Transference interpretation and psychotherapy outcome: a systematic review of a no-consensus relationship

Meltem Yılmaz,1 Kutlu Kağan Türkarslan,2 Ludovica Zanini,1 Dilara Hasdemir,3 Grazia Fernanda Spitoni,1Vittorio Lingiardi11Department of Dynamic and Clinical Psychology, and Health Studies, Sapienza University of Rome, Italy; 2Department of Psychology, Atılım University, Ankara, Turkey; 3Steve Hicks School of Social Work, University of Texas at Austin, TX, USA
Research in Psychotherapy: Psychopathology, Process and Outcome 2024; volume 27:744; doi:10.4081/ripppo.2024.744

2024 Apr 18

ABSTRACT

Despite its well-established importance in psychoanalytic theory, there is a scarcity of empirical evidence on the relation-ship between a therapist’s transference interpretation (TI) and therapeutic outcome. The current scientific literature shows no consensus on the existence and nature of such an association. Therefore, the present study aimed to systematically review the literature on the link between TI and outcomes in psychody-namic psychotherapies. The American Psychological Associa-tion PsycInfo, MEDLINE, and the Web of Science Core Collection were selected as the primary databases for the litera-ture search. Studies were included if they measured the fre-quency/concentration of TIin psychodynamic psychotherapy [e.g., transference focused psychotherapy (TFP), supportive-ex-pressive therapy] orcompared a treatment group (e.g., high in TI and TFP) with a control group (e.g., low in TI supportive ther-apy) in an adult population with psychiatric symptoms. Out of 825 retrieved abstracts, 25 articles (21 studies) were included in the final synthesis. 13 out of 21 (62%) studies showed a signif-icant improvement in at least one therapy outcome measure fol-lowing the use of TI. The present systematic review also revealed high heterogeneity across studies in terms of TI meas-urement, outcome assessment (e.g., psychiatric symptoms, dy-namic change, interpersonal functioning, therapeutic alliance), study design (e.g., experimental, quasi-experimental, naturalis-tic), patient population (e.g., anxiety disorders, personality dis-orders), and types of treatment (e.g., TFP, supportive-expressive therapy), preventing researchers from asserting solid conclu-sions. The results strongly highlight the urgent need for high-quality research to understand which types of patients, how, and when TIs could be effective throughout the therapy process.

Key words: transference, interpretation, psychodynamic psy-chotherapy, outcome.

Introduction

As one of the most frequently applied and empirically sup-ported psychotherapies, psychodynamic psychotherapy has been shown to be an effective treatment for various psychopatholo-gies and personality disorders (Leichsenring & Rabung, 2011; Shedler, 2010; Steinert et al., 2017). Psychodynamic psy-chotherapy comprises several essential techniques, including transference interpretation (TI), which has been regarded as a fundamental mechanism that brings about a change in patients’ psychological functioning (Cooper, 1987; Freud, 1912; Gab-bard, 2004; Hobson & Kapur, 2005; Leichsenring et al., 2006).

Freud was the first clinician who described transference as “new editions or facsimiles of the impulses and phantasies which are aroused and made conscious during the progress of the analysis; but they have this peculiarity, which is character-istic for their species, that they replace some earlier person by the person of the physician. To put it another way: a whole series of psychological experiences are revived, not as belonging to the past, but as applying to the person of the physician at the present moment” (Freud, 1905, p. 116). Despite the lack of con-sensus on the current definition of transference within the psy-choanalytic and psychodynamic fields, transference, in its most general form, refers to the unconscious repetition and projection of patterns of impulses, feelings, thoughts, and behaviors toward the therapist that were continuously experienced during the in-teraction with significant early others (Hobson & Kapur, 2005; Levy & Scala, 2012).

Based on the assumption that pervasive enactment of past relational dynamics within the therapeutic relationship could provide access to the unconscious, Freud and the psychoanalysts who followed him stressed the importance of the analysis and working through of transference for the success of a psychoan-alytic treatment (Freud, 1912; Klein, 1952; Langs, 1973; Zetzel, 1956). Freud (1917), following his acceptance of transference as a key element of analysis rather than an obstacle, argued that transference should not be interpreted before it becomes a re-sistance in the treatment process. According to Strachey (1934), “mutative interpretation” should make patients aware that the fantasy object (e.g., projected and distorted perception of the therapist based on prior representations) and the real analyst (e.g., the therapist with their actual behavior and attitude) differ from each other.

In later years, Winnicott (1949) highlighted the necessity of the therapists’ recognition of their countertransference, including negative feelings such as hate, as it is of utmost importance to capture and interpret both positive and negative transference rather than defending themselves against their negative feelings by denial or reaction formation (e.g., flexing the analytic frame). On the other hand, Klein (1952) pointed out that transference could be observed in all therapeutic material (“total situations”) brought by patients, including everyday life narratives, con-scious associations, or unconscious projections, from which therapists can extract information on how patients act out in transference. Klein also favored the use of early TIs by targeting primitive unconscious fantasies and the split part-objects inter-nalized in early years, which are then projected onto the analyst, as she purported that transference starts to be formed from the very first moment of the encounter with the therapist. On the contrary, Kohut (1971) was opposed to early TIs as they could hamper the full development of self-object transferences. Kohut (1984) further argued that TI must be a two-step process com-prising an empathic understanding of inner dynamics and their genetic interpretations.

Contemporary psychoanalysis and psychoanalytic therapies continue to hold the centrality of transference and TI (Banon et al., 2013; Gabbard, 2004; Kernberg et al., 2008). For example, Kernberg et al. (2008) developed transference focused psy-chotherapy (TFP), which aims to help patients identify and ad-dress the emotional and interpersonal dynamics that contribute to borderline personality structure (i.e., identity diffusion, im-mature defensive functioning, distorted perception of reality) and develop more integrated and adaptive ways of thinking, feel-ing, and behaving. TFP argues that patients with borderline per-sonality disorder (BPD) have polarized (good versus bad) views of both themselves and others, which eventually lead to mal-adaptive behaviors (e.g., impulsivity, demandingness, amend-ments to negate painful mental states). During therapy, patients are expected to enact their problematic dyadic perceptions in their interactions with the therapist, who interprets these inter-actions progressively so that patients can integrate their conflict-ing views of themselves and others.

Empirical literature on transference interpretation

Since the 1970s, psychodynamic psychotherapy process and outcome research have investigated the effects of TI via natura-listic, quasi-experimental, and experimental studies (Høglend, 2014; Levy & Scala, 2012). TI is proposed to be an active agent in facilitating therapeutic alliance, managing patients’ feelings and thoughts, and working on their psychic organization and identity integration (Crits-Christoph & Gibbons, 2021; Gabbard & Horowitz, 2009; Giovacchini, 1979; Joseph, 1985; Tyndale, 1999). Experimental studies, such as the First Experimental Study of Transference Interpretations (FEST) (Høglend et al., 2006), opened up new horizons in psychotherapy research re-garding their rigorous study design and thought-provoking re-sults. FEST investigated the effects of TIs on outcome variables by comparing patients who received frequent TIs to patients whose therapists were requested to refrain from interpreting transference during one-year psychodynamic psychotherapy (Høglend et al., 1993, 2008; Ulberg et al., 2021). On the other hand, observational studies assessed the frequency or proportion of TIs given during different phases of therapy sessions prima-rily by using standardized rating scales (e.g., therapist interven-tion rating system, transference work scale) (Ogrodniczuk et al., 1999; Piper et al., 1986).

