Using therapeutic metacommunication and systematic monitoring of the working alliance in adolescent psychotherapy: a clinical case study

ABSTRACT

The quality of the working alliance as assessed by the patient is the most robust predictor of the therapeutic outcome. However, the working alliance is much less studied in youth psychotherapy than among adults, even though it might be a more complex phenomenon. This clinical case study presents examples of how a problematic working alliance is addressed through therapeutic metacommunication and how strengthening the therapeutic alliance contributes to the outcome in the psychotherapy of an adolescent girl. The results of this study encourage further studies of working alliances in adolescent psychotherapy apart from child and adult psychotherapy.

KEYWORDS:

• Working alliance
• therapeutic metacommunication
• rupture resolution
• adolescent psychotherapy

Introduction

One of the clearest results in psychotherapy research is the impact of the quality of the working alliance as assessed by a patient (Doran, 2016). However, some data suggest that psychotherapists are generally not very skillful in estimating the patient’s view of the working alliance or the progress of the therapy, and they tend to overestimate them (Hersoug et al., 2001). It has been suggested that a lack of reliable feedback often hinders the progress of the therapeutic skills of the therapists (Chow et al., 2015). There are also empirical data suggesting that if the therapist receives feedback about the progress of the treatment and the quality of the relationship, it provides a potential for better results and less premature terminations (Miller et al., 2015). The working alliance in adolescent psychotherapy has been studied much less than in adult psychotherapy, even though it may be a much more complex phenomenon in adolescent settings (Hawley & Garland, 2008).

Therapeutic metacommunication is a concept used to refer to therapeutic conversations about the therapeutic relationship, aiming to repair alliance ruptures when they occur (Safran et al., 2014). Research about therapeutic metacommunication, alliance ruptures, and repairing alliance ruptures and their unique characteristics in adolescent psychotherapy settings is very scarce, however. In this patient-focused clinical case study, the aim is to examine how feedback about the working alliance is used to strengthen the alliance, how therapeutic metacommunication is used to resolve problems in the working alliance and to discuss advancing research and clinical work in the field of working alliances in adolescent psychotherapy.

A comprehensive body of evidence shows that, across different orientations and measures, a working alliance is a consistent predictor of psychotherapy outcomes (Flückiger et al., 2018). Previous meta-analyses about the working alliance in child and adolescent psychotherapy have yielded similar results to those about adults (Karver et al., 2006; Shirk & Karver, 2003), but a more recent meta-analysis by McLeod (2011) found a smaller effect size (0.14). However, there seemed to be a lot of variation in how the alliance-outcome relationship was moderated across theoretical and methodological variables, implying that more detailed research is needed. There is some evidence about slight differences in how the therapeutic alliance affects the treatment effects in adult and adolescent psychotherapy. Contrary to adult psychotherapy studies, in adolescents, the outcome seems to associate more strongly with the working alliance measured later in the treatment than earlier on. This might imply that, in adolescents, the working alliance develops more slowly than with adults, or then again, it might reflect that adolescents lack reflective skills to assess the working alliance, and good outcomes lead to better assessments about the working alliance. Also differing from adult studies, in adolescents, the assessment of the therapist seems to be more strongly associated with the outcome than assessments of the adolescents themselves (Shirk & Karver, 2003).

Working with adolescents and creating and maintaining a therapeutic relationship with an adolescent is more complicated than working with adults also, because of the developmental stage of adolescence between childhood and adulthood and especially great changes happening during that period in relationship between adolescent and their parents and in their own autonomy and agency. Complicated and changing relationship with adolescent’s parents, withdrawal and struggle for autonomy often amplifies and complicates the occurring transference and countertransference patterns, and how they are pulled to be enacted in the therapeutic relationship (Friedman & Laufer, 2018).

The importance of the relationship between therapist and the client has been a central idea inside the psychoanalytic theory since the beginning. Freud first introduced the idea of positive transference as a way of allowing therapist and the client to collaborate working through the transference and resistance, and he later also suggested that there might be a partly transference-free bond between therapist and the client (Strachey, 1964). After Freud, the roots of current understanding about the importance of working alliance within psychotherapy process are considered to be in work of Sterba (1934). He set the foundation on the discussion about working alliance, considered the therapeutic relationship emerging from the positive transference towards the therapist crucial for successful treatment, and also emphasized that the therapist contributes substantially to establishing and maintaining the relationship.

