Authors: Laura Råman, Mia Scheffers, Janet Moeijes, Bertus F. Jeronimus
Categories: Basic Research Article, Body image, body experience, sexual abuse, child maltreatment, adverse events, embodiment, Imagen corporal, experiencia corporal, abuso sexual, maltrato infantil, eventos adversos, corporalidad
Source: European Journal of Psychotraumatology
Authors: Laura Råman, Mia Scheffers, Janet Moeijes, Bertus F. Jeronimus
Background: The body is the medium through which humans experience the world, and the body is key to most suffering, healing, and clinical mental diagnoses. Body attitude refers to the affective, cognitive, and behavioural aspects of embodiment, which typically is more negative in clinical samples.
Objective: We examine how adult body attitude is associated with self-reported childhood abuse and neglect. We hypothesised that child sexual abuse is associated stronger with a negative adult body attitude than emotional or physical abuse/neglect would. Second, we expected that the association between body attitude and childhood sexual abuse was gender equivalent. Third, we expected a more positive body attitude in men than women after childhood physical abuse/neglect or emotional abuse/neglect.
Method: Body attitude was measured with the Dresden Body Image Questionnaire (DBIQ-NL) and the severity and type of childhood trauma with the Childhood Trauma Questionnaire-Short Form (CTQ-SF) in 749 Dutch adults aged 18-77. We fit multiple regression models and focused on childhood abuse and neglect with moderate to severe intensity.
Results: Childhood sexual abuse (∼15%), physical neglect (∼14%), emotional abuse (∼20%) and emotional neglect (∼30%) are associated with a more negative body attitude, while childhood physical abuse (moderate/severe, ∼6%) associated with a slightly more positive adult body attitude. Body attitude associations with childhood abuse/neglect were similar for both genders (no moderation).
Conclusion: Child maltreatment seems to precede the development of a more negative adult body attitude and more negative body experiences compared to individuals without child maltreatment.
We experience the world through our body and also most of our suffering, healing, and well-being (Geuter, 2016; Radley, 2000). ‘Out-of-tune embodiment’ and experiential issues such as body dissociation, somatic complaints, and/or body shame and dissatisfaction, are part and parcel of many psychiatric disorders; especially stress and trauma disorders, such as anxiety and mood disorders, and somatic symptom disorder (American Psychiatric Association, 2013; Rønberg, 2019; Scheffers, van Busschbach et al., 2017). Depression, for example, is marked by bodily experiences, next to depressed affect and thoughts (APA, 2013). Childhood maltreatment (CM, i.e. abuse/neglect) proves to be a key pathway to disrupted body experience (Bodicker et al., 2022) and is reported by up to ∼40% of adults around the world (Stoltenborgh et al., 2014). Also, European community estimates indicate a high prevalence of the five types of child maltreatment that are often distinguished (see Table 1), and which have strong adverse effects on lifespan development (Norman et al., 2012). Reductions in child maltreatment are therefore a key societal issue (Gilbert et al., 2009; Sethi et al., 2013). Table 1.Five types of childhood maltreatment (CM) and their definitions and European community prevalences.CM typesAbbr.Prev.DefinitionAbuse CEA+CPA+CSA. Prime examples are verbal abuse, humiliation, and acts that scare or terrorise a child.EmotionalCEA∼29%Acts of threat against a minor child that (could have) caused conduct, cognitive, affective or other mental disturbance, such as verbal abuse, excessive demands on a child’s performance that may lead to negative self-image and disturbed behaviour. Verbal assaults on a child’s sense of worth or well-being or any humiliating or demeaning behaviour directed towards a child by an adult or older person. Examples incl. disregard for a child’s requests or needs and manipulation of emotions, e.g. withholding affection, intentional social deprivation, intimidation/threats, or gaslighting.PhysicalCPA∼23%Actual or attempted infliction of physical pain or injury by adult or older person with or without use of an object or weapon and including use of severe corporal punishment. Examples include beating, shaking, choking, slapping, biting, and throwing objects.SexualCSA∼10%Unwanted and/or coercive (attempted) sexual contact and exposure to age-inappropriate sexual material, environments, or exploitation, often conducted by adult or older person. A dependent, developmentally immature child/adolescent (aged <18) is exposed to sexual activities which they do not fully comprehend, for which they cannot give consent, and/or which violate social taboo or family roles.Neglect CEN+CPN, primarily inadequate health care, supervision, protection from hazards, and/or deprivation of basic needs (clothing/food) and inattentiveness to a child’s emotional and developmental needs.EmotionalCEN∼18%Acts of deprivation against a minor child that caused or could have caused conduct, cognitive, affective or other mental disturbance, such as caretakers who fail to meet basic emotional and psychological needs, including love, belonging, nurturance, and support. One example is parents who fail to arrange appropriate education or failure in providing support, warmth.PhysicalCPN∼16%Failure to provide for a child’s basic survival needs, such as nutrition, clothing, shelter, hygiene, and medical care (e.g. caries). Physical neglect may also involve inadequate supervision of a child and other forms of reckless disregard of the child’s safety and welfare, such as seeking needed medical care.Body AttitudeAffective and cognitive evaluations of the body i.e. thoughts, perceptions, and beliefs, and the behavioural consequences.Acceptance ‘I wish I had a different body’. Reflects how much one accepts their body.Vitality ‘I am physically fit.’ The ability to live, grow, and develop, and to have energy to be vigorous and active.Self-aggrandisement‘I use my body to attract attention’. The way how the body is actively used in social interactions to enhance self-esteem.Physical contact‘I do not like people touching me’. The importance physical contact has in one’s relationships with other people.Sexual fulfilment‘I am very satisfied with my sexual experiences’. Evaluation of positive aspects of one’s own sexual experience and satisfaction with it.Note. The child maltreatment (CM) type definitions were adapted from Bernstein et al. (2003) and Myroniuk et al. (2023). European community prevalences from Gilbert et al. (2009) and Sethi et al. (2013). The five body attitude dimensions (DBIQ-NL) were derived from Pöhlmann et al. (2014) and Pöhlmann et al. (2008). Abbrev. = Abbreviation. Prev. = Prevalence. CEA = Child Emotional Abuse, CPA = Child Physical Abuse, CSA = Child Sexual Abuse, CEN = Child Emotional Neglect, CPN = Child Physical Neglect.