Existing studies exploring TI and its effects have mainly as-sessed psychodynamic functioning, psychiatric symptoms, thera-peutic alliance, interpersonal functioning, and the number of dropouts as outcome measures. The specific characteristics of pa-tients (e.g., the quality of object relations), quantity and quality of TIs, immediate patient responses, and therapist-patient relationship have been found to be important factors impacting the potential of TIs to create change (Banon et al., 2001; Piper et al., 1991).

Another highlight of TI-outcome research is that TI is a “high-risk, high-gain phenomenon”, meaning its interpretation is likely to follow two paths: an increase in patient defensiveness along with ruptures in the therapeutic alliance or an increase in insight and relational functioning (Gabbard et al., 1994; Hersoug et al., 2014; McCullough et al., 1991). For example, frequent interpretation of transference is shown to bring a less favorable outcome as well as a weakened therapeutic alliance across dif-ferent forms of analytic therapies, including ones with high-level personality organization patients (Crits-Christoph & Gibbons, 2021; Luyten et al., 2012; Piper et al., 1991). In contrast, the main results of the FEST revealed both treatments to have sim-ilar effects on the outcome at termination and 3-year follow-up. Moderator analyses showed that patients with low-quality of ob-ject relations and/or personality disorders benefited from unique positive effects of TI compared to patients with high-level per-sonality structure (Hersoug et al., 2014; Høglend et al., 2006; Høglend et al., 2008).

Lastly, the disaccord observed in research findings might result from the wide variety of factors that might influence the link between TI and outcome, such as its frequency, intensity, accu-racy, timing, valence, and content (Ulberg et al., 2014). For instance, it is known that therapists tend to increase the frequency of their TI when patients are more defensive and when a rupture is experienced within the relationship, which does not resolve the strains in the alliance (Høglend, 2004). Another reason would be the difficulty of making an empirically succinct defi-nition of TI. Hobson and Kapur (2005, p. 281) suggested three distinguishing characteristics among TIs employed in the re-search studies: “(a) how the interpretations are anchored, (b) to which features of patient-therapist interaction they are directed; and (c) the kind of patient-therapist engagement they appear to foster”. It is therefore suggested to remain cautious when mak-ing generalizations on the therapeutic effectiveness of TIs; what is considered a TI may vary in different empirical studies.

Aim of the systematic review

Studies carried out so far have provided conflicting evidence concerning the relationship between TI and outcome variables, leading to confusion for empirically informed clinicians. To date, there have been prior efforts to review the literature on the relationship between interpretations of any kind and outcome (Antichi et al., 2022; Crits-Christoph & Gibbons, 2021; Zilcha-Mano, Fisher, et al., 2023; Zilcha-Mano, Keefe, et al., 2023), andTI and therapy outcome (Brumberg & Gumz, 2012; Hø-glend, 2004, 2014; Levy & Scala, 2012). However, to the best of our knowledge, the current study constitutes the only and the most up-to-date (Brumberg & Gumz, 2012) review of the liter-ature employing a rigorous systematic approach with a specific focus on objectively measured TI (Crits-Christoph & Gibbons, 2001; Høglend, 2004, 2014; Levy & Scala, 2012). As a response to the pressing need to comprehensively and systematically sum-marize and describe the current literature on the relationship be-tween TI and outcome variables, the present systematic review aims to contribute to efforts to close this gap in the literature.

Methods

Information sources and search procedure

The systematic search of the current review was conducted on February 28, 2022. Due to the time gap between the conclu-sion of our systematic search and the submission of the study, the articles published after this date were hand-searched and as-sessed for eligibility (Diamond et al., 2023). The American Psy-chological Association PsycInfo, MEDLINE, and the Web of Science Core Collection were selected as the primary databases for the literature search. The following words were entered as search terms: ((transference OR patient-therapist relation* OR therapist-patient relation*) AND interpretation AND (psycho-dynamic OR psychoanalytic* OR dynamic* OR analytic* OR supportive-expressive) AND (psychotherapy OR treatment OR therapy) OR transference-focus*). During the identification phase, the ‘abstract’ option was selected in search engines.

Prior systematic reviews and meta-analyses on TI were also checked through PROSPERO and the Campbell Collaboration to access the latest compiles of relevant information. Through the Cochrane Central Register of Controlled Trials, further clin-ical trials were checked. Lilliengren’s list (2017), a regularly up-dated compendium of psychodynamic clinical trials, was also consulted specifically for controlled trials investigating psycho-dynamic psychotherapies. The database for unpublished study searches (e.g., doctoral dissertations, conference papers, and preprints) was selected as Proquest. Four review articles were especially helpful in tracking the previous theoretical and em-pirical work on TI (Brumberg & Gumz, 2012; Høglend, 2004, 2014; Levy & Scala, 2012). Prominent researchers in the field of transference and its interpretation were contacted to identify additional applicable studies. A bibliographic review of the in-cluded studies was performed to avoid missing any relevant studies.

Eligibility criteria

During the full-text reading, the articles were selected based on the inclusion and exclusion criteria. The inclusion criteria for the current systematic review were as follows: publication date from 1970 onward, written in English, inclusion of patients older than 18, inclusion of more than ten subjects in the study, and in-vestigation of a psychodynamic therapy of psychopathology other than psychosis with its relation to therapy outcome. Naturalistic and experimental designs (e.g., quasi-experimental), but not case studies and qualitative designs, were included. Studies with unsuitable publication types (e.g., theoretical papers, man-uals, review articles) were extracted. When there were multiple publications based on the same research study, only one study that included a detailed description of the study design was in-cluded and presented in the review (e.g., eleven articles from the FEST Study between 2006 and 2020).

Screening, selection process, and data extraction

The screening and selection of the articles were performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Page et al., 2021). The systematic review process is summarized in detail in Figure 1. Following the acquisition of the complete list of articles identified with the search terms, duplicates were removed both manually and viaZotero. The identification of relevant studies was performed by one reviewer (DH) to winnow down the large pool of studies.

Abstracts were then screened, which was followed by the retrieval and selection of articles based on the eligibility criteria. Abstract screening and selection of eligible articles were inde-pendently undertaken by two authors (DH, KKT, and MY, KKT, respectively), who reached substantial inter-rater agreements with Cohen’s κ=.78 and κ=.77. In cases of disagreement regard-ing the inclusion of a publication, reviewing authors discussed their decisions until they reached a consensus. If not possible, a third reviewer was involved to reach a final decision.

Two reviewers (MY and KKT) extracted the following data reported in the eligible studies on an electronic sheet: authors, publication year, search strategy (i.e., via a search engine, hand search), title, study design (experimental control), sample size, gender proportion, mean age (or age range, depending on the data availability), quality index, patient population, number of therapists, intervention type, treatment frequency, treatment du-ration, the measure of transference, operationalization of trans-ference, outcome measure, and results.

