International Journal of Cognitive Therapy

, Volume 11, Issue 2, pp 222–233 | Cite as

Therapeutic Alliance and Outcome in Complicated Grief Treatment

  • Kim Glickman
  • M. Katherine Shear
  • Melanie M. Wall
Regular Article


In this study, we examined the role of the therapeutic alliance in complicated grief treatment (CGT), an efficacious psychotherapy for complicated grief. Using data from a previously reported randomized controlled trial, we explored whether the alliance–outcome relationship differs by treatment group among treatment completers assigned to either CGT (n = 35) or interpersonal psychotherapy (IPT) (n = 34). Early alliance (at week 4) was associated with a reduction in grief symptoms in CGT but not IPT. The difference in the alliance–outcome relationship was accounted for by goal/task items on the Working Alliance Inventory. The contribution of a positive alliance, particularly the agreement on goals and tasks, may be more significant in CGT, a highly structured treatment, than IPT in reducing symptoms of complicated grief.


Therapeutic alliance Complicated grief treatment Psychotherapy outcome 

Complicated grief treatment (CGT) has shown to be efficacious in three randomized controlled trials (Shear et al. 2005; Shear et al. 2014; Shear et al. 2016). The specific factors that contribute to its efficacy are not fully understood. In a previous study, we identified three potential mediators of treatment outcome: a reduction in guilt/self-blame related to the deceased, negative thoughts about the future, and avoidance of reminders of the loss (Glickman et al. 2016). These results were consistent with prior research showing that treatments that target dysfunctional thoughts and avoidance behavior are more effective in treating complicated grief than those that do not (Boelen et al. 2007; Bryant et al. 2014; Rosner et al. 2014; Shear et al. 2014; Eisma et al. 2015).

Non-specific treatment factors such as the therapeutic alliance have not yet been examined in CGT and may also contribute to its efficacy. Therapeutic alliance, commonly defined as the agreement between patient and therapist on therapeutic goals, consensus on treatment tasks, and the relationship bond (Bordin 1979), has been identified in the psychotherapy literature as one of the most salient components of treatment (Wampold and Imel 2015). The most recent meta-analysis of alliance in individual psychotherapy found an aggregate correlation between alliance and outcome of.275 (Horvath et al. 2011). Recent studies have also shown a significant correlation between alliance and outcome across treatment type, patient characteristics, measurement scales, raters, time of alliance (early, mid, or late), outcome measures, research design, and researcher allegiance (Horvath et al. 2011; Fluckiger 2012). Only a handful of published studies on the alliance–outcome relationship have failed to find evidence of this association (Gaston et al. 1991; Barber et al. 1999; Feeley 1999). However, Feeley et al. (1999) argued that the alliance effect may be caused by early symptom improvement rather than the reverse. Researchers have cautioned that unless we can show temporal precedence (that alliance develops prior to symptom change), we cannot conclude that a positive alliance plays a role in symptom reduction (Johansson and Hoglend 2007; Kazdin 2007; Murphy 2009). More recent studies on the alliance–outcome relationship have controlled for prior symptom change (Klein et al. 2003; Zuroff and Blatt 2006). Falkenstrom et al. (2013) examined the impact of alliance on symptom change session by session and found that there was a reciprocal relationship between alliance and outcome (within the patient) whereby alliance predicted change in symptoms and prior change in symptoms predicted change in alliance. The authors concluded that even though alliance may be enhanced by symptom improvement, it is not only a “bi-product” of change in outcome but also a causal factor.

Research spanning several decades has demonstrated an association between alliance and outcome across treatment type (Krupnick et al. 1996; Zuroff and Blatt 2006; Capaldi et al. 2016). More recently, there is an emerging body of literature showing that within a given study, treatment type moderates the alliance–outcome relationship. Beckner et al. (2007) found that alliance predicted improvements in depression in CBT but not in emotion-focused therapy. Arnow et al. (2013) found that the alliance–outcome relationship was significant in both CBT and brief supportive psychotherapy for chronically depressed patients but was stronger in CBT. Ormhaug et al. (2014) found that alliance predicted reduction in PTSD symptoms in CBT but not in treatment as usual. These studies suggest that the alliance may operate differently across treatments.