We aimed to examine how childhood maltreatment types are associate with adult body attitude, such as body acceptance, vitality, self-aggrandisement, physical contact and sexual fulfilment (e.g. Scheffers, van Duijn et al., 2017). From a biopsychosocial perspective, our body attitude is a combined outcome of biological factors (e.g. body size), psychological characteristics (e.g. negative affect, low self-esteem, perfectionism), childhood experiences (e.g. abuse and neglect, see Table 1), and sociocultural pressure from caretakers, family, peers, and media (Nichols et al., 2018). Childhood abuse and neglect can have enduring adverse effects on brain development (Hein & Monk, 2017) and adult well-being (van der Kolk, 2014), such as a 1.5–3.4 times higher risk of fulfilling common clinical diagnoses (Norman et al., 2012) and an 86% higher risk of premature mortality (D'Arcy-Bewick et al., 2022). Child maltreatment influences development via the multimodal systems that underlie adult self-awareness of the body and mind (Ogden et al., 2006; Tsakiris, 2017). We know child maltreatment types are associated with differences in emotion dynamics (Myroniuk et al., 2024) and specific somatic complaints (Eilers et al., 2023), among others. We also know young women typically report greater body dissatisfaction than men (Lacroix et al., 2023), and that child maltreatment predicts greater body dissatisfaction among young women (Momeñe et al., 2023). We examined whether differences in adult body attitude are associated with different types of child abuse/neglect. Below we first introduce embodied experiences and the concept of body attitude before, we present our models and synthesise and discuss our results.
One’s sense of self develops in social interactions between the infant and caregivers (Shai & Fonagy, 2014), where the body serves as a prime medium via touch, movement, and dyadic synchronisation (Ogden et al., 2006). Body experience evolves through different developmental processes, expanding a body-centred experience of the self (body–object) over time into an experience of the self as subject (body-subject; e.g. Joraschky (1988)) when children start to compare themselves with others (Lacroix et al., 2023). A multimodal internal body representation emerges during development from bottom-up and top-down processes that align the system levels (Kearney & Lanius, 2022; Tsakiris, 2017). The development of a healthy embodied self requires people to be able to distinguish between self and other (Joraschky, 1988), and to experience body ownership, which emerges from interactions between the developing individual and the environment, including caregivers. The quality of these childhood interactions may lead to different outcomes, depending on whether the interaction supports healthy development, or causes disruptions (Sigelman & Rider, 2021). Child maltreatment is the key example of disruptive interaction that can lead to trauma symptoms in adulthood, and is considered to be a distal risk factor for the development of a negative cognitive–affective body image (see meta-analysis by Bodicker et al., 2022), in support of the expected link between child abuse/neglect and adult body attitude.
Researchers use body experience as an umbrella term that captures at least the domains of body awareness, body satisfaction, and body attitude (Emck & Scheffers, 2019; Piran, 2019). Body awareness is defined as the perception of one’s bodily states, processes, and actions presumed to originate from sensory, proprioceptive, and interoceptive signals accessible to conscious awareness (Mehling et al., 2009). Body satisfaction refers to the degree of contentment with body appearance or functionality (Orlandi et al., 2006). In this paper, we focus on body attitude (Fazio & Olson, 2003), which refers to affective and cognitive evaluations of the body, as well as the behavioural consequences (Scheffers, Hoek et al., 2017), and thus captures a subjective and holistic view of an individual’s relationship with their body (Pöhlmann et al., 2008). Body image, in contrast, is a more narrow concept, that focuses on visual and perceptual aspects such as one’s dissatisfaction with specific body parts or one’s appearance, or the discrepancy between one’s perceived and ideal body size. By studying the broader concept of body attitude we aim for a deeper understanding of how individuals relate to their bodies (embodiment), including their emotional and cognitive investment towards the self. This approach allows for a more comprehensive assessment of body-related issues.
Maltreatment in childhood can profoundly influence body attitude through the disruption of a child’s developing sense of self and bodily integrity (Scheffers, 2018). Experiences of abuse or neglect often lead to an altered state of psychological arousal and chronic stress, which can result in a distorted body attitude as a protective mechanism, such as chronic detachment or self-objectification (Kearney & Lanius, 2022; van der Kolk, 2014). The two most harmful characteristics of traumatic events are the violation of the body-boundaries and of physical integrity (Scheffers, 2018; Scheffers, Hoek et al., 2017; van der Kolk, 2006). Previous studies have shown a negative body attitude to be associated with anxiety and depression (Röhricht et al., 2002) and trauma and eating disorders (Dyer et al., 2013), but we enrich the literature with this study of body attitude in a population sample.
Child abuse has an interpersonal nature and includes violation of physical integrity during vulnerable developmental periods and often circumstances a child cannot escape (Maercker et al., 2022). Sexual and physical abuse have been studied more often than emotional and physical neglect (Mennen et al., 2010; Stoltenborgh et al., 2014). Although there has long been awareness of emotional abuse and neglect being child maltreatment types with different downstream consequences, only a few validated instruments distinguish child maltreatment subtypes, which typically co-occur (Mennen et al., 2010; Myroniuk et al., 2024). The most used instrument to study child maltreatment types is the Childhood Trauma Questionnaire-Short Form (CTQ-SF; Bernstein et al., 2003), which distinguishes the child maltreatment types mentioned and studied in this paper (see Table 1). Child maltreatment has been studied widely in clinical populations because of its increased risk of common mental disorders (Borgmann et al., 2014; Dyer et al., 2013; Rohde et al., 2008).
