(800) 681-3114

Childhood abuse: Differential gender effects on mental health and sexuality

M. Abrams a,b,* , M. Milisavljevi ́c c , A. ˇSoˇski ́c c

a Psychology for NJ, LLC, 120, Mountain Park Road, Clifton, NJ 07013, USA 
b New York University, 70, Washington Square South, New York, NY, 10012, USA
c University of Belgrade, Studentski trg 1, 11000 Belgrade, Serbia

Available online 24 July 2019


  • Childhood abuse;
  • Gender differences;
  • Atypical sexuality;
  • Psychopathology;
  • BPD;
  • PTSD


Objectives. — Childhood abuse is linked to many maladaptive outcomes in adulthood, but its effects on adult sexuality are rarely explored. The goal of this study was to explore adult correlates of childhood abuse, related to both sexual fantasies and behavior, as well as mental health. Moreover, the relationship of these outcomes and gender was explored.
Methods. — Surveys exploring sexual activity and fantasies, and psychopathological symptoms were conducted online on two groups of adults — those not abused in childhood, and those abused during their pre-teen age by close family members (sample of 349 participants).
Results. — Atypical sexual fantasies were more common in the abused than in the non-abused males, while the same relationship was not observed in the females. Similar tendencies, albeit not as strong, were seen in the case of sexual behaviors. On the other hand, both man and abused women were more prone to developing psychological symptoms, in comparison to non- abused group. However, this relationship was more pronounced in the females. Moreover, high tendency for borderline personality disorder was registered in both, abused males and females, but tendency for posttraumatic stress disorder was only increased in the females.
Conclusions. — Findings supported the hypothesis that gender moderates the outcomes of child- hood abuse, with the abused males experiencing more disturbances in the sexuality domain, and the females experiencing more psychological symptoms. This is in accordance with findings claiming that male sexuality is more likely to be influenced by developmental events, while the females tend to experience more psychological symptoms in the face of childhood abuse.
© 2019 Elsevier Masson SAS. All rights reserved.

DOI of original article:https://doi.org/10.1016/j.sexol.2019.07.001.
La version en franc ̧ais de cet article, publiée dans l’édition imprimée de la revue, est également disponible en ligne : https://doi.org/10.1016/j.sexol.2019.07.001.
* Corresponding author. 120, Mountain Park Road, Clifton, NJ 07013, USA.E-mail address: mike.abrams@nyu.edu (M. Abrams).
1158-1360/© 2019 Elsevier Masson SAS. All rights reserved.


Early life abuse by a family member is associated with a range of pathologies throughout the lifespan. The effects of childhood abuse on sexuality have not been thoroughly explored, yet studies suggest that childhood abuse can influ-ence adult sexuality. Although much of the sexual behavior is innate, in humans it remains flexible and is shaped by devel-opmental variations, traumas, and cultural demands (e.g., Bowlby, 1969). One study demonstrated that nearly half of the clients attending sex therapy have a history of child- hood sexual abuse (Berthelot et al., 2014). Childhood sexual abuse was correlated with sexual avoidance and compulsiv- ity (often co-existing) (Vaillancourt-Morel et al., 2015). In females, early life sexual abuse has been associated with revictimization in adulthood, as well as anxiety, fear and sui-cidal tendencies (Beitchman et al., 1992; Messman-Moore & Long, 2000). Further, Abrams and Stefan (2012) observed in a clinical sample that women who were severely sexually abused as youths are prone to sexual masochism, self- destructive lifestyles, and borderline personality disorder (BPD). Females also seem to experience more psychopatho- logical effects of childhood sexual abuse (Rind et al., 1998). In males, childhood abuse seems to be related with adult sexual dysfunction and paraphilias (Abrams, 2016; Seibel et al., 2009). Some findings suggest that male sexuality is more likely to be affected by disturbances during the key age of sexual development (Bowlby, 1969; Harlow and Harlow, 1962).