The term working alliance was introduced by Greenson (1965). He defined the working alliance as rational, non-neurotic part of the therapeutic relationship, distinct from the transference neurosis. Nevertheless, he considered working alliance no less important than working with transference. Later on, the relational school of psychoanalysis developed especially in North America, has been emphasizing the relationship between the patient and the therapist more and more, sometimes in a way that more traditional psychoanalysts see as forgetting the importance of dynamic unconscious altogether (Mills, 2005). However, relational theorists themselves consider therapeutic relationship and unconscious transference intertwined. Transference is considered to emerge as an adaptation to specific therapist and to specific relationship, and the therapist has a notable degree of influence on how the relationship and the transference shapes up. Countertransference is not considered as a inference to therapy, but as an important tool for gaining awareness and deepening the therapeutic work, and therapist’s openness, subjectivity and self-awareness are valued (Michell & Aron, 1999).

Safran and Muran (2006) have characterized that the concept of therapeutic alliance was particularly useful within psychoanalytic tradition, when psychoanalysis emphasized technical aspects over relational aspects, and tended to understand relational aspects on therapeutic relationship rising merely from the patient’s transference. Concept of alliance provided possibility for greater therapist flexibility and responsiveness. According to them, after the relational shift and emphasis on authenticity, flexibility and mutual influence on relationship, the alliance has been taken more as a granted or as a basic assumption in psychoanalysis. Indeed, the traditional conceptualization of the working alliance may overemphasize the rational collaboration between therapist and the client and disregard transferential and unconscious aspects about the relationship. Bordin’s (1979) classical definition has become popular among psychotherapy researchers partly, because it avoids this controversy of whether collaboration is unconscious or conscious. In Bordin’s definition, quality of the alliance consists of three parts, the bond, meaning how well the client feels understood and respected by the therapist, and therapist and client agreement on therapy tasks and goals.

New interest in concept of working alliance, characterized by Safran et al. (2014) as the second generation of the alliance research has focused on understanding the factors that lead to the development of the alliance and the processes that involve repairing the alliance ruptures. This research has contributed to more systematic understanding about distinctive ways to manage the therapeutic relationship. Even with our current understanding about two-person psychology and complexity and two-tied nature of the transference and countertransference, studying the alliance distinctively can give us more systematic and detailed information about aspects contributing to maintaining the bond between therapist and the patient while therapeutic work itself is often painful and difficult. According to Safran et al. (2011), there are three different ways to identify and study alliance ruptures: patient and therapist self-reports of ruptures within the session, changes in measures between sessions, and observer-based methods.

In adult psychotherapy settings, there is a growing pool of empirical research data providing information about the importance of alliance ruptures and reparation in clinical practice. For example, a large naturalistic study (N = 1104) in a primary care psychotherapy setting in Sweden found that ruptures in the working alliance were associated with inferior outcomes, but a successful resolution of the ruptures seemed to avert this negative association. In longer treatments, the rupture-repair pattern was even associated with better treatment outcomes than the patterns without ruptures at all. However, only just over half of the identified ruptures were repaired (Larsson et al., 2018). McLaughlin et al. (2014) have yielded similar results about the importance of rupture resolution also in a manualized 10-session psychotherapy for PTSD consisting of prolonged exposure. There is no empirical quantitative research data about alliance ruptures and their repairment in adolescent psychotherapy, but there is limited available research data about alliance development trajectories within psychotherapy for anxiety in children and adolescents. Kendall et al. (2009) studied alliance development trajectories in children with anxiety disorders within family-based psychotherapy with and without exposure tasks. They found that alliance ratings assessed by the therapist, child and parents grew across different treatment modalities. In a study by Chu et al. (2014) about CBT for child and adolescent anxiety, however, alliance ratings by the therapist indicated systematic growth, where ratings made by children and adolescents themselves were quite stable across the treatment.

There are some qualitative research data about how alliance ruptures are displayed within the psychotherapy relationship in adolescent psychotherapy and how therapists address them and aim to repair them. For example, Binder et al. (2008) have interviewed psychotherapists working with adolescents and children about how they try to re-establish contact with a patient when facing an alliance rupture. The strategies they found were exploring reasons for rupture from the adolescent’s point of view, confirming ambivalence or handling it directly as a choice, establishing a language for fluctuations in the adolescent’s experience of motivation and distress, interpreting not wanting therapy as a sign of autonomy needs and self-protection, and exploring reasons for the rupture from the therapist’s subjective point of view. Diamond et al. (1999) have studied interventions aiming to improve poor alliances in Multidimensional Family Therapy for adolescents. They found five reliably codable alliance-building interventions, orienting the adolescent to the collaborative nature of therapy, formulating personally meaningful goals, attending to the adolescent’s experience, presenting themselves as an ally, and addressing trust, honesty, and confidentiality. In their data, attending to the adolescent’s experience, presenting themselves as an ally, and helping the adolescent to formulate personally meaningful goals in the first three sessions of the therapy were associated with improving the working alliance.

