ArticlePDF Available

Abstract and Figures

The objective of this study was to evaluate the clinical- and cost-effectiveness of Intensive Short-Term Dynamic Psychotherapy (ISTDP) for generalized anxiety disorder (GAD). We further aimed to examine if a key clinical process within the ISTDP framework, termed the level of mobilization of unprocessed complex emotions (MUCE), was related to outcome. The sample consisted of 215 adult patients (60.9% female) with GAD and comorbid conditions treated in a tertiary mental health outpatient setting. The patients were provided an average of 8.3 sessions of ISTDP delivered by 38 therapists. The level of MUCE in treatment was assessed from videotaped sessions by a rater blind to treatment outcome. Year-by-year healthcare costs were derived independently from government databases. Multilevel growth models indicated significant decreases in psychiatric symptoms and interpersonal problems during treatment. These gains were corroborated by reductions in healthcare costs that continued for 4 years post-treatment reaching normal population means. Further, we found that the in-treatment level of MUCE was associated with larger treatment effects, underlining the significance of emotional experiencing and processing in the treatment of GAD. We conclude that ISTDP appears to reduce symptoms and costs associated with GAD and that the ISTDP framework may be useful for understanding key therapeutic processes in this challenging clinical population. Controlled studies of ISTDP for GAD are warranted.
Content may be subject to copyright.
RESEARCH ARTICLE
Intensive ShortTerm Dynamic Psychotherapy for generalized
anxiety disorder: A pilot effectiveness and processoutcome
study
Peter Lilliengren
1
|Robert Johansson
2
|Joel M. Town
3,5
|Steve Kisely
4
|Allan Abbass
5
1
Department of Psychology, Stockholm
University, Stockholm, Sweden
2
Department of Clinical Neuroscience,
Karolinska Institute, Stockholm, Sweden
3
Plymouth University Peninsula Schools of
Medicine and Dentistry, Plymouth, UK
4
School of Medicine, The University of
Queensland, Queensland, Australia
5
Centre for Emotions and Health, Dalhousie
University, Nova Scotia, Canada
Correspondence
Peter Lilliengren, Hornsbruksgatan 11, 117 34
Stockholm, Sweden
Email: peter.lilliengren@affekta.se
Abstract
The objective of this study was to evaluate the clinicaland costeffectiveness of Intensive Short
Term Dynamic Psychotherapy (ISTDP) for generalized anxiety disorder (GAD). We further aimed
to examine if a key clinical process within the ISTDP framework, termed the level of mobilization
of unprocessed complex emotions (MUCE), was related to outcome. The sample consisted of 215
adult patients (60.9% female) with GAD and comorbid conditions treated in a tertiary mental
health outpatient setting. The patients were provided an average of 8.3 sessions of ISTDP deliv-
ered by 38 therapists. The level of MUCE in treatment was assessed from videotaped sessions by
a rater blind to treatment outcome. Yearbyyear healthcare costs were derived independently
from government databases. Multilevel growth models indicated significant decreases in psychi-
atric symptoms and interpersonal problems during treatment. These gains were corroborated by
reductions in healthcare costs that continued for 4 years posttreatment reaching normal popu-
lation means. Further, we found that the intreatment level of MUCE was associated with larger
treatment effects, underlining the significance of emotional experiencing and processing in the
treatment of GAD. We conclude that ISTDP appears to reduce symptoms and costs associated
with GAD and that the ISTDP framework may be useful for understanding key therapeutic pro-
cesses in this challenging clinical population. Controlled studies of ISTDP for GAD are warranted.
KEYWORDS
costeffectiveness, emotion processing, generalized anxiety disorder, naturalistic, shortterm
dynamic therapy
1|INTRODUCTION
Generalized anxiety disorder (GAD) is a common psychiatric condition
characterized by uncontrollable worry and accompanying mental and
physical symptoms such as restlessness, fatigue, difficulties concentrat-
ing, irritability, muscle tension, and sleep disturbance (American
Psychiatric Association, 2000). GAD has a lifetime prevalence of 4.3
to 5.9% (Tyrer & Baldwin, 2006), and the condition is associated with
considerable individual and societal costs due to lost work productivity
and high use of medical resources (Hoffman, Dukes, & Wittchen, 2008).
GAD is further associated with high rates of comorbid mood, anxiety,
and personality disorders, as well as increased risk for substance depen-
dence (Grant et al., 2005; Kessler et al., 2008). The condition typically
has a chronic course; only about 3841% of patients seeking treatment
recover within 5 years and the presence of a personality disorder
contributes to an even worse prognosis (Yonkers, Dyck, Warshaw, &
Keller, 2000). Due to its chronic course and high rates of comorbidity,
GAD has been estimated to be one of the most expensive psychiatric
disorders to treat (Hoffman et al., 2008).
While pharmacological treatment options for GAD exist, psycho-
logical treatments may be preferable and better tolerated by patients
(Tyrer & Baldwin, 2006). Currently, cognitivebehavior therapy has
the strongest evidence base (Cuijpers et al., 2014; Hofmann & Smits,
2008), but several studies indicate that treatments based on psychody-
namic principles may also be effective in this population (Andersson
et al., 2012; CritsChristoph, Gibbons, Narducci, Schamberger, &
Gallop, 2005; Leichsenring et al., 2009; Levy Berg, Sandell, & Sandahl,
2009). Traditionally, psychodynamic treatments have placed a stronger
emphasis on insight into problematic relationship patterns and
processing of wardedoff emotions compared to cognitivebehavior
Received: 6 December 2016 Revised: 16 May 2017 Accepted: 23 May 2017
DOI: 10.1002/cpp.2101
Clin Psychol Psychother. 2017;24:13131321. Copyright © 2017 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/cpp 1313
therapy (CritsChristoph, 2002; Blagys & Hilsenroth, 2000). This may
be particularly important in GAD patients because research indicates
that emotional avoidance and dysregulation (Borkovec, Alcaine, &
Behar, 2004; Mennin, Heimberg, Turk, & Fresco, 2005), as well as
insecure attachment (Cassidy, LichtensteinPhelps, Sibrava, Thomas,
& Borkovec, 2009; Marganska, Gallagher, & Miranda, 2013), may
underlie the disorder. These factors are also likely to contribute to
the high rates of comorbid personality disorders and a more chronic
course of GAD.
In the present study, we aim to evaluate the clinicaland cost
effectiveness of Intensive ShortTerm Dynamic Psychotherapy
(ISTDP; Davanloo, 1990, 2005; Abbass, 2015) for patients with GAD.
ISTDP is an emotionfocused, experiential psychodynamic treatment
model with a growing evidence base (Abbass, Town, & Driessen,
2012; Lilliengren, Johansson, Lindqvist, Mechler, & Andersson, 2016;
Town & Driessen, 2013). Previous studies have indicated that ISTDP
may be effective in complex psychiatric conditions, such as personality
disorders (e.g., Abbass, Sheldon, Gyra, & Kalpin, 2008; Solbakken &
Abbass, 2015), treatment resistant depression (Town, Abbass, Stride,
& Bernier, 2017) and medically unexplained symptoms (Abbass et al.,
2010; Chavooshi, Mohammadkhani, & Dolatshahee, 2016), all of which
overlap considerably with GAD. Although ISTDP has not previously
been evaluated specifically for GAD, we hypothesize that ISTDP will
be associated with significant reductions in symptoms and health care
utilization in this population.
A secondary aim of this study is to examine if a key clinical process
within the ISTDP framework, termed the level of mobilization of
unprocessed complex emotions (hereafter called MUCE for short), is
related to treatment outcome in this population. Based on detailed
study of several hundred videotaped cases, Davanloo (1990) proposed
that a central therapeutic element in ISTDP is the patient's experienc-
ing of his true feelings about past and present individuals in his life.
These feelings typically relate to adverse relational events (e.g., inter-
personal traumas and/or attachment ruptures), which generate both
anxiety and secondary avoidant behaviors (i.e., defenses and resis-
tance) in new relational contexts, including the therapy relationship.
Davanloo (2000, 2001) further described various degrees of activation
of such unprocessed emotions, and he termed the more intense levels
unlocking of the unconsciousbecause the patient typically makes
spontaneously linkages to past experiences when the feelings are
activated and experienced.
Insession emotional arousal and processing may be a key mecha-
nism of change across therapy models and disorders (Greenberg &
Pascualleone, 2006; Solbakken, Hansen, & Monsen, 2011; Whelton,
2004). For example, a number of processoutcome studies have
indicated that the degree of insession emotional activation is related
to outcome in psychodynamic therapies (e.g., Diener, Hilsenroth, &
Weinberger, 2007; Fisher, Atzilslonim, BarKalifa, Rafaeli, & Peri,
2016; Town, Lomax, Abbass, & Hardy, 2017; Town, Salvadori,
Falkenstrom, Bradley, & Hardy, in press), as well as in emotionfocused
therapy (Goldman, Greenberg, & Pos, 2005; Pos, Greenberg, Goldman,
& Korman, 2003). Previous research on ISTDP has also indicated that a
high degree of MUCE in treatment (i.e., an unlocking of the uncon-
scious) is associated with better outcome in mixed patient samples
(Johansson, Town, & Abbass, 2014; Town, Abbass, & Bernier, 2013).
