ABSTRACT
Objective
Although two meta‐analyses have examined the relationship between rupture repair and treatment outcome, no meta‐analysis to date has investigated the association between alliance ruptures and treatment outcome. Numerous primary studies have explored this relationship; however, the methods used to assess ruptures vary considerably. These include direct measures, such as ratings of the occurrence, frequency, or intensity of ruptures by observers, patients, or therapists, and indirect measures, such as fluctuations in alliance ratings. This methodological heterogeneity may contribute to inconsistent findings across studies. The present meta‐analysis sought to clarify the association between alliance ruptures and treatment outcomes by including only studies that assessed ruptures using direct measures.
Methods
After applying inclusion and exclusion criteria, four publications remained reporting 18 effect sizes of the associations between ruptures and outcomes. Those provided data from N = 301 clients undergoing psychotherapy. Multilevel models with a three‐level structure with restricted maximum likelihood to estimate the aggregated effect size of ruptures on outcome were conducted.
Results
The results showed a significant aggregated effect size of z = −0.22, SE = 0.04, 95 CI [−0.303, −0.133]; t(17) = −5.40, p < 0.001, r = −0.21, indicating that alliance ruptures were associated with poorer therapy outcomes.
Conclusions
A small but consistent negative association between alliance ruptures and treatment outcomes was found. An individual patient data meta‐analysis would be an important future endeavor in the field of alliance ruptures, by including more studies and a homogenous analytic approach to estimate rupture–outcome associations and including rupture–repair as a moderator.
1.
The therapeutic alliance, defined as the collaboration on the tasks and goals of therapy as well as the affective bond between client and therapist (Bordin 1979), has been found to be predictive of treatment success (Flückiger et al. 2018). However, the alliance is not a fixed phenomenon, but is rather characterized by rupture and repair episodes, which are common across theoretical frameworks and therapists of all skill levels (Eubanks‐Carter et al. 2012). Alliance ruptures are defined as breakdowns in the collaboration between therapist and client on tasks or goals or a strain in the emotional bond (Eubanks and Muran 2022). Ruptures can be marked by withdrawal, confrontation or a combination of both (Safran and Muran 2000). In the case of withdrawal ruptures, the client and/or therapist moves away from the other, or the work of therapy, while in confrontation ruptures, the client and/or therapist moves against each other or the work of therapy (Eubanks et al. 2018). Alliance ruptures can have detrimental effects on therapy outcome or may lead to dropout (e.g., Rubel et al. 2018).
Ruptures can be addressed by means of repair strategies. Repair strategies are interventions that aim to resume collaboration with the client on tasks or goals of therapy and strengthen the emotional bond (Eubanks and Muran 2022). Two meta‐analyses have been conducted exploring the association between rupture repair processes and treatment outcomes (Eubanks et al. 2018; Safran et al. 2011). Both meta‐analyses found a moderate positive association between rupture repair and treatment outcomes, with pooled effect sizes of r = 0.24 (Safran et al. 2011) and r = 0.29 (Eubanks et al. 2018). Although many studies on alliance ruptures have been published in recent years, the field still lacks a meta‐analytic investigation of the direct relationship between alliance ruptures and treatment outcomes. Understanding the size and consistency of the association between alliance ruptures and poorer treatment outcomes is essential to contextualize the benefits of rupture repair.
Studies investigating ruptures rely on a variety of measurement instruments, which can be distinguished conceptually and operationally as direct or indirect measures (Eubanks et al. 2018). Direct rupture measures are specifically designed to assess rupture processes—such as withdrawal or confrontation—either through self‐report or observer‐based coding. Self‐report measures, such as the Post‐Session Questionnaire (PSQ; Muran et al. 1992), ask clients or therapists to explicitly report on the occurrence or intensity of ruptures (e.g., “Did you experience any tension, misunderstanding, conflict, or disagreement in your relationship with your therapist/patient during the session? If yes, please rate how tense or upset you felt about this during the session.”). Observer‐based methods, such as the Rupture Resolution Rating System (3RS; Eubanks et al. 2018; Eubanks and Muran 2022), involve independent coders assessing videotaped therapy sessions for withdrawal and confrontation ruptures (e.g., Babl et al. 2022). Direct measures often allow for more fine‐grained, episode‐level assessment, capturing subtle or transient ruptures that may be repaired within the same session (Babl et al. 2024; Coutinho et al. 2014).
