Exploring the roles of compassion and post-traumatic stress disorder on global distress after sexual trauma

Objectives: Recovery from sexual trauma can be complex and multi-faceted. Most current psychological treatment protocols for trauma use a cognitive model of post-traumatic stress disorder (PTSD). However, sexual trauma may include specific complexities beyond that of a cognitive model of PTSD, such as relational factors. The distress experienced after sexual abuse may involve variables not exclusive to a PTSD model. Compassion focused therapy (CFT) is an approach that incorporates evolutionary, relational and social perspectives. This study explored the relationships between variables associated with CFT, PTSD and distress in survivors of sexual abuse to determine the role of CFT-related variables. Methods: 155 adults who had experienced sexual abuse or any unwanted sexual experience at any point in their lives completed online questionnaires pertaining to various CFT variables (self-compassion, receiving compassion from others, having a fear of compassion from others, having a fear of


I N TRODUC T ION
Trauma is conventionally understood in terms of post-traumatic stress disorder (PTSD), which lends itself to being treated with cognitive and behavioural therapies which can directly target PTSD symptomatology.Standard PTSD protocols are based on cognitive processes (Brewin et al., 1996) and focus on managing fear responses and memory processing (Zayfert & Becker, 2007).Treatment recommendations are, therefore, in line with cognitive and behavioural models (National Institute for Health and Care Excellence (NICE), 2018).Whilst the use of existing PTSD protocols has been found to be effective for some populations diagnosed with PTSD after sexual violence (Smallwood, 2020), a substantial number of individuals treated for PTSD drop out of, or do not respond to PTSD treatment (Taylor et al., 2012), 64% of whom have some experience of childhood physical or sexual abuse (Zayfert et al., 2005).Furthermore, Rothbaum et al. (1992) found that 47% of sexual abuse survivors continued to experience PTSD symptomatology after treatment, and for some, symptoms may even worsen over time (Lambert, 2010).Saraiya and Lopez-Castro (2016) suggest that, despite progress in the treatment of PTSD, many individuals remain symptomatic following therapies.They also argue that it is important to examine affective disturbances that involve emotions beyond that of fear or threat, as encapsulated in the diagnosis of Complex Trauma or Complex-PTSD which considers relational, affect and self-concept based aspects (World Health Organization (WHO), 2019).Such emerging diagnostic conceptualizations provide a rationale for developing additional models for trauma outside the PTSD symptomatology frameworks.Although PTSD has been found to be common and severe in survivors of sexual abuse (Dworkin et al., 2023), the impact of sexual trauma appears to include complexities beyond the those captured through diagnostic models (Chivers-Wilson, 2006;Domino et al., 2020).
Sexual trauma may be experienced differently to other life-threatening events as it involves an intrusion on the body and often impacts individuals both in terms of fear and their interpersonal and intimate relationships (Herman, 1997).Related problems include difficulties in social functioning and in the availability of social support after abuse (Frank, Turner & Stewart, 1980;Calhoun, Atkeson & Ellis, 1981;Baker, Skolnik, Davis & Brickman, 1991, cited in: Petrak, 2002), as well as 'negative internal representations about safety, power, trust, esteem and intimacy' (McCann, Sakheim & Abrahamson, supported.Future research may explore experimental designs utilizing CFT in this population, as well as further investigations on the roles of these specific CFT variables.

