Authors: Rune Johansen (Division of Mental Health and Substance Abuse, Diakonhjemmet Hospital, Oslo, Norway), Hanne-Sofie Johnsen Dahl (Department of Psychology, Faculty of Social Sciences, University of Oslo, Oslo, Norway), Anna Grasdal Lunde (Division of Mental Health and Addiction, Institute of Clinical Medicine, University of Oslo, Oslo, Norway), Rebekka Thorkildsen Sem (Division of Mental Health and Addiction, Institute of Clinical Medicine, University of Oslo, Oslo, Norway), Mathilde Sem Thorkildsen (Division of Mental Health and Addiction, Institute of Clinical Medicine, University of Oslo, Oslo, Norway), Sarah Christine Fahs (Department of Child- and Adolescent Psychiatry, University of Oslo, Oslo, Norway; Faculty of Health Sciences, VID Specialized University, Oslo, Norway), Randi Ulberg (Department of Child- and Adolescent Psychiatry, University of Oslo, Oslo, Norway; Research Unit, Division of Mental Health, Vestfold Hospital Trust, Tønsberg, Norway; Child and Adolescent Mental Health Research Unit, Division of Mental Health and Addiction, Oslo University Hospital, Oslo, Norway)
Categories: Original Research, adolescent psychotherapy, depression, psychoanalytic, psychodynamic, transference
Source: Frontiers in Child and Adolescent Psychiatry
Authors: Rune Johansen, Hanne-Sofie Johnsen Dahl, Anna Grasdal Lunde, Rebekka Thorkildsen Sem, Mathilde Sem Thorkildsen, Sarah Christine Fahs, Randi Ulberg
Psychodynamic therapy is based on the understanding that past experiences, relationships, and unconscious conflicts can influence a person's current behavior and mental state. Transference is a central construct in psychodynamic psychotherapy, which refers to the emotions and attitudes that the patient attributes to the therapist. These emotions shape the patient's perception of both the therapist and the therapeutic process. Emphasizing and exploring how transferential phenomena manifest within the therapeutic relationship is assumed to have a beneficial therapeutic effect. The primary aim of the present qualitative study was to identify examples of such transference interventions in short-term psychoanalytic psychotherapy with adolescents diagnosed with depression.
Examples of transference interventions were identified from randomly selected sessions, as well as from two therapies with good outcomes.
Transference interventions were used to a moderate extent and were primarily aimed at encouraging patients to reveal and discuss their thoughts and feelings about the therapist and the therapy.
Transference interventions in psychotherapy with adolescents with depression seemed mostly to be preparatory and directly relevant to the young person's present, everyday life. The interventions were often posed as open question.
Psychodynamic therapy recognizes that a person’s present behavior and mental state can be shaped by past experiences, relationships, emotions, and unconscious conflicts.
Short-term psychodynamic psychotherapy (STPP) is a time-limited treatment built on general psychodynamic principles. The manualized STPP described by Cregeen and colleagues (1) consists of 45-min weekly sessions for 28 weeks. In STPP, the therapeutic relationship provides the framework for both the patient–therapist dialog and the therapist's interventions. The overall aim of the therapy is to alleviate depressive symptoms, help the young person through the developmental phase in relation to peers and parents, prepare for the transition to the workforce, and prevent psychiatric problems in adulthood. In STPP, the therapist's interventions are intended to guide the adolescent through an uncovering process, in which more of the young person's unconscious motives and fantasies are revealed.
In the United Kingdom-based RCT entitled IMPACT (Improving Mood with Psychoanalytic and Cognitive Therapies (2), STPP was implemented as a treatment modality for adolescents diagnosed with depression. STPP was compared with cognitive behavioral therapy and a brief psychosocial intervention, and was found to be as effective as the two other treatment modalities.
Transference, free association, emotions, and the unconscious are core concepts in psychoanalysis and psychoanalytic psychotherapy (PDT). Transference interventions (TIs)—including interpretation of avoidance, defense mechanisms, and resistance—are central to exploring deeper layers of the psyche and are thought to facilitate improved insight, as well as symptomatic, relational, and functional improvement. The concept of transference was introduced by Sigmund Freud, who also described it in the treatment of an adolescent (3, 4).