In the abovementioned studies, it was posited that the emphasis in CBT on transparency and structure with respect to collaborative tasks and goal setting might have contributed to the stronger alliance–outcome association. Arnow et al. (2013) examined between group differences in alliance dimensions and found higher mean ratings on tasks and goals subscales of the WAI in the CBT condition compared to brief supportive psychotherapy. However, this study did not examine whether the subscales were related to outcome.

To our knowledge, no studies have examined the role of alliance in CGT or whether specific dimensions of the alliance play a stronger role in predicting outcome in CGT than others. CGT is a structured approach that uses procedures from cognitive behavioral therapy, interpersonal psychotherapy, and motivational interviewing to foster the natural process of adaptation to loss and to resolve problematic thoughts, feelings, and behaviors that interfere with adaptation. One premise of CGT is that we do not grieve well alone, and for this reason, establishment of a companionship alliance is a core treatment component. This companionship alliance is also used to support collaboration in the other key treatment components.

The objective of the present study is to examine whether the alliance–outcome relationship differs by treatment type and if so, whether specific dimensions of the alliance account for the difference. Data was utilized from treatment completers of a randomized controlled trial (RCT), in which complicated grief treatment (CGT) was significantly more effective than interpersonal psychotherapy (IPT) in reducing symptoms of CG and grief-related impairment in functioning (Shear et al. 2005).

To carry out this objective, we posed the following questions: (1) Does the therapeutic alliance, measured early in treatment and controlling for early symptom change, predict outcome? (2) Does the alliance–outcome relationship differ by treatment group? And (3) does the relationship between specific dimensions of the alliance and outcome differ by treatment group? While a handful of recent studies have shown treatment type to moderate the alliance–outcome relationship, the preponderance of evidence has shown that a positive alliance is associated with symptom change in all types of psychotherapy. Both CGT and IPT emphasize a close and supportive relationship between patient and therapist, where agreement on treatment goals and tasks is sought and the therapist takes the role of ally and treatment companion, celebrating accomplishments and providing support for situations that are challenging (Markowitz and Weissman 2004; Shear 2013). We therefore hypothesized that alliance would predict outcome in both groups with no differential effect based on treatment type. As there were no prior studies about how the different dimensions of alliance relate to outcome across treatments, this aim was exploratory.


Overview of Parent Study

A detailed description of the clinical trial is available in a previously published paper (Shear et al. 2005). Briefly, recruitment of participants was carried out through media advertisement, professional and self-referral. Participants were included if they had a loss at least 6 months prior, a score of > 30 on the Inventory of Complicated Grief (ICG) (Prigerson et al. 1995) and identified grief as their most important problem. Medication was permitted as long as the participant had been on it for 3 months and at a stable dose for greater than 6 weeks. Participants were randomly assigned to receive 16 sessions of CGT or IPT. Therapists (either masters or doctoral level) were trained in either CGT or IPT and received ongoing supervision. Therapy sessions were audio taped for adherence. Self-report measures and questionnaires administered by independent evaluators (blinded to treatment assignment) were used to collect data at baseline, during treatment and post-treatment follow-up. Response to treatment was defined as a rating of 1 or 2 (very much or much improved) on the Clinical Global Impression of Improvement Scale (CGI). Additional outcome measures included the Inventory of Complicated Grief (ICG), Beck Anxiety Inventory (BAI), Beck Depression Inventory (BDI), and Work and Social Adjustment Scale (WSAS). Therapeutic alliance was measured by the Working Alliance Inventory (WAI).

Each treatment was delivered in 16 weekly sessions. For a detailed description of each treatment, see Shear et al. (2005). Briefly, IPT treatment is based on a model that postulates a bidirectional relationship between interpersonal problems and mood. The therapist and patient collaborate to select one or two of four interpersonal problem areas that are thought to be associated with mood symptoms including grief, interpersonal disputes, role transition, and interpersonal deficits. Most of the IPT treatments focused on grief alone or with one of the other problem areas.