K E Y W O R D S
CFT, compassion, sexual abuse, sexual trauma, trauma

Practitioner points
• Recovery from sexual trauma may be distinct from that of other types of traumas and may not be fully understood through the PTSD model.• The compassion focused therapy (CFT) model provides a potential theoretical model for treating trauma after sexual abuse.• An analysis of CFT-related variables showed self-criticism to play a significant role in distress after sexual abuse.• The results from this study highlight specific aspects of the CFT model which clinicians may wish to consider in work with sexual abuse survivors.1988, cited in: Petrak, 2002, p. 34), p. 34).Sexual trauma involves heightened levels of despair, hopelessness and shame (Petrak, 2002).If sexual trauma has occurred in a context where it could have been prevented by others (such as caregivers or organizational systems), survivors may develop distrust of others (Herman, 1997).Difficulties with trust can have an additional and nuanced impact on therapy as the inherent power imbalance within a therapeutic relationship can be re-triggering (Scott & Stubley, 2022).It is, therefore, vital that survivors of sexual trauma are offered treatment that considers such specific repercussions.As a result, the cognitive models of PTSD alone may not sufficiently explain the psychological and relational impact of sexual traumas.
There is a theoretical argument for the use of compassion focused therapy (CFT) in sexual trauma work, as the CFT model was developed to treat high levels of shame and self-criticism and support the regulation of the threat-based system following trauma (McLean et al., 2017).CFT is a metatheoretical model that encompasses evolutionary, attachment, interpersonal, self-concept and socialconcept perspectives as well as the cognitive and behavioural theory (Gilbert, 1998(Gilbert, , 2010(Gilbert, , 2017)).This meta-theoretical approach provides a potential framework for the interpersonal and shame-based difficulties exhibited in survivors of sexual trauma (McLean et al., 2017).Furthermore, compassion based intervention has qualitatively been found to be helpful in alleviating emotional distress in those with Complex Trauma (Willis et al., 2023).Lee et al. (2001) argue the need to address emotional responses in PTSD other than fear, with a particular emphasis on shame and guilt which can be particularly disabling as they affect 'the experience of the self and social behaviour', as well as help-seeking behaviour (Andrews, 1995(Andrews, , 1998;;Gilbert, 1997, cited in: Lee et al., 2001, p. 451) and may, therefore, contribute to later psychopathology.There is an emerging understanding of trauma as linked to shame, self-criticism, self-compassion and fear of compassion which is explored in this current study.Saraiya and Lopez-Castro (2016) scoping review of 47 studies investigating PTSD showed an association between shame and PTSD and also provided evidence to support the utility of shame reduction as a treatment target.There is an argument for examining shame in relation to trauma, in combination with fear-based responses as it often has 'equally debilitating effects, including being related to inadequacy, self-criticism and secondary emotions, such as anger' (Irons & Lad, 2017, p. 1).Shame and self-criticism are also found to be interpersonally reinforced difficulties (Gilbert, 2010), with shame being comprised of 'a range of feelings like anger, anxiety, humiliation, embarrassment and disgust' (Lee, 2012, p. 68).Furthermore, Adshead describes persistent shame, as well as fear reactions, as key to understanding PTSD, but that these different types of reactions may require different treatments (Adshead, 2000).Shame and guilt are also associated with PTSD symptom severity and treatment outcomes, as well as increased suicide risk among individuals with PTSD (Cunningham, 2020).Bhuptani and Messman-Moore (2019) suggest that shame and self-blame are common following sexual assault, and Au et al. (2017) found that CFT intervention effectively reduced shame and self-blame and increased compassion in adults with PTSD.
In clinical settings, self-criticism is a construct that is often co-occurring with shame (Gilbert, 2009a;Lee, 2012).It has been found to mediate the relationship between experiences of sexual violence and mental health outcomes in women (McAllister, 2019) and mediate the relationship between sexual abuse and non-suicidal self-injury (Glassman et al., 2007).Self-compassion has also been found to be negatively associated with trauma-related stress and PTSD (McLean et al., 2018), as well as negatively associated with memories interpreted as being shameful in children with experiences of trauma (Minimol & Lucila, 2019).Mindful self-compassion appears to be helpful for those who have experienced intimate partner abuse (including sexual abuse; Tesh et al., 2015), and self-compassion has been found to increase well-being after an imagined sexual assault in women (Allen et al., 2020).Differences in the experience of self-compassion were also found to exist across individuals depending on whether abuse was suffered in childhood or adulthood (Kjose, 2019).These distinctions are, therefore, important to consider in therapeutic work.There is also a positive association between the fear of self-compassion and PTSD in those who have experienced childhood sexual abuse (Miron et al., 2016).Fear of compassion has also been found to mediate the relationship between childhood maltreatment and PTSD in female survivors of childhood abuse (Boykin et al., 2018).
These studies suggest that the CFT model could be useful in understanding distress experienced by survivors of sexual abuse.However, no research has explicitly investigated the role of these CFTrelated variables in distress experienced for this group.Psychological treatment for sexual trauma often requires idiographic and integrative approaches tailored to the individual based on clinical judgement (Petrak & Hedge, 2002).Therefore, understanding the role of various CFT mechanisms that might be involved with sexual trauma would help determine how to develop and tailor treatment more effectively.It would, therefore, be helpful to explore how these CFT processes link to distress experienced after sexual abuse.