Throughout the history of PDT, up to the present day, the importance of transference interventions has been maintained (1). The development of transference interventions as a valuable concept, along with empirical findings of interventions across various personality pathologies, has been summarized in the literature (5, 6). These concepts are also established in STPP, but with a somewhat different focus. In Malan’s (7) original work on short-term psychotherapy, he sought to systemize psychodynamic theory into two “triangles.” The triangle of conflict has three One represents feelings or impulses, another represents anxiety, and the third represents psychological defenses. Symptoms and maladaptive patterns are understood as defensive solutions to manage anxiety about underlying feelings. The triangle of persons explains where these conflicts emerge in One corner represents important past relationships, the second represents current relationships, and the third represents the therapeutic relationship or transference. Kernberg (5) developed and empirically tested a treatment model for more severe personality disorders. Named Transference Focused Therapy, it is a highly structured, twice-weekly psychodynamic therapy that focuses almost exclusively on transference and the patient's moment-to-moment experience of the therapist (5).
The aim of transference interventions has changed throughout the history of psychoanalytic and psychodynamic therapy, and there now appears to be a greater consensus on grounding these interventions in the here-and-now therapeutic process. Definitions of the concept of transference have included a broad variation in what constitutes transference interventions. In updated work on personality pathologies, the focus in short-term therapies is largely the same as in long-term therapies. The main difference lies in the shorter time and more focused goals, with an emphasis on working with transference interpretations (8). TIs can be defined as a specific treatment technique believed to initiate a chain of events leading to insight and maturational dynamic change. Transference work (TW) refers to such transference interventions and the subsequent interaction between therapist and patient, and involves analysis of the patient–therapist relationship. TW is considered to be one of the key ingredients in PDT (4, 9, 10).
However, there is a lack of empirical data on the effect of transference interventions in PDT in adolescent populations. Establishing timely and effective treatment approaches for adolescent depression is therefore considered an important undertaking (9).
The First Experimental Study of Transference Work-In Teenagers (FEST-IT)—a randomized controlled trial of psychotherapy conducted for adolescents diagnosed with depression in Norway—demonstrated that a moderate application of TIs yielded particularly strong benefits (11). In a recent review of the evidence base for psychosocial and combination treatments for child and adolescent depression, the IMPACT and FEST-IT studies were highlighted as core contributions (30).
To explore TW in the present nested qualitative study, we used the five categories of transference interventions described in the First Experimental Study on Transference Interpretations (4, 12, 13), a study conducted on adults, with examples of transference interventions from time-limited (40 weekly sessions) PDT with adults.
Categories 1–3 may be considered preparatory TIs, primarily involving clarification and confrontation (6). These TIs aim to emphasize and explore the interaction in the therapy room and increase the patient's awareness of the therapeutic relationship. Categories 4–5 are considered proper transference interpretations, with Category 5 including genetic interpretations.
Drawing on data from the IMPACT study (“Improving mood with psychoanalytic and cognitive therapies” (2), Della Rosa and colleagues conducted a qualitative study of four adolescents patients with depression, each seen for four sessions, focusing on their responses to the use of TIs directed at the ending of the therapy. They identified two main types of response to this transference work, which they classified as “dramatizing” and “downplaying” responses. They concluded that “it may be worth considering whether transference interpretations linked to endings are more appropriate with certain kind of adolescent patients than others, and whether it is important to maintain a flexible approach to the exploration of transference in relation to endings and separations when working with adolescents” (14).
From the same study, Cirasola and colleagues (15) examined how TW affected alliance ruptures. They examined 16 sessions from four adolescent therapies and reported that TW was commonly employed and most frequently targeted immediate issues within the therapeutic relationship, rather than linking them to external relationships. Overall, this form of TW made a positive, albeit modest, contribution to the repair of alliance ruptures.
To the best of our knowledge, apart from these two papers from the IMPACT study, there are no other studies reporting on the use of TIs and their impact on in-session processes and long-term outcome in adolescent psychotherapy. Therefore, further research is needed to deepen the understanding of how TIs are applied in adolescent psychotherapy and to examine both their immediate and long-term effects.
In the present study, data were drawn from FEST-IT, a randomized controlled trial investigating the effects of transference interventions in STPP for adolescents with depression. The study design has been described in greater detail by Ulberg et al. (11, 16).
