Sex and gender-based differences in current literature

The vast majority of DBT treatment studies involving patients with BPD have focused on women, with a particular lack of systematic investigation of sex and/or gender differences in terms of treatment response, excluding cisgender males, as well as other gender identities (Wupperman and Edwards, Reference Wupperman, Edwards and Nadal2017). This is particularly concerning, given the important distinctions between sex and gender, paired with an overwhelming tendency for many in the field to confound the two. Sex refers to a set of physical and physiological characteristics in humans and animals, including chromosomes, gene expression, hormone levels and function, and reproductive anatomy. In the past, sex has often been categorized as male or female, although variation among these biological characteristics supports the notion that there are more than two discrete sexes (Canadian Institutes of Health Research, 2020). In contrast, gender is socially constructed and refers to the roles, behaviours and expressions of girls, boys, men, women and gender-diverse people. Gender identity exists along a continuous spectrum, and is neither binary (e.g. man/woman) nor static (Canadian Institutes of Health Research, 2020). While there is an abundance of research support for the efficacy and effectiveness of DBT among women with BPD, it is often unclear whether these studies refer to sex assigned at birth or gender, and thus whether samples consist exclusively of cisgender women (i.e. gender corresponds with sex assigned at birth) due to a lack of adequate characterization of sex and gender variables in previous research.

A 2012 Cochrane systematic review of psychological treatments for BPD concluded that DBT appeared to be more effective than treatment as usual in reducing several BPD symptoms, including suicidal behaviours, NSSI, anger, depression and anxiety, and increased overall mental health status and level of functioning. However, only 2.7% of participants in the DBT studies included in the review were men, and no single study had balanced sex or gender proportions, nor were clear parameters for definitions of gender or sex provided. As such, the authors concluded that the main findings of the review may not necessarily apply to men and proceeded to express that ‘future studies should focus on male patients with BPD who have been neglected in BPD treatment studies so far’ (Stoffers-Winterling et al., Reference Stoffers-Winterling, Voellm, Rücker, Timmer, Huband and Lieb2012; Wetterborg et al., Reference Wetterborg, Dehlbom, Långström, Andersson, Fruzzetti and Enebrink2020). A handful of small studies have been conducted to rectify this gap in the literature regarding men (for instance, see Wetterborg et al., Reference Wetterborg, Dehlbom, Långström, Andersson, Fruzzetti and Enebrink2020), although well-powered, randomized clinical trials are lacking. Furthermore, it is also unclear whether the men included in the existing studies are cisgender or include other male-identifying individuals. Given the imbalanced rates of study participants who are cisgender women relative to male and/or other gender identities, it is possible that their corresponding studies inadvertently obfuscated possible gender differences in clinical presentations and treatment response.

There are several potential reasons for the apparent imbalance of sex and gender representation among present DBT treatment studies. For example, there is some evidence to suggest that women are more likely to participate in scientific studies and clinical trials than men, and thus study samples may not be a true representation of the gender distribution (Wupperman and Edwards, Reference Wupperman, Edwards and Nadal2017). Relatedly, as women with BPD are more likely than men with BPD to seek help and use mental health services (Goodman et al., Reference Goodman, Patil, Steffel, Avedon, Sasso, Triebwasser and Stanley2010), data collected from such settings may not represent the true gender distribution of BPD in the general population. There may also be an imbalanced representation of women in clinical samples because of a diagnostic bias; past research has shown that women are more likely to be diagnosed with BPD than men, even when the same symptoms are reported by both groups (Becker, Reference Becker2019; Bjorklund, Reference Bjorklund2006). From a treatment perspective, this disparity is particularly concerning as previous research points to divergent manifestations of BPD traits and differences in psychiatric co-morbidities across sexes, although the difference between sex and gender is not usually clearly differentiated (Barnow et al., Reference Barnow, Herpertz, Spitzer, Stopsack, Preuss, Grabe and Freyberger2007; Sansone and Sansone, Reference Sansone and Sansone2011; Tadić et al., Reference Tadić, Wagner, Hoch, Başkaya, Von Cube, Skaletz and Dahmen2009). Presumably, this could have implications for case identification, treatment planning, and treatment outcomes.

