Development of Multidimensional Model of Change and Emotional Safeness Therapy – the backstage view

Posted by:

Photo by Hans-Peter Gauster

In this article I’m presenting the organizational background behind the development of Multidimensional Model of Change (MMC) and Emotional Safeness Therapy (EST), before discussing their content in the next paper.

Multidimensional Model of Change was initially thought as a metatheoretical framework for the integration of Acceptance and Commitment Therapy (ACT) and Compassion Focused Therapy (CFT). Originally, the model was designed for the treatment of mood disorders with co-occurring personality problems, but its applicability has been subsequently extended to other clinical areas. The model has already been applied in the development of a modular group treatment for recurrent depression, called Emotional Safeness Therapy (EST), successfully implemented and run at the Akershus University Hospital in 2015-2018.


The Multidimensional Model of Change and Emotional Safeness Therapy are results of six years’ work of several people. The conceptual work on the model was kicked off in 2012 during the ACBS World Conference in Washington DC. The clinical part of the work started three years later, in 2015, at the Akershus University Hospital in Norway.

The development of the model wouldn’t be possible without the involvement of those whom it was meant to serve. Thus, I express my gratitude to the patients of the District Psychiatry Center in Jessheim who participated in our prototype EST groups and whose continuous feedback shaped our work on the model. It wouldn’t be an exaggeration to say that MMC and EST were co-created by patients.

I also express my gratitude to members of my EST research team at the Akershus University Hospital, who often voluntarily devoted their spare time to progress the work. Hereby I thank psychiatrist Jan Anders Borning and clinical psychologists: Eline Wøien Hansen, Karoline Halvorsen, Charlotte Kristine Krogh, Mari Morell, Kim Dahl Skreslett and Sjur Kvadsheim.

My work was heavily influenced by (many) years of collaboration with the former president of the Association for Contextual and Behavioral Science and Director of the Portland Psychotherapy Center, Jason Luoma, PhD, and his work on the use of ACT with shame-prone clients.

I also thank my co-members of the Executive Board of the ACBS Compassion-Focused SIG who contributed to this work by inspiration and intellectual challenge: Dennis Tirch, PhD; Laura Silberstein-Tirch, PhD; Russell Kolts, PhD and Melissa Platt, PhD. For years we have shared a common goal: making science of compassion an integral part of the CBS family.


The work on the Multidimensional Model of Change has been inspired by the fact that numerous depressed clients demonstrate high levels of self-criticism, shame and self-contempt during therapy, and that this subpopulation seems to benefit from a slightly different therapeutic approach than traditional ACT or CBT. A highly self-critical client may use any therapeutic situation, such as seemingly neutral psychoeducation or the failure to do homework, against herself to fuel shame, self-blame and self-disgust. Such self-defeating attitudes not only impede recovery from the current depressive episode and hamper the ongoing therapeutic process but tend to be stable also through symptom-free periods, contributing to the client’s vulnerability and the risk of relapse. Moreover, self-contempt seems to be strongly associated with relational problems and a generally poor quality of interpersonal functioning, which in turn diminishes social support – an important contributor to recovery and prevention.

In order to address clients’ self-harshness, many ACT therapists already include compassion-based strategies into treatment of highly self-critical and self-defeating clients. Yet, they do it without having any conceptual guidance on how to integrate such strategies with traditional ACT.

It is a common clinical practice to eclectically combine strategies originating from different theoretical traditions into a single intervention. It’s OK when it works for the client. Existing research suggests, for example, that self-compassion may be a potential agent of change in Acceptance and Commitment Therapy. But there is also a real risk that ACT and CFT strategies may undermine each other if used in an incoherent way.
Compassion-based strategies are aimed at the transformation of emotional experience. The question is: Are they compatible with the acceptance of emotions promoted by ACT, and if so – in which contexts?

At the beginning of a session the therapist may for example encourage the client to “stay with difficult feelings”, and ten minutes later ask her “How does it feel when you look at yourself having these difficult feelings?” – a move that instantly changes and transforms the whole experience. Isn’t it a bit confusing for the client? What is actually the direction of this session: shaping the willingness to experience difficult emotions or enhancing the ability to get rid of them?