The main aim of the RCT was to investigate the effects of transference interventions in adolescents diagnosed with major depressive disorder.
The participants were 70 adolescents with an ongoing moderate to severe unipolar depression, diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV; 31). The participants were of both sexes and aged 16–18 years. Comorbidity was expected to be common among adolescents and was not an exclusion criterion. Adolescents with developmental disorders, learning disabilities, drug addiction and psychotic disorders were excluded.
Patients were referred to outpatient clinics in the southeastern region of Norway. A total of 70 adolescents were included in the study. At the start of the treatment, the two randomized groups were similar. The mean age was 17.5 years, with the majority being girls (82%). 42% of participants lived with both parents, while 49% resided with one parent or commuted between parents. All participants fulfilled criteria for severe depression, and 8% had a diagnosed personality disorder, as determined by a semistructured interview. Patients in both groups received STPP. One group received therapy with a moderate degree of TIs, while the other received therapy without TI (16). The therapy involved 28 weekly sessions of 45 min each. Assessments were conducted at pretreatment, during treatment, post-treatment, and 1-year follow-up.
Given that STPP with adolescents requires therapists to be experienced in treatment of children and adolescents, all therapists were specialists in psychiatry or clinical psychology with at least 2 years of training in psychodynamic psychotherapy. Before inclusion of participants, the therapists completed a 1-year course in psychodynamic psychotherapy, both with and without the use of TIs. Therapists treated patients in both treatment groups. Randomization was blind to evaluators and researchers; only the therapists knew whether the patients were randomized to the transference group or the non-transference group.
The level of depressive symptoms was assessed by clinicians using the Montgomery–Åsberg Rating Scale (MADRS) (17). Patients also completed the Beck Depression Inventory (BDI) (18). Both measures were administered at pretreatment, after session 12 and 20, at post-treatment (session 28), and at the 1-year follow-up.
Assessing symptoms related to diagnoses alone does not necessarily provide a comprehensive picture of the effects of therapy. Therefore, the Psychodynamic Function Scale (PFS) was also used (score range 0–100, with scores above 70 indicating good functioning) based on a semistructured interview focusing on relational and intrapsychic functioning. The PFS provides information about how patients manage their relationships with others, solve problems, and understand their own functioning, rather than focusing solely on symptoms. The PFS has been shown to be useful in therapy with adolescents (19, 20). PFS scores were obtained before treatment, after treatment, and at the 1-year follow-up.
The Transference Work Scale (TWS) is a rating tool designed to categorize and analyze transference interventions. The tool is considered particularly useful for evaluating transference work in studies combining quantitative and narrative data. The TWS consists of 26 items divided into seven subscales. These subscales aim to identify, categorize, and assess the timing, content, and valence of the TI. Finally, the patient's response to the transference work is also rated. The items are scored on a Likert scale from 0 to 4, where 0 represents “not at all” and 4 represents “very much” (13, 21). TIs are categorized on the basis of five TI categories and the TWS items are scored either “Yes” or “No.” It has been reported that interrater reliability was good to excellent for most of the 26 questions in the TWS items (13).
The main findings of the study were that transfer interventions had a positive effect on treatment outcomes at the 1-year follow-up (Cohen's d = 0.53 for BDI, d = 0.69 for MADRS). Adolescents in the non-transfer group experienced slightly faster improvement up to the 12th session. From the 12th session to the 1-year follow-up, their improvement remained stable. In contrast, patients in the transfer group continued to improve, showing further reductions in depressive symptoms up to the 1-year follow-up assessment (11).
The participants were comprehensively and reliably assessed for symptoms consistent with personality disorders (22). A small but notable additional effect of transference interventions was found in the transference group among adolescents with personality pathology (including emotionally unstable personality disorder). Patients with a higher number of personality disorder criteria at the start of treatment showed a better treatment effect, and the difference between the groups persisted up to 1 year after treatment termination. Adolescents with severe depression and comorbid cluster B personality symptoms appeared to benefit more from STPP with transference interventions than adolescents without such personality pathology (23). Participants forged healthier relationships with family and peers, cultivated a more optimistic future outlook, and reported enhanced self-perception and relational expectations—factors that appear pivotal for continued development (24). Patients who discontinued treatment early displayed heterogeneous, multidimensional dropout reasons, implicating factors such as therapeutic effort, session structure, and the quality of therapist–patient communication (25).