In order to explore the relationship of childhood abuse and outcomes in adulthood, a study was conducted inves- tigating the psychological symptoms and atypical sexual fantasies and acts in adults who were abused during child- hood, and those who reported no abuse. According to the extant studies, it was expected that gender would moderate the outcome of abuse — males are expected to experience more sexual disturbances, and females more psychological symptoms.



A sample of 349 people was surveyed online. Participants were separated in two groups — 149 participants reported being abused in their childhood (cut-off age was 12 — abuse was experienced in pre-teen age, age particularly sensitive in sexual development (Bowlby, 1969)), while 200 partici- pants reported no childhood abuse. Nature, severity and duration of the abuse were not specified. Current findings suggest that the type of abuse might not be the crucial fac- tor influencing its outcomes (Cecil et al., 2017). Since abuse by immediate family members is considered the most trau- matizing form of abuse (Courtois, 1988; Palmer et al., 1999) only individuals abused at homes by immediate family mem- bers (parents or siblings) were included in the study. Among abused participants, there were 78 males, and 71 females, and in the non-abused sample there were 107 males and 93 females.


The survey included questions about the psychological symp- toms and atypical sexual desires (both, in fantasy and acted upon). The lists of fantasies/behaviors/symptoms were pro- vided, and the participants were asked to check each one that applies to them. The psychological symptoms espe- cially focused on those associated with posttraumatic stress disorder (PTSD) and BPD (based on the DSM-5 criteria), as studies suggest that these two disorders are com- monly associated with the childhood abuse (Bounoua et al., 2015; Roberts et al., 2012). However, the symptom items were worded non-clinically. The list of psychopathologi- cal symptoms included: depression, panic attacks, phobia, anger, sadness, intrusive thoughts, anxiety, feeling mis- understood, feeling betrayed, loneliness or feeling alone, splitting (swinging from idealization to anger), other symp- toms, and no symptoms. The list of sexual behaviors and fantasies was comprised from most common paraphilias from the website Fetlife.com (a web based social network for people interested in BDSM, fetish and kink), chosen paraphilias from the comprehensive list developed by John Money (Money, 1984), and insights gained from interviews with BDSM experts, sex therapists and clients (Abrams, 2016). The list included: orgies, bondage, domination, cuck- olding, crossdressing, nudism, verbal abuse, sex assault, masochism, sadism, swinging, and submissiveness, and no such fantasies/experiences.


Participants were recruited via CrowdFlower (now Figure 8), an Internet service that promotes the ‘‘crowdsourcing’’ services of its participant base. Crowdflower, like Amazon’s Mechanical Turk (MTurk), has been shown to be a reliable source of online survey data (Zhai et al., 2013). Participants were reimbursed for their completion of the survey with a predefined amount of money. Only the highest rated of par- ticipants (based on their reliability in prior surveys or work tasks) were included in the study.
Participants were presented with the primary aim of the study, technical details, and the option to withdraw at any time. The participants were provided with a resource they could contact in case the survey triggers any issues. They were asked to affirm that they were answering candidly with the caution that the gathered information will be used clinically.


In all analyses there were two independent variables, both categorical and binary: gender (with levels male and female) and presence of pre-teen abuse (levels were abused and not abused). There were three groups of dependent variables exploring sexual and psychological manifestations, derived from the questions regarding:

  • fantasies about atypical sexual activities (13 variables);
  • previous engagement in these sexual activities (13 varia-bles);
  • presence of emotional psychopathological symptoms (12 variables).

Two additional variables regarding presence of psy- chopathological syndromes were calculated. All mentioned variables were categorical and binary — yes/no type (fan- tasy/behavior/symptom present or absent).


Three new ratio variables were created, by calculating the percent of sexual fantasies, sexual acts actually per- formed, and psychological symptoms that each participant had endorsed. On these new variables, analyses of variance were performed.