Assessment of the quality of included studies

The assessment of study quality was performed by one of the authors (LZ) using the Newcastle-Ottawa Scale, through which a quality index ranging from 0 to 7 was obtained. Details on the ap-plied criteria and the quality indices derived for each study can be found in Supplementary Tables 1 and 2, respectively.

Results

The first search strategy identified 1593 potentially eligible articles; following the removal of duplicates, 825 studies re-mained. Through the application of inclusion and exclusion cri-teria, a full-text assessment was carried out on 49 articles. Of these, 19 articles met all inclusion/exclusion criteria; 6 further publications were identified through a hand search. To avoid redundancies, articles that presented previously published data were considered one study. The present systematic review is therefore based on the data derived from 21 studies (25 articles in total); Table 1 provides an overview of their characteristics (Clarkin et al., 2001; Clarkin et al., 2007; Connolly et al., 1999; Doering et al., 2010; Fischer-Kern et al., 2015; Hoglend et al., 1993; Hoglend et al., 2006; Klein et al., 2003; Levy et al., 2006; Malan, 1976; Marmar et al., 1989; Marziali, 1984; Milbrath et al., 1999; Ogrodniczuk et al., 1999; Perez et al., 2016; Piper et al., 1986; Piper et al., 1991; Piper et al., 1999; Ryum et al., 2010; Sahin et al., 2018; Schut et al., 2005). In the following sections, all included articles have been described according to the general charac-teristics of the studies, characteristics of the interventions, measurement of transference, measurement of outcome, and type of experimental control.

General characteristics of the studies

A total of 21studies examining the relationship between TI and therapy outcomes were retrieved; of these, three included data also presented in other articles (Doering et al., 2010; Fis-cher-Kern et al., 2015; Høglend et al., 2006).

Figure 1. Study selection procedure based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement.Non-commercial use only.

The publication dates of the included studies ranged from 1976 to 2023, and the quality indices varied between 4 and 7. In total, 1013 participants were included in the current system-atic review, 832 of whom were female. Participants’ ages ranged from 18 to 65 years, and the sample size ranged from 10 to 106 participants. In terms of sex distribution, the majority of the studies included both males and females, whereas only five stud-ies exclusively recruited women (Clarkin et al., 2001; Doering et al., 2010; Fischer-Kern et al., 2015; Milbrath et al., 1999; Perez et al., 2016). In four out of the five female-only studies, the sample was comprised of females with a diagnosis of BPD (Clarkin et al., 2001; Doering et al., 2010; Fischer-Kern et al., 2015; Perez et al., 2016); the remaining study recruited females with normal or pathological grief (Milbrath et al., 1999). When considering patient characteristics in all 21 included studies, eight recruited patients with mixed diagnoses, 7 recruited BPD patients, and the remaining recruited patients diagnosed with MDD (Connolly et al., 1999), panic disorders (Klein et al., 2003), pathological grief (Marmar et al., 1989; Milbrath et al., 1999), cluster C personality disorders (Ryum et al., 2010), and avoidant personality disorder (Schut et al., 2005).

Measure of transference

Details regarding the definition, operationalization, and measurement of TI within the included studies are specified in Table 2 (Buchheim et al., 2017; Clarkin et al., 2001, 2007; Con-nolly et al., 1999Diamond et al., 2023; Doering et al., 2010; Fis-cher-Kern et al., 2015; Hoglend et al., 1993; Hoglend et al., 2006; Klein et al., 2003; Levy et al., 2006; Malan, 1976; Mar-mar et al., 1989; Marziali, 1984; Milbrath et al.,1999; Ogrod-niczuk et al., 1999; Perez et al., 2016; Piper et al., 1986; Piper et al., 1991; Piper et al., 1999; Ryum et al., 2010; Sahin et al., 2018; Shut et al., 2005); the table highlights how most articles refer to not only varying definitions of TI but also different methods of TI measurement.

In general, transference was either measured, manipulated, or assumed (i.e., TI being by definition a part of the therapy model, such as TFP). In seven studies, transference was assumed by the authors due to the use of a form of psychodynamic psy-chotherapy that presumed the use of TI within its process; of these, six made use of TFP (Clarkin et al., 2001; Clarkin et al., 2007; Doering et al., 2010; Fischer-Kern et al., 2015; Levy et al., 2006; Perez et al., 2016), whereas the remaining study used object-relational psychotherapy (Sahin et al., 2018). All seven studies where transference was only assumed were conducted on BPD patients.

On the other hand, all 14 studies that included measurements of TI used different rating scales or methods of TI assessment. Several studies described utilizing a measure of score, fre-quency, proportion, or average number of TIs either based on a rating scale or a rater’s coding. For example, Ogrodniczuk et al.(1999) used the Therapist Intervention Rating System to calcu-late TI frequency. Other than one study that also measured TI frequency (Marziali, 1984), others evaluated TI delivery as the proportion of TIs over other forms of interpretation (Malan, 1976; Piper et al., 1986; Piper et al., 1991) or an average of TI frequency across specific sessions (Connolly et al., 1999). Fi-nally, several studies utilized Likert-type scores (e.g., 0-not at all or 5-great deal) from rating scales measuring TI (Klein et al., 2003; Marmar et al., 1989; Piper et al., 1999; Ryum et al., 2010).

The remaining articles made use of other assessment scales, such as the Vanderbilt Psychotherapy Process Scale (Piper et al., 1999), the Transference Focus Factor in the Interactive Process Assessment (Klein et al., 2003), the Therapist Action Scale (Marmar et al., 1989), the Psychodynamic Intervention Rating Scale (Milbrath et al., 1999), and the Inventory of Therapeutic Strategies (Ryum et al., 2010). Importantly, the studies also var-ied in terms of whether TI was evaluated internally by the ther-apist or externally through a rater. In the latter case, ratings were based on either transcripts of the sessions (Connolly et al., 1999; Marziali, 1984; Schut et al., 2005) or the therapist’s notes (Malan, 1976); for further details, see Table 2.

Finally, two articles actively manipulated TI by subdividing the sample into patients to whom TI was delivered and patients to whom TI was withheld (Høglend et al., 1993; Høglend et al., 2006). Both studies had similar characteristics: they included a sample of patients with mixed diagnoses, made use of dynamic psychotherapy once per week, and included blind raters who had the role of evaluating whether TI was delivered or withheld.

Characteristics of the interventions

Irrespective of transference being measured, assumed, or manipulated, all 21 studies included some form of psychody-namic psychotherapy (see column “Type of Dynamic Interven-tion” in Table 1). In general, the number of therapists per study that performed the interventions ranged from 3 to 31, with one study failing to specify the number of therapists (Malan, 1976). On the other hand, treatment frequency was highly standard-ized, with all 21 studies using a frequency of either once or twice per week. The range of treatment duration was instead quite wide, varying from 6 weeks to over a year, with six studies specifying that treatment duration was standardized to 1 year. As can be seen in Table 1, the interventions with the longest duration were often less standardized, and their characteristics were less clear. For example, a study by Malan (1976) reported 400 sessions; however, details regarding the frequency of the sessions or the number of therapists performing the interven-tions were unclear.

