Abstract
In this article we present a video-based explorative study of three consecutive mentalization-based group therapy (MBT-G) sessions at a Norwegian Mental Health Centre. MBT-G has its roots in group analytic psychotherapy, but is modified to avoid the chaotic and destructive processes often encountered in groups consisting of severely personality disordered patients. Among the most notable differences from the Foulkesian group analytic approach are the therapists handling of authority and a ‘not-knowing stance’ (as opposed to an interpretative stance). Contrary to our expectations we found the typical chaotic borderline group processes that MBT-G was designed to avoid. An in-depth examination of the data, employing qualitative Thematic Analysis, revealed that the therapists failed to establish themselves as authoritative leaders of the group and misconstrued the ‘not-knowing stance’. We present transcripts of some typical sequences and discuss their clinical and therapeutic implications.
Keywords
Introduction
In a companion article we have outlined the theoretical rationale for mentalization-based group therapy (MBT-G) for borderline patients (Karterud, 2015). In this article, we present the results of a qualitative study of a MBT group at a Norwegian Mental Health Centre (MHC). The aim of the study was to gain more knowledge about what kind of problems therapists would encounter when conducting MBT groups in an ordinary MHC, as part of the general public mental health services, without any connection to academic or university settings.
Material
The object for the study was a slow-open group that met 1.5 hours every week at a Mental Health Centre in Norway, as part of an established MBT programme. The choice of this particular group was based on convenience; it was part of the only MBT programme that existed in the same region as the university site of the first author. As such, we do not claim it to be representative for MBT groups in general.
The group consisted of six female patients (age range 21–46) with borderline-type problems (being assessed by SCID-II). Thus, all patients had problems associated with emotion regulation, interpersonal relations and unstable sense of self. All had severely compromised social functioning, e.g. being unable to work or study. Four of the patients had been members of the group for a year, while the last two had attended the group for 10 and four sessions prior to data collection. All patients had supplementary individual therapy.
The group was conducted by two male therapists of psychodynamic orientation. One had been trained in group analysis for three years, supplemented with a two year course in MBT-G. The other had followed a one year course in MBT-G. The group therapists met with the individual therapists at team meetings every second week. The treatment team received external expert supervision about four times a year.
The researchers had no direct contact with the group and knew none of the patients in advance. The therapists informed the group about the research project, and patients read and signed an informed consent form before participating in the study. The therapists had previously video-recorded their sessions for supervisory purposes, so the research project did not represent any structural deviation from existing routines. The research was approved by the Regional Health Research Ethical Committee.
Methods
While designing the study we decided to go beyond the question of whether the therapists adhered to the MBT manual or not (Karterud, 2012). We wanted to be open for hitherto unknown phenomena and therefore advanced no propositions as to what we might find. We based our employment of Thematic Analysis (TA) on the accounts of Howitt and Cramer (2007) and Braun and Clarke (2006).
TA is basically a method for pattern finding. Themes derived from a TA organize and describe data. As a consequence of our theoretical framing, this study called for a detailed analysis of some aspects of the data (i.e. therapists interventions based on MBT principles), at expense of a rich description of the data as a whole.
Three subsequent therapy sessions, summing up to 4.5 hours of video data, were transcribed verbatim into about 59,000 words. All patients were given fictional names, and all content that could compromise their anonymity were altered.
As familiarization with data is vitally important in TA, the first author transcribed the data himself. The analysing process moved through transcription, repeated read-throughs of the transcripts, several meetings and discussions with supervisors, a feedback session with the therapists of the group in question, and ultimately ended while the first draft of the report was written (Inderhaug, 2013).