Studies on the relationship between child maltreatment and body attitude in clinical populations have emerged recently (Scheffers, Hoek et al., 2017) but there is no data on their relationship in adult population samples, although associations with related concepts have been established (Bodicker et al., 2022). Thus, although a history of child maltreatment is linked to a variety of social, psychological and somatic problems later in life (Clemens et al., 2018; Sack et al., 2010; Scheffers, Hoek et al., 2017), the question remains whether and how these early traumatic experiences associate with adult body attitude in non-clinical adult populations. The differential associations between types of abuse/neglect and specific adult somatic complaints (Eilers et al., 2023) and emotion dynamics (Myroniuk et al., 2024) also suggest diverging relationships for body attitude. Moreover, studies of clinical populations suggest that sexual abuse leads to more, and more severe, problems in body experience and body satisfaction than other types of maltreatment (Sack et al., 2010). This paper aimed to explore the relationship between exposure to five child maltreatment types and adult body attitude in a sample of Dutch adults from the general population.
Research on gender differences in body image and body attitude concepts is inconsistent. Gender differences in body appreciation across forty studies in the general population indicated that men tend to be more appreciative of their bodies than women (He et al., 2020). A meta-analysis of gender differences in self-esteem concluded that in non-clinical populations men tend to have more self-esteem than women do (Bleidorn et al., 2016). The prevalence of child sexual abuse in women seems to be at least twice that in men (Sawyerr & Adam-Bagley, 2023), and the consequences of child sexual abuse in women received more attention than those in men (O'Leary et al., 2017). Clinical samples with early child maltreatment in women showed severely impaired self-reported body attitude (on the Dresden Body Image Questionnaire (DBIQ-NL; Scheffers, van Duijn et al., 2017)). There are indications that sexual abuse amongst men is underreported (O'Leary et al., 2017), as men tend to struggle with disclosure (often delayed by 20 years or more) and non-disclosure, compromised masculinity, shame, and confused sexuality (O'Leary et al., 2017), and clinical and societal bias (Gruenfeld et al., 2017). In the few studies that examined child sexual abuse in men, adverse long-term effects were observed, similar to women, including serious disturbances of mental functioning (O'Leary et al., 2017). Since sexual abuse violates the body-boundaries and physical integrity in both genders and other population studies of child sexual abuse did not observe salient gender differences (Myroniuk et al., 2024), we expected that child sexual abuse affected body attitude in both women and men in comparable ways. We also explore gender differences in the association between physical and emotional abuse/neglect and body attitude and expected affected men to report less negative body attitude than women do, on average, based on the aforementioned reviews.
In this paper, we examine the relationships between five different child maltreatment types (see Table 1) and adult body attitude. We hypothesise (H1) that sexual abuse has a more severe and negative impact on body attitude than physical or emotional abuse or physical or emotional neglect. Second, we examine gender differences in how the five types of child maltreatment are associated with adult body attitude. We expect (H2a) that sexual abuse is associated with similar differences in body attitude in both women and men. For physical (H2b) and emotional abuse and neglect (H2c), however, we hypothesise that men report a more positive body attitude than neglected/abused women do. We hope that our study results help propel our understanding of the link between different types of child maltreatment and body attitude and help identify the vulnerabilities that negatively influence lifespan development.
Data were derived from a Dutch crowdsourcing research project called HowNutsAreTheDutch (HND), which was designed to study mental health as a dimensional and dynamic phenomenon, characterised by both vulnerabilities and strengths (see van der Krieke et al., 2016). We selected adults (aged ≥18 years) who completed the Dutch version of the Dresden Body Image Questionnaire (DBIQ-NL) measuring body attitude (*N *= 811) and the Childhood Trauma Questionnaire-Short Form (CTQ-SF), which left us with a final sample of 749 participants. Comparisons between included participants and those who did not complete these instruments are provided in the supplement.
Dutch citizens were invited to participate in the HND project at www.HoeGekIs.nl by means of radio broadcasts, television, newspapers, magazines, podium discussions, and social media. Participants had to register on the website and create an account and could then complete multiple mental health questionnaires (or topical modules) and/or take part in a 30-day ecological momentary assessment (EMA, see van der Krieke et al., 2016). The HND platform was launched on December 19th, 2013, and the data set for the present study was extracted on December 19th, 2017. The first four questionnaires pertained to sociodemographic variables and affect/mood and well-being. After completing these, all other questionnaires became available, including the body attitude and childhood trauma questionnaires. The completion of all questionnaires was voluntary and could be completed in any order. The participants had no specific information about the kinds of studies that would be conducted on the collected data. The HND study was approved by the Medical Ethical Committee of the University Medical Center Groningen, Netherlands (registration numbers M13.147422 and M14.160855). All details and procedures are described in van der Krieke et al. (2016).
Child maltreatment was measured with the 28-item Childhood Trauma Questionnaire-Short Form (CTQ-SF, Bernstein et al., 2003). The CTQ-SF examines recollections of five dimensions of childhood trauma subdivided into physical, emotional, and sexual abuse, and physical and emotional neglect, with five items each. The frequency of each item is scored on a 5-point Likert scale from 1 (never true) to 5 (very often true). Clinical cut-off values are given for four trauma severity levels for each none or minimal (CEA: 5-8; CPA: 5-7; CSA: 5; CEN: 5-9; CPN: 5-7); mild (CEA: 9-12; CPA: 8-9; CSA: 6-7; CEN: 10-14; CPN: 8-9); moderate (CEA: 13-15; CPA: 10-12; CSA: 8-12; CEN: 15-17; CPN: 10-12) and severe trauma (CEA: ≥16; CPA: ≥13; CSA: ≥13; CEN: ≥18; CPN: ≥13) (see Bernstein & Fink, 1998). Although a retrospective questionnaire on child maltreatment may suffer from both underreporting and overreporting, previous work has found different scores for groups known to vary on childhood trauma, thereby demonstrating known-group validity (Thombs et al., 2009). Besides, the CTQ-SF has shown convergence with a childhood trauma interview, indicating similarity in self-rated versus clinician-rated reports of trauma (Bernstein & Fink, 1998) and with data from knowledgeable informants (Bernstein et al., 2003).
