The influence of self-compassion on perceived responsibility and shame following acquired brain injury

ABSTRACT Primary objective The purpose of this study was to investigate the influence of perceived personal responsibility for an acquired ABI (ABI) on shame, and whether self-compassion moderates this relationship. We hypothesized that people who perceived themselves to be responsible for their injury would have high levels of shame and poorer recovery outcomes. Research design A mixed-methods design was employed using both standardized measures and a series of open questions. Methods and procedures 66 participants with ABI were included in the analysis. Data were analyzed using descriptive statistics, correlations, multiple regression, and thematic analysis. Main outcomes and results Significant relationships were found between self-compassion, shame, anxiety, and depression, but perceived responsibility for ABI was not correlated with any examined variables. Due to issues with the measurement of responsibility, it was not possible to complete all proposed forms of analysis. The thematic analysis revealed the ways participants’ injuries affected their perceived level of functioning, its consequences for sense of self, shame, and self-compassion. Conclusions This study concluded that people with ABI might experience shame with respect to the injury’s impact on functioning. Study limitations and implications for providing therapeutic interventions such as Compassion Focused Therapy and Acceptance and Commitment Therapy are discussed.


Introduction
The term 'compassion' describes the practice of understanding others' distress with a desire to alleviate suffering, a process that can help people learn to care for themselves during distress (1,2). However, when people experience negative events and distress, they tend to treat themselves less kindly than they would another in the same situation (3). People who experience shame due to fearing they are different from others can struggle to be self-compassionate, increasing sensitivity to criticism from the self and others and negatively impacting psychological wellbeing (4). Skills of compassion form the basis of several third-wave psychological therapies (5,9), including Compassion-Focused Therapy (CFT, 2), Acceptance and Commitment Therapy (ACT, 6), and Mindfulness-Based Therapies (e.g. MBCT,7,8). Third-wave therapies follow the first and second waves of thought-based psychological therapies developed in the 20 th century, such as Behavioral Therapy and Cognitive-Behavioral Therapy (CBT). The third-wave movement defines a shift away from the cognitive focus on what we think and feel, to a focus on how we relate to what we think and feel (5).
Reduced self-compassion and high shame and selfcriticism have been linked to poorer psychological wellbeing in various cancer populations (10,11). Similar research has shown that blaming oneself for a breast (12), lung (13), or colorectal cancer (14) diagnosis is associated with poorer psychological wellbeing, stigma, shame, guilt, and depression. These studies illustrate that shame and responsibility for illness or condition shame may equate to lower selfcompassion, which has detrimental effects on psychological wellbeing (15).
In general, acquired brain injuries (ABI) can lead to selfcriticism and lower levels of self-compassion (16,17), and these individuals can respond well to CFT (18)(19)(20). However, no research has considered why self-compassion levels are lower, and whether this is linked to self-blame as seen in other health population. The degree to which an individual could be considered to have played a role in the acquisition of their injury can be considered on a continuum. For example, individuals with encephalitis from a viral infection may feel less responsible than an individual who acquired TBI during dangerous sports (21). Thus, in a similar way in which this has been shown in health samples, perception of causation of an injury could influence subsequent psychological wellbeing, as in other health populations (12)(13)(14).
Findings of neurorehabilitation studies form a basis for these conceptions, where the importance of a positive sense of self for adjustment and psychological wellbeing during ABI recovery is emphasized (22,23), and discrepancies between pre-and postinjury self can lead to problems with adjustment (24,25). Lewington (26) and Jones and Morris (27) highlighted that it is difficult to adjust to receiving care from parents following ABI, part of which involves shame and self-criticism about being a "disappointment". Participants in Jones & Morris' study also found that directing blame toward a parent for an ABI led to increased psychological distress. And indeed, these adjustment processes can be affected by factors such as time since injury (23).
Other research indicates that a person is viewed less sympathetically if they are perceived to have contributed to their injury (21). An important link can be drawn here with a finding that being well-liked by others leads to positive self-evaluations (28), indicating the powerful impact of others' judgments for sense of self. Interestingly, Hart et al. (29) found that selfblame for injury was actually linked to reduced depression. This may illustrate that taking responsibility could be a coping mechanism, as doing so in a non-critical way may suggest use of a self-compassionate perspective. Indeed, if an individual considers the process to be positive and something they could emotionally recover and grow from, they are more likely to retain their sense of self and wellbeing (30). Consequently, this study hopes to explore the importance and rehabilitation implications of encouraging self-compassion and acceptance at an appropriate time since injury and in the right stage of adjustment to manage expectations and possible shame, whilst also acknowledging strengths.