Analyses were also performed to examine the influence of abuse on each of the sexual fantasies, the sexual behaviors and the psychological symptoms. The association between each dependent variable and the presence of pre-teen abuse was examined in the males and females separately. Since both, independent and dependent variables were categori- cal, one-sided Fisher exact tests were used for each of the dependent variables. Fisher’s exact test was used because a significant portion of cells had counts below 10, making Pearson Chi-Square unsuitable. In the case of psychopatho- logical syndromes, Chi-Square was used, as conditions for its use were satisfied. Odds ratios and confidence intervals (95%) were also calculated and are provided in the tables.


Abuse and atypical sexual (and paraphilic) fantasies

Analysis of variance revealed significant effects of abuse (F(1, 345) = 5200, P = .000) and gender (F(1, 345) = 6512, P = .023) on the percent of sexual fantasies endorsed. The participants who reported childhood abuse and the males were more likely to engage in more atypical sexual fantasies. The analysis of gender, abuse, and fantasies showed no significant three-way effects. However, five of 13 varia- bles were associated with early life abuse in the male participants, and no such association was found in the females: orgies (P = .011), bondage (P = .027), verbal abuse (P = .036), swinging (P = .009), and having none of the fan- tasies (P = .031) (non-abused men marked this option more often). Only for cuckolding the association was significant for both genders (P = .024 for the males and P = .014 for the females). The effect of abuse in all cases was such that the fantasy was more common in abused participants Table 1.

Abuse and unusual sexual (and paraphilic) behavior

As it was expected, the frequency of atypical sexual expe- riences was in all cases lower than that of fantasies, which could have affected the possibility of detecting potential relationships. Analysis of variance on the percent of expe- riences marked showed neither significant effect of gender, nor of abuse (interaction of factors was also non-significant). Comparison of the male and female samples showed that there was a different pattern of results in the two groups.

The abused male participants had more frequent atypical sexual experiences overall (significant (P = .013) effect on the variable ‘‘no such experiences’’), compared to the men who were not abused in pre-teens. On the other hand, the abused women were more likely to report being sexually assaulted, than the women who were not abused (P = .043), which is the only result that differs from the general trend. Percentages for each of the experiences are available at Table 2.

Abuse and clinical symptoms

Analysis of variance on the frequency of psychological symp- toms showed the significant interaction of gender and abuse (F(1, 345) = 7987, P = .005). The increase in the number of symptoms marked by the abuse sample was more pro- nounced in the females, than in the males (Graph 1). Main effects of abuse (F(1, 345) = 43,596, P = .000) and gender (F(1, 345) = 11,597, P = .001) were also significant — more psychological symptoms were marked by the abused parti- cipants, and the females.

Analysis by gender revealed more about the relationship of abuse and psychological symptoms. The two-way analysis showed that all symptoms were significantly more common in the abused females. In the male participants significant increases were present in 4 out of 13 variables, but the effect was smaller than in the females: depression, anger, recurrent unwanted thoughts, and feeling betrayed by close people. All frequencies are shown in Table 3.

Psychopathological syndromes

Participants were asked about a number of psychological symptoms, clusters of which approximated the DSM-5 diag- nostic criteria for BPD and PTSD. This survey was not meant to be a diagnostic instrument, and therefore it might be more precise to discuss the tendency toward PTSD and BPD. Tendency for PTSD was characterized by the presence of at least 3 of the following 5 symptoms: depression, panic attacks, phobias, intrusive thoughts, and recurrent anxiety. Tendency for BPD was assessed by detecting the presence of 4/7 following symptoms: depression, often feeling mis- understood, excessive anger, periods of prolonged sadness, often feeling betrayed, often feeling alone or lonely, and splitting. As expected, tendencies toward both syndromes were observed significantly more among the abused partici- pants of both genders based on Pearson Chi-Square (PTSD: 2(1, n = 349) = 23.75, p = .000, odds ratio: 3.7, CI 95%: 2.1—6.4; BPD: 2(1, n = 349) = 28.19, p = .000, odds ratio: 4.4, CI 95%: 2.5—7.8). Tendency for PTSD was increased by abuse in the females (2(1, n = 349) = 22.53, p = .000, odds ratio: 5.5, CI 95%: 2.6—11.4), but not in the males. However, tendency for BPD was elevated in all abused participants, female (2(1, n = 349) = 21.97, p = .000, odds ratio: 6.4, CI 95%: 2.8—14.8) and male (2(1, n = 349) = 7.63, p = .006, odds ratio: 3, CI 95%: 1.3—6.7). Consistent with other findings,the relationship was stronger in the female sample.