The internal consistency of the Dutch translation of the CTQ-SF administered to community and psychiatric populations varied in terms of Cronbach α between .63–.95 (Thombs et al., 2009). In the present sample, the subscales proved reliable (CPN α = .59, CPA α = .79, CSA α = .87, CEA α = .88, and CEN α = .92). Both Thombs et al. (2009) and Spinhoven et al. (2014) dropped the ‘molested’ item in the Dutch version, resulting in four rather than five items for the subscale sexual abuse. Although ‘molested’ more clearly refers to sexually abusive behaviour in English, the translated Dutch word does not necessarily have this sexual connotation. We chose to keep the dropped item (item 24) in the present study, to be able to compare our results to those in other studies using the English CTQ-SF or translations to other languages. Furthermore, this enables us to use the recommended cut-off scores (Bernstein & Fink, 1998). Good internal consistency of the sexual abuse subscale suggests that this was not a problem.
Body attitude was assessed using the Dutch version of the 35-item Dresden Body Image Questionnaire (DBIQ-NL; Scheffers, van Duijn et al., 2017). Dresden Körperbildfragebogen is an umbrella instrument derived from three German questionnaires of body ‘Fragebogen zum Körperbild’ (FKB20), ‘Fragebogen zur Bewertung des eigenen Körpers’ (FBeK), and the ‘Frankfurter Körper Konzept Skalen’ (FKKS, see Scheffers, van Duijn et al., 2017). The DBIQ-NL measures the multidimensional character of body attitude with five dimensions determined by factor body vitality, body acceptance, self-aggrandisement, physical contact, and sexual fulfilment, see Table 1 for items (Pöhlmann et al., 2008; Pöhlmann et al., 2014). Responses were scored on a five-point Likert scale from (1) ‘not at all’ to (5) ‘entirely’ and higher scores indicate a more positive body attitude. The internal consistency of the Dutch translation (DBIQ-NL) administered in a non-clinical sample varied from α = .83 for self-aggrandisement to α = .92 for sexual fulfilment (Scheffers, van Duijn et al., 2017). The alphas for the subscales in the present study body acceptance α = .81, physical contact α = .86, self-aggrandisement α = .91, vitality α = .91, and sexual fulfilment α = .92. Finally, scale invariance across sex and age were reported by Scheffers, van Duijn et al. (2017) who concluded that the questionnaire is invariant in non-clinical groups.
The sociodemographic covariates age (continuous) and gender (0 = female, 1 = male) were included as covariates in the analyses because of their correlation with the CTQ-SF (Witt et al., 2017) and DBIQ-NL (Scheffers, van Duijn et al., 2017) and because of their distribution in the sample (see Table 2). Table 2.General sample characteristics. Total (n= 749) Women (n= 579) Men (n= 170) MeanSDn MeanSDn MeanSDn Age 44.83(14.6) 42.60(14.3) 52.46(12.7) Category18–25 9612.8% 9015.5% 63.5%26–35 14018.7% 12521.6% 158.8%36–45 12216.3% 9917.1% 2313.5%46–55 17723.6% 13723.7% 4023.5%56–65 16822.4% 10217.6% 6638.8%66–75 435.7% 254.3% 1810.6%≥76 30.4% 10.2% 21.1%EducationLower 324.3% 203.5% 127.1%Medium 11415.2% 9115.7% 2313.5%High 60380.5% 46880.8% 13579.4%Relationship statusSingle 21528.7% 17129.5% 4425.9%Couple 53471.3% 40870.5% 12674.1%Body attitude^a^Total3.33(0.58) 3.28(0.59) 3.47(0.50) Child maltreatment^b^Sexual abuse6.19(2.75) 6.31(2.99) 5.79(1.64) Emotional abuse9.39(4.47) 9.63(4.62) 8.56(3.82) Physical abuse5.82(2.06) 5.82(2.14) 5.85(1.78) Emotional neglect12.48(4.82) 12.34(4.90) 12.94(4.50) Physical neglect6.95(2.44) 6.91(2.51) 7.09(2.18) Note*.* ^a^Dresden Body Image Questionnaire-Dutch translation, ^b^Childhood Trauma Questionnaire-Short Form. Participants were predominantly women (77.3%).
All analyses were carried out using IBM SPSS Statistics 29.0.1.0. For descriptives, the sum scores for child maltreatment types were computed. Bernstein and Fink (1998) categorised the severity of the trauma exposure of the child maltreatment types (see Table 1) into four categories, namely, none to minimal, mild, moderate and severe trauma. We recoded the categories into a 0 = none to mild exposure and 1 = moderate to severe exposure to examine differences in group means between participants with moderate to severe abuse and neglect (1) in terms of adult body attitude, to facilitate analytic ease and communication clarity, albeit at the cost of reduced statistical power (Iacobucci et al., 2015). This can be offset by our larger sample. We estimated the differences between (sub-)groups with t-tests. Mean differences between the subgroups (means divided by standard deviation) were expressed as Cohen’s d and group differences were considered to be small between 0.20 and 0.49, moderate between 0.50 and 0.79, and large if >0.80 (Cohen, 1992).
To explore the relationship between exposure to different types of trauma and body attitude, a multiple linear regression analysis was conducted. Childhood maltreatment subscale scores were taken as predictor variables of body attitude in adulthood. In the adjusted analysis gender and age were included as covariates.