A limited number of themes which adequately reflected the textual data were identified through an analytic process of coding and integration of codes into themes. In the coding phase, the data was read over and over again, coding small segments with intuitive verbal descriptors. The codes were modified from one reading to the next, in the light of the developing apprehension of the totality of the data. In turn, themes that integrated substantial sets of the codes were formulated, and at last, the themes were compared to the items of the MBT-G Adherence and Competence scale (Karterud, 2012). The analytic process progressed from a descriptive phase, where the data was organized to show patterns without considering theoretical concerns, to an interpretative phase, where MBT theory served as an organizing principle for formulation of themes. In writing the present article, we shortened and sharpened the results by conflating some of the themes presented in the initial report (Inderhaug, 2013).
The authors held a meeting with the therapists after having landed on some core themes in the analysis. When we had provided them with our feedback, the meeting took the form of a reflective dialogue where the therapists could help us adjust, develop, or confirm our understanding of the data.
Results
Early on, it was apparent that the therapists met substantial challenges in dealing with this group. Borderline pathology suffused the discourse and the group processes, evident both in what patients told about their life outside therapy and in the way they acted in the group. The content of the stories was often highly disturbing, describing chaotic family lives, substance abuse, violence, sexual transgressions and destructive acting out. The main therapeutic challenge appeared to be taking the reins in a group bordering on chaos. However, the therapists did not stand out as authoritative leaders. Instead, some patients had taken central positions in the group and dominated the discourse and the process.
The following four themes from the Thematic Analysis addressed the bulk of the therapists’ challenges: (1) structuring the process; (2) interruptions/getting a word in; (3) determining importance; and (4) promoting pseudomentalization. Comparing these challenges to the MBT-G adherence and competence scale, we found that they related conceptually to two MBT-G principles. Theme one and two could be related to managing authority, and theme three and four to exploration, curiosity and not-knowing stance. The therapists seemed to have downplayed their role as experts on psychotherapy and group dynamics. On a phenomenological level, the tendencies of the therapists abdicating from the leadership and holding a mere supporting role in the group could be associated with the not-knowing stance in MBT. In the following we will elucidate the themes and related principles, starting with managing authority.
Before text examples are presented, it should be mentioned that one of the therapists (T1) was a recurrent character in the chosen transactions. T1 was more involved in relevant transactions, while the other (T2) adhered less to MBT principles. Furthermore, a few patients were clearly dominant, both in terms of shaping the group dynamics and in mere verbal activity. The tendencies we want to highlight here stand out most clearly in the transactions involving these patients, and necessarily, they are overrepresented in the text examples. As space limitations do not allow us to present longer transactions, we jump straight into complex and at times chaotic transactions. The reader may find it difficult to grasp what the group is talking about. In fact we were tempted to correct the transcript in order to make it more comprehensive for the reader. However, this would distort the discourse as it actually took place. The discourse was often chaotic and difficult to understand. Still, the main points we want to make concerns the process more than the content of the transactions, and we believe the relevant processes are well illustrated in the text examples.
In MBT-G, the therapists should take a firm stance by expressing in simple words the tasks and goals of the group (e.g. what is expected from the group members, why the therapists will work in certain manners, etc.). The challenges associated with managing authority in this group will be presented through the following themes: 1) structuring the process; and 2) interruptions/getting a word in.
Structuring the Process
Each session should optimally start with ‘minding the group’ and ‘building bridges’ to the previous group session. It is a strategy that aims to compensate for the lack of self-structure among the members and to promote a better group cohesion. It reminds the members of what happened, emphasizing that the recapitulation is a result of the therapists minding endeavour (‘this is what we have thought since last time’), and also conveys what the therapists have found important to mention. Next, the therapists should ask for initiatives, preferably difficult interpersonal events to be explored in the group. The next structuring challenge for the therapists is terminating the work around one patient and opening up for another protagonist. Finally, the therapists should terminate the session in an organized way. In the following, we provide examples from the opening phase, the mid phase and the end of the group.
