The effect of childhood trauma on complex posttraumatic stress disorder: the role of self-esteem

ABSTRACT

Background: Due to the short time that complex posttraumatic stress disorder (CPTSD) has been an independent diagnosis, few studies have explored the role that self-esteem might play in the relationship between childhood trauma and CPTSD.

Objective: The current study aimed to explore the impact of childhood trauma on CPTSD and the role of self-esteem in this relationship.

Methods: Study 1 involved a questionnaire survey in which a total of 360 young Chinese adults with childhood trauma participated. Study 2 used experimental research by manipulating short-term self-esteem to explore the effect of self-esteem on CPTSD. A total of 80 young Chinese adults with childhood trauma participated.

Results: The results of Study 1 showed that childhood trauma positively predicts adulthood CPTSD, while self-esteem partially mediates the relationship between the two. It also found that self-esteem has a greater mediating effect on the disordered self-organization (DSO) dimension of CPTSD than on the PTSD dimension. In Study 2, participants in the high manipulated-self-esteem group reported fewer CPTSD symptoms than those in the low manipulated-self-esteem group.

Conclusion: Overall, this study emphasized the role of self-esteem in CPTSD among individuals with a childhood trauma history. In practice, we provided a potential optimization direction for CPTSD clinical treatment, suggesting a method of self-esteem reconstruction.

HIGHLIGHTS

• Self-esteem mediates the relationship between childhood trauma and CPTSD.

• Short-term self-esteem decrease may contribute to CPTSD symptoms.

• Self-esteem has greater effects on disordered self-organization than PTSD symptoms.

Antecedentes: Debido a corto periodo en el que el trastorno de estrés postraumático (TEPTC) es un diagnostico independiente, pocos estudios han explorado el papel que pudiera jugar la autoestima en la relación entre trauma infantil y TEPTC.

Objetivo: El presente estudio tuvo como finalidad explorar el impacto del trauma infantil en el TEPTC y el papel de la autoestima en esta relación.

Métodos: El Estudio 1 implicó una encuesta por cuestionarios en la que participaron un total de 360 chinos adultos jóvenes con trauma infantil. El Estudio 2 utilizó una investigación experimental manipulando la autoestima a corto plazo para explorar el efecto de la autoestima en el TEPTC. Participaron un total de 80 chinos adultos jóvenes con trauma infantil.

Resultados: Los resultados del Estudio 1 mostraron que el trauma infantil predijo positivamente el TEPTC en la adultez, mientras que la autoestima media parcialmente la relación entre los dos. Se encontró también que la autoestima tiene un gran efecto mediador en la dimensión de la alteración en la auto-organización (DSO por sus siglas en inglés) del TEPTC que en la dimensión TEPT. En el Estudio 2, los participantes del grupo de alta autoestima manipulada reporto menos síntomas de TEPTC que aquellos del grupo de baja autoestima manipulada.

Conclusión: En general, este estudio enfatizo el papel de la autoestima en el TEPTC entre los individuos con historia de trauma infantil. En la práctica, proporcionamos una posible dirección de optimización para el tratamiento clínico del TEPTC, sugiriendo un método de reconstrucción de la autoestima.

背景：由于复杂性创伤后应激障碍（CPTSD）成为独立诊断的时间很短，很少有研究探讨自尊在童年创伤和 CPTSD 之间的关系中可能发挥的作用。

目的：本研究旨在探讨童年创伤对 CPTSD 的影响以及自尊在这种关系中的作用。

方法：研究 1 涉及一项问卷调查，共有 360 名有童年创伤的中国年轻人参与。研究2通过操纵短期自尊的实验研究来探讨自尊对CPTSD的影响。共有80名有童年创伤的中国年轻人参加。

结果：研究1的结果表明，童年创伤正向预测成年期CPTSD，而自尊部分中介两者之间的关系。研究还发现，自尊对 CPTSD 的自我组织障碍（DSO）维度的中介作用比对 PTSD 维度的中介作用更大。在研究2中，高自尊操纵组的参与者比低自尊操纵组的参与者报告的 CPTSD 症状更少。

结论：总体而言，本研究强调了自尊在有童年创伤史的个体 CPTSD 中的作用。在实践中，我们为CPTSD临床治疗提供了一个潜在的优化方向，即对自尊的重建。

KEYWORDS:

• Complex posttraumatic stress disorder
• childhood trauma
• self-esteem
• disordered self-organization
• identity

PALABRAS CLAVE:

• Trastorno de estrés postraumático complejo
• trauma infantil
• autoestima
• autoorganización desordenada
• identidad

关键词:

