A recent study, led by Dr. Andrzej Werbart, interviewed seven psychoanalytic therapists about their experiences in therapy with young adult patients who did not improve. The results, published in Psychotherapy Research, demonstrate that therapists experience a “split picture” of the “nonimproved” patient and, in their confusion, struggle to strike a balance between distance and closeness which results in an overall sense of lost control in the therapy process.
“The more the therapist tried to deepen the relationship to help the patient reach her unformulated experiences, the more the patient withdrew, adding to the therapist’s confusion, therapists reported. Over time, the therapists became frustrated and unable to find a way to move further with their patients, being stuck in a struggle.”

There is extensive evidence suggesting psychotherapy is effective for a wide-range of mental health concerns. Yet, a substantial number of clients do not experience improvement in psychotherapy. Worse still, some experience what is referred to as “deterioration” during treatment, or a worsening of their distress.
While evidence suggests that therapists can improve outcomes by identifying instances of unsuccessful treatment, research also demonstrates that it is difficult for therapists to detect their treatment failures. Werbart and colleagues highlight the importance of understanding the specific factors that lead to treatment improvement, stagnation, or deterioration, thereby justifying the merit of this project examining therapists experiences of “unimproved” psychotherapy cases.
Using quantitative inclusion criteria and qualitative analysis techniques, the researchers pose the following questions in this study:
- “How do therapists describe their work in these particular cases and themselves as the particular patient’s therapist?”
- “How do the therapists describe their patients, the therapeutic relationship, and the therapy outcome?”
- “Which factors and processes seem to have been crucial for the unsuccessful outcome from the view of the therapist?”
They also investigated “if there are any particular characteristics of the therapists’ experiences already observable at the outset of treatment, as reported in baseline interviews.”
Archival data from the Young Adult Psychotherapy Project (YAPP) was used. This study featured naturalistic, longitudinal data of psychoanalytic psychotherapy with young adults (who were mostly self-referred) in Stockholm, Sweden. The patients participated in therapy for a mean of 22.3 months, and outcome data were assessed at termination, at 1.5-years, and at a 3-year follow-up.
The researchers used selection criteria to exclusively study cases of non-improvement, defined as “patients who reported a symptom level in the clinical range at pretreatment, and no reliable symptom reduction or even deterioration at termination.” The cut off between nonimproved and clinically improved cases was guided by the Global Severity Index (GSI) outcome measure. Psychotherapists were recruited from a previous study that examined patients’ perspectives of their nonimproved psychotherapy experience.
In this study, eight patients worked with seven psychotherapists. Four of the seven therapists were female, and three were male. The average age of the therapist sample was 52.9 years.
After engaging in semi-structured interviews with therapists and coding therapist accounts through grounded theory, the researchers identified a core category that connected additional subcategories and domains. This core category, titled “Having Half of the Patient in Therapy,” was created based upon direct quotations from the interviewed therapists.
The core category “Having Half of the Patient in Therapy,” captured a process in which the therapist experienced parts of the patient as obscure or absent at the beginning of therapy—that essential parts of their life circumstances or problems were excluded in their presentation. Despite initially feeling drawn into the patient and their story, the therapist felt unable to bridge the distance. This ultimately resulted in a perceived sense of the therapist’s loss of control in the therapy process and an inability to achieve a favorable balance of closeness and distance.
Nine subcategories are mapped around this core category. These nine subcategories were gathered into two thematic domains: 1) Experiences of the therapeutic process and 2) Experiences of therapy outcomes.
Experiences of the therapeutic process
The therapists initially described being interested and involved in work with the patient. They did not struggle to locate empathy for the patient and experienced the patient as extraordinarily verbal, capable, and fond of therapy, despite having a traumatic background. Werbart and co-authors write:
“The therapist experienced herself as unusually alert, free or creative, and could describe a liberating feeling that things will turn out and clear up with time and that they were working successfully. It was easy to like the patient and feel empathy, and the therapist felt important, as the patient dared to open up and show confidence in an unusual way.”
From the beginning, therapists felt a distance from the patient, or from parts of the patient’s problem, that increased over time. Therapists understood the patients as having difficulties with allowing for closeness in their professional relationship and interpersonal relationships outside of therapy. One therapist participant stated: “She has a certain distance; she has a hard time letting her feelings flow and expresses nothing strong about her bond to me either.”