The present study
Literature suggests that self-compassion is a protective factor for psychological wellbeing and adjustment, and can reduce shame and self-criticism in ABI populations (18). The perceived responsibility a person has for their ABI may also impact shame and psychological wellbeing (29). Therefore, this study aimed to investigate whether self-compassion moderates the effect of perceived responsibility on shame in this population. Thus, self-compassion could be a protective mechanism against the shame that results from responsibilityfor injury. It also assessed the impact of these concepts on psychological wellbeing, including anxiety and depression, based on findings from previous research (17,25,29).
Lack of insight into one's difficulties post injury may serve as a protective factor by acting as a "buffer" to protect the patient from a potentially difficult reality (31)(32)(33) and this was therefore controlled for. Time since injury was also controlled for to consider the influence of stage of adjustment (23).
The definitions of abstract concepts such as self-compassion and shame can vary (34), making it difficult for standardized measures to accurately capture these ideas. Therefore, this study also included open questions for participants to explore these concepts and these were analyzed using thematic analysis.

Research questions
The study investigated the following research questions; (1) Does self-compassion moderate the impact of PR on shame in ABI?; Hypothesized that as self-compassion increases, the impact of PR on shame would reduce, resulting in a significant interaction between selfcompassion and PR. (2) Are there relationships between self-compassion, selfawareness, shame, perceived responsibility (PR), and anxiety and depression for people with ABI? (Hypothesized relationships illustrated in Table 1). (3) What are the experiences of self-compassion, shame, and perceived responsibility for people following ABI?

Participants and procedures
This study employed a cross-sectional design, collecting quantitative and qualitative data through self-report questionnaires. The dependent variable was shame, and the predictor variables were perceived responsibility, self-compassion, gender, time since injury, and level of awareness into the impact of the injury on daily functioning. Approval for the study was granted by the London-Surrey NHS Research Ethics Committee. Individuals with ABI took part in anonymous online (35) or paper-based surveys between September 2017 and February 2018. They were recruited from six National Health Service (NHS) Trusts, the Brain Injury Rehabilitation Trust (BIRT) and six voluntary and charitable ABI organizations across Yorkshire and Northern Lincolnshire. The researcher visited services to meet potential participants, assess inclusion criteria, collect informed consent and assist with questionnaire administration, or provide a digital link. Following survey completion, participants were debriefed by the researcher or shown a debriefing page.
The inclusion criteria specified that participants must: have experienced an ABI after the age of 18; have English as a first language; and currently be in contact with an active rehabilitation setting for their injury, or in contact with a community or voluntary service related to ABI.
Exclusion criteria included: diagnosis of a degenerative condition, learning disability or neurodevelopmental condition; lack of capacity to consent to take part in the study; lack of ability to comprehend or produce speech to levels necessary for the tasks; and diagnosis of Post-Traumatic Stress Disorder in relation to ABI.

Demographics
Information was collected about age, gender, relationship status, education, time since injury, contact with service type, and length of service contact. Specific medical diagnoses were not collected because participants were recruited from large community sources, and access to accurate medical records was not always available. The study also focused on participants' personal perspective into their condition, and therefore diagnosis was not collected as this was not directly related to hypotheses.