To determine whether (H1) sexual abuse leads to a more negative body attitude than the other maltreatment types, several dummy variables were used. We noted that child maltreatment intensity (0/1) was indicated by ‘moderate’ or ‘severe’ exposure on a particular subscale, a dummy code that was used to estimate differences between (a) abuse (CPA/CEA), (b) neglect (CPN/CEN), and (c) no abuse/neglect. The presence of sexual abuse (CSA) was the reference group. In this analysis, the participants who suffered both neglect and abuse in the absence of sexual abuse were excluded.
In the second multiple regression analysis, everyone who reported childhood sexual abuse was included as the reference group (CSA). Dummy coding was used for all possible combinations of abuse and neglect (CPA, CEA, CPN and CEN) and the participants without child maltreatment experiences.
To test (H2a, H2b, H2c) for an interaction effect of gender and child maltreatment in the prediction of body attitude, a multiple linear regression analysis with the interaction terms (type of CM × gender) as independent variables and body attitude as dependent variable was conducted. Significance was defined as p ≤ .05 following convention.
Sample characteristics are given in Table 2. Participants were aged 18–77 years. Men were about a decade older than women, on average (Mage = −8.10, p<.001). Approximately 80% of the participants were highly educated. Secondary education was the highest completed level for 15%, and basic education for 4%.= 52.5 ± 12.7 vs. 42.6 ± 14.3, t(747)
The prevalence of the different types of child maltreatment in the sample is presented in Table 3. The cut-off scores for moderate maltreatment intensity were used for all subscales (Bernstein & Fink, 1998), which resulted in 43.1% of the participants who reported child maltreatment, most commonly emotional neglect (31.1%), followed by emotional abuse (19.5%), sexual abuse (15.2%) and physical neglect (13.9%). The prevalence of physical abuse (5.7%) was comparatively low. Table 3.Frequency and severity of child maltreatment.DummySeverity categorySexual abusenEmotional abusenPhysical abusenEmotional neglectnPhysical neglectn None or minimal52170%41555%66989%23331%52870% Mild11415%18825%375%28338%11716%0 63585%60380%70694%51669%64586% Moderate8511%659%213%10714%7410% Severe294%8111%223%12617%304%1 11415%14620%436%23331%10414%Note*. N* = 749. Child maltreatment was assessed with the Dutch translation of the Childhood Trauma Questionnaire-short form (CTQ-SF). See for definitions Table 1.
To examine whether sexual abuse stood out as a predictor of body attitude we fit a multiple regression analysis to predict adult body attitude with differences in five child maltreatment types (Model F(7, 741) = 11.87, p < .001). The five trauma types explained 10% of differences in body attitude (R² = .10). Sexual abuse, physical abuse and emotional neglect were significant predictors of body attitude differences, but physical neglect and emotional abuse were unrelated (see Table 4). There was an unexpected positive significant association between physical abuse and body attitude. Gender showed a significant main effect (B = 0.161, 95% CI [0.060, 0.262], p = .002). Age seemed unrelated to differences in body attitude.
Table 4.Associations between child maltreatment subtypes and body attitude. Crude analysisAdjusted analysis^a^ 95% CI 95% CI VariableB^b^SELLULpB^b^SELLULpPhysical abuse0.040.010.010.06.0020.030.010.010.06.005Emotional abuse−0.020.01−0.03−0.00.026−0.010.01−0.030.00.142Physical neglect−0.010.01−0.030.01.290−0.010.01−0.040.01.191Emotional neglect−0.020.01−0.03−0.00.019−0.020.01−0.03−0.01.004Sexual abuse−0.030.01−0.05−0.01<.001−0.030.01−0.04−0.01<.001Gender (♀ = 0) 0.160.050.060.26.002Age 0.000.00−0.000.00.522Note*. n *= 749. Results of multiple linear regression analyses with subscale scores of the Childhood Trauma Questionnaire-Short Form (CTQ-SF) as independent variables and body attitude as dependent variable. ^a^Adjusted for gender and age. ^b^Unstandardised regression coefficient. CI = Confidence Interval. SE = Standard Error. LL = Lower limit. UL = Upper limit. *p *= significance.
To examine whether sexual abuse predicted a more negative body attitude than nonsexual types of child maltreatment we used dummy coding for at least moderate severity (a) abuse (physical and emotional), (b) neglect (physical and emotional), and (c) no abuse/neglect. Sexual abuse was the reference category. The group having suffered both abuse and neglect without sexual abuse were excluded (resulting in n = 665 included). Indeed, child maltreatment predicted 6% of body attitude (R² = .061, F(5, 659) = 8.62, p<.001) and those without child maltreatment and neglected participants showed differences in body attitude (see Table 5). Abuse was unrelated to body attitude. Gender scores indicated that men scored significantly higher on body attitude than women.
Table 5.Associations between sexual abuse versus other types of abuse versus neglect and body attitude. Crude analysisAdjusted analysis^a^ 95% CI 95% CI VariableBSELLULpBSELLULpSexual AbuseReference group Abuse0.280.140.010.56.0450.260.14−0.010.54.060Neglect0.200.080.060.35.0060.180.080.030.32.020No CM0.330.060.210.44<.0010.320.060.200.43<.001Gender (♀ = 0) 0.170.050.060.27.002Age 0.000.00−0.000.00.887Note*. N* = 665. ^a^Adjusted for gender and age, CM = child maltreatment B = regression estimate. CI = Confidence interval. SE = standard error. LL = lower level or boundary. UL = upper level or boundary. *p *= significance. ♀ = women.
The model in Table 5 also shows that an individual without child maltreatment reported a more positive average body attitude (+0.32 points) than an individual who experienced sexual abuse. An individual having suffered abuse (excluding sexual abuse) as a minor, scored a more positive average body attitude (+0.26 points) than an individual who has suffered childhood sexual abuse, but the difference was not statistically significant. An individual having suffered neglect in childhood also reported a more positive average body attitude (+0.18 points) than an individual who has suffered sexual abuse. This value was significant.