The most used and standardized form of therapy was TFP, with six articles having performed this type of intervention. These six studies have several common characteristics. Firstly, they all had a treatment frequency of twice per week, and five of the six studies also had an identical treatment duration of 1 year (Clarkin et al., 2001, 2007; Doering et al., 2010; Fischer-Kern et al., 2015; Levy et al., 2006); the remaining study that performed TFP therapy reported an average of 76.60 sessions for each participant (Perez et al., 2016). Finally, in all six studies, the sample consisted of BPD patients.

Of the remaining articles, two refer to using dynamic psy-chotherapy (Høglend et al., 1993; Høglend et al., 2006). As op-posed to those that used TFP, the studies that included dynamic psychotherapy had a lower treatment frequency of once per week and a highly variable duration that ranged from 9 weeks to 1 year. It should be noted that both studies included patient samples with mixed diagnoses; articles that recruited their sam-ple based on a specific diagnosis refer to using a more special-ized form of dynamic psychotherapy that is more focused on the characteristics of their sample. For example, panic-focused dy-namic psychotherapy was used on patients with panic disorders (Klein et al., 2003), and dynamic psychotherapy for stress re-sponse syndromes was used on patients with pathological grief (Milbrath et al., 1999).

Another form of therapy that was performed on samples of patients with a specific diagnosis was the supportive-expres-sive therapy (Connolly et al., 1999; Schut et al., 2005); in this case, the articles were focused on patients with major depres-sive disorder and avoidant personality disorder, respectively. The characteristics of the remaining types of psychotherapies are summarized in Table 1 (Clarkin et al., 2001; Clarkin et al., 2007; Connolly et al., 1999; Doering et al., 2010; Fischer-Kern et al., 2015; Hoglend et al., 1993; Hoglend et al., 2006; Klein et al., 2003; Levy et al., 2006; Malan, 1976; Mar-mar et al., 1989; Marziali, 1984; Milbrath et al., 1999; Ogrod-niczuk et al., 1999; Perez et al., 2016; Piper et al., 1986; Piper et al., 1991; Piper et al., 1999; Ryum et al., 2010; Sahin et al., 2018; Schut et al., 2005). Finally, studies that compared psychodynamic psychotherapy with another form of therapy that excluded TI delivery are described in the following paragraph.

Type of experimental control

In general, three forms of experimental control were used: pre-post intervention, comparison with another form of therapy, or manipulation of TI delivery. The only study not to include a form of experimental control assessed therapeutic alliance in a sample of patients diagnosed with pathological grief undergoing brief dynamic psychotherapy to validate the California Thera-peutic Alliance Rating System (Marmar et al., 1989).

The remaining 20 studies assessed the difference in one or more outcomes at different time points: before, during therapy, or at follow-up. Other than a pre-post form of experimental control, six studies also compared psychodynamic therapy, in which TI delivery is assumed, with another form of therapy in which the interpretation of transference is not a central focus of the intervention, such as in dialectical behavior therapy (Clarkin et al., 2007; Levy et al., 2006; Sahin et al., 2018), dy-namic supportive treatment (Clarkin et al., 2007), psychody-namic supportive therapy (Levy et al., 2006), and cognitive therapy (Ryum et al., 2010). In two studies, rather than per-forming different forms of therapy, patients were subdivided into a group that underwent TFP and one in which therapy was simply carried forth by experienced community psychothera-pists, in which it was assumed that TI would not be used (Do-ering et al., 2010; Fischer-Kern et al., 2015).

Finally, as previously mentioned, the last form of experi-mental control was the manipulation of TI delivery, in which TI was either actively delivered or withheld, with the assis-tance of external raters evaluating the process (Høglend et al., 1993; Høglend et al., 2006).

Measure of outcome

13 out of 21 (62%) included studies were able to observe a significant improvement in at least one therapy outcome measure following the use of TI in therapy. However, despite this result, no studies measured the same combination of out-comes, and the percentage given should be interpreted cau-tiously (e.g., within 13 studies, there are negative results on specific measures along with positive results on the remaining measures) (see column Outcome Measure in Table 1).

Due to the differing characteristics of the patient popula-tions in the included studies, the outcome measures that were taken into consideration were highly heterogeneous. For ex-ample, Klein et al. (2003) observed a reduction in panic symp-toms in patients with panic disorder following a 12-week panic-focused dynamic psychotherapy, whereas patients with cluster C personality disorders seemed to benefit in terms of interpersonal problems from a low dose of TI (Ryum et al., 2010). On the other hand, studies with mixed patient popula-tions found improvements in global outcomes, such as Malan’s Global Outcome Scale (Malan, 1976; Marziali, 1984).

The most consistent and wide-ranging results have been observed in BPD patients who underwent a yearlong TFP treat-ment. These patients significantly improved in terms of suicide attempts, hospitalization, impulsivity, aggression, irritability, and anger (Clarkin et al., 2001; Clarkin et al., 2007; Doering et al., 2010; Perez et al., 2016); most importantly, the same pa-tients also showed a reduction in BPD diagnostic criteria and an improvement in personality organization (Doering et al., 2010). Interestingly, two of these studies also observed an im-provement in the capacity for reflective functioning, as meas-ured by the Reflective Functioning Scale based on the Adult Attachment Interview (Fischer-Kern et al., 2015; Levy et al., 2006).

On the contrary, several studies either failed to find signif-icant results on specific outcome measures due to TI in therapy or even observed a negative effect of TI on therapy outcomes. For example, Piper et al. (1991) found an inverse relationship between the proportion of TIs and both therapeutic alliance and therapy outcome in patients with a history of high-quality object relations. The result was supported by further studies that failed to demonstrate a difference in therapeutic outcomes when TI was delivered or withheld from a sample of patients with mixed diagnoses (Høglend et al., 2006). Moreover, some studies show a significant negative effect of a high number of TIs on long-term dynamic outcomes (Høglend et al., 1993), on defensive style and family functioning (Ogrodniczuk et al., 1999), on levels of depression in patients suffering from major depressive disorder (Connolly et al., 1999), on alliance vari-ables such as patient commitment and patient working capacity (Marmar et al., 1989), and on global functioning (Schut et al., 2005). Finally, Milbrath et al. (1999) observed that TI was not correlated with any measures assessing symptomology and functioning.

It is important to note that the studies often listed numerous variables that were not ultimately statistically analyzed or in-cluded as an outcome within the results; nonetheless, all re-ported outcomes for each individual study, including those that resulted as significantly improved, are specified in Table 1 (Clarkin et al., 2001; Clarkin et al., 2007; Connolly et al., 1999; Doering et al., 2010; Fischer-Kern et al., 2015; Hoglend et al., 1993; Hoglend et al., 2006; Klein et al., 2003; Levy et al., 2006; Malan, 1976; Marmar et al., 1989; Marziali, 1984; Milbrath et al., 1999; Ogrodniczuk et al., 1999; Perez et al., 2016; Piper et al., 1986; Piper et al.,1991; Piper et al., 1999; Ryum et al., 2010; Sahin et al., 2018; Schut et al., 2005).