In clinical literature about the therapeutic alliance with adolescents, the notion that adolescents are often referred to treatment by someone else is often presented as a complicating factor. It is seen that adolescents often lack the cognitive maturity and psychological insight to understand their problem, the need to change, and the idea that therapy would be useful in the change process (DiGiuseppe et al., 1996). In regard to Bordin’s (1979) classical definition, the literature on adolescents has focused heavily on the bond (Zack et al., 2007). Empathy and genuineness, nonjudgmental acceptance and understanding are considered essential for creating a therapeutic bond with adolescents (Oetzel & Scherer, 2003). Fitzpatrick and Irannejad (2008) have suggested, that with adolescents who are already ready for change, the agreement about tasks and goals of the therapy should be more in focus than with adolescents who are not yet ready for change. This emphasizes the need for flexibility and therapist capacity for individualizing the treatment for each adolescent (Clark, 2013).

A great deal of work on understanding alliance ruptures and resolving them has been done by Jeremy Safran and his colleagues. They have used the concept of therapeutic metacommunication to describe how the therapist, when encountering a rupture, tries to bring the conversation to what is happening in the therapeutic relationship. The aim of the metacommunication is to bring the rupture, the relationship, and what is happening within the relationship to mutual conscience and to achieve a more outside and reflective position about the ongoing relational cycle (Safran et al., 2014). Therapists should use their awareness of their subjective interpersonal reactions about the interaction and the relationship as a source of information about ruptures, and to use that information in order to mutually explore explicitly relational aspects that are being unconsciously enacted, and to accept responsibility about therapist’s own contributions to rupture in open and non-defensive manners (Safran & Kraus, 2014). Safran and Muran (2006) argue that alliance ruptures can all be understood as transference-countertransference enactments. This formulation allows us to consider therapeutic metacommunication as one specific way to address countertransference and working with alliance ruptures in order to restore the collaborative connection, as a specific way to work through transference-countertransference enactments.

Safran et al. (2011) have summarized the most common interventions therapists use when addressing alliance ruptures. According to them, they include repeating the therapeutic rationale, changing the task or goals, clarifying misunderstandings at a surface level, exploring relational themes associated with the rupture, linking the alliance rupture to common patterns in a patient’s life, and providing new relational experience. Later, Eubanks-Carter et al. (2014) developed a coding system to identify alliance ruptures, which also includes a further developed system for identifying rupture-resolution strategies from therapy transcriptions. In the Rupture Resolution Rating System (3RS), they describe ten resolution strategies: clarifies a misunderstanding, changes tasks or goals, illustrates tasks or provides a rationale for treatment, invites the patient to discuss thoughts or feelings with respect to the therapist or some aspect of therapy, acknowledges their contribution to a rupture, discloses their internal experience of the patient–therapist interaction, links the rupture to larger interpersonal patterns between the patient and the therapist, links the rupture to larger interpersonal patterns in the patient’s other relationships, validates the patient’s defensive posture, and responds to a rupture by redirecting or refocusing the patient.

In this patient-focused clinical case study, the aim was to examine the establishment and strengthening of the working alliance in one long, non-manualized integrative psychotherapy process of an adolescent girl with major depression, self-injurious and suicidal behavior, and other comorbid problems. The working alliance is understood in accordance with Bordin’s (1979) classical definition as the bond itself, and the agreement on tasks and goals of therapy. The beginning of the treatment was characterized by a struggle in creating and maintaining the working alliance, and interaction showed markers of alliance ruptures. A specific aim is to show an example of how mutual insecurity in a therapeutic relationship and emerging transference-countertransference pattern complicated working collaboratively and steered the tasks and goals of the therapy in the wrong direction. Also, the aim was to examine the therapeutic interaction on sessions where the alliance strengthened and examine how therapeutic metacommunication and alliance-correcting interventions and rupture-resolution strategies described by Safran et al. (2011) and Eubanks-Carter et al. (2014) may have affected the strengthening of the working alliance, and how systematic monitoring of the working alliance may have helped therapist and the client to work through emerging transference-countertransference patterns and to establish collaborative working alliance. Implications on clinical practice and possible future research on adolescent psychotherapy are discussed.