• 复杂性创伤后应激障碍
• 童年创伤
• 自尊
• 自我组织障碍
• 同一性

1. Introduction

In 2018, the International Classification of Diseases (ICD-11) of the World Health Organization (WHO) officially included complex posttraumatic stress disorder (CPTSD) as an independent disorder, paralleling PTSD under the classification of stress-related disorders (WHO, 2018). Although CPTSD has recently been listed as an independent diagnosis, it has been discussed by researchers for many years. Clinical professionals have found that some trauma types can have additional effects on PTSD. For example, Herman’s research found that beyond PTSD symptoms, female victims who have experienced domestic and sexual violence (including childhood sexual abuse) also develop six other core symptoms, including affective dysregulation, changes in consciousness, self-perception disorders, misperception of perpetrators, relationship problems, and value system transformation (Herman, 2015). The ICD-11 further concludes that there are two significant differences between CPTSD and PTSD. The first distinction lies in the trauma types. PTSD tends to appear after single trauma events (e.g. natural disasters, accidents), while CPTSD is often caused by long-term or recurring events (e.g. domestic violence and repeated sexual or physical abuse; WHO, 2018). In addition, CPTSD symptoms are more easily induced by some childhood adversities, such as domestic conflicts, school learning difficulties and harm from relationships (Daniunaite et al., 2021).

Symptom structure is another important difference between CPTSD and PTSD. Except for the three core symptom clusters of ICD-11 PTSD (i.e. re-experiencing, avoidance, and sense of threat), CPTSD, defined according to the ICD-11, should also fulfill the diagnostic requirements of the newly proposed disordered self-organization (DSO) dimension (WHO, 2018). DSO includes three symptom clusters: affect dysregulation, negative self-concept (NSC), and interpersonal problems. More specifically, affective dysregulation refers to the disturbance of emotional regulation, which is usually manifested as intensified emotional reactions, impulsive violent behavior, self-destructive behavior, or dissociative symptoms under pressure; NSC refers to the long-term belief that one is weakened, failed, or worthless, accompanied by feelings of shame, guilt or failure associated with traumatic events; interpersonal problems are reflected in the difficulty maintaining relationships and being close to others (WHO, 2018). These three clusters are mainly characterized by distortion of personal identity and severe emotional dysregulation (Brewin et al., 2017).

Since the ICD-11 included CPTSD as an independent disorder, studies on ICD-11 CPTSD have gradually emerged, and childhood trauma has been demonstrated to be an important predictor. Childhood trauma generally refers to a variety of traumatic events or circumstances before the age of 18 that may cause mental or physical harm, including interpersonal violence, accidents, serious injuries, and many traumatic network events (such as traumatic experiences involving people children are close to; McLaughlin et al., 2013). Numerous studies have demonstrated that childhood traumatic experiences increase the risk of developing CPTSD in adulthood (Karatzias et al., 2020; Parra et al., 2017; van Dijke et al., 2015). Individuals who have experienced childhood trauma are more likely to develop adverse health behaviors, social adjustment problems, and psychiatric disorders (e.g. mood disorders and stress-related disorders), and this effect on physical and mental functioning may persist into adulthood or throughout life (Dye, 2018; Krammer et al., 2016; Perkonigg et al., 2016; Taylor et al., 2004; Thoma et al., 2021). However, although empirical evidence has confirmed the predictive role of childhood trauma in adulthood CPTSD, the underlying mechanism is still unclear.

Self-esteem is a potential mediator in the relationship between childhood trauma and CPTSD. Studies have shown that childhood adversity can lead to changes in personal self-worth and may negatively affect the development of self-concepts (Kim & Cicchetti, 2006). For example, childhood interpersonal trauma (especially interpersonal violence and sexual abuse) can directly promote the formation of negative self-concepts (Hyland et al., 2017), so individuals who have suffered childhood trauma are more likely to be self-deprecating and self-loathing, thus developing lower emotional self-esteem (Gilbert, 2015; Weindl & Lueger-Schuster, 2018). Therefore, childhood trauma may affect an individual’s mental health by impairing their self-esteem. Moreover, some studies have shown that self-esteem can partially mediate the relationship between specific childhood traumatic experiences (such as sexual abuse) and PTSD (Murphy et al., 2014; Turner et al., 2010). The predictive role of self-esteem on PTSD has also been generally confirmed: low self-esteem is regarded as a risk factor for PTSD, and high self-esteem can reduce the risk of PTSD (Adams & Boscarino, 2006; Frazier et al., 2011). Moreover, there are some associations between low self-esteem and DSO symptoms (especially NSC symptoms) in CPTSD, since low self-esteem and NSC are both closely related to negative self-evaluation and the destruction of self-worth (Mann et al., 2004; WHO, 2018). Although self-esteem and self-concept seem to have many overlaps in definition, they are not synonymous. Self-esteem is usually limited to the evaluative aspects of the self, while self-concept could be used for all self-descriptions, which do not necessarily involve judgments of worth (Watkins & Dhawan, 1989). Considering that self-esteem is associated with both PTSD and DSO symptoms of CPTSD and can also be affected by childhood trauma, we assumed that self-esteem itself may mediate the association between childhood trauma and CPTSD.