Aims and hypotheses
This study examined whether the CFT model was a useful adjunct to the cognitive model of PTSD for survivors of sexual trauma in those who self-reported experiences of sexual abuse across childhood and/or adulthood.CFT-related variables included in this study were self-compassion, compassion from others, fear of compassion from others, fear of compassion from the self, shame and self-criticism.Global distress was used as a mental health outcome, as this is frequently used within service settings for victims of sexual trauma [such as community mental health teams (Leach et al., 2005) and specialist trauma services (Tavistock & Portman NHS Foundation Trust, 2021)].As the diagnostic criteria for PTSD and global distress are highly related concepts, the authors planned to examine each CFT-related variable to understand CFT's unique role for global distress.

I Primary hypothesis:
The CFT variables would explain a small but significant increment in distress beyond PTSD symptomatolog y alone.II Secondary hypothesis: The CFT variables would be related to global distress irrespective of PTSD.III Exploratory hypothesis: To explore the relative contributions of each CFT-related variable and PTSD subscale to global distress.

M E T HOD Participants
Participants were recruited online via an advert disseminated through appropriate charities and social media accounts on various popular platforms.Participants were offered to opt into a £100 voucher prize draw on completing the study.A free CFT and sexual trauma resource, co-developed with a leading author in the field of CFT and checked by two Experts by Experience (EbE), was offered to anyone who visited the survey website, irrespective of their decision to take part.A power analysis calculation gave a recommendation of N = 103 [G*Power (Faul et al., 2009): Effect size f 2 = .15,α = .05,power = .80].
Participants were 155 adults who had experienced sexual trauma at any point in their life.The participants reported a variety of gender and/or sexual identities and/or practices.Even though the study was advertised as taking place in the United Kingdom, participants were from all over the world.The sample was predominantly aged 18-50 (83.2%), lived in the UK (62.6%) and of white ethnicities (86.1%) (Table 1).49.7% of the sample identified as heterosexual and 80.9% as female (Table 2).95.5% of the sample had received mental health support and 84.5% had received some form of psychological therapy (Table 3).Therapies received by participants included were counselling (17.9%),CBT (13.5%), mindfulness (7.6%) and EMDR (7.3%).3.6% had received CFT.23.2% of participants had received four or more years of psychological therapy, and 17.4% received 1-2 years of therapy.The main diagnoses received were depression (24.1%),PTSD/C-PTSD (20.9%) and anxiety (17.1%) (Table 4).

Design
The study used a cross sectional design.Questionnaires related to CFT, PTSD and global distress were administered online via Qualtrics XM between the dates 15 November 2021 and 14 February 2022.

Procedure
Participants were asked to download a participant information sheet and electronically sign an informed consent form.The survey generated a unique participant ID.Screening questions were used for language and age eligibility.Where eligibility was not met, participants were directed to a page exiting the survey and signposting risk information relevant to their age.They were also offered the CFT resource.Eligible participants were directed to a single question defining sexual trauma and asking if their sexual trauma had occurred in childhood, adulthood or both.No further details about their trauma experiences were required.Participants were then directed to take part in the survey which took approximately 20-30 min to complete.Participants were then asked to provide optional information relating to demographics and mental health to check for heterogeneity of the sample.They were then offered the opportunity to relay any feedback on the survey and opt into the prize draw.Participants were then directed to a page thanking them for their time, providing further information on CFT, a link to the free CFT resource and information related to risk and sexual trauma services.They could download a de-brief sheet re-iterating this information.