Additionally, they described the patient as exhibiting a growing aversion to closeness, as depicted by this therapist’s statement:
“The more therapy meant to her, and the more I meant to her, the worse, the more dangerous the therapy became for her and the more she needed to turn me into a no-body. During that time I was functioning pretty much as an extension of the furniture in the room, I was part of the fitting-up, so to say, I was not a living person.”
Therapists perceived the patients as becoming increasingly threatened by questions, confrontations, and interventions, resulting in “fruitless battles” and unsuccessful attempts at collaboration. Ultimately, the therapist reports experiencing a loss of control in the process, feeling perplexed, overinvolved, and more apt to forego their professional stance.
Experiences of therapy outcomes
The therapists in the sample described believing that the patient gained increased insight into their life, despite the unimproved outcome of the case. They felt that the patient’s symptoms decreased in strength and that patients acquired new ways to manage their life.
Additionally, they reported some favorable therapy outcomes such as perceiving the patient’s improved trust in the therapeutic relationship as well as a change in interpersonal functioning. However, therapists understood the patient’s core problems to remain at termination, and that the perceived improvements had not led to substantial change.

“The right side of Figure 1 represents therapy outcomes, as generally described by the therapists at termination. Therapy resulted in increased insight and Mitigated problems, thus leading the therapists to conclude that the therapy had Favorable outcomes. At the same time, all therapists concluded that the patient’s Core problems remain. There is a remarkable lack of any connection or interaction between these irreconcilable outcome subcategories.”
These results were interpreted by the researchers alongside the results of their sister study, wherein they examined patient perspectives of unimproved psychotherapy, captured by the title “Spinning One’s Wheels.” While patients experienced a too passive therapist, therapists described a half-present patient.
“Thus, there is a marked difference between the patients’ and the therapists’ experiences of the therapeutic process in cases of nonimprovement. We interpreted experiences of nonimproved patients as an unbalanced therapeutic alliance, with a good-enough emotional bond, but not enough agreement concerning therapy goals and tasks,” the researchers write.
Werbart and colleagues continue to describe that the patients were likely to have experienced difficulty approaching and bringing up emotionally laden subjects, and, in the end, therapists were unable to help them manage this. In this way, the researchers discuss how therapists may have overestimated the patient’s functioning while underestimating the scope of the patient’s problems.
“Both patient and therapist described from their different viewpoints that the therapist did not understand the patient, which might have added to the patient’s experience of an artificial relationship.”
Further, the therapist struggled to adapt their approach to the patient’s tailored needs and level of functioning, despite their active attempts to meta-communicate, perhaps as a result of their overestimating the patient’s functioning. Therapists, convinced they needed only more time, or more work, but were generally on the right track, may not have considered adopting a new understanding or focus.
Werbart and coauthors explain how this false conviction leads therapists to potentially interpret unsuccessful therapy as patient resistance:
“They do not attribute the limited progress in therapy to their own limited understanding of the patient’s problems, but rather to the patient’s lack of will to open up and try harder. Taken together, this resulted in an inability to adapt their technique and to address their interventions to the patients’ core problems.”
While the patient described their interest and involvement in the case, this may have compromised their ability to maintain an effective balance of closeness and distance successfully. The researchers hypothesize that “it possible that the therapists’ restricted awareness of their countertransference contributed to difficulties in taking a ‘third position’ together with the patient and to challenge the patient’s pseudo-mentalization.”
There are several implications of this research. Two notable highlights by Werbart and colleagues center around the prevention of suboptimal therapy outcomes. First, they assert that therapists must be observant of contradictions and incompatibilities in their early assessment of patients, the therapeutic relationship, and the therapy process.
Specifically, when therapists experience positive and stimulating collaboration in conjunction with feeling a distance from the patient, this may be an indication that the therapist is not entirely in touch with the patient’s functioning and experiences.
“If the therapist one-sidedly focuses on the more well-functioning parts, there is a risk of no therapeutic change. This kind of incompatibilities and split tendencies in the therapist’s experiences may be difficult to recognize by novice therapists and experienced therapists alike, and has to be addressed in psychotherapy training and supervision,” Werbart and researchers write.
Finally, they encourage continuous assessment of patient functioning throughout the therapy process to inform and guide therapists’ approaches and adaptations of interventions. The therapist must also be willing to reconsider their initial assessment.
****
Werbart, A., von Below, C., Engqvist, K., & Lind, S. (2018). “It was like having half of the patient in therapy”: Therapists of nonimproved patients looking back on their work. Psychotherapy Research, 1-14. (Link)