CGT used an attachment theory model of grief and mourning. It specifically targeted relief of complicating problems and facilitation of the natural healing process. CGT includes seven core modules: (1) history taking and psychoeducation, (2) self-regulation, (3) aspirational goals, (4) building support, (5) imaginal revisiting: the story of the death, (6) situational revisiting, and (7) memories and pictures. Sessions are structured and include attention to both loss and restoration-related issues. For example, a core loss-related procedure is an “imaginal revisiting” exercise in which the patient tells the story of when s/he first learned of the death in the present tense and then listens to an audiotape of this at home in-between sessions. Aspirational goals work is a core restoration-related procedure. Grief monitoring, situational revisiting, imaginal conversation, and memories work are all loss- and restoration-oriented. The manual used in our NIMH-funded treatments is available at


The current study examined treatment completers only (n = 69) including n = 35 for CGT and n = 34 for IPT in order to test hypotheses about the alliance–outcome relationship among those who complete the full course of treatment. Completers did not differ from drop-outs on any of the baseline measures. For CGT, 13/49 (27%) dropped out of treatment, and for IPT, 12/46 (26%) dropped out of treatment.

Dependent Variables

Clinical Global Improvement (CGI)

The CGI scale (Guy 1976) is a single Likert-type clinician rating from 1 to 7 where 1 through 3 indicate very much, much, and minimally improved respectively; 4 indicates no change; and 5 through 8 indicate minimally, much, and very much worse, respectively. Independent evaluators completed CGI ratings after the post-treatment assessment interviews. A CGI rating of 1 or 2 (very much or much improved) qualified the participant as a treatment responder.

Inventory of Complicated Grief (ICG)

The 19-item Inventory of Complicated Grief assesses symptoms of CG. This scale has been utilized in various studies of CG and has good internal validity and reliability (alpha = .94) and 6-month test–retest reliability (r = .80). In the initial study, a score of 25 defined the upper quartile of scores and was associated with significant impairment in functioning (Prigerson et al. 1995). This measure was administered at baseline, weeks 1–16, post-treatment, and 6-month follow-up.

Work and Social Adjustment Scale (WSAS)

The WSAS is a modification of a scale developed by Hafner and Marks, consisting of 0 to 8 point ratings of the extent to which grief symptoms interfere in five areas of daily functioning: work, home management, private leisure, social leisure, and family relationships. It is a well-validated and widely used measure (Mundt et al. 2002). This measure was administered at baseline, weeks 1–16, post-treatment, and 6-month follow-up.

Independent Variable

Working Alliance

Alliance was measured by the Working Alliance Inventory short form (WAI), a 12-item self-report instrument that measures three aspects of the therapeutic relationship: (1) perceived agreement between client and therapist on therapy goals (e.g., “my therapist and I are working towards mutually agreed upon goals”); (2) perceived agreement between client and therapist on the tasks of therapy (e.g., “my therapist and I agree about the things I will need to do in therapy to help improve my situation”); and (3) the interpersonal bond between client and therapist (e.g., “my therapist and I trust one another”). Each item is rated on a seven-point scale with response choices ranging from never to always. This measure is widely used and has good performance characteristics (α = .92) (Horvath 1989; Byrd et al. 2010; Falkenstrom et al. 2015a). The WAI was collected at weeks 4, 8, and 12. The week 4 alliance score was used in this study as a measure of early alliance in order to establish temporal precedence with symptom change.

Statistical Analysis

Comparisons between completers in each treatment group on baseline characteristics used Fisher’s exact test for categorical variables and two-tailed t tests for continuous variables. Two-tailed t tests were used to examine associations between treatment group and alliance at week 4, treatment group and each individual item of the WAI, and treatment group and a composite goal/task subscale of the WAI. Statistical significance was set at p ≤ .05. Effect size (ES) for the difference in alliance scores between groups was calculated by dividing the mean difference between groups by the pooled standard deviation.

Regression analyses were performed to examine whether early alliance (week 4), predicted subsequent change in outcome, controlling for early change in outcome (week 1 to 4), and treatment group. Linear regression was used for modeling subsequent change in ICG and WSAS from week 5 to week 16. Logistic regression was used for modeling the end of study CGI at week 20 while controlling for early change in grief symptoms. Similar analyses were conducted to examine the relationship between the goal/task subscale of the WAI and outcome, controlling for early change in outcome.

A test for interaction between treatment group and alliance (week 4) on change in outcome (week 5 to 16) was also included in the regression models to examine whether the relationship between alliance and outcome differed by treatment group, controlling for early symptom change (week 1 to 4). Similar analyses were conducted for the goal/task subscale.

All analyses were performed using STATA version 11 software.