The current study aimed to explore the impact of childhood trauma on CPTSD and focused on the role of self-esteem in their relationship. We conducted two studies to fulfill this goal. Study 1 used questionnaires to understand their relationship in a naturalistic setting. We hypothesized that childhood trauma would positively predict CPTSD, while self-esteem would mediate their relationship. Study 2 manipulated the level of self-esteem of participants with childhood trauma in the laboratory to explore the potential impact of self-esteem on CPTSD. We hypothesized that high self-esteem priming would be associated with a level of CPTSD decrease and that low self-esteem priming would be associated with a level of CPTSD increase.

2. Study 1

2.1. Methods

2.1.1. Participants and procedures

In this study, participants spent approximately 10 min completing an online questionnaire. Each participant voluntarily consented before participation and received 8 RMB as a reward after the study. They were also told in advance that if they had any uncomfortable feelings after the study, the researcher would provide them with professional psychological services. The present study was approved by the Ethics Committee on Human Research Protection of East China Normal University.

Five hundred and thirty-seven Chinese college students aged 18–25 years filled out the online questionnaire after reading our recruitment advertisement. The valid sample was obtained after excluding 62 invalid data points due to failing the attention check items or inaccurate responses. A general population of 475 valid samples remained. Among the general population, 360 had experienced or witnessed at least one childhood traumatic event (any item of the Life Events Checklist for the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders [DSM-5] scored greater than or equal to 4), and they were identified as the population who reported having a childhood trauma history. Among the participants with childhood trauma (n = 360), 154 were male and 206 were female, with an average age of 20.71 years (SD = 1.69).

2.1.2. Measures

The Life Events Checklist for DSM-5 (LEC-5; Weathers et al., 2013) was used to assess childhood traumatic experiences with the instruction ‘consider your life before you were 18 years old’. Because Chinese college students are currently unlikely to experience war-related or terroristic events in mainland China, we deleted three items (‘combat or exposure to a war zone’, ‘captivity’ and ‘severe human suffering’). This modification was similar to previous research conducted by Tian et al. (2020), which also used a large college student sample in China. We also deleted the item ‘serious accident at work, home, or during recreational activity’; as most of the college students in China have not participated in work, this item was not fully appropriate to our sample. In addition, we added two items (‘severely bullied’ and ‘emotional abuse’) because the prevalence rates of these two types of childhood trauma are high in China (Zhou et al., 2019; 2022). Each item was rated on a 6-point Likert scale ranging from 0 to 5 (0 = doesn’t apply, 1 = not sure, 2 = part of my life, 3 = learned about it, 4 = witnessed it, and 5 = happened to me). In Study 1, we used LEC-5 for two purposes. During the stage of recruiting participants, we used LEC-5 as an inclusion criterion to screen out our target sample, that is, people who reported having a childhood trauma history. Any LEC-5 item scored greater than or equal to 4 (i.e. witnessing or experiencing any traumatic event) was considered to meet the inclusion criterion, and subsequent data analyses were based on these samples. During data analysis, we calculated the total LEC-5 score as a measure of the independent variable, childhood trauma.

The International Trauma Questionnaire (ITQ; Cloitre et al., 2018) is used to assess ICD-11 CPTSD. It includes two subscales of PTSD and DSO. Each subscale contains nine items, six assessing symptoms and three assessing dysfunction. All items are rated on a 5-point Likert scale from 0 to 4 (0 = not at all, 1 = a little bit, 2 = moderately, 3 = quite a bit, 4 = extremely). Probable PTSD or DSO can be independently diagnosed when each symptom item is scored 2 or above and at least one of three dysfunction items is scored 2 or above. Probable CPTSD should meet the criteria for both probable PTSD and DSO. In this study, we used a Chinese version of the ITQ revised by Ho et al. (2019). The total score of twelve PTSD and DSO symptom items was calculated to measure CPTSD symptoms. The scale had good internal consistency reliability in this study, Cronbach's α = .92; the Cronbach's α of the PTSD subscale was .90, and that of the DSO subscale was .90.

The Rosenberg self-esteem questionnaire (RSE; Rosenberg, 1965) is used to measure self-esteem. Ten items are rated on a 4-point Likert scale from 1 to 4 (1 = strongly disagree, 2 = disagree, 3 = agree, and 4 = strongly agree). According to Tian (2006), we changed the 8th item (‘I wish I could have more respect for myself’) from reverse scoring to normal scoring, which was more in accordance with the Chinese cultural context because it was often regarded as a traditional virtue of modesty. A higher total score indicated a higher self-esteem level. The scale had good internal consistency reliability in this study, Cronbach's α = .91.

