sign in sign up
<<
Home , Focusing (psychotherapy), Isolation (psychology)

J Contemp Psychother (2015) 45:89–98 DOI 10.1007/s10879-014-9286-8

ORIGINAL PAPER

Compassion Focused Therapy to Counteract , Self-Criticism and Isolation. A Replicated Single Case Experimental Study for Individuals With Social

K. Boersma • A. Ha˚kanson • E. Salomonsson • I. Johansson

Published online: 6 November 2014 Ó Springer Science+Business Media New York 2014

Abstract Most forms of psychological distress encom- Introduction pass both the relation to the self in the form of shame and self-criticism, as well as the relation to others in the form of Most forms of psychological distress encompass both the distance and isolation. These are often longstanding and relation to the self in the form of shame and self-criticism, pervasive problems that permeate a wide range of psy- as well as the relation to others in the form of distance and chological disorders and are difficult to treat. This paper isolation. Shame is a painful self-conscious associ- focuses on how problems with shame and self-criticism can ated with the perception of having a personal attribute, be addressed using focused therapy (CFT). In a characteristic or behavior that others find unattractive or pilot study we tested the effectiveness of CFT with a single undesirable and that may result in rejection or being put case experimental design in six individuals from down (Gilbert 1998). Included in a shameful experience are social anxiety. The aim was to establish whether CFT lead oftentimes self-critical and self-attacking cognitions (Gil- to increases in self-compassion, and reductions in shame, bert and Miles 2000). These cognitions are an important self-criticism and social anxiety. Moreover, the aim was to target for intervention as they trigger, perpetuate and investigate to what extent participants were satisfied and intensify emotional reactivity. For example, Longe et al. experienced CFT as helpful in coping with social anxiety (2010) compared the neuronal correlates of self-criticism and in increasing self-compassion. Taken together the and self-reassurance using fMRI and showed that self- preliminary results show that CFT is a promising approach. critical thinking engaged brain regions signaling greater CFT was effective for 3 of 6 participants, probably effec- error processing and behavioral inhibition. Self-reassur- tive for 1 of 6 and more questionably effective for 2 of 6 ance on the other hand, engaged brain regions similar to participants. These results add to the empirical evidence expressing compassion and towards others. that CFT is a promising approach to address problems with Indeed, compassion and empathy towards the self could self-compassion. This research body is as of yet small, and be seen as the antidote of self-criticism and shame and it more studies are needed. appears that highly self-critical individuals have a very hard time mobilizing self-compassion. For example, there Keywords Compassion focused therapy Á Self-criticism Á are indications that highly self-critical individuals have Shame Á Social anxiety difficulties attenuating their physiological reactivity through compassionate self-soothing (Rockliff et al. 2008). In this study, practicing self-soothing imagery was gener- ally related to increased heart rate variability and decreased cortisol levels, but this was not the case for those with high levels of self-criticism and low levels of social safeness. K. Boersma (&) Á A. Ha˚kanson Á E. Salomonsson Á Moreover, several treatment studies show that high pre- I. Johansson treatment levels of self-criticism in general predict worse Department of Law, Psychology and Social Work, O¨ rebro University, O¨ rebro, Sweden treatment outcome (Cox et al. 2002; Marshall et al. 2008). e-mail: [email protected] These results indicate that self-criticism is hard to change 123 90 J Contemp Psychother (2015) 45:89–98 and that the mere (self) instruction to be more self-reas- perspective with specific cognitive, behavioral, mindful- suring and empathetic may not suffice for those who are ness and compassion focused imagery exercises that foster chronically self-critical. Over and above the direct patho- self-care in the form of compassion for self and for others. genic effect that self-criticism has in firing up the threat The aim of CFT is to, through these efforts, increase system, it seems that self-critical individuals may be unable empathy and for one’s own distress, increase to access and generate self-directed warmth and soothing. mindful awareness without judgment or blame, increase the This means that psychological interventions should not ability to refocus and activate safety-signaling processing only focus on decreasing self-criticism but also on helping systems, generate compassionate (warmth and self-critical individuals access and build positive emotional affiliation) and increase compassionate attention, thinking experiences in the form of warmth, soothing and self- and behavior. reassurance. While CFT has been extensively described only a few, The concept of self-compassion has received increasing small, empirical studies have been published (e.g. Gilbert clinical and scientific . While there are several and Procter 2006; Laithwaite et al. 2009; Judge et al. 2012; definitions, one of the most commonly cited is Neff’s Lucre and Corten 2013; Gale et al. 2014). The results of definition (Neff 2003), describing self-compassion as being these studies are promising but there is a need to replicate kind and understanding towards oneself in instances of and extend these results to other populations and settings. or failure rather than being harshly self-critical; perceiving For example, no empirical data has yet been published on one’s experience as part of the larger human experience the effectiveness of CFT on social anxiety disorder, rather than seeing them as shameful and isolating failures although reports and accounts have been made of it being and holding painful thoughts and in mindful used clinically (Henderson 2011). Yet, CFT might be an awareness rather than over identifying with them. These especially suited treatment for social anxiety disorder given concepts draw largely on Buddhism, but are, in many ways, its large shame and self-criticism component. For example, also related to themes stressing the Cox et al. (2004) showed that self-criticism was robustly importance of self- and self- (Barnard associated with social phobia in a large general population & Curry, 2011). However, the concept of self-compassion mental health survey. Moreover, Gilbert (2000) points out can be said to stretch beyond just self-acceptance and self- the great overlap between social anxiety and shame prob- kindness to incorporate a sense