See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/318874687

Short-Term Group Therapy for Complicated Grief: The Relationship Between

Patients’ In-Session Reflection and Outcome

Article  in  Psychiatry Interpersonal & Biological Processes · April 2017

DOI: 10.1080/00332747.2016.1220231

CITATIONS READS

3 326

6 authors, including:

David Kealy Carlos Sierra Hernandez

University of British Columbia - Vancouver Simon Fraser University
110 PUBLICATIONS   353 CITATIONS    14 PUBLICATIONS   65 CITATIONS   

SEE PROFILE SEE PROFILE

William Piper Anthony Joyce

University of British Columbia - Vancouver University of Alberta
172 PUBLICATIONS   5,668 CITATIONS    156 PUBLICATIONS   4,968 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Anthony Joyce on 23 March 2018.

The user has requested enhancement of the downloaded file.

 Psychiatry
Interpersonal and Biological Processes

ISSN: 0033-2747 (Print) 1943-281X (Online) Journal homepage: http://www.tandfonline.com/loi/upsy20

Short-Term Group Therapy for Complicated Grief:

The Relationship Between Patients’ In-Session
Reflection and Outcome

David Kealy, Carlos A. Sierra-Hernandez, William E. Piper, Anthony S. Joyce,

Rene Weideman & John S. Ogrodniczuk

To cite this article: David Kealy, Carlos A. Sierra-Hernandez, William E. Piper, Anthony S. Joyce,
Rene Weideman & John S. Ogrodniczuk (2017) Short-Term Group Therapy for Complicated Grief:
The Relationship Between Patients’ In-Session Reflection and Outcome, Psychiatry, 80:2, 125-138

To link to this article: http://dx.doi.org/10.1080/00332747.2016.1220231

Published online: 02 Aug 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=upsy20

Download by: [The University of British Columbia Library] Date: 02 August 2017, At: 13:37
 Psychiatry, 80:125–138, 2017 125
Copyright Ó Washington School of Psychiatry
ISSN: 0033-2747 print / 1943-281X online
DOI: 10.1080/00332747.2016.1220231

Short-Term Group Therapy for Complicated Grief:

The Relationship Between Patients’ In-Session
Reflection and Outcome
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017

David Kealy, Carlos A. Sierra-Hernandez , William E. Piper, Anthony S. Joyce,

Rene Weideman, and John S. Ogrodniczuk

Objective: The objective of the present study was to examine the nature of
patients’ work in two types of short-term group psychotherapy. The study sought
to investigate the relationship between patients’ psychodynamic work versus
supportive work in group psychotherapy and treatment outcome at termination
and at 6-month follow-up. Psychodynamic work refers to reflection regarding
intrapsychic motivations, defenses, and relational patterns, and supportive work
refers to practical problem solving. Method: Participants were 110 patients who
completed two forms of group therapy for complicated grief: interpretive therapy
and supportive therapy. Two types of patients’ in-session activity—psychody-
namic work and supportive work—were rated by group therapists in both treat-
ments. Pre-post and follow-up outcome domains included general symptoms, grief
symptoms, and life dissatisfaction/severity of target objectives. Results: There was
no significant difference in the nature of patients’ therapeutic work between
interpretive and supportive groups. Psychodynamic work was associated with
pre-post improvement in grief symptoms. Psychodynamic work was also asso-
ciated with further improvement in grief symptoms at 6-month follow-up, along
with improvement in broader symptom domains. Supportive work was not asso-
ciated with any pre-post or follow-up benefit. Conclusion: The findings provide
evidence that psychodynamic work—focused on the development of insight and
self-reflection—in group psychotherapy can contribute to further benefit after the
completion of treatment. This finding cut across two approaches to short-term
group therapy for complicated grief, suggesting that it may reflect a general
curative mechanism of group treatments.

The nature of patients’ work in psychother- ideally leading to durable long-term benefit.
apy can have important implications for One particular kind of therapeutic work with
treatment outcome, with productive work the potential for continued returns is the