Discussion

Despite constituting the key technical element that differen-tiates psychodynamic psychotherapies from other forms of ther-apies (e.g., cognitive-behavioral therapies), empirical evidence supporting the efficacy of TI lags behind its use and centrality in clinical practice (Cutler et al., 2004). Overall, this systematic review strived to contribute to the previous efforts to close the gap in the literature, highlighting the substantial differences in the characteristics of the studies that have explored the relation-ship between TI and therapy outcomes. Nonetheless, the ob-served results suggest that although the use of TI within dynamic psychotherapies is often associated with significant benefits, there are many factors at play in determining whether this tech-nique is ultimately beneficial, or in some cases, even detrimen-tal, to the therapeutic process.

The current systematic review analyzed 21 studies that ex-plored the relationship between TI and therapy outcomes. The main finding of this study is the high heterogeneity observed in the designs of the studies, ranging from the operationalization of transference to the types of measures that were considered therapy outcomes. Regarding the conceptualization and opera-tionalization of transference, most individual studies used vary-ing definitions of TI, which were also reflected in the use of different methods of measuring, manipulating, or controlling for this variable within their study designs (see Table 2). The het-erogeneity and lack of standardization also extended to the char-acteristics and descriptions of the interventions and the characteristics of the patients undergoing therapy.

Nonetheless, despite the high variability, 62% of included studies observed a statistically significant improvement in ther-apy outcomes linked to TI, suggesting that its inclusion within psychotherapy may bring notable benefits to patients. It is also important to note that the studies that failed to observe this result often did not consider a measure of outcome change at all (Mar-mar et al., 1989) or found more complex results that do not nec-essarily exclude the benefits of TI use within therapy. Instead, these results suggest that improper and excessive use of TIs may be detrimental and that the delicate relationship between trans-ference and therapy outcome is interconnected to several other factors, such as frequency and proportion of TIs, as well as the patients’ diagnosis and specific characteristics.

While Ogrodniczuk et al. (1999) found an inverse relation-ship between the frequency of TIs and both therapeutic alliance and favorable outcomes in individuals with low quality of object relations, their prior study (Piper et al., 1991) and the FEST re-sults showed the opposite. The incongruence between these re-sults might be due to the nature of the treatment characteristics [i.e., the proportion of TI was 6% in Ogrodniczuk et al. (1999) and in Piper et al. (1999) it was 12%]. On the one hand, the neg-ative correlation between the proportion of TI and therapeutic outcome (e.g., interpersonal functioning, psychiatric symptoms) could be due to the interaction between the short duration of therapy and the patient’s quality of object relations. Due to their more mature personality organization (i.e., integrated identity) and defensive functioning (i.e., neurotic or obsessional defenses such as repression or isolation of affect), patients with high-qual-ity of object relations are more likely to repress negative internal object representations and build more neutral and/or positive therapeutic alliance with their therapists, starting from the very beginning of the treatment (Caligor et al., 2007; Conversano et al., 2023). More time might be needed with high-quality object relation patients for their repressed relational representations to unfold and reveal themselves within the therapeutic relationship in the form of transference, which could also be observable to the therapist to capture and interpret them. On the other hand, patients with low quality of object relations reveal their trans-ference dynamics, characterized by negative and aggressive en-actments, much sooner than patients with high quality of object relations. It could also be asserted that patients with low quality of object relations who are characterized by intense splitting and disavowal of traumatic experiences unconsciously act on their object relations within the therapeutic relationship, and the ther-apist’s interpretations of these dynamics along with the thera-pist’s stable presence negate the expected enactment of destructive and persecutory fantasies. This process might serve as a corrective emotional experience.

Limitations

The current systematic review has several limitations. Firstly, there was a small number of available studies. For this reason, the inclusion criteria that were applied in the selection process were not specific to a particular patient population, type of therapy, or method of TI measurement. This may have led to less standardi-zation and more heterogeneity between studies; however, it also allowed for a more accurate portrayal of the current state of the literature on the subject as a whole. Furthermore, due to the lack of consensus in defining and operationalizing TI, it is important to remain cautious when attempting to generalize any observed results; what is considered TI in one study may vary in another.

Due to the aforementioned extreme variability regarding the measurement of transference, type of intervention, type of ex-perimental control, and measurement of outcome, it was also chosen not to conduct a meta-analysis. Indeed, it seemed more critical and appropriate to systematically review the numerous methods and designs used throughout the years to explore the effects of TI in dynamic therapy before statistically testing its potential benefits via a meta-analysis. Finally, although the cur-rent study was limited to quantitative studies, several qualitative case studies in the field of TI may be of great interest (Banon et al., 2001; Goodman, 2011; Henriksen et al., 2021; Ulberg et al., 2014). In particular, when considering the mixed and complex results observed through the current systematic review, single-case designs and qualitative studies would allow researchers to explore the dynamics underlying TI use throughout the therapy process in greater detail; perhaps a systematic review limited to qualitative studies that investigates the relationship between TI and therapy outcome may be the next step.

Suggestions for future research

Based on the results of the present study, several suggestions could be given for future research. Understanding how TI and its specific components work and influence the therapy process and outcome requires more complex models. In this respect, prospective studies should investigate TI both as an independent variable that brings improvement in therapy outcomes when cer-tain moderating factors (e.g., patient variables, insight) are in-cluded in the model and as a mediator itself to scrutinize its role as a core mechanism of therapeutic change in psychodynamic psychotherapy (Kazdin, 2007).

Moreover, since TI is a multi-factorial clinical phenomenon, including its timing, dosage, accuracy, and impact on the pa-tient’s psyche, it surpasses frequency measurement as commonly practiced in previous studies (Luborsky et al., 1988; Luborsky & Crits-Christoph, 1990; Schut et al., 2005; Silberschatz et al., 1986). As Malan (1976, pp. 210-211) suggested, “a single cor-rect and well-timed interpretation may be all that is needed for a successful result…it seems to be much commoner that such an interpretation needs to be given on a number of different oc-casions in different contexts, and thus to be to some extent ‘worked through’ before therapeutic effects can be permanent”. Thus, it would be erroneous to conclude that the less frequent the TI within a psychotherapy process, the better the outcome. For example, FEST studies found a negative relationship be-tween the number of TIs and outcomes for patients with high-quality of object relations, differentiated treatment, and control groups based on the frequency of TI per session. Although it was valuable as a starting point, future research should move beyond the assessment of frequency by investigating clinically relevant factors such as timing, accuracy, and impact of TIs both in ses-sion and throughout therapy.

Another important point to consider within the context of psychotherapy research is allegiance bias, which refers to the possible bias of researchers toward adherence to a specific psy-chotherapy approach while conducting research and assessing the efficacy of their psychotherapy approach (Leichsenring et al., 2017). Allegiance bias might lead to an inflated effect size for the observed effect. Most of the existing studies on TI dom-inantly include researchers and clinicians from the psychody-namic approach. Most of the existing studies on TI include researchers and clinicians from the psychodynamic approach. In future studies, researchers could collaborate with colleagues using different methods to reduce allegiance bias to a minimum. Finally, to assess the real-life effectiveness of TIs and the gen-eralizability of the findings, conducting pragmatic controlled tri-als (e.g., flexibility in the interpretation of the intervention, minimal exclusion criteria) may be an aim for future research on the effects of TI on therapy outcomes (Godwin et al., 2003).

