A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Counseling Psychology
This content is subject to copyright. Terms and conditions apply.
The Anger-Depression Mechanism in Dynamic Therapy:
Experiencing Previously Avoided Anger Positively Predicts Reduction
in Depression via Working Alliance and Insight
Joel M. Town
1
, Fredrik Falkenström
2
, Allan Abbass
1
, and Chris Stride
3
1
Department of Psychiatry, Dalhousie University
2
Department of Behavioral Sciences and Learning, Linköping University
3
The Institute of Work Psychology, University of Sheffield
A central tenet of psychodynamic theory of depression is the role of avoided anger. However empirical
research has not yet addressed the question of for which patients and via what pathways experiencing anger
in sessions can help. The therapeutic alliance and acquisition of patient insight are important change
processes in dynamic therapy and may mediate the anger–depression association. This study was embedded
into a randomized trial testing the efficacy of Intensive Short-Term Dynamic Psychotherapy (ISTDP) for
treatment resistant depression. In-session patient affect experiencing (AE) was coded for every available
session (475/481) by blinded observers in 27 patients randomized to ISTDP. Dynamic Structural Equation
Modeling was used to examine within-person associations between variation in depression scores session-
by-session and both patient ratings (alliance) and observer ratings (AE and insight) of the treatment process.
Alliance and insight were independent mediators of the effect of anger on next-session depression.
However, the relative importance of these two indirect effects of anger on depression was conditional
on pretreatment patient personality pathology (PP). In patients with higher PP, in-session anger was
negatively related to depressive symptoms next session, with this effect operating through higher alliance. In
patients with low PP, in-session anger was negatively related to depressive symptoms next session, with this
effect operating through enhanced patient insight. These findings highlight an anger–depression mechanism
of change in dynamic therapy. Depending upon patient personality, either an “insight pathway”or a
“relational pathway”may promote the effectiveness of facilitating arousal and expression of patients’in-
session feelings.
Public Significance Statement
This study highlights the importance of addressing avoided feelings of anger when treating depression in
dynamic therapy. The effectiveness of this approach involves monitoring the development of the
therapeutic alliance and acquisition of patient insight, according to a patient’s personality functioning.
Keywords: depression, psychodynamic, insight, anger, alliance
Supplemental materials: https://doi.org/10.1037/cou0000581.supp
Looking beyond the results of efficacy studies for informing
treatment recommendations for Major Depressive Disorder (MDD),
psychotherapy research exploring mechanisms of change aims to
test the clinical theories that therapists are recommended and
trained to use in practice. The conceptualization of depression
as a psychological state of inverted anger is a central principle
when treating MDD in psychodynamic therapies. Studies have
shown that depressed patients commonly report suppressing anger
This article was published Online First September 30, 2021.
Fredrik Falkenström https://orcid.org/0000-0002-2486-6859
Allan Abbass https://orcid.org/0000-0003-1285-5770
Chris Stride https://orcid.org/0000-0001-9960-2869
We have no conflict of interest to disclose. The data reported in this
article have been previously published and were collected as part of a
larger data collection. The findings from the data collection have been
reported in separate articles. MS 1 (Town, Abbass, et al., 2017) focuses on
change in PHQ-9 scores at baseline, 3 to 6 months. MS 2 (Town et al.,
2020) focuses on change in HAM-D, PHQ-9, GAD-7, IIP-32, PHQ-15
scores from baseline to 18 months and a cost-effectiveness analysis. MS 3
(Town et al., 2022) focuses on the associations between scores on PHQ-9,
ATOS affect experiencing scale, ATOS insight scale, Agnew Relation-
ship Measures (ARM-5) over weekly sessions.
Correspondence concerning this article should be addressed to Joel M.
Town, Department of Psychiatry, Dalhousie University, Abbie J. Lane
Building, 7
th
Floor, Room 7507, 5909 Veteran’s Memorial Lane, Halifax,
NS B3H 2N1, Canada. Email: joel.town@dal.ca
Journal of Counseling Psychology
© 2021 American Psychological Association 2022, Vol. 69, No. 3, 326–336
ISSN: 0022-0167 https://doi.org/10.1037/cou0000581
326
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
(Gilbert et al., 2004) and turning anger inwards correlates with
higher levels of depressive symptoms (Painuly et al., 2005).
However, to date psychodynamic mechanisms research has not
studied the association between patients experiencing anger in-
sessions and subsequent levels of depression symptoms.
A Psychodynamic Model of Depression
There are multiple psychodynamic accounts of depression (Blatt,
1974;Bowlby, 1973;Freud, 1957;Klein, 1935). Although there is
not a unified psychodynamic theory of depression, we offer a
synthesis of key ideas. Psychodynamic theory posits that experi-
ences of actual or perceived loss in relationships lead to feelings of
anger toward the other, and to intolerable guilt about the anger. The
individual attempts to cope by unconsciously defending against the
anger and guilt by turning the anger against the self, resulting in
depressive symptoms. As a result, patients present with chronic
irritability, self-reproach, and aggression toward the self (Busch,
2009;Freud, 1957). Feelings of sadness related to the loss of a
wished-for state can similarly fail to be adequately acknowledged.
Instead, a person can experience a persistent state of hopelessness or
pathological mourning that prevents them from moving forward
(Bowlby, 2008). The individual’s subjective experience of loss and
depression is related to their self-other representations and to
features of their personality (Blatt, 1998). The preponderance of
negative representations of self and others, and the negative feelings
such as anger that results are themselves defended against, at great
cost to the individual. The experience of loss that lies at the heart of
depression is preserved in autobiographical affective memory struc-
tures, and resolution of depression is assumed to require affective
arousal and experiencing (AE) to access these structures. Emotions
related to past adverse events can then be processed in a new and
different way. Emerging neurobiological findings (Lane, 2018)
support this assumption.
This theoretical understanding of etiological factors in depres-
sion, suggests a model for changing depressogenic thoughts, feel-
ings, and behaviors in dynamic therapy. Such a model should
account for the role of patients’personality functioning, on the
putative process of AE. The putative in-session process will involve
a central focus on anger within relationships as a conflicted affective
state. Alongside anger, examining other feelings such as sadness and
guilt about anger, associated to adverse relational experiences,
would reflect the perspective of a multidimensional role of affect
in treating depression.
Affect–Depression Change Mechanism in
Dynamic Therapy
As discussed, dynamic therapy for MDD assumes that the
activation and subsequent dysregulation of conflicted emotions
such as anger precedes the emergence of depressive symptoms
(unprocessed affect →emergence of depression). Empirical studies
of emotional processing as a psychotherapy change mechanism
(Peluso & Freund, 2018) and specifically in dynamic therapy
(Diener et al., 2007) have been reviewed, indicating a positive
association between increased experiencing and outcomes. Two
recent studies, demonstrate the findings of a positive association
between AE and outcome in dynamic therapy (Fisher et al., 2016;
Keefe et al., 2019) and provide strong evidence that AE contributes
to improved outcomes in dynamic therapy rather than being a
product of ongoing symptom change.
In a precursor to the present study, using a single-case repli-
cation design, Town, Salvadori, et al. (2017) replicated these
findings in dynamic therapy for MDD. However, current research
does not describe the mechanism by which in-session processing
of anger, or other attachment related affects, drives change in
depressive symptoms.
Moderators and Mediators of an Affect–Depression
Association
Personality
Depressions differ depending on the patient’s personality,
among other factors (Gabbard & Simonsen, 2007), so that the
interrelationship between depressive symptoms and personality
have important implications for treatment. Empirical studies have
shown the magnitude of treatment effects are moderated by a
variety of primary manifestations of personality organization—
including pretreatment levels of personality pathology (Koelen
et al., 2012); attachment style (Diener & Monroe, 2011); degree of
object relations (Piper et al., 2001); alexithymia (Ogrodniczuk
et al., 2011); and self-criticism (Blatt et al., 2010). It is assumed
that patients with higher levels of personality organization are
likely to have more adaptive psychological structures, allowing
them to more readily utilize dynamic interventions in therapy to
activate mechanisms of change such as AE, with the resultant
therapeutic changes.