Materials and methods

Measures

Fluctuations in working alliance was measured using the Session Rating Scale (SRS; Duncan et al., 2003). SRS is a questionnaire developed for routine use in clinical practice for monitoring working alliance after each session, and used as a indicator of a therapy at risk for a failure, and as an intervention and a way of getting to discussion about the progress of the therapy. SRS is normally used together with similar outcome measure, Outcome Rating Scale (ORS), and measures are used as a tool to talk about the working alliance and therapy progress in a certain way. In this case, however, SRS was used solely for measuring the working alliance, and the patient was not willing to discuss about her assessments of the alliance. SRS consists of four visual and continuously assessed items, based on classical definition of the elements of the working alliance defined by Bordin (1979), emotional bond, shared goals, and mutual understanding of the methods or tasks of the treatment. The fourth item measures patient’s overall experience of the session. On each item the respondent places a mark on a 10 cm line representing his or her assessment of the item. Values are scored by measuring on what place on the line the mark is. Each item can get values between 0 and 10, and the total score can vary from 0 to 40.

For assessing change in psychological well-being during psychotherapy, The Young Person’s Clinical Outcomes in Routine Evaluation (YP-CORE; Twigg et al., 2009) was used. YP-CORE is a pan-theoretical self-report measure developed to assess global level of distress of children and adolescents aged 11–16 years. It consists of 10-items assessed on a 5-point likert scale ranging from 0 to 4.

For assessing the symptoms of depression, a short modified version developed from Beck Depression Inventory to use among Finnish adolescents was used (RBDI; Raitasalo, 2007). It is a self-assessment inventory including 13 items. For each item, respondent assesses as a five-point likert scale how well the item describes his/her current feelings. Total scores can range from 0 to 39.

At the beginning of therapy, the patient (Sara) was a 16-year-old girl diagnosed with major depression, post-traumatic stress disorder, and anxiety disorder. Before psychotherapy, Sara had been in treatment for two years, including regular conversations with a psychiatric nurse. Despite the regular appointments, her depression and anxiety got worse, and after a suicide attempt and a short treatment period at an adolescent psychiatric inpatient ward, she was referred to a secondary care adolescent psychiatric outpatient clinic. After the diagnostic assessment, she was referred to start a psychotherapy assessment with me. I worked in a specialized care adolescent psychiatry clinic within public healthcare. At the moment, I was doing my four-year specialization training in integrative psychotherapy, and the sessions were recorded for supervision. The therapy orientation was long, non-manualized psychotherapy integrating different theoretical orientations freely. In the beginning, sessions were conducted twice a week, but after six months, the session frequency was changed to weekly sessions. In the termination phase of the therapy, approval for the study design was granted from the research unit of Tampere University Hospital, and the patient was asked to provide informed consent to use recordings and other collected data for research purposes.

When I first met Sara, she was very critical of herself, and she was socially anxious and unable to attend school. She had regular anxiety attacks, and she was depressed. She felt that her appearance and the way she presented herself was conspicuously weird and horrible. In order to cope with her anxiety, she was cutting herself regularly. After the psychotherapy assessment, she presented herself as motivated, although she said that she finds the psychotherapy setting somewhat forced and inhumane. Her own self-worded goal for therapy was to find concrete ways to control her anxiety (a sentence I nowadays think many adolescents have learned to phrase as a treatment goal inside the psychiatric treatment system without actually thinking what they really want for themselves), and I thought some very concrete cognitive behavioral and mindfulness-based interventions would be suitable to start with.

The beginning of the therapy was, however, characterized by difficulties in working collaboratively, and the working alliance seemed poor. Sara refused my suggestions for homework assignments and seemed anxious and sometimes passive-aggressive during sessions. She rarely presented with her own concerns for discussion but was somewhat reluctant to engage in discussion about subjects I brought up. I felt powerless and anxious and feared that Sara would drop out of therapy prematurely. My anxiety grew, as she talked with criticism and distrust about previous encounters with mental health professionals. I was afraid that at some point, if not already, I would be a similar disappointment to her as the professionals before me. When looking back the beginning of the therapy, I would conceptualize Sara’s passivity and withdrawal rising from transference-based assumption about me expecting and demanding her to be able speak about her experience clearly and to be able to work and progress in therapy quickly, clearly and efficiently. Also, I would consider that she had as well a transference-based assumption, that I would not really be interested about her experience and would not be able to understand it. Sara’s withdrawal enacted in me anxiety and need to work efficiently, offering quick interventions that would be concrete and easy to comprehend.