Figure 1 The percent of the psychological symptoms marked by non-abused and abused males and females.


In this study, relationship of childhood abuse with adult atyp- ical sexual acts or fantasies and with symptoms of PTSD and BPD was explored. A distinct gender difference in the impact of the abuse was detected, consistent with some previous research (Abrams, 2016; Abrams and Stefan, 2012). In accor- dance with our hypotheses, childhood abuse in this study was related to the increase in atypical sexual fantasies or behav- iors, primarily in the males, and with the increase of BPD and PTSD symptoms in both genders, but more pronounced in the females.

Results of this study confirmed the expectations regard- ing moderating effect of gender on effect of abuse on sexuality — in fantasy and behavior. Specifically, the abused males reported having more atypical sexual desires (fan- tasies). Although results of Anova do not show interaction of gender and abuse, detailed analysis of the frequency of paraphilic and unusual fantasies shows that they tend to increase more in the abused men than in the abused women, with the exception of cuckolding, which increased simi- larly in both abused males and females. This is particularly interesting, as cuckolding is the rarest occurring fantasy. It could be hypothesized that this fantasy may stem from the complicated relationship toward sexuality, described by Vaillancourt-Morel et al. (2015). Significantly, the fantasy of participating (versus actual participation) in orgies, swing- ing, nudism, and especially cuckolding tend to be associated with sexual masochism (Abrams, 2016).

Atypical sexual tendencies were more often left in fan- tasy than brought to action. This influenced the lower number of significant effects, yet the general tendencies remained the same. These activities were more common in the males, as the non-abused males more often reported not having any of these experiences. The noteworthy result was the increased number of the females who reported being sexually assaulted in abused group. Since the question regarding the atypical acts was not limited to adulthood, it is possible that sexual assault in question was the very childhood abuse they have previously reported. However, in some participants this could be the consequence of lifestyle and risky sexual behaviors influenced by early-life abuse (Norman et al., 2012).

The relationship between psychopathological symptoms and syndromes and abuse was more pronounced, with all symptoms and syndromes being more common in those abused in childhood. Furthermore, there was an interaction between abuse and gender — the abused females experi- enced a greater number of different psychological symptoms than the abused males.

According to the results, childhood abuse is likely to be related to distinct consequences in adulthood, both in sexuality and psychological well-being. However, it appears that gender moderates these outcomes. Interest in atypical sexual experiences is higher in the male sample, with practi- cally no association in the women. This is also in accordance with the fact that male sexuality seems to show more plas- ticity in response to developmental disturbances (Bowlby, 1969; Chivers et al., 2004; Harlow and Harlow, 1962). When it comes to emotional or personality symptoms, both abused men and women were affected; however, the effects were stronger in the female sample, which is in accordance with previous studies (Rind et al., 1998).


The current study supports the notion that early life abuse permeates different areas of adult functioning, yet males and females face somewhat different consequences. Further studies should address the question whether these outcomes differ in relationship to the type of abuse, its severity and duration, as well as the relationship between the abuser and the victim. Moreover, while the occurrence of atypical sexual interests was higher in the abused sample, the fact that there were no meaningful increases in atypical sexual acts requires more study. Further research should focus on exploring the sexual function of people abused in childhood, and whether they experience heightened distress from their atypical sexuality.

Disclosure of interest

The authors declare that they have no competing interest.