of inter-connectedness to lems in that both can be viewed as submissive strategies in humanity at large, equanimity, as well as and mean- contexts where people feel vulnerable to loss of social ing in the face of life’s difficulties (Neff et al. 2007). standing, attractiveness, rejection and/or criticism. Indeed, Several-related-treatment approaches have been devel- this study showed that there is a strong between oped recently, as a means to specifically and directly target social anxiety and shame. In addition, there is evidence that compassion (see Singer and Botz 2013). These therapies people with social anxiety have problems with self-com- share in many ways the strengths of mindfulness oriented . For example, Werner et al. (2012) studied self- therapies such as Acceptance and Commitment Therapy, compassion in a clinical sample of individuals diagnosed Dialectical Behavior Therapy and Mindfulness Based with social anxiety disorder. The results showed that Stress Reduction (Barnard and Curry 2011), but stand out individuals with social anxiety disorder had significantly in their explicit focus on increasing compassion. One of lower levels of self-compassion than healthy controls and these, Compassion Focused Therapy (CFT; also referred to that low self-compassion was associated with greater as Compassionate Mind Training (CMT)) focuses on for both negative and positive evaluation. Thus, the large developing a warm, compassionate and accepting attitude self-criticism component and the low level of self-com- towards the self and others in order to specifically coun- passion suggest that targeting self-compassion may be teract shame, self-criticism and isolation (Gilbert 2014). particularly important in individuals with disorders that While originally designed and tested for pervasive prob- include problems with social anxiety. lems with self-criticism in chronically depressed patients Therefore, as a pilot study, we aimed to investigate the (Gilbert and Procter 2006), CFT has been extended to effect of CFT on self-compassion, shame and self-criticism conceptualize problems with shame and self-criticism in in individuals with social anxiety. As the application of this populations with for example eating disorder, psychosis treatment in this population is a largely unexplored area, and social anxiety disorder (e.g. Gale et al. 2014; Gumley we make use of a replicated single case experimental et al. 2010; Werner et al. 2012). CFT integrates influences design (Kazdin 2011). While this design provides the from cognitive behavioral theories, , opportunity to draw valid inferences of treatment effec- Buddhism, attachment theory and evolution theory (Gilbert tiveness, the requirement to collect data on large and 2010). It combines thorough psycho education on human homogeneous groups is circumvented. Thus, this type of (emotional) functioning and the brain from an evolutionary design is well suited to test initial feasibility and get a 123 J Contemp Psychother (2015) 45:89–98 91 preliminary indication on effectiveness. Specifically, this Intervention pilot study aims to answer the following questions: Does CFT lead to an increase in self-compassion, and reductions The intervention consisted of eight, weekly, individual in shame, self-criticism and social anxiety? To what extent sessions of 1 h. The content of the intervention was are participants satisfied and experience the treatment as inspired by the book ‘Compassionate-Mind Guide to helpful? Building Social Confidence’ (Henderson 2011) and similar self-help books (Tirch 2012; Goss 2011) that are based on compassion-focused therapy as described by Paul Gilbert Methods (see Gilbert and Procter 2006; Gilbert 2010, 2014). A main adaptation consisted of transforming self-help format of Design and Analyses Henderson (2011) to a face-to-face therapy format. Fur- thermore, somewhat more explicit focus was put on A replicated single case experimental design was used with addressing safety behaviors and avoidance (using valuing weekly repeated measurement during a three-week base- and guiding principles of self-validation, acceptance, val- line period and an eight-to-nine week treatment period. ued direction and compassion from respectively ACT and Using the baseline as a benchmark, 6 participants function DBT). Details of the content of the protocol are presented as their own control and the primary analysis is a com- in Appendix 1. Each session contained psycho education parison of daily diary data during the baseline and sub- and Socratic dialogue around a theme, in session training sequent treatment phase. This analysis is complemented relating this theme to the participant’s own situation, and with analysis of change of scores on standardized measures in-session experiential mindfulness and imagery exercises from baseline to post treatment, and a short term focused on compassion. Participants were given a written (2–4 week) follow up. psycho education, information about common obstacles that may arise during homework assignments, a training Participants and Procedure dairy and instructions for homework assignments after each session. In the training diary, participants were asked to fill Participants were recruited through advertisements posted in how much they practiced each of the exercises during on bulletin boards throughout campus as well as on a the week in between sessions. If participants encountered student internet forum. Inclusion criteria were experiencing problems related to training during the week, they were social anxiety (operationalized as fulfilling the DSM-IV able to contact their therapist by email or read the docu- criteria for social phobia according to the Social Phobia ment on ‘‘Common barriers’’ that they brought with them Screening Questionnaire (SPSQ; Furmark et al. 1999) and from the sessions. Therapists were two psychology students being a university student. To prevent inclusion of indi- who were in their final part of their professional clinical viduals in need of specialist psychological care, persons training program. They received weekly supervision of a with serious depressive symptoms (operationalized as [27 licensed psychotherapist and followed a manual. Audio points and/or [2 on question on suicidal ideation of the files with meditation and imagery exercises were included Montgomery Asberg Rating Scale (MADRS-S; in the training program and were sent to participants by Svanborg and A˚ sberg 1994), substance abuse and serious e-mail in concurrence with each session. mental disorder were excluded from participation. Fur- thermore, participants could not be in currently ongoing Measures psychological treatment or recently (last 3 months) having initiated psychopharmacological treatment. After screening The Social Phobia