To our knowledge, this assumption has only been tested in one
study. Keefe et al. (2019) examined the moderating effect of
personality disorder traits on the relationship between emotional
expression and symptom improvement in panic focused psychody-
namic psychotherapy. They found that patients with more primitive
personality organizations, indicated by meeting two or more DSM
criteria for borderline personality disorder, showed no beneficial
AE-outcome relationship. Posthoc observations of the AE-outcome
association in dynamic therapy for MDD also suggested that
impairment in personality functioning could account for the lack
of significant process-outcome associations (Town, Salvadori,
et al., 2017).
Based upon these theories and empirical data, AE may be better
modeled as having an indirect effect on outcomes via multiple
(mediator) variables, with the identity of the most active mediator
conditional on patient personality characteristics. The early devel-
opment of talking therapy highlighted two candidate mediators: The
role of insight through interpretation, versus supportive or relation-
ship aspects of treatment. Both patient insight (Jennissen et al.,
2018) and the therapeutic alliance (Flückiger et al., 2018) have since
been established as predictors of psychotherapy outcome. In psy-
choanalysis, acquisition of insight was initially viewed as the
primary vehicle for change; however, to broaden therapy to fit
different patients, treatments began to emphasize relational factors
(Alexander & French, 1946). For patients with personality
impairment, who have limited anxiety tolerance and utilize more
primitive defenses, “supportive”interventions that cement the
relationship are assumed to be particularly important. Whereas
dynamic techniques promoting insight have been described as
ANGER-DEPRESSION MECHANISM IN DYNAMIC THERAPY 327
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
“expressive”interventions and are recommended to the degree a
patient has adequate ego capacity and ability to reflect upon
relationships.
Therapeutic Alliance
Recent research on patterns of alliance development over psy-
chotherapy sessions suggests that different patterns do exist across
patients and one size does not fit all (Zilcha-Mano & Errázuriz,
2015). Patients presenting with higher levels of personality difficul-
ties may depend more on the alliance for positive treatment out-
comes (Falkenström et al., 2016;Zilcha-Mano & Errázuriz, 2015).
These studies indicate that subgroups of patients based on person-
ality factors may benefit from distinct patterns of alliance develop-
ment. Furthermore, alliance-outcome associations may also differ
between modalities when individual patient characteristics are
controlled for (Bedics et al., 2015;Zilcha-Mano et al., 2015). These
findings point to an interaction between therapeutic alliance,
patient characteristics, and other therapeutic ingredients to pre-
dict outcomes.
Insight
Psychodynamic theorists have described the putative function of
insight as enabling patients to find new solutions or more adaptive
ways of behaving, which in turn lead to improvements in symptoms
(Gabbard, 2014). Empirical studies of dynamic therapy have shown
that insight increases over treatment (Gibbons et al., 2009), and is
generally associated to symptom change (Gibbons et al., 2009;
Johansson et al., 2010). Secondary analyses of three randomized
controlled trials of dynamic therapy found that patient insight into
dynamic patterns acted as a mediator of outcomes (Johansson et al.,
2010;Kallestad et al., 2010)—and that improved insight is neces-
sary for long-term treatment effects (Høglend & Hagtvet, 2019).
Rather than attempting to confirm or deny past polarized positions
on insight as the primary active ingredient in dynamic therapy, it is
more likely that in some circumstances, and for specific patients,
eliciting insight is especially impactful.
Present Study
Time-limited Intensive Short-Term Dynamic Psychotherapy
(ISTDP; Abbass, 2015;Davanloo, 2000) for MDD is a 20-session
treatment that is efficacious and cost-effective for treatment resistant
depression in one study conducted in Canada (Town, Abbass, et al.,
2017;Town et al., 2020). ISTDP focuses on mobilizing and
experiencing complex emotional states, including unacknowledged
anger toward attachment figures. Through recognizing and
experiencing emotions, the patient is hypothesized to rely less on
implicit tendencies toward defensive avoidance of emotions that
perpetuate depressive symptoms.
Collectively, the current theoretical and empirical literature points
to several key findings regarding putative processes of change in
dynamic therapy relevant to ISTDP and the optimal treatment for
depression. First, studies by Fisher et al. (2016) and Keefe et al.
(2019) demonstrate that in-session patient AE is an independent
predictor of improvement in symptom difficulties rather than a
consequence of improvements. However, to quantify dynamic
theory that AE is a treatment mechanism in depression, this
session-to-session process-outcome association should be replicated
for change in depression symptoms. Furthermore, the putative role
of patients experiencing anger requires confirmation and secondary
analyses should quantify the role of guilt about anger and sadness.
Second, while empirical research has shown that pretreatment levels
of patient personality functioning can moderate the effect of therapy
on treatment outcomes, only one study provides evidence that
personality characteristics may affect capacity for in-session AE
(Keefe et al., 2019). Third, although empirical findings suggest that
developing a patient–therapist alliance and the acquisition of patient
insight are important in dynamic therapy, it is much less clear if, and
potentially how, these variables interact with in-session patient AE
to facilitate change in symptoms. Multiple measures of personality
functioning exist beyond a categorical nosology of personality
disorder. As patients with high levels of personality organization
typically exhibit fewer primitive defenses, experience fewer inter-
personal problems, and better capacity for self-reflection and insight
on emotions (McWilliams, 2011), we believe insight is more likely
to be an active ingredient in therapy for these patients. On the other
hand, for patients with lower personality organization who experi-
ence difficulties in reality-testing (seen with primitive defenses such
as projection and splitting) and affect regulation (heightened alex-
ithymia), we expect a more central role for a strong alliance in
mediating the helpfulness of experiencing anger. There is accumu-
lating evidence to suggest that the therapeutic alliance is a particu-
larly important change process for patients with a greater burden of
personality problems (e.g., Falkenström et al., 2016). In these
patients, we believe a conscious therapeutic alliance can be under-
stood as a marker of sufficient restructuring of primitive defense and
difficulties observing emotions. AE can then lead to greater inte-
gration of emotions through development of a strong therapeutic
relationship, rather than emotions being disowned using primitive
defenses.
Hypotheses
The following hypotheses were made a priori:
Hypothesis 1: The association between in-session AE of
anger and self-reported depressive symptoms in the next
7 days will be moderated by patient pretreatment personality
pathology (PP), such that patients with higher PP will be
more likely to report a weaker negative relationship between
AE and depression.
Hypothesis 2: The relationship between AE of anger and
depression will operate indirectly, via two mediators, the
therapeutic alliance and patient insight, with these indirect
effects conditional: (a) we expect that the alliance will be the
more critical mediator for patients with higher pretreatment PP
and, (b) insight will be the more critical for patients with lower
pretreatment PP
We will test each hypothesis using three models of AE: The
primary analyses will be conducted on ratings of patient’s
experiencing of anger, secondary analyses will be conducted
on ratings of patient’s experiencing of sadness and also guilt
about anger.
328 TOWN, FALKENSTRÖM, ABBASS, AND STRIDE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
Method
Participants
This study combines the collection of new observer rated process
data with secondary analysis of self-report process and outcome data
for participants receiving time-limited ISTDP collected as part of the
Halifax Depression Study (Town, Abbass, et al., 2017). The origi-
nal superiority trial used a single blind randomized parallel group
design to examine the efficacy of ISTDP versus secondary care
treatment provided by community mental health teams (CMHTs),
for treatment resistant depression (TRD). The original trial protocol
was registered with ClinicalTrials.gov (ID: NCT01141426) and
both studies approved by the Nova Scotia Health Authority
Research Ethics Board (NSHA-RS/2013-049). All participants
provided written informed consent.
The Halifax Depression Study included eligible patients aged
18–65 years, with a primary diagnosis of major depressive disorder
according to DSM-IV criteria. Patients met study criteria for TRD by
having had at least one trial of antidepressants at the adequate
recommended therapeutic dose; a current depressive episode duration
of 6 or more weeks; inadequate response to treatment (assessed by
17-item HAM-D score ≥16); not having started further medication or
changed dose of existing medication in the previous 6 weeks; and not
having received treatment in the previous 2 years at any of the four
CMHTs. 60 participants were allocated to ISTDP or CMHT treat-
ment in a 1:1 ratio (i.e., 30 patients randomly assigned to each group).
The final sample for the present study, which utilizes data from just
the ISTDP group, was 27, after two participants failed to start ISTDP
and one participant received only one session. The mean age of the
participants in the ISTDP group was 38.9 years (SD =11.87); 17
(56.7%) were women; all were White; 25 (86.2%) had comorbid
personality disorder of which 21 (70%) met criteria for a Cluster C
personality disorder; 28 (93.3%) had a comorbid Axis I disorder.
