Edited by: Lorys Castelli, University of Turin, Italy
Reviewed by: Claudia Cormio, National Cancer Research Institute Giovanni Paolo II, Italy; Gabriele Roberto Cassullo, University of Turin, Italy
*Correspondence: Susan G. Simpson, Division of Education, Arts and Social Sciences, School of Psychology, Social Work and Social Policy, University of South Australia, Magill Campus, GPO Box 2471, Adelaide, SA 5001, Australia e-mail:
This article was submitted to Psychology for Clinical Settings, a section of the journal Frontiers in Psychology.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Schema Therapy has shown promising results for personality disorders but there is a limited evidence base for group schema therapy (ST-g) with mixed personality disorders. The aim of this study was to explore the feasibility, acceptability, and preliminary effectiveness of ST-g in a sample of eight participants with mixed personality disorders (with a predominant diagnosis of avoidant personality disorder) and high levels of comorbidity. Treatment was comprised of 20 sessions which included cognitive, behavioral, and experiential techniques. Specific schema-based strategies were chosen for a diagnostically mixed group of personality disorder clients. Six participants attended until end of treatment and two dropped-out before mid-treatment. All outcome measures showed changes with large effect sizes in avoidant personality disorder symptom severity, depression and anxiety levels between pre-therapy and follow-up. Four participants achieved a loss of personality disorder diagnosis at the end of therapy. By follow-up, five participants had achieved a loss of diagnosis, suggesting that participants derived ongoing benefits from the group even after treatment ended. Six participants no longer met criteria for depression at the end of treatment and this was maintained for all participants at 6-month follow-up. At follow-up, the majority of participants showed clinically significant change on the Global Symptom Index (GSI). For the Schema Mode Inventory (SMI) maladaptive modes, the majority of participants showed improvement at follow-up. At follow-up, 40% of participants showed clinically significant change on the SMI adaptive modes. Qualitative feedback indicates that the group helps to normalize participants' psychological experiences and difficulties and promotes self-expression and self-disclosure, while reducing inhibition. Preliminary results suggest that short-term ST-g may benefit those with mixed personality disorders, but generalizability is limited by the small sample size and lack of control group.
Personality disorders are highly prevalent in clinical settings. Approximately one-third of clients in outpatient clinical settings are diagnosed with a personality disorder (Zimmerman et al.,
Schema Therapy (ST) is one of the third wave innovative therapies that have developed specifically for treating personality disorders and other complex, chronic clinical presentations. ST has proven to be an efficacious and cost effective treatment for Borderline Personality Disorder (BPD) (Giesen-Bloo et al.,
The emergence of group therapy protocols has been an important development in the growth of ST. It is thought that group therapy provides important curative factors, including corrective emotional learning experiences, a unique opportunity for clients to practice new behavioral and coping skills in a de-shaming environment, and opportunities for vicarious learning (Farrell et al.,
Preliminary evidence supports the use of ST in group treatment for BPD; however, evidence supporting the use of group schema therapy with patients with other personality disorders is sparse. This paper described a pilot study using short-term group schema therapy (ST-g) in a case series of eight patients with Cluster A, B, and C personality disorders and high levels of comorbidity.
The present study utilized a single group pre- and post-pilot study in order to investigate the outcome of ST-g for a group of participants with mixed personality disorders in an outpatient university clinic. The aim of this study was to explore the feasibility, acceptability, and preliminary effectiveness of ST-g. This pilot study is a requisite initial step in informing the feasibility of a larger scale study investigating the efficacy of ST-g in patients with mixed personality disorders.
