Therapeutic alliance and early change in depression: benefits of enhancing working alliance at the initial sessions of short-term supportive–expressive psychodynamic psychotherapy

Results: There was no significant increase or decrease in the severity of depression on the baseline (χ2 = 3.54, P=0.14) and during evaluation sessions (χ2 = 0.85, P=0.65), but participants showed a significant improvement in quality of life once the sessions had started (χ2 = 8.95, P=0.01). The mean scores on all three working alliance components showed a slight increase over three sessions and the scores on the bond subscale showed a significant increase (χ2 = 11.56, P=0.003).


INTRODUCTION
Working alliance was originally a construct emerging from the psychoanalytic approach.However, it soon went beyond psychoanalysis and became counted as a common factor in psychotherapies; it is currently emphasized by all schools of psychotherapy [1].
In psychoanalytic psychotherapy, transference is triggered in order to gain access to pathogenic materials, which are then interpreted in order to gain insight into the patient.These are the two main techniques in psychotherapy, but they are not sufficient for causing change in the patient.Greenson believes that for neurotic patients to enter analysis and effectively try to overcome their conflicts, they must experience another type of relationship with the therapist, known as a working alliance, in addition to the transference relationship [2].
Numerous studies indicate a correlation between working alliance and improved effectiveness of various psychotherapies (including individual therapies, group therapy and family therapy) [3][4][5][6][7][8].Freud considered working alliance to be a form of positive transcendence in patients, which helps them to trust the therapist's professional ability.It means that working alliance that Freud positioned within the general concept of transference was not well differentiated from other elements of transference [9].The emphasis on the existence of a relationship based on reality, in addition to the transference relationship, and a differentiation of working alliance from transference gradually emerged in the works of Alexander [10], Sterba [11], Greenson [2], and Zetzel [12].The concept of working alliance prevalent in the literature on psychoanalysis today slightly differs from Freud's viewpoint.Understanding the importance of the relationship between the therapist and the patient, Greenson held that unlike a transference relationship, working alliance is a relatively non-neurotic and rational rapport between the patient and the therapist.The rational and goal-oriented aspect of the patient's feelings towards the therapist is activated by his motivation to overcome his illness and his conscious willingness to cooperate with the therapist [2].
In order to further explain this concept, Bordin proposed a theory of the components of a working alliance.He considered working alliance to be the existence of an agreement between the patient and the therapist over therapeutic goals, the method of attaining these goals, and the humanistic bond between the therapist and the patient which allows the patient to develop trust in the therapist [13].
In classic psychoanalysis, effective working alliance usually arises gradually, after an analysis of transference resistance [2].However, in short-term psychodynamic approaches, the therapist actively helps to create a working alliance from the initial sessions and tries to create a balance between resistance and working alliance [14].One of the approaches of short-term psychodynamics, which highly emphasizes the importance of working alliance, is supportive-expressive (SE) psychoanalysis psychotherapy.Luborsky believes that attaining helping alliance, which is formed using supportive techniques, is one of the factors bringing about change during the process of psychotherapy.He maintains that patients experience helping alliance in two ways: they see the therapist as a person capable of helping them, and view the process of therapy as a joint effort with the therapist to attain therapeutic goals [15].SE highlights the therapist's efforts to improve the helping alliance, as an effective therapeutic factor, from the initial sessions.For Luborsky, there is no doubt about the importance of working alliance in the effectiveness of psychotherapy, although there are conflicting findings as well.
Numerous studies have been conducted on the influence of working alliance on therapy.Most of these studies evaluated working alliance over one or two sessions and investigated whether it had any effect on symptom improvement.However, working alliance can fluctuate over sessions, and evaluating it in only one or two sessions impairs the credibility of the research.Conflicting findings on the effect of the working alliance are partly due to this issue [16].Therefore, in order to understand the mechanism of action in working alliance, it seems that this construct must be separately examined in different phases of psychotherapy.
So far, studies have been carried out on the effect of working alliance in the initial sessions of psychotherapy.They have shown that working alliance has a stronger relationship with therapeutic outcomes in the initial sessions than in the middle sessions [17][18][19].Nevertheless, it is not clear whether the therapist's efforts in strengthening the components of working alliance from the first contact with the patient only affect the therapeutic outcomes or whether they also influence the patient's symptoms from the begin-ning.In other words, can good working alliance result in an early improvement in symptoms?Such early improvement raises hope in the patient and is one of the predicting factors of the effectiveness of psychotherapy [3,7,20].
Previously, effective components of working alliance were examined in the initial sessions of cognitive-behavioral therapy (CBT).For instance, agreement on therapeutic goals and the method of attaining them at the onset of therapy are correlated with a decrease in symptoms at the end of psychotherapy in patients with depression; however, the quality of rapport between the patient and the therapist in the initial sessions does not correlate with the therapeutic outcome [20].
In CBT, techniques for changing various symptoms are mostly used from the first session.It is clear that incorporating a technique which targets a patient's symptoms strengthens working alliance in the patient [21] and brings about changes in the patient's symptoms during the initial sessions.However, unlike CBT, there are no therapeutic techniques in analytic psychotherapies which focus on disorder symptoms, and the emphasis is mostly on the context of symptoms rather than on symptoms themselves.Compared with cognitive psychotherapists, analytic therapists do not have a tool for altering disorder symptoms in the initial sessions.Nevertheless, we should keep in mind that working alliance in itself is a therapeutic factor, and strengthening it in the initial sessions may partly compensate for this limitation and independently bring about changes in patients' symptoms.
The present study aimed to establish whether the use of supportive techniques and improving working alliance in the initial sessions of short-term SE psychotherapy result in a change in symptoms in patients with a depressive disorder.