Audio–visual recordings of treatment provided within CMHTs were
not available therefore it was not possible to collect observer rated
process data from this treatment arm for the purposes of this study.
Treatment
The ISTDP model is a brief psychotherapy format that helps the
patient identify and address the emotional factors that culminate into,
exacerbate, and perpetuate depression. The treatment provided is
discussed in detail in our earlier study (Town, Abbass, et al., 2017).
ISTDP was provided according to a 20-session time limited, individual
format, and delivered according to published recommendations
(Abbass, 2015;Davanloo, 2000).ThemeannumberofISTDPsessions
completedintheRCTwas16.1(SD =6.68) across 30 patients. In the
present study sample, 475 sessions were available from a total of 481,
nested within 27 patients. Any missing data was due to a problem
recording the treatment session. ISTDP therapists were licensed
professionals with supervised experience practicing ISTDP (mean
experience =10.25 years, range =4–20 years). The integrity of the
ISTDP intervention as a form of dynamic therapy was established by
trained independent researchers (Town, Abbass, et al., 2017).
Outcome Measure
The primary outcome measure for this study was the nine-item
Patient Health Questionnaire, PHQ-9 (Kroenke et al., 2001). The
PHQ-9 is a brief self-report questionnaire for measuring the severity
of symptoms of depression, demonstrating good reliability and
validity in psychometric studies (Kroenke et al., 2001). Internal
consistency was high for the PHQ-9 (Cronbach’sα=.900) The
PHQ-9 was completed by each patient at baseline, and before each
psychotherapy session.
Moderator Variable
Central impairments in personality functioning have been
described in interpersonal relationships and underlying difficulties
in mental representations of self and other (Kernberg, 1984;Pincus,
2005). A composite measure of personality pathology (PP) was thus
derived from three reliable and validated self-report scales: Toronto
Alexithymia Scale-20 (TAS-20; Bagby et al., 1994); the Inventory
of Interpersonal Problems 32-item (IIP-32; Horowitz et al., 2000);
and the Defense Style Questionnaire (DSQ-40; Andrews et al.,
1993). These scales relate to domains of functioning common across
personality pathology: Affective, social-interpersonal and cognitive
style, respectively (Mischel & Shoda, 2008). Alexithymia, interper-
sonal functioning and defense style are considered distinct but
overlapping dimensions of personality functioning. The decision
to combine measures in a composite score enables an examination of
one global metric of personality functioning rather than multiple-
related scales. These moderator scales were assessed at baseline,
prior to study randomization.
The IIP-32 was completed to assess severity of interpersonal
problems. Previous research has demonstrated this version of the
IIP-32 has a 7-day test–retest reliability coefficient of r=0.78
(Horowitz et al., 2000) and good convergent validity with other
self-report personality measures (Morse & Pilkonis, 2007). The
IIP-32 had an internal consistency of α=.85.
The TAS-20 is a 20-item patient self-report measure that was used
to assess the degree to which a participant could be considered
alexithymic. Alexithymia is defined as impairment in the ability to
understand, process, and describe emotions. The convergent, dis-
criminant, and concurrent validity of the TAS-20 have been shown
to be good (Bagby et al., 1994). The TAS-20 was internally
consistent in this study (α=.81).
The DSQ-40 is a patient self-report measure assessing patients’
conscious awareness of their characteristic style of dealing with
conflict. It yields three higher order factors relating to mature,
neurotic, and immature defense styles. Previous research has re-
ported the psychometric properties, including high internal consis-
tency and temporal stability appropriate in a state measure (Andrews
et al., 1993). The findings from one meta-analysis showed the DSQ
three-factor structure has discriminant validity for MDD (Calati
et al., 2010). In this study, the DSQ-40 immature scale had an
internal consistency of α=.70.
Process Measures
Self-Report Measures
Participant rated therapeutic alliance data were collected using the
five-item Agnew Relationship Measure (ARM-5), completed imme-
diately after each therapy session. It is a short-form version of the
28-item measure developed to represent an overall alliance score
(Agnew-Davies et al., 1998). The ARM-5 had an internal
ANGER-DEPRESSION MECHANISM IN DYNAMIC THERAPY 329
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
consistency of α=.89. Previous demonstrated ARM-5 has accept-
able psychometric properties (Cahill et al., 2011).
Observer Rated Measure
The Affective Experiencing Scale (AES) and Insight Scale (IS),
taken from the Achievement of Therapeutic Objectives Scale
(ATOS; McCullough et al., 2003), were used in the present study
to measure patient emotional arousal and patient insight, respec-
tively. Previous studies have found that the ATOS has adequate
psychometric properties. Using generalizability analyses, Berggraf
et al. (2014) demonstrated that ATOS is sensitive to differences
among patients and differences were found among subscales within
patients. They reported a generalizability coefficient of .90 and .88
on the AES and IS, respectively, indicating the scales can be used
across patient samples. Evidence of the validity of the ATOS
subscales include studies that examined the theoretically derived
factor structure (Ryum et al., 2014), predicted relationships with
other process variables (Town et al., 2012) and outcome variables
(Berggraf et al., 2014).
For AE and insight ratings, each session is divided into 10-min
segments and rated using audio–visual session recordings. Insight is
defined as recognition of links between maladaptive patterns of
anxiety, defense, and feelings, as operationalized in the Triangle of
Conflict (Malan, 1979). At higher levels of Insight, connections
between the Triangle of Conflict and Triangle of Person (Malan,
1979) are seen. Using the IS, raters consider the clarity of patient’s
description of maladaptive patterns and ability to describe why and
how the patterns are maintained. For AE, raters consider three
components of emotional arousal grounded in behavioral examples:
Peak degree of arousal, duration of the affective response, and relief
in the experience of the feeling. AE is considered adaptive when
feelings about another’s perceived or actual actions can be tolerated
without a preponderance of defensive affect or anxiety. A score is
then awarded between 1 and 100, with higher scores reflecting
greater Insight and fuller AE. For the purposes of this study, the
original ATOS manual was modified to standardize the coding for
ISTDP material (ATOS-I; Town et al., 2014). Judges were trained
and instructed to rate three specific affect categories on the AES:
AES Anger. Ratings of anger were defined as a patient ex-
pressing in-session, and to some degree experiencing, angry feelings
toward another. Anger was typically rated when patients cognitively
identifying reactive anger related to a perceived theme of an unmet
attachment need, trauma or abuse. This could relate to current, past
or the therapeutic relationship (transference) with the therapist.
Healthy anger was differentiated from maladaptive expressions of
primarily anxiety or defense, which may take the form of a discharge
of tension, a tantrum, or self-criticism. Higher ratings required
greater evidence of in-session bodily arousal.
AES Sadness. Sadness was defined as an emotional experience
related to the actual or perceived loss of a wished-for state within an
important relationship. Ratings of adaptive sadness are easily
confused with tears associated with hopelessness, helplessness,
shame or heightened anxiety. Sadness related to the impact of a
patient understanding the damaging impact of behavioral or inter-
personal patterns (defenses) was coded on a different ATOS scale.
AES Guilt About Anger. The adaptive components of guilt
come when patient experiences regret, typically over imagined
thoughts of doing harm toward someone they care about. Some
of the components include a verbal report of regret, patients
describing constriction in the upper chest, a wish to reverse what
was done, a desire for reunification and showing caring/tender
feelings toward the target of their anger. Guilt about the anger
will typically be accompanied by tears as experiencing increases.
Adaptive guilt is differentiated from thoughts of self-loathing and
shame, dominating the person in a self-critical or punitive manner.
Procedure
Judges and Training
Observer ratings of AE and Insight were conducted by 10 judges,
four Bachelor Honors level psychology students, two psychology
Masters students, and four clinical psychology PhD students. Judges
were provided 16–20 hr of training on four of the ATOS scales,
including the AES and IS, by an experienced ATOS rater. Judges
then rated a series of training tapes to assess rater reliability against
expert generated ratings. To participate, all judges were required to
achieve a reliability criterion of greater or equal to .70.