Screening Questionnaire (SPSQ; Fur- and assessment six persons could be offered participation mark et al. 1999) was used to screen for social anxiety. The in the project. SPSQ can be used to get an indication of social phobia The 6 participants (5 females and 1 male; age range against the criteria in DSM-IV. While formal diagnosis was 20–32) had studied 1–5 semesters (median = 2.5). All not part of this study, the SPSQ was used to benchmark the reported that they experienced social anxiety and level of difficulties that participants experience and whe- more or less throughout their whole life, and all fulfilled ther they could be judged to belong to a clinical population the DSM-IV criteria for social phobia according to the or not from pre to post measurement. The SPSQ is reported SPSQ. One participant (P6) completed three of the eight to have high concurrent validity (r = 0.79 with SIAS) as sessions (38 %) and then chose to discontinue. The well as internal consistency (a = 0.90) (Furmark et al. reported cause was lack of motivation and time. However, 1999). this participant consented to remain in data analysis and fill The Self-Compassion Scale (SCS; Neff 2003) was used out post measurement. to measure self-compassion. The measure contains 26 123 92 J Contemp Psychother (2015) 45:89–98 items that are scored on a five-point Likert scale the question; all referring to the past week). Participants (1 = ’’almost never’’ to 5 = ’’almost always’’; range filled out the diary at home and brought it completed to 26–136). While the scale can be divided into 5 subscales, each training session. If the participant forgot the diary was this study uses the total score on the scale. In accordance filled out on the spot before the session started. Therapists with the original article, mean scores on each of the sub- did not have access to dairy scores during treatment. scales were added to form a total score. SCS has been shown to have good psychometric properties (Neff 2003). Statistical Analyses The Social Interaction Anxiety Scale (Mattick and Clarke 1998) was used to measure the degree of anxiety Weekly dairy ratings on shame, self-criticism and self- experiences in social interactions. SIAS contains 20 items compassion were graphically displayed and the percentage that are scored on a five-point Likert scale (0 = ’’strongly of data points exceeding the median of baseline phase disagree’’ to 4 = ’’agree’’; range 0–80). The internal con- (PEM) approach was used to complement visual inspection sistency of the SIAS (a = 0.93) and comparisons between of changes in level and trend between baseline and treat- SIAS and several other instruments that measure social ment (Ma 2006; Kazdin 2011). In the PEM approach, the anxiety have shown medium to high correlations and good baseline median is used as a benchmark to judge change validity (Mattick and Clarke 1998). during the treatment phase against. The percentage of The Mongomery Asberg Depression Rating Scale treatment phase data points above the baseline median (if (MADRS-S; Svanborg and A˚ sberg 1994) was used to increase is expected such as with self-compassion) or screen for the degree of depressive symptoms where below the baseline median (if decrease is expected such as potential participants with severe depression and/or sui- with shame and self-criticism) is calculated. As criteria for cidal ideation were excluded (see ‘participants and proce- interpretation, 90–100 % of treatment phase data points dure’). Depression scores of the 6 participants ranged from below the median signified a strong effect, 70–90 % a 10 to 19 (signifying slight to mild depression). The moderate effect and \70 % a questionable or no effect MADR-S was not used for further assessment of treatment (Scruggs et al. 1986). As this study uses a relatively short effect. baseline, the mean baseline score was used as a benchmark, A treatment evaluation at the end of treatment period instead of the median. was used to assess treatment satisfaction. Three questions To investigate whether there was a reliable pre to post were asked: ‘How satisfied are you with treatment?’ difference between the scores on standardized measures for (scoring alternatives: ‘very dissatisfied’, ‘quite dissatis- self-compassion and social anxiety, the Reliable Change fied’, ‘neither satisfied, nor dissatisfied’, ‘quite satisfied’, index (RC, Jacobson and Truax 1991) was calculated. For ‘very satisfied’); ‘To what degree do you consider you have this calculation, the standard deviation and test–retest improved your ability to cope with discomfort in social reliability of the SCS and the SIAS were obtained from situations?’ and ‘To what degree do you consider you have previous research (Mattick and Clarke 1998; Neff 2003). improved your ability to be self-compassionate?’ (scoring Furthermore, clinically significant improvement on social alternatives for both questions: ‘not at all’, ‘a little’, anxiety (SPSQ) was determined for each individual using a ‘somewhat’, ‘a lot’, ‘very much’). post-test score within two standard deviations (SDs) of the A weekly diary was used to establish a baseline and to mean of the normal population as a benchmark, in accor- track the process of participants’ change in shame (a dance with Jacobson and Truax (1991). Swedish normative composite of ‘How often have you felt ashamed of some- data (Furmark et al. 2000) were used (SPSQ M = 9.1, thing you thought, said or did?’ and ‘How discomforting SD = 7.3). have these feelings of shame been?’), self-criticism (a composite of ‘How often have you felt you failed or thought you failed?’; ‘How often did you become angry Results frustrated/disappointed with yourself when you felt you failed or felt you did not meet the standard you set for Figure 1 graphically displays participants’ weekly dairy yourself?’ and ‘How discomforting has it been to get ratings on shame, self-criticism and self-compassion. angry/frustrated/disappointed with yourself and have self- Table 1 shows pre to post changes on standardized mea- critical thoughts and feelings?’) and self-compassion (a sures of self-compassion and social anxiety. Participant 1 composite of ‘As I experienced difficulties, I have tried to showed consistent improvements of moderate size in diary show myself warmth and comfort myself’ and ‘How often scores from baseline to treatment on all three variables as have you managed to feel warmth and compassion towards well as reliable and clinical improvements on all pre to post yourself?’) across treatment. The items were scored on a measures. In the treatment evaluation, this participant rated five-point Likert scale (0–4 with end points depending on a high treatment satisfaction (‘very satisfied’) and a high 123 J Contemp Psychother (2015) 45:89–98 93