2.1.3. Data analysis

Statistical software IBM SPSS Statistics version 28 for Windows and Mplus 8.3 were used for data analysis. SPSS was used for descriptive statistics and correlation analysis, while Mplus 8.3 was used for mediation analysis. In the descriptive statistics, we reported the prevalence of childhood trauma and CPTSD in both the general population (n = 475) and participants with childhood trauma (n = 360). The correlation analysis and mediation analysis were only performed in the participants with childhood trauma.

Correlation analyses were performed to test the relationship among childhood trauma, self-esteem, and CPTSD in SPSS. Although bias exists in cross-sectional mediation analyses (Maxwell & Cole, 2007), analyses of cross-sectional correlations with well-founded theories can still advance knowledge of potential relationships between variables (Shrout, 2011). In this study, since childhood trauma was a preexisting and more stable variable, it was suitable to serve as an independent variable. Previous research has also indicated that self-esteem may affect PTSD and DSO (Frazier et al., 2011; Mann et al., 2004), so self-esteem and CPTSD were used as mediating and dependent variables, respectively. We conducted a mediation analysis by Mplus 8.3 to test whether self-esteem plays a mediating role in the relationship between childhood trauma and CPTSD (also in PTSD and DSO). The bootstrap sample size was set to 1000.

2.2. Results

2.2.1. Descriptive statistics

According to the diagnostic criteria of the ITQ, the prevalence rates of probable CPTSD and PTSD were 4.4% and 2.5%, respectively, in the general population (n = 475). Among 360 participants with trauma, the prevalence rates of probable CPTSD and PTSD were 5.3% and 3.3%, respectively. Among participants with childhood trauma, transportation accidents (48.1%), physical assault (41.6%), and natural disasters (41.4%) had the highest prevalence rates among all traumatic events. The prevalence rates of all traumatic events are shown in Table 1.

Table 1. Prevalence of childhood trauma.

Traumatic events	Prevalence (general population)	Prevalence (participants with trauma)
Nature disaster	31.4%	41.4%
Fire or explosion	14.7%	19.4%
Transportation accident	36.4%	48.1%
Exposure to toxic substance	4.8%	6.4%
Life-threatening illness or injury	21.1%	27.7%
Physical assault	31.6%	41.6%
Assault with a weapon	15.2%	20.0%
Sexual assault	2.9%	3.9%
Other unwanted or uncomfortable sexual experience	11.2%	14.7%
Serious injury, harm, or death you caused to someone else	6.1%	8.0%
Emotional abuse	27.6%	36.3%
Severely bullied	24.4%	32.2%
Sudden violent death	8.4%	11.1%
Sudden accidental death	18.7%	24.7%

Note: The general population refers to 475 valid samples after excluding invalid data; the participants with trauma refers to 360 samples that meet the LEC criteria.

The criterion for calculating the prevalence of each traumatic event was that one item scored greater than or equal to 4 (4 = participants witnessed trauma, 5 = personally experienced trauma).

2.2.2. Correlation analysis

The results of correlation analyses showed that childhood trauma was negatively correlated with self-esteem and positively correlated with CPTSD, PTSD and DSO. Self-esteem was negatively correlated with adulthood CPTSD, PTSD and DSO. There was no significant correlation between sex or age and other variables (see Table 2).

Table 2. Correlation between childhood trauma, self-esteem and CPTSD.

	Age	Sex	Childhood trauma	Self-esteem	PTSD	DSO
Age	1
Sex	−.101	1
Childhood trauma	−.022	−.033	1
Self-esteem	.071	−.036	−.312**	1
PTSD	.039	−.078	.522**	−.302**	1
DSO	−.048	.020	.586**	−.643**	.633**	1
CPTSD	−.005	−.032	.613**	−.524**	.902**	.905**

Note: ** p < .01.

2.2.3. The mediating role of self-esteem

The results showed that for PTSD, the direct effect of childhood trauma on PTSD was significant (p < .001, 95% CI = [.37, .57]), and the direct effect size was .47, accounting for 90.8% of the total effect. Additionally, the mediating effect of self-esteem was significant (p = .006, 95% CI = [.02, .09]), with an effect size of .05, accounting for 9.2% of the total effect. For DSO, the direct effect of childhood trauma on DSO was significant (p < .001, 95% CI = [.34, .50]), and the direct effect size was .43, accounting for 72.9% of the total effect. Additionally, the mediating effect of self-esteem was significant (p < .001, 95% CI = [.11, .21]), with an effect size of .16, accounting for 27.1% of the total effect (see Table 3).