In this group the therapists omitted any reflection upon the last meeting. They went straight on to organize themes and events. The patients had obvious problems with this phase. They took long to call in their events, and if they did, they usually were not able to keep it brief and started to give detailed accounts as if the therapeutic work had already started. The therapists struggled hard to organize the phase, as illustrated by this sequence from about eight minutes into a session.
Sarah is presenting a theme that she may want to work with in the group and the therapists are trying to reach a conclusion in order to move on in the planning of the session:
. . . and dating is like, so stiff! Getting home to take a shower, dress up, and fair enough I like that, I do, but . . .
(attempting to break in) But wait a minute, wait, Sarah, wait! Sarah! Sarah, wait a minute . . .
It’s so, it’s so stiff! (laughs).
Hold on Sarah! (smiles) ee . . .
(scattered laughter).
But is it important?
Is, is, shall we . . . is, is your relationship to him an important subject to you, and how it has been?
No, we don’t have a relationship yet.
No okay, but okay, it’s eh . . . I’m not speaking of a relation as a romantic relationship, but considering, you have a relation to him because you’ve met him, and it means something to you. And it is a relation in that respect, is what I meant. Becau . . . that is, not a relation as in a romantic relation, but your relation, considering that he, is a man who arouse some feelings.
Oh yes, but I fear that I will end up with the same thing I had with Chris, because, I already get a sense that . . .
Yes, yes (gestures: wait, stop, enough already).
This sounds like an important event.
It seems to be one of those warning signs, with all this, like fixation on appearance.
Yes, but hey, but wait a minute.
Yeah! I’m waiting!
(laughs).
Yes (laughs).
There’s no way to stop you, when you’ve started!
(laughs loudly, puts her hand in front of her face) Aah! I feel like I’m . . . (unintelligible).
But perhaps, eh like, perhaps it’s important for you to have a look on this matter?
It would be even more abnormal if you did not feel like this! If not you would have gone blindly into something with someone, one that you might as well had found completely uninteresting, just to see if there was anything to it.
Yes, but I’m thinking it’s better . . .
Hold on! Okay! (gestures: stop, wait).
Now we’re working!
Now we’re working again, let us . . . s-s-say this is a theme.
Yes, it sounds very interesting.
Are there anyone else who have eh . . . anything?
The length of the opening phases varies considerably among groups. In well-organized groups it is fairly rapid and patients and therapists come to some kind of agreement within five to 10 minutes. 10 minutes after the sequence above, the group had not yet started working, bearing witness of problems with the alliance and agreement on the basic group principles.
In the next sequence the therapists are terminating a turn with a designated patient. T1 has been trying to stop Sarah for about four minutes in order to move on to a new patient. Sarah is clearly mocking him, and she is fully aware that she has been ignoring him for some time.
I’m thinking we might be stopping there.
Yes, we’ve stopped several times! (Smiles and imitates T1’s ‘stopping-gesture’). We’ll stop here! We’ll stop here!
(unintelligible comments and laughter).
I’ll zip my mouth now (straightens herself up in the chair).
You get eager, Sarah.
Yeeah, but, eh . . .
But that’s a good thing!
But you’ll have to praise me, I’ve been good and worked out a great deal.
Yees (hesitant).
I think you’re doing great.
And I’ve been working on loads of other stuff in the group too.
I get really impressed with you, when you’re doing well in your job, and you’ve been working and have regular activities and stuff!
Yes.
It’s got to be just wonderful!
Yes . . .
(laughs).
But then I feel sorry for myself some times.
(laughs).
Who doesn’t?
Yeah, okay! Shall we move on?
(laughs).
Not to render it commonplace though.
No, I know! I know, I get what you’re saying, everybody feels the same. I’m normal! (looks at T2).
Mmmm (nods).
I’m one among many (smiles) I’m not . . .
(laughs loudly).
I’m not on my spaceship anymore!
(laughs).
(laughs).
(smiles) I’m getting the feeling that eh, I’ve been repeating myself a number of times now.
(laughs on).
okay, zip! (‘zips’ her mouth).