In addition to the lack of evidence on sex-based differences in DBT outcomes, there may also be significant gender-based differences in terms of treatment response that warrant further investigation (Nolen-Hoeksema, Reference Nolen-Hoeksema2012). For example, a recent study using a multimethod assessment approach, which incorporated gender-related minority stress, found that BPD was the most common personality disorder among both transgender males and females (Anzani et al., Reference Anzani, Panfilis, Scandurra and Prunas2020). However, the exact prevalence of BPD among transgender and gender diverse (TGD) individuals is controversial and varies considerably across studies, ranging from 1 to 33% (Cole et al., Reference Cole, O’Boyle, Emory and Meyer1997; Duišin et al., Reference Duišin, Batinić, Barišić, Djordjevic, Vujović and Bizic2014; Goldhammer et al., Reference Goldhammer, Crall and Keuroghlian2019; Heylens et al., Reference Heylens, Elaut, Kreukels, Paap, Cerwenka, Richter-Appelt and De Cuypere2014; Madeddu et al., Reference Madeddu, Prunas and Hartmann2009; Meybodi et al., Reference Meybodi, Hajebi and Jolfaei2014). According to Goldhammer and colleagues (Reference Goldhammer, Crall and Keuroghlian2019), this inconsistency may be explained by methodological differences between these studies, including different eligibility criteria for participation, measures used to assess gender identity and psychiatric co-morbidities, cultural norms regarding gender minorities in the various countries in which these studies were conducted (e.g. Iran, Italy, Serbia, Sweden, United States), as well as selection bias and limitations in study design. Nevertheless, studies on the prevalence of BPD and other personality disorders in the general population have reported similar or lower prevalence than studies on TGD populations (Huang et al., Reference Huang, Kotov, De Girolamo, Preti, Angermeyer, Benjet and Kessler2009; Lenzenweger et al., Reference Lenzenweger, Lane, Loranger and Kessler2007; Meaney et al., Reference Meaney, Hasking and Reupert2016). Regardless of why incidence rates of BPD may differ among TGD individuals, there remains a scarcity of literature on the treatment response to DBT among TGD individuals with BPD.

Quantitative review of sex and gender characteristics reported in current DBT literature

Adopting a systematic approach, we aimed to identify the number of DBT studies involving persons with BPD that report on both sex and gender constructs in order to quantify the number of women, men, and TGD individuals  included in clinical samples in the literature. This included conducting a broad review to quantify the gender and sex variance in the DBT treatment literature on Ovid Medline, PsycINFO, Wiley Cochrane Library (CENTRAL, CDSR, HTA), WHO, and ClinicalTrials.gov. The search terms utilized were ‘borderline personality’ and ‘dialectical behavior therapy’ and its variations (Dialectical Behavior Therapy/ or (dialectical adj2 (behavio?r* or therap*)).ti,ab,id,kw. or (DBT).ti AND Borderline Personality Disorder/ or *Borderline States/ or ((borderline or border-line) adj2 (personalit* or disorder*)).ti,ab,id,kw. After duplicates were removed, a total of 1044 studies remained. Clinical studies investigating the effect of DBT in individuals with BPD were included based on the consensus of two authors working independently (S.C., A.F.). Due to the variety of outcomes reported, this was then again reviewed by two more experienced authors (M.S., T.H.) to ensure that criteria were strictly met. Here, 253 studies met these specifications; where 6 were cross-sectional, 136 were self-controlled longitudinal, 37 were longitudinal group-compared studies, and 74 studies were longitudinal case-control with randomization. Of these, only two case-control randomized studies used double blind randomization.

The 253 studies pooled together 17,103 patients, of which 1931 (11.3%) were males and 12,029 were females (70.3%), while 56 studies did not clearly report either gender or sex, leaving 3143 patients (18.4%) with no clear gender or sex declared. Eighty nine studies solely reported data for females, 102 studies reported data for two sexes (i.e. male and female), and 6 studies reported data for males only. Importantly, even in the 126 (27.39%) studies which included two sexes (male and female), females were heavily over-represented. Only two studies reported including TGD individuals, although this remained unclear (e.g., listed as “other” and/or “LGBTQ”). Not only does this brief overview of the existing literature highlight how gender and sex differences have been neglected in DBT research, thereby omitting vital information on differences across gender in terms of BPD presentation and treatment response, it also emphasizes the need to differentiate between sex and gender in future treatment studies. Additionally, while the above is presumed to be primarily cisgender women, this was rarely explicitly reported. The noted lack of differentiation between sex and gender precludes the possibility of knowing whether cisgender and transgender women respond similarly to treatment.