Fig. 1 Graphical display of 4 4 weekly diary ratings of self- criticism, shame and self- 3 3 compassion. PEM = Percentage of treatment 2 2 phase data points above (if increase is expected as with 1 1 self-compassion) or below the baseline mean (if decrease is 0 0 expected as with shame and 123456789101112 123456789101112 self-criticism) shame P1 (PEM 89%) shame P2 (PEM 11%) self-cricism P1 (PEM 89%) self-cricism P2 (PEM 11%) self-compassion P1 (PEM 89%) self-compassion P2 (PEM 56%)

4 4

3 3

2 2

1 1

0 0 1234567891011 123456789101112 shame P3 (PEM 100%) shame P4 (PEM 22%) self-cricism P3 (PEM 75%) self-cricism P4 (22%) self-compassion P3 (PEM 50%) self-compassion P4 (44%)

4 4

3 3

2 2

1 1

0 0 12345678910111213 123456 shame P5 (PEM 90%) shame P6 (PEM 67%) self-cricism P5 (PEM 50%) self-cricism P6 (PEM 67%) self-compassion P5 (PEM 100%) self-compassion P6 (PEM 100%) degree of perceived improvement (‘improved a lot’) in and being ‘somewhat’ improved in ability to cope with ability to cope with social anxiety and in self-compassion. social anxiety and in self-compassion. Participant 4 showed Participant 2 showed some possible but questionable no clear changes in diary scores but clinically significant changes in dairy scores on self-compassion but reliable improvements on pre to post measures for social anxiety, as change on this variable and an additional reliable change well as reliable changes on self-compassion. This partici- on social anxiety on pre to post measures. This participant pant rated being ‘quite satisfied’ with treatment and being rated a high treatment satisfaction (‘very satisfied’) and a ‘somewhat’ improved in ability to cope with social anxiety high degree of perceived improvement in ability to cope and in self-compassion. Participant 5 showed consistent with social anxiety and in self-compassion (‘improved a changes of moderate and strong size in diary scores on two lot’). Participant 3 showed changes of strong to moderate out of three variables as well as reliable improvements on size in diary scores on two out of three variables but no two out of three pre to post measures. This participant rated reliable improvements on any of the pre to post measures. a high treatment satisfaction (‘very satisfied’) and a This participant rated being ‘quite satisfied’ with treatment ‘somewhat’ respectively high degree (‘improved a lot’) of

