In an attempt to further conceptualize CSB, many studies regarding psychiatric comorbidities and psychological traits have been conducted. In the studies reviewed CSB has been reported to have correlations with perfectionism [45], paraphilias and sexual dysfunction [30,46], mood disorders (e.g., anxiety and depression) [26,35,38,41,47,48], experience of childhood sexual abuse [47,49,50], OCD [35,41], Autism Spectrum Disorder [42], general emotional dysregulation [31], substance abuse [43,44], maladaptive schemata [51], excitement seeking in general [45,52], compulsivity and impulsivity [47,49,53], high sexual arousal/excitation [25], higher neuroticism [47], avoidant coping mechanisms [54], high rates of psychiatric comorbidity in general [38], high distress rates and negative emotions [14], boredom [48], general childhood adversities and trauma [30,33,43,49,55,56,57,58,59], pathological buying [37], insecure attachment style [60,61,62] and risky sexual behavior [32] in both genders.

Gender differences have not been studied extensively in the reviewed studies. Twenty-two studies calculated at least some analyses to compare men and women. As the studies are mostly of a correlative nature, a variety of correlative gender differences were found.

Of the reviewed studies, four studies focused on sexual behavior [52,63] and paraphilic interests [23,30] and reported some gender effects. For example, Rettenberger et al. investigated a large online sample (n = 1749, 56.5% female) and reported lower scores on sexual excitation (SES) and CSB (HBI) for women and consequently higher scores on sexual inhibition dimensions as well. Interestingly, CSB seemed connected to sexual excitation, “fun-seeking”, low scores on threat performance consequences and conscientiousness for both genders [52].

In contrast to this finding, Engel et al. [23] reported some gender effects regarding sexual behavior correlations. For women, CSB was associated with partnered sexual activity, TSO-masturbation and time of pornography consumption, while there were no significant correlations between CSB symptom severity and partnered sexual activity in men. However, there were significant but weaker correlations between pornography consumption and TSO-masturbation. When compared to women, men had higher rates of pornography consumption (e.g., 6.64 h weekly in men compared to 1.05 h in women; d = 0.41) and elevated rates of problematic cybersex (d = 0.96) [23]. This ties in with the finding of Bőthe et al. who reported a moderate positive association between CSB and Problematic Pornography Use (PPU) in women but a strong association in men [64]. In a second study, Bőthe et al. reported a greater relationship between impulsivity and CSB (β = 0.41 and β = 0.42) than between CSB and compulsivity (β = 0.21 and β = 0.16) among men and women, respectively [53]. In a recent study by Efrati et al. [63], CSB was connected to engaging only in online sexual activities and female gender in an adolescent sample.

When investigating the association between CSB, paraphilic behaviors and sexual functioning, gender differences concerning coercive fantasies and behavior [23] and general paraphilic interests [30] were reported. Paraphilic interests were also associated with CSB symptom severity in a study with Italian university students, with higher HDSI scores in all paraphilic interest groups (except transvestitism). Additionally, gender differences relative to the nature of paraphilic interests were reported: men exhibited higher prevalence of exhibitionism, sadism, voyeurism and frotteurism, while women exhibited higher prevalence of masochism and fetishism [30]. In accordance with this finding, coercive fantasies were found to be more prevalent in men compared to women (p < 0.001), as was the past behavior of forcing someone to perform sexual behaviors (p < 0.001) or using child abuse images (p < 0.001) [23]. When compared with the general population, women with CSB reported more fantasies about forced sexual actions as well as actual past behavior of forcing someone to perform sexual acts. The same was true for men [23].

Another focus of past research was psychiatric comorbidity with 10 studies reporting gender effects on some psychiatric correlates, such as depression [23,28,35], anxiety [28,35], OCD [35,41], Autism Spectrum Disorder (ASD) [42], ADHD symptoms [64], sexual dysfunction [30,46] and general stress proneness [15,65].

Regarding depression and OCD, the picture seems relatively consistent: In all four studies [23,28,35,41], depressive symptoms, anxiety and OCD symptoms were more strongly associated with CSB symptom severity in men than compared to women. For example, Levi et al. investigated the impact of compulsivity, depression, anxiety and impulsivity on problematic online sexual activities in individuals who use the internet for finding sexual partners. In their sample, OCD, depression and anxiety contributed more to CSB in men (40%) than in women (20%) [35].