As we continued working, I noticed that the ambiance of the situation and our relationship and collaboration seemed to go worse every time I did try to bring in some concrete interventions or exercises, or even when I tried to work in a concrete way, for instance by trying to do a chain analysis about a situation where Sara had injured herself. Also, all efforts to normalize some of Sara’s experiences, or give psychoeducational talks about how experiences Sara was feeling usually are understood seemed to be poisonous for our collaboration. Then again, exploring Sara’s relationships with family members, showing empathy and validating her feelings seemed to be inducing more collaboration, at least for a while. Talking about our mutual relationship, engaging in therapeutic metacommunication, also seemed to ease the ambiance charged with anxiety, but I noticed that my own anxiety narrowed my mind, and all the moments that I should have or could have been seizing seemed to pass. As an intervention for difficulties in the working alliance, I suggested we could start monitoring the working alliance as perceived by Sara, beginning with session number 36.

After session number 36, fluctuations in the working alliance were measured using the Session Rating Scale (SRS). The SRS is a questionnaire developed for routine use in clinical practice to monitor the working alliance after each session; it is used to indicate whether the therapy is at risk for failure and is also used as an intervention and to promote discussion about the progress of the therapy. The SRS is normally used together with a similar outcome measure, the Outcome Rating Scale (ORS), and measures are used as tools to talk about the working alliance and the therapy progress in a certain way. In this case, however, the SRS was used solely to measure the working alliance, as Sara was not willing to discuss her assessments of the alliance.

Data analysis

From the SRS data, particular sessions in which there was a substantial improvement in the working alliance as reported by the patient were identified (Sessions 43, 44, 47, and 48; see Figure 1). These sessions were transcribed, and from the transcriptions, I searched for therapist actions that could be identified as rupture resolution strategies, which were described by Safran et al. (2011) and Eubanks-Carter et al. (2014). Transcriptions were translated from Finnish to English by the author.

Figure 1. Patient evaluations of quality of working alliance measured with SRS from therapy sessions 36–72.

Transcriptions

Session 43 included rupture resolution strategies. The following is a transcribed excerpt from the session:

Therapist: About what would you like to talk about today?

Sara: I don’t …

Therapist: How’s perfectionism as a subject?

Sara: I don’t know what to say about it.

Therapist: It somehow came up now with this situation.

Sara: It is just a thing that I have, and I can’t help it.

Therapist: Does it feel like I’m blaming you for it? (Therapist invites the patient to discuss thoughts or feelings with respect to the therapist or some aspect of therapy.)

Sara: Maybe, yeah.

Therapist: It is not my intention at all … I understand that you can’t, or if you could just change it you surely would have done it. I see that you suffer about it. And that I’m bringing it up and talking about it is not at all meant to blame you. (Therapist clarifies a misunderstanding.)

Sara: Uhum.

Therapist: But I feel, I feel sadness and compassion toward you. (Therapist discloses her internal experience of the patient–therapist interaction.) I see how it’s somehow repetitive. It comes, you suffer, but you can’t do anything about it. It looks like a chain or like a train that just goes and you can’t get out. But it’s probably also quite an unpleasant subject to talk about … (Therapist validates the patient’s defensive posture.) Because we have been talking about you blaming yourself for blaming yourself … Is it a beneficial subject, does it make sense to talk about it, what do you think? (Therapist invites the patient to discuss thoughts or feelings with respect to the therapist or some aspect of therapy.)

Sara: Well, I don’t know. Mostly it irritates me, I guess.

Therapist: Are you irritated about it right now? (Therapist invites the patient to discuss thoughts or feelings with respect to the therapist or some aspect of therapy.)

Sara: Well, not that much now.

Therapist: I wonder, what would be a subject you yourself would feel beneficial or sensible? (Therapist responds to a rupture by redirecting or refocusing the patient.)

Sara: I don’t know.

Therapist: I think that at this point, it would be easiest for me to come up with a subject we would then start to talk about, something I consider important … but I notice that I’m thinking that, could it be that then we would be talking about something you find irritating or less important, and that we would be going in the wrong direction? (Therapist discloses their internal experience of the patient–therapist interaction.)