In contrast, indirect rupture measures infer rupture processes from fluctuations or declines in alliance ratings rather than measuring ruptures explicitly. For example, the Working Alliance Inventory (WAI; Tracey and Kokotovic 1989) has frequently been used as an indirect rupture measure, where drops in session‐level alliance scores are interpreted as potential rupture events (Eubanks‐Carter et al. 2012). Comparisons between WAI‐based indicators and the 3RS suggest that alliance fluctuations capture some rupture events but may miss more subtle or transient disruptions, such as withdrawal ruptures or ruptures that are repaired within a session (Babl et al. 2024; Coutinho et al. 2014). Importantly, these findings primarily concern indirect, alliance‐fluctuation measures and do not necessarily generalize to self‐report measures designed to explicitly assess rupture experiences, such as the PSQ.
This conceptual and operational distinction highlights that while some items on alliance measures may overlap with rupture experiences, direct measures are specifically tailored to identify rupture processes at the session or episode level, whereas indirect measures infer ruptures from broader alliance changes.
Based on the gaps in previous research on alliance ruptures, the current meta‐analysis aims to establish a link between ruptures (assessed with direct measures) and outcome. It was hypothesized that alliance ruptures would be negatively associated with psychotherapy outcomes.
2. Methods
The current meta‐analysis followed the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA; Page et al. 2021) guidelines.
2.1. Search Strategy
Through the research platform EBSCOhost, the database PsycINFO was comprehensively screened. The search terms were applied in a multi‐field format in the following manner: alliance (field 1) AND rupture (field 2) AND psychotherapy OR therapy OR counseling OR intervention OR treatment (field 3). All three fields must be mentioned in the abstract and search options were specified to All Journals, Journal Article and English language. This primary search resulted in a list of 232 publications on January 16, 2024.
2.2. Eligibility Criteria
Inclusion and exclusion criteria of the meta‐analysis were as follows: (a) be an empirical study (not a review or theoretical paper), (b) not be a case study or task analysis, (c) involve a quantitative and direct measure of alliance ruptures, (d) involve a quantitative outcome measure, (e) implement a psychotherapeutic intervention (in contrast to primary medical care (or other types of intervention), (f) include a population with a sample mean age of at least 18 years, (g) not be duplicated data, and (h) report an effect size of the association between ruptures and psychotherapy outcome.
2.3. Selection Process
Screening was conducted in two sequential stages: (1) title‐and‐abstract screening and (2) full‐text screening. Both stages were carried out independently by three master's‐level psychology students. Following each stage, discrepancies regarding study eligibility were resolved through discussion with the supervisor until consensus was reached.
The initial search of the PsycINFO database using the predefined search strategy yielded 232 publications. All records underwent title‐and‐abstract screening, followed by full‐text screening for potentially eligible studies by three independent coders. Inclusion and exclusion decisions were based on the predefined eligibility criteria. Interrater reliability was assessed using Fleiss' κ, a chance‐corrected measure of agreement for multiple raters. For the title/abstract screening stage, agreement was moderate/substantial (Fleiss' κ = 0.49) and 161 articles were excluded. The remaining 71 articles went for the full‐text screening stage, agreement was moderate (Fleiss' κ = 0.34). An additional 44 studies were excluded. Of the 27 remaining studies, four reported effect sizes for the association between ruptures and psychotherapy outcomes and were thus included in the meta‐analysis. For the flow‐chart of the studies included in this meta‐analysis, see Figure 1.
Figure 1.
Flow chart of the studies included in the meta‐analysis.
Finally, in an effort to find relevant studies that might have been uploaded on other databases or were inadvertently omitted due to our search terms, we manually examined the reference lists of the 27 eligible publications. This secondary screening did not yield additional studies. General descriptive data of the studies meeting the inclusion criteria for this meta‐analysis are provided in Table 1.
Table 1.