Baseline Characteristics

The completer sample was 84% female, 74% Caucasian, 22% African-American, and 3% Asian, with a mean age of 48.4 years. Mean ICG score was 44.8. Mean WSAS score was 21.2. Comorbid mood or anxiety disorders were common. Forty-six percent had current major depressive disorder (MDD), 52% current posttraumatic stress disorder (PTSD), 10% current panic disorder (PD), 3% current social phobia disorder, and 6% current obsessive–compulsive disorder (OCD). Thirty-four out of 69 (49%) were taking antidepressant medication. There were no significant differences in demographic or clinical baseline characteristics between the two randomized groups at baseline.

Treatment Outcome

As reported in a previously published manuscript (Shear et al. 2005), the response to CGT was significantly better than the response to IPT. Among treatment completers, participants in CGT showed a greater reduction in grief-related symptoms and impairment (CGI odds ratio [OR] 4.01 (2.1), p = .007; ICG β = .62 (.23), p = .009; WSAS β = .49 (.24), p = .041). Effect size (ES) of the difference between groups was large (Cohen’s d = .91) for the ICG and moderate for the WSAS (d = .50).

Alliance in CGT vs. IPT

Alliance at week 4 was higher in the CGT group than in IPT (M = 70.4, SD = 9.5 in CGT vs. M = 65.5, SD = 11.6 in IPT, p = .053). The effect size of the difference between the groups was moderate (Cohen’s d = .46).

As shown in Table 1, we found a higher rate of endorsement in CGT for three specific items on the WAI: Q6, “My therapist and I are working towards mutually agreed upon goals” (M = 6.3, SD = 1.0 in CGT vs. M = 5.6, SD = 1.4 in IPT, p = .016); Q8, “We agree on what is important for me to work on” (M = 6.2, SD = .87 in CGT vs. M = 5.6, SD = 1.2 in IPT, p = .022); and Q11, “We have established a good understanding of the kind of change that would be good for me” (M = 6.1, SD = 1.1 in CGT vs. M = 5.1, SD = 1.7 in IPT, p = .006). The effect size differences for these items were all moderate to high (Cohen’s d = .57, .55, and .65, respectively). When combined into a goal/task subscale, the difference between the treatments was significant (M = 18.54, SD = 2.4 in CGT vs. M = 16.26, SD = 3.93 in IPT, p = .005) with a moderate to high effect size (Cohen’s d = .66).
Table 1

Working Alliance Inventory (WAI) by treatment type (measured at 4 weeks)


CGT (n = 35)

IPT (n = 34)


M (SD)

M (SD)



Q1. My therapist and I agree about the things I will need to do in therapy to help improve my situation

5.85 (1.34)

5.20 (1.49)



Q2. What I am doing in therapy gives me new ways of looking at my problem

5.31 (1.32)

4.91 (1.42)



Q3. I believe my therapist likes me

5.65 (1.35)

5.47 (1.56)



Q4. My therapist does not understand what I am trying to accomplish in therapy

2.40 (2.10)

2.17 (1.66)



Q5. I am confident in my therapist’s ability to help me

5.97 (1.20)

5.56 (1.35)



Q6. My therapist and I are working towards mutually agreed upon goals

6.29 (1.02)

5.56 (1.42)



Q7. I feel that my therapist appreciates me

5.74 (1.31)

5.20 (1.45)



Q8. We agree on what is important for me to work on

6.20 (.87)

5.62 (1.18)



Q9. My therapist and I trust one another

6.17 (1.15)

5.88 (1.01)



Q10. My therapist and I have different ideas on what my problem is

2.09 (1.48)

2.38 (1.37)



Q11. We have established a good understanding of the kind of change that would be good for me

6.06 (1.14)

5.09 (1.66)



Q12. I believe that the way we are working with my problem is correct

5.60 (1.31)

5.41 (1.44)



WAI total score

70.37 (9.53)

65.35 (11.63)



Goal/task subscale

18.54 (2.40)

16.26 (3.97)



Each item on the WAI is scored from 1 = never to 7 = always. Items Q4 and Q10 were reverse scored when calculating the total WAI score. Items that significantly differed by treatment type are in italics. These items (Q6, Q8, and Q11) formed the composite variable for the goal/task subscale