Because a core feature of GAD seems to be avoidance of negative
emotions triggered in interpersonal contexts (Behar, DiMarco, Hekler,
Mohlman, & Staples, 2009; Timulak & McElvaney, 2016), and some
research suggest that emotional experiencing reduces GAD symptoms
(e.g., Sugiura & Sugiura, 2015), activating and processing emotional
states within therapy sessions may be particularly important in this
population. Thus, for this study, we hypothesize that a higher degree
of MUCE in treatment will be associated with greater symptom reduc-
tion for GAD patients receiving ISTDP.
2|METHODS
2.1 |Patient sample
The sample for this study was drawn from a large naturalistic study
conducted at the Centre for Emotions and Health (CEH), Dalhousie
University, Halifax (Abbass, Kisely, Rasic, Town, & Johansson, 2015).
CEH is a tertiary psychotherapy service receiving referrals from emer-
gency departments, family practice offices, medical and surgical spe-
cialties, as well as secondary and tertiary mental health services in
the province of Nova Scotia, Canada. The larger study included a total
of 890 patients that were referred to CEH over a 9year period of
which 215 fulfilled criteria for DSMIV GAD (American Psychiatric
Association, 2000). Diagnoses were assessed clinically by the treating
therapist and supervising psychiatrist using referral information, the
initial treatment interview, and observation during the treatment
courses.
Characteristics of the current sample are presented in Table 1. The
sample consisted of primarily females (60.9%) in their young adult
(1935) years (80.0%). Besides the diagnosis of GAD, there was a high
rate of comorbidity with other DSMIV AxisI diagnoses, most notably
with somatoform disorder (61.4%) and depression (35.8%). More than
half of the sample (55.8%) met criteria for a personality disorder, most
commonly in Cluster C (36.3%).
All patients gave their written informed consent for video
recording of their treatment sessions as part of standard care at the
CEH. The original study was approved by the local hospital ethics
review board and registered in ClinicalTrials.gov as identifier number
NCT01924715.
Key Practitioner Message:
Intensive ShortTerm Dynamic Psychotherapy (ISTDP)
was found to reduce psychiatric symptoms,
interpersonal problems and healthcare costs in patients
with Generalized Anxiety Disorder (GAD)
The effect of ISTDP was associated with the level of
mobilization of unprocessed complex emotions in
treatment
Psychotherapy of GAD may be enhanced by including
emotionfocused interventions targeting the effects of
attachment trauma
1314 LILLIENGREN P. ET AL.
2.2 |Treatment
ISTDP was originally developed by Habib Davanloo (1990, 2000) and
has recently been described in detail by Abbass (2015). The model
holds that symptoms such as depression, anxiety, somatoform, and
interpersonal problems stem from adverse experiences in key attach-
ment relationships, giving rise to anxiety and maladaptive defenses in
response to activation of unprocessed complex emotions in new situ-
ations. The treatment includes a graded format,which is applied with
patients who have low anxiety tolerance and/or capacity for
experiencing and processing emotions, as well as a standard format,
which is used when patients evidence high levels of emotional detach-
ment and distancing (Abbass, 2015).
At the CEH, patients received ISTDP beginning with an extended
initial interview termed a trial therapy(Davanloo, 1990). This 23hr
long initial meeting aims at determining patients' main problem areas
and their capacity for intensive emotionfocused work, which informs
the selection of treatment modality (i.e., graded or standard format of
ISTDP). Patients who continued treatment after the trial therapy were
typically provided once a week sessions of 50 min. Treatment was
not timelimited; rather, termination was determined by patients'
response to treatment and mutually agreed upon by patient and thera-
pist. Patients in the current sample attended an average of 8.3 sessions
(SD = 12.7, range 170). Adjunct medication was monitored throughout
the study by the treating therapist and/or supervising psychiatrist.
2.3 |Therapists
The ISTDP treatments were delivered by a total of 38 therapists of
which 22 worked as psychiatry residents at CEH, 11 were licensed
health professionals, and five were other trainees learning ISTDP at
the Centre. The therapists had variable levels of training in ISTDP, with
one therapist being a highly experienced ISTDP trainer and supervisor,
considered an expert in the field. During the study, all therapists partic-
ipated in weekly smallgroup supervision and attended weekly didactic
courses, both involving detailed review of video recorded treatment
sessions, led by the experienced ISTDP trainer. A portion of the
videotaped treatments in the current sample (n= 42; 19.5%), was rated
for adherence using a 4point scale that was previously developed for
a randomized controlled trial (Abbass et al., 2008). These ratings
suggested that therapist adherence to the ISTDP model was accept-
able (M= 3.3; SD = 0.7).
2.4 |Measures
2.4.1 |Psychiatric symptoms
Patients' selfreported psychiatric symptoms were assessed at base-
line, after the initial trial therapy and at the end of the complete treat-
ment course using the Brief Symptom Inventory (BSI; Derogatis &
Melisaratos, 1983). This instrument measures the patient's subjective
experience of symptoms during the last week and includes 36 items
rated on 5point Likert scales ranging from 0 (not at all)to4(very
much). The total score of the instrument represents the Global Severity
Index that was used in this study as an indicator of patients' overall
symptom load. The internal consistency (i.e., Cronbach's α) of the total
score ranged between .90.95 across all three assessment points in
the present sample.
2.4.2 |Interpersonal problems
The level of interpersonal difficulties was assessed with the Inventory
of Interpersonal Problems (IIP; Horowitz, Rosenberg, Baer, Ureño, &
Villaseñor, 1988), which was also distributed at baseline, after the ini-
tial trial therapy and at the end of treatment. This inventory consists
of 32 items selfrated on 5point Likert scales ranging from 0 (not at
all)to4(very much) and the sum of all items comprises a measure of
overall interpersonal distress. In the present sample, Cronbach's αfor
the total score ranged between .79.88 across the measurement
points used.
2.4.3 |Mobilization of unprocessed complex emotions
The degree of mobilization of unprocessed complex emotions (MUCE)
in treatment was measured using a 5point rating scale. This scale rate
patients' insession emotional arousal and expressiveness, as well as
signs of activation of unconscious anxiety, defenses, and indications
of the unconscious therapeutic alliance (UTA). The points of the scale
are derived from Davanloo's (1990, 2000, 2001) extensive video
based case research (explicated by Abbass, 2015) and include the
following:
1. Low Rise. At this level, the therapy process is characterized by an
intellectual dialogue with very little emotional activation. There
is little to no signs of unconscious anxiety or defenses being acti-
vated in the process.
2. High Rise. At this level, there are clear signs of activation of unpro-
cessed, complex, attachmentrelated emotions with the therapist.
This manifests clinically as unconscious anxiety and defenses that
begin to crystallizeor coalesce in the form of distancing behav-
iors such as eye avoidance, nervous smile, and turning of the body
away from the therapist or the use of various character defenses
such as passivity. At this level of rise, the UTA becomes activated
and the patient may intellectually draw parallels between past and
present relationalemotional events.
3. Partial Breakthrough. At this degree, the complex feelings are at a
high level but the main conscious experience is grief about trauma
TABLE 1 Sample characteristics (n= 215)
n%
Gender Female 131 60.9
Male 84 39.1
Age 1935 172 80.0
3660 24 11.2
>60 19 8.8
AxisI comorbidity Somatoform disorder 132 61.4
Depression 77 35.8
Dysthymic disorder 58 26.9
Panic disorder 58 26.9
Substance abuse 14 6.5
AxisII diagnoses Any personality disorder 120 55.8
Cluster A 6 2.8
Cluster B 55 25.6
Cluster C 78 36.3
LILLIENGREN P. ET AL.1315
or loss or the adverse effects of the defenses. This experience
leads to a minor drop in defenses, and the UTA will be active lead-
ing to a link to recent or past attachment trauma.
4. Partial Unlocking. At this level the patient has directly experienced
aspects of complex feelings (i.e. some grief, rage and guilt about
rage) with the therapist, which results in a drop in anxiety and sec-
ondarily a drop in defenses. At this point the UTA is at a higher
level than the defenses and the patient makes clear linkages to
past attachment trauma.
5. Major Unlocking. This degree of mobilization is characterized by a
strong passage of the complex feelings toward the therapist
including positive feelings, rage, guilt about the rage, and grief.
The event results in a strong rise in the UTA, and the mental image
of the therapist transfers to the image of a past person. For exam-
ple, the patient may feel a rage to choke the therapist, but after
the urge has passed, the image of the therapist becomes the
image of a past person such as the father. This results in an expe-
rience of guilt about the rage and the emergence of loving feelings
toward the parent coupled with a marked drop in anxiety and
other symptoms.