Main characteristics of the studies included in the meta‐analysis.
| Authors (year) | Peer‐Reviewed | Treatment type/treatment conditions | Average treatment length (in sessions) | n clients | Mean client age (SD) | % Female clients | n therapists | Mean therapist age (SD) | Client main diagnosis |
|---|---|---|---|---|---|---|---|---|---|
| Atzil‐Slonim et al. (2021) | Yes | Psychodynamic psychotherapy | 15 | 58 | 39.06 (13.67) | 58.9% | 52 | N/A (N/A) | Mostly anxiety and affective disorders |
| Watson et al. (2017) | Yes | Sex offender treatment program | 36 | 71 | 45.06 (15.07) | 0.0% | 17 | N/A (N/A) | Sex offense |
| Gersh et al. (2017) | Yes | Two conditions: Cognitive analytic therapy or Befriending treatment | 16 | 44 | 18.2 (2.81) | 81.8% | 14 | N/A (N/A) | Borderline personality disorder |
| Muran et al. (2009) | Yes | Three conditions: Short‐term psychodynamic therapy or BRT or CBT | 30 | 128 | 41.33 (10.52) | 53.1% | 70 | 38.14 (1.89) | Personality disorders |
Abbreviations: BRT = brief relational therapy, CBT = cognitive behavioral therapy.
2.4. Data Extraction
Data from eligible publications was extracted independently by three master‐level psychology students using a data extraction form specifically designed for the purpose of this study. The extracted information of the three coders was subsequently compared, and any disagreements were discussed collaboratively with the supervisor until consensus was reached. No formal interrater reliability statistic was calculated.
The data extraction form comprised the following information: authors, year of publication, title of the study, sample characteristics including sample size, mean age, percentage of female clients, primary diagnosis, and prevalence of personality disorder diagnoses; therapist characteristics including mean age and percentage of female therapists; therapy length; number of sessions per week; method used to measure alliance ruptures; and timing of the rupture assessment. Additionally, information regarding the method used to assess treatment outcome, timing of outcome assessment, comparison groups, and reported effect sizes was extracted.
Some aspects of the extraction process required researcher judgment, particularly the classification of how alliance ruptures were operationalized across studies. Specifically, effect sizes reflected rupture intensity, rupture frequency, or the presence versus absence of a rupture. Rupture intensity was the most common operationalization. Judgment was also required when interpreting study descriptions to determine the relevant timing and measurement procedures for ruptures and outcomes.
2.5. Statistical Analysis
The analyses were conducted within the free environment R (R Core Team 2021) with the package metafor (Viechtbauer and Cheung 2010). We used a multilevel meta‐analytic model with a three‐level structure with a restricted maximum likelihood method to estimate the aggregated effect size of ruptures on outcome. In this model, sampling variance (Level 1) was nested within effect sizes (k = 18; Level 2), nested within studies (s = 4; Level 3). Of the four included studies, three reported effect sizes as Pearson's r. The Adjusted R 2 originally reported in the other study was converted into Pearson's r by use of an online converter (Lin n.d.). A multilevel forest plot was used to illustrate the distribution of the effect sizes.
Most studies reported multiple effect sizes, for instance when measuring rupture frequency from both patients' and therapists' perspectives or when measuring several outcome variables. Thus, to assess the overall heterogeneity across the included effect sizes, a Q analysis was implemented (Higgins and Thompson 2002). We have harmonized effect directions across studies. We decided to code effect directions so that a negative correlation coefficient (r) consistently represents a negative association between alliance ruptures and psychotherapy outcomes; that is, higher levels of alliance ruptures are associated with poorer treatment outcomes. Where necessary, correlations were reversed to ensure that the direction of effects was comparable across studies, regardless of the original scoring direction of the outcome measures (Table 2). The correlation coefficients were transformed into Fisher's z and the analysis was performed based on those z scores (Borenstein et al. 2010). Then, the z was transformed back into correlation coefficients in order to allow for easier interpretation.
Table 2.