CGT complicated grief treatment, IPT interpersonal psychotherapy

Association between Alliance and Outcome

The regression equations, controlling for earlier change in outcome (weeks 1 to 4) on the ICG and WSAS revealed no statistically significant relationship between the total alliance score and subsequent change in outcome on any of the measures (CGI OR = 1.04, SE = .03, p = .100; ICG β = .09, SE = .12, p = .448; WSAS β = .21, SE = .12, p = .088). There was, however, a statistically significant interaction between alliance and treatment group on outcome (ICG β = .50, SE = .23, p = .032). The interaction indicated that there was a significant positive relationship between alliance and change in grief symptoms within the CGT arm (ICG β = .39, SE = .18, p = .029), but a null relationship between alliance and outcome in the IPT arm (ICG β = − .12, SE = .15, p = .436); see Fig. 1.
Fig. 1

Scatterplot showing relationship between alliance at week 4 and subsequent change in grief symptoms (Inventory of Complicated Grief)

Similarly, the analysis of the goal/task subscale and outcome revealed no statistically significant association between the goal/task subscale and subsequent change in outcome in the overall sample. However, there was a statistically significant interaction between the goal/task subscale and treatment group on both the ICG and WSAS (ICG β = .59, SE = .26, p = .028; WSAS β = .72, SE = .26, p = .008), indicating a significant goal/task–outcome relationship in CGT (ICG β = .45, SE = .23, p = .044; WSAS β = .66, SE = .23, p = .003) but not in IPT (ICG β = − .14, SE = .14, p = .308; WSAS β = − .06, SE = .14, p = .674).


This study examined the role of therapeutic alliance in relieving symptoms of CG among treatment completers of a randomized controlled trial comparing CGT to IPT. Contrary to our prediction, early alliance did not predict subsequent change in CG symptoms in the overall sample; however, there was an interaction effect of treatment group and alliance on outcome, whereby early alliance predicted outcome in CGT but not in IPT. The difference in the alliance–outcome relationship by treatment type was accounted for by a goal/task dimension of the alliance.

The finding that the alliance–outcome relationship varied by treatment type is consistent with several prior studies (Beckner et al. 2007; Arnow et al. 2013; Ormhaug et al. 2014). Prior research showing a stronger alliance–outcome relationship in more structured treatments was similar to our findings. Beckner et al. (2007) speculated that the transparent process, structure, clear goals, and expectations of CBT may account for stronger alliance–outcome associations. Arnow et al. (2013) surmised that the structural features of CBT compared to brief supportive therapy enhanced patients’ sense that the treatment would be helpful and that they had shared goals with the therapist. Ormahaug et al. (2014) noted that the alliance may be particularly important in treatments that include activating procedures such as exposure where “the alliance could function as a catalyst for engagement and involvement with active therapy components,” p. 60.

Structure, clear goals, transparency, and support for activating elements of the treatment may explain the greater impact of alliance on outcome in CGT vs. IPT. This possibility is supported by the difference we observed in items on the alliance measure that refer to perceived agreement on treatment goals and tasks. Patients may feel themselves to be collaborators when the therapist lays out the theoretical model of the treatment, the goals, and the activities in a collaborative manner. While this is done in both treatments, CGT differs in being more structured both in terms of the format of each session and the exercises (e.g., grief monitoring, imaginal and situational revisiting, between session homework, etc.). Each session follows a format that is explicitly shared with the patient at the beginning of the session. This structure may have an emotion regulation effect and may allow better learning in the sessions. It may also provide a clearer roadmap for what can be accomplished each week and how each step fits into the broader goals for treatment.

With a clear understanding of where the treatment is headed and the steps necessary to facilitate change, patients may find it easier to follow through with treatment tasks (particularly those that are very challenging) and therefore achieve better outcomes. For example, in CGT, the imaginal and situational revisiting exercises are core treatment components that for some patients can be very difficult. Imaginal revisiting requires that the patient describe the moment she learned of the death in great detail and then listen to an audio tape of the retelling at home. Situational revisiting involves confronting situations that remind the patient of the death or the deceased such as places, people, or activities that stir up painful emotions. Having a clear understanding between patient and therapist about the goals of treatment and the tasks necessary to achieve them may provide the support needed for patients to tolerate these activating elements and benefit from them.

The content of psycho-education is another factor that may explain the findings in this study. CGT helps patients to understand the difference between normal grief and complicated grief and provides information about addressing obstacles that can interfere with adaptation to loss. Patients feel considerable relief when they feel that someone understands what they have been going through and that there is a treatment specifically designed to help them. In IPT, complicated grief symptoms are explained as a form of depression rather than as a distinct syndrome requiring a unique treatment approach. This explanation may not seem as validating or generate as much hope and/or motivation in the patient as in CGT.