123 94 J Contemp Psychother (2015) 45:89–98

Table 1 Pre post raw scores, reliable change index and clinically research body consists as of yet of relatively small, pilot significant change studies, and more studies are clearly needed. Participant 1 2 3 4 5 6 Discussion of Results Self-compassion (SCS) Pre 15.8 17.75 10.3 15.7 8.08 9.45 Not all participants improved on all variables. For example, Post1 20.5 21.25 13.4 19.35 16.25 14.7 improvement on self-criticism was not as consistent or Post2 21.7 22.2 12.45 21 15.55 – convincing as on self-compassion. This could reflect the RC 4.20* 3.17* 1.53 3.78* 5.35* 3.74* established difficulty of changing self-critical thinking (e.g. Social anxiety (SIAS) Cox et al. 2002). It could also reflect that the measure used Pre 60 32 34 31 62 25 was not sensitive enough to pick up true changes. Indeed, a Post1 40 29 27 27 60 19 clinical observation was that the psycho education and self- Post2 36 24 36 26 49 – monitoring elements related to self-criticism did lead to RC -5.09* -1.70 0.42 -1.01 -2.75* -1.27 participants shifting their perspective. Several participants Social anxiety (SPSQ) commented on that they were surprised to find out that they Pre 41 18 20 29 38 35 were at many times unduly harsh and critical towards Post1 25 17 14 22 29 25 themselves. However, it could be that the treatment dose Post2 21 12 20 19 34 was not enough to effectively change self-critical thinking. CSI Y Y N Y N N Indeed, as self-critical thinking often is chronic and long Pre = pretest, Post1 = post measurement directly after intervention, standing, more extended treatment may be necessary to Post2 = post measurement, 2–4 weeks after intervention. influence it. Not in the least it may be important to address RC = reliable change index relating to change between Pre and hindrances and blocks to feeling positive affiliative emo- Post2. CSI = clinically significant improvement. SCS = self-com- tions and letting go of self-criticism. For some individuals passion scale; SIAS = social interaction anxiety scale; SPSQ = social phobia screening questionnaire. *p\ 0.05 feelings of warmth and affiliation can be hard and even be frightening. These have been shown to be highly correlated to self-criticism (Gilbert et al. 2011). The treatment in this study did include some focus on exploring perceived improvement in ability to cope with social possible blocks and fears but could have been more explicit anxiety and in self-compassion. Participant 6 showed a and extensive in addressing this. Future studies could also change of strong size in diary scores on self-compassion as include measurement of fears of compassion, and thereby well as reliable change on self-compassion on pre to post study its possible moderating role in treatment effect. measures. This participant rated being ‘quite satisfied’ with Most systematic improvements were seen on self-com- treatment and being ‘somewhat’ improved in ability to passion where 5 of 6 participants showed reliable cope with social anxiety and in self-compassion. improvements on the pre to post measure. This could be an indication of the sensitivity of the measure, picking up on subtle changes, but could also reflect that this variable was Discussion most systematically addressed in treatment. Indeed, CFT includes a clear focus on psycho education with the aim to This pilot study tested the effectiveness of compassion ‘deshame’ and increase empathic understanding of oneself focused therapy (CFT) using a single case experimental and ones problems. A clinical observation is that the psy- design with six individuals suffering from social anxiety. cho educational elements that focus on how our evolved The aim was to establish whether CFT lead to increases in hard wiring sets us up for difficult helped par- self-compassion, and reductions in shame, self-criticism ticipants to gain a new perspective on that their responses and social anxiety. Moreover, the aim was to investigate to are’not their fault’. This helped to bring about a clear what extent participants were satisfied and experienced motivational shift towards empathetic understanding of CFT as helpful in coping with social anxiety and in ones difficulties and trying to help and support oneself. increasing self-compassion. In summary, the results indi- Results on improvements in social anxiety were more cate that CFT was effective for 3 of 6 participants (P1, P2 mixed. Two participants showed reliable improvements on and P5), probably effective for 1 of 6 (P4) and more socials anxiety as measured with SIAS and three showed questionably effective for 2 of 6 participants (P3 and P6). clinically reliable improvement on social anxiety as mea- Taken together, these results add to the growing body of sured with SPSQ. This may indicate that the SIAS and empirical evidence that CFT may be seen as a promising SPSQ pick up on different aspects of social anxiety, but approach to address problems with self-compassion. This could also signal that the participants included in this study 123 J Contemp Psychother (2015) 45:89–98 95 were heterogeneous when it comes to the severity of their of the effect and relate this to the effect of other treatments. problems. There was some indication that participants with For example, the percentage of treatment phase data points more severe problems had larger treatment effects. This above the mean approach that was used as a complement to could be due to floor effects on the measures but also to the visual analysis is insensitive to magnitude of data points face validity of this approach for those with obvious self- above the mean. Also, this method does not consider trend compassion and self-criticism difficulties. Future studies and variability in data points of treatment phase. could include measurement of treatment credibility to pick Another shortcoming is that, while all participants had up on