Rating Procedure
Judges worked in pairs to rate an entire ISTDP treatment course for
a participant. They were given the ATOS-I manual (Town et al.,
2014), written instructions identifying the participant code and anon-
ymized tape number, to be coded. Judges viewed sessions in 10-min
segments, pausing between each to independently generate a ratingon
the relevant scales including the ATOS-AES for each affect category.
Consensus was then reached on the final scores to be awarded through
discussion. Coding drift was monitored through regular meetings to
review exemplar material alongside established coding criteria.
Judges’pairings were also rotated to minimize the possibility of
further drift. Interrater reliability was calculated between the two raters
using a two-way random effects model [ICC 2,1] for each 10-min
segment. The judges demonstrated ICC values in the good range
(.61–.80) based on Shrout and Fleiss (1979), on the IS, insight =.771,
and in the excellent range (>.81) across each affect category on the
AES, anger =.863, sadness =.831, guilt =.863.
Statistical Analyses
The data in this study consisted of repeated observations (ses-
sional data) for each patient, with study variables collected at each
session therefore having within and between patient variance com-
ponents. Given the large number of sessions (20), Dynamic Struc-
tural Equation Modeling (Asparouhov et al., 2018), which combines
aspects of time-series analysis (traditionally used in single-case
designs with a large number of time points) with Multilevel Model-
ing and Structural Equation Modeling offered the most suitable
analytic structure for testing our hypotheses. (Schultzberg &
Muthén, 2018).Figure 1 shows a path diagram of the final moder-
ated mediation model.
We did not model therapist effects, since Falkenström et al.
(2020) recently showed using Monte Carlo simulations that this
does not affect estimates of within-patient effects, and with small
number of therapists it may increase bias. Because of the complexity
of the models and the relatively small sample, we built the models in
steps starting with separate bivariate regression models, then putting
330 TOWN, FALKENSTRÖM, ABBASS, AND STRIDE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
these together into mediation models, and finally testing the mod-
erated mediation models. This sequential approach enabled trouble-
shooting so that any large deviations between results from the
simpler models and the more complex ones would be detected
and checked early on. However, since results changed very little
between these steps, we present results only for the final moderated
mediation models. Recommendations for interpretations of effects
are provided within the Online Supplement.
Models were estimated using Bayesian imputation, in which each
missing value has its own posterior distribution. This approach
assumes data is Missing At Random (Asparouhov & Muthén, 2010).
Power Analysis/Test of Estimator Performance
Due to the small sample and the complex models analyzed, we ran a
series of Monte Carlo simulations to check statistical power and
estimator performance. For the primary within-patient paths (i.e.,
AE →insight/alliance, and insight/alliance →depression), statistical
power to find small standardized effects (β=.10) was between 54%
and 68%, while for medium-sized effects (β=.20) power was 98%–
100%. Average coefficient bias was small (1.2–2.9%), and coverage
of 95% credible intervals was excellent (93.8–95.1%). Power for
indirect effects was 35%–36% for small effects (a ×b=0.01) and
97%–98% for medium-sized effects (a ×b=0.04). Statistical power
was reasonably good even with an Nas small as 27 because power for
within-person effects are determined not just by N,butalsobyT—the
number of repeated measurements.
Although the excellent coverage of the 95% credible intervals
should ensure correct Type-I error rate, we reran the simulations
with all population coefficients set to zero to test the “empirical α
level,”that is, the proportion of times the estimator yields a
statistically significant coefficient estimate despite it being zero
in the population. This should be close to 5%. Results showed
that the largest αlevel for any within-patient coefficient was 6.1%,
for between-patient coefficients it was at most 3.8%, and for indirect
effect estimates it was 0.3%. Thus, there was no indication of
increased levels of spurious coefficients.
Results
Table 1 shows descriptive statistics for all included variables.
Although 475/481 sessions were rated using the ATOS, the fre-
quency of ratings on the AES and IS do not correspond exactly to the
number of rated sessions as observable evidence of these processes
was not evident in every session. There were moderate to strong
intercorrelations among the IIP-32, TAS-20, and DSQ-40 immature
Figure 1
Path Diagram of Moderated Mediation Model With Alliance and Insight as Med-
iators of the Affect (A) →Depression Path, With Personality Pathology as the
Moderator
Note. The model is a two-level Dynamic Structural Equation Model, with random intercepts u1
for Alliance, u2for Insight, and u3for Depression.Affect, Alliance, and Insight are all entered for
session t−1, while Depression is entered for session t. The moderator Personality Pathology is
allowed to impact the paths from A to the mediators Alliance and Insight (Paths m1 and m2), as
well as the direct effect on Depression (Path m3). Latent centering is used for all endogenous
variables, while manual centering is used for the exogenousone (A). Primary moderatedmediation
paths are labeled and black, grayscale arrows are auxiliary (control and model setup) paths.
ANGER-DEPRESSION MECHANISM IN DYNAMIC THERAPY 331
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
subscale (.33 <r<.66). These were combined into the PP index by
first standardizing each variable and then taking their arithmetic
mean. This PP index had an internal consistency of α=.77. The
correlations between the three affects anger, guilt, and sadness were
moderate (.47 <r<.49). We chose to enter these as separate
predictors given the potential theoretical and clinical importance if
results are different among these affects.
Direct Effects of Anger, Guilt, and Sadness on
Next-Session Depression
There was a moderated direct effect of anger on next-session
depression, PP ×anger 0.08, SD =0.03, p=.004, 95% CI [0.03,
0.14], with simple slopes analysis showing that only at low PP (one
SD below average) was there a direct effect of more experience of
anger predicting less severe depressive symptoms, −0.10, SD =
0.04, p=.03, 95% CI [−0.19, −0.02]. For guilt and sadness, the
direct effect was statistically nonsignificant, and there was no
moderation effect (all p>.40). However, this does not preclude
the possibility of mediation effects (Hayes, 2009).
The Indirect Effect of Experiencing In-Session
Anger on Next-Session Depression
Table 2 shows the results from the moderated mediation model
including both insight and alliance as mediators of the anger →
depression effect. As an omnibus test of all coefficients involved in
the moderated mediation, we compared the Deviance Information
Criterion (DIC) between a model in which all of these coefficients (8
in total) were set to zero and one in which all coefficients were freely
estimated. This comparison favored the model in which the moder-
ated mediation parameters were estimated (DIC
null
−DIC
est
=
45.16). PP significantly moderated the paths from anger →insight,
interaction effect =−0.09, SD =0.04, p=.03, 95% CI [−0.16,
−0.01], and anger →alliance, interaction effect =0.09, SD =0.03,
p<.001; 95% CI [0.04, 0.14]. Simple slopes analysis showed that at
low (1 SD below the mean) and at mean PP, insight was a significant
mediator in the hypothesized direction, low PP indirect effect =
−0.02, SD =0.01, p=.03, 95% CI [−0.04, −0.00]
1
, mean PP
indirect effect =−0.01, SD =0.01, p=.03, 95% CI [−0.03, −0.00],
but at high PP (1 SD above the mean) insight was not a significant
mediator (p=.43). For alliance, the opposite was the case, with
significant mediation only at high PP, indirect effect =−0.01,
SD =0.01, p<.05, 95% CI [−0.03, −0.00]. Figure S1 (see Online
Supplement) shows the indirect effects with 95% credible intervals
from −2 standard deviations below to +2 standard deviations
above mean PP and Figure S2 shows the simple slope estimates
by personality pathology.
There was also significant moderation of the direct effect of
experiencing anger on next-session depression, interaction effect =
0.09, SD =0.03, p<.001, 95% CI [0.04, 0.14]. This time, with both
mediators included in the model, simple slopes analysis showed that
at low PP the direct effect was significantly negative, direct effect =
−0.10, SD =0.04, p=.03, 95% CI [−0.19, −0.01], that is,
indicating that experience of anger positively predicted improve-
ment in depressive symptoms by the next session. However, at high
PP the direct effect was positive, indicating that more experience of
anger predicted deterioration in depressive symptoms by the next
session, direct effect =0.08, SD =0.04, p=.03, 95% CI [0.01,
0.16]. Figure S3 (see Online Supplement) shows the direct effect
with 95% credible intervals from −2 standard deviations below to
+2 standard deviations above mean PP.