The scale was rated by an experienced ISTDP supervisor based on
direct observation of interviews or video recordings presented in
supervision. Ratings were conducted on the session as a whole,
reflecting the highest degree of mobilization observed on video. All
intake interviews were rated and a variable number of sessions there-
after per treatment. When ratings of several sessions from the same
treatment were available, we used the highest achieved value to indi-
cate the peak intensity of the treatment for the MUCE variable. The
rater was blind in that all ratings were conducted without knowing
patients' scores on the selfreport scales. The psychometric properties
of the MUCEscale have yet to be evaluated, but a preliminary study of
reliability based on 30 videorecorded sessions with three raters
yielded acceptable interrater agreement (κ= or >0.66).
2.5 |Healthcare costs
All included patients resided in the province of Nova Scotia and had
valid provincial health card numbers; hence, all of their health care ser-
vice costs were recorded in the provincial health care registry and
accessible via government databases. Independent professionals
linked to the government databases extracted mean hospital costs
and physician costs for the period of 1year duration prior to start of
psychotherapy (baseline) and 1, 2, 3, and 4 years afterwards (for more
details of this aspect of the study, see Abbass et al., 2015).
2.6 |Data analyses
This study utilizes longitudinal data collected in a naturalistic setting
with no control group. Also, the selfreport data is hierarchically struc-
tured with repeated measurements nested within patients, who are
nested within therapists. Therefore, we decided to analyze our out-
come data using multilevel growth modeling (MLM; e.g., Singer &
Willett, 2003).
First, because some therapists treated several patients, we exam-
ined the possible presence of therapist effects. Unconditional models
with random intercepts at therapist level were fitted for both BSI
and IIP. This enables the calculation of an intraclass correlation (see
Wampold & Serlin, 2000), which may be interpreted as the percentage
of the total variability that may be attributable to therapist differences.
The examination indicated that about 0.4% of the total variability in
BSI and 4.8% of the total variability in IIP was due to therapist differ-
ences; however, neither of the variance components at therapist level
was significant (BSI, z= 0.218, p= .828; IIP, z= 0.755, p= .450). Thus,
because there was little indication of significant betweentherapist
variability in this sample, we decided to proceed with twolevel analy-
ses including repeated measurements at level 1 and patients at level 2.
Additionally, there was a large portion of missing data in this nat-
uralistic data set; of the total of 215 GAD patients, 131 (60.9%) com-
pleted at least one assessment and could thus be included in the
MLM estimations. One reason for missing data was that the selfreport
measures were phased in during the 9year study period. Another was
that some of the referred patients who lived further away only
attended the initial extended trial therapy session after which recom-
mendations were provided to local mental health professionals. As a
result, these patients only completed baseline measures. Because for-
mal dropout from treatment was not recorded, we did not know when
data was missing due to measure implementation issues, travel dis-
tance, patients' unwillingness to fill in measures, treatment dropout,
or therapeutic resolution of problems. In order to explore if missing
data was related to sample characteristics, we compared patients
who had no or only one data point on either BSI or IIP (n= 131) with
patients two or more assessments (n= 84). We found no significant
differences in terms of gender (χ
2
= 3.668, p= .055), age (mean
diff. = 5.166, SE = 2.743, t(213) = 1.884, p= .063), number of AxisI
diagnoses (mean diff. = 0.338, SE = 0.178, t(213) = 1.893,
p= .060), presence of an AxisII diagnosis (χ
2
= 0.659, p= .417), mean
baseline score on BSI (mean diff. = 0.113, SE = 0.126, t(126) = 0.893,
p= .375) or IIP (mean diff. = 0.164, SE = 0.115, t(117) = 1.346,
p= .158).
Given this, we decided to proceed with MLM using maximum like-
lihood estimation on all available data. This type of estimation provides
unbiased estimates under the less restrictive assumption that missing
data are missing at random (MAR; Enders, 2011). MAR allows the
probability of missing data be dependent on any observed or unob-
served variable (e.g., initial symptom level or travel distance) but not
on the wouldbe score at the point of missingness. By definition, it is
impossible to directly test if MAR holds. However, because there were
several possible reasons for missing data in our dataset and we found
no relationship between observed sample characteristics and
missingness, the assumptions of MAR seemed justified.
Main model building started with estimating unconditional growth
models for changes in BSI and IIP over the course of treatment
(hypothesis 1). Time was coded 0 for baseline, 0.25 for posttrial
assessment, and 1 for termination, based on the estimation that the
average waiting time from the trial therapy to posttrial assessment
was roughly a third of the length of an average full treatment course.
The models were estimated with random intercepts and slopes and
an unstructured variancecovariance structure was assumed. The
1316 LILLIENGREN P. ET AL.
unconditional growth model examining changes in symptoms over
time constitutes Model 1.
Next, in order to explore if the level of MUCE predicted outcome
(hypothesis 2), conditional growth models were then estimated for BSI
and IIP, respectively. MUCE was entered both as a main effect (to con-
trol for possible differences in baseline score) and in interaction with
time. In order to get a more interpretable model output, the MUCE
variable was grand meancentered (Curran & Bauer, 2011), meaning
that the sample mean score was subtracted from each patients rating.
This yields a variable that treats betweenpatient variation in the pre-
dictor as deviations from zero; thus, the intercept in the conditional
model (Model 2) represents the baseline score for patients with an
average degree of MUCE during treatment.
Lastly, we examined control predictors. Variables were entered
onebyone both as main effects and in interaction with time. Sample
characteristics such as gender (coded 0 = female, 1 = male), age (con-
tinuous variable) and presence of an AxisII diagnosis (coded 0 = not
present, 1 = present) all proved nonsignificant and were therefore
discarded. AxisI comorbidity (coded 0 = only GAD, >1 = number of
additional diagnoses) was significantly related to initial symptom levels,
but did not interact with time, and was therefore kept as a main effect.
The number of sessions was highly skewed because over a third of the
sample (n= 77, 35.8%) only attended the initial trial therapy session.
We therefore decided to dummy code this variable (i.e., 0 = trial
therapy only, 1 = additional sessions) to control for the influence of
having additional sessions beyond the initial trial therapy. This proved
nonsignificant and was therefore not included in our final model.
For the healthcare data, pairwise ttests were calculated indepen-
dently by the government linked database holders comparing baseline
year with year 1 posttreatment. Observed healthcare costs from years
2, 3, and 4 posttreatment were also provided and are presented
below; however, due to data access limitations, we were not provided
statistical tests using this data.
All statistical calculations were performed with the SPSS (v. 20)
software package, and because the study was considered
exploratory, we did not include correction for familywise error rate
(i.e., Bonferroni; Bender & Lange, 2001).
3|RESULTS
3.1 |Change in symptoms over time
The results of our growth model estimations are presented in Table 2.
Model 1 represents the unconditional growth models for symptom
changes over time. Baseline BSI (Intercept) was estimated to be 1.63
with a significant change rate of 0.51 BSI units over the course of
treatment (Slope). The intercept for IIP was 1.58 with a significant
downward slope of 0.41 IIP units during therapy. For both BSI and
IIP, there is significant residual variance in intercepts and slopes, indi-
cating that there is unexplained betweenpatient variance in both
symptom levels at start of treatment and in rate of change over time.
The covariance between the intercept and the slope was significant
for IIP (estimate = 0.10, p= .042), indicating that patients who started
treatment with more interpersonal difficulties at baseline had
somewhat steeper IIP slope trajectories. In terms of prepost effect
sizes,
1
a moderate to large effect was observed for the BSI (Cohen's
d= 0.79), while a moderate prepost effect (d= 0.54) was observed
for IIP.
3.2 |Impact of MUCE on rate of change in symptoms
The average degree of MUCE across all treatments was 2.95
(SD = 0.90), indicating that patients in this sample experienced at least
aPartial Breakthroughof complex emotions in treatment on average.
Model 2 constitutes our final conditional model including the impact of
MUCE on rate of change in symptoms. The intercept in the model rep-
resents patients with GAD only (AxisI comorbidity = 0), and having
additional AxisI diagnoses was related to more symptoms at start of
therapy (BSI, estimate = 0.12, p= .003; IIP, estimate = 0.10,
p= .014). MUCE was unrelated to patients' baseline levels of both
BSI and IIP but proved to be a significant predictor of the rate of
change over time for both BSI (estimate = 0.17, p= .035) and IIP
(estimate = 0.17, p= .037). Thus, in line with hypothesis 2, patients
who reached higher degrees of mobilization of unprocessed complex
emotions during treatment experienced more symptom and interper-
sonal change.