Study design
The results reported here are part of the findings of a single-case experimental study on the effectiveness of SE psychotherapy on the components of vulnerability to depression.
The study was conducted on 6 female outpatients with major depressive disorders, who had all volunteered to participate in the study between April 2015 and June 2015 in a private psychotherapy clinic.In order to control the effect of gender and other psychiatric disorders, all participants were women; they did not have psychotic symptoms, comorbidity of other Axis I disorders, or personality disorders; and none of them were receiving pharmacotherapy.Four had previous episodes of depression.Two subjects had received psychotherapy, and one had received medication during previous episodes.None of the subjects were treated with medications or other psychotherapy at the time of the study.
The process was explained and written consent was obtained.Participants were selected using the Structured Clinical Interview for DSM-IV.After an initial evaluation, target variables were measured over three consecutive weeks (baseline 1,2 and 3).Then, therapeutic -intervention was commenced for all participants and changes were measured over the first three sessions (three weeks) of psychotherapy (assessment 1, 2, and 3).In addition, to working alliance, depression severity symptoms and the participants' quality of life were examined.

Psychotherapy and psychotherapist
The therapeutic approach used was short-term SE psychotherapy.SE is a modern analytical psychotherapy, which includes many major aspects of a psychodynamic therapy [22] and usually lasts for 16 to 24 sessions.It helps patients to understand the pattern of dysfunctional relationships recurring in most of their relationships.It greatly emphasizes the importance of working alliance -the term "supportive" in SE refers to techniques utilized for improving the alliance.The initial phase of this therapy was allocated to determining therapeutic goals, forming a therapeutic relationship, and evaluating the pattern of interpersonal relationships [15].In these sessions, almost no interpretation or therapeutic intervention was made, but numerous supportive techniques were employed with the aim of forming a working alliance.The therapists use these supportive techniques accord-ing to SE psychotherapy treatment guidelines: they support the patient's wish to achieve their goals, and convey a sense of understanding and acceptance of the patient; they help the patient maintain vital defenses and activities which bolster their level of functioning, and give a realistic view about the treatment goals that are likely to be achieved [15].This phase usually lasts two to five sessions.In the present study, this phase lasted at least three sessions for all participants.Sessions were held once a week.
Intervention sessions were conducted by a therapist with a doctoral degree in clinical psychology.This therapist had three years of experience in practicing SE psychotherapy.The intervention sessions were supervised by a person with a doctoral degree in clinical psychology with full psychoanalytical training and a 10-year experience in SE psychotherapy.