Informed consent was obtained from all patients and ethics approval was granted. Patients were referred to the group from the University of South Australia psychology clinic, private psychologists, and non-government organizations. Patients were included if they met criteria for at least one personality disorder as assessed following the DSM-IV TR (American Psychiatric Association,
1 | Avoidant | Major depression, anxiety | 3 | 28 | Student | Honors degree | Single | 19 |
2 | Avoidant | Major depression, anxiety | 4 | 47 | Unemployed | Secondary school | Single | 16 |
3 | Avoidant, schizoid and dependent | Major depression, anxiety | 3 | 37 | Unemployed | Secondary school | Single | 19 |
4 | Avoidant | Major depression, anxiety | 1.5 | 25 | Student | Honors degree | Single | 20 |
5 | Avoidant | Major depression, anxiety, somatization | 1 | 29 | Unemployed | Diploma | De facto relationship | 20 |
6 | Avoidant | Major depression, anxiety | 2 | 42 | Unemployed | Year 9, secondary school | Widowed | 17 |
7 | Borderline | Major depression, anxiety | 3 | 27 | Hospitality | Completing bachelor degree | Single | 1 |
8 | Borderline | Major depression | 2 | 35 | Health services | Bachelor degree | Relationship | 8 |
The treatment was adapted from van Vreeswijk and Broersen's (
Our adapted model had a strong focus on experiential techniques and mode work for a diagnostically mixed group of personality disorder patients (with a predominant diagnosis of Avoidant personality disorder). ST-g consisted of twenty, 60-min sessions that were run weekly (for 5 months). It was a closed therapy group with provision of five individual sessions during the course of therapy. Sessions were recorded and participants who missed sessions were required to watch the recording before the next session. All sessions were delivered by the same two therapists. Group therapists received one hour of weekly supervision from a doctoral level Clinical Psychologist and accredited schema therapy trainer and supervisor. Consultation was also provided by the fourth author (van Vreeswijk).
The reparenting approach allowed participants to develop a sense of interconnectedness and safety in the initial stages, and in the later stages therapists encouraged participants to develop independence, autonomy, and healthy assertiveness skills in order to meet their emotional needs. Therapists helped participants to develop skills in recognizing their own needs better and to find healthier ways of getting these needs met. In the later stage of treatment, the focus was on motivating and encouraging behavioral change in the present. The “re-familying” component involves the group being encouraged to function as a family where the therapists become a parent figure by reparenting the client, and other group members adopt supportive sibling roles (Farrell and Shaw,
We selected specific schema mode focused techniques on the basis of predominant schema modes. Treatment strategies focused on reducing highly avoidant coping mechanisms (i.e., Avoidant/Detached Protector) by using experiential and physical movement exercises to bypass coping. Participants were encouraged to recognize and label the Detached Protector mode when it was active during sessions and when talking about avoidant behavior that had taken place between sessions. To challenge excessively high standards and self-criticism (Demanding Parent), group chair work exercises and group role plays were used. Limited reparenting and “re-familying” exercises were used with child modes (e.g., group imagery). Due to a lack of emotional awareness and emotional tolerance within the group there was an additional focus on increasing awareness, tolerance, and expression of emotions (i.e., Vulnerable Child mode, Healthy Adult mode) e.g., emotion-focused and acceptance exercises. To further increase awareness of schemas and modes and to facilitate emotion regulation, schema-focused mindfulness exercises and regular “mode check-points” (a brief mode awareness exercise) were also incorporated (Kristeller et al.,
Both group therapists had more than 12 months of experience in providing ST to clients with personality disorders and other complex difficulties under the supervision of an accredited schema therapy trainer and supervisor, who regularly observed their psychological therapy skills and checked for treatment fidelity. Both group therapists had attended over 6 days of training in ST.
The Millon Clinical Multiaxial Inventory (MCMI-III; Millon et al.,
The Young Schema Questionnaire: Short form, second version (YSQ-S2) (Young,
The Schema Mode Inventory (SMI; Young et al.,
The Symptom Checklist 90-R (SCL-90-R; Derogatis et al.,
The YSQ-S2, SMI, and the SCL-90-R were administered at pre-treatment, at mid-treatment (session 10), at end of treatment, and at 6-month follow-up. The MCMI-III was only administered at pre-treatment, at the end of treatment, and at 6-month follow-up. Clinical interview was conducted at pre-therapy and at the end of therapy to assess for personality disorder criteria. To gain an understanding of participants' qualitative experience of participating in ST-g, a focus group was conducted at the end of treatment.
The focus group provided participants with an opportunity to discuss relevant aspects of their experience participating in the ST-g. The focus group was facilitated by the project supervisor. The focus group session was audio and video recorded.
Following visual inspection of the data, a repeated measures analysis of variance (ANOVA) was conducted on all questionnaire measures across the four trial periods (pre, mid, post, and follow-up), and the results of the test are shown in Table
Two participants dropped out of therapy. The first participant dropped out of the group at session three. Feedback provided by this participant suggested that this was due to high levels of shame related to returning to the group after missing a session, and difficulty tolerating the distress associated with working on schemas and maladaptive coping strategies whilst moving house and working full-time. The second participant dropped out at session 16. The participant indicated that this was due to strong feelings of guilt associated with having missed multiple group sessions which triggered a sense of shame and self-criticism for “not putting in enough effort.”