3.3 |Change in healthcare costs
The changes in healthcare costs from baseline to 4 years post
treatment are presented in Table 3. Overall, physician costs decreased
every year and were significantly lower (p= .042) directly the first year
posttreatment compared to the baseline year. Physician costs tra-
versed the normal per patient average cost of in Nova Scotia $595
(CAD) per year. The same pattern of yearbyyear reduction was
observed for hospital costs, but the first year posttreatment was not
statistically different than the 1 year baseline, likely due to data skewing
TABLE 2 Growth models estimating change in symptoms during
treatment (n= 131)
Model 1 Model 2
BSI IIP BSI IIP
Baseline score
Intercept 1.63** 1.58** 1.27** 1.29**
MUCE
a
0.07 0.04
AxisI comorbidity 0.12** 0.10*
Rate of change
Slope 0.51** 0.41** 0.46** 0.37*
MUCE
a
0.17*0.17*
Variance components
Residual variance 0.13** 0.09** 0.13** 0.09**
Intercept 0.35** 0.31** 0.31** 0.28**
Slope 0.22** 0.31** 0.20*0.28**
Covariance .06 .10*.05 .09
Note. BSI = Brief Symptom Inventory; IIP = Inventory of Interpersonal
Problems; MUCE = mobilization of unprocessed complex emotions.
a
The variable is grandmean centered.
**p< .01.
*p < .05.
LILLIENGREN P. ET AL.1317
and sample size because the absolute cost difference was large.
Hospital costs also traversed the normal mean of $1,389 per patient
(Abbass et al., 2015). The average cost difference per patient from
1 year pretreatment to the 4 years posttreatment is $16,205.
4|DISCUSSION
This study provides preliminary evidence for the clinicaland cost
effectiveness of ISTDP for patients with GAD and comorbid conditions
in a tertiary, outpatient setting. In line with our expectations, we found
significant, moderate to large, reductions in psychiatric symptoms and
interpersonal problems during treatment. The observed prepost effect
sizes are comparable to those reported in other psychodynamic treat-
ment studies of GAD (e.g., CritsChristoph et al., 2005; Leichsenring
et al., 2009; Levy Berg et al., 2009), while the treatments in our study
averaged only 8.3 sessions. As an independent outcome measure, we
also found healthcare cost reductions following treatment where
patients went from high to within normal population cost ranges.
The treatments at CEH were not timelimited, and we observed
great variation in treatment length, which warrants some further dis-
cussion. Based on his case research, Davanloo (2005) has proposed
that the optimal number of sessions in ISTDP depends on patient fac-
tors such as level of resistance and anxiety tolerance. With motivated
and highly responsive patients, the treatment process may be com-
pleted within 510 sessions, whereas for patients with high levels of
resistance the treatment may be 3040 sessions. Patients with low
anxiety tolerance (e.g., those with fragile character structure, somatiza-
tion, and severe depression) need the graded format of ISTDP and
typically require more treatment to achieve symptom remission. Ther-
apist's level of experience and skill may also influence treatment length
with junior therapists requiring more treatment sessions.
It was notable that about one third of the patients only attended
the initial extended session (e.g., the trial therapy) due in large part
to travel distances to the treatment center. Nonetheless, research
has indicated that a single ISTDP trial therapy session may yield signif-
icant benefits, particularly when a high level of MUCE is achieved in
the session (Abbass, Town, Ogrodniczuk, Joffres, & Lilliengren, 2017).
We did not find a significant difference between patients who only
attended the trial therapy session compared to those who attended
more sessions. While this lack of a difference may be an artifact due
to loss of data or other factors, it may also indicate that some GAD
patients can be successfully treated in a very brief course of ISTDP.
Nevertheless, providing that the average termination scores on both
BSI and IIP were still above normal means for the whole sample, it is
not likely that a single trial therapy session, nor our 8.3 sessions aver-
age, is a sufficient dose of treatment for most patients with GAD. In a
recent randomized controlled trial, Town, Abbass, et al. (2017) found
moderate to large effects in favor of a timelimited, 20session format
of ISTDP compared to a secondary care treatmentasusual control
condition for patients with treatment resistant depression. Such a
timelimited format of ISTDP could also be tested in a randomized
design for patients with GAD, making it possible to compare results
with other treatments of fixed duration.
As for our second hypothesis, we found that the degree of in
treatment MUCE, which is integral to the ISTDPmodel, appears to
be associated with its clinical effectiveness in GAD. This result is in line
with the general finding that patients' insession emotional activation
may be a therapeutic ingredient across treatment modalities (Diener
et al., 2007; Greenberg & Pascualleone, 2006; Whelton, 2004). It is
also in line with previous studies examining the impact of unlocking
of unconsciouson outcome in ISTDP for mixed patient populations
(Abbass et al., 2017; Johansson et al., 2014; Town et al., 2013), as well
as a recent study where insession emotional arousal was found to
predict subsequent reduction in symptom distress 1 week later in
depressed patients who recovered from ISTDP (Town et al., in press).
The average level of MUCE in our sample suggests that most
patients experienced a Partial Breakthroughin the treatment. This
is achieved when complex emotions are mobilized to a high degree,
but the patient mainly experiences grief related to trauma, loss, or
the negative impact of their defensive strategies. Theoretically, this is
not the optimal level of MUCE but may lead to significant symptom
changes (Abbass, 2015; Davanloo, 1990, 2000, 2001). Our result sug-
gests that that even higher levels of MUCE, which include the patient
directly experiencing complex emotions (grief, pain, rage and guilt
about rage), with clear links to their relational history, are related to
larger changes in both symptoms and interpersonal problems. This is
also in line with qualitative accounts of patients undergoing ISTDP
(Town, Abbass, et al., 2017). Still, our results are correlational and do
not determine causality; hence, it may be that treatment improvements
led to higher levels of MUCE or that change in both variables are
byproducts of some other process. Nonetheless, we believe our result
points to specific intreatment processes in need of further study.
In terms of treatment implications, we believe that the ISTDP
framework and concept of MUCE may complement other models of
the disorder. For example, several contemporary theoretical models
of GAD posit that a core feature of the disorder is emotional dysregu-
lation (Behar et al., 2009; Timulak & McElvaney, 2016) and scholars
from different theoretical backgrounds seem to agree that worrying
and rumination in GAD may serve the defensive function of avoiding
negative emotions (Borkovec et al., 2004; CritsChristoph, 2002;
Timulak & McElvaney, 2016). In ISTDP, rumination within the therapy
session is typically regarded as a defense against complex emotions
rising within the therapy relationship. When the patient's defensive
operations are identified, clarified, and challenged by the therapist, it
leads to an intensification of complex feelings in the session that may
TABLE 3 Change in healthcare costs from baseline to 4 years post
treatment (n= 215)
Timeline
Mean physician
costs (SD)
Mean hospital
costs (SD)
Baseline year
a
$748 (685) $5,294 (32,726)
Year 1 (n= 215) $661 (1,278)* $2,365 (11,197)
Year 2 (n= 179) $595 (622) $1,672 (5,790)
Year 3 (n= 150) $500 (531) $765 (2,698)
Year 4 (n= 107) $437 (573) $968 (3,209)
Note. All costs in Canadian dollars (CAD); Year 1 was compared with base-
line year using pairwise ttests.
a
Prior to start of Intensive ShortTerm Dynamic Psychotherapy.
*p< .05.
1318 LILLIENGREN P. ET AL.
trigger a breakthrough of painful emotions related to the patient's rela-
tional history. Thus, the MUCE construct specifically targeted within
the ISTDP framework refers to a particular set of complex emotions
that rise in the context of attachment relationships, including the ther-
apy relationship. Because GAD has been linked with adverse childhood
experiences and insecure attachment (Cassidy et al., 2009; Marganska
et al., 2013), the insession visceral processing of intense, conflicted
emotions related to attachment trauma may be particularly important
in this population.
4.1 |Strengths and limitations
Some of the strengths of this study include the use of a large sample of
GAD patients with a high rate of comorbid conditions, treated in a nat-
uralistic setting by therapists with various levels of training and exper-
tise; thus, the study context is representative of realworld clinical
conditions. Further, the use of routine video recording of sessions for
supervision facilitated treatment fidelity and allowed for the direct
assessment of the intreatment emotional process of interest.
The naturalistic setting of this study is also associated with specific
limitations. First, because there was no control group, we cannot rule
out the possibility that outcomes were due to time passage, regression
to the mean, or other factors. However, it seems unlikely that the pas-
sage of time alone would produce similar outcomes given the brevity
of treatment and the chronic nature of GAD (Hoffman et al., 2008;
Tyrer & Baldwin, 2006; Yonkers et al., 2000). Additionally, diagnoses
were assessed without the use of structured instruments and, because
the original study included a broad range of patients, symptom mea-
sures were nonspecific to GAD. These limitations should be
addressed in future studies, preferably using a randomized controlled
design.
Further, many patients in the study had missing selfreport out-
come data and there were several possible reasons for why data were
missing for any particular patient. To accommodate for this, we
employed maximum likelihood estimation using all available data,
which yields unbiased estimates under the assumption of MAR
(Enders, 2011). Still, our results need to be interpreted with caution
because we cannot exclude the possibility that the results were
affected by the amount of missing data.
While the healthcare data were derived independently from gov-
ernment databases, the limited data access did not allow us to conduct
statistical comparisons between years 2, 3, and 4 in followup
(although the observed values suggest continued cost reductions).
Also, we cannot know whether patients sought additional treatments
from other health care providers such as psychologists who are not
covered by these administrative health data.