The aim of the present qualitative study was to identify examples of TIs in STPP with adolescents diagnosed with depression. In particular, we sought to investigate the use of TIs over the course of treatment and to analyze the patient–therapist dialog following these interventions. Examples of the five different categories of TIs are presented from clinical work with adolescents in the study, followed by extended details of the effects of transference work in two therapies.
Two experienced psychodynamic therapists listened to 20 randomly selected audio-recorded sessions 12 from every third patient in FEST-IT. They rated the use of TIs with TWS scoring on a 5-point Likert scale. Interrater agreement was calculated in SPSS. TIs were identified, transcribed, and categorized by RTS under the supervision of RU.
In addition, two good-outcome cases randomized to transference therapy and completing the full course of therapy (28 weeks) were selected to exemplify the use of TI throughout therapy. These cases were chosen based on the absence of a diagnosable major depressive disorder at the 1-year follow-up and low scores on depression BDI (1 and 3) and MADRS (2 and 4). Sessions 3, 12, 20, and 28 from each therapy were analyzed to identify TIs and patient responses, as these time points coincided with the administration of quantitative measures. Material from these sessions was transcribed and rated with the TWS to identify, categorize, and evaluate transfer work in therapy. Two medical students (AL and MT) were trained in TWS scoring by Randi Ulberg (13). The students rated the transcripts individually.
The two students also listened to all therapy sessions in both treatments to obtain a comprehensive picture of patient histories and treatment courses. This was important for assessing whether there were circumstances in the patients’ lives, or systematic differences in the use of TI, that were not captured in the TWS-rated sessions (26).
TIs were identified in 15 of the 20 audio-recorded sessions. The level of TIs was moderate in the transference group and low in the non-transference group (Table 1).
Intra-class correlation between the two raters (single measure) across the whole sample was.89 (95% CI 0.75 to 0.96, p = 0.000) (11, 27).
In the following, we will present examples of TIs from these randomly selected sessions to illustrate the five different TI categories.
We now present two vignettes to illustrate a more detailed and thorough use of TIs during the course of therapy.
Anna is a 16-year-old girl who lives in Oslo, the capital of Norway. She lives with her mother, father, brother, and their dog. As a child, she moved from the southern part of Norway to Oslo. At the time of treatment, Anna was in a transitional phase from secondary school to upper secondary school and was unhappy, especially when at school. She found it difficult to connect with peers and had few friends. During these difficult times, she struggled with self-harm and isolation. She had a MADRS score of 28 and a self-reported BDI-II score of 33. On the PFS, she was rated 47.5, which indicates serious interpersonal struggles. According to DSM IV criteria, she was diagnosed with major depressive disorder and was rated with a high level of suicidality. She also met criteria for an unspecified personality disorder (14 criteria; e.g., schizoid and avoidant traits).
In the two initial therapy sessions, Anna appeared deeply depressed. She had recently become aware that a close family member was seriously ill and might pass away in the near future. Emotions related to this sad news were explored. The focus was also on how to take care of herself in this stressful situation.
At post-treatment, Anna had an average PFS score of 73.2, MADRS of 6, and BDI-II of 0. One year after treatment, her PFS was 69.6, MADRS was 6, and BDI-II was 2.
Betty is an 18-year-old girl attending upper secondary school in a non-urban area. Her parents are divorced. At the start of therapy, she lived with her father. She described a strained relationship with her mother. In addition to this, she had a long history of difficult relationships with peers. Before treatment, she was measured with a MADRS score of 14, but rated herself with a BDI-II score of 31. On PFS, she had a score of 65. According to DSM IV, she was diagnosed with recurrent depression, major depressive disorder, panic disorder, agoraphobia, and social phobia.
Initially, therapy focused on Betty's depressive symptoms and the associated loss of functioning. In the first two therapy sessions, Betty focused on her relationship with her parents, difficulties with friendships, and issues related to sexuality.
At post-treatment, Betty had an average PFS score of 79.6, MADRS of 22, and BDI-II of 21. One year after treatment her PFS score was 80, MADRS was 4, and BDI-II was 3.