A related possibility is that elements in CGT that foster cognitive change (e.g., debriefing after the imaginal revisiting) impact both the alliance and the outcome. Patients who begin to reframe negative thoughts about the future and reconsider self-blame related to the death may feel more confident in the therapist’s ability to help them, which may in turn contribute to positive outcomes. Prior research shows that cognitive change contributes to symptom reduction in the treatment of depression (Lorenzo-Luacas et al. 2015) and complicated grief (Glickman et al. 2016).

This study has several limitations. We examined alliance only as it pertained to treatment completers. Inclusion of treatment drop-outs in an intent-to-treat analysis may produce different results. In fact, it seems likely that alliance ratings would be lower among patients who dropped out than in treatment completers. As such, we were most interested in knowing what contributed to improvement among patients who received the full dose of treatment.

Another limitation is that alliance was measured by only one instrument, the patient-rated WAI. Our results suggest that patient perceptions of the relationship with the therapist matter in CGT. This finding supports prior research, which has shown that self-reported alliance is most predictive of outcome (Horvath and Symonds 1991; Zuroff and Blatt 2006). Because we did not use the therapist version, we cannot conclude that the same is true for therapist perceptions of the alliance. Future research should examine the alliance–outcome relationship using a therapist-rated and observer-rated measure of the WAI.

Additionally, the WAI specifically assesses factors that are important in cognitive–behavioral treatments (i.e., agreement on goals and tasks) with less of a focus on the interpersonal bond between patient and therapist such as feelings of being liked or feelings of emotional safety with the therapist. It is possible that using another measure of alliance would produce different results, and this should be explored in future research. Using the long form rather than the short form of the WAI may be another way to examine the relationship between different aspects of the working alliance and outcome as it contains more questions in each domain. Future research might also examine the relationship between the change in alliance and change in symptoms over time or even session-by-session. The 6-item version of the WAI (Falkenstrom et al. 2015b) would be a useful way to measure alliance at frequent intervals.

The study is limited by being a secondary analysis of a study not designed to explore alliance in depth. Therefore, client characteristics such as attachment style or comorbid personality disorder that might affect alliance (Lorenzo-Luacas et al. 2014) were not included and could be explored in future studies of alliance in CGT.

Our findings are the first to examine therapeutic alliance in the treatment of complicated grief. The identification of early alliance (particularly the goal/task dimension) as a predictor of outcome in CGT suggests that in a short-term treatment for an identified problem, clear goal-setting between patient and therapist may play a key role in facilitating positive change.


