BMC Public Health 2009;9:133. Bernstein DP, Fink L, Handelsman L, Foote J, Lovejoy M, Wenzel K, et al. Initial reliability and validity of a new retrospective measure of child abuse and neglect. Am J Psychiatry 1994;151:1132e6. Bornstein MH, Lamb ME. Developmental science: an advanced text book. 6th ed. New York: Psychological Press; 2010. Briere J, Elliott DM. Prevalence and psychological sequelae of self-reported childhood physical and sexual abuse in a general population of men and women. Child Abus Negl 2003;27:1205e22. Cloninger CR, Svrakic DM, Przybeck TR. A psychobiological model of temperament and character. Arch Gen Psychiatry 1993;50:975. rbara JFRS, Santos CAST, Gomes WA, de Souza HL. Costa MCO, de Carvalho RC, Ba ^ncia contra crianças e adolescentes, segundo registro de conO perfil da viole ~es de viole ^ncia. Cie ^ncia Saúde selhos tutelares: vítimas, agressores e manifestaço Coletiva 2007;12:1129e41. Couper MP. Review: web surveys: a review of issues and approaches. Public Opin Q 2000:464e94. Daskalakis NP, Bagot RC, Parker KJ, Vinkers CH, De Kloet E. The three-hit concept of vulnerability and resilience: toward understanding adaptation to early-life adversity outcome. Psychoneuroendocrinology 2013;38:1858e73. De Carvalho HW, Bisol LW, Ottoni LG, Lara DR. The affective and emotional temperament composite scale: psychometric analysis including anxiety and instability subscales. In: Abstract book of the third international conference on neurobiology, psychopharmacology and treatment guidance; 2013. p. 176. DuMont KA, Widom CS, Czaja SJ. Predictors of resilience in abused and neglected children grown-up: the role of individual and neighborhood characteristics. Child Abus Negl 2007;31:255e74. Ellis BJ, Boyce WT, Belsky J, Bakermans-Kranenburg MJ, van Ijzendoorn MH. Differential susceptibility to the environment: an evolutionaryneurodevelopmental theory. Dev Psychopathol 2011;23:7e28.
7
Flory JD, Yehuda R, Grossman R, New AS, Mitropoulou V, Siever LJ. Childhood trauma and basal cortisol in people with personality disorders. Compr Psychiatry 2009;50:34e7. Frizzo MN, Bisol LW, Lara DR. Bullying victimization is associated with dysfunctional emotional traits and affective temperaments. J Affect Disord 2013;148: 48e52. Gosling SD, Vazire S, Srivastava S, John OP. Should we trust web-based studies? A comparative analysis of six preconceptions about internet questionnaires. Am Psychol 2004;59:93. Grassi-Oliveira R, Stein LM, Pezzi JC. Translation and content validation of the Childhood Trauma Questionnaire into Portuguese language. Rev Saude Publica 2006;40:249e55. € J, Honkalampi K, Koivumaa-Honkanen H, Hintikka J, Haatainen K, Tanskanen A, Kylma et al. Gender differences in the association of adult hopelessness with adverse childhood experiences. Soc Psychiatry Psychiatr Epidemiol 2003;38:12e7. Heins M, Simons C, Lataster T, Pfeifer S, Versmissen D, Lardinois M, et al. Childhood trauma and psychosis: a case-control and case-sibling comparison across different levels of genetic liability, psychopathology, and type of trauma. Am J Psychiatry 2011;168:1286e94. Lara DR, Akiskal HS. Toward an integrative model of the spectrum of mood, behavioral and personality disorders based on fear and anger traits: II. Implications for neurobiology, genetics and psychopharmacological treatment. J Affect Disord 2006;94:89e103. Lara DR, Bisol LW, Brunstein MG, Reppold CT, de Carvalho HW, Ottoni GL. The Affective and Emotional Composite Temperament (AFECT) model and scale: a system-based integrative approach. J Affect Disord 2012a;140:14e37. Lara DR, Ottoni GL, Brunstein MG, Frozi J, de Carvalho HW, Bisol LW. Development and validity data of the Brazilian Internet Study on Temperament and Psychopathology (BRAINSTEP). J Affect Disord 2012b;141:390e8. Lara DR, Pinto O, Akiskal K, Akiskal HS. Toward an integrative model of the spectrum of mood, behavioral and personality disorders based on fear and anger traits: I. Clinical implications. J Affect Disord 2006;94:67e87. Li X, Wang Z, Hou Y, Wang Y, Liu J, Wang C. Effects of childhood trauma on personality in a sample of Chinese adolescents. Child Abus Negl 2014;38:788e96. Mathews G, Deary IJ, Whiteman MC. Personality traits. 2nd ed. Cambridge: Cambridge University Press; 2003. McGloin J, Widom CS. Resilience among abused and neglected children grown up. Dev Psychopathol 2001;13:1021e38. Nederhof E, Schmidt MV. Mismatch or cumulative stress: toward an integrated hypothesis of programming effects. Physiol Behav 2012;106:691e700. Rademaker AR, Vermetten E, Geuze E, Muilwijk A, Kleber RJ. Self-reported early trauma as a predictor of adult personality: a study in a military sample. J Clin Psychol 2008;64:863e75. Samplin E, Ikuta T, Malhotra AK, Szeszko PR, DeRosse P. Sex differences in resilience to childhood maltreatment: effects of trauma history on hippocampal volume, general cognition and subclinical psychosis in healthy adults. J Psychiatr Res 2013;47:1174e9. Sansone RA, Pole M, Dakroub H, Butler M. Childhood trauma, borderline personality symptomatology, and psychophysiological and pain disorders in adulthood. Psychosomatics 2006;47:158e62. Spertus IL, Yehuda R, Wong CM, Halligan S, Seremetis SV. Childhood emotional abuse and neglect as predictors of psychological and physical symptoms in women presenting to a primary care practice. Child Abuse Negl 2003;27(11): 1247e58. ^a F, Souza-Formigoni MLO. Childhood trauma in substance use Tucci AM, Kerr-Corre disorder and depression: an analysis by gender among a Brazilian clinical sample. Child Abus Negl 2010;34:95e104. Turner CF, Ku L, Rogers SM, Lindberg LD, Pleck JH, Sonenstein FL. Adolescent sexual behavior, drug use, and violence: increased reporting with computer survey technology. Science 1998;280:867e73. Widom CS, DuMont K, Czaja SJ. A prospective investigation of major depressive disorder and comorbidity in abused and neglected children grown up. Arch Gen Psychiatry 2007;64:49e56. Zhang T, Chow A, Wang L, Dai Y, Xiao Z. Role of childhood traumatic experience in personality disorders in China. Compr Psychiatry 2012;53(6):829e36.
Please cite this article in press as: Sudbrack R, et al., What doesn't kill you makes you stronger and weaker: How childhood trauma relates to temperament traits, Journal of Psychiatric Research (2015), http://dx.doi.org/10.1016/j.jpsychires.2015.01.001