Conclusions

To sum up, accumulated evidence suggests that TI should be used meticulously with consideration of various patient and therapy process factors and further empirically investigated by means of appropriate measurement tools and experimental ma-nipulation. Current results show that TI brings favorable out-comes in psychodynamic psychotherapy, such as significant decrease in symptom severity and maladaptive behaviors, as well as improvements in psychodynamic functioning and inter-personal relationships. Being one of the fundamental techniques of psychodynamic psychotherapy, TI must receive significantly more attention and research effort, which would reinforce the current evidence base for psychodynamic practice.

Author Contributions and other additional information

Contributions: MY, KKT, DH, were involved in the data extraction process and the writing of the manuscript; MY, KKT, LZ, prepared the figures and tables, and interpreted the results; LZ, summarised and wrote the results.

All co-authors were involved in the concep-tualization and the revision of the article for content and language, and agreed on the final version.

Conflict of interest: the authors declare no potential conflict of in-terest. Ethics approval and consent to participate: not applicable.

Funding: this research did not receive any grant from funding agen-cies in the public, commercial, or non-profit sectors. Availability of data and materials: the available data is the list of the studies pertaining separately to each PRISMA step. Since the current study is a systematic review, no original data from any study is included. Received: 17 December 2023.

Accepted: 18 April 2024. Citation: Yılmaz, M., Türkarslan, K.K., Zanini, L., Hasdemir, D., Spitoni, G.F., & Lingiardi, V., (2024). Transference interpretation and psychotherapy outcome: a systematic review of a no-consensus relationship. Research in Psychotherapy: Psychopathology, Process and Outcome, 27(1), 744. doi: 10.4081/ripppo.2024.744

References

Antichi, L., Giannini, M., & Loscalzo, Y. (2022). Interpretation in psychodynamic psychotherapy: a systematic review. Psy-chodynamic Practice, 28(3), 254-274. doi: 10.1080/ 14753634.2022.2046140.

Banon, E., Evan-Grenier, M., & Bond, M. (2001). Early transfer-ence interventions with male patients in psychotherapy. Jour-nal of Psychotherapy Practice and Research, 10(2), 79-92.

Banon, E., Perry, J. C., Semeniuk, T., Bond, M., de Roten, Y., Her-soug, A. G., & Despland, J.-N. (2013). Therapist interventions using the psychodynamic interventions rating scale (PIRS) in dynamic therapy, psychoanalysis and CBT. Psychotherapy Research, 23(2), 121-136. doi: 10.1080/10503307.2012. 745955.

Brumberg, J., & Gumz, A. (2012). [Transference interpretations and how they work: a systematic review]. Zeitschrift fur Psy-chosomatische Medizin und Psychotherapie, 58(3), 219-235. doi: 10.13109/zptm.2012.58.3.219. [Article in German].

Caligor, E., Kernberg, O. F., & Clarkin, J. F. (2007). Handbook of dynamic psychotherapy for higher level personality pathol-ogy. American Psychiatric Publishing, Inc.

Clarkin, J. F., Foelsch, P. A., Levy, K. N., Hull, J. W., Delaney, J. C., & Kernberg, O. F. (2001). The development of a psycho-dynamic treatment for patients with borderline personality disorder: a preliminary study of behavioral change. Journal of Personality Disorders, 15(6), 487-495. doi: 10.1521/pedi. 15.6.487.19190.

Clarkin, J. F., Levy, K. N., Lenzenweger, M. F., & Kernberg, O. F. (2007). Evaluating three treatments for borderline person-ality disorder: a multiwave study. American Journal of Psy-chiatry, 164, 922-928. doi: 10.1176/appi.ajp.164.6.922.

Connolly, M., Crits-Christoph, P., Shappell, S., Barber, J., Luborsky, L., & Shaffer, C. (1999). Relation of transference interpretations to outcome in the early sessions of brief sup-portive-expressive psychotherapy. Psychotherapy Research, 9(4), 485-495. doi: 10.1080/10503309912331332881.

Conversano, C., Di Giuseppe, M., & Lingiardi V. (2023) Case re-port: changes in defense mechanisms, personality functioning, and body mass index during psychotherapy with patients with anorexia nervosa. Frontiers in Psychology, 14, 1081467. doi: 10.3389/fpsyg.2023.1081467.

Cooper, A. M. (1987). Changes in psychoanalytic ideas: transfer-ence interpretation. Journal of the American Psychoanalytic Association, 35(1), 77-98. doi: 10.1177/00030651870350 0104.

Crits-Christoph, P., & Gibbons, M. B. C. (2021). Psychotherapy process-outcome research: advances in understanding causal connections. In Bergin and Garfield’s Handbook of Psy-chotherapy and Behavior Change (pp 263-296).

Crits-Christoph, P., & Gibbons, M. B. C. (2001). Relational in-terpretations. Psychotherapy: Theory, Research, Practice, Training, 38(4), 423-428. doi: 10.1037/0033-3204.38.4.423.

Cutler, J. L., Goldyne, A., Markowitz, J. C., Devlin, M. J., & Glick, R. A. (2004). Comparing cognitive behavior therapy, interpersonal psychotherapy, and psychodynamic psychother-apy. American Journal of Psychiatry, 161(9), 1567-1573. doi: 10.1176/appi.ajp.161.9.1567.

Diamond, D., Keefe, J. R., Hörz-Sagstetter, S., Fischer-Kern, M., Doering, S., & Buchheim, A. (2023). Changes in attachment representation and personality organization in transference-focused psychotherapy. American Journal of Psychotherapy, 76(1), 31-38. doi: 10.1176/appi.psychotherapy.20220018.

Doering, S., Hörz, S., Rentrop, M., Fischer-Kern, M., Schuster, P., Benecke, C., Buchheim, A., Martius, P., & Buchheim, P. (2010). Transference-focused psychotherapy v. treatment by community psychotherapists for borderline personality disor-der: randomised controlled trial. British Journal of Psychiatry, 196(5), 389-395. doi: 10.1192/bjp.bp.109.070177.

Fischer-Kern, M., Doering, S., Taubner, S., Hörz, S., Zimmer-mann, J., Rentrop, M., Schuster, P., Buchheim, P., & Buch-heim, A. (2015). Transference-focused psychotherapy for borderline personality disorder: change in reflective function. British Journal of Psychiatry, 207(2), 173-174. doi: 10.1192/ bjp.bp.113.143842.

Freud, S. (1905). Fragment of an analysis of a case of hysteria (1905 [1901]). The Standard Edition of the Complete Psycho-logical Works of Sigmund Freud, Volume VII (1901-1905): A Case of Hysteria, Three Essays on Sexuality and Other Works. Available from: https://www.sas.upenn.edu/~cavitch/ pdf-library/Freud_SE_Dora_complete.pdf.

Freud, S. (1912). The dynamics of transference. In Classics in Psychoanalytic Techniques, 12, 97-108. Freud, S. (1917). Introductory lectures on psycho-analysis. The Standard Edition of the Complete Psychological Works of Sigmund Freud, 16,241-463.