We have presented examples of different categories of TIs, as well as examples of their use in STPP with adolescents experiencing depression. Positive effects of TIs in adult psychotherapy have been documented in the research literature (6). Della Rosa and colleagues (14) discussed whether the use of TIs in adolescent psychotherapy is appropriate, suggesting that TW with adolescents should not be emphasized, as investing in the therapeutic relationship and exploring earlier relationships with parental figures—which is often the focus of transference work—may compete with adolescents’ developmental task of establishing their identity in relation to the peer group. Further, Cirasola and colleagues (15), drawing on the IMPACT study, reported that factors improving the effectiveness of TW included the therapist's ability to remain focused on transference dynamics in the here and now, as well as their capacity to affirm and flexibly adjust their use of TW techniques. In contrast, a rigid approach to TW and the fostering of dependency between the adolescent and the therapist or the therapy itself hindered the resolution of alliance ruptures.
The beneficial effects of psychoanalytic/psychodynamic psychotherapy with TIs for adolescents with depression are demonstrated in the RCT FEST-IT (11, 23). Moreover, neither the studies based on IMPACT (14, 15) nor the present study supports the view that TIs offered in STPP dwell on more theoretical and irrelevant relationship between the adolescent patient and the therapist or on past relationships. Rather, results from both RCTs exploring TIs in adolescents with depression show interventions that are directly relevant to the young person's present, everyday life.
Based on the definition of TI categories (1 to 5), the present study has provided examples of the use of TIs in research material. As Meehan (6) noted, there are different views as to whether all categories can be considered “real” TIs, or whether the lower categories function more as preparation and clarification. In this study, the five TI categories include two preparatory categories (1 and 2). Category 2 was frequently used. Category 3 was more demanding, as it encouraged the patient to reveal feelings and thoughts about what they believed the therapist thought about them. Categories 4 and 5 corresponded to interventions often understood as transference interpretations in a stricter sense (6); these categories were rare in the available clinical material.
In the extended examples from two adolescents, the therapists mainly used Category 1 and 2 TIs, only rarely Category 3, and no Category 4 or 5 TIs. The therapists often encouraged the patient to explore difficult feelings in relation to the therapist and other people outside of therapy. These findings invite discussion about differences in the use of TIs in therapeutic work with adolescents compared with adults. It may be more difficult to use more “proper” TIs with adolescents. This might reflect therapist-related explanations, and the lack of information about such influences represents a limitation in the study. In our material, it appears that even when therapists did not formulate a proper grammatical question, their interventions effectively ended in a question mark, typically conveyed through intonation. Whether this is specific to this group of therapists is unknown, but there may be cultural aspects to this rather humble way of offering interventions. Using Schwartz’s (28) cultural value orientations, Norway appears particularly high on egalitarianism and low on hierarchy, with strong norms of equality, flat structures, modesty, small power distance, and informality with authorities. In general, harmony is valued and open conflict is avoided. Interventions and interpretations presented as questions may represent a plausible microlevel enactment of these macrolevel cultural value orientations.
The complexity underlying the results should be taken into consideration. When discussing and exploring the effects of TI in time-limited youth psychotherapy, it is necessary to be clear about what definition of TI is applied and to describe the technical use of TIs (frequency and level of transference intervention). Based on the present study, no conclusions can be drawn regarding whether these effects will last over time or what impacts TIs may have in other PDT-treatment modalities (e.g., longer treatments) or with other diagnostic groups. Our study includes a limited number of STPP therapies. Furthermore, the use of TI in youth psychotherapy may be culturally sensitive, and our examples may therefore not be representative of TIs in STPP in other countries. For future studies, youth-specific instruments such as MADRS for youth (MADRS-Y) should be considered (29).
There is a need to overcome barriers to individualized care and to integrate emerging evidence into practice. The findings may expand our understanding of the efficacy of Tis in adolescent depression and support the inclusion of transference work in the design of tailored therapeutic protocols for this vulnerable population.
In the clinical material examined from a study of STPP with adolescents experiencing depression, the therapists offered TIs to a moderate degree. In the present paper, the five categories of TIs and subsequent TWs were exemplified. Most of these interventions were preparatory (Categories 1–2), encouraging the patient to reveal thoughts and feelings about the therapy and the therapist. The material also showed less use of confrontation and traditional interventions (Categories 4–5).