Measures
Changes in the severity of depression symptoms and the patients' quality of life were evaluated using the Beck Depression Inventory (BDI-II) and WHOQOL-BREF, respectively, over three consecutive weeks, comparing the baseline with the end of each session in the initial phase of the intervention.BDI-II, created by Aaron Beck, consists of 21 questions and measures the severity of emotional, affective, motivational, cognitive and physical symptoms of depressive disorder, with scores ranging from 0 to 63.The internal consistency, split-half validity and test-retest reliability of the Persian version of this instrument were reported at 0.85, 0.81 and 0.73, respectively.Moreover, correlations between BDI-II and depression measured by the Hamilton Rating Scale and the Oxford Happiness Questionnaire were reported to be 0.66 and -0.55 [23].
The short version of the World Health Organization Quality of Life scale (WHOQOL-BREF) is a frequently used instrument for self-evaluation of the quality of life.This instrument comprises 29 questions and examines quality of life in four health-related domains.It has been standardized for Iranian society, with reliability and validity measures indicating the acceptability of its structural factors [24].
We also administered the Working Alliance Inventory-12, which is one of the most common instruments for investigating changes in patients' working alliance in the initial phase of therapy.This instrument consists of 12 items and measures three factors: goals (agreement between the patient and the therapist on therapeutic goals), tasks (agreement on the method of attaining the therapeutic goals), and bond (quality of the therapeutic relationship), with higher scores indicating a stronger therapeutic alliance.The reliability and validity of this instrument have been confirmed [25].

Statistical analysis
A common statistical method used in singlecase studies is visual analysis of the graph of changes.In the present study, the significance of the trend of changes in participants was also examined.To this end, considering the small sample size and a violation of the basic assumptions of ANOVA, the Friedman test was employed.Finally, statistical analysis was conducted using the SPSS software for Windows (version 23).

RESULTS
Table 1 presents demographic information of the participants as well as target variables on the baseline and after the intervention sessions.Table 2 shows changes in the scores of depression severity and quality of life during the study, as well as components of the working alliance during the intervention in the six participants.The results of the Friedman test indicated no significant change in the severity of depression on the baseline (χ 2 = 3.54, P=0.14) and during evaluation sessions (χ 2 = 0.85, P=0.65).Moreover, no changes were observed in the quality of life on the baseline (χ 2 = 0.3, P=0.86), while it was significantly higher during evaluation sessions (χ 2 = 8.95, P=0.01).The mean ranks of quality of life scores were 1.25, 1.83 and 2.92 in the first, second and third sessions, respectively.The results of the Wilcoxon post-hoc test indicated a significant difference between the first and third evaluation sessions (Z = -2.88,P=0.012).
Table 2 presents working alliance mean scores during the intervention sessions.The mean scores of all three components showed a slight increase over three sessions.The results of the Friedman test revealed that changes in task (χ 2 = 0.5, P=0.77) and goal scores (χ 2 = 1.44,P=0.48) were not significant.However, scores of the bond sub-scale showed a significant increase (χ 2 = 11.56,P=0.003).The mean ranks of bond scores were 1.08, 1.92 and 3 in the first, second and third sessions, respectively (Z = -3.32,P=0.003).