The Effect of Experiencing Guilt in the Session
When guilt was used as predictor, the omnibus test again favored the
moderated mediation model over the null model (DIC
null
−DIC
est
=
38.49). The moderator effects for insight was statistically significant,
interaction effect =−0.12, SD =0.05, p=.02, 95% CI [−0.22,
−0.02], with simple slopes analysis indicating the same pattern as for
anger with significant mediation at low, indirect effect =−0.03,
SD =0.02, p=.01, 95% CI [−0.07, −0.01], and mean PP, indirect
effect =−0.02, SD =0.01, p=.01, 95% CI [−0.05, −0.01], but not at
high PP (p=.09). For alliance, the moderation by PP was nonsignifi-
cant (p=.98). When the analysis was rerun without the moderation of
PP ×alliance, mediation was not quite significant for the guilt →
alliance →depression, indirect effect =−0.01, SD =0.01, p=.054,
95% CI [−0.02, 0.00]. The direct effect was also nonsignificant
(p=.65) and there was no moderation for the direct effect (p=.68).
The Effect of Experiencing Sadness in the Session
For Sadness, the results were very similar to the results for guilt,
again with the omnibus test favoring the moderated mediation
model (DIC
null
−DIC
est
=38.28). The sadness ×PP →Insight
moderation was statistically significant, interaction effect =
−0.12, SD =0.06, p=.04, 95% CI [−0.24, −0.00], with simple
Table 1
Descriptive Statistics for Variables Used in Analysis
Variable NMSDMin Max
Patient health questionnaire—9439 14.33 6.28 0.00 27.00
Agnew relationship measure 470 6.17 0.80 2.60 7.00
Insight 472 42.10 6.61 17.00 80.00
Anger 420 37.54 8.58 20.75 83.00
Grief 273 41.00 10.71 15.00 81.50
Guilt 298 39.38 10.93 15.00 87.00
Toronto alexithymia scale 26 60.54 11.72 36.00 83.00
Inventory of interpersonal problems 27 1.54 0.52 0.62 2.75
Defense style questionnaire (immature) 26 3.85 0.90 2.58 5.46
1
Due to rounding decimal places, credible intervals may include .00, for
instance when the coefficient is negative and the upper limit is very close to
zero.
332 TOWN, FALKENSTRÖM, ABBASS, AND STRIDE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
slopes analysis showing significant mediation at low, indirect effect
=−0.03, SD =0.01, p=.01, 95% CI [−0.06, −0.00], and mean PP,
indirect effect =−0.01, SD =0.01, p=.01, 95% CI [−0.04, −0.00],
but not at high PP (p=.64). The sadness ×PP →alliance
moderation was nonsignificant (p=.40), and reestimating without
this interaction showed that the indirect effect was not quite
significant, indirect effect =−0.01, SD =0.01, p=.056, 95%
CI [−0.02, 0.00]. Also, the direct effect was nonsignificant
(p=.79) and there was no moderation of the direct effect ( p=.19).
Discussion
We aimed to test a psychodynamic theory of change in depres-
sion, by examining the effect of a patient experiencing, and ex-
pressing feelings of anger in sessions, on levels of depression
symptoms at the next session. Prospectively embedding this study
into an RCT design importantly allowed us to establish a measure-
ment timeline that enabled the anger–depression mechanism
(↑anger →↓depression) to be elaborated by testing with which
patients and via what pathways does experiencing negative feelings
promote reduced depressive symptoms.
Anger–Depression Mechanism of Change
This is the first study to demonstrate that in dynamic therapy for
MDD, patients experiencing anger in-session positively predicts the
degree of reduction in depressive symptoms 7 days later. Consistent
with dynamic theory, we found that this association was conditional
on the moderating role of patient personality functioning. This result
underscores our central hypothesis that facilitating AE of anger to
reduce depression is more accurately understood through the lens of
differences in patients’personality functioning (PP ×anger →
depression).
Personality Factors: A Relational Path for Some,
an Insight Path for Others
A second key new finding disproves the view of a single pathway
of change in dynamic therapy for depression. The current moderated-
mediation findings extend clinical theory (McWilliams, 2011;
Westen et al., 2006) by describing two pathways for personalizing
dynamic therapy based upon patients’personality functioning. For
patients who typically experience difficulties holding a balanced and
integrated sense of self and others, following the mobilization of
emotions in-session, a relational path, evidenced by an enhanced
alliance such as an improving bond with the therapist and clearer task
agreement, can be tracked to indicate a positive therapeutic process
(high PP ×↑anger →↑alliance →↓depression). On the other hand,
for patients with generally more positive and stable perceptions of self
and others, an insight-based path, that helps them to experience and
express their feelings is beneficial when it allows for a deeper
emotional insight (low PP ×↑anger →↑insight →↓depression).
The importance of insight is consistent with the principle of
patients needing to consciously extract meaning from an emotional
response (Lane, 2018) and previous findings associating outcomes
in dynamic therapy to increased understanding into dynamic pat-
terns (Johansson et al., 2010;Kallestad et al., 2010). The proposed
relational pathway of change supports the suggestion that alliance
may interact with therapist technique and other process variables to
predict outcomes (Beutler et al., 2012). This is in line with the work
Table 2
Moderated Mediation Results for the Effect of Affect Experiencing on Depression (PHQ-9) Moderated by Personality Pathology (PP)
Moderated mediation
Anger Guilt Sadness
bSDp 95% CI bSDp 95% CI bSDp 95% CI
AE →ARM (a) 0.01 0.03 .82 [−0.06, 0.07] 0.08 0.04 .09 [−0.02, 0.16] 0.06 0.04 .12 [−0.02, 0.14]
AE ×PP →ARM 0.09 0.03 <.001 [0.04, 0.14] −0.00 0.04 .98 [−0.08, 0.07] 0.04 0.04 .40 [−0.05, 0.12]
ARM →PHQ-9 (b) −0.15 0.05 <.01 [−0.24, −0.06] −0.13 0.05 <.01 [−0.23, −0.04] −0.12 0.05 .01 [−0.21, −0.03]
AE →insight (a) 0.13 0.05 <.01 [0.04, 0.23] 0.22 0.06 <.001 [0.10, 0.33] 0.16 0.06 <.01 [0.04, 0.26]
AE ×PP →insight −0.09 0.04 .03 [−0.16, −0.01] −0.12 0.05 .02 [−0.22, −0.02] −0.12 0.06 .04 [−0.24, −0.00]
Insight →PHQ-9 (b) −0.08 0.04 .03 [−0.15, −0.01] −0.10 0.04 <.01 [−0.18, −0.03] −0.10 0.04 <.01 [−0.18, −0.02]
AE →PHQ-9 (c) −0.01 0.03 .82 [−0.07, 0.06] 0.02 0.04 .65 [−0.07, 0.11] −0.01 0.04 .79 [−0.09, 0.07]
AE ×PP →PHQ-9 0.09 0.03 <.001 [0.04, 0.14] −0.02 0.04 .68 [−0.08, 0.06] −0.04 0.04 .19 [−0.12, 0.04]
Conditional indirect effects, AE →ARM →PHQ-9 (a ×b)
a
, by personality pathology
Low PP 0.01 0.01 .22 [−0.01, 0.03]
Mean PP −0.00 0.00 .82 [−0.01, 0.01] −0.01 0.01 .05 −0.02, 0.00 −0.01 0.01 .06 −0.02, 0.00
High PP −0.01 0.01 <.05 [−0.03, −0.00]
Conditional indirect effects, AE →Insight →PHQ-9 (a ×b), by personality pathology
Low PP −0.02 0.01 .03 [−0.04, −0.00] −0.03 0.02 .01 −0.07, −0.01 −0.03 0.01 .01 −0.06, −0.00
Mean PP −0.01 0.01 .03 [−0.03, −0.00] −0.02 0.01 .01 −0.05, −0.01 −0.01 0.01 .01 −0.04, −0.00
High PP −0.00 0.00 .43 [−0.02, 0.01] −0.01 0.01 .09 −0.03, 0.00 −0.00 0.01 .64 −0.02, 0.01
Conditional direct effect (c)
Low PP −0.10 0.04 .03 [−0.19, −0.01]
Mean PP −0.01 0.03 .79 [−0.02, 0.01]
High PP 0.08 0.04 .03 [0.01, 0.16]
Note.AE=ATOS affect experiencing scale; ARM =agnew relationship measure; Insight =ATOS insight scale; PHQ-9 =patient health questionnaire for
depression; PP =personality pathology.
a
When the moderator was nonsignificant, the model was reestimated without the moderator, and the result for
unmoderated mediation is presented on the row for Mean Severity.