When looking at the differences in body attitude between the groups that have suffered (a) sexual abuse, (b) all possible combinations of physical/emotional abuse/neglect, and (c) no abuse/neglect, the overall regression was statistically significant (R² = .059, F~(4, 744) ~= 11.75, p< .001).
An individual having suffered any type of child maltreatment other than sexual abuse, on average, reports a more positive body attitude (+0.16 points) than those exposed to sexual abuse. An individual with no experience of child maltreatment, on average, scores a more positive body attitude (+0.32 points) than those who experienced child sexual abuse. Both, the dummy variable abuse and neglect in one and no child maltreatment, were found to be significant predictors of body attitude. Gender showed a significant main effect in which men were more positive on average, while age seemed unrelated (see Table 6). Table 6.Associations between sexual abuse versus combinations of abuse & neglect and body attitude. Crude analysisAdjusted analysis^a^ 95% CI 95% CI VariableBSELLULpBSELLULPSexual AbuseReference group Abuse & Neglect0.180.070.050.31.0080.160.070.030.29.018No CM0.330.060.210.45<.0010.320.060.200.44<.001Gender (♀ = 0) 0.170.050.070.27<.001Age 0.000.00−0.000.00.676Note*. N* = 749. ^a^Adjusted for gender and age, CM = child maltreatment. B = regression estimate. CI = Confidence interval. SE = standard error. LL = lower level or boundary. UL = upper level or boundary. ♀ = women.
To explore gender differences in the associations, first, an independent samples t-test was performed to compare the scores of the child maltreatment types between women (*n *= 579) and men (*n *= 170). As shown in Table 7, women reported more sexual abuse (*d *= 0.19) and emotional abuse (d = 0.24) than men did. The other child maltreatment scales did not differ. Table 7.Independent samples t-test between women and men for five childhood maltreatment (CTQ-SF) types. WomenMen 95% CI^a^ (n = 579)(n = 170)dftpCohen’s dLLULCSA (M, SD)6.31(2.99)5.79(1.64)7472.16.0130.1890.0170.360CEA (M, SD)9.63(4.62)8.56(3.82)7472.77.0060.2410.0700.413CPA (M, SD)5.82(2.14)5.85(1.78)747−0.21.834 CEN (M, SD)12.34(4.90)12.94(4.50)747−1.43.153 CPN (M, SD)6.19(2.51)7.09(2.18)747−0.86.392 Note. CEA = Child Emotional abuse. CEN = Child Emotional neglect. CPA = Child Physical abuse. CPN = Child Physical Neglect. CSA = Child Sexual abuse. CI = Confidence interval. *df *= degrees of freedom. *p *= significance of test. *t *= t-test, see method section. LL = lower likelihood. UL = Upper likelihood.
To determine whether gender moderated the influence of child maltreatment on body attitude (H2a-c), interaction terms were calculated for gender and the types of child maltreatment. In Table 8 we present the results of Model 2 which predicted 10.2% of body attitude (R² = .102, F~(7, 737) ~= 7.63, p<.001). Model 2 was expanded with interaction terms for gender, which accounted for an additional 0.2% of the variation in the body attitude scores, but these interaction terms were statistically nonsignificant. Childhood sexual abuse, physical abuse, emotional neglect and gender reached significance as main effects. Childhood emotional abuse and physical neglect were not significant predictors, as was the case also in Model 1 (not including the interaction terms). Table 8.Multiple childhood abuse/neglect predicting adult body attitude and potential moderation by gender.Model 95% CI BSELLULp1.CSASexual abuse −0.030.01−0.04−0.01<.001 CEAEmotional abuse −0.010.01−0.030.00.136 CPAPhysical abuse 0.030.010.010.06.005 CENEmotional neglect −0.020.01−0.03−0.01.004 CPNPhysical neglect −0.010.01−0.040.01.192 Gender (♀= 0) 0.170.050.070.27<.0012. CSA −0.030.01−0.04−0.01.003 CEA −0.010.01−0.030.01.258 CPA 0.030.010.010.06.020 CEN −0.020.01−0.03−0.00.020 CPN −0.020.01−0.040.00.102 Gender 0.160.050.060.26.002 CSA * gender −0.010.03−0.070.05.698 CEA * gender −0.010.02−0.040.03.770 CPA * gender 0.020.03−0.040.08.594 CEN * gender −0.010.01−0.030.02.748 CPN * gender 0.030.03−0.020.08.268Note*. N *= 749. Childhood Trauma Questionnaire-Short Form (CTQ-SF) was used to measure maltreatment, see method section for details. B = regression estimate. SE = standard error. CI = Confidence interval. LL = Lower level. UL = Upper level. *p *= significance.
Table 9 and Figure 1 show the mean scores for body attitude and the mean scores for the five subscales with the sample divided into three groups (a) sexual abuse present, (b) abuse and/or neglect present (nonsexual), (c) no child maltreatment. Figure 1.Body Attitude (DBIQ-NL) total mean score and scores per subscales.Note. Body attitude total score and scores per five subscales in three groups as measured with the DBIQ-NL. DBIQ-NL = Dresden Body Image Questionnaire-Dutch translation. Table 9.Total score per subgroup and DBIQ-NL subscale scores per subgroup. Sexual abuse^a^Mean (SD)Abuse and/or neglect^b^Mean (SD)No CM^c^Mean (SD)N114209426Body attitude^d^ mean score3.10^e,f^(0.65)3.27^g^(0.62)3.42(0.51)Body acceptance^d^3.25^e^(0.90)3.41^g^(0.89)3.67(0.76)Physical contact^d^3.37^e^(0.86)3.48^g^(0.86)3.69(0.70)Sexual fulfilment^d^3.11^e^(1.00)3.36(0.98)3.36(0.90)Self-aggrandisement^d^2.64^e^(0.62)2.79(0.67)2.89(0.59)Vitality^d^3.19^e,f^(0.85)3.41^g^(0.79)3.58(0.71)^a^Sexual abuse = Child sexual abuse all cases, ^b^Abuse and/or neglect = all cases having suffered nonsexual child maltreatment, ^c^No CM = Child maltreatment not present, ^d^Measured with Dresden Body Image Questionnaire-Dutch translation, ^e^significant difference between Sexual abuse and No CM (p < .05), ^f^significant difference between Sexual abuse and Abuse and/or neglect (p < .05), ^g^significant difference between abuse and/or neglect groups and those without CM (p < .05).