Characteristics of the effect sizes reported in the included studies of the meta‐analysis.
| Authors (year) | Rupture measure | Timing rupture measure | Outcome (measure) | Timing outcome measure | Outcome rater | Reported ES | r (negative r values represent a negative rupture–outcome association) |
|---|---|---|---|---|---|---|---|
| Atzil‐Slonim et al. (2021) | PSQ‐T (presence of rupture) | After each session | Symptomatology (HSCL) | Before each session | Patient | r = 0.32 | −0.32 |
| Watson et al. (2017) | PSQ‐P (presence of rupture) | ¾ through treatment | Treatment gain (TRRG‐SV) | Post‐treatment | Other | Adjusted R 2 = −0.00 | −0.00 |
| Gersh et al. (2017) |
3RS (number of ruptures) (rupture intensity) |
Sessions 3, 9, 15 | Interpersonal problems (SAS‐SR) | Pre–post change scores | Patient |
Early session (3): r = 0.32 |
−0.32 |
| Mid session (9): r = 0.28 | −0.28 | ||||||
| Late session (15): r = −0.30 | 0.30 | ||||||
| Early session (3): r = 0.25 | −0.25 | ||||||
| Mid session (9): r = 0.16 | −0.16 | ||||||
| Late session (15): r = −0.24 | 0.24 | ||||||
|
3RS (number of ruptures) (rupture intensity) |
Symptomatology (BPDSI‐IV) | Pre–post change scores | Other |
Early session (3): r = 0.18 |
−0.18 |
||
| Mid session (9): r = 0.42 | −0.42 | ||||||
| Late session (15): r = 0.43 | −0.43 | ||||||
| Early session (3): r = 0.17 | −0.17 | ||||||
| Mid session (9): r = 0.36 | −0.36 | ||||||
| Late session (15): r = 0.40 | −0.40 | ||||||
| Muran et al. (2009) | PSQ‐P (rupture intensity) | After sessions 1–6 (average) | Symptomatology (SCL‐90‐R & GAS) | Post‐treatment | Patient | r = −0.15 | −0.15 |
|
PSQ‐T (rupture intensity) |
Symptomatology (SCL‐90‐R & GAS) | Post‐treatment | Therapist | r = −0.08 | −0.08 | ||
|
PSQ‐P (rupture intensity) |
Interpersonal problems (IIP & WISPI) |
Post‐treatment | Patient | r = −0.35 | −0.35 | ||
|
PSQ‐T (rupture intensity) |
Interpersonal problems (IIP & WISPI) |
Post‐treatment | Therapist | r = −0.32 | −0.32 |
Abbreviations: BPDSI‐IV = borderline personality disorder severity index‐IV, ECR = experience in close relationships scale, GAS = global assessment scale, HSCL = hopkins symptom checklist, IIP = inventory of interpersonal problems, PSQ‐P = post‐session questionnaire patient version, PSQ‐T = post‐session questionnaire therapist version, SAS‐SR = social adjustment scale—self‐report, SCL‐90‐R = symptom checklist‐90‐revised, TRRG‐SV = treatment readiness responsivity gain scale—short version, WISPI = wisconsin personality inventory.
The magnitudes of the correlations were classified as follows: small effect size (r = 0.10), medium effect size (r = 0.30), and large effect size (r = 0.50) (Cohen 1992). In an effort to check for publication bias, a funnel plot which depicts the standard error (vertical axis) and effect sizes for each study (horizontal axis; Fernández‐Castilla et al. 2020) was created and analyzed. A clear asymmetry towards one side of the mean represents outliers and would be indicative of publication bias. No such pattern could be detected in our sample.
3. Results
3.1. Study and Effect Size Characteristics
The four studies that met the inclusion criteria for the meta‐analysis provided data from N = 301 clients undergoing treatment, including a total of 18 correlational effect sizes of the association between alliance ruptures and psychotherapy outcome. The studies included treatments from different theoretical frameworks, such as psychodynamic psychotherapy, cognitive‐analytic therapy, befriending treatment, brief relational therapy, cognitive behavioral therapy and a sex offender treatment program. Main diagnoses ranged from anxiety and affective disorders to paraphilia and personality disorders.
Studies used either the 3RS or PSQ as direct measures of rupture (most of the time repeatedly in the early stages of treatment), alongside a range of outcome measures (most of the time post‐treatment scores). Assessment timing varied across studies, as reported in Table 2. Treatment length ranged from 15 to 36 weekly sessions (26 on average), as presented in Table 1. All studies were peer‐reviewed. Furthermore, characteristics of the effect sizes included in this meta‐analysis are provided in Table 2. The study quality of the included studies ranged from 67% to 83% (the criteria and quality rating per study can be found in the Supporting Information S1).