Self-compassion
Self-compassion was measured using the Short Self-Compassion Scale (36), a shortened version of The Self-Compassion Scale developed by Neff et al. (1). The SCS consists of 26 items related to self-kindness, mindfulness, and common humanity. This scale has previously been administered to adults with ABI to measure the effectiveness of CFT interventions (37), and has high reliability (α = .93; 38).

Self-awareness
The Awareness Questionnaire (AQ, 39) measured insight. This 17-item form uses a 5-point Likert scale to rate participants' degree of difficulty across several tasks and functions.
Responses are compared to ratings of identical items by a friend or staff/family member, and discrepancy between the two scores indicated participants' self-awareness, with less awareness indicated by a larger discrepancy. The AQ has adequate construct validity (p = .34-.39), excellent internal consistency (a = .88) and excellent to adequate test-retest reliability for the participant (ICC = .80), and other person forms (ICC = .66), respectively (40). The AQ has been used extensively with participants with ABI (41,42).

Perceived responsibility
Participants were presented with the item "Please rate how much you think you are responsible for your brain injury on a percentage scale between 0-100, where 0 is not responsible at all, and 100 is fully responsible." This design is similar to an investigation into the role of PR in the development of posttraumatic stress disorder (PTSD) for people with ABI following road traffic accidents (43,44), and was chosen due to a scarcity of available PR in ABI measures.

Shame
The State Shame and Guilt Scale (SSGS) measured shame (46), a 15-item self-report measure of guilt, shame, and pride. This measure has been used in a study that investigated shame in individuals with cancer (13). In young adult samples, this measure had high internal consistency, test-retest reliability, and predictive and convergent validity (47).

Psychological wellbeing
Psychological wellbeing was measured using the Hospital Anxiety and Depression Scale (HADS; 48), a 14-item selfreport measure of anxiety and depression that has been used with people with ABI (18,20), and has proven validity for the subscales (49).

Qualitative questions
The following questions were used to further explore participants' experiences of perceived responsibility, shame, and selfcompassion.
(1) How positively or negatively do you view yourself following your injury? (2) How different is this to how you would have viewed yourself prior to your injury? (3) How kindly do you treat yourself? (4) How did your injury happen? (5) How responsible do you think other people were for any part of your injury happening? (6) How responsible do you think you were for any part of your injury happening? (7) What have been your biggest achievements or areas of growth/development since your injury?

Statistical analyses
Data were analyzed using IBM SPSS Statistics version 24.0 for Windows. Descriptive statistics were used to analyze the demographic data, psychological wellbeing, and the variance in the level of perceived responsibility for injury. A significance level of 5% was used for all statistical tests. Pearson's correlation coefficient was used to examine the correlational relationships between the predictor variables. Statistical comparisons were not carried out because PR was measured on a scale rather than allocating participants to defined groups of "responsible" and "not responsible." Linear multiple regression analyzes were completed to explore the degree to which the demographic and predictor variables could explain participants' scores of shame, and to carry out a moderation analysis. The moderating effect of selfcompassion was analyzed by examining the change in R 2 when an interaction between PR and self-compassion was added to a multiple regression model containing PR, self-compassion, gender, time since injury, and self-awareness. Forty-six of 66 participants (69.7%) completed the measure of self-awareness, and thus two regression analyzes were fitted; one with the full range of data available, and one with participants with selfawareness data available.
Qualitative data were analyzed using thematic analysis. The Braun and Clarke (50) analysis guidelines were followed to ensure data was of good quality prior to theme development. Responses which were very limited or unrelated to the questions were excluded from theme development in the final analysis.
Pearson's Chi-Squared tests were carried out for the categorical variables (age, gender, relationship status, level of education, and time since injury) to test for differences between participant settings, and revealed a significant difference between participants' setting and level of education (p = .027). There were no other relationships between setting and remaining demographic characteristics (p > .05). A One Way ANOVA revealed that time since injury was not significantly different between settings (F(4, 61) = 1.267, p > .05).