In addition to the direction of causality previously mentioned,
there are several other limitations to our examination of the associa-
tion between MUCE and outcome. The degree of MUCE was assessed
by a single rater using a scale that has not yet been adequately evalu-
ated psychometrically, increasing the risk of bias. Also, we used a single
rating per treatment as an indicator of the peak intensity of MUCE,
which may not be an optimal representation of this process given the
likelihood of intreatment variability in MUCE over time. Still, our
results indicate that a single rating of MUCE was related to outcome,
which suggests that the construct may a capture important clinical
phenomena in the treatment of GAD.
5|CONCLUSION
In conclusion, this study offers some convergent preliminary data that
ISTDP may be beneficial for patients with GAD. Further, the results
indicate that the degree of insession emotional activation is related
to greater treatment effects. This is in line with the notion that avoid-
ance of negative emotional states is a core feature maintaining GAD
and suggests that activation and processing of complex emotions
related to adverse attachment experiences, central to the ISTDP
framework, may be an important clinical focus in this population. We
conclude that ISTDP for GAD should be investigated in a randomized
controlled design, accounting for time passage and other factors.
Future processoutcome studies should use repeated assessments of
MUCE and sessional outcomes in order to investigate its impact in
more detail.
NOTE
1
Withingroup effect sizes were calculated using the observed means and
SDs (controlling for the prepost correlation; e.g., Morris & DeShon,
2002, equation 8) and interpreted using criteria proposed by Cohen
(1992).
REFERENCES
Abbass, A. (2015). Reaching through resistance: Advanced psychotherapy
techniques. Kansas City, MO: Seven Leaves Press.
Abbass, A., Campbell, S., Hann, S. G., Lenzer, I., Tarzwell, R., & Maxwell, D.
(2010). Cost savings of treatment of medically unexplained symptoms
using Intensive Shortterm Dynamic Psychotherapy (ISTDP) by a hospi-
tal emergency department. Archives of Medical Psychology,3,3443
http://amphome.org/archives/Nov2010.pdf#page=9
Abbass, A., Kisely, S., Rasic, D., Town, J. M., & Johansson, R. (2015).
Longterm healthcare cost reduction with Intensive Shortterm
Dynamic Psychotherapy in a tertiary psychiatric service. Journal of
Psychiatric Research,64, 114120. http://doi.org/10.1016/j.
jpsychires.2015.03.001
Abbass, A., Sheldon, A., Gyra, J., & Kalpin, A. (2008). Intensive shortterm
dynamic psychotherapy for DSMIV personality disorders: A random-
ized controlled trial. The Journal of Nervous and Mental Disease,196(3),
211216. http://doi.org/10.1097/NMD.0b013e3181662ff0
Abbass, A., Town, J., Ogrodniczuk, J., Joffres, M., & Lilliengren, P. (2017).
Intensive ShortTerm Dynamic Psychotherapy trial therapy. The Journal
of Nervous and Mental Disease (online publication)., http://doi.org/
10.1097/NMD.0000000000000684
Abbass, A., Town, J. M., & Driessen, E. (2012). Intensive ShortTerm
Dynamic Psychotherapy: A systematic review and metaanalysis of out-
come research. Harvard Review of Psychiatry,20(2), 97108. http://doi.
org/10.3109/10673229.2012.677347
American Psychiatric Association. (2000). Diagnostic and statistical manual
of mental disorders (4th ed., text rev.). Washington, DC: American
Psychiatric Association.
Andersson, G., Paxling, B., RochNorlund, P., Ostman, G., Norgren, A.,
Almlöv, J., Silverberg, F. (2012). Internetbased psychodynamic ver-
sus cognitive behavioral guided selfhelp for generalized anxiety
disorder: A randomized controlled trial. Psychotherapy and Psychoso-
matics,81(6), 344355. http://doi.org/10.1159/000339371
Behar, E., DiMarco, I. D., Hekler, E. B., Mohlman, J., & Staples, A. M. (2009).
Current theoretical models of generalized anxiety disorder (GAD):
LILLIENGREN P. ET AL.1319
Conceptual review and treatment implications. Journal of Anxiety Disor-
ders,23(8), 10111023. http://doi.org/10.1016/j.janxdis.2009.07.006
Bender, R., & Lange, S. (2001). Adjusting for multiple testingwhen and
how? Journal of Clinical Epidemiology,54(4), 343349. http://www.
ncbi.nlm.nih.gov/pubmed/17530590
Blagys, M. D., & Hilsenroth, M. J. (2000). Distinctive features of shortterm
psychodynamicinterpersonal psychotherapy: A review of the compar-
ative psychotherapy process literature. Clinical Psychology: Science and
Practice,7(2), 167188. http://doi.org/10.1093/clipsy.7.2.167
Borkovec, T. D., Alcaine, O., & Behar, E. (2004). Avoidance theory of worry
and generalized anxiety disorder. In R. G. Heimberg, C. L. Turk, & D. S.
Mennin (Eds.), Generalized anxiety disorder: Advances in research and
practice (pp. 77108). New York: Guilford Press.
Cassidy, J., LichtensteinPhelps, J., Sibrava, N. J., Thomas, C. L., & Borkovec,
T. D. (2009). Generalized anxiety disorder: Connections with self
reported attachment. Behavior Therapy,40(1), 2338. http://doi.org/
10.1016/j.beth.2007.12.004
Chavooshi, B., Mohammadkhani, P., & Dolatshahee, B. (2016). Efficacy of
Intensive ShortTerm Dynamic Psychotherapy for medically unex-
plained pain: A pilot threearmed randomized controlled trial
comparison with mindfulnessbased stress reduction. Psychotherapy
and Psychosomatics, 123125. http://doi.org/10.1159/000441698
Cohen, J. (1992). A power primer. Psychological Bulletin,112(1), 155159.
http://doi.org/10.1037/00332909.112.1.155
CritsChristoph, P. (2002). Psychodynamicinterpersonal treatment of gen-
eralized anxiety disorder. Clinical Psychology: Science and Practice,9(1),
8184. http://doi.org/10.1093/clipsy.9.1.81
CritsChristoph, P., Gibbons, M. B. C., Narducci, J., Schamberger, M., &
Gallop, R. (2005). Interpersonal problems and the outcome of interper-
sonally oriented psychodynamic treatment of GAD. Psychotherapy:
Theory, Research, Practice, Training,42(2), 211224. http://doi.org/
10.1037/00333204.42.2.211
Cuijpers, P., Sijbrandij, M., Koole, S., Huibers, M., Berking, M., & Andersson,
G. (2014). Psychological treatment of generalized anxiety disorder: A
metaanalysis. Clinical Psychology Review,34(2), 130140. http://doi.
org/10.1016/j.cpr.2014.01.002
Curran, P. J., & Bauer, D. J. (2011). The disaggregation of withinperson and
betweenperson effects in longitudinal models of change. Annual
Review of Psychology,62, 583619. http://doi.org/10.1146/annurev.
psych.093008.100356
Davanloo, H. (1990). Unlocking the unconscious. New York: Wiley & Sons.
Davanloo, H. (2000). Intensive shortterm dynamic psychotherapy: Selected
papers of Habib Davanloo. Chichester, NY: Wiley & Sons.
Davanloo, H. (2001). Extended major direct access to the unconscious.
European Psychotherapy,2(1), 2570.
Davanloo, H. (2005). Intensive shortterm dynamic psychotherapy. In B. J.
Sadock, V. A. Sadock, & H. I. Kaplan (Eds.), Kaplan and Sadock's compre-
hensive textbook of psychiatry (pp. 26282652). Philadelphia, PA:
Lippincot Williams & Wilkins.
Derogatis, L. R., & Melisaratos, N. (1983). The brief symptom inventory: An
introductory report. Psychological Medicine,13, 595605. https://doi.
org/10.1017/S0033291700048017
Diener, M. J., Hilsenroth, M. J., & Weinberger, J. (2007). Therapist affect
focus and patient outcomes in psychodynamic psychotherapy: A
metaanalysis. The American Journal of Psychiatry,164(6), 936941.
http://doi.org/10.1176/appi.ajp.164.6.936
Enders, C. K. (2011). Analyzing longitudinal data with missing values. Reha-
bilitation Psychology,56(4), 267288. https://doi.org/10.1037/
a0025579
Fisher, H., Atzilslonim, D., BarKalifa, E., Rafaeli, E., & Peri, T. (2016). Emo-
tional experience and alliance contribute to therapeutic change in
psychodynamic therapy. Psychotherapy,53(1), 105116. http://doi.
org/10.1037/pst0000041
Goldman, R. N., Greenberg, L. S., & Pos, A. E. (2005). Depth of emotional
experience and outcome. Psychotherapy Research,15, 248260.
https://doi.org/10.1080/10503300512331385188
Grant, B. F., Hasin, D. S., Stinson, F. S., Dawson, D. A., June Ruan, W., Gold-
stein, R. B., Huang, B. (2005). Prevalence, correlates, comorbidity,
and comparative disability of DSMIV generalized anxiety disorder in
the USA: Results from the National Epidemiologic Survey on Alcohol
and Related Conditions. Psychological Medicine,35, 17471759.
http://doi.org/10.1017/S0033291705006069
Greenberg, L. S., & Pascualleone, A. (2006). Emotion in psychotherapy: A
practicefriendly research review. Journal of Clinical Psychology: In Ses-
sion,62(5), 611630. http://doi.org/10.1002/jclp
Hoffman, D. L., Dukes, E. M., & Wittchen, H.U. (2008). Human and eco-
nomic burden of generalized anxiety disorder. Depression and Anxiety,
25(1), 7290. http://doi.org/10.1002/da.20257
Hofmann, S. G., & Smits, J. A. J. (2008). Cognitivebehavioral therapy for
adult anxiety disorders: A metaanalysis of randomized placebocon-
trolled trials. The Journal of Clinical Psychiatry,69(4), 621632. http://
doi.org/10.4088/JCP.v69n0415
Horowitz, L. M., Rosenberg, S. E., Baer, B. A., Ureño, G., & Villaseñor, V. S.