In this study, we examined associations between body attitude and five types of child maltreatment in 749 adults from the Dutch general population (see Tables 1 and 2). Participants who experienced childhood sexual abuse reported more severe adult problems in body attitude than other child maltreatment types, when analysing the five trauma types as separate variables, or all five types combined, when especially sexual abuse helped to differentiate between adult body attitude, whereas emotional and physical abuse provided no additional information. Body attitude was also lower in participants who reported emotional abuse and emotional and physical neglect (vs. participants without). Three additional key observations were that childhood physical abuse (moderate/severe) was associated with a more positive adult body attitude. Associations between child maltreatment types and adult body attitude were similar for both genders (i.e. no moderation). Emotional neglect was most common and reported by ∼30% of the respondents, and emotional abuse by ∼20%, sexual abuse by ∼15%, physical neglect by ∼14%, physical abuse by ∼6%. These five observations will now be discussed in more detail below.
Body attitude, including cognitive, affective and behavioural components, was poorer in adults with child maltreatment experiences than among those without. Maltreatment seems to disturb the healthy development of the self, which starts in infancy and is first and foremost situated in the body in the early phase (Tsakiris, 2017). As outlined in the introduction, a healthy development of the embodied self requires people to be able to distinguish between self and other, thus establishing stable body boundaries (Joraschky, 1988). Maltreatment in the early attachment period can destabilise these boundaries, complicating this differentiation process, as is evident in disturbances in social information processing networks (e.g. Hein & Monk, 2017; van der Kolk, 2014). Child sexual abuse and physical abuse are considered a violation of the developing body boundaries (see Table 1), and are seen as disruptive to healthy development. A lower body attitude score was found in the group that had suffered childhood sexual abuse than in the group that had suffered nonsexual child maltreatment types or no child maltreatment.
Previous work also detected that body dissatisfaction was stronger among women who survived sexual trauma than those with nonsexual trauma or without trauma (Sack et al., 2010). However, Sack et al. (2010) did not specify at what age the participant experienced the traumatic event, as often is the case in retrospective studies. A survey of adolescents (15-year olds) found that sexual abuse history was associated with poorer body image and lower self-esteem in the victims compared to non-victims (Mignot et al., 2018). Our study was restricted to adults, while maltreatment was limited to childhood and adolescence (Bernstein et al., 2003). None of the reviewed studies collected more specific information on the age when child maltreatment occurred, even though age gives a good indication of the developmental phase of a child which seems to influence the consequences of (sexual) abuse (e.g. Andersen et al., 2008). Sexual development in adolescence is an example of such a specific phase, when sexual thoughts, feelings, and behaviours change and mature (Sigelman & Rider, 2021). Positive sexuality development is considered beneficial for well-being outcomes (Maes et al., 2023). Variation in outcomes among adults exposed to childhood maltreatment may be explained (in part) by the timing of maltreatment or specific windows of vulnerability in development, although the literature on such age windows of vulnerability for child maltreatment remains inconclusive (Schaefer et al., 2022).
Sexual abuse and physical abuse are violations of the body boundaries and physical integrity. In the current study sexual abuse correlated, as expected, negatively with body attitude, whereas physical abuse showed surprisingly positive correlations. Different attitudes towards the maltreatment subtypes may play a role. In past decennia, a fundamental change can be detected, especially in the attitude towards physical abuse (Bullinger et al., 2020). The boundaries between right and wrong are mostly clear-cut when considering child sexual abuse. On the other hand, some forms of physical abuse, such as spanking or corporal punishment, have long been culturally considered an accepted parental disciplinary method (Benjet & Kazdin, 2003). Since 2007, the Dutch Civil Code has stated that caregivers are not allowed to use physical violence when educating or correcting children’s behaviour. The cultural change towards no acceptance of parental disciplinarian behaviours falling under physical abuse unfolded over years before it became part of the Civil Code. The low occurrence of physical abuse in the current sample is hopefully an effect of the change in legislature and the general attitude, as child maltreatment rates have been steadily decreasing in the past decades (e.g. Bullinger et al., 2020).
The positive correlation between physical abuse and body attitude may be explained by a study on the consequences of spanking, which reported little evidence that mild and incidental physical punishment would have strong negative consequences for the development of the child (Benjet & Kazdin, 2003). One explanation might be that childhood physical abuse always contains interaction, however negative. From a developmental perspective, even negative interaction, presumably with the caregivers, might satisfy a need for interaction or satisfy a sense of being seen. Alternatively, once children become adults, they can no longer be so easily physically hurt and dominated, which increases their sense of control. Previous work connected physical abuse to adaptive and positive adult outcomes such as higher positive affect and lower anxiety and depression (Myroniuk et al., 2024; Sudbrack et al., 2015). In the present dataset, physical abuse is also associated positively with well-being domains such as personal autonomy (Jeronimus et al., 2024).
With regard to gender differences, we hypothesised that victims of sexual abuse would show no difference in body attitude score, while for the other four child maltreatment types we expected that women would report a lower body attitude, based on studies in which women reported worse on body image related domains than men (Cash et al., 2004; He et al., 2020; Longobardi et al., 2022). In our sample, none of the interaction terms were found to be significant. Thus, the association between any of the physical or emotional maltreatment types and body attitude was not dependent on gender, as was also the case for sexual abuse.