Dependent & predictor variables
A Shapiro-Wilk test of normality revealed that anxiety (p = .350) and depression (p = .118) levels were normally distributed, while self-compassion, PR, and insight were not normally distributed (p < .001). Scaled score means and corresponding standard deviations were calculated for all variables and can be found in Table 3. The level of shame (mean = 11.51; SD = 4.82) was higher for this sample than a normative sample of students without ABI, in which the mean was 6.71 (SD = 2.60) (47), suggesting that participants in this study had higher levels of shame than people without ABI.
Twenty (30.3%) participants did not complete the AQ due to the need for another person to provide ratings, particularly for those recruited from voluntary or charitable organizations. An independent samples T-test revealed a significant difference between participants' setting and whether they completed the AQ (p = .003). This indicated that participants recruited from inpatient and community settings were more likely to complete the AQ. 54.5% of participants rated themselves as 0% responsible for their injury, indicating a floor effect for this variable.

Are the variables related?
Relationships between the predictor variables were examined through Pearson's Product Moment correlation coefficient and are presented in Table 4. Due to the non-normal distribution of some predictor variables, bootstrapped P values are included. Shame was significantly positively correlated with depression and anxiety, and shame was significantly negatively correlated with self-compassion.
A significant negative correlation was evident between selfawareness and anxiety, and a negative relationship at the .1 level (p = .054) was shown between shame and self-awareness, suggesting that the more insight a person had into their abilities, the greater the possibility of them experiencing shame. A negative relationship at the .1 level (p = .055) was demonstrated between anxiety and self-compassion; participants who felt anxious were less likely to be self-compassionate. A significant negative relationship was also found between PR and depression, though, no other significant relationships were found between PR and the main dependent variables. Finally, depression and anxiety were significantly positively correlated.

Is shame related to an individual's perceived level of responsibility for their injury, and is this relationship moderated by self-compassion?
An independent samples T-test examined potential differences between participants who did and did not complete the AQ, and revealed no significant differences between these groups for all variables (p > .05). As the aim of the study was to investigate the influence of self-awareness on the other variables, and there were no significant differences between participants who did not complete the AQ, the main model of  regression included only the 46 participants who provided selfawareness information.
A significant regression coefficient was found for selfawareness and shame (β = -.190, SE = -.090, t = -2.102, p = .043), indicating that as self-awareness decreased (i.e. the discrepancy of the AQ increased) so did shame. The results also illustrated a significant regression coefficient for self-compassion and shame (β = -1.615, SE = .515, t = -3.138, p = .003). This demonstrates that as self-compassion levels increased, shame decreased. The regression coefficient for PR was not statistically significant (β = -.003, SE = .017, t = -.158, p = .875), meaning that PR did not share a relationship with any of the measured variables. The results from the regression model including all participant data illustrated similar results. A representation of the regression analysis is illustrated in Table 5.

Qualitative analysis
Thematic analysis was utilized to examine the responses of 42 (63.6% response rate) participants to the open questions. Of the participants who responded to the qualitative questions, 22 were female (52.4%). Four super-ordinate themes were identified with sub-ordinate themes in each: achievement and growth, sense of self, self-compassion, and functional impact of injury. These are shown in Table 6 with corresponding subordinate themes and quotes. Three of the super-ordinate themes related to participants' experiences of themselves due to and following their injury, and one described areas of growth since the injury. Correlation is significant at the.1 level (2-tailed)

Confidence
Commenting on how the brain injury impacted the person's confidence 'I was more confident and self-assured and took more risks.'

Independence
Commenting on how the brain injury impacted the person's level of independence

'[I] have to rely on others to do things I can't do now.'
Mental Health Commenting on how the brain injury impacted mental health difficulties 'Unable to do things I could do before as I suffer with tiredness and anxiety when I go out or in large crowds.'