Where were we?
The following sequence is from the end of a session. Rose is stopped in the middle of a frantic effort to pack in more details about her life in the very last minute. The result is that the group ends abruptly, not leaving room for any summary or closing remark from the therapists.
(draws his breath) okay, it’s, it’s . . .
We’re watching some detective series on TV, right? And Kirsten comes in halfway through, because her girl wouldn’t sleep. She had to lay down with her and sing, until she fell asleep! (frowns).
Clock is ticking.
Where the hell is normality then?
Well, so you say. Clock is ticking. And we could be looking at many things here, cause I can hear that . . .
Could sit for one more hour!
Sounds as if there is a lot being stirred up here and it gives rise to many episodes that have provoked you. Eeeh, and I think it’s great that you bring it up . . . Think it’s . . .
It’s just whining.
Is it just whining?
So now it’s all about doing a turnaround, now, to begin thinking about myself again. And that’s something I have been thinking of a lot, that’s something I find so hard now, because that’s the way I am. Now I’m all grumpy and irritable, and to make matters worse, daddy just fell, because he was moving last Saturday. (Looks at her wrist-watch) I’ll hurry up!
(laughs).
He fractured his kneecap! And boy do I feel bad for not visiting my mum in the Easter holidays, because we had visitors at home the whole time. And Thomas is calling me constantly, and Jesus I’m driven crazy! So I think I have to go off to somewhere. Go to Hawaii or someplace like that.
Spain . . .
Be gone for a month. Or two, or three, or four.
And why not?
(laughing) take off or take a stand.
Okay! That’s it for today.
Structuring the process provides ample opportunities for stimulating metacognition, e.g. by not focusing on the content, but on structural principles that are decisive for the very process of dialogue and mentalizing in a group (e.g. ‘were we able to tune in on this story?’). By discussing and practising these principles, the members ideally may come to internalize and own them, and thus, this part of the discourse may contribute to the alliance. In this group, the therapists did not stimulate (or had they given up?) this kind of higher order reflection.
Interruptions/Getting a Word in
There were a high rate of interruptions on the part of the patients, and the therapists often had difficulty getting a word in. In advance of the following sequence, T1 had made many unsuccessful attempts to slow down the ongoing interaction. He opens with excusing himself for interrupting Sarah, but ends up being ignored and not getting his point through.
I wond . . . I’m wondering, I think I interrupted you earlier, Sarah, eh . . . if that’s the case I’m sorry
(shakes her head).
(to Rose, while T1 speaks) Not to be nasty.
No! But I see what you’re saying.
(to T1) what?
I thought . . . uh, you said something like ‘what can youth understand?’
Yes.
(draws his breath) Wha, how . . .
I think she may have too high expectations of them.
Now I tell you what, I don’t think so! That’s the least you can expect from someone who’s turned 20 years old!
Nooo . . .
But . . . but youth, they are all different too!
Yes, wha . . .
That’s, they are . . .
You must bear in mind that people express love differently!
Yeah!
Like Jesus, you can’t, like I, understand that you may have . . .
But! Yes that’s exact . . . that’s exactly what they . . .
What kinda expectations, if you have expectations it’s just the same as being let down.
But wha . . .
You’re better off having no expectations and let them show you love in their own way!
And that’s what I think I’m doing too, cause that’s how I am, and that has nothing to do with the kids, and it can be something that . . . maybe something is supposed to take place and I think ‘no, it’s not gonna happen anyway’.
(draws his breath).
That’s something to remind them of too. What you actually do for them! Because they’re . . . most kids are certainly, spoiled!
Yes but how are they supposed to know, and we acknow . . . or express love when you don’t tell them what you want then?! It’s like you have . . .
I have told them!
Yes but, once then!
No, I have told them several times after that too.
(says something that drowns in Rose’s speech).
T1 might have had something useful to say in this sequence, something that could clarify what they were talking about, but he does not get a word in.