Compassion
The Compassionate Engagement and Action Scales (Gilbert et al., 2017) is a measure of compassion competencies across three 'flows'-to others, from others and to self.Scales focusing on self-compassion and compassion experienced from others, respectively, were used, totalling 26 items.Self-compassion showed good internal consistency (Cronbach's α = .80),whilst compassion from others showed excellent internal consistency (Cronbach's α = .90).Both scales showed good validity (Gilbert et al., 2017).Within this study, self-compassion showed acceptable internal reliability (Cronbach's α = .73),and compassion from others showed good internal reliability (Cronbach's α = .83).

Fear of compassion
The Fears of Compassion Scale (Gilbert et al., 2011) measures how fearful people are of compassion across three 'flows'-to other, from others and to self.Scales focusing on fear of compassion from others and from the self, respectively, were used, totalling 28 items.Both scales showed good internal consistency (Cronbach's α = .87and α = .85,respectively).Within this study, both scales showed excellent internal reliability (Cronbach's α = .93and α = .95,respectively).

Shame
The External and Internal Shame Scale (EISS)-short form (Ferreira et al., 2020) is an eight-item scale that measures shame across two subscales for external and internal shame.This scale showed good internal consistency on both subscales (Cronbach's α = .80and α = .82,respectively).It also showed good internal consistency on the total scale (Cronbach's α = .89).This scale showed good discriminant validity and strong validity (Ferreira et al., 2020).Within this study the scale showed good internal reliability (Cronbach's α = .88).

Global distress
The Clinical Outcomes in Routine Evaluation-Outcome Measure (CORE-OM; Evans et al., 2002) is a measure of global distress created to assess efficacy and effectiveness across multiple disciplines offering psychological therapies.It has four domains relating to subjective well-being, problems/symptoms, life functioning and risk/harm, containing 34 items.The scale showed acceptable to good internal consistency for all the domains (Cronbach's α = .77,α = .90,α = .86and α = .79,respectively).It also showed excellent internal consistency for the total scale (Cronbach's α = .94).Convergent validity was reported as good (Evans et al., 2002).Within this study the scale showed good internal reliability (Cronbach's α = .86).

Analysis
Statistical analysis was conducted using IBM SPSS Statistics (version 25) and interpreted using .05significance levels.There was no missing data due to the online requirement in completing questionnaires.As a result, the total score of the CORE-OM was used in analysis rather than the calculated CORE-OM score.The means for the study sample were compared against the original scale studies samples (taken from the papers introducing each scale), to check for generalizability of the sample.Tests of normality and goodness of fit were conducted.Pearson's correlations between all variables were assessed to check the strength of the relationships between variables.Hierarchical multiple regressions were conducted exploring the relative contributions of two models on global distress: the first examining PTSD on global distress and the second exploratory model examining PTSD with CFT on global distress.Further hierarchical multiple regression analyses were conducted to ascertain the relative contribution of each CFT-related variable and each PTSD subscale.The differences for global distress across therapy experience, and age at which abuse occurred, were also examined.

Ethical considerations
Considerations were taken with regards to the sensitivity of the topic and online recruitment.Information pertaining to risk and signposting further support was provided at the end of the study and after the CORE-OM due to the nature of the questions.EbE feedback was sought on all participant facing documentation and the survey, including language, and terms, such as 'sexual trauma', 'sexual abuse' and 'unwanted sexual experiences'.These three terms were used interchangeably based on the advice that participants may relate to different terms.Participants were made aware that they could withdraw data up to a week after completing the study and/or opt out at any point during the study.They were made aware that incomplete data would be saved but could be withdrawn, if preferred.Collection of sensitive demographic data was informed through consultation with EbE's and TransPlus (2022).Ethical approval was granted from the Royal Holloway, University of London ethics department on 11 November 2021.For accessibility, e-reader friendly versions of all documentation and graphics descriptions were made available on Qualtrics XM and e-reader text was added to all social media posts.