Physical abilities
Commenting on how the brain injury impacted physical abilities 'I was a keen cyclist and used to push myself to do that. Did lots of walking and always took the longest route.'

Relationships /Social
Commenting on how the brain injury impacted relationships and social life 'I am more insular, lacking confidence and more reluctant to socialize with other than with very close family and friends.'

Responsibility
Feeling responsible in some way Perceived view that the individual or another is personally responsible for their injury in some way

Discussion
An initial hypothesis proposed that as PR increased, so would shame, anxiety, and depression. The findings of this study largely did not support this hypothesis as PR was only significantly associated with depression, and not the main predictor and dependent variables. This, therefore, meant that the study was unable to carry out all of the proposed statistical analyses. There was a trend for shame, self-compassion, and anxiety decreasing as PR increased, though not to a significant level. A potential explanation here is that responsibility is a difficult concept to capture through a single item, as it can have different implications depending on the type and causation of ABI. Indeed, Hart et al (29) illustrated that participants who could be considered objectively responsible for their injury were more likely to blame others. As authors did not make judgments on the level of responsibility, it is unclear whether the given levels of PR by participants would have been the same as how others would have judged their responsibility. Moreover, some of these results were based on correlational analyses. It is important to highlight that while correlational and regression analyses can demonstrate relationships between variables, causal effects cannot be detected. For example, good self-compassion skills could be the reason someone feels less shameful, or low shame may have contributed to an individual's ability to be self-compassionate.
This study asked participants to describe their experiences of ABI and PR, and inclusion criteria were made broad to consider the perspective of a wide range of ABI on perceived responsibility. As such, participants with particular types of ABI, such as those caused by involvement in fights, might be more likely to assume greater responsibility for injury. However, the study is unable to comment on the relationships between a specific type of ABI and the variables as this information was not collected. As such, statistical comparisons between ABI types were not carried out, though further research might make use of this design.
There is a possibility that some participants viewed taking responsibility to be a positive experience which actually resulted in a reduction in shame and mental health difficulties, also shown by Hart et al. (29). An association could be drawn here with the concept of "locus of control" (LOC), as people with "internal" LOC (who believe they have control over their own life, 52) are less likely to be depressed than people with "external" LOC (who attribute their lives as in control of others or concepts such as fate, 53). Studies examining LOC in ABI found an association between external LOC and depression and decreased quality of life (54,55). Participants who blamed others for their injury in Hart et al.'s study (29) might, therefore, have had an external LOC, explaining high depression levels in this group. From this, taking responsibility could mean the person is acknowledging the control they have over their lives and the ability to recover, perhaps bringing them closer to acceptance, resulting in less shame. This would explain the significant negative correlation between PR and depression, as perhaps taking responsibility allowed participants to acknowledge their role, and use this as an experience from which to recover. However, it is important to highlight that trends between PR and the predictor and dependent variables were not significant; thus, reliable conclusions cannot yet be drawn about this.
Shame was significantly positively correlated with anxiety and depression, supporting initial hypotheses and linking to previous findings that shame is associated with poorer psychological wellbeing (10,11). This illustrates how people with ABI experience similar associations between shame and wellbeing as those without injuries, making it appropriate to use established shame-targeting psychological therapies. Indeed, shame amongst participants in this study was higher than in studies of individuals without brain injuries (51). The finding that anxiety, and to a lesser extent shame (P value significant at the .1 level), increased with the self-awareness a participant had into their condition also supported initial hypotheses. This could indicate the protective nature of poorer selfawareness for difficult psychological experiences in ABI populations, as less insight into functioning makes it less likely for a person to perceive the injury's potential negative impacts (56). Toglia and Kirk (57) distinguished between insight and the psychologically motivated symptom of "denial," which functions to protect a person from stressors. It can be difficult to identify which process is at play, providing a rationale for a thorough assessment of selfawareness for ABI psychological rehabilitation. However, as the AQ relies on another person to rate the participant, reliability may be affected due to difficulty establishing how well the person knew the participant. Additionally, participants who provided qualitative information were likely to have more insight to be able to provide in-depth descriptions about their experiences, making it difficult to incorporate the qualitative data into discussions about selfawareness.
Self-compassion was significantly negatively correlated with shame, and with anxiety at the .1 level, indicating partial support for correlational hypotheses. This showed that high selfcompassion was associated with less shame and anxiety, providing evidence for the protective nature of self-compassion (1). This reinforces the suitability of therapies that target selfcompassion to improve anxiety in ABI populations.