Not-knowing Stance
We will present two aspects of the therapists’ challenges associated with the not-knowing stance. These challenges are: 1) determining importance; and 2) that therapists may unwittingly promote pseudomentalizing by stimulating patients to mentalize on the basis of chaotic narratives replete with multiple and fuzzy problems.
Determining Importance
The therapists conducted the group by spending a long time (20–30 minutes) on the opening phase of each session. Partly, this could be understood as a consequence of the therapists not holding an expert position in the group, because a major challenge was confusion among the patients about which topic to present.
Before the following segment, Sarah has just been narrating her emerging relationship with a new man. We understand from what she was saying earlier in the session that relationships are problematic to her in many ways, but as T1 asks her what she is afraid will happen, she answers that she is afraid to get stomach trouble and be embarrassed. T1 draws on what Sarah said earlier when he asserts that there is another, perhaps more important aspect to the situation.
And I’m thinking like (smiles) . . .
But . . . eh is that what you are most anxious about now, because . . .
I’m mostly . . .
I . . . eh was a little surprised with your answer, cause you you . . .
Mmmm . . .
I would have thought you’d say ‘I’m afraid that . . .’
Not s . . .
. . . I’ll engage in a relationship that eh . . . is not making me feel, not making me feel good, and that is not good for me, and that may lead to me starting drinking again. That’s what I would have thought was your answer.
No, that’s no problem.
You’re not afraid of that?
No like, I’m such a badass when it comes to that stuff, I . . . I’ll just ‘pyssh’ (gestures: tossing something over her shoulder), enough of that, but . . . eh . . .
What do you think about that I thoug . . . eh . . . about my understanding of the situation?
I understand that, because that’s what’s normal, but I feel that by now I’ve been working so hard with myself. I’m that independent, that I have . . . I’m standing on my own two feet, dammit. I used to be terribly emotionally dependent on Chris, but no more of that. I refuse to be emotionally dependent on another human being, or to be dependent on another human being in any way. Eeeem . . . and I have so many other things in my life that I’ve filled my life with, so there is no . . . even if I like meet him a couple of times more and find it’s not working out, I’m thinking, well well, but at least it’s one more lesson learned, like it’s . . .
What are you others thinking about this?
Like it’s not . . . life’s not over.
No.
What are you thinking?
Well, it sounds like terrific and you seem so incredibly strong.
How can you figure out how much he is drinking then?
We see that T1 is cautiously questioning Sarah’s focus on physical and somatic aspects of the situation. Instead of being more persistent, he calls on the group for support. T1 is clearly on to something, he knows that there are important mental aspects to this. He hints towards alternative perspectives, but is also hesitant and ‘not-knowing’. He asks the group what they are thinking; apparently hoping they will pick up on his cue. Unfortunately the other patients (or T2) seldom follow him, and he ends up not getting his point through.
Promoting Pseudomentalization
One of the most frequent interventions from T1 was, on the basis of patient narratives, to ask the group what they were thinking about the situation being described, what they believed was going on in the minds of the subjects of the narrative, what they were feeling, and why they may have been behaving the way they were. From a mentalizing point of view, these are perfectly legitimate questions. However, as a consequence of poorly structured processes, the information the patients were asked to use as a basis for mentalization was often extremely vague. Upon T1’s questions, this vagueness led patients to make general assumptions, suggest stereotypical explanations of what was going on, and often simply suggest solutions to the fuzzy problems they were presented with. What seemed to be efforts to mentalize were often not grounded in an appreciation of real events, but rather on phantasies and loose assumptions. In other words, the patients’ mentalization was often not based on the results of conjoint explorations of subjective realities, and appeared therefore more like pseudomentalization.
The sequences we have already quoted contain several examples of pseudomentalization. In the following example, Trudy has just finished a complex story, which becomes interpreted by clichés. The behaviour is believed to be a consequence of unresolved grief/sorrow without any evidence of it.