(1988). Inventory of interpersonal problems: Psychometric properties
and clinical applications. Journal of Consulting and Clinical Psychology,
56(6), 885892. http://www.ncbi.nlm.nih.gov/pubmed/3204198
Johansson, R., Town, J. M., & Abbass, A. (2014). Davanloo's intensive short
term dynamic psychotherapy in a tertiary psychotherapy service: Over-
all effectiveness and association between unlocking the unconscious
and outcome. PeerJ,2, e548. http://doi.org/10.7717/peerj.548
Kessler, R. C., Gruber, M., Hettema, J. M., Hwang, I., Sampson, N., & Yon-
kers, K. A. (2008). Comorbid major depression and generalized
anxiety disorders in the National Comorbidity Survey followup. Psy-
chological Medicine,38(3), 365374. https://doi.org/10.1017/
S0033291707002012
Leichsenring, F., Salzer, S., Jaeger, U., Kachele, H., Kreische, R., Leweke, F.,
Leibing, E. (2009). Shortterm psychodynamic psychotherapy and
cognitivebehavioral therapy in generalized anxiety disorder: A random-
ized, controlled trial. American Journal of Psychiatry,166(8), 875881.
http://doi.org/10.1176/appi.ajp.2009.09030441
Levy Berg, A., Sandell, R., & Sandahl, C. (2009). Affectfocused body psy-
chotherapy in patients with generalized anxiety disorder: Evaluation
of an integrative method. Journal of Psychotherapy Integration,19(1),
6785. http://doi.org/10.1037/a0015324
Lilliengren, P., Johansson, R., Lindqvist, K., Mechler, J., & Andersson, G.
(2016). Efficacy of experiential dynamic therapy for psychiatric condi-
tions: A metaanalysis of randomized controlled trials. Psychotherapy,
53(1), 90104. http://doi.org/10.1037/pst0000024
Marganska, A., Gallagher, M., & Miranda, R. (2013). Adult attachment, emo-
tion dysregulation, and symptoms of depression and generalized
anxiety disorder. American Journal of Orthopsychiatry,83(1), 131141.
http://doi.org/10.1111/ajop.12001
Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2005). Prelim-
inary evidence for an emotion dysregulation model of generalized
anxiety disorder. Behaviour Research and Therapy,43(10), 12811310.
http://doi.org/10.1016/j.brat.2004.08.008
Morris, S. B., & DeShon, R. P. (2002). Combining effect size estimates in
metaanalysis with repeated measures and independentgroups
designs. Psychological Methods,7(1), 105125. http://doi.org/
10.1037/1082989X.7.1.105
Pos, A. E., Greenberg, L. S., Goldman, R. N., & Korman, L. M. (2003). Emo-
tional processing during experiential treatment of depression. Journal
of Consulting and Clinical Psychology, 71, 10071016. http://dx.doi .
org/10.1037/0022006X.71.6.1007
Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis. Model-
ing change and event occurrence. New York, NY: Oxford University Press
https://doi.org/10.1093/acprof:oso/9780195152968.001.0001
Solbakken, O. A., & Abbass, A. (2015). Intensive shortterm dynamic resi-
dential treatment program for patients with treatmentresistant
1320 LILLIENGREN P. ET AL.
disorders. Journal of Affective Disorders,181,6777. http://doi.org/
10.1016/j.jad.2015.04.003
Solbakken, O. A., Hansen, R. S., & Monsen, J. T. (2011). Affect integration
and reflective function: Clarification of central conceptual issues.
Psychotherapy Research,21(4), 482496. http://doi.org/10.1080/
10503307.2011.583696
Sugiura, Y., & Sugiura, T. (2015). Emotional intensity reduces later general-
ized anxiety disorder symptoms when fear of anxiety and negative
problemsolving appraisal are low. Behaviour Research and Therapy,71,
2733. http://doi.org/10.1016/j.brat.2015.05.015
Timulak, L., & McElvaney, J. (2016). Emotionfocused therapy for general-
ized anxiety disorder: An overview of the model. Journal of
Contemporary Psychotherapy,46(1), 4152. http://doi.org/10.1007/
s1087901593107
Town, J. M., Abbass, A., & Bernier, D. (2013). Effectiveness and cost effec-
tiveness of Davanloo's Intensive ShortTerm Dynamic Psychotherapy:
Does unlocking the unconscious make a difference? American Journal
of Psychotherapy,67(1), 89108.
Town, J. M., Abbass, A., Stride, C., & Bernier, D. (2017). A randomised con-
trolled trial of Intensive ShortTerm Dynamic Psychotherapy for
treatment resistant depression: The Halifax depression study. Journal
of Affective Disorders, (Online publication), 214,1525. http://doi.org/
10.1016/j.jad.2017.02.035
Town, J. M., & Driessen, E. (2013). Emerging evidence for intensive short
term dynamic psychotherapy with personality disorders and somatic
disorders. Psychiatric Annals,43(11), 502507.
Town, J. M., Lomax, V., Abbass, A., & Hardy, G. (2017). The role of
emotion in psychotherapeutic change for medically unexplained
symptoms. Psychotherapy Research,13. http://doi.org/10.1080/
10503307.2017.1300353
Town, J. M., Salvadori, A., Falkenstrom, F., Bradley, S., & Hardy, G. (in press).
Is affect experiencing therapeutic in depressive disorders: The relation-
ship between affect experiencing and distress in Intensive ShortTerm
Dynamic Psychotherapy. Psychotherapy.
Tyrer, P., & Baldwin, D. (2006). Generalised anxiety disorder. Lancet,368,
21562166. http://doi.org/10.1016/S01406
Wampold, B. E., & Serlin, R. C. (2000). The consequence of ignoring a
nested factor on measures of effect size in analysis of variance.
Psychological Methods,5(4), 425433. http://doi.org/10.1037/1082
989X.5.4.425
Whelton, W. J. (2004). Emotional processes in psychotherapy: Evidence
across therapeutic modalities. Clinical Psychology & Psychotherapy,
11(1), 5871. http://doi.org/10.1002/cpp.392
Yonkers, K. A., Dyck, I. R., Warshaw, M., & Keller, M. B. (2000). Factors
predicting the clinical course of generalised anxiety disorder. British
Journal of Psychiatry,176(6), 544549. http://doi.org/10.1192/
bjp.176.6.544
How to cite this article: Lilliengren P, Johansson R, Town JM,
Kisely S, Abbass A. Intensive ShortTerm Dynamic Psychotherapy
for generalized anxiety disorder: A pilot effectiveness and process
outcome study. Clin Psychol Psychother. 2017;24:13131321.
https://doi.org/10.1002/cpp.2101
LILLIENGREN P. ET AL.1321
... If not, therapists can interpret when the patient is in proximity of the insight, but has reached the limit of their own exploratory process. It is also thought best to have the patient be affectively connected to the topic as well (e.g., see Strachey, 1934) and, more generally, the exploration of affect seems to be positively associated with outcome (e.g., Lilliengren et al., 2017). What is somewhat more variable across specific dynamic therapies is the recommended level of tentativeness of the intervention (i.e., presenting the interpretation as a hypothesis or as a more definitive statement). ...
... However, there was some evidence that psychodynamic therapy for generalized anxiety disorder may be significantly less effective than alternate treatments at follow-up. It should be noted that a recent pilot study of Intensive Short-Term Dynamic Therapy for generalized anxiety disorder (Lilliengren et al., 2017) was associated with significant symptom reductions as well as reductions in overall healthcare costs at 4 years post-treatment. Therefore, more research on generalized anxiety disorder appears warranted. ...
... From a broader economic standpoint, additional documentation of the cost-effectiveness of psychodynamic therapy would be beneficial. The existing studies (e.g., Abbass et al., 2015;Lilliengren et al., 2017) appear promising, especially in terms of reducing long-term healthcare costs. ...
Chapter
The phrase “psychodynamic therapy” refers to the family of treatment approaches based on the early work of Sigmund Freud. It is one of the major forms of contemporary psychotherapy practice and has undergone a great deal of theoretical and technical modification over the past 120 years. Its main techniques have been catalogued and many forms of psychodynamic therapy have been manualized. There is mounting evidence not only in favor of its efficacy, but also in terms of theory-based mechanisms of change and the research generativity of its concepts (e.g., attachment theory).