3.2. Meta‐Analytic Models
The results of the three‐level model showed a significant aggregated effect size of z = −0.22, SE = 0.04, 95 CI [−0.303, −0.133]; t(17) = −5.40, p < 0.001, r = −0.21, demonstrating a significant negative association between alliance ruptures and treatment outcomes. In Figure 2, a forest plot is presented displaying the effect sizes with their sampling variance, within‐study variability and between‐study variability.
Figure 2.
Forest plot of the pooled correlational (r) effect sizes representing the association between ruptures and outcomes. The gray confidence interval represents the sampling variance (Level 1). The black confidence interval represents the within‐study variance (Level 2).
The Q test indicated that heterogeneity across effect sizes was not statistically significant, Q(17) = 19.58, p = 0.30. Variance decomposition showed that 20.02% of the total variance was attributable to within‐study variability across reported outcomes (Level 2), whereas < 1% was attributable to between‐study variability (Level 3). These estimates descriptively suggest greater variability within studies than between studies; however, this interpretation should be treated with caution. Given the small number of included studies (s = 4), the between‐study variance estimate is likely unstable, and the analysis has limited power to detect between‐study heterogeneity.
The funnel plot displayed in Figure 3 did not show evidence of substantial publication bias. The absence of outliers and the plots tending towards the same direction (Figure 2) indicate a small but consistent effect size estimation.
Figure 3.
Funnel plot of aggregated effect sizes (k = 18) within studies (s = 4).
4. Discussion
This is the first meta‐analysis analyzing the association between alliance ruptures and outcome in psychotherapy. We found a small but consistent negative association between alliance ruptures and client outcome, drawing exclusively from studies that employed direct measures of ruptures.
In line with our hypothesis and previous findings (e.g., Rubel et al. 2018), the results suggest that ruptures during the therapeutic process are associated with worse therapy outcomes. Importantly, previous research has suggested that the consequences of alliance ruptures may depend on how they are addressed within therapy (Eubanks et al. 2018; Safran et al. 2011). However, the present meta‐analysis did not examine rupture repair and therefore cannot determine whether repair moderates the association between ruptures and outcome. Future research should investigate whether repaired and unrepaired ruptures differ in their associations with treatment outcome and under which conditions rupture repair may contribute to therapeutic change.
The present findings should be interpreted in light of substantial heterogeneity across study characteristics. Variations in clinical populations (e.g., diagnosis, severity) or therapeutic modality (e.g., CBT, psychodynamic) could influence the magnitude of rupture–outcome associations. Certain populations or treatment approaches may be more sensitive to the impact of ruptures, and ruptures that are assessed more frequently or at the session level may be more strongly related to outcomes.
An additional source of heterogeneity concerns the operationalization of alliance ruptures. Although all included studies used direct rupture measures, the synthesized effect sizes did not uniformly represent the same rupture characteristic. Most effect sizes reflected rupture intensity, whereas others reflected rupture frequency or the mere occurrence of ruptures. Consequently, the pooled estimate should be interpreted as reflecting the overall burden of alliance ruptures rather than a specific rupture dimension. Future research should examine whether rupture intensity, frequency, and occurrence differ in their associations with psychotherapy outcome.
The interpretability of the results is further complicated by the diversity of outcome measures, assessment timing (e.g., before each session, pre–post change scores), and frequency (e.g., repeated measures vs. a single timepoint) used across the included studies, which precluded meaningful comparisons.
While our pooled estimate suggests a small‐to‐moderate negative association between ruptures and outcome, this effect likely reflects an average across diverse contexts, and variability across studies may not be fully captured due to limited sample size. Although we restricted inclusion to studies using direct rupture measures, future meta‐analyses with a larger number of studies are needed to enable a more statistically robust synthesis and to explore sources of heterogeneity through moderator (e.g., outcome type, outcome rater), subgroup or sensitivity analyses.
Future research should include the 23 studies that fulfilled all other inclusion criteria but did not report effect sizes for the association between ruptures and psychotherapy outcome. Incorporating these studies, for example, through an individual patient data (IPD) meta‐analysis, would increase statistical power and provide a more comprehensive understanding of the rupture–outcome relationship. An IPD meta‐analysis would further allow to disentangle the effects of repaired and unrepaired ruptures on psychotherapy outcome, by indeed including rupture repair as a moderator of the rupture effect on outcome which unfortunately was not possible in this traditional meta‐analysis due to the presentation of the results in the original studies.