This study was funded by Psychology for NJ, LLC, a clinical research corporation that found the study to meet all of its ethical guidelines. Authors report no financial or other conflicts of interest.

All procedures performed in studies involving human par- ticipants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent was obtained from all individual par- ticipants included in the study. All participants were cautioned about the content of the study, had the right to withdraw at any time, and were offered consultations with a licensed psychologist at any point in the process. The study was conducted anonymously.


Abrams M. Sexuality and its disorders. Thousand Oaks. CA: Sage Publications, Inc; 2016.

Abrams M, Stefan S. Sexual abuse and masochism in women :etiology and treatment. J Evidence-Based Psychotherapies2012;12(2):231.

Beitchman JH, Zucker KJ, Hood JE, DaCosta GA, Akman D, Cassavia E. A review of the long-term effects of child sexual abuse. ChildAbuse Negl 1992;16(1):101—18.

Berthelot N, Godbout N, Hébert M, Goulet M, Bergeron S. Prevalenceand correlates of childhood sexual abuse in adults consulting forsexual problems. J Sex Marital Ther 2014;40(5):434—43.

Bounoua N, Felton JF, Long K, Stadnik RD, Loya JM, MacPhersonL, et al. Childhood emotional abuse and borderline personalityfeatures: the role of anxiety sensitivity among adolescents. PersMent Health 2015;9(2):87—95.

Bowlby J. Attachment and Loss: Volume I: Attachment. London: TheHogarth Press and the Institute of Psycho-Analysis; 1969.

Cecil CA, Viding E, Fearon P, Glaser D, McCrory EJ. Disentanglingthe mental health impact of childhood abuse and neglect. ChildAbuse Negl 2017;63:106—19.

Chivers ML, Rieger G, Latty E, Bailey JM. A sex difference in thespecificity of sexual arousal. Psychol Sci 2004;15(11):736—44.

Courtois CA. Healing the incest wound: adult survivors in therapy.New York: Norton; 1988.

Harlow HF, Harlow MK. The effect of rearing conditions on behavior.Bull Menninger Clin 1962;26(5):213.

Messman-Moore TL, Long PJ. Child sexual abuse and revictimizationin the form of adult sexual abuse, adult physical abuse, and adultpsychological maltreatment. Journal of interpersonal violence2000;15(5):489—502.

Money J. Paraphilias: phenomenology and classification. Am J Psy-chother 1984;38(2):164—79.

Norman RE, Byambaa M, De R, Butchart A, Scott J, Vos T. The long-term health consequences of child physical abuse, emotionalabuse, and neglect: a systematic review and meta-analysis. PLoSMed 2012;9(11):e1001349.

Palmer S, Brown R, Rae-Grant N, Loughlin JM. Responding to chil-dren’s disclosure of familial abuse: what survivors tell us. ChildWelfare 1999;2(78):259—82.

Rind B, Tromovitch P, Bauserman R. A meta-analytic examination ofassumed properties of child sexual abuse using college samples.Psychol Bull 1998;124(1):22.

Roberts AL, Rosario M, Corliss HL, Koenen KC, Austin SB. Elevatedrisk of posttraumatic stress in sexual minority youths: media-tion by childhood abuse and gender nonconformity. Am J PublicHealth 2012;102(8):1587—93.

Seibel SL, Rosser BRS, Horvath KJ, Evans CD. Sexual dysfunction,paraphilias and their relationship to childhood abuse in men whohave sex with men. Int J Sex Health 2009;21(2):79—86.

Vaillancourt-Morel MP, Godbout N, Labadie C, Runtz M, Lussier Y,Sabourin S. Avoidant and compulsive sexual behaviors in maleand female survivors of childhood sexual abuse. Child Abuse Negl2015;40:48—59.

Zhai H, Lingren T, Deleger L, Li Q, Kaiser M, Stoutenborough L,et al. Web 2.0-based crowdsourcing for high-quality gold stan-dard development in clinical natural language processing. J MedInternet Res 2013;15(4):101—17.