Yeah, I know you may have heard people saying that “anything that works is ACT”. Well, not exactly. Efficacy is a rather poor criterion for distinguishing one treatment from another. Taking a look at the theoretical underpinnings might be a more fruitful way to see the demarcation lines. Many therapists deliver treatments that are quite distant from ACT still calling them “ACT”. It just means that there’s sometimes a difference between what people do and what they name it.

When we look at ACT, the concept of experiential avoidance is to be found at its heart. Furthermore, we find the acceptance of experience not only in the name of the treatment but also among the core therapeutic processes. Yes, it’s about the acceptance of experience, not transformation. ACT changes clients’ problematic and harmful responses to difficult emotions, not the emotional experience itself, at least not primarily. The change in ACT is supposed to occur on the behavioral level in the first place, with the transformation of emotional experience as a possible but not required outcome.

The core therapeutic strategies of Compassion Focused Therapy go in an opposite direction as this approach is based on the evolutionary model of emotion regulation. CFT focuses explicitly on the transformation of emotions in the first place, with behavioral change as a consequence. This therapy is aimed at eliciting very specific emotional experiences and sets them up as an operational therapeutic goal and the mediator of behavioral change. ACT interventions are never aimed at eliciting any specific experience, as it would be viewed as the opposite of experiential openness.

So, eclectically throwing strategies from those two approaches into one session is like putting your legs in two different trains, with the hope that they will go in the same direction. The timetable says they won’t. The MMC model was created to prevent the split. Yes, I personally think that ACT and CFT can be integrated and can complement each other boosting the intervention, otherwise the work on the model wouldn’t make any sense. But the word “integration” does not mean “simply blending”.

In what way may MMC be useful for you as a clinician?

Well, imagine a client, Mary, who was referred to you because of problematic anger. The day before the session Mary had an important and difficult exam at school. She planned to sleep a bit longer on that day in order to be in a better shape during the exam. Unfortunately, her new mobile phone woke her up at 6 a.m. because of default alarm settings. (1) She got angry at the phone and threw it against a cupboard. Later, on her way to school, she wanted to call her mum, but discovered that the phone was broken. (2) At that point she got mad at herself for destroying the phone. (3) Next time she got angry when her pen stopped writing during the test and she lost precious time trying to borrow another one. (4) Right after the exam she felt angry at herself, because she worked really hard to prepare for the exam, but she chose wrong tasks. (5) On the way home, she got mad at two guys who harassed her sexually and deliberately touched her bottom on a bus. (6) Later that day after dinner, she felt intense anger while thinking how her sister persuaded her to choose her current school, although she never felt interested in it.

The six instances of Mary’s anger should be approached by the therapist in totally different ways. However, ACT does not provide therapists with conceptual lenses to see the functional difference between those types of anger at the first glance. How would you choose your therapeutic moves with Mary? What would you target: cognitive antecedents, emotional reactions or maladaptive responses to those reactions? And how to tell adaptive from maladaptive, without looking at each of the situations in the context of personal values and specific behavioral goals?

There is a huge difference between getting mad at someone who harasses you sexually and getting mad at yourself after failing a test. The difference is obvious to any seasoned clinician, but what exactly makes both situations so different at the conceptual level? Anyway, not values.

The MMC model sorts the client’s seemingly identical emotional reactions into four distinct dimensions of psychosocial functioning, which helps to easily see the difference between e.g. your anger at a mobile phone and anger at yourself. So, if you wear the MMC glasses while working with Mary, the choice of your therapeutic responses for different types of anger would be easier.

The Multidimensional Model of Change is not a new treatment strategy. It just provides clinicians with a conceptual framework for a coherent integration of empirically proven but theoretically distant and potentially incompatible treatment strategies. The model was meant to help clinicians avoid being simultaneously driven in opposite directions and undermining own work by the lack of coherence. The model should be viewed purely as a pragmatic clinical tool and not as a distinct theory. It is based solely on preexisting basic and clinical research, and on well-established scientific and conceptual work of others.

MMC and EST are strongly rooted in clinical practice and have been built bottom-up, starting from the exploration of clients’ goals and expectations for therapy, to analysis of the content of established treatments and assessment tools. This work still continues, and you may also contribute to the further development of the model. Contact me if you’re interested in collaboration.

EST Project

The empirical part of the EST project started in 2015 at the District Psychiatry Center in Jessheim, which is an organizational part of the Akershus University Hospital. It was preceded by conceptual work.
In 2015 the Center decided to launch a new treatment modality for depressed patients and a team of clinicians (see: Acknowledgments) led by me took up the task. The team consisted of licensed clinical psychologists and a psychiatrist.