Sara: I don’t know.

Therapist: Like am I pulling this in my own direction … Have you been feeling like that lately? (Therapist links the rupture to larger interpersonal patterns between the patient and the therapist.)

Also, session 44 included rupture-resolution strategies. See this transcribed section from the session:

Therapist: Often, therapy might be a place or a space or a situation where you can also express thoughts and feelings which you can’t in another situation or other ways, or aren’t possible in other ways … Sometimes it might be that just feeling and expressing feelings like that could be beneficial and, like, help with them … So, this could be a situation where you could – like the feelings that you can’t bring up at home – if you want, you could experience them and talk about them here … That here, you don’t have to think about whether you are insulting your parents or … Because here we are considering your experience, and when we talk about your parents, we are talking about your mental image about your parents and about your feelings and experiences and dealing with them. Like basically, this is a different place or space where you can deal with those feelings a bit differently than is possible in other situations … (Therapist illustrates tasks or provides a rationale for treatment.) I don’t know if this makes any sense to you?

Sara: Yeah, it does.

Therapist: In here you could also be irritated, angry, disappointed, or furious, feel those feelings that you can’t feel in another situation. (Therapist illustrates tasks or provides a rationale for treatment.)

Session 47 included a therapeutic conversation that can be described as an attempt at rupture

resolution strategies. See the transcribed section:

Sara: Yeah, that just frightens me, like what I counted, that would be 10 different people who I had to talk with in these past few years, and not at all have they all been sensible. Not at all have they all understood me.

Therapist: So you have an experience that all of them don’t know what they are doing [Sara: Yeah] and then you are worried based on that, your own experience.

Sara: Yeah.

Therapist: How often do you have that kind of experience, like a bad experience?

Sara: Very much.

Therapist: You have said, at least, that one psychiatric nurse was a bad experience?

Sara: Yeah well, that was the worst. But there have been others that I just haven’t … They just haven’t, like, understood me.

Therapist: Have you felt like that in here? Like at which point and how much? Like in here with me? (Therapist invites the patient to discuss thoughts or feelings with respect to the therapist or some aspect of therapy.)

Sara: Not in here with you, but …

Therapist: Not at any point?

Sara: Not really … But [a longer discussion about bad experiences in different stages of assessment and treatment in adolescent psychiatric services.]

Therapist: Like those experiences where you have been misunderstood, and those where you have felt like abandoned or rejected, kind of like not getting what you came for or what you need. (Therapist links the rupture to larger interpersonal patterns in the patient’s other relationships.)

In other sessions assessed (sessions 40–52), no therapeutic metacommunication and rupture resolution strategies could be found.

In the end, this psychotherapy process lasted for almost three years and included 125 sessions. The process could be assessed as successful, as during the therapy Sara could return to school, and move away from her childhood home for independent living as a young adult. At the end of the therapy, she was not experiencing any clinically meaningful psychiatric symptoms. The global level of distress assessed with YP-CORE started to decline after the working alliance was strengthened, and depressive symptoms measured with RBDI declined during therapy (see Table 1).

Table 1. Scores of outcome measures.

Session number	YP-CORE	RBDI
Prior therapy		20
#19	28
#44	23
#45	21	18
#47	21
#55	14	6
#59	25	17
#64	20
#69	14	10
#78	8	1
#97	17
#111	17
#116	9
#121	6	0

Discussion

This study aimed to examine the establishment and strengthening of the working alliance and to show an example about how the collaborative working alliance was ruptured via emerging transference-countertransference enactment, and how working alliance was strengthened using therapeutic metacommunication and rupture-resolution strategies described by Safran et al. (2011) and Eubanks-Carter et al. (2014), and systematic monitoring of the working alliance.

Clinical discussion

When I first started working with Sara, we were both feeling anxiety and insecurity with our therapy relationship. Sara’s difficulty in engaging collaborative work was induced at least partly by her insecurity and criticism towards herself. This came evident in a conversation, where we discussed our relationship and the way I felt all concrete interventions seemed to frustrate her, even though she had said at the beginning she wanted to learn concrete ways to control her anxiety. It can also be assumed that her criticism towards herself was reflected to me creating a transference-induced assumption about my critical and demanding attitude towards Sara. Sara’s passivity aroused insecurity and anxiety in me. I was happy to hear Sara had concrete and clear goals for therapy, but I did not understand that goal-directed work would be too anxiety-provoking for Sara before our therapy relationship would be more secure. Me acting on my countertransference and trying to ease my own anxiety by providing concrete and goal-directed interventions was enacting our mutual transference-countertransference pattern. We ended up to a vicious cycle, where my goal-directed work provoked insecurity and passivity in Sara, and her passivity provoked anxiety (and more need for goal-directedness) in me. This gives us some clinical implications to discuss.