David Kealy, PhD, William E. Piper, PhD, Rene Weideman, PhD, and John S. Ogrodniczuk, PhD, are affiliated with
the Department of Psychiatry, University of British Columbia. Carlos A. Sierra-Hernandez, MA, is affiliated with the
Department of Psychology, Simon Fraser University. Anthony S. Joyce is affiliated with the Department of Psychiatry,
University of Alberta.
Address correspondence to David Kealy, PhD, Department of Psychiatry, University of British Columbia, #420 –
5950 University Blvd., Vancouver, BC V6T 1Z3. E-mail: david.kealy@ubc.ca
 126
effort to develop insight and self-reflection
(Connolly Gibbons, Crits-Christoph, Barber,
& Schamberger, 2007; Jones, 2000). Indeed,
the development of insight and self-reflection
is central to psychodynamic therapy, which
emphasizes the understanding of complex
and often implicit mental phenomena—
including defense mechanisms and conflicted
motivations—in perpetuating current pro-
blems (Gabbard, 2005; Lacewing, 2014;
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
Messer & Williams, 2007). Furthermore,
the process of working in therapy toward
self-understanding may be more salient than
the acquisition of a particular insight (Con-
nolly Gibbons et al., 2007). Many psychody-
namic clinicians would thus agree with
Lessing’s dictum that “the search for truth
is more precious than its possession” (Ein-
stein, 1954/1982, p. 335). Indeed, it has been
suggested that a core process of psychody-
namic therapy is the initiation of self-reflec-
tive abilities in order to facilitate insight
development beyond termination—regard-
less of whether a discrete insight comes to
fruition during the treatment itself (Jones,
2000; Lacewing, 2014; Ogden & Gabbard,
2010). This perspective, emphasized in recent
literature regarding mentalization in psy-
chotherapy, suggests that patients work in
therapy to discover and hone their capacity
to understand and reflect upon the inner
workings of their mind (Allen, 2013; Fonagy
& Allison, 2014; Lacewing, 2014). This
reflective therapeutic work—focused on
inner conflicts, motivations, and relational
processes—can be considered patients’ psy-
chodynamic work, in contrast to efforts
aimed at practical adaptation or problem
solving, which we term supportive work.
We define psychodynamic work as the
patient’s attempt to understand issues in
terms of wishes, fears, and defenses—includ-
ing affects, cognitions, and relational pat-
terns that have motivational properties.
Psychodynamic work thus involves the effort
to reflect upon one’s psychological function-
ing in terms of motivation and/or conflict. By
contrast, supportive work is defined as the
patient’s attempt to adapt to or solve
Psychodynamic Work in Group Psychotherapy
problems in terms of practical activities
rather than through intrapsychic reflection.
It may be the conduct of psychodynamic
work during therapy—essentially practicing
reflective abilities—that contributes to contin-
ued post-treatment benefits. Patients might
extend this work after termination, applying
an intrapsychic perspective more broadly for
the purpose of achieving further insight and
problem resolution. Thus, patients’ in-session
psychodynamic work could yield benefits that
surpass the discrete insights or problems
addressed during their treatment. This may be
particularly relevant for short-term therapies,
where the acquisition of insight or the solving
of specific problems can be limited by time
constraints. Patients would derive consider-
ably greater benefit from short-term therapies
if such treatments can foster broadly applic-
able reflective abilities rather than the solving
of particular problems.
The relationship between psychody-
namic work and therapy outcome may be
especially salient for disorders that involve a
considerable degree of psychological conflict.
One such condition is complicated grief (CG).
While the loss of a loved one through death
involves significant emotional pain—with
reactions such as shock, sadness, anger, and
preoccupied longings for the lost person—CG
is an intense and unrelenting experience of
anguish that interferes with the ability to
mourn and recover. Individuals who experi-
ence CG often have prolonged difficulty
accepting the loss, intrusive memories, anger
and bitterness, and disengagement from nor-
mal life activities (Shear & Shair, 2005).
Although the pathway from normal bereave-
ment to CG is unique among individuals, the
emotional responses to the relationship
between the bereaved and the deceased—
including the ongoing psychological bond
with the deceased—has been consistently iden-
tified as contributing to CG (Bowlby, 1980;
Lobb et al., 2010; Maccallum & Bryant,
2013; Piper, Ogrodniczuk, Joyce, & Weide-
man, 2011; Shear & Shair, 2005). In the case
of CG, recovery may thus be facilitated
through efforts aimed at understanding
 Kealy et al.
thwarted yearnings, maladaptive defenses, and
ambivalence in close relationships.
Psychotherapy, in both individual
and group formats, has been found to be
efficacious in the treatment of CG, with
treatment effects tending to persist at fol-
low-up (Wittouck, Van Autreve, De Jae-
gere, Portzky, & van Heeringen, 2011).
The present study examined the role of
patients’ psychodynamic work, compared
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
with problem solving or supportive work,
in two forms of group psychotherapy for