Table 3. Indirect effect for the mediation model.

Model pathways	Estimate	Lower 95% CI	Higher 95% CI
Childhood Trauma→Self-esteem→PTSD	.048	.018	.088
Childhood Trauma→Self-esteem→DSO	.159	.106	.210

Note: All values above are standardized.

In conclusion, self-esteem played a partial mediating role in both PTSD and DSO, with a larger effect on DSO symptom clusters. The mediation model is shown in Figure 1.

Figure 1. Mediating effect of self-esteem on the relationship among childhood trauma, PTSD and DSO.Note: The coefficients in the figure are all standardized. ** p < .01.

Self-esteem mediated the relationship between childhood trauma and CPTSD symptoms and had a more strongly negative association with DSO than with PTSD symptoms.

3. Study 2

The main purpose of Study 2 was to examine whether short-term self-esteem manipulation affected the CPTSD symptoms of individuals with childhood trauma in a laboratory setting. The hypothesis was that participants in the low self-esteem group would show higher CPTSD levels than those in the high self-esteem group.

3.1. Methods

3.1.1. Participants

We recruited 80 participants with childhood trauma, and the inclusion criterion was the same as that in Study 1 (any item of the LEC-5 scored greater than or equal to 4). Participants who participated in Study 1 could still participate in Study 2 if they met the inclusion criteria. Among all the participants, 26 were men and 54 were women, with an average age of 20.48 years (SD = 1.53).

The participants were randomly assigned to two manipulation groups, 40 people (27 women) in the high self-esteem priming group, with an average age of 20.40 years (SD = 1.34), and 40 people (27 women) in the low self-esteem priming group, with an average age of 20.55 years (SD = 1.71).

3.1.2. Measures

The Life Events Checklist for DSM-5 (LEC-5; Weathers et al., 2013) was the same as in Study 1.

The State Self-Esteem Questionnaire (SSES) was used to measure state self-esteem and was first compiled by Heatherton and Polivy (1991). We adopted the Chinese version revised by Lan (2008). The scale includes three dimensions: performance self-esteem, social self-esteem, and appearance self-esteem. All 20 items were scored on a 5-point Likert scale ranging from 1 (not at all) to 5 (extremely). Specifically, 14 items were reverse scored. Higher total scores referred to a higher state of self-esteem. The scale had good internal consistency reliability in this study, with Cronbach’s α = .92.

The International Trauma Questionnaire (ITQ; Cloitre et al., 2018) was similar to that used in Study 1. Based on the purpose of this study, we slightly revised the instructions of this questionnaire to ‘These questions are about how you are feeling right now’. The scale had good internal consistency reliability in this study, Cronbach’s α = .92; the Cronbach’s α of the PTSD subscale was .88, and that of the DSO subscale was .93.

The Self-Esteem Manipulation Paradigm was adapted from the Success or Failure Manipulation Task revised by Lan (2008). First, participants were asked to complete as much of an IQ test as possible within 15 min. The questions were selected from the Raven Advanced Graphical Reasoning Test, which had a total of 32 questions. Then, participants were told that their test results would be compared with the existing data pool of 100 participants of the same age, which actually did not exist. After they finished the test, participants automatically received fake result feedback set by the researcher in advance. The high self-esteem priming group received success feedback, while the low self-esteem priming group received failure feedback. We recruited 20 participants to perform the pilot study, and the results showed the effectiveness of this priming paradigm in this study, as the self-esteem level of the high group was significantly higher than that of the low group (t = 3.21, p = .005, Cohen's d = 11.42).

3.1.3. Procedure

To run the self-esteem manipulation paradigm effectively, we used a univariate between-subject design and conducted the study in a single-blind way. Participants were blinded to the real purpose of the experiment and the grouping method, whereas the experimenter conducted the whole study according to standardized instructions. First, the participants were randomly divided into two groups (a high self-esteem priming group and a low self-esteem priming group) and were told that the purpose of this study was to explore the mechanism of intelligence level on mental health. Then, the participants were guided to complete a pretest questionnaire (including the SSES and ITQ) and the self-esteem manipulation paradigm in sequence. Afterward, participants were asked to complete the posttest questionnaire, which included the SSES and ITQ. Since the manipulation task contained the deceiving procedure, the researcher revealed the truth after the experiment and emphasized that there was no peer comparison data pool to eliminate the potential negative effects on the participants.

All participants voluntarily consented before participation and received 20 RMB as a reward after the study. They were also told in advance that professional psychological services were available after the study if needed. The whole experiment took approximately 25 min. The present study was approved by the Ethics Committee on Human Research Protection of East China Normal University.