Qualitative data
A mixed-methods approach was chosen for this study to allow for exploration of participants' experiences of the variables that were difficult to fully capture by quantitative data collection alone. The thematic analysis emphasized that participants found it difficult to be self-compassionate ("I'm hard on myself") but easier to be compassionate toward others ("I am generally more understanding toward others"), a common occurrence across many populations (3). As such, it would be important to encourage skills of compassion for people with ABI, particularly focusing on compassion toward the self. A link could be drawn here with the theme of functioning, where participants illustrated how their injury had affected areas of their lives such as their cognitive abilities, employment and independence, evidenced by quotes such as "[I] have to reply on others to do things I can't now." Difficulties with selfcompassion would likely contribute toward self-criticism about one's altered abilities following ABI, and explain correlations between self-compassion, shame, and anxiety. This demonstrates the importance of having therapeutic conversations to explore a person's perception of themselves and their functioning following ABI, as presence of self-criticism about this could hamper recovery.
Another theme explored the sense of self, as participants described how their sense of usefulness had changed following injury, illustrating further discrepancies between ideal and preinjury selves. For example, some participants commented on the sadness they felt about their current self, "I view myself badly now"; and others described a sense of loss, "I feel worthless and like I am causing more unnecessary stress and worry to my family." This could explain the higher than average levels of shame shown in the quantitative analysis, as participants struggled to be kind toward themselves about their altered abilities. Likewise, Gracey et al. (23) highlighted the importance of having a positive sense of self for adjustment and recovery in ABI. This study, therefore, strengthens the validity of the available literature for targeting shame and sense of self during psychological therapy for adjustment in ABI rehabilitation.
Regardless, participants were readily able to describe their biggest areas of growth or achievement since the injury to form the theme of "growth and achievement," in areas of their lives such as physical achievements ("learning to walk"), mental achievements ("My memory is improving a little") and employment. This suggests that though participants acknowledged their difficulties and feelings of shame, they remained able to reflect on the positive aspects of their recovery. This also links back to the 'Motivation to improve' and 'Positive sense of self' sub-ordinate themes within "Sense of self." Here participants described feeling able to retain a positive sense of self despite challenges, and be motivated to make the adaptations needed. This provides optimistic implications for the ability of people with ABI to consider their strengths and motivations, an important skill for approaches such as CFT and ACT which help individuals to focus on what is within their control to improve their wellbeing.