First, my experience working with Sara is in line with the notion by Fitzpatrick and Irannejad (2008), about that goal-directed work might not be suitable with adolescents in an early phase of the treatment if they are not yet ready for change, and that working with the bond might be more crucial at this point. Fitzpatrick and Irannejad (2008) emphasize readiness for the change, and I argue that with adolescents who are seemingly ready for change but are insecure about the relationship, working with the bond would be primary. Sara also expressed at some point being ambivalent about renunciation of the depression. In several psychotherapeutic theories from Freud’s psychoanalytic theory to more recent third-wave cognitive behavior therapies, symptoms are often seen as functional, serving the cohesion and continuity of the self and as a form of avoiding psychic pain (Hayes et al., 1996). This creates ambivalence towards healing, and in psychotherapy with adolescents this might be emphasized, as adolescent’s identity formation is still in process, and continuous experience about self is not as clear as in adults, and motivation for facing the emotional pain is often more limited. Therefore, the therapeutic bond and the safety that it can provide for the patient for a terrifying situation of facing emotional pain, might be even more crucial with adolescents than with adults.

Second, the case of Sara shows that even if an adolescent can articulate specific goals for psychotherapy, working towards those goals is not always straightforward. Compared to adults, adolescents on average lack cognitive, verbal and reflective skills for understanding and verbalizing their notion of the problem (DiGiuseppe et al., 1996), and therefore face more challenges in verbalizing and understanding their very own, perhaps contradictory wishes and motives for psychotherapy. Outspoken goals for treatment might also, in fact, be adopted from parents, other professionals or other sources (Hawley & Garland, 2008) without really being internalized by the adolescent as their own. The case of Sara illustrates how focusing on the therapeutic bond, validating and appreciating adolescent’s feelings and perspectives and expressing genuine empathy towards the adolescent seems to be even more important than when working with adults, as described by Oetzel & Scherer (2003).

The SRS was used in this therapy process differently than normally suggested. This is partly due to Sara’s choosing to not share her feelings and ideas about why she rated the alliance the way she did. This implies that feedback about the therapy process can be collected and used beneficially in many ways. When working with adults, the aim of using SRS would be, in my opinion, more about orienting the patient to think about the alliance, and to facilitate discussion about different facets of the alliance. In adolescent psychotherapy, where the therapeutic alliance can be a more complex and difficult phenomenon (Friedman & Laufer, 2018; Hawley & Garland, 2008), the integrative and innovative use of therapeutic skills and therapeutic tools can be needed and be crucial. We used SRS as a communicative tool to gather mutual information about our interaction, at the point of our therapy process where discussion about alliance was not yet possible. With adolescents, direct discussion about aspects of the therapeutic relationship and working alliance may sometimes be particularly difficult because of power imbalance, insecurity and difficulty to be directly critical towards adults, or for example because of limited verbal ability and lack of suitable words.

I think SRS was, in this case, used also as a communicative tool from me towards Sara. I introduced the SRS as an intervention for the situation where our collaboration did not seem to work and explained that it would be important for me to know what Sara really thinks about our interaction. I also explained that I understand that it is not easy to talk about this kind of thing directly and that we could use SRS as an ancillary to ease it. I believe that in this case, using SRS gave Sara a signal, that I truly was interested in her experience. In addition of being validating and encouraging, this may have worked to de-escalate the emerging transference-countertransference enactment and creating the possibility for a more reflecting position towards our interaction and relationship. We eventually stopped using the SRS at session number 72, as Sara told me that she did not feel we would need it anymore and that she feels that we are at a point, where she can tell me directly if there is a problem with something we do or discuss.

After working with Sara, I have been sometimes using SRS as an intervention for problematic working alliance or therapeutic impasse with adolescent clients. Often results have been good, but not always. In my clinical experience, sometimes the working alliance with the adolescent is so fragile and insecure, that giving feedback and criticism is not possible even indirectly using a tool such as SRS. Sometimes I have not been able to introduce it in a way, that would motivate the adolescent enough for seeing the value and giving feedback and filling the SRS after each session has become frustrating and pointless for the client.