ANGER-DEPRESSION MECHANISM IN DYNAMIC THERAPY 333
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
of Ulvenes et al. (2012) demonstrating that the effectiveness of an
affect focus in dynamic therapy can in part be understood through
the role of the alliance. Zilcha-Mano (2017) suggested that the
alliance is therapeutic through providing a corrective emotional
experience (Alexander & French, 1946). In light of these new
findings, we propose that the putative role of the alliance may
work through a more complex change mechanism involving AE:
Experiencing and expressing feelings in the therapy relationship can
sometimes generate a corrective emotional experience. Our data
showed that this seemed to happen for patients high on personality
pathology, as demonstrated by AE predicting improving alliance
which in turn, predicted symptomatic improvement.
While the efficacy of dynamic therapy has been demonstrated in
patients with impairments in personality in the setting of MDD
(Abbass et al., 2011), some patients do not benefit. We found that
when anger did not affect the mediators, insight or therapeutic
alliance, it appears that the improvements in depression following
increased experience of anger are only evident in low personality
pathology patients, with possible negative effects of anger
experiencing in higher personality pathology patients. This might
suggest that one means of optimizing treatment outcomes in
dynamic therapy for depression, specifically in patients with
more severe personality difficulties, is studying how to more
consistently mobilize feelings while also activating a strong alliance
for some patients for whom otherwise effects may be delayed or
potentially negative. To do so, therapists should attend to the in-
session impact of alexithymia, syntonic defenses, and potentially
problematic interpersonal processes. An alternative interpretation is
that in the context of a strong therapeutic alliance, anger experienc-
ing is related to decreased depression (Høglend et al., 2011).
The Role for a Broader Affect–Depression Mechanism
Secondary analyses conducted in this study, found that the
effectiveness of dynamic therapy for depression involves patients
experiencing and expressing a range of mixed feelings about close
relationships, although the magnitude of the associations were the
greatest with anger. Processing the trauma of ruptured attachment
bonds includes sadness about losses and painful guilt when faced
with anger toward loved ones. The smaller number of available
observations for guilt and sadness may have contributed to the
somewhat weaker results for these variables.
Existing findings on the relative importance of patients experienc-
ing different affects in dynamic therapy are mixed. An RCT of
dynamic therapy for anorexia (Friederich et al., 2017) found that
both anger and sadness were significantly associated to outcomes. In
two studies, an RCT of panic focused psychodynamic therapy
(Keefe et al., 2019) and an observational study of STPP for depres-
sion (Kramer et al., 2014), patients experiencing sadness but not
anger were responsible for the majority of the process-outcome
association. Across this research, the relative degree to which
treatments targeted the anger–depression mechanism is unclear,
so drawing conclusions should be done with caution.
It is possible that the temporal sequence in which emotions are
explored in therapy is also important. Transforming emotions in
sequential phases during therapy has been proposed as a model that
could span theoretical approaches (Pascual-Leone & Greenberg,
2007). The absence of a moderating effect of high personality
pathology on the indirect effects of patients experiencing either
guilt about anger or sadness through the alliance, reflects a differ-
ence compared to the mechanisms through which anger appears to
work in therapy. These findings indicate that alliance mediates the
positive effects of experiencing guilt and sadness on depression for
all patients, regardless of pretreatment personality pathology. One
interpretation for these results, in line with the role of temporal
phases of processing emotions, is that after an unlocking of anger,
defenses are sufficiently restructured such that the effects of high
pretreatment personality pathology is diminished. In contrast, it
appears that postsession patient insight is only an important media-
tor of change, regardless of the nature of the affect type, in cases with
lower personality pathology. Given previous findings that both
improved insight and affect awareness are important for patients
with low quality of object relations in longer-term psychodynamic
therapy (Høglend & Hagtvet, 2019), future research may explore
differences in mechanism between short-term dynamic therapies
and longer-term models.
Study Strengths and Limitations
The current process-outcome study was prospectively embedded
into the Halifax Depression RCT (Town, Abbass, et al., 2017),
allowing for the collection of detailed session-to-session process
and outcome data, establishing a timeline for testing causality.
Limitations include: A primarily White sample mostly meeting
criteria for a Cluster C personality disorder, in that the results may
not extend to more diverse populations, particularly given the impor-
tance of culture in emotional expression. Ratings of AE and insight
were simultaneously rated by the same judges potentially inflating
their correlation. While the sample size of treated patients is small
(N=27), the Monte Carlo simulation demonstrated that the study had
sufficient power to find small-to-medium-sized effects due to the large
number of repeated measures data collected, at least for anger which
had a greater number of observed data points than guilt and sadness.
In the majority of psychotherapy studies, it is assumed that
process-outcome results are generalizable to the entire treatment
process, despite only coding portions of sessions. Furthermore,
studies are often limited by the validity of patient self-report when
attempting to measure implicit emotional processes. In contrast, the
present study used: A reliable and validated rating system for
measuring patient AE and insight; ratings were conducted indepen-
dently by assessors with excellent interrater reliability; sessions were
rated in their entirety in a random sequence; and significantly, there
was a negligible amount of missing data with 99% of sessions rated.
With the benefits of this study design and complex analytic strategy,
we believe that the current findings go a long way toward being able
to offer a more reliable empirical picture of how depression changes
in dynamic therapy than has previously been possible.
References
Abbass, A. (2015). Reaching through Resistance. Seven Leaves Press.
Abbass, A., Town, J., & Driessen, E. (2011). The efficacy of short-term
psychodynamic psychotherapy for depressive disorders with comorbid
personality disorder. Psychiatry,74(1), 58–71. https://doi.org/10.1521/
psyc.2011.74.1.58
Agnew-Davies, R., Stiles, W. B., Hardy, G. E., Barkham, M., & Shapiro,
D. A. (1998). Alliance structure assessed by the Agnew Relationship
Measure (ARM). British Journal of Clinical Psychology,37(2), 155–172.
https://doi.org/10.1111/j.2044-8260.1998.tb01291.x
334 TOWN, FALKENSTRÖM, ABBASS, AND STRIDE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
Alexander, F., & French, T. M. (1946). Psychoanalytic therapy: Principles
and application. Ronald Press.
Andrews, G., Singh, M., & Bond, M. (1993). The defense style question-
naire. Journal of Nervous and Mental Disease,181(4), 246–256. https://
doi.org/10.1097/00005053-199304000-00006
Asparouhov, T., Hamaker, E. L., & Muthén, B. (2018). Dynamic structural
equation models. Structural Equation Modeling,25(3), 359–388. https://
doi.org/10.1080/10705511.2017.1406803
Asparouhov, T., & Muthén, B. (2010). Bayesian analysis using Mplus:
technical implementation [Unpublished report]. https://www.statmodel
.com/download/Bayes3.pdf
Bagby, R. M., Taylor, G. J., & Parker, J. D. (1994). The twenty-item Toronto
Alexithymia scale—II. Convergent, discriminant, and concurrent validity.
Journal of Psychosomatic Research,38(1), 33–40. https://doi.org/10
.1016/0022-3999(94)90006-X
Bedics, J. D., Atkins, D. C., Harned, M. S., & Linehan, M. M. (2015). The
therapeutic alliance as a predictor of outcome in dialectical behavior
therapy versus nonbehavioral psychotherapy by experts for borderline
personality disorder. Psychotherapy: Theory, Research, & Practice,52(1),
67–77. https://doi.org/10.1037/a0038457
Berggraf, L., Ulvenes, P. G., Hoffart, A., McCullough, L., & Wampold, B. E.
(2014). Growth in sense of self and sense of others predicts reduction in
interpersonal problems in short-term dynamic but not in cognitive therapy.