CG. This study used data from a compara-
tive trial that investigated the impact of
group composition in short-term interpre-
tive and supportive group therapies for CG
(Piper, Ogrodniczuk, Joyce, Weideman, &
Rosie, 2007). Although there were no out-
come differences between these two
approaches, a composition effect was
found: The higher the percentage of
patients in a group with relatively mature
relationships, the better the outcome for all
patients in the group. Follow-up analyses
found nearly three-quarters of patients to
have achieved clinically significant
improvement at 6 months post-treatment
(Piper, Ogrodniczuk, Joyce, & Weideman,
2009). Although overall work in therapy
was found to be related to pre-post out-
come in a previous trial of group therapy
for CG (Piper, Ogrodniczuk, Joyce, Weide-
man, & Rosie, 2002), the relationship
between patients’ psychodynamic work
and outcome remains unclear. Moreover,
the potential role of psychodynamic work
with regard to post-treatment gains has not
been examined. The present study sought
to examine whether psychodynamic work
—compared to supportive work—was
related to patients’ progress in short-term
group therapy for CG, both post-treatment
and at 6-month follow-up.
Two objectives guided the present study.
The first was to examine the differential rela-
tionship between patients’ psychodynamic and
supportive work with outcome in short-term
interpretive and supportive group therapies for
CG. We hypothesized that psychodynamic
127
work—focused as it is on the development and
practice of reflective abilities—would account
for a greater degree of the outcome variance
than supportive work. The second objective
was to investigate the degree to which psycho-
dynamic and supportive work were predictive
of patients’ further progress between termina-
tion and follow-up at 6 months post-treatment.
Extending the hypothesis of our first objective,
we hypothesized that psychodynamic work
would be responsible for a greater degree of
improvement during the follow-up period than
patients’ use of therapy for problem solving and
adaptation-oriented efforts (i.e., supportive
work).
METHOD
A detailed description of the design
and methodology of the study is presented
in Piper et al., 2007. Patients at two Cana-
dian outpatient psychiatric clinics (Univer-
sity of Alberta Hospital, Edmonton,
Alberta, and Vancouver General Hospital,
Vancouver, British Columbia) who met
criteria for CG were included in the
study. The presence of CG was determined
by the patient scoring 10 or higher on one
of the following self-report measures: (1) a
set of seven pathologic grief items each
ranging from 0 (never or rarely) to 3
(very often; Prigerson et al., 1995); (2)
the Intrusion subscale of the Impact of
Events Scale (IES; Horowitz, Wilner, &
Alvarez, 1979); or (3) the Avoidance sub-
scale also from the IES. The items from the
IES range from 1 (not at all) to 4 (often).
Scores from these measures were related to
at least one significant loss that occurred
at least 3 months previously. A score of
2.0 or higher on one of the six subscales
of the Social Adjustment Scale-Self Report
(SAS-SR; Weissman & Bothwell, 1976)
was also required to meet criteria for
CG. Items from the SAS-SR are rated on
a 5-point scale, with higher scores indicat-
ing poorer social functioning. The above
criteria were used to include patients with
 128
at least moderate grief symptoms and
social role dysfunction and to rule out
immediate grief reactions. Participants
had to be older than 18 years, not engaged
in any concurrent psychosocial treatments,
and willing and capable of engaging in
therapy within a group environment.
Patients with comorbid problems that
would interfere with group participation
or require alternative treatment—such as
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
acute psychosis or mania—were excluded
from the study. Ethics approval was
obtained from the local institutional
review board, and all participants pro-
vided consent prior to the commencement
of the study.
Groups were formed to reflect varia-
tions of group composition based on
patients’ quality of object relations
(QOR). Patients’ QOR was assessed using
a 1-hour semi-structured interview that
explores lifelong relational patterns (Azim,
Piper, Segal, Nixon, & Duncan, 1991).
Patients were designated as either high-
QOR (scores of 4.2 or greater) or low-
QOR (scores of 4.1 or lower; possible
range = 1 to 9). High QOR reflects mature,
reciprocal tendencies in relationships, while
low-QOR indicates fraught relational pat-
terns characterized by inordinate depen-
dence and anxiety. Patients were then
allocated to the following groups: (a) six
homogeneous, low-QOR groups, (b) five
heterogeneous, mixed-QOR groups, and
(c) seven homogeneous, high-QOR groups.
Because of the impossibility of perfect
group assignments, a percentage score was
assigned to reflect group composition by
indicating the percentage of high-QOR
patients in the group (M = 55%;
SD = 34%; range = 0% to 100%).
Patients
A total of 135 patients with CG com-
menced short-term group therapy for CG,
forming 18 different treatment groups. Of
these, 110 patients completed treatment—
Psychodynamic Work in Group Psychotherapy
defined as attending 8 or more of the 12
sessions—including 52 patients in interpre-
tive group therapy and 58 in supportive
group therapy. The majority of patients