(expires) Hoi!
Mmmm.
But not that exact episode though, the condom stuff, because then . . . I had already caught him cheating, so it was already over, but I had the key to his apartment, so that was no sweat. And eventually she got fired.
But . . . there . . . there’s been a lot going on here.
Mmm, (laughs a little) there’s usually some action.
A-a-and, what are you (the group) thinking? Eeeh . . . How do you think Trudy has been doing?
(Trudy laughs).
You do have a fierce temper.
Yes I do.
It’s because you’re not able to free yourselves from the sorrow, I think. So you get angry instead. There’s always something lying under the anger.
But I’ve always had my temp . . .
There’s always something lying under the anger!
Calling on the group to mentalize is a technique that underscores the principle of therapist–patient equality: the therapists are often not in a better position than the patients to make sense of mental states. However, the therapists are often in a better position to know what information is needed to properly grasp what is going on. The preparation of scenes is an important task in MBT-G. Therapists should assist group members in clarifying relevant scenes, and then stimulate curiosity about what went on in the minds of the involved people. Our impression in this group was often that if the therapists themselves were to make any sense out of the ‘hotchpotch’ presented in many of the sequences, their mentalization would be just as much guesswork, that is, pseudomentalization.
Discussion
In this study of a MBT group at a MHC, we found chaotic group processes of types that are well known from the literature (Pines, 1990; Marziali and Munroe-Blum, 1994; Hummelen, Wilberg and Karterud, 2007). The group was conducted by two therapists who had previous training in group analysis and MBT. We could observe that one of them (T1) consistently tried to apply MBT principles. However, the effect upon the group seemed minimal with respect to constructing scenarios that were useful for conjoint exploration and mentalizing.
We found that several themes from the Thematic Analysis were related to two principles in the MBT-G manual: managing authority and the not-knowing stance. The therapists struggled to manage their authority in an efficient fashion, and also exhibited what we conceived as an exaggerated not-knowing attitude. That is, the therapists were seen to adhere little to the item ‘managing authority’, while the ‘not-knowing stance’ was overplayed. The therapists seemed to be trapped between two models. With one leg in each camp, they did not practise group analytic psychotherapy, nor did they practise MBT-G.
The findings of this study illustrate some difficulties therapists encounter when trying to establish themselves as leaders of a certain communicational structure. We observed how the therapists were overheard, not paid attention to, interrupted, and at times ridiculed. As outlined in the companion article (Karterud, 2015), this is a form of discourse where no development of social cognition can take place. Some patients had become very influential in shaping this discourse and the therapists feared aggressive outbursts if they were to challenge it. Consequently, trapped in a lack of therapeutic alliance, they became reticent about their endeavour to transform the therapeutic process to conform to MBT-G standards.
We suggest that some of the difficulties that the therapists encountered can be explained by competing theoretical models in their minds, being embodied and acted out through a more mentalization-based approach by T1 and a more psychodynamic approach by T2. We still contend that group analysis and MBT-G are not contradictory models of group psychotherapy (Karterud, 2011). We have witnessed several fruitful hybrids. A working hypothesis is that the better mental functioning the group members exhibit, the more appropriate are group analytic principles, while increasing mentalizing difficulties call for MBT techniques.
The study concerns three meetings in a MCH MBT group and the findings may not be valid for this group over time. We may have observed a stressful period in the life of the group. We should neither generalize the findings to other MCH MBT groups. Nevertheless we recognize a theoretical and therapeutic tension here that may have relevance for many groups. Therefore we have taken the opportunity to exploit these group experiences for the purpose of clarifying crucial theoretical and practical issues. Through this study, we have also become more alert to sources of confusion on the themes of authority and structure versus the not-knowing stance, and it has helped us to clarify issues that earn more consideration in MBT-G training, and possibly in group analytic training as well.