The project was planned to contain two phases: (1) development and implementation of a new group treatment for depression and (2) a randomized controlled trial of its effectiveness. The project has been approved as a clinical trial by the Regional Committee for Medical and Health Research Ethics.

During the first phase of the project our team ran eight prototype EST groups. The duration of the groups varied from 8 to 12 weeks, as different variants were tried out in order to establish the optimal length. The first group started in December 2015 and the last one was concluded in January 2018. During this period our team delivered a total of 62 group sessions, each with the duration of 120 min plus a 15 min mid-session coffee break. Thus, we have delivered 124 hours of thoroughly monitored treatment.

The first two groups were run under the working name “ACT-2” or “ACT Plus”, as their content heavily relied on the Hexaflex model (ACT) with the self-compassion component added to different sessions. The working name was change to Emotional Safeness Therapy after modularization of treatment. If branding is the key future of treatments for you, you can still call it ACT because it is derived from and built upon ACT.

All groups were led by me and one or two co-therapists. Sessions were usually followed by a short (20-30 min) debriefing with the co-therapists. Additionally, the whole team had regular 2,5-hour meetings every week to discuss the situation in groups and the ongoing project. Work on the translation of materials and assessment tools into Norwegian, as well as on designing handouts and exercises was done in the spare time.

The goal of the first phase of the project was to test feasibility, safety and acceptability of the treatment before moving over to randomization (phase 2). Thus, all participants were continually and carefully monitored with clinical risk assessment tools, symptom measures of depression (BDI-II) and anxiety (BAI), and general measures of mental health (OQ-45).

In the first four groups each session was followed by an anonymous qualitative and quantitative survey on the perceived usefulness and feasibility of the current session and the treatment as whole, together with an evaluation of group leaders. The ongoing feedback was discussed during team meetings and swiftly responded to during the next session. As a result, the content of treatment, including techniques and materials, was modified and changed from session to session, often under high time pressure. In the last four groups the evaluation was collected only during the last session.

Each group consisted of 7-10 adult participants referred by other therapists within the same unit. Our team did not do diagnostic pre-assessments of the candidates as this was the role of the referring clinicians. The minimal treatment acceptability criterion was set up at 20 percent of drop-out, which was quite an ambitious goal taking into account that all participants had individual therapy available as an alternative option. Not showing up to more than two sessions during the whole treatment was considered as a drop-out, regardless the reason for absence.

The second phase of the project started in November 2017 with a plan to randomly allocate patients to EST and to standard treatment. Right after the recruitment started, due to organizational issues, the District Psychiatric Center in Jessheim withdrew its former consent to randomization, leaving us only with the option to continue the project without the control condition.
At that point, our team decided that over two years of delivering the treatment had proven Emotional Safeness Therapy to be safe, feasible and well-accepted by patients, but of unknown clinical efficacy. Thus, further testing of the treatment without a control group and in exactly the same clinical context was concluded unnecessary. We decided to terminate EST groups at the Center and move on to extending the applicability of the Multidimensional Model of Change to other clinical settings and areas, including internet-based treatment programs and self-help groups.

Multidimensional Model of Change

The overview of the model itself is presented in a separate article (coming soon), so If you want to learn more, just stay attuned.
If you have any questions about the possibility of implementing MMC/EST at your place, training or research collaboration feel free to contact me.


  • Gilbert, P. (2010). Compassion focused therapy: Distinctive features. Routledge.
  • Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (2011). Acceptance and commitment therapy: The process and practice of mindful change. Guilford Press.
  • Luoma, J. B., & Platt, M. G. (2015). Shame, self-criticism, self-stigma, and compassion in Acceptance and Commitment Therapy. Current Opinion in Psychology, 2, 97-101.
  • Malicki, S., & Ostaszewski, P. (2014). Experiential avoidance as a functional dimension of a transdiagnostic approach to psychopathology. Postępy Psychiatrii i Neurologii, 23(2), 61-71.
  • Norcross, J. C., & Goldfried, M. R. (Eds.). (2005). Handbook of psychotherapy integration. Oxford University Press.

CC BY-NC-ND 4.0 This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.


About the Author:

licensed clinical psychologist, psychotherapist, researcher, peer-reviewed ACT trainer
  Related Posts
  • No related posts found.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.