More generally, the temporal ordering of rupture and outcome processes remains unclear across much of the included literature. Because most studies assessed associations between alliance ruptures and treatment outcomes without explicitly examining their temporal sequencing, the observed relationships should not be interpreted as evidence that ruptures necessarily cause poorer outcomes. Alternative explanations remain possible, including the possibility that clients who are progressing less favorably may be more likely to experience difficulties within the therapeutic relationship and subsequent rupture experiences. As a result, the direction of influence between ruptures and treatment outcomes cannot be established from the available evidence. Future research should employ longitudinal designs with repeated assessments of both rupture processes and clinical outcomes to clarify their temporal and potentially reciprocal relationships.
In conclusion, this meta‐analysis adds to the existing literature on alliance ruptures by providing evidence for a negative association between ruptures and therapy outcome using direct rupture measures. Reporting standards with regard to rupture data in this young but promising field need to be established in order to lay ground for comparisons between studies.
Ethics Statement
This study is a meta‐analysis based exclusively on data reported in previously published studies. No new data involving human or animal participants were collected; therefore, ethical approval was not required.
Supporting information
Supporting File
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
References
- Atzil‐Slonim, D. , Juravski D., Bar‐Kalifa E., et al. 2021. “Using Topic Models to Identify Clients' Functioning Levels and Alliance Ruptures in Psychotherapy.” Psychotherapy 58, no. 2: 324–339. 10.1037/pst0000362. [DOI] [PubMed] [Google Scholar]
- Babl, A. , Berger T., Gómez Penedo J. M., Grosse Holtforth M., Caspar F., and Eubanks C. F.. 2022. “Disaggregating Between‐ and Within‐Patient Effects of Ruptures and Resolutions on the Therapeutic Alliance and Symptom Severity.” Psychotherapy 59, no. 4: 567–571. 10.1037/pst0000457. [DOI] [PubMed] [Google Scholar]
- Babl, A. , Rubel J., Gómez Penedo J. M., Berger T., grosse Holtforth M., and Eubanks C. F.. 2024. “Can Session‐by‐Session Changes in Self‐Reported Alliance Scores Serve as a Measure of Ruptures in the Therapeutic Alliance?” Journal of Consulting and Clinical Psychology 92, no. 2: 129–133. 10.1037/ccp0000861. [DOI] [PubMed] [Google Scholar]
- Bordin, E. S. 1979. “The Generalizability of the Psychoanalytic Concept of the Working Alliance.” Psychotherapy: Theory, Research & Practice 16: 252–260. [Google Scholar]
- Borenstein, M. , Hedges L. V., Higgins J. P. T., and Rothstein H. R.. 2010. “A Basic Introduction to Fixed‐Effect and Random‐Effects Models for Meta‐Analysis.” Research Synthesis Methods 1, no. 2: 97–111. 10.1002/jrsm.12. [DOI] [PubMed] [Google Scholar]
- Cohen, J. 1992. “A Power Primer.” Psychological Bulletin 112, no. 1: 155–159. 10.1037/0033-2909.112.1.155. [DOI] [PubMed] [Google Scholar]
- Coutinho, J. , Ribeiro E., Sousa I., and Safran J. D.. 2014. “Comparing Two Methods of Identifying Alliance Rupture Events.” Psychotherapy 51, no. 3: 434–442. 10.1037/a0032171. [DOI] [PubMed] [Google Scholar]
- Eubanks, C. F. , and Muran J. C.. 2022. “Rupture Resolution Rating System (3RS): Manual Version 2022”. 10.13140/RG.2.2.29780.17282. [DOI]
- Eubanks, C. F. , Muran J. C., and Safran J. D.. 2018. “Alliance Rupture Repair: A Meta‐Analysis.” Psychotherapy 55, no. 4: 508–519. 10.1037/pst0000185. [DOI] [PubMed] [Google Scholar]