these variations. sufficient levels of social anxiety to meet DSM-IV criteria As in many that have a cognitive and for social phobia according to the SPSQ, some of the behavioral and/or a mindfulness orientation, emphasis in participants scored low on the weekly ratings, possibly CFT is put on active engagement and training on the part of leading to floor effects and difficulty in detecting change. the patient. In general, all participants showed active This could in part be due to that the diary was constructed engagement in both the daily mindfulness and compas- for the purpose of the study, and therefore not formally sionate imagery training as well as in the other CBT ori- tested on its sensitivity to pick up on change, but it could ented exercises such as such observing and writing down also be due to recruitment from a non-clinical population critical thoughts and letting go of safety behaviors. How- with more variable problem levels. In addition, while the ever, an interesting observation was that the degree of threats to internal validity are relatively well addressed homework compliance appeared to covary with variation in within a single case experimental design, a shortcoming of treatment effect. Specifically, participants’ training fre- this design is that the generalizability of the results remains quency of mindfulness and compassion focused imagery unclear. Due to the non-clinical population and the respectively other CBT exercises showed that the partici- restrictions of the design in this pilot study, external pants with the best results (P1, 2, 4 and 5) adhered to validity remains to be established. This being said, the 80–95 % respectively 88–97 % of these homework exer- results show that CFT may be a promising approach for cises while the participants with more questionable results individuals with problems with social anxiety and further adhered to 46 % respectively 67 % (P3) and 82 % studying is justified to replicate and extend these results to respectively 52 % (P6) of the exercises. While, of course, other, preferably clinical, populations and settings. Even- caution should be taken in interpreting the direction of tually, this approach needs testing in RCTs and further these relations, it does point to the potential importance of investigation of processes of change. The next step may be addressing obstacles for homework compliance. to test CFT in a randomized controlled trial, preferably including individuals formally diagnosed with social anx- Limitations iety disorder. Lastly, it should be kept in mind that the treatments were It should be kept in mind that this is a pilot study and it has performed by two different and not so experienced clini- some obvious shortcomings. Due to time constraints the cians. While they had extensive support in a manualized baseline measurements were relatively short and a longer protocol and patient materials as well as clinical supervi- baseline measurement would have been preferable. More- sion, this could also be linked to the different pattern of over, all measurement was conducted with self-report, improvements in patients. increasing the risk for common method variance. Also, no follow up data beyond 2–4 weeks after treatment was Theoretical Discussion available and it is therefore not sure whether results are maintained over time. This is important, not only because The main hypothesis of CFT is that emotional suffering is treatment effects may be lost to follow up, but also because exacerbated and perpetuated by constant self-attacking and the opposite may be the case and treatment effects may be shame and stands in the way of people being able to gen- delayed. For example, changing perspective on one’s uinely sooth and support themselves as well as receive and constant self-attacking cognitions and starting to genuinely seek support from others. It is theorized that self-criticism support oneself and seek out others for comfort and support activates the threat regulation system and that there may require time. It could be that changes on scores of self- may be two separate but interactive positive affect regu- critical thinking improve gradually over time, rather than lating systems (Gilbert et al. 2008). One is focused on abruptly during treatment. doing/achieving and anticipating rewards, while the other Another point to be kept in mind while drawing conclu- focuses on social signals of affiliation and care. Develop- sions is that, while the data give indications of reliable ment of these systems takes place during the developmental improvements with some clinical significance, group studies periods of the individual through by caretakers on the effectiveness of CFT need to establish the magnitude providing warmth, reassurance and soothing and thus 123 96 J Contemp Psychother (2015) 45:89–98 creating experiences and memories of safeness for the decreasing shame and self-criticism. These problems are individual. This enables the individual to understand and not specific to social anxiety disorder and this approach feel safe with their own emotions and be able to respond may therefore be used and tested as a transdiagnostic with self-soothing in response to stress. It is hypothesized treatment approach to address shame and self-criticism in a that soothing and affiliation systems are underdeveloped for wide range of psychological disorders. self-critical individuals. Indeed, evidence suggests that lack of self-compassion is linked to insecure attachment and childhood maltreatment, and may mediate the effect of Appendix 1 Session content description insecure attachment on (Tanaka et al. 2011). Also, evidence shows that in self-critical Session 1 Psycho education on shyness, how the brain individuals, high levels of negative affect are accompanied evolved through evolution and sets us up with sensitivity to by low levels of ‘safe and content’ positive