R E SU LTS
Participant characteristics are shown in Tables 1 and 2. Nearly half of the sample (49%) had experienced sexual abuse both in childhood and adulthood, whereas 22.6% had experienced abuse in childhood and 28.4% had experienced abuse in adulthood.Descriptive statistics for the CFT-related variables, PTSD and global distress were compared against the original scales samples (Table 5).
As anticipated, the study sample scored lower in compassion for the self, and compassion from others, and higher in fear of compassion from others, fear of compassion from the self, shame, self-criticism, PTSD and global distress than the original scales samples.Normality, skewness and kurtosis were also examined for the outcome of global distress.The sample data was found to be normally distributed (Kolmogorov-Smirnov = .06,df = 155, p = .20),with no skewness (skewness = .17,SE = .20)and slightly mesokurtic (kurtosis = −.85,SE = .39).Parametric assumptions for the inferential statistical tests were met.A normal distribution was indicated on a histogram, normal probability-probability plot and scatter plot.The Cooks distance (M = .007)indicated an absence of outliers, which was confirmed by a boxplot.

Regression of global distress on PTSD and CFT
Prior to the hierarchical multiple regression of global distress on PTSD and CFT-related variables, no transformation of the data was required.Multicollinearity was checked due to the high correlations among the CFT-related variables and PTSD.The collinearity statistics showed that the tolerance was greater than .10 on all the independent variables, and the variance inflation factor was less than 10 for all independent variables, suggesting that multicollinearity was not a concern.Both regression models were shown to be statistically significant ( PTSD and self-criticism had a significant unique positive relationship to global distress (PTSD: B = .64,β = .70,t(147) = 10.58,p < .001and Self-Crit B = .43,β = .22,t(147) = 3.44, p < .01).The difference in the means of model 1 and 2 are also statistically significant (ANOVA: F(1,153) = 273.16,p < .001and F(6,147) = 45.67,p < .001,respectively), suggesting that including CFT-related variables makes an important contribution to the overall model.
A further analysis was run to determine whether self-criticism and each of the significant PTSD subscales (hyperarousal, negative alterations in cognition and mood and re-experiencing) predicted global distress.All subscales, were found to significantly contribute to the variance in global distress: R 2 = .69,adjusted R 2 = .68,R 2 ∆ = .03,F(1,150) = 82.38,p < .001.

Therapy experience and age of abuse
The differences for global distress across therapy experience and age at which abuse occurred were found to be statistically non-significant (ANOVA: F(1,153) = .03,p = .858and F(1,153) = 2.05, p = .155,respectively).