In regard to ASD, Schöttle et al. [42] measured CSB in 96 individuals with ASD (58% male) and 96 matched healthy controls. They reported that ASD patients had a higher HBI sum score than healthy controls and that more ASD individuals met the HBI cut-off. They found that men with ASD reported more CSB than their healthy control counterparts, while there was no such difference in females.

Bőthe et al. found gender differences in the association between ADHD symptoms and Problematic Pornography Use (PPU): They found a moderate association for men but a weak association for women [64]. The authors conclude that for men with ADHD, pornography use could be some kind of self-medication, whereas women with ADHD symptoms do not use pornography but other sexual behaviors instead. This could be due to a more normative approach to pornography among men. Two studies found an association between CSB in response to stress [15,65], with women reporting greater stress vulnerability. Sexual dysfunctions were associated with CSB in both genders [30]. Furthermore, Burri et al. reported that women with higher sexual compulsivity (measured with SCS) reported worse sexual functioning [46].

Another branch of correlative studies was conducted in the realm of personality traits, with three studies reporting gender differences on personality traits in individuals with CSB [26,36,65]. For example, Shimoni et al. conducted an analysis of variance and found that personality factors contributed significantly to the variance of CSB by explaining as much as 11.7% of the variance. Interestingly, in their study, greater neuroticism was indicative of CSB in men but not women [36]. In contrast to this finding, Carvalho et al. reported from their cluster analysis that women with higher neuroticism were more likely to be sorted into a “control cluster”, which is indicative of CSB [26]. When studying gender differences in a clinical population, Reid et al. [65] found no gender differences in the NEO domain neuroticism, but in the domains conscientiousness (competence: w < m; dutifulness: w < m; achievement striving: w < m) and impulsivity (w > m in excitement-seeking), thus indicating that conscientiousness seems more relevant for men, while impulsivity seems to be a more relevant factor for women with CSBD.

Often discussed antecedents to CSB include childhood adversities and trauma, which were investigated in ten of the reviewed studies [30,33,43,49,55,56,57,58,59] of which six investigated gender differences. One study [30] found a correlation between the HDSI and a history of sexual abuse (β = 0.144 and p = 0.001) in women but not in men. Similarly, in a drug user outpatient group, Diehl et al. [43] found that women had higher childhood maltreatment rates than men (sexual abuse: OR: 2.9; physical abuse: OR: 3.7). In contrast to this finding, Slavin et al. [49] found an interaction between Childhood Sexual Abuse and Adolescent/Adult Sexual Abuse (CSA/AASA) and gender. In this sample, CSA/AASA impacted CSB in men more strongly than in women. Likewise, Fontanesi et al. [55] reported an effect of post-traumatic stress symptoms on CSB with the male gender being a risk factor in the development of CSB. In contrast to these findings, some studies reported no gender effects. While Meyer et al. [56] reported a correlation of CSB (measured with SAST) and sexual abuse (r = 0.26 **) in their sample (812 individuals, 62% females), they did not find any gender effects. Similarly, Vaillancourt et al. [58,59] reported a correlation of CSB and CSA frequency without gender effects.

One study used exploratory factor analysis to investigate the latent structure of CSB in men and women. This study concluded that CSB in males seemed to be more consistent with a taxon rather than with a continuum. Female CSB on the other hand seemed to be more dimensional and showed evidence of continuous latent structure [29].

With the introduction of CSBD into the ICD-11, the interest in research has sparked in this topic [66]. Up until now, CSBD has been measured with a variety of questionnaires and indices. In the studies reviewed for this article, the HBI [16] was used most often, followed by the SCS [21], HBCS [17], HDSI [18] and SAST [20]. The variety of indices reflects the debate on the different concepts that were used in the past to describe CSBD (e.g., addiction concept, compulsive behavior, impulsive behavior) (compare [2,66]). The different indices are one possible explanation for the wide range (2.8% up until 35.1% in convenience samples; 3.3% up until 8.3% in national surveys; 3.3% up until 40.3% in clinical samples) of estimated prevalence.

While studies using convenience or online sampling include more women in the last years, studies with clinical samples do still focus more strongly on men. In convenience samples, men show higher scores in indices measuring CSB. This is true for a wide variety of measures, although the variety makes it harder to compare results. It seems quite stringent that men who score higher in measures of CSB also report more masturbation frequency and more sexual behaviors than women. The same is true for sexual behavior, especially masturbation, in men in the general population [67].