Limitations
This study relied on self-report, which can be susceptible to bias. The challenge of limited self-report questions was partly overcome through the collection of qualitative data. However, participants did not comment in depth to some questions, particularly those related to responsibility. Future research could use semi-structured interviews to allow the researcher to be adaptable with questioning and to allow greater depth from follow-up questions. Moreover, there are few available measures for self-compassion and shame, and these measures can differ in their definitions of the concepts and have not been used extensively with neurological populations. It was for this reason that a mixed-methods approach was chosen to allow exploration of the examined variables to capture further detail.
This study centered on responsibility for ABI, a new but important area with sparse literature. As responsibility is a difficult concept to capture and measure, adopting a mixedmethods approach was deemed the most effective way to explore this area. Regardless, the method of measuring PR was limited, which meant that all of the proposed statistical analyses could not be completed. Information about participants' type of ABI was not collected in this study, and as mentioned, collecting this information may have allowed for further exploration of the relationships between the variables.
The researchers recognize that not all participants fully completed the self-awareness measure (AQ), which is limited in that a second party is needed to complete the questionnaire. Some participants were understandably unable to recruit a second person to fill this out, particularly those from community settings, meaning that the study has limited data for this variable which may affect the results. However, this measure has been widely recommended and used amongst brain injury populations to measure awareness, and it was considered more important to use a valid measure to try to measure insight as accurately as possible, rather than eliminating this variable from the study.
It should also be acknowledged that participants may have interpreted the quantitative and qualitative questions about PR to be asked whether they intentionally "blamed" themselves. While responsibility should capture the objective role a person played in their injury, self-blame describes the subjective process of an individual believing and feeling guilty about the personal role they had in their injury causation (44). Responsibility does not always lead to selfblame; a person might be responsible for their injury but not blame themselves, or vice versa. This study managed this by asking participants to describe their experience of injury responsibility to consider subjective beliefs, and the majority of participants answered in depth which suggests understanding. However, some participants may have interpreted this question as self-blame, possibly impacting the validity of this measure. Further research could use semistructured open-ended interview questions to further explore PR, and be flexible with questioning if there are misinterpretations.

Conclusions and future directions
This study is the first of its kind to explore the relevance of a person feeling personally responsible for ABI and their ability to practice self-compassion. It has given strength to the literature base that people with ABI experience shame, anxiety, and depression, and also struggle to be kind to themselves. The thematic analysis revealed that participants had changes to their sense of self and worth potentially due to their injury changing their functioning, and had feelings of sadness and shame about this.
This study could not reliably conclude that responsibility directly contributed to shame and poorer psychological wellbeing in ABI populations. Despite this finding, there are thoughts for further research into responsibility and ABI as this area is very new. Longitudinal designs could examine whether the time since injury impacts responsibility and consider the role of responsibility in longer-term adjustment, similar to how Bennett et al. (12) found that blaming oneself for cancer diagnoses negatively impacted adjustment processes. The statistical flexibility of the responsibility measure could be improved using a forced choice rather than rating scale, to avoid floor effects. Additionally, responsibility could have been explored deeper by asking open questions instead of asking participants to rate their PR.
More research could explore responsibility and LOC (54), and also explore if people feel they are viewed and treated differently by others since their injury, highlighting the implications of this for their own sense of self and wellbeing (21,28). This could be broadened by studying shame and compassion levels of family members of individuals with ABI, and the implications for family therapy as part of psychological rehabilitation.
Overall, people with ABI likely experience shame and selfcompassion in a unique way, due to the role the injury has played in their functioning. Therefore, a rationale is provided to use third-wave psychological approaches and therapies which encourage self-compassion and acceptance (6). This study also demonstrates the need to tailor these therapies to incorporate factors associated with difficulties adjusting to ABI, such as the impact on the sense of self and functioning, to make them appropriate for this population.
• Dashes (M-dash) should be clearly indicated in manuscripts by way of either a clear dash (-) or a double hyphen (--

References
References should be presented in a separate section at the end of the document, in accordance with Vancouver system guidelines (see Citing Medicine, 2nd ed.). The references should be listed and numbered based on the order of their first citation. Every reference should be assigned its own unique number. References should not be repeated in the list, with each mention given a different reference number, nor should multiple references be combined under a single reference number. Digits in parentheses (e.g. (1,2)) should be used for in-text citations. Citations should precede terminal (e.g., periods, commas, closed quotation marks, question marks, exclamation point) and nonterminal punctuation (e.g., semicolons, colons).