The participants who dropped out reported significantly fewer avoidant personality disorder symptoms at the start of treatment compared with those who completed group therapy:
Visual inspection of the data reveals that at the end of treatment, four clients no longer met criteria for avoidant personality disorder based on scores on the MCMI-III (Millon et al.,
Visual inspection of the SMI group mean scores indicates a trend for the Detached Protector, Compliant Surrenderer and Vulnerable Child to reduce across treatment with a slight increase in the Vulnerable Child from post to follow-up (Figure
First we determined Pearson correlations among scores of the GSI, YSQ-S2 (total score), SMI adaptive modes and SMI maladaptive modes before therapy. As can be seen in Table
1. GSI | |||||||
2. YSQ-S2 | 0.60 | ||||||
3. SMI maladaptive | 0.82 |
0.69 | |||||
4. SMI adaptive | −0.68 | −0.90 |
−0.7 | ||||
5. Depression | −0.06 | 0.31 | 0.08 | 0.13 | |||
6. Anxiety | −0.13 | −0.60 | −0.5 | 0.27 | −0.81 | ||
7. Avoidant | 0.77 | 0.63 | 0.90 |
−0.63 | 0.22 | −0.60 |
GSI | 53 (6.54) | 49.83 (5.71) | 46 (6.72) | 45.50 (6.57) | 6.37, |
YSQ-S2 | 52.87 (5.84) | 47.5 (6.37) | 40.63 (5.26) | 37.90 (5.23) | 27.43, |
SMI maladaptive | 36.68 (4.15) | 33.08 (1.89) | 30.46 (3.15) | 30.74 (2.87) | 8.49, |
SMI adaptive | 6.02 (0.91) | 6.88 (1.23) | 7.35 (1.10) | 7.27 (1.12) | 3.56, |
Depression | 89.50 (3.56) | 60.5 (13.63) | 53.33 (20.67) | 19.86, |
|
Anxiety | 94.17 (6.62) | 74 (36.89) | 73 (36.11) | 2.23, |
|
Avoidant | 107.5 (8.43) | 77.67 (11.52) | 77 (11.22) | 36.33, |
As the study aimed to investigate change in the maladaptive schemas of the participants across treatment phases, the results of the repeated measures ANOVA for the YSQ-S2 indicated a significant time effect:
Effect size (
Pre to post | 2.20 | 2.96 | 0.76 | 2.91 | 1.06 | 1.32 | 1.69 |
Pre to follow-up | 2.70 | 3.07 | 0.82 | 2.44 | 1.14 | 1.22 | 1.66 |
Post to follow-up | 0.52 | 0.06 | 0.03 | 0.41 | 0.08 | (0.07) | (0.09) |
Reliable change and clinical significance were calculated for the SMI and the GSI of the SCL-90-R. For the SMI adaptive modes, from pre-therapy to post-therapy, two (33.33%) participants recovered, one (16.67%) improved, and three (50%) remained the same. For the SMI maladaptive modes, from pre-therapy to post-therapy, one (16.67%) participant recovered, and five (83.33%) improved. For the GSI, from pre-therapy to post-therapy, two (33.33%) participants recovered and four (66.67%) improved. By follow-up, three (50%) participants recovered, two (33.33%) improved and one (16.67%) remained unchanged (one participant improved from pre-therapy to post-therapy, but due to an increase in scores from post-therapy to follow-up was considered unchanged overall from pre-therapy to follow-up).
The video recorded data (of the focus group) was viewed a number of times and themes were identified and agreed upon by all authors. Themes were selected on the basis of frequency. Illustrative verbatim quotes are shown with the relevant themes below. Data analysis revealed four overarching themes: (1) the normalizing effect of ST-g, (2) ST-g actively challenges schemas, (3) disinhibition effect of ST-g, and (4) motivational influence of ST-g.
Nearly all participants reported that ST-g normalized their schemas and associated emotional experiences. For instance:
“Everyone has schemas.”
“When I see that I can be compassionate toward others who have schemas, I then start to challenge my own.”