- Eubanks‐Carter, C. , Gorman B. S., and Muran J. C.. 2012. “Quantitative Naturalistic Methods for Detecting Change Points in Psychotherapy Research: An Illustration With Alliance Ruptures.” Psychotherapy Research 22, no. 6: 621–637. 10.1080/10503307.2012.693772. [DOI] [PubMed] [Google Scholar]
- Fernández‐Castilla, B. , Jamshidi L., Declercq L., Beretvas S. N., Onghena P., and Van den Noortgate W.. 2020. “The Application of Meta‐Analytic (Multi‐Level) Models With Multiple Random Effects: A Systematic Review.” Behavior Research Methods 52, no. 5: 2031–2052. 10.3758/s13428-020-01373-9. [DOI] [PubMed] [Google Scholar]
- Flückiger, C. , Del Re A. C., Wampold B. E., and Horvath A. O.. 2018. “The Alliance in Adult Psychotherapy: A Meta‐Analytic Synthesis.” Psychotherapy 55, no. 4: 316–340. 10.1037/pst0000172. [DOI] [PubMed] [Google Scholar]
- Gersh, E. , Hulbert C. A., McKechnie B., Ramadan R., Worotniuk T., and Chanen A. M.. 2017. “Alliance Rupture and Repair Processes and Therapeutic Change in Youth With Borderline Personality Disorder.” Psychology and Psychotherapy: Theory, Research and Practice 90, no. 1: 84–104. 10.1111/papt.12097. [DOI] [PubMed] [Google Scholar]
- Higgins, J. P. T. , and Thompson S. G.. 2002. “Quantifying Heterogeneity in a Meta‐Analysis.” Statistics in Medicine 21, no. 11: 1539–1558. 10.1002/sim.1186. [DOI] [PubMed] [Google Scholar]
- Lin, H. , n.d. “Effect Size Converter.” Accessed July 13, 2026. https://www.escal.site/.
- Muran, J. C. , Safran J. D., Gorman B. S., Samstag L. W., Eubanks‐Carter C., and Winston A.. 2009. “The Relationship of Early Alliance Ruptures and Their Resolution to Process and Outcome in Three Time‐Limited Psychotherapies for Personality Disorders.” Psychotherapy: Theory, Research, Practice, Training 46, no. 2: 233–248. 10.1037/a0016085. [DOI] [PubMed] [Google Scholar]
- Muran, J. C. , Safran J. D., Samstag L. W., and Winston A.. 1992. “ Patient and Therapist Postsession Questionnaires”, Version 1992 . Beth Israel Medical Center. [Google Scholar]
- Page, M. J. , McKenzie J. E., Bossuyt P. M., et al. 2021. “The PRISMA 2020 Statement: An Updated Guideline for Reporting Systematic Reviews.” BMJ 372: 71. 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- R Core Team . 2021. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. [Google Scholar]
- Rubel, J. A. , Zilcha‐Mano S., Feils‐Klaus V., and Lutz W.. 2018. “Session‐to‐Session Effects of Alliance Ruptures in Outpatient CBT: Within‐ and Between‐Patient Associations.” Journal of Consulting and Clinical Psychology 86, no. 4: 354–366. 10.1037/ccp0000286. [DOI] [PubMed] [Google Scholar]
- Safran, J. D. , and Muran J. C.. 2000. “Resolving Therapeutic Alliance Ruptures: Diversity and Integration.” Journal of Clinical Psychology 56, no. 2: 233–243. 10.1002/(sici)1097-4679(200002)56:2<233::aid-jclp9>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
- Safran, J. D. , Muran J. C., and Eubanks‐Carter C.. 2011. “Repairing Alliance Ruptures.” Psychotherapy 48, no. 1: 80–87. 10.1037/a0022140. [DOI] [PubMed] [Google Scholar]
- Tracey, T. J. , and Kokotovic A. M.. 1989. “Factor Structure of the Working Alliance Inventory.” Journal of Consulting and Clinical Psychology 1, no. 3: 207–210. 10.1037/1040-3590.1.3.207. [DOI] [Google Scholar]
- Viechtbauer, W. , and Cheung M. W. L.. 2010. “Outlier and Influence Diagnostics for Meta‐Analysis.” Research Synthesis Methods 1, no. 2: 112–125. 10.1002/jrsm.11. [DOI] [PubMed] [Google Scholar]
- Watson, R. , Thomas S., and Daffern M.. 2017. “The Impact of Interpersonal Style on Ruptures and Repairs in the Therapeutic Alliance Between Offenders and Therapists in Sex Offender Treatment.” Sexual Abuse 29, no. 7: 709–728. 10.1177/1079063215617514. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting File
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.