Psychotherapy Research,24(4), 456–469. https://doi.org/10.1080/
10503307.2013.840401
Beutler, L. E., Forrester, B., Gallagher-Thompson, D., Thompson, L., &
Tomlins, J. B. (2012). Common, specific, and treatment fit variables in
psychotherapy outcome. Journal of Psychotherapy Integration,22(3),
255–281. https://doi.org/10.1037/a0029695
Blatt, S. J. (1974). Levels of object representation in anaclitic and introjective
depression. The Psychoanalytic Study of the Child,29(1), 107–157.
https://doi.org/10.1080/00797308.1974.11822616
Blatt, S. J. (1998). Contributions of psychoanalysis to the understanding and
treatment of depression. Journal of the American Psychoanalytic Associ-
ation,46(3), 722–752. https://doi.org/10.1177/00030651980460030301
Blatt, S. J., Zuroff, D. C., Hawley, L. L., & Auerbach, J. S. (2010). Predictors
of sustained therapeutic change. Psychotherapy Research,20(1), 37–54.
https://doi.org/10.1080/10503300903121080
Bowlby, E. (2008). Loss-Sadness and Depression: Attachment and Loss
(Vol. 3). Random House.
Bowlby, J. (1973). In Attachment and loss: Vol. 2. Separation, anxiety
and anger. The Hogarth Press and the Institute of Psycho-Analysis.
Busch, F. N. (2009). Anger and depression. Advances in Psychiatric
Treatment,15(4), 271–278. https://doi.org/10.1192/apt.bp.107.004937
Cahill, J., Stiles, W. B., Barkham, M., Hardy, G. E., Stone, G., Agnew-
Davies, R., & Unsworth, G. (2011). Two short forms of the Agnew
relationship measure: The ARM-5 and ARM-12. Psychotherapy Research,
22(3), 241–255. https://doi.org/10.1080/10503307.2011.643253
Calati, R., Oasi, O., De Ronchi, D., & Serretti, A. (2010). The use of the
defence style questionnaire in major depressive and panic disorders: A
comprehensive meta-analysis. Psychology and Psychotherapy: Theory,
Research and Practice,83(Pt 1), 1–13. https://doi.org/10.1348/1476083
09X464206
Davanloo, H. (2000). Intensive short-term dynamic psychotherapy. Wiley.
Diener, M. J., Hilsenroth, M. J., & Weinberger, J. (2007). Therapist affect
focus and patient outcomes in psychodynamic psychotherapy: A meta-
analysis. The American Journal of Psychiatry,164(6), 936–941. https://
doi.org/10.1176/ajp.2007.164.6.936
Diener, M. J., & Monroe, J. M. (2011). The relationship between adult
attachment style and therapeutic alliance in individual psychotherapy: A
meta-analytic review. Psychotherapy: Theory, Research, & Practice,
48(3), 237–248. https://doi.org/10.1037/a0022425
Falkenström, F., Ekeblad, A., & Holmqvist, R. (2016). Improvement of the
working alliance in one treatment session predicts improvement of
depressive symptoms by the next session. Journal of Consulting and
Clinical Psychology,84(8), 738–751. https://doi.org/10.1037/ccp0000119
Falkenström, F., Solomonov, N., & Rubel, J. A. (2020). Do therapist effects
really impact estimates of within-patient mechanisms of change? A Monte
Carlo simulation study. Psychotherapy Research,30(7), 885–899. https://
doi.org/10.1080/10503307.2020.1769875
Fisher, H., Atzil-Slonim, D., Bar-Kalifa, E., Rafaeli, E., & Peri, T. (2016).
Emotional experience and alliance contribute to therapeutic change in
psychodynamic therapy. Psychotherapy: Theory, Research, & Practice,
53(1), 105–116. https://doi.org/10.1037/pst0000041
Flückiger, C., Del Re, A. C., Wampold, B. E., & Horvath, A. O. (2018). The
alliance in adult psychotherapy: A meta-analytic synthesis. Psychother-
apy: Theory, Research, & Practice,55(4), 316–340. https://doi.org/10
.1037/pst0000172
Freud, S. (1957). Mourning and melancholia. In The standard edition of the
complete psychological works of Sigmund Freud, Volume XIV (1914–
1916): On the history of the psycho-analytic movement, papers on
metapsychology and other works (pp. 237–258). Hogarth Press.
Friederich, H.-C., Brockmeyer, T., Wild, B., Resmark, G., de Zwaan, M.,
Dinkel, A., Herpertz, S., Burgmer, M., Löwe, B., Tagay, S., Rothermund,
E., Zeeck, A., Zipfel, S., & Herzog, W. (2017). Emotional expression
predicts treatment outcome in focal psychodynamic and cognitive beha-
vioural therapy for anorexia nervosa: Findings from the ANTOP study.
Psychotherapy and Psychosomatics,86(2), 108–110. https://doi.org/10
.1159/000453582
Gabbard, G. O. (2014). Psychodynamic psychiatry in clinical practice.
American Psychiatric Pub.
Gabbard, G. O., & Simonsen, E. (2007). The impact of personality and
personality disorders on the treatment of depression. Personality and
Mental Health,1(2), 161–175. https://doi.org/10.1002/pmh.21
Gibbons, M. B. C., Crits-Christoph, P., Barber, J. P., Wiltsey Stirman, S.,
Gallop, R., Goldstein, L. A., Temes, C. M., & Ring-Kurtz, S. (2009).
Unique and common mechanisms of change across cognitive and dynamic
psychotherapies. Journal of Consulting and Clinical Psychology,77(5),
801–813. https://doi.org/10.1037/a0016596
Gignac, G. E., & Szodorai, E. T. (2016). Effect size guidelines for individual
differences researchers. Personality and Individual Differences,102,74–
78. https://doi.org/10.1016/j.paid.2016.06.069
Gilbert, P., Gilbert, J., & Irons, C. (2004). Life events, entrapments and
arrested anger in depression. Journal of Affective Disorders,79(1–3), 149–
160. https://doi.org/10.1016/S0165-0327(02)00405-6
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation
analysis in the new millennium. Communication Monographs,76(4),
408–420. https://doi.org/10.1080/03637750903310360
Høglend, P., & Hagtvet, K. (2019). Change mechanisms in psychotherapy:
Both improved insight and improved affective awareness are necessary.
Journal of Consulting and Clinical Psychology,87(4), 332–344. https://
doi.org/10.1037/ccp0000381
Høglend, P., Hersoug, A. G., Bøgwald, K.-P., Amlo, S., Marble, A., Sørbye,
Ø., Røssberg, J. I., Ulberg, R., Gabbard, G. O., & Crits-Christoph, P.
(2011). Effects of transference work in the context of therapeutic alliance
and quality of object relations. Journal of Consulting and Clinical
Psychology,79(5), 697–706. https://doi.org/10.1037/a0024863
Horowitz, L. M., Alden, L. E., Wiggins, J. S., & Pincus, A. L. (2000).
Inventory of interpersonal problems. The Psychological Corporation.
Jennissen, S., Huber, J., Ehrenthal, J. C., Schauenburg, H., & Dinger, U.
(2018). Association between insight and outcome of psychotherapy:
Systematic review and meta-analysis. The American Journal of Psychia-
try,175(10), 961–969. https://doi.org/10.1176/appi.ajp.2018.17080847
Johansson, P., Høglend, P., Ulberg, R., Amlo, S., Marble, A., Bøgwald,
K.-P., Sørbye, O., Sjaastad, M. C., & Heyerdahl, O. (2010). The mediating
role of insight for long-term improvements in psychodynamic therapy.
Journal of Consulting and Clinical Psychology,78(3), 438–448. https://
doi.org/10.1037/a0019245
ANGER-DEPRESSION MECHANISM IN DYNAMIC THERAPY 335
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
Kallestad, H., Valen, J., McCullough, L., Svartberg, M., Høglend, P., & Stiles,
T. C. (2010). The relationship between insight gained during therapy and
long-term outcome in short-term dynamic psychotherapy and cognitive
therapy for cluster C personality disorders. Psychotherapy Research,20(5),
526–534. https://doi.org/10.1080/10503307.2010.492807
Keefe, J. R., Huque, Z. M., DeRubeis, R. J., Barber, J. P., Milrod, B. L., &
Chambless, D. L. (2019). In-session emotional expression predicts symp-
tomatic and panic-specificreflective functioning improvements in panic-
focused psychodynamic psychotherapy. Psychotherapy: Theory, Research,
& Practice,56(4), 514–525. https://doi.org/10.1037/pst0000215
Kernberg, O. (1984). Severe personality disorders: Psychotherapeutic strat-
egies. Yale University Press.