Participants explained that ST-g increases understanding of self through observation of others and allows direct experience in practicing challenging schemas just by attending. Some verbatim quotes that illustrate this are the following:
One participant with the social isolation schema reported “When I noticed others sharing and expressing themselves, it helped me feel safe enough to participate. Seeing others with similar schemas helps me feel I am similar to others.”
Another participant with an entitlement schema reported “I notice others' needs more.”
A participant with the self-sacrifice schema said “I began to think about what I need from other group members.”
Another theme frequently endorsed by participants was that the group actually encourages and evokes self-expression. For example:
“Seeing others being vulnerable and not judging each other, allowed me to feel safe to express my feelings and needs.”
A majority of participants noted that ST-g motivated them to make behavioral changes. For instance:
“Seeing others make progress has more impact on me (compared to individual therapy)”
“others' progress spurs me on.”
“There is a lot of emphasis on the steps needed to make (behavioral) changes and build my healthy self.”
This study aimed to investigate the feasibility, acceptability and preliminary efficacy of ST-g in a sample of eight participants with mixed personality disorders and comorbidity. A large effect size was found for the group between pre-therapy and follow-up for all measures. All outcome measures showed changes with large effect sizes in avoidant personality disorder symptom severity, and depression and anxiety levels between pre-therapy and follow-up. Four participants achieved a loss of personality disorder diagnosis at the end of therapy. By follow-up, five participants had achieved a loss of diagnosis, suggesting that participants derived ongoing benefits from the group even after treatment ended. Six participants no longer met criteria for depression at the end of treatment and this was maintained for all participants at 6-month follow-up. Clinically significant change on the GSI at follow-up showed that a majority of participants had recovered. For the SMI maladaptive modes, the majority of participants showed improvement at follow-up. Clinically significant change for the SMI adaptive modes at follow-up showed that 40% of the participants had recovered.
One of the main aims of ST is to help patients to reduce maladaptive coping modes so that they can begin to find healthy ways of enabling their emotional needs to be met, both by others and themselves (Young et al.,
In this study two participants dropped out of therapy. Comparison with attrition rates in other group studies suggests that this is may be a relatively low attrition rate and is comparable with similar studies using ST (Davis et al.,
The clinical improvements demonstrated in this study indicate that ST delivered in a group setting may hold promise for participants with different personality disorders and high levels of comorbidity. It has been hypothesized that there may be specific factors operating in a group setting that challenge schemas at a group-process level (Simpson et al.,
The group also motivated participants to make behavioral changes that they had been unable to make during individual therapy. Participants felt motivated by directly observing other participants' progress. Some of the themes identified through the focus group have been previously identified in the literature (e.g., Farrell et al.,
Limitations of this study include a small sample size and a lack of a longer follow-up period. Moreover, the fact that a mixed personality disorder group was used reduced our number of participants further. If our study had focused on a single diagnostic group, this may have allowed greater standardization of assessments and facilitated the use of specific cut-off points for inclusion in the study, thereby improving the methodological robustness of the study. However, this may also have reduced the generalizability of this study to general clinical populations seen within health and hospital-based clinics that are often composed of heterogeneous client groups. In addition, due to the fact that this was a small pilot study that was conducted to determine the feasibility and efficacy of short-term ST-g with a mixed personality disorder population, a comparative control group was not included. As such, it is not possible to determine from these findings the extent to which the results are due to participants' response to therapy in general or to ST-g specifically or to other variables. Although DSM-IV criteria were utilized when interviewing and diagnosing psychiatric disorders, a formal diagnostic interview such as the SCID-I and II would have been a more robust means of assessment at pre-treatment. Additionally, in this study, we did not set a predetermined MCMI-III cut off point for inclusion. Finally, although treatment fidelity was rated by accredited schema therapists regularly throughout the study, no formal ST-adherence rating was done. Future studies in this area could improve on our procedure by assigning random tapes to raters not involved in the study. If it was a large trial, we would complete intention-to-treat analyses by including participants who dropped-out in all analyses.
This naturalistic pilot study allows greater exploration of the level of change possible over a 20 week schema therapy group with a mixed personality disorder sample. Naturalistic designs are clinically useful and have high ecological validity, which can allow results to be generalized to patients generally seen in typical clinical settings (i.e., community mental health teams, hospital wards) (Lincoln and Guba,
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.