DISCUS SION
This study examined the role of a variety of CFT-related variables in self-reported sexual traumaassociated distress.When controlling for the effect of other variables, the CFT-related measures contributed a small but significant amount of the variance in distress.However, when examining the individual roles of each CFT-related variable, self-criticism was the only variable to have a significant contribution, alongside the relevant PTSD subscales, to the variance in global distress.We may, therefore, deduce that self-criticism provides a significant role in predicting global distress.Whilst the remaining CFT-related variables were not found to predict global distress, it is important to note that the measurement for self-criticism used in this study is rooted in the CFT theory (Baião et al., 2015).However, it is difficult to extrapolate self-criticism explored here as exclusive to the CFT model.It is also important to note, the FSCRS showed questionable internal consistency which highlights the difficulty in isolating selfcriticism as a construct.Furthermore, developments in the CFT theory suggest there may be multiple subtypes of self-criticism such as the fear of 'fakery' or being an 'imposter', tying into evolutionary motivations for deception (Gilbert, 2009b, p. 316), reflecting the wider challenges of operationalizing self-criticism in this study.
When PTSD was compartmentalized into its subscales; hyperarousal, negative alterations in cognition and mood, re-experiencing and avoidance, all but avoidance played a significant role in predicting distress for this client population.This may make clinical sense as avoidant coping, although unhelpful, can develop after trauma (Zayfert & Becker, 2007).
Examination of the roles of each of the variables may provide insights into which aspects of the CFT model are important in a therapeutic context.Self-criticism, fear of compassion (from others and from the self), and shame showed the strongest relationships to distress.This supports previous findings that showed that self-criticism and fear of compassion played important roles as mediators between sexual abuse and various mental health outcomes (Boykin et al., 2018;Glassman et al., 2007;McAllister, 2019), as well as evidence for the positive association between fear of compassion and PTSD (Miron et al., 2016).Whilst shame showed a strong relationship with global distress, it did not predict it.However, these results should be viewed in the context of the measures used; the eight-item EISSshort form was used with the aim to minimize participant distress as the study was completed without clinician support; short-form scales have been found to be associated with reduced levels of internal consistency (Romero et al., 2012, cited in;Russell & Daniels, 2018).Compassion (from the self and from others) had small significant inverse relationships with global distress in this study, which reflects evidence for the scale as highly correlated with other aspects of compassion, but not with variables related to psychopathology (Gilbert et al., 2017).Contradictory to this, previous findings from other studies highlighted the significant negative associations between self-compassion and psychopathology (McLean et al., 2018;Minimol & Lucila, 2019).Therefore, self-compassion may be quite distinct from receiving compassion from others.It is important to consider the latter construct as this study captures people's perceptions of responding to compassion from others.However, a review by Mascaro et al. describes the distinctions between observational and self-report measures in measuring compassionate acts from others as providing 'insight into the subjective perception' versus 'intersubjective evidence of compassion itself' (Mascaro et al., 2020, p. 8) Actual compassion from others may, therefore, be better recorded through naturalistic observations.
It is important to consider challenges with construct validity, as various aspects of CFT may be difficult to operationalize and measure unlike the cognitive processes outlined in other evidence-based theories (Strauss et al., 2016), such as the cognitive model of PTSD.CFT is not a specific theory, but rather a therapeutic approach.The CFT model as defined by Gilbert ( 2010) is underpinned by evolutionary, motivational and affective processes that may be harder to objectively define and measure and thus determine their impact through research.
Many of the participants had experienced sexual abuse both in childhood and adulthood, highlighting that sexual abuse prevalence is difficult to characterize by age groups.The results showed no significant differences in distress across groups categorized by age of abuse occurrence, which contrasts with previous findings that point to differences in self-compassion across age groups (Kjose, 2019).This phenomenon, however, would require further specific research, as our results may be skewed by the high proportion of the sample having experienced abuse in both childhood and adulthood (49%).It was also noted that many participants had experienced some form of psychological support for up to 4 years or more and it is possible that this may have impacted the results.For example, prior experience in therapy may foster the capacity to accept compassion from others or alleviate the fear of compassion due to exposure to compassion from a therapist.This is highlighted by Willis et al.where compassion intervention for complex trauma-alleviated fears, blocks and resistances to compassion (Willis et al., 2023).Therefore, with samples who had not experienced therapy, we may expect the fear of compassion to be highly correlated with global distress.Previous experiences of therapy may also promote self-compassion, as has been shown by Minimol and Lucila (2019) and Tesh et al. (2015), which may have played a role in the small correlations in compassion (from others and from self ).It may also be the case that shame and self-criticism require more long-standing treatment than other CFT-related variables, as more entrenched and interpersonally reinforced difficulties (Gilbert, 2010); individuals high in shame and self-criticism have been found to do less well in standard treatments (Black et al., 2013;Rector et al., 2000).This may provide some explanation for the higher correlations found in shame and self-criticism with global distress in this sample.
It was also noted that the majority of the sample had been given one or more psychiatric diagnoses with the majority of these being depressive, post-traumatic stress and anxiety-based disorders.This was to be expected given the high risk for psychiatric disorders associated with sexual assault (Dworkin, 2020).However, an important consideration lies in whether people had received treatment corresponding to these diagnoses.The main therapies reported were counselling and CBT, which may be indicative of the sample mainly accessing UK services.However, it may also indicate that for a large proportion of this sample, cognitive and behavioural models had been used to specifically target depression, anxiety and PTSD.Targeted prior treatment for these diagnoses, therefore, may have impacted these results.
Conceptual limitations regarding operationalizing and measuring compassion have been discussed, and future research may wish to continue in its consideration of these challenges when devising scales.There is also another consideration regarding the assumption that measuring compassion related variables will accurately translate to the CFT model.Therefore, future experimental research into this therapy and its underlying constructs is important.Sample factors highlight important limitations of this study with regards to heterogeneity in the demographic representation of the sample.The sample was predominantly white, female, UK-based and heterosexual, suggesting that the results may not reflect the established diversity of patient populations.Future research may want to consider alternative purposive or quota-based sampling methods, as well as barriers to sexual trauma research engagement.Recruitment through social media presents an additional potential confounder as it assumes that participants feel able to approach a study advertisement acknowledging sexual trauma, which may not be the case for everyone.It is also important to consider that generalizability may be difficult to determine, as the parameters of what constitutes this population are difficult to define; the study examined those with self-reported experiences of sexual trauma, rather than those who met criteria for PTSD or Complex-PTSD.This decision was made in light of the challenges of capturing a full picture of the impact of sexual trauma through diagnostic means (Chivers-Wilson, 2006;Domino et al., 2020) and is in part why research examining models outside of the PTSD diagnostic paradigm is critical and timely.