... If not, therapists can interpret when the patient is in proximity of the insight, but has reached the limit of their own exploratory process. It is also thought to be best to have the patient be affectively connected to the topic as well (e.g., see Strachey, 1934) and, more generally, the exploration of affect seems to be positively associated with outcome (e.g., Lilliengren, Johansson, Town, Kisely, & Abbas, 2017). What is somewhat more variable across specific dynamic therapies is the recommened level of tentativeness of the intervention (i.e., presenting the interpretation as a hypothesis or as a more definitive statement). ...
... From a broader economic standpoint, additional documentation of the cost-effectiveness of psychodynamic therapy would be beneficial. The existing studies (e.g., Abbass, Kisely, Rasic, Town, & Johansson, 2015;Lilliengren et al., 2017) appear promising, especially in terms of reducing long-term healthcare costs. ...
Chapter
The phrase “psychodynamic therapy” refers to the family of treatment approaches based on the early work of Sigmund Freud. It is one of the major forms of contemporary psychotherapy practice and has undergone a great deal of theoretical and technical modification over the past 120 years. Its main techniques have been cataloged and many forms of psychodynamic therapy have been manualized. There is mounting evidence not only in favor of its efficacy, but also in terms of theory-based mechanisms of change and the research generativity of its concepts (e.g., attachment theory).
... ISTDP has been found efficacious relative to control conditions and bonafide treatments when applied with a broad range of common mental disorders (Lilliengren, Johansson, Lindqvist, Mechler, & Andersson, 2016). It has been found effective for anxiety disorders including posttraumatic stress disorder (Roggenkamp, Abbass, Town, Kisely, & Johansson, 2019), mixed anxiety and depressive disorders (Solbakken & Abbass, 2016), generalized anxiety disorder (Lilliengren, Johansson, Town, Kisely, & Abbass, 2017), and social anxiety disorder (SAD; Rahmani, Abbass, Hemmati, Mirghaed, & Ghaffari, 2020): In the first three of these studies the treatment was also found to be cost-effective. ISTDP has been shown to produce structural personality changes as well as reduce symptoms in patients with mixed anxiety disorders (Rocco et al., 2014). ...
... ISTDP was effective in reducing measures of SAD anxiety compared with nontreated controls. This study adds further to evidence for ISTDP's effectiveness for anxiety disorders (Lilliengren et al., 2017;Rahmani et al., 2020;Rocco et al., 2014). In comparing ISTDP without challenge to the standard ISTDP format we found the two to be equally effective suggesting that emotional processing maybe more important than challenge or other interventions. ...
Article
Objectives Intensive short‐term dynamic psychotherapy (ISTDP) requires the technique of challenge to defenses when treating resistant patients. As the technique of challenge is difficult for some therapists to practice, it leads us to question whether challenge can be replaced by clarification of defenses without losing treatment effectiveness. This study compared ISTDP with two different technical emphases while treating social anxiety disorder (SAD). Method Forty‐two subjects with DSM‐5 SAD were randomly assigned to either a waitlist control, 10 sessions of ISTDP with the use of challenge or 10 sessions of ISTDP without the use of challenge. Results ISTDP led to significant, sustained symptom reduction on the Liebowitz Social Anxiety Scale (LSAS‐SR) compared with the control group. There were no significant outcome differences between standard ISTDP and ISTDP where challenge was restricted. Conclusion ISTDP is efficacious for SAD. ISTDP may be effective for SAD without the use of challenge elements.
... 93). Patients had presented different treatment times and oscillations in the severity of psychopathological symptoms, which can be explained by the fact that DP is conducted considering the patients' capacity to process and tolerate feelings, which reflects directly on the length of therapy [23,41,42]. Since the focus of DP is on raising awareness of unconscious thoughts and feelings related to traumatic events, the patients needed time to slow down the automatic processing of trauma stimuli and enhance emotion regulation in facing their experiences [25]. ...
Article
Full-text available
In Portugal, forest fires are responsible for disasters that tend to be repeated annually, leading to dramatic consequences, such as those that have occurred in 2017, with the destruction of hundreds of houses and the deaths of dozens of people. Firefighters who are exposed to these potentially traumatic events are considered a high-risk group for the development of stress-related disorders. The aim of this study was to monitor the progress of two firefighters with symptoms of post-traumatic stress disorder (PTSD) treated through dynamic psychotherapy (DP) and to assess the feasibility of implementing this intervention within fire departments. A female firefighter and a male firefighter, with similar sociodemographic characteristics and PTSD symptom severity, were selected to verify the treatment applicability for both genders. The symptomatology changes were assessed through a set of instruments (PHQ-15, PCL-5, BSI, DASS, and CALPAS-P) applied every three months over 15 months (including pre-treatment, treatment period, and post-treatment). DP seemed to be an effective treatment for PTSD symptoms, with patients showing a state of increasing improvement even after the end of treatment. The acceptability to firefighters, the treatment adherence, the therapeutic alliance, and the reduction in PTSD symptoms suggest feasibility for implementing this intervention inside the Portuguese fire departments.
... Although some reports have been uncontrolled, naturalistic studies, treatments are reported to yield significant improvements in symptoms. One study reported similar outcomes to medications in the same clinic (Ferrero et al. 2007), and another study reported reduced health care costs (Lilliengren et al. 2017). Many of these treatments posit that anxiety is related to conflictual interpersonal attachment patterns and incomplete processing of past traumatic or other emotional events. ...
Chapter
Psychological treatment of generalized anxiety disorder.
... An interesting non-pharmacological option is psychotherapeutic support, such as the short-term psychodynamic psychotherapy (STPP) [11] or the mindfulness-based intervention [12]. Our group previously demonstrated that the STPP, a form of psychotherapy proposed by Edmond Gilliéron and widely used in the treatment of several psychiatric diseases [13][14][15], reduced headache days and decreased the chance of relapse into medication overuse, when added to pharmacological therapy in MOH patients [11]. ...
Preprint
Full-text available
Background: Chronic migraine (CM) is an ensemble disorder in which pain, comorbidities, and medication overuse headache (MOH) interact to determine the degree of disability. No data support the choice of a monotherapy rather than a multidisciplinary preventive approach. The aim of this study is testing the superiority of multidisciplinary approach, i.e. Short-term Psychodynamic Psychotherapy (STPP) plus drug of choice, vs. monotherapy, i.e. OnabotulinumtoxinA. Methods: We consecutively recorded data from CM patients who underwent STPP or OnaBoNT-A, with a 3-month follow-up schedule. Headache days and analgesics intake were monitored as primary outcome measures to determine the benefit. Propensity score matching (PSM) was used to eliminate discrepancies between groups. Discriminant function analysis was used to pinpoint predictive factors associated to the clinical response. Results: 96 patients with CM (64% with MOH) were treated with STPP and 54 (59% with MOH) with OnaBoNT-A. At baseline, OnaBoNT-A patients had more failed preventive therapies, more years of illness and chronicity, and were older, while STPP patients were more depressed and had an higher HIT-6 (p<0.001). Both STPP and OnaBoNT-A patients showed a significant reduction of headache days (STPP: -14 vs. OnaBoNT-A:-14.3) and analgesics intake (STPP: -12,3 vs. OnaBoNT-A -13.5 pills/month), respectively. MOH diminished more in STPP, adherence was higher in OnaBoNT-A. Results were confirmed after that PSM balanced the two groups for all these variables that resulted different (but age). DFA predicts responders in pooled and OnaBoNT-A group. Conclusions: OnaBoNT-A monotherapy produced similar results of a multidisciplinary approach with psychotherapy plus medication. STPP provided a better short term response but short- and long-term adherence was higher in OnaBoNT-A.
Article
Objective: The objective of this study was to test the superiority of multidisciplinary approach, that is, Short-Term Psychodynamic Psychotherapy (STPP) plus drug of choice, versus monotherapy, that is, OnabotulinumtoxinA (OnaBoNT-A). Method: We consecutively recorded data from chronic migraine (CM) patients, with or without medication overuse headache (MOH), who underwent STPP or OnaBoNT-A, with a 3-month follow-up schedule. Headache days and analgesics intake were monitored as primary outcome measures. Propensity score matching (PSM) was used to eliminate discrepancies between groups. Discriminant function analysis (DFA) was used to pinpoint predictive factors associated with the clinical response. Results: 96 patients with CM (64% with MOH) were treated with STPP and 54 (59% with MOH) with OnaBoNT-A. At baseline, OnaBoNT-A patients had more failed preventive therapies, more years of illness and chronicity, and were older; STPP patients were more depressed and had a higher HIT-6. Both STPP and OnaBoNT-A patients showed a significant reduction of headache days (STPP: -14 vs. OnaBoNT-A:-14.3) and analgesics intake (STPP: -12,3 vs. OnaBoNT-A -13.5 pills/month), respectively. MOH diminished more in STPP, adherence was higher in OnaBoNT-A. Results were confirmed after PSM balancing of the groups for those variables that resulted as different (but age). Conclusion: OnaBoNT-A monotherapy produced similar results to psychotherapy plus medication, after correcting for baseline differences.