  1. Arnow, B. A., Steidtmann, D., Blasey, C., Manber, R., Constantino, M. J., Klein, D. N., et al. (2013). The relationship between the therapeutic alliance and treatment outcome in two distinct psychotherapies for chronic depression. Journal of Consulting and Clinical Psychology, 81(4), 627–638. Scholar
  2. Barber, J., Luborsky, L., Christoph-Crits, P., Thase, M. E., Weiss, R., Frank, A., et al. (1999). Therapeutic alliance as a predictor of outcome in treatment of cocaine dependence. Psychotherapy Research, 9(1), 54–73.CrossRefGoogle Scholar
  3. Beckner, V., Vella, L., Howard, I., & Mohr, D. (2007). Alliance in two telephone-administered treatments: relationship with depression and health outcomes. Journal of Consulting and Clinical Psychology, 75(3), 508–512.CrossRefPubMedGoogle Scholar
  4. Boelen, P., de Keijser, J., van den Hout, M., & van den Bout, J. (2007). Treatment of complicated grief: a comparison between cognitive-behavioral therapy and supportive counseling. Journal of Consulting and Clinical Psychology, 75(2), 277–284.CrossRefPubMedGoogle Scholar
  5. Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research, and Practice, 16, 252–260.CrossRefGoogle Scholar
  6. Bryant, R. A., Kenny, L., Joscelyne, A., Rawson, N., Maccallum, F., Cahill, C., et al. (2014). Treating prolonged grief disorder: a randomized clinical trial. JAMA Psychiatry, 71(12), 1332–1339. Scholar
  7. Byrd, K. R., Patterson, C. L., & Turchik, J. A. (2010). Working alliance as a mediator of client attachment dimensions and psychotherapy outcome. Psychotherapy Theory Research Practice Training, 47(4), 631–636.CrossRefGoogle Scholar
  8. Capaldi, S., Asnaani, A., Zandberg, L. J., Carpenter, J. K., & Foa, E. B. (2016). Therapeutic alliance during prolonged exposure versus client-centered therapy for adolescent posttraumatic stress disorder. Journal of Clinical Psychology.
  9. Eisma, M. C., Boelen, P. A., van den Bout, J., Stroebe, W., Schut, H. A., Lancee, J., et al. (2015). Internet-based exposure and behavioral activation for complicated grief and rumination: a randomized controlled trial. Behavior Therapy, 46(6), 729–748. Scholar
  10. Falkenstrom, F., Granstrom, F., & Holmqvist, R. (2013). Therapeutic alliance predicts symptomatic improvement session by session. Journal of Counseling Psychology, 60(3), 317–328. Scholar
  11. Falkenstrom, F., Hatcher, R. L., & Holmqvist, R. (2015a). Confirmatory factor analysis of the patient version of the working alliance inventory—short form revised. Assessment, 22(5), 581–593. Scholar
  12. Falkenstrom, F., Hatcher, R. L., Skjulsvik, T., Larsson, M. H., & Holmqvist, R. (2015b). Development and validation of a 6-item working alliance questionnaire for repeated administrations during psychotherapy. Psychological Assessment, 27(1), 169–183. Scholar
  13. Feeley, M., DeRubeis, R., & Gelfand, L. (1999). The temporal relation of adherence and alliance to symptom change in cognitive therapy for depression. Journal of Consulting and Clinical Psychology, 67(4), 578–582.CrossRefPubMedGoogle Scholar
  14. Fluckiger, C., Del Re, A. C., Wampold, B. E., Symonds, D., & Horvath, A. O. (2012). How central is the alliance in psychotherapy? A multilevel longitudinal meta-analysis. Journal of Counseling Psychology, 59(1), 10–17.CrossRefPubMedGoogle Scholar
  15. Gaston, L., Marmar, C. R., Gallagher, D., & Thompson, L. W. (1991). Alliance prediction of outcome beyond in-treatment symptomatic change as psychotherapy processes. Psychotherapy Research, 1, 104–112.CrossRefGoogle Scholar
  16. Glickman, K., Shear, M. K., & Wall, M. M. (2016). Mediators of outcome in complicated grief treatment. Journal of Clinical Psychology.
  17. Guy, W. (1976). CGI clinical global impressions ECDEU Assessment Manual for Psychopharmacology (Revised). (pp. 217–222): National Institute of Mental Health.Google Scholar
  18. Horvath, A. (1989). Development and validation of the Working Alliance Inventory. Journal of Counseling Psychology, 36(2), 223–233.CrossRefGoogle Scholar
  19. Horvath, A., & Symonds, B. D. (1991). Relation between working alliance and outcome in psychotherapy: a meta-analysis. Journal of Counseling Psychology, 38(2), 139–149.CrossRefGoogle Scholar
  20. Horvath, A., Fluckiger, C., Del Re, A. C., & Symonds, D. (2011). Alliance in individual psychotherapy. Psychotherapy, 48(1), 9–16.CrossRefPubMedGoogle Scholar
  21. Johansson, P., & Hoglend, P. (2007). Identifying mechanisms of change in psychotherapy: mediators of treatment outcome. Clinical Psychology and Psychotherapy, 14, 1–9.CrossRefGoogle Scholar