3.1.4. Data analysis

The statistical software IBM SPSS Statistics 28 was used for data analysis. There were no missing data. Independent sample t tests were used to analyze between-group differences in CPTSD symptoms, and paired samples t tests were conducted to analyze within-group differences in CPTSD symptoms. To avoid the distracting effect of the baseline levels of the pretest ITQ, we used a univariate analysis of covariance to analyze the differences in participants’ posttest CPTSD symptom scores. All data met the assumptions of the covariance analysis in advance.

3.2. Results

3.2.1. Validity of the self-esteem manipulation paradigm

The average scores of pre- and posttest self-esteem in different groups are shown in Table 4. There was no significant between-group difference in pretest self-esteem (t = −1.09, p = .278), but there were significant between-group differences in posttest self-esteem scores (t = 4.79, p < .001). Taking the manipulation group as an independent variable, the pretest SSES score as a covariate, and the posttest SSES as a dependent variable, we conducted a covariance analysis to verify the validity of the self-esteem manipulation paradigm. The results showed that after controlling for premeasured self-esteem, the self-esteem level of the high group was significantly higher than that of the low group, F(1,77) = 43.36, p < .001, η²_p= .36. Thus, the manipulation effect was valid.

Table 4. Paired samples t-tests on pretest scores and posttest scores.

		Pretest scores		Posttest scores
Manipulation group	Variable	Mean	SD	Mean	SD	t	p
High self-esteem group (n = 40)	Self-esteem	59.65	13.86	67.78	12.29	−4.67	<.001
	CPTSD	17.65	11.13	15.60	9.87	2.56	.014
	PTSD	7.45	5.07	7.15	5.11	0.63	.532
	DSO	10.10	6.92	8.45	6.08	2.77	.009
Low self-esteem group (n = 40)	Self-esteem	62.83	12.07	55.25	11.07	4.11	<.001
	CPTSD	16.10	9.91	17.73	10.16	−2.13	.039
	PTSD	7.45	5.26	8.55	5.60	−2.04	.048
	DSO	7.78	5.54	8.65	5.33	−2.30	.027

3.2.2. The influence of self-esteem on CPTSD

Pre- and posttest scores of CPTSD, PTSD and DSO in the high and low SSES groups are shown in Table 4. The clustered column charts are shown in Figure 2.

Figure 2. CPTSD, PTSD, DSO scores in different manipulation groups. Note: * p < .05, ** p < .01, ns = non-significant. The between-group differences in the pretest and posttest scores of each variable were insignificant.

Participants in the high manipulated-self-esteem group reported fewer CPTSD symptoms in post-test than in pre-test, while participants in the low manipulated-self-esteem group reported higher CPTSD symptoms in post-test than in pre-test.

Within-group differences in CPTSD, PTSD and DSO scores are shown in Table 4. The posttest score of CPTSD was significantly lower than the pretest CPTSD score in the high self-esteem group, while the posttest score of CPTSD was significantly higher than the pretest CPTSD score in the low self-esteem group. Self-esteem enhancement did not affect PTSD, but self-esteem decline increased PTSD symptoms. In addition, self-esteem affected DSO symptoms in individuals who experienced childhood trauma.

Independent samples t tests were used to test between-group differences. For CPTSD, PTSD and DSO, the between-group differences in the pretest and posttest scores of each variable were non-significant (all p > .100). To avoid the distracting effect of the baseline levels of each variable, we used a univariate analysis of covariance to analyze the differences in participants’ posttest scores for each variable. Setting the group as an independent variable, the pretest CPTSD score as a covariate, and the posttest CPTSD score as a dependent variable, a one-way analysis of covariance was performed to examine the influence of different levels of manipulated self-esteem on self-reported CPTSD symptoms. The results showed that after controlling for pretest CPTSD, the estimated marginal mean of posttest CPTSD changes was 14.95 (95% confidence interval [CI]: 13.47–16.42) in the high group and 18.38 (95% CI: 16.90–19.85) in the low group, and the CPTSD scores in the high group were significantly lower than those in the low group, F(1,77) = 10.67, p = .002, η²_p= .12. The results indicated that self-esteem can affect the perception of CPTSD symptoms in individuals who have experienced childhood trauma.

Further covariance analysis was performed on the PTSD and DSO dimensions. The results showed that after controlling for pretest PTSD, the difference in posttest PTSD between the two manipulation groups was significant, with the PTSD score in the high group being significantly lower than that in the low group, F(1,77) = 3.99, p = .049, η²_p= .05. After controlling for the pretest DSO, there was a significant difference in the posttest DSO between the two groups, with the DSO score in the high group being significantly lower than that in the low group, F(1,77) = 9.57, p = .003, η²_p= .11.