Importance of gender-related considerations

There is a significant gap in the literature regarding gender-diverse individuals with BPD. Given the unique stressors impacting the TGD community, this is a critical area for future research. One such avenue for this research includes exploring an integration between the biosocial model upon which the theory of DBT is predicated, and the gender minority stress model, which theorizes that the increased risk of mental health problems among TGD individuals is a consequence of the marginalizing nature of their social environment (Testa et al., Reference Testa, Habarth, Peta, Balsam and Bockting2015). This marginalization includes heightened rates of discrimination, oppression, microaggressions, non-affirmation and victimization occurring across societal, institutional and interpersonal domains (Chodzen et al., Reference Chodzen, Hidalgo, Chen and Garofalo2019; Craig and Austin, Reference Craig and Austin2016; Jäggi et al., Reference Jäggi, Jellestad, Corbisiero, Schaefer, Jenewein, Schneeberger and Garcia Nuñez2018; Tan et al., Reference Tan, Treharne, Ellis, Schmidt and Veale2019), which may lead to internalized stigma, identity concealment, and rejection-related cognitions, behaviours and emotions such as shame. Ultimately, these factors create chronic stress, including both external and internal invalidation.

According to the biosocial model, the pervasive pattern of dysregulation in BPD is the result of a transactional effect between biological vulnerabilities to emotion (i.e. high sensitivity and reactivity, as well as a slow return to baseline) and an invalidating social environment (Gill and Warburton, Reference Gill and Warburton2014; Linehan, Reference Linehan1993). Given complex and insidious interactions with gender-related minority stressors, TGD individuals may be at a greater risk of experiencing such a similar pattern of dysregulation due to a higher likelihood of experiencing the adverse effects of an invalidating environment (e.g. less support and/or rejection from family and friends; verbal, physical, and/or sexual violence; Aparicio-García et al., Reference Aparicio-García, Díaz-Ramiro, Rubio-Valdehita, López-Núñez and García-Nieto2018; Johns et al., Reference Johns, Lowry, Andrzejewski, Barrios, Demissie, McManus and Underwood2019; Shipherd et al., Reference Shipherd, Maguen, Skidmore and Abramovitz2011; Sloan et al., Reference Sloan, Berke and Shipherd2017). Any resulting heightened emotional dysregulation is therefore a reasonable outcome of chronic invalidation transacting with a vulnerable, marginalized identity and existence, notwithstanding one’s biological and/or emotional vulnerability (Goldhammer et al., Reference Goldhammer, Crall and Keuroghlian2019; Sloan et al., Reference Sloan, Berke and Shipherd2017). Moreover, BPD is also highly correlated with trauma and post-traumatic stress disorder (PTSD), often of a complex nature (i.e. C-PTSD; Ford and Courtois, Reference Ford and Courtois2021). Given that PTSD affects a disproportionate number of TGD individuals compared with the general population (Reisner et al., Reference Reisner, White Hughto, Gamarel, Keuroghlian, Mizock and Pachankis2016), it is not surprising that core symptoms of BPD are highly prevalent in TGD populations.

As previously noted, the true prevalence and presentation of BPD among TGD individuals continues to be an area in need of further clarification, particularly given increased social risk factors for symptoms typically considered characteristic of BPD in TGD populations. For instance, many core symptom areas of the disorder, including emotional reactivity, rejection sensitivity, suicidality and NSSI, have been determined to be elevated in TGD populations due to chronic exposure to minority stressors (Goldhammer et al., Reference Goldhammer, Crall and Keuroghlian2019). This is not to say that TGD individuals are necessarily at greater risk of developing BPD. Rather, given the overlap between core BPD symptoms and minority stress reactions among TGD individuals, these two constructs are often highly conflated, which in turn likely creates more invalidation, thereby furthering the transactional nature of these vulnerabilities. Minority stress exposure occurs in the context of BPD (i.e. as an added layer of an individual’s invalidating environment) and also in its absence. As the gender minority stress model and biosocial model align, and because TGD individuals experience fundamental invalidation which leaves them susceptible to experience concerns akin to those DBT was designed to treat, DBT may be an effective treatment option to address these concerns (Sloan et al., Reference Sloan, Berke and Shipherd2017).