DISCUSSION
It was clearly observed in this study that despite a slight increase in the severity of depression in at least four participants, patients' quality of life and working alliance, especially the bond component, were improved during the initial sessions of SE psychotherapy.
Strengthening working alliance in the initial phase of therapy can be affected by an early change in disorder symptoms [26].However, almost no decrease was observed in depression symptoms in this study.Therefore, the improvement observed in the therapeutic alliance is probably a result of utilizing supportive techniques.This finding is in line with other studies, which had shown a relationship between the therapist's technique (such as empathic relationship, cooperative interaction with the patient, and explaining evaluation results to the patient) and an improvement of working alliance in patients, and their sense of depth and a positive assessment of the evaluation sessions [27].A systematic review of 25 studies on the influence of a therapist's technique on the formation of working alliance indicated that using techniques which transfer feelings of supportiveness, interest, trustworthiness and flexibility to patients improves the patients' understanding of their problems as well as the working alliance [28].
Among the components of the working alliance only the mean score of task showed a slight decrease.This finding contradicts the results of a similar study on cognitive therapy, which showed improvements on the task and goal components during initial therapy sessions [20].This contradictory result may be due to differences in the two approaches.Most cognitive techniques are planned as focused entirely on the patient's symptoms, and the patient clear-ly understands this logic.However, in analytic therapies, it is difficult for patients to understand the process of treatment and its relationship with their symptoms, at least in the initial sessions.Moreover, in contrast to CBT, working alliance in psychodynamic psychotherapies is more than a prerequisite for the effectiveness of treatment; psychodynamic therapists help patients re-experience suppressed emotions in a communication space based on the working alliance [29].So, psychotherapy interventions are merged with working alliance and the findings of this paper are consistent with the hypothesis that therapeutic alliance is more than just a prerequisite for the effectiveness of treatment.But what is the mechanism of this impact?
The mean scores of depression severity saw no decrease on the baseline and during the initial psychotherapy sessions in any participant (except for one score in one participant), and no early change occurred in the depression symptoms.Nevertheless, the quality of life scores saw a significant increase.The mean quality of life scores of five participants were above the baseline.The feeling of improvement in quality of life and satisfaction with life, in addition to strengthening the emotional relationship and attachment to the therapist, is a noteworthy finding.Previous studies had revealed the influence of a positive relationship with the therapist in improving the quality of life in patients with mental health [30] but also those with physical health problems [31,32].According to these studies, if patients experience a strong therapeutic relationship with their therapists, they are more likely to benefit from the advantages of therapy in everyday life and have a higher general satisfaction with their lives [33,34].Based on these explanations, there is no doubt about the importance of paying attention to the working alliance in working with depressed patients, whose mental suffering is partly caused by social isolation and loneliness.Previous studies indicated that working alliance determines approximately 15% of the variance of the outcome of therapy in depressed patients [16], which is more than the effect of other factors, including therapist adherence or competence on the outcome of psychotherapy [20].
What the current study adds is that patients can also benefit from strengthening working al-liance in the initial sessions.Regarding this, we can hypothesize that strengthening working alliance in the initial sessions, even if limited to the bond component, can improve a sense of satisfaction with life in depressed patients; the change that seems to be related to experience is an empathic and trustworthy relationship.Zilcha-Mano [35], explaining the possible mechanisms of the therapeutic effect of alliance, believes that changes in the person's interactions and changes in his interpersonal world affect his wellbeing.Although this study did not show any improvement in depression symptoms during the early phase of treatment, it did show that strengthening the working alliance in the early sessions of treatment can be linked to an improved wellbeing and better quality of life in depressed patients.

LIMITATIONS
The present research was a single-case study.One of the most important limitations of such studies lies in the generalization of their results to other patients and psychotherapies.Singlecase studies are used solely for observing changes in participants, and it is not possible to extrapolate the effects of variables to other patients.As a result, the findings of this study must be interpreted with caution.The single-case method is an appropriate approach for psychoanalytic studies; it examines the results of clinical interventions and changes in patients in different phases of the therapy and, despite lower associated costs than a randomized-controlled trial (RCT), it can answer questions related to the outcomes of psychotherapies [37].Examining everyday therapeutic interventions of analytic psychotherapists through single-case studies can considerably enrich the available information on the mechanism of change in these psychotherapies.

CONCLUSIONS
Psychoanalytic studies have seen significant improvements since Sigmund Freud first introduced the term.However, more studies are needed on the mechanisms of change and var-iables in these therapeutic approaches.It is believed that no evidence-based explanation can as yet be proposed as to how and why analytical psychotherapy causes a change in patients [36].A noteworthy finding of the present study for analytical psychotherapists is that although there is no possibility for intervention or a quick change in symptoms in the initial sessions of most short-term analytical therapies (akin to what happens as a result of behavioral activation in the initial sessions of CBT), improving working alliance in the initial sessions and creating an environment in which the patient feels support and empathic understanding, can lead to a change in the patient's mental experience of quality of life.This feeling of improvement is different from the early improvement in symptoms, but it may influence treatment outcome and enable pre-term termination of therapy.More studies are required to confirm these hypotheses.

Figures 1 and 2 3 Figure 1 . 3 Figure 2 .
Figures 1 and 2 illustrate changes in BDI-II and quality of life scores of the six participants, respectively.Based on these figures, it is clear that participants 1, 2, 4 and 5 showed a slight in-

Table 1 .
Demographic information and outcome measures of clinical sample BDI-II, Beck Depression Inventory; QOL, quality of life.

Table 2 .
Descriptive analysis of variables