affect (Gilbert social threat, how emotions are regulated in different sys- et al. 2008). Therefore, in CFT the ability to treat oneself tems (threat, soothing and achievement), compassion and and others with compassion is seen as a key regulator of mindfulness. In session, collaborative, case-conceptuali- emotion, not only pivotal to down regulate negative emo- zation to connect emotion regulation systems to the par- tion but also to increase the experience of positive emotions ticipants’ own lives. In session mindfulness soothing and well-being. breathing exercise. Homework: daily practice in soothing Where cognitive behavioral psychotherapies tradition- breathing. ally mainly focus on decrease of symptoms and direct Session 2 Psycho education on shame, self-criticism and change of maladaptive cognitions and behaviors, CFT adds barriers to feeling compassion. Further conceptualization an explicit focus on compassion and integrates and utilizes of the participant’s problems focusing on the threat system, a variety of psychotherapeutic methods that aim to directly coping strategies to regulate anxiety symptoms and com- increase ability to experience self-soothing, warmth and passionate understanding of oneself. Homework: daily communion. The methods used in CFT are not necessarily monitoring with focus on identifying self-criticism, daily ‘new’, but rather integrated within a new theoretical practice in soothing breathing. framework of emotion theory and put in a uniting Session 3 Psycho education on the function of critical humanistic context of compassion for self and others as a thoughts and on imagery. In session experiential exercise point of departure and as a basis for developing adaptation on how imagery can help to create warm, helpful and and well-being. In CFT compassion for self and others has compassionate feelings while negative thoughts create an extended meaning that incorporates a sense of inter- negative feelings. Homework: daily monitoring and connectedness to humanity at large, equanimity, as well as mindfulness of self-critical automatic thoughts, daily hope and meaning in the face of life’s difficulties. It is practice of imagery exercise ‘‘Safe Place’’, daily practice in therewith, like other psychotherapies such as ACT and soothing breathing. DBT, based a Buddhist life philosophy that departs from Session 4 Psycho education on self-validation. In ses- common western conceptualizations of mental health sion training in generating more compassionate thoughts as problems as pathologies, and ‘normalizes’ them as part of an alternative to self-critical thoughts. In session imagery the human experience. exercise on feeling compassion from others. Homework: While effectiveness of CFT is one main question that daily training in generating compassionate thoughts as needs to be further addressed, another question is what alternative to self-critical thoughts, daily practice of mechanisms may be underlying the effect. It is hypothe- imagery exercise ‘‘Receiving compassion from others’’, sized that CFT deactivates the self-perpetuated activation daily practice in soothing breathing. of the threat system in a direct way by decreasing self- Session 5 Psycho education on safety behaviors. In critical thinking and shame. But it is also hypothesized that session training on identification of safety behaviors in an ability to feel safe and secure can be trained and become relation to own shyness and social anxiety. Homework: a buffer, an increased resilience, through activation of self- daily monitoring and challenging of safety behaviors, daily soothing and warmth. Thus, compassion may be hypothe- practice of imagery exercise ‘‘Feeling compassion for sized to regulate emotion through its effect on rumination others’’, daily practice in soothing breathing. (brooding) as well as through its effect on positive emotion. Session 6 Psycho education on how life values can Future studies could include process measures on rumina- motivate and help people to cope with difficult emotions in tion and positive affect to investigate mechanisms of effect. order to reach a long term goals. In session training on In conclusion, this study adds to the as of yet small body mapping important values using a life compass. Introduc- of evidence indicating the effectiveness of compassion tion of guiding principles validation, acceptance, direction focused therapy for increasing self-compassion, and and compassion as an aid in facing difficult situations in 123 J Contemp Psychother (2015) 45:89–98 97 daily life. Homework: exposure to a difficult situation anxiety, depression, and self-other blame. Personality and using guiding principles, daily practice of imagery exercise Individual Differences, 29(4), 757–774. Gilbert, P., & Procter, S. (2006). Compassionate mind training for ‘‘Feeling compassion for oneself and others’’, daily prac- people with high shame and self-criticism: overview and pilot tice in soothing breathing. study of a group therapy approach. Clinical Psychology and Session 7 In session work on how to integrate com- , 13(6), 353–379. passion in one’s life and action and how to use compassion Goss, K. (2011). The Compassionate Mind-Guide to Ending Over- eating: Using Compassion Focused Therapy to overcome to meet difficult situations. Homework: exposure to a dif- Bingeing & Disordered Eating. Oakland: New Harbinger ficult situation using guiding principles, assignment on Publications. writing a compassionate letter to self, daily practice in Gumley, A., Braehler, C., Laithwaite, H., MacBeth, A., & Gilbert, P. soothing breathing. (2010). A Compassion Focused Model of Recovery after Psychosis. International Journal of , 3, Session 8 Collaborative summary of the intervention. In 186–201. session work on making a plan in order to continue to Henderson, L. (2011). The compassionate mind guide to building social evolve and meet