Future directions

At the most basic level, as evidenced by the significant lack of reporting on sex and gender variables in previous research, future studies should aim to better characterize their participants in order to increase our understanding and generalizability of DBT treatment outcome results. Given the symptom severity of BPD, elevated social risk factors for symptoms associated with invalidation experienced by TGD individuals, and the pre-existing vulnerabilities TGD individuals already face, further research examining treatment response using standard DBT is essential to improve care for this vulnerable population. Future studies exploring differences in DBT outcomes across sex and gender groups are critical, particularly individuals assigned male, as well as TGD individuals, as this may improve clinical decision making and, ultimately, treatment outcomes. Furthermore, given that there are 256 possible BPD symptom combinations (Fowler et al., Reference Fowler, Carlson, Orme, Allen, Oldham, Madan and Frueh2020), it is worth investigating whether these varying presentations lead to differences in treatment response, and whether sex and gender play an influencing role in both the varying presentations and subsequent treatment response. More generally, whether treatment response occurs at the same rate of progress across genders, and whether genders respond differentially to individual or group therapy components of this treatment should also be explored.

Standard DBT is intended to be tailored to the unique needs and experiences of an individual patient based on case conceptualization (Koerner and Linehan, Reference Koerner and Linehan1997). However, as previously stated, DBT was developed with a primary sample of women. This suggests that there is significant evidence that DBT produces positive change among women with BPD, but there is currently limited evidence of the efficacy of this treatment in other groups (i.e. cisgender men, TGD individuals). While it may be argued that DBT clinicians can presumably tailor treatment to each individual’s needs across other sexes or gender identities, this is an overly simplistic assumption that treatment will be as effective for them as it is for women with BPD without evidence to substantiate this. More importantly, this assumes that clinicians have adequate knowledge of unique needs and challenges of TGD individuals, such as gender-related minority stress, and would be adept at integrating this into treatment in a therapeutically effective manner. Indeed, it is well documented that many healthcare providers do not have sufficient training or knowledge on TGD healthcare needs, which contributes to the ongoing mental health disparities experienced by the community (Korpaisarn and Safer, Reference Korpaisarn and Safer2018; Lykens et al., Reference Lykens, LeBlanc and Bockting2018; Rider et al., Reference Rider, McMorris, Gower, Coleman, Brown and Eisenberg2019; Safer, Reference Safer2016; Safer et al., Reference Safer, Coleman, Feldman, Garofalo, Hembree, Radix and Sevelius2016).

Deleterious impacts of minority stressors on mental health, such as pervasive discrimination, victimization and rejection experienced by many TGD individuals are well-documented (Griffin et al., Reference Griffin, Casanova, Eldridge-Smith and Stepleman2019; Hendricks and Testa, Reference Hendricks and Testa2012; Jäggi et al., Reference Jäggi, Jellestad, Corbisiero, Schaefer, Jenewein, Schneeberger and Garcia Nuñez2018), although adequate treatment options addressing the consequences of minority stress-related experiences remain scarce (Busa et al., Reference Busa, Janssen and Lakshman2018). Whether TGD individuals meet diagnostic criteria for BPD or core symptoms of BPD as a result of a chronically invalidating environment (Goldhammer et al., Reference Goldhammer, Crall and Keuroghlian2019), is a critical area for future study. It is possible that minority stress may need to be explicitly targeted in treatment to produce meaningful change. A basic treatment assumption is that patients with BPD struggle to navigate their world due to a lack of intrapersonal and interpersonal skills, and treatment is focused – in part – on ameliorating these deficits to help patients navigate the world more effectively (Chapman, Reference Chapman2006). Preliminary studies of cognitive behavioural approaches incorporating a minority stress framework have shown promising results in sexual and gender minorities (Craig and Austin, Reference Craig and Austin2016; Flentje, Reference Flentje2019; Pachankis et al., Reference Pachankis, Hatzenbuehler, Rendina, Safren and Parsons2015). Sloan et al. (Reference Sloan, Berke and Shipherd2017) have proposed a framework for tailoring DBT to TGD individuals, including adapting both acceptance and change strategies, as a preliminary step in addressing treatment gaps for gender-diverse individuals, although further empirical evaluation and support is needed. Future clinical trials should consider including a minority stress lens to DBT, both in terms of psychoeducation and application of DBT skills to minority stressors. Future research should also investigate whether the severity of minority stress impacts treatment response among TGD individuals.