CONC LUSIONS
The results from this study highlight the role of CFT targeting self-criticism as a useful adjunct to the PTSD cognitive model of therapy for those who are recovering from sexual abuse.Whilst the cognitive model of PTSD alone may be sufficient in the treatment for sexual trauma, the CFT model may support aspects of recovery that are based in affect and internalized processes such as shame and self-criticism, which may be harder to access through cognitive conceptualizations of trauma.Whilst self-criticism appeared to play a critical role in distress, it is not exclusive to CFT, and clinicians may wish to consider incorporating this aspect into their work more broadly, irrespective of the modality used.Future research investigations may wish to examine experimental designs utilizing CFT for survivors of sexual trauma and in-depth qualitative explorations of self-criticism as part of the CFT model, for this population.

AC K NOW L E DGE M E N T S
We wish to thank the participants who gave up their time to take part in this study.We also thank the Experts by Experience and TransPlus for their time in consultation.We would also like to express our gratitude to Professor Paul Gilbert, Dr Lisa McLean, Dr Deborah Lee, and the members of the Compassionate Mind Foundation community for guidance in this research area.

F U N DI NG I N FOR M AT ION
This research was funded by the Royal Holloway, University of London, Clinical Psychology Research Department.

CON F L IC T OF I N T ER E S T S TAT E M E N T
None of the authors have any competing interests to declare.

DATA AVA I L A BI L I T Y S TAT E M E N T
Due to the sensitive nature of the research supporting data is not available.

E T H IC A L A PPROVA L
Approval was granted from Royal Holloway, University of London ethics department.ID: 2743.

R E F ER E NC E S
20448260, 0, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/bjc.12465by Test, Wiley Online Library on [18/06/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

identity Sexual orientation Relationship status N = 155 Responses Percent N = 155 Responses Percent N = 155 Responses Percent
20448260, 0, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/bjc.12465by Test, Wiley Online Library on [18/06/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Participant characteristics for age, country and ethnicity.
T A B L E 1T A B L E 2 Participant characteristics for gender identity, sexual orientation and relationship status.

participants had received previous mental health support Whether previous mental health treatment was medical or psychological Whether participants had received/ were receiving psychological therapy
Frequencies for mental health support received.Frequencies for therapy received and known diagnoses.
T A B L E 3 20448260, 0, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/bjc.12465by Test, Wiley Online Library on [18/06/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Means and standard deviations for all variables in current study sample and in the original scales samples.
T A B L E 5 20448260, 0, Downloaded from https://bpspsychub.onlinelibrary.wiley.com/doi/10.1111/bjc.12465by Test, Wiley Online Library on [18/06/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Table7): Model 1 (global distress on PTSD) showed that PTSD had a significant effect on global distress (R 2 = .64,adjusted R 2 = .64,F(1,153) = 273.16,p < .001).Model 2 (global distress on PTSD and CFT-related variables) showed that the exploratory model contributed to 67% (adjusted R 2 ) of the variance in global distress (R 2 = .69,adjusted R 2 = .67,F(6,147) = 3.43, p < .01).CFT-related variables, therefore, contributed an additional 4.4% to the total variance (model 2, R 2 ∆ = .044),supporting the hypothesis that CFT-related variables may play a role in addition to PTSD in predicting distress after sexual trauma.The partial regression coefficients showed that only Results from hierarchical regression analysis.
T A B L E 7Note: Regression is significant at corresponding level: ***p < .001,**p<.01,*p<.05.T A B L E 6Correlations for all variables.