difficulties with compassion for self and confidence: Using compassion-focused therapy to overcome others. shyness and social anxiety. Oakland: New Harbinger Publications. Jacobson, N. S., & Truax, P. (1991). Clinical significance: a statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12–19. Judge, L., Cleghorn, A., McEwan, K., & PaulGilbert, P. (2012). An References Exploration of Group- Based Compassion Focused Therapy for a Heterogeneous Range of Clients Presenting to a Community Barnard, L. K., & Curry, J. F. (2011). Self-compassion: conceptual- Mental Health Team. International Journal of Cognitive Ther- izations, correlates & interventions. Review of General Psychol- apy, 5(4), 420–429. ogy, 15(4), 289–303. Kazdin, A. E. (2011). Single-Case Research Designs (2nd ed.). New Cox, B. J., Fleet, C., & Stein, M. B. (2004). Self-criticism and social York: Oxford University Press. phobia in the US national comorbidity survey. Journal of Laithwaite, H., O’Hanlon, M., Collins, P., Doyle, P., Abraham, L., Affective Disorders, 82(2), 227–234. Porter, S., et al. (2009). Recovery After Psychosis (RAP): a Cox, B. J., Walker, J. R., Enns, M. W., & Karpinski, D. C. (2002). Compassion Focused Programme for Individuals Residing in Self-criticism in generalized social phobia and response to High Security Settings. Behavioural and Cognitive Psychother- cognitive-behavioral treatment. Behavior Therapy, 33(4), apy, 37(5), 511–526. 479–491. Longe, O., Maratos, F. A., Gilbert, P., Evans, G., & Volker, F. (2010). Furmark, T., Tillfors, M., Everz, P., Marteinsdottir, I., Gefvert, O., & Having a word with yourself: neural correlates of self-criticism Fredrikson, M. (1999). Social phobia in the general population: and self-reassurance. Neuroimage, 49, 1849–1856. prevalence and sociodemographic profile. Social Psychiatry and Lucre, K. M., & Corten, N. (2013). An exploration of group Psychiatric Epidemiology, 34(8), 416–424. compassion-focused therapy for personality disorder. Psychol- Furmark, T., Tillfors, M., Stattin, H., Ekselius, E., & Fredrikson, M. ogy and Psychotherapy: Theory, Research and Practice, 86(4), (2000). Social phobia subtypes in the general population 387–400. revealed by cluster analysis. Psychological Medicine, 30, Ma, H. (2006). An alternative method for quantitative synthesis of 1335–1344. single-subject researches: percentage of data points exceeding Gale, C., Gilbert, P., Read, N., & Goss, K. (2014). An Evaluation of the median. Behavior Modification, 30(5), 598–617. the Impact of Introducing Compassion Focused Therapy to a Marshall, M. B., Zuroff, D. C., McBride, C., & Bagby, M. (2008). Standard Treatment Programme for People with Eating Disor- Self-criticism predicts differential response to treatment for ders. Clinical Psychology & Psychotherapy, 21(1), 1–12. major depression. Journal of Clinical Psychology, 64(3), Gilbert, P. (1998). What is shame? Some core issues and controver- 231–244. sies. In P. Gilbert & B. Andrews (Eds.), Shame: interpersonal Mattick, R. P., & Clarke, J. C. (1998). Development and validation of behaviour, psychopathology and culture (pp. 3–38). New York: measures of social phobia scrutiny fear and social interaction Oxford University Press. anxiety. Behaviour Research and Therapy, 36(4), 455–470. Gilbert, P. (2000). The relationship of shame, social anxiety and Neff, K. D. (2003). Development and validation of a scale to measure depression: the role of the evaluation of social rank. Clinical self-compassion. Self and Identity, 2(3), 223–250. Psychology & Psychotherapy, 7(3), 174–189. Neff, K. D., Kirkpatrick, K. L., & Rude, S. S. (2007). Self- Gilbert, P. (2010). Compassion Focused Therapy. London: compassion and adaptive psychological functioning. Journal of Routledge. Research in Personality, 41(1), 139–154. Gilbert, P. (2014). The origins and nature of compassion focused Rockliff, H., Gilbert, P., & McEwan, K. (2008). A pilot exploration of therapy. British Journal of Clinical Psychology, 53, 6–41. heart rate variability and salivary cortisol responses to compas- Gilbert, P., McEwan, K., Matos, M., & Rivis, A. (2011). Fears of sion-focused imagery. Clinical Neuropsychiatry, 5(3), 132–139. compassion: development of three self-report measures. Psy- Scruggs, T. E., Mastropieri, M. A., Cook, S. B., & Escobar, C. (1986). chology and Psychotherapy: Theory Research and Practice, 84, Early interventions for children with conduct disorders: a 239–255. quantitative synthesis of single-subject research. Behavioral Gilbert, P., McEwan, A., Mitr, R., Franks, L., Richter, A., & Rockliff, Disorders, 11, 260–271. H. (2008). Feeling safe and content: a specific affect regulation Singer, T. & Botz, M. (Eds.) (2013). Compassion. Bridging Science system? Relationships to depression, anxiety, stress and self- and Practice. Saarbrucken, Germany: Satzweiss Print Web criticism, The Journal of Positive Psychology, 3(3), 182–191. Software GmbH. ISBN: 978-3-9815612-0-3 Gilbert, P., & Miles, J. N. V. (2000). Sensitivity to Social Put-Down: Svanborg, P., & A˚ sberg, M. (1994). A new self-rating scale for it’s relationship to perceptions of social rank, shame, social depression and anxiety states based on the Comprehensive

123 98 J Contemp Psychother (2015) 45:89–98

Psychopathological Rating Scale. Acta Psychiatrica Scandina- Werner, K., Jazaieri, H., Goldin, P., Ziv, M., Heimberg, R., & Gross, via, 89, 21–28. J. J. (2012). Self compassion and social anxiety disorder. Tanaka, M., Wekerle, C., Schmuck, M. L., & Paglia-Boak, A. (2011). Anxiety, Stress & Coping: An International Journal, 25(5), The linkages among childhood maltreatment, adolescent mental 543–558. health, and self-compassion in child welfare adolescents. Child Abuse & , 35(10), 887–898. Tirch, D. (2012). The Compassionate-Mind Guide to Overcoming Anxiety: Using Compassion-Focused Therapy to Calm , , and Fear. Oakland: New Harbinger Publications.

123