Chapter
This chapter describes the evolution of short-term psychodynamic psychotherapy. Its conceptual origins can be traced to The Budapest School of Psychoanalysis where Sandor Ferenczi investigated innovations in technique. Franz Alexander made vital contributions with his ideas about corrective emotional experiences. Ferenczi’s disciples Alice and Michael Balint emigrated to London and joined the Tavistock Clinic. The Tavistock became the hub of the British object relations school which included the Balints, David Malan, Donald Winnicott and John Bowlby. Bowlby transformed key ideas of object relations into attachment theory. Starting in the 1960s, the experiential dynamic psychotherapies began to flourish with Habib Davanloo, Diana Fosha, Leigh McCullough and Allan Abbass.
Article
Interpersonal dysfunction is posited to maintain worry and generalized anxiety disorder (GAD). It has been suggested that the low remission rates in psychotherapy for GAD may be attributable, in part, to inadequately addressing interpersonal dysfunction. This paper systematically reviewed the literature examining the moderating role of interpersonal dysfunction on GAD psychotherapy outcomes and change in interpersonal dysfunction over the course of GAD treatment. Thirteen studies were identified, seven of which examined the relationship between interpersonal dysfunction or distress and treatment outcome and nine investigated change in interpersonal dysfunction over the course of psychotherapy. The majority of studies indicated that interpersonal dysfunction improves following psychotherapy. However, there is preliminary evidence that not all subscales of interpersonal dysfunction improve, including subscales relevant to GAD pathology such as overly-nurturant dysfunction. Further, greater interpersonal dysfunction predicted worse treatment outcomes. As such, interpersonal dysfunction may hinder treatment success and further research is needed to delineate for whom additional or integrated interpersonal interventions may be needed. Approaches to target interpersonal dysfunction in GAD are discussed.
Article
Full-text available
Objective: Research has shown that Intensive Short-Term Dynamic Psychotherapy (ISTDP) can effectively decrease pain intensity and improve quality of life in patients with Medically Unexplained Pain (MUP). Understanding that not all patients with MUP have access to in-person ISTDP, this study aims to investigate the efficacy of an Internet-Delivered ISTDP (ID-ISTDP) for individuals with MUP using Skype™ in comparison with Treatment as Usual (TAU). Method: In this randomized controlled trial, 100 patients were randomly allocated into ID-ISTDP (n = 50) and TAU (n = 50) groups. Treatment intervention consisted of sixteen weekly, hour-long therapy sessions. The primary outcome was perceived pain assessed using the Numeric Pain Rating Scale (NPRS). The secondary outcome included Depression Anxiety Stress Scale-21 (DASS-21), Emotion Regulation Questionnaire (ERQ), Mindful Attention Awareness Scale (MAAS) and Quality of Life Inventory (QOLI). Blind assessments were conducted at the baseline, post-treatment and at a six-month follow-up. 2 Results: In the intention-to-treat analysis, pain symptoms in the intervention group were significantly reduced (p < 0.001), while a reduction was not observed in the TAU group (p = 0.651). Moreover, there were significant decreases in depression, anxiety and stress, as well as a greater increase in emotion regulation functioning, mindfulness and quality of life observed in the intervention group six-month after the treatment compared to the TAU condition. Conclusion: The results of this pilot trial demonstrates that sixteen weeks of ISTDP delivered by Skype™ can significantly improve pain intensity and clinical symptoms of MUP.
Article
Full-text available
Accumulating evidence suggests that the therapeutic alliance and clients' contact with emotions during therapy sessions can be effective in reducing their suffering outside of sessions. However, the complex associations among these determinants are not yet clear. Using data collected in therapy on a session-by-session basis, this study explored (a) the temporal associations between emotional experience and the therapeutic alliance; (b) the temporal associations between emotional experience and clients' level of functioning; and (c) the direct and indirect associations among emotional experience, the therapeutic alliance, and functioning. Clients (N = 101) undergoing psychodynamic therapy completed a functioning and distress measure prior to each session, and reported on their emotional experience and perceived alliance strength following each session. Longitudinal multilevel models indicated that higher therapeutic alliance scores at the end of 1 session predicted a greater emotional experience in the next session but that emotional experience did not predict subsequent levels of alliance. The results provided evidence of reciprocal prediction in which a previous emotional experience predicted a subsequent change in functioning and vice versa. Finally, the alliance predicted emotional experience, which, in turn, predicted functioning; hence, alliance strength indirectly predicted clients' level of functioning. Findings indicate that emotional experience and the therapeutic alliance are important determinants of the therapeutic process, which contribute to predict clients' improvement in functioning within psychodynamic treatment. (PsycINFO Database Record
Article
This study examined the effects of trial therapy interviews using intensive short-term dynamic psychotherapy with 500 mixed sample, tertiary center patients. Furthermore, we investigated whether the effect of trial therapy was larger for patients who had a major unlocking of the unconscious during the interview compared with those who did not. Outcome measures were the Brief Symptom Inventory (BSI) and the Inventory of Interpersonal Problems (IIP), measured at baseline and at 1-month follow-up. Significant outcome effects were observed for both the BSI and the IIP with small to moderate preeffect/posteffect sizes, Cohen's d = 0.52 and 0.23, respectively. Treatment effects were greater in patients who had a major unlocking of the unconscious compared with those who did not. The trial therapy interview appears to be beneficial, and its effects may relate to certain therapeutic processes. Further controlled research is warranted.
Article
Objectives: Evidence of the contribution of emotional processes to the emergence, maintenance, and experience of medically unexplained symptoms (MUS) suggests that clinical approaches which target these processes could be beneficial. In this study, qualitative methods were used to examine patients' perspectives and subjective experiences of emotional processes in the context of a psychotherapy assessment and treatment service for MUS provided in a hospital emergency department (ED). Methods: Seven semi-structured interviews were conducted with ED patients presenting with MUS who received a course of intensive short-term dynamic psychotherapy treatment. Results: Interpretative phenomenological analysis was employed with three superordinate themes emerging: Barriers to examining emotional processes; reflections on the therapeutic process; psychological change; and improved well-being. Obstacles to clinical engagement in treatment for MUS were described in relation to patients' and therapists' ability to identify, address, and utilize emotion processes. Specific elements of this work were identified as integral components of the psychotherapy change process for MUS. Conclusions: Directly observing the physical effects of emotional experiencing in MUS provides sensory evidence that can enable patients to make mind-body connections. Psycho-emotional processes warrant further study to explore the applicability to other conceptual models for assessing and treating MUS.
Article
Background While short-term psychodynamic psychotherapies have been shown effective for major depression, it is unclear if this could be a treatment of choice for depressed patients who have not sufficiently responded to existing treatments and commonly have chronic and complex health issues. Method This superiority trial used a single blind randomised parallel group design to test the effectiveness of time-limited Intensive Short-Term Dynamic Psychotherapy (ISTDP) for treatment resistant depression (TRD). Patients referred to secondary care community mental health teams (CMHT) who met DSM-IV criteria for major depressive episode, had received antidepressant treatment 6 weeks, and had Hamilton Depression Rating Scale (HAM-D) scores of  16 were recruited. The effects of 20 sessions of ISTDP were judged through comparison against secondary care CMHT treatment as usual (TAU). The primary outcome was HAM-D scores at 6 months. Secondary outcomes included dichotomous measures of both remission (defined as HAM-D score  7) and partial remission (defined as HAM-D score  12). Results Sixty patients were randomised to 2 groups (ISTDP=30 and TAU=30), with data collected at baseline, 3, and 6 months. Multi-level linear regression modelling showed that change over time on both depression scales was significantly greater in the ISTDP group in comparison to TAU. Statistically significant between-group treatment differences, in the moderate to large range, favouring ISTDP, were observed on both the observer rated (Cohen’s d = 0.75) and self-report measures (Cohen’s d = 0.85) of depression. Relative to TAU, patients in the ISTDP group were significantly more likely after 6 months to achieve complete remission (36.0% vs. 3.7%) and partial remission (48.0% vs. 18.5%). Limitations It is unclear if the results are generalizable to other providers, geographical locations and cultures. Conclusions Time-limited ISTDP appears an effective treatment option for TRD, showing large advantages over routine treatment delivered by secondary care services.
Article
Although the consequences of ignoring a nested factor on decisions to reject the null hypothesis of no treatment effects have been discussed in the literature, typically researchers in applied psychology and education ignore treatment providers (often a nested factor) when comparing the efficacy of treatments. The incorrect analysis, however, not only invalidates tests of hypotheses, but it also overestimates the treatment effect. Formulas were derived and a Monte Carlo study was conducted to estimate the degree to which the F statistic and treatment effect size measures are inflated by ignoring the effects due to providers of treatments. These untoward effects are illustrated with examples From psychotherapeutic treatments.