  22. Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology, 3, 1–27.CrossRefPubMedGoogle Scholar
  23. Klein, D. N., Schwartz, J. E., Santiago, N. J., Vivian, D., Vocisano, C., Castonguay, L. G., et al. (2003). Therapeutic alliance in depression treatment: controlling for prior change and patient characteristics. Journal of Consulting and Clinical Psychology, 71(6), 997–1006. Scholar
  24. Krupnick, J. L., Sotsky, S. M., Simmens, S., Moyer, J., Elkin, I., Watkins, J., et al. (1996). The role of the therapeutic alliance in psychotherapy and pharmacotherapy outcome: findings in the National Institute of Mental Health Treatment of Depression Collaborative Research Program. Journal of Consulting and Clinical Psychology, 64(3), 532–539.CrossRefPubMedGoogle Scholar
  25. Lorenzo-Luacas, L., DeRubeis, R., & Webb, C. A. (2014). Client characteristics as moderators of the relation between the therapeutic alliance and outcome in cognitive therapy for depression. Journal of Consulting and Clinical Psychology, 82, 362–373. Scholar
  26. Lorenzo-Luacas, L., German, R. E., & DeRubeis, R. J. (2015). It’s complicated: the relation between cognitive change procedures, cognitive change, and symptom change in cognitive therapy for depression. Clinical Psychology Review, 41, 3–15. Scholar
  27. Markowitz, J. C., & Weissman, M. M. (2004). Interpersonal psychotherapy: principles and applications. World Psychiatry, 3(3), 136–139.PubMedPubMedCentralGoogle Scholar
  28. Mundt, J. C., Marks, I. M., Shear, M. K., & Greist, J. H. (2002). The Work and Social Adjustment Scale: a simple measure of impairment in functioning. British Journal of Psychiatry, 180, 461–464.CrossRefPubMedGoogle Scholar
  29. Murphy, R. (2009). How do psychological treatments work? Investigating mediators of change. Behaviour Research and Therapy, 47(1), 1.CrossRefPubMedPubMedCentralGoogle Scholar
  30. Ormhaug, S. M., Jensen, T. K., Wentzel-Larsen, T., & Shirk, S. R. (2014). The therapeutic alliance in treatment of traumatized youths: relation to outcome in a randomized clinical trial. Journal of Consulting and Clinical Psychology, 82(1), 52–64. Scholar
  31. Prigerson, H., Bierhals, A., Stanislav, K., Reynolds III, C., Shear, K., Day, N., et al. (1995). Inventory of complicated grief: a scale to measure maladaptive symptoms of loss. Psychiatry Research, 59, 65–79.CrossRefPubMedGoogle Scholar
  32. Rosner, R., Pfoh, G., Kotoucova, M., & Hagl, M. (2014). Efficacy of an outpatient treatment for prolonged grief disorder: a randomized controlled clinical trial. Journal of Affective Disorders, 167, 56–63. Scholar
  33. Shear, K. (2013). Complicated grief treatment: instructional manual for NIMH grants (introductory phase). New York: Center for Complicated Grief, Columbia University.Google Scholar
  34. Shear, K., Frank, E., Houck, P., & Reynolds III, C. (2005). Treatment of complicated grief: a randomized controlled trial. JAMA, 293(21), 2601–2608.CrossRefPubMedPubMedCentralGoogle Scholar
  35. Shear, M. K., Wang, Y., Skritskaya, N., Duan, N., Mauro, C., & Ghesquiere, A. (2014). Treatment of complicated grief in elderly persons: a randomized clinical trial. JAMA Psychiatry, 71(11), 1287–1295. Scholar
  36. Shear, M. K., Reynolds 3rd, C. F., Simon, N. M., Zisook, S., Wang, Y., Mauro, C., et al. (2016). Optimizing treatment of complicated grief: a randomized clinical trial. JAMA Psychiatry, 73(7), 685–694. Scholar
  37. Wampold, B. E., & Imel, Z. E. (2015). The great psychotherapy debate : the evidence for what makes psychotherapy work ((Second edition. ed.). New York: Routledge.Google Scholar
  38. Zuroff, D., & Blatt, S. (2006). The therapeutic relationship in the brief treatment of depression: contributions to clinical improvement and enhanced adaptive capacities. Journal of Consulting and Clinical Psychology, 74(1), 130–140.CrossRefPubMedGoogle Scholar

Copyright information

© International Association of Cognitive Psychotherapy 2018

Authors and Affiliations

  • Kim Glickman
    • 1
  • M. Katherine Shear
    • 2
  • Melanie M. Wall
    • 3
  1. 1.Department of Social Work, School of Health Sciences and Professional ProgramsYork College of the City University of New YorkNew YorkUSA
  2. 2.Columbia University School of Social WorkColumbia University College of Physicians and SurgeonsNew YorkUSA
  3. 3.Division of Biostatistics in the Department of Psychiatry, Department of Biostatistics in the Mailman School of Public HealthColumbia UniversityNew YorkUSA

Personalised recommendations