Gabbard, G. O. (2004). Long-term psychodynamic psychother-apy: a basic text. American Psychiatric Publishing, Inc. Gabbard, G. O., & Horowitz, M. J. (2009). Insight, transference interpretation, and therapeutic change in the dynamic psy-chotherapy of borderline personality disorder. American Jour-nal of Psychiatry, 166(5), 517-521. doi: 10.1176/appi.ajp. 2008.08050631.

Gabbard, G. O., Horwitz, L., Allen, J. G., Frieswyk, S., Newsom, G., Colson, D. B., & Coyne, L. (1994). Transference interpre-tation in the psychotherapy of borderline patients: a high-risk, high-gain phenomenon. Harvard Review of Psychiatry, 2(2), 59-69. doi: 10.3109/10673229409017119.

Giovacchini, P. L. (1979). Treatment of primitive mental states. Jason Aranson. Godwin, M., Ruhland, L., Casson, I., MacDonald, S., Delva, D., Birtwhistle, R., Lam, M., & Seguin, R. (2003). Pragmatic con-trolled clinical trials in primary care: the struggle between ex-ternal and internal validity. BMC Medical Research Methodology, 3(1), 28. doi: 10.1186/1471-2288-3-28.

Goodman, G. (2011). An attachment-based pathways model de-picting the psychology of therapeutic relationships. In A. M. Columbus (Ed.), Advances in psychology research. (pp. 105-125). Nova Science Publishers. Henriksen, A. K., Ulberg, R., Tallberg, B. P. U., Løvgren, A., & Johnsen Dahl, H. S. (2021). Conflicted anger as a central dy-namic in depression in adolescents—a double case study. In-ternational Journal of Environmental Research and Public Health, 18(12), 6466. doi: 10.3390/ijerph18126466.

Hersoug, A. G., Ulberg, R., & Høglend, P. (2014). When is trans-ference work useful in psychodynamic psychotherapy? Main results of the first experimental study of transference work (FEST). Contemporary Psychoanalysis, 50(1-2), 156-174. doi: 10.1080/00107530.2014.880314. Hobson, R. P., & Kapur, R. (2005). Working in the transference: clinical and research perspectives. Psychology and Psy-chotherapy: Theory, Research and Practice, 78(3), 275-295. doi: 10.1348/147608305X25801.

Høglend, P. (2004). Analysis of transference in psychodynamic psychotherapy: a review of empirical research. Canadian Journal of Psychoanalysis/Revue Canadienne de Psych-analyse, 12, 280-300. Høglend, P. (2014). Exploration of the patient-therapist relation-ship in psychotherapy. American Journal of Psychiatry, 171(10), 1056-1066. doi: 10.1176/appi.ajp.2014.14010121.

Høglend, P., Amlo, S., Marble, A., Bøgwald, K.-P., Sørbye, Ø., Sjaastad, M. C., & Heyerdahl, O. (2006). Analysis of the pa-tient-therapist relationship in dynamic psychotherapy: an ex-perimental study of transference interpretations. American Journal of Psychiatry, 163(10), 1739-1746. doi: 10.1176/ajp. 2006.163.10.1739.

Høglend, P., Bøgwald, K.-P., Amlo, S., Marble, A., Ulberg, R., Sjaastad, M. C., Sørbye, Ø., Heyerdahl, O., & Johansson, P. (2008). Transference interpretations in dynamic psychother-apy: do they really yield sustained effects? American Journal of Psychiatry, 165(6), 763-771. doi: 10.1176/appi.ajp.2008. 07061028.

Høglend, P., Sørlie, T., Heyerdahl, O., Sørbye, Ø., & Amlo, S. (1993). Brief dynamic psychotherapy. Journal of Psychother-apy Practice and Research, 2(3), 230-241. Joseph, B. (1985). Transference: the total situation. International Journal of Psychoanalysis, 66, 447-454. Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychol-ogy, 3(1), 1-27. doi: 10.1146/annurev.clinpsy.3.022806. 091432.

Kernberg, O. F., Yeomans, F. E., Clarkin, J. F., & Levy, K. N. (2008). Transference focused psychotherapy: overview and update. International Journal of Psychoanalysis, 89(3), 601-620. doi: 10.1111/j.1745-8315.2008.00046.x. Klein, C., Milrod, B. L., Busch, F. N., Levy, K. N., & Shapiro, T. (2003). A preliminary study of clinical process in relation to outcome in psychodynamic psychotherapy for panic disorder. Psychoanalytic Inquiry, 23(2), 308-331. doi: 10.1080/ 07351692309349035.

Klein, M. (1952). The origins of transference. International Jour-nal of Psychoanalysis, 33, 433-438. Kohut, H. (1971). The analysis of the self. University of Chicago Press.

Kohut, H. (1984). How does analysis cure?. University of Chicago Press. Langs, R. J. (1973). The patient’s view of the therapist: reality or fantasy? International Journal of Psychoanalytic Psychother-apy, 2(4), 411-431.

Leichsenring, F., Abbass, A., Hilsenroth, M. J., Leweke, F., Luyten, P., Keefe, J. R., Midgley, N., Rabung, S., Salzer, S., & Steinert, C. (2017). Biases in research: risk factors for non-replicability in psychotherapy and pharmacotherapy research. Psychological Medicine, 47(6), 1000-1011. doi: 10.1017/ S003329171600324X.

Leichsenring, F., Hiller, W., Weissberg, M., & Leibing, E. (2006). Cognitive-behavioral therapy and psychodynamic psy-chotherapy: techniques, efficacy, and indications. American Journal of Psychotherapy, 60, 233-259. doi: 10.1176/appi. psychotherapy.2006.60.3.233.

Leichsenring, F., & Rabung, S. (2011). Long-term psychodynamic psychotherapy in complex mental disorders: Update of a meta-analysis. British Journal of Psychiatry, 199(1), 15-22. doi: 10.1192/bjp.bp.110.082776.

Levy, K. N., Clarkin, J. F., Yeomans, F. E., Scott, L. N., Wasser-man, R. H., & Kernberg, O. F. (2006). The mechanisms of change in the treatment of borderline personality disorder with transference focused psychotherapy. Journal of Clinical Psy-chology, 62(4), 481-501. doi: 10.1002/jclp.20239.

Levy, K. N., & Scala, J. W. (2012). Transference, transference in-terpretations, and transference-focused psychotherapies. Psy-chotherapy, 49(3), 391-403. doi: 10.1037/a0029371.

Lilliengren, P. (2017). Comprehensive compilation of randomized controlled trials (RCTs) involving psychodynamic treatments and interventions. Luborsky, L., & Crits-Christoph, P. (1990). Understanding trans-ference: the core conflictual relationship theme method. Basic Books.

Luborsky, L., Crits-Christoph, P., Mintz, J., & Auerbach, A. (1988). Who will benefit from psychotherapy? Predicting therapeutic outcomes. Basic Books. Luyten, P., Fonagy, P., Lowyck, B., & Vermote, R. (2012). As-sessment of mentalization. In Handbook of mentalizing in mental health practice (pp. 43-65). American Psychiatric Pub-lishing, Inc.