CSB rates seem especially high in student populations. For example, in their university student sample, Castro-Calvo et al. [28] found rates of 10.1% of CSBD out of which about 30.0% were female. Comparably, Kingston et al. [29] reported 9.2% of women in their university sample meeting the HBI cut-off. Similarly, even up to 19.4% of students met CSBD criteria measured with the HD and HBCS [31]. We only found one study that did not report a gender difference for the measure in CSBD [48] measured with the SAST.

At the same time, the picture seems inconclusive regarding clinical samples. Clinical samples mostly target other clinical populations and survey CSBD as a correlate. Rates of CSBD have been investigated in samples of patients with OCD [41], ASD [42] and those with alcohol and drug associated disorders [43,44]. While prevalence in the OCD sample was comparable to prevalence found in large non-clinical samples (5.4% in men and 1.4% in women [41]), prevalence seemed to be much higher in individuals using drugs [43,44]. It is important to note that the SAST was used in the drug-using sample using a cut-off of six, which is validated for the Brazilian population. In other samples the cut-off of 13 is usually used. Still, this makes the numbers harder to compare.

In this context, it is also important to mention that high frequency sexual behavior is not identical with CSB or CSBD. Bőthe et al. [39] found in their study that individuals who were in the high frequency sexual behavior group (in this case pornography consumption) differ from people with high frequency problematic sexual behavior. These groups were distinguishable according to their HBI scores, which verifies that there is such a thing as unproblematic high frequency pornography consumption and that it is not to be confused with CSBD. Vaillancourt-Morel [27] reported a similar finding in their study by using the SCS scores to differentiate between clusters (recreational pornography use, highly distressed non-compulsive use and compulsive use).

New models of CSBD are currently emerging (e.g., the Sexhavior Cycle of Hypersexuality [68], the Moral Incongruence Model [66] and the Integrated Model of CSBD [2]) and it is important to test these models and their assumptions. In these models, gender differences regarding sexual response are not explicitly addressed. They are mostly addressed as different psychosocial influences on women and men.

In our review, 22 studies reported at least some analyses comparing men and women. As the studies are mostly of a correlative nature up to now, a variety of correlative gender differences were found in the study.

In a DSM-IV trial [17], Reid et al. found higher scores on the HBI in men. On the other hand, Öberg Görts et al. [15] demonstrated higher burden on women. In our review, almost all studies reported higher scores in CSBD indices in men. It is possible, that women only seek help when they experience their behavior as threatening to their private and professional lives, thus waiting longer to reach out to an outpatient clinic. Possibly, it is easier for men to report their behaviors because high frequency sexual behaviors are culturally more expected from men [69], which results in the over-reporting of sexual behaviors in men and underreporting of sexual behaviors in women (“sexual double standard”) [76]. This might concern the CSBD population as well. Here, it is to be discussed if it is helpful to use the same cut-off scores on indices for men and women. Due to the sexual double standard [76] and generally higher frequency of sexual behavior in men, it is possible that current measurements will result in the underreporting of Compulsive Sexual Behavior in women when we measure it the same manner as in men (compare [77]). Women are still especially scarce in clinical studies, which limits the clinical implications of these findings. Additionally, as questionnaire data are used dimensionally in many studies, the implications for individuals with diagnosed CSBD are also limited.

With the inclusion of CSBD in ICD-11, there is hope that researchers will start using comparable questionnaires and cut-offs in their future studies. In addition, the release of binding diagnostic categories will verify that the same construct is being investigated. Another possible development will be that CSBD will be more prevalent because mental health professionals will be able to use the diagnosis to treat their patients. For this, reliable therapy options are still required. As we have demonstrated that CSB seems to have at least somewhat different expressions in men and women, it is recommended that gender specific treatment foci should be implemented.

As there are some indications that women with CSB have other concerns than men with these behaviors (e.g., they worry more about STIs and pain), researchers and clinicians should investigate the needs and concerns of women even if they seem to be less often affected by CSBD. It might be useful to explicitly ask for sexual behaviors (and also compulsive sexual behaviors) when women present themselves for other disorders that pertain to impulse control, substance abuse or mood disorders. On the other hand, it seems to be advisable to ask about paraphilic fantasies in men as well as women.