4. Discussion

The current study aimed to gain insight into the role of self-esteem in the development of CPTSD by combining questionnaire and experimental designs. The results of Study 1 showed that childhood trauma was positively associated with CPTSD in adulthood, while self-esteem was negatively associated with CPTSD. Moreover, self-esteem played a partial mediating role in the relationship between childhood trauma and CPTSD, and the mediating effect was greater on DSO than on PTSD symptoms. In Study 2, we found that participants with childhood trauma in the high manipulated-self-esteem group reported fewer CPTSD symptoms than those in the low manipulated-self-esteem group. High self-esteem priming reduced only DSO symptoms, while low self-esteem priming increased both PTSD and DSO scores. Overall, these findings indicated that short-term changes in self-esteem may contribute to CPTSD symptoms at a nonclinical level.

Study 1, using a questionnaire survey, found a positive predictive effect of childhood traumatic experiences on adulthood CPTSD, which was consistent with the findings of previous studies. Childhood traumatic events, especially childhood interpersonal trauma, are the strongest predictors of CPTSD (Cloitre et al., 2013; Haselgruber et al., 2020; Karatzias et al., 2016). The effects of childhood interpersonal trauma are mostly long-term and cumulative (Hyland et al., 2017; Weinstein et al., 2000). We further found that self-esteem mediated the relationship between childhood trauma and CPTSD. These results indicated that the effect of childhood trauma on CPTSD symptoms was related to the destruction of the self-concept and the degradation of the self-evaluation, as they were both closely related to self-esteem. Cook et al. (2005) noted that repeated hurtful behaviors or rejection by significant others may lead children to develop defective, helpless, and unloved self-perceptions. Children are more likely to blame themselves when confronting negative events and fail to properly ask for or respond to outside social support. Additionally, trauma victims are prone to degrade themselves with shame, guilt, and disgust and exhibit social inhibition behaviors, such as social withdrawal and fear (Hyland et al., 2017). These negative effects all lead to the aggravation of DSO symptoms.

The experimental manipulation of self-esteem in Study 2 revealed that the short-term enhancement of self-esteem levels may contribute to reducing CPTSD symptoms among individuals with childhood trauma. Moreover, the influence of self-esteem enhancement decreased only DSO symptoms, while the influence of self-esteem diminishment elevated both PTSD and DSO symptoms. The possible reasons why enhancement of self-esteem was only associated with a decline in DSO but not PTSD are as follows. Combined with the results in Study 1, low self-esteem was more strongly associated with DSO than with PTSD. Therefore, changes in self-esteem might be more likely to affect DSO. In addition, interventions for PTSD take more time to devote to improving self-esteem (Verhagen et al., 2023). Self-esteem priming in this study might not have been sufficient to affect PTSD. The negative relationship between self-esteem and PTSD has been indicated by previous studies, as low self-esteem might increase the likelihood of PTSD (Adams & Boscarino, 2006; Murphy et al., 2014). As for the relationship between low self-esteem and CPTSD, especially the DSO dimension, low self-esteem is closely related to emotional problems and interpersonal behavior problems (Lee & Hankin, 2009; Zhang & Cao, 2011). On the one hand, compared with individuals with high self-esteem, individuals with low self-esteem suffer from more negative self-concept and low self-evaluation, which might lead to negative emotions, mood disorders (such as anxiety and depression), and difficulties in emotion regulation (Kuster et al., 2012; Lee & Hankin, 2009; Martin-Storey & Crosnoe, 2012; Zhang et al., 2017). On the other hand, individuals with low self-esteem are inclined to experience more social avoidance, isolation and other interpersonal difficulties than individuals with high self-esteem, thus making them more likely to engage in problem behaviors, such as dropping out of school, aggressive behavior, and substance abuse (Leary et al., 1995; Zhang & Cao, 2011).

There were several limitations that can be improved in the future. First, self-esteem was measured with different instruments in the two studies. In Study 1, we used a scale to measure dispositional self-esteem, while in Study 2, we manipulated participants’ state self-esteem in the short term. Although some studies have suggested that traits of self-esteem and state self-esteem can be dynamically converted to each other (De Ruiter et al., 2017; Heatherton & Polivy, 1991), future research may consider adopting regular intervention (e.g. a daily diary method) or longitudinal studies to explore the impact of stable self-esteem changes on CPTSD among individuals with childhood trauma. An interesting topic that can be explored in longitudinal studies is the interaction between self-esteem and CPTSD. For individuals who have experienced childhood trauma, is low self-esteem not only a risk factor for CPTSD but also the victim of CPTSD? Although there is currently no empirical evidence in the field of CPTSD, it is not new in the field of depression (Sowislo & Orth, 2013; van Tuijl et al., 2020). The scar model believes that it is not self-esteem that causes depression (as the vulnerability model declares) but that depression erodes self-esteem. From this perspective, low self-esteem is not a preexisting risk factor for depression but rather the result of a previous depression episode, thus serving as a mediator in two episodes of depression (Kraemer et al., 2001). Future studies can further explore the transaction between self-esteem and CPTSD in people with childhood trauma by tracking their long-term changes. Second, the current study only recruited young Chinese adults aged 18–25 years old with childhood trauma. Future studies could broaden the sample to include other age and clinical groups. Third, we assessed CPTSD symptoms by the ITQ. Although the ITQ has been demonstrated to be an effective tool for measuring CPTSD (Cloitre et al., 2018), the natural restrictions of self-report questionnaires still exist. Structured interviews can be used to investigate symptoms in the future. Finally, although participants were similar in age and we asked them to recall only traumatic experiences before the age of 18, the exact time the reported experience occurred was not measured. The effect of the specific age of traumatic experience on CPTSD needs to be considered in future studies.