Malan, D. H. (1976). Toward the validation of dynamic psy-chotherapy. A replication. Springer New York. Marmar, C. R., Weiss, D. S., & Gaston, L. (1989). Toward the val-idation of the California therapeutic alliance rating system. Psychological Assessment: a Journal of Consulting and Clin-ical Psychology, 1(1), 46-52. doi: 10.1037/1040-3590.1.1.46.

Marziali, E. A. (1984). Prediction of outcome of brief psychother-apy from therapist interpretive interventions. Archives of Gen-eral Psychiatry, 41(3), 301-304. doi: 10.1001/archpsyc. 1984.01790140091011. McCullough, L., Winston, A., Farber, B. A., Porter, F., Pollack, J., Vingiano, W., Laikin, M., & Trujillo, M. (1991). The rela-tionship of patient-therapist interaction to outcome in brief psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 28(4), 525-533. doi: 10.1037/0033-3204.28.4.525.

Milbrath, C., Bond, M., Cooper, S., Znoj, H., Horowitz, M., & Perry, J. (1999). Sequential consequences of therapists’ inter-ventions. Journal of Psychotherapy Practice and Research, 8, 40-54. Ogrodniczuk, J. S., Piper, W. E., Joyce, A. S., & McCallum, M. (1999). Transference interpretations in short-term dy-namic psychotherapy. Journal of Nervous and Mental Dis-ease, 187(9), 571-578. doi: 10.1097/00005053-199909000-00007.

Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoff-mann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., Chou, R., Glanville, J., Grimshaw J.M., Hróbjartsson, A., Lalu, M. M., Li, T., Loder, E. W., Mayo-Wilson, E., McDonald, S., McGuinness, L.A., Stewart, L. A., Thomas, J., Tricco, A. C., Welch, V. A., Whiting, P., & Moher, D. (2021). The PRISMA 2020 statement: an updated guide-line for reporting systematic reviews. International Journalof Surgery, 88, 105906.Perez, D. L., Vago, D. R., Pan, H., Root, J., Tuescher, O., Fuchs, B. H., Leung, L., Epstein, J., Cain, N. M., Clarkin, J. F.,Lenzenweger, M. F., Kernberg, O. F., Levy, K. N., Silber-sweig, D. A., & Stern, E. (2016). Frontolimbic neural circuitchanges in emotional processing and inhibitory control asso-ciated with clinical improvement following transference-fo-cused psychotherapy in borderline personality disorder.Psychiatry and Clinical Neurosciences, 70(1), 51-61. doi:10.1111/pcn.12357.

Piper, W. E., Azim, H. F. A., Joyce, A. S., & McCallum, M. (1991). Transference interpretations, therapeutic alliance, and outcome in short-term individual psychotherapy. Archives of General Psychiatry, 48(10), 946-953. doi: 10.1001/archpsyc. 1991.01810340078010.

Piper, W. E., Debbane, E. G., Bienvenu, J.-P., de Carufel, F., & Garant, J. (1986). Relationships between the object focus of therapist interpretations and outcome in short-term individual psychotherapy. British Journal of Medical Psychology, 59(1), 1-11. doi: 10.1111/j.2044-8341.1986.tb02659.x.

Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S., McCallum, M., Rosie, J. S., O’Kelly, J. G., & Steinberg, P. I. (1999). Predic-tion of dropping out in time-limited, interpretive individual psychotherapy. Psychotherapy: Theory, Research, Practice, Training, 36, 114-122. doi: 10.1037/h0087787.

Ryum, T., Stiles, T. C., Svartberg, M., & McCullough, L. (2010). The role of transference work, the therapeutic alliance, and their interaction in reducing interpersonal problems among psychotherapy patients with Cluster C personality disorders. Psychotherapy: Theory, Research, Practice, Training, 47, 442-453. doi: 10.1037/a0021183.

Sahin, Z., Vinnars, B., Gorman, B. S., Wilczek, A., Åsberg, M., & Barber, J. P. (2018). Clinical severity as a moderator of out-come in psychodynamic and dialectical behavior therapies for borderline personality disorder. Personality Disorders: The-ory, Research, and Treatment, 9(5), 437-446. doi: 10.1037/ per0000276.

Schut, A. J., Castonguay, L. G., Flanagan, K. M., Yamasaki, A. S., Barber, J. P., Bedics, J. D., & Smith, T. L. (2005). Therapist interpretation, patient-therapist interpersonal process, and out-come in psychodynamic psychotherapy for avoidant person-ality disorder. Psychotherapy: Theory, Research, Practice, Training, 42(4), 494-511. doi: 10.1037/0033-3204.42.4.494.

Shedler, J. (2010). The efficacy of psychodynamic psychotherapy. American Psychologist, 65(2), 98-109. doi: 10.1037/ a0018378. Silberschatz, G., Fretter, P. B., & Curtis, J. T. (1986). How do in-terpretations influence the process of psychotherapy?. Journal of Consulting and Clinical Psychology, 54(5), 646-652. doi: 10.1037/0022-006X.54.5.646.

Steinert, C., Munder, T., Rabung, S., Hoyer, J., & Leichsenring, F. (2017). Psychodynamic therapy: as efficacious as other em-pirically supported treatments? A meta-analysis testing equiv-alence of outcomes. American Journal of Psychiatry, 174(10), 943-953. doi: 10.1176/appi.ajp.2017.17010057.Strachey, J. (1934).

The nature of the therapeutic action of psy-cho-analysis. International Journal of Psycho-Analysis, 15, 127-159.Tyndale, A. (1999). How far is transference interpretation essentialto psychic change? In Psychoanalytic Psychotherapy in the Independent Tradition. Routledge. Ulberg, R., Amlo, S., & Høglend, P. (2014). Manual for transfer-ence work scale; a micro-analytical tool for therapy process analyses. BMC Psychiatry, 14(1), 291. doi: 10.1186/s12888-014-0291-y.

Ulberg, R., Hummelen, B., Hersoug, A. G., Midgley, N., Høglend, P. A., & Dahl, H.-S. J. (2021). The first experimental study of transference work–in teenagers (FEST–IT): a multicentre, ob-server- and patient-blind, randomised controlled componentstudy. BMC Psychiatry, 21(1), 106. doi: 10.1186/s12888-021-03055-y.Winnicott, D. W. (1949). Hate in the countertransference. Inter-national journal of psychoanalysis, 30(2), 69-74. Zetzel, E. R. (1956). Current concepts of transference. Interna-tional Journal of Psychoanalysis, 37, 369-375.

Zilcha-Mano, S., Fisher, H., Dolev-Amit, T., Keefe, J. R., & Bar-ber, J. P. (2023). A systematic review of the association be-tween interpretations and immediate, intermediate, and distal outcomes. Psychotherapy, 60(3), 266-282. doi: 10.1037/ pst0000479.

Zilcha-Mano, S., Keefe, J. R., Fisher, H., Dolev-Amit, T., Veler-Poleg, N., & Barber, J. P. (2023). Is the use of interpretations associated with treatment outcome? A systematic review and meta-analytic answer. Clinical Psychology: Science and Practice, 31(1). doi:10.1037/cps0000158.