Klein, M. (1935). A contribution to the psychogenesis of manic-depressive
states. The International Journal of Psycho-Analysis,16, 145–174.
Koelen, J. A., Luyten, P., Eurelings-Bontekoe, L. H., Diguer, L., Vermote, R.,
Lowyck, B., & Bühring, M. E. (2012). The impact of level of personality
organizationon treatment response: A systematic review. Psychiatry,75(4),
355–374. https://doi.org/10.1521/psyc.2012.75.4.355
Kramer, U., Pascual-Leone, A., Despland, J.-N., & De Roten, Y. (2014).
Emotion in an alliance rupture and resolution sequence: A theory-building
case study. Counselling & Psychotherapy Research,14(4), 263–271.
https://doi.org/10.1080/14733145.2013.819932
Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of
a brief depression severity measure. Journal of General Internal Medicine,
16(9), 606–613. https://doi.org/10.1046/j.1525-1497.2001.016009606.x
Lane, R. D. (2018). From reconstruction to construction: The power of
corrective emotional experiences in memory reconsolidation and enduring
change. Journal of the American Psychoanalytic Association,66(3), 507–
516. https://doi.org/10.1177/0003065118782198
Malan, D. H. (1979). Individual psychotherapy and the science of psycho-
dynamics. Butterworth Heinemann.
McCullough, L., Larsen, A. E., Schanche, E., Andrews, S., Kuhn, N.,
Hurley, C., & Wolf, J. (2003). Achievement of therapeutic objectives
scale: ATOS scale [Unpublished manual]. Harvard Medical School.
McWilliams, N. (2011). Psychoanalytic diagnosis: Understanding person-
ality structure in the clinical process. Guilford Press.
Mischel, W., & Shoda, Y. (2008). Toward a unified theory of personality:
Integrating dispositions and processing dynamics within the cognitive-
affective processing system. In O. P. John, R. W. Robins, & L. A. Pervin
(Eds.), Handbook of personality: Theory and research (pp. 208–241).
Guilford Press.
Morse, J. Q., & Pilkonis, P. A. (2007). Screening for personality disorders.
Journal of Personality Disorders,21(2), 179–198. https://doi.org/10
.1521/pedi.2007.21.2.179
Ogrodniczuk, J. S., Piper, W. E., & Joyce, A. S. (2011). Effect of alexithymia
on the process and outcome of psychotherapy: A programmatic review.
Psychiatry Research,190(1), 43–48. https://doi.org/10.1016/j.psychres
.2010.04.026
Painuly, N., Sharan, P., & Mattoo, S. K. (2005). Relationship of anger and
anger attacks with depression: A brief review. European Archives of
Psychiatry and Clinical Neuroscience,255(4), 215–222. https://doi.org/
10.1007/s00406-004-0539-5
Pascual-Leone, A., & Greenberg, L. S. (2007). Emotional processing in
experiential therapy: Why “the only way out is through..”Journal of
Consulting and Clinical Psychology,75(6), 875–887. https://doi.org/10
.1037/0022-006X.75.6.875
Peluso, P. R., & Freund, R. R. (2018). Therapist and client emotional
expression and psychotherapy outcomes: A meta-analysis. Psychother-
apy: Theory, Research, & Practice,55(4), 461–472. https://doi.org/10
.1037/pst0000165
Pincus, A. L. (2005). A contemporary integrative interpersonal theory of
personality disorders. In M. F. Lenzenweger & J. F. Clarkin (Eds.), Major
theories of personality disorder (pp. 282–331). Guilford Press.
Piper, W. E., McCallum, M., Joyce, A. S., Rosie, J. S., & Ogrodniczuk, J. S.
(2001). Patient personality and time-limited group psychotherapy for
complicated grief. International Journal of Group Psychotherapy,
51(4), 525–552. https://doi.org/10.1521/ijgp.51.4.525.51307
Ryum, T., Støre-Valen, J., Svartberg, M., Stiles, T. C., & McCullough, L.
(2014). Factor analysis of the Achievement of Therapeutic Objectives Scale
(ATOS) in short-term dynamic psychotherapy and cognitive therapy.
Psychological Assessment,26(3), 925. https://doi.org/10.1037/a0036570
Schultzberg, M., & Muthén, B. (2018). Number of subjects and time points
needed for multilevel time-series analysis: A simulation study of dynamic
structural equation modeling. Structural Equation Modeling,25(4), 495–
515. https://doi.org/10.1080/10705511.2017.1392862
Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing
rater reliability. Psychological Bulletin,86(2), 420–428. https://doi.org/10
.1037/0033-2909.86.2.420
Town, J. M., Abbass, A., Stride, C., & Bernier, D. (2017). A randomised
controlled trial of intensive short-term dynamic psychotherapy for treat-
ment resistant depression: The Halifax Depression Study. Journal of
Affective Disorders,214,15–25. https://doi.org/10.1016/j.jad.2017.02.035
Town, J. M., Abbass, A., Stride, C., Nunes, A., Bernier, D., & Berrigan, P.
(2020). Efficacy and cost-effectiveness of intensive short-term dynamic
psychotherapy for treatment resistant depression: 18-Month follow-up of
the Halifax depression trial. Journal of Affective Disorders,273, 194–202.
https://doi.org/10.1016/j.jad.2020.04.035
Town,J.M.,Chafe,D.,&Pienkos,E.(2014).Achievement of therapeutic
objectives scale: A revised manual for rating ISTDP. Dalhousie Univeristy.
Town, J. M., Falkenström, F., Abbass, A., & Stride, C. (2022). The anger-
depression mechanism in dynamic therapy: Experiencing previously
avoided anger positively predicts reduction in depression via working
alliance and insight. Journal of Counseling Psychology,69(3), 326–336.
https://doi.org/10.1037/cou0000581
Town, J. M., Hardy, G. E., McCullough, L., & Stride, C. (2012). Patient
affect experiencing following therapist interventions in short-term
dynamic psychotherapy. Psychotherapy Research,22(2), 208–219.
https://doi.org/10.1080/10503307.2011.637243
Town, J. M., Salvadori, A., Falkenström, F., Bradley, S., & Hardy, G. (2017).
Is affect experiencing therapeutic in major depressive disorder? Examin-
ing associations between affect experiencing and changes to the alliance
and outcome in intensive short-term dynamic psychotherapy. Psychother-
apy: Theory, Research, & Practice,54(2), 148–158. https://doi.org/10
.1037/pst0000108
Ulvenes, P. G., Berggraf, L., Hoffart, A., Stiles, T. C., Svartberg, M.,
McCullough, L., & Wampold, B. E. (2012). Different processes for
different therapies: Therapist actions, therapeutic bond, and outcome.
Psychotherapy: Theory, Research, & Practice,49(3), 291–302. https://
doi.org/10.1037/a0027895
Westen, D., Gabbard, G. O., & Blagov, P. (2006). Back to the Future: Personality
Structure as a Context for Psychopathology. In R. F. Krueger & J. L. Tackett
(Eds.), Personality and psychopathology (pp. 335–384). Guilford Press.
Zilcha-Mano, S. (2017). Is the alliance really therapeutic? Revisiting this
question in light of recent methodological advances. American Psycholo-
gist,72(4), 311–325. https://doi.org/10.1037/a0040435
Zilcha-Mano, S., & Errázuriz, P. (2015). One size does not fit all: Examining
heterogeneity and identifying moderators of the alliance-outcome associ-
ation. Journal of Counseling Psychology,62(4), 579–591. https://doi.org/
10.1037/cou0000103
Zilcha-Mano, S., Roose, S. P., Barber, J. P., & Rutherford, B. R. (2015).
Therapeutic alliance in antidepressant treatment: Cause or effect of
symptomatic levels? Psychotherapy and Psychosomatics,84(3), 177–
182. https://doi.org/10.1159/000379756
Received December 11, 2020
Revision received July 21, 2021
Accepted July 31, 2021 ▪
336 TOWN, FALKENSTRÖM, ABBASS, AND STRIDE
This document is copyrighted by the American Psychological Association or one of its allied publishers.
Content may be shared at no cost, but any requests to reuse this content in part or whole must go through the American Psychological Association.
Content uploaded by Joel M Town
Author content
All content in this area was uploaded by Joel M Town on Oct 05, 2021
Content may be subject to copyright.