Despite these limitations, the current study made several theoretical and clinical contributions. Theoretically, the findings of our study propose a potential mechanism for how individuals with childhood traumata develop CPTSD in adulthood, which emphasizes the important effect of self-esteem on CPTSD. In particular, a decline in self-esteem might elevate both PTSD and DSO symptoms. To date, few theoretical accounts of CPTSD development have been presented. Hyland et al. (2023) proposed the memory and identity theory of CPTSD recently, emphasizing two major processes that contribute to CPTSD: intrusive memories and negative identities. The latter process includes powerless and worthless, which are related to low self-esteem. Furthermore, this theory holds that negative identities are caused by trauma exposure and individual vulnerability and subsequently contribute to PTSD and DSO symptoms. Our research supports this theory and indicates that low self-esteem might be an important negative identity that can affect CPTSD symptoms.

Regarding practice implications, our findings indicated that both PTSD and CPTSD were associated with low self-esteem; thus, trauma therapy for both PTSD and CPTSD should include improvements in self-esteem. Previous research also indicated that although changing negative identity is not a part of standard treatments for PTSD, it is effective in improving symptoms in people with high trauma exposure (Hyland et al., 2023). Moreover, CPTSD has been listed as an independent diagnosis for a short time. The current clinical treatments for CPTSD mostly follow the treatment methods for PTSD, such as eye movement desensitization and reprocessing and cognitive behavior therapy (Corrigan & Hull, 2015). However, the effectiveness of using evidence-based therapies for PTSD to treat CPTSD is questionable (Corrigan & Hull, 2015; Dorrepaal et al., 2014). Cook et al. (2005) proposed a ‘Six Core Components of Complex Trauma Intervention’ system including safety, self-regulation, self-reflective information processing, traumatic experience integration, relationship engagement, and positive affect enhancement. Regarding positive affect enhancement, the improvement of self-esteem, self-worth and positive self-appraisal are especially emphasized. In 2012, the International Society of Traumatic Stress Studies (ISTSS) published guidelines on CPTSD intervention for adults, advocating a phased approach to CPTSD treatment. The first stage is called the stabilization phase, which aims to help patients establish a sense of security and is accompanied by training in relevant skills, such as emotion recognition and expression and human social skills. This stage may help the patient regain a sense of security and control and increase self-efficacy over the trauma. Combined with our study, future CPTSD interventions can take the reconstruction of self-cognition and self-evaluation into account to improve intervention effects.

5. Conclusion

Study 1, using questionnaires, found that self-esteem mediated the relationship between childhood trauma and CPTSD symptoms and that self-esteem had a more strongly negative association with DSO than with PTSD symptoms. Study 2 manipulated the level of self-esteem in the laboratory and found that participants in the high manipulated-self-esteem group reported fewer CPTSD symptoms than those in the low manipulated-self-esteem group. These findings advance our understanding of how childhood trauma contributes to CPTSD symptoms.

Acknowledgments

We extend our sincere gratitude to Yihan Yang for her help in data collection.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Due to privacy involved in this particular sample, the data sets used and analyzed during the current study are available from the corresponding author on a reasonable request.

Correction Statement

This article has been republished with a minor change. This change does not impact the academic content of the article.

Additional information

Funding

This work was supported by the Shanghai Morning Star Project Sailing Program (No. 22YF1411600), ‘Chenguang Program’ supported by Shanghai Education Development Foundation and Shanghai Municipal Education Commission (No. 22CGA26), the National Social Science Foundation of China (No. 22CSH092), the Research Project of Shanghai Science and Technology Commission (No. 20dz2260300), and the Fundamental Research Funds for the National Central Universities, Youth Interdisciplinary Innovation Team Project of Humanities and Social Sciences in East China Normal University (No. 2022QKT005).