Reference numbers should not be placed in parentheses.
Checklist: what to include (1) Author details. Please ensure everyone meeting the International Committee of Medical Journal Editors (ICJME) requirements for authorship is included as an author of your paper. Please include all authors' full names, affiliations, postal addresses, and e-mail addresses on the cover page. Where appropriate, please also include ORCiDs and social media handles (Facebook, Twitter or LinkedIn). One author will need to be identified as the corresponding author, with their e-mail address normally displayed in the published article. Authors' affiliations are the affiliations where the research was conducted. If any of the named coauthors moves affiliation during the peer-review process, the new affiliation can be given as a footnote. Please note that authorship may not be changed after acceptance. Also, no changes to affiliation can be made after your paper is accepted. Read more on authorship here.
(2) Structured abstract. Author agreement/Use of third-party material Authors are responsible for obtaining permission to reproduce copyrighted material from other sources and are required to sign an agreement for the transfer of copyright to the publisher. As an author, you are required to secure permission if you want to reproduce any figure, table or extract text from any other source. This applies to direct reproduction as well as "derivative reproduction" (for which you have created a new figure or table which derives substantially from a copyrighted source). Please see our page on requesting permission to reproduce work(s) under copyright for more guidance. Authors are required to sign an agreement for the transfer of copyright to the publisher. All accepted manuscripts, artwork, and photographs become the property of the publisher. Guidelines for medicine and health publications Disclosure of interest Please include your disclosure statement under the subheading "Disclosure of interest." If you have no interests to declare, please state this (suggested wording: The authors report no conflict of interest). For all NIH/Wellcomefunded papers, the grant number(s) must be included in the declaration of interest statement. Read more on declaring conflicts of interest here.

Clinical Trials Registry
In order to be published in a Taylor & Francis journal, all clinical trials must have been registered in a public repository at the beginning of the research process (prior to patient enrollment). Trial registration numbers should be included in the abstract, with full details in the methods section. The registry should be publicly accessible (at no charge), open to all prospective registrants, and managed by a not-for-profit organization. For a list of registries that meet these requirements, please visit the WHO International Clinical Trials Registry Platform (ICTRP). The registration of all clinical trials facilitates the sharing of information among clinicians, researchers, and patients, enhances public confidence in research, and is in accordance with the ICMJE guidelines.

Complying with ethics of experimentation
Please ensure that all research reported in submitted papers has been conducted in an ethical and responsible manner, and is in full compliance with all relevant codes of experimentation and legislation. All papers which report in vivo experiments or clinical trials on humans or animals must include a written statement in the Methods section. This should explain that all work was conducted with the formal approval of the local human subject or animal care committees (institutional and national), and that clinical trials have been registered as legislation requires. Authors who do not have formal ethics review committees should include a statement that their study follows the principles of the Declaration of Helsinki.

Consent
All authors are required to follow the ICMJE requirements on privacy and informed consent from patients and study participants. Please confirm that any patient, service user, or participant (or that person's parent or legal guardian) in any research, experiment, or clinical trial described in your paper has given written consent to the inclusion of material pertaining to themselves that they acknowledge that they cannot be identified via the paper; and that you have fully anonymized them. Where someone is deceased, please ensure you have written consent from the family or estate. Authors may use this Patient Consent Form, which should be completed, saved, and sent to the journal if requested.

Health and safety
Please confirm that all mandatory laboratory health and safety procedures have been complied within the course of conducting any experimental work reported in your paper. Please ensure your paper contains all appropriate warnings on any hazards that may be involved in carrying out the experiments or procedures you have described, or that may be involved in instructions, materials, or formulae.
Please include all relevant safety precautions; and cite any accepted standard or code of practice. Authors working in animal science may find it useful to consult the International Association of Veterinary Editors' Consensus Author Guidelines on Animal Ethics and Welfare and Guidelines for the Treatment of Animals in Behavioral Research and Teaching. When a product has not yet been approved by an appropriate regulatory body for the use described in your paper, please specify this, or that the product is still investigational.