All communication, including introducing any kind of intervention, clinical tool or scale such as SRS happens in context of the therapeutic relationship, and its unique transference-countertransference patterns and enactments. In some other context, the way I introduced the SRS could also be acting on the countertransference in a harmful way, for example confirming transferential experience and expectation about therapist steering away from the real connection to distancing clinical instruments, and not being interested about the unique experience of the client. In psychoanalytic theories emphasizing relational thinking, transference is always considered to emerge as an adaptation to specific relationship and context (Michell & Aron, 1999). In psychoanalytic psychotherapy, use of scales, measures or clinical tools is often disregarded, but I personally consider that SRS, or other kind of scale or intervention may be implemented, but therapist needs to be responsive of the client’s needs in the context and aspire to be aware of one’s own countertransference. Also, it is crucial to be aware of that there is no certainty about how any kind of communication is being interpreted by the client in specific context and relationship with specific transference-countertransference patterns, and the therapist needs to be ready to accept responsibility about the own contribution to the possible rupture in an open and non-defensive manner (Safran et al., 2014). In psychoanalytic and psychodynamic psychotherapy, we value the idea about the therapist being aware of their own feelings and countertransference and using them productively and responsively in a current context. However, we know from the research that psychotherapists in general are not always aware of problems in working alliance (Hersoug et al., 2001), and that unmanaged countertransference is connected to countertransference-based behavior (Ligiéro & Gelso, 2002) and more ruptures and less resolutions in working alliance (Tishby & Wiseman, 2022). At least in some instances, using systematic ways to gather information about changes in working alliance may help the therapist to become more aware not only about the ruptures, but also about their own countertransference, and help to manage it more effectively.

Implications for research

In this particular psychotherapy process, using the therapeutic metacommunication and rupture-resolution strategies described by Safran et al. (2011) and Eubanks et al. (2014) seemed to be connected with improving the working alliance. The working alliance as reported by the patient improved substantially after sessions including metacommunication. The psychological well-being of the patient started to improve after the working alliance was established. This is in line with the body of research from adult psychotherapy, implying that a good working alliance leads to better results, not vice versa (see Martin et al., 2000). It could be possible that, in child and adolescent psychotherapy, where evidence about the predictive value of the working alliance is not as well established (Shirk & Karver, 2003), there might be more variation in how reflective patients are about the relationship and its effects, and the treatment benefits might come from more variable sources. Also, in research, child and adolescent psychotherapy are often pooled together, even though it might be more beneficial to research adolescent psychotherapy as a different process than child psychotherapy. This view is supported, for example, by the research data suggesting that adolescent-onset depression seems to be a somewhat distinct phenomenon from child-onset depression (Shanahan et al., 2011).

The results of this study have similar implications as the study by Diamond et al. (1999): The attending psychotherapeutic alliance seemed to improve it. How the therapeutic alliance and alliance ruptures should be attended most beneficially in adolescent psychotherapy is still rarely studied. Studies of adolescents are scarce, and there does not seem to be an established research methodology. This study implies that the Rupture Resolution Rating System (3RS) could be as promising a system for assessing therapeutic alliance ruptures and their resolution on adolescents as it is on adults. It is notable that in the handful of studies that qualitatively assessed how therapists are addressing alliance ruptures in child and adolescent psychotherapy (Binder et al., 2008; Diamond et al., 1999), the classification was not very different than in the 3RS. The SRS and ORS are mainly considered to be therapeutic tools to help improve outcomes, but according to this study, they could also be used more as a quick and easy way to gather data about fluctuations in therapeutic alliances in research settings.

This study has limitations. As a single-case study, its results cannot be generalized to any other cases. Also, as the data were originally not collected for research purposes, data collection has not been done in a very systematic manner. Also, the data did not allow coding 3RS rupture markers the way they are supposed to be coded, as data included only audio recordings, and 3RS was used adaptively to identify therapist strategies for dealing with problems with alliance. However, this study might create ideas about further research on therapeutic alliances, alliance ruptures, and rupture resolution in adolescents, and the implication is that we need more qualitative and quantitative research on these matters in adolescent psychotherapy, separately from child or adult psychotherapy.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Sami Juhani Eloranta

Sami J. Eloranta Clinical psychologist and psychotherapist in a private practice in Tampere, Finland and a PhD student in Tampere University. He also teaches clinical skills at the degree program in psychology of Tampere University.