The maltreatment subtypes have not received equal attention in the past. Research on the prevalence and effects of sexual abuse emerged some four decades ago and was soon followed by research on physical abuse and only later on the other types of maltreatment (Stoltenborgh et al., 2014). When the five child maltreatment types were included separately as predictors of body attitude, emotional neglect, sexual abuse and physical abuse were found to be significant predictors of body attitude. In a German general population sample aged 14–21, Wendler-Bodicker et al. (2023) found emotional neglect and emotional abuse to be significant predictors of body dissatisfaction. In our adult sample, as in the youth sample of Wendler-Bodicker et al. (2023), the occurrence of emotional neglect was most frequent compared to the other four maltreatment subtypes.
One characteristic difference between abuse and neglect is that abuse can be a onetime event, whereas neglect is more structural and happens over a longer period of time. Shai and Fonagy (2014) postulate that the development of the subjective self depends on the verbal and non-verbal interaction with the caregiver. In childhood emotional neglect there is a lack in the amount of contact between the child and caregiver, the child’s emotions are not seen and validated, or the presence of the caregiver is completely missing (Table 1). Thus, emotional neglect is the omission of essential and necessary components for normal development. A possible explanation for an altered body attitude after emotional neglect could be its effect on the attachment relationship. Associations between attachment styles and body appreciation in adolescents with and without eating disorders indicated that safe attachment is related to a higher body appreciation than dysfunctional attachment styles, which was observed both in the clinical group and among ‘healthy’ controls (Laporta-Herrero et al., 2022). Evidently, child attachment styles have consequences for the development of adult body experience in a broader sense and specifically the attitudes and behaviours one applies to one’s body later in life. Based on the high prevalence of emotional neglect in retrospective reports and the harmful consequences for attachment and further development, neglect as a type of child maltreatment deserves more attention.
As different types of maltreatment often co-occur, the influence of the separate forms of maltreatment on body attitude should be interpreted with caution. Clemens et al. (2018) studied the effects of child maltreatment on long-term physical health and found a cumulative effect when a combination of two or three maltreatment types was experienced. Body attitude could also be negatively affected by the cumulative effect of multiple child maltreatment types, as we observed a substantial overlap in the maltreatment subtypes in our sample.
Strengths of the present study include the large community sample including men and women of a wide age range. The study of child maltreatment is built on predominantly female-samples and a strength of our study lies in the crowdsourcing data collection method in the HND-project, which allowed participants to fill out the questionnaires in any order they liked (reduces order effects), and participants perceived themselves to be anonymous, which reduces social desirability and self-representation bias. At the time of filling out the questionnaires participants had no knowledge of which questionnaires would later be combined in research. The online crowdsourcing strategy may result in recruitment bias, and the HND sample comprised predominantly women and higher-educated individuals, but was in many ways more diverse and representative of mental health states or elderly Dutch than the large population samples in the Netherlands (e.g. Lifelines, NESDA), see van der Krieke et al. (2016) for details.
Sexuality is often a problematic body-related issue in patients with mental disorders as well as somatic disorders and instruments measuring attitudes towards sexuality are scarce. The subscale of sexual fulfilment in Dresden Body Image Questionnaire measures just that and the subscale emerged as a separate factor in the original German version (Pöhlmann et al., 2008) as well as in the Dutch version (Scheffers, van Duijn et al., 2017). Filling out a self-report questionnaire via the online crowdsourcing strategy has enabled the collection of data in this area in the Dutch adult population.
One limitation is the cross-sectional design which does not allow causal inferences. A further limitation is the use of retrospective self-report questionnaires for data collection, which may result in recall bias, and manifest in several ways (e.g. Baldwin et al., 2024). Young adults report their childhood experiences closer in time than older adults and could therefore be expected to have less memory bias. The lack of distance to the experiences at a younger age might also be problematic if hurtful memories are repressed or denied. Older adults have had more time to reflect on their upbringing and the relationship with their caregivers, which might help them to come to terms with the possible shortcomings, accepting the past for what it is. On the other hand anonymity may help the respondent answer the questions truthfully by diminishing the fear factor of speaking against one’s caregivers or by lifting the need to protect one’s childhood memories, regardless of age. After all, in the field of child maltreatment anonymous retrospective self-reports are a highly valuable source of information (Clemens et al., 2018).
The DBIQ has been studied for its psychometric properties in Western cultures only. It is possible that in non-Western cultures the concept of body attitude based on the present items would take a different factorial structure. Results thus cannot be generalised. The influence of culture on body attitude should be explored in future studies.
Our study lacked exact information on the age one was maltreated, which is a shared feature in studies using the Childhood Trauma Questionnaire-Short Form (CTQ-SF). This questionnaire determines the presence or absence of abuse and neglect during childhood and adolescence and gives an indication of the severity per maltreatment subtype. The CTQ-SF does not capture information on whether an experience was ongoing, or during which age-related developmental phase it took place. In future studies, more specific information on age and duration of child maltreatment is relevant to identify ‘windows of sensitivity’, such as the first five years of life. However, child maltreatment occurring during, or across, different developmental stages, may affect body attitude in different ways.
To conclude, in the past child maltreatment has been shown to influence different domains of well-being. The current study indicated a high risk that the domain of body attitude is also affected. Child maltreatment can be seen as a risk factor for the development of a more negative body attitude. Although body attitude is a subjective evaluation of one’s own feelings, thoughts and beliefs about one’s body, these evaluations are formed during interactions with others. Harmful childhood interactions, such as abuse and neglect, may result in a problematic body attitude far into the future. It is important that this link gains more attention. Experiences of child maltreatment are common as is shown in our non-clinical sample. And after all, body attitude deserves more attention because it is a central theme in every human being.