were White (n = 92; 84%) and female
(n = 87; 79%), with an average age of
45.2 years (SD = 11.7, range = 22–74).
Thirty-six percent (n = 40) were living
with a partner, 17% (n = 19) were sepa-
rated or divorced, 12% (n = 14) were
widowed, and 21% (n = 23) were single.
Fifty-nine percent (n = 65) were educated
beyond high school and 45% (n = 49) were
employed. The majority (n = 67; 61%)
reported receiving previous psychiatric
treatment, and most reported current use
of psychotropic medication (n = 97;
88%). Patients had experienced various
types of losses, including that of a parent
(n = 61; 56%), sibling (n = 12; 11%),
partner (n = 13; 12%), child (n = 11;
10%), friend (n = 4; 4%), grandparent
(n = 2; 2%), or other (n = 7; 6%). The
average time since the loss was 7.4 years
(SD = 10.49, range = 0.25–62.5).
All of the 110 completers were
assessed for DSM-IV diagnoses (American
Psychiatric Association, 2000) using the
Structured Clinical Interview for DSM-5
(SCID) Screen Patient Questionnaire–
Extended (First, Gibbon, Williams, Spitzer,
& Multi-Health Systems Staff, 2001) and
independently confirmed by an intake
assessor and a psychiatrist. Forty percent
(n = 44) received an Axis I diagnosis, the
most common being major depression
(n = 27; 25%), post-traumatic stress disor-
der (n = 10; 9%), and panic disorder
(n = 6; 6%). Thirty-seven percent (n = 41)
received Axis II diagnoses determined by
the computer-administered SCID II Patient
Questionnaire and the Computer-Assisted
SCID II Expert System (First, Gibbon, Spit-
zer, Williams, & Benjamin, 2000), includ-
ing avoidant (n = 21; 19%), obsessive-
compulsive (n = 13; 12%), and borderline
(n = 10; 9%) personality disorders.
 Kealy et al.
Therapists
The therapy groups were led by four
therapists: a 45-year-old male psychologist, a
50-year-old male psychologist, a 46-year-old
female social worker, and a 62-year-old female
nurse. Each had substantial experience practi-
cing group therapy (13, 18, 19, and 25 years,
respectively). The four therapists conducted 5,
4, 6, and 3 groups, respectively. Each therapist
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
conducted both types of groups and endorsed
a practice orientation that encompassed inter-
pretive and supportive interventions within a
broad psychodynamic/interpersonal theoreti-
cal tradition. Therapy sessions were recorded
for review during supervision meetings follow-
ing each session between therapists and one of
the investigators of the study (WEP). Meetings
focused on adherence to the treatment manual,
difficulties with the group or group members,
and any other issues concerning the provision
of therapy.
Therapies
The two forms of manualized short-
term group therapy were interpretive psy-
chotherapy and supportive psychotherapy,
each involving once-weekly 90-minute sessions
over a 12-week period. Nine interpretive and
nine supportive groups were provided, with an
average of six completers per group. Despite
structural similarity, the two therapy
approaches differ substantially in terms of
their overall objectives, session objectives, and
therapist techniques. A comprehensive descrip-
tion of these treatments, along with evidence
supporting their efficacy for CG may be found
in Piper et al. (2011). The following brief sum-
mary outlines the main objectives and pro-
cesses of each.
Interpretive Psychotherapy
The primary objective of CG-focused
interpretive therapy is to enhance patient
insight about repetitive conflicts (intrapsychic
and interpersonal) and trauma that are
129
associated with the loss(es) and assumed to
impede a normal mourning process. A related
objective is to help the patients develop toler-
ance for ambivalent feelings toward the peo-
ple whom they have lost. During the sessions,
the therapist attempts to create a climate of
tolerable tension wherein conflicts can be
examined using here-and-now experience.
The therapist accomplishes this by encoura-
ging patients to explore uncomfortable emo-
tions, including those evoked through the
interpersonal milieu of the group. The thera-
pist withholds immediate praise and gratifica-
tion but remains active in offering interpretive
comments regarding patients’ wishes,
defenses, and relational concerns. Moreover,
patients’ transference reactions to the thera-
pist are addressed interpretively. Interventions
in interpretive therapy are thus oriented
toward facilitating insight regarding patients’
intrapsychic experiences, interpersonal pat-
terns, and emotions surrounding their rela-
tionships to those whom they have lost. In
terms of therapeutic work, interpretive inter-
ventions are expressly aimed at fostering psy-
chodynamic work.
Supportive Psychotherapy
The main objective of supportive ther-
apy is to improve patients’ immediate adap-
tation to their life situations. With regard to
CG, supportive therapy aims to help patients
adapt to the loss, restore normal activities,
and enhance social engagement. This
approach is based on the assumption that
the provision of support and problem solving
can help patients achieve improvements in
symptoms and social functioning. The thera-
pist attempts to create a climate of gratifica-
tion wherein patients can share common
experiences and feelings and receive praise
(reinforcement) for their efforts at coping.
The therapist is thus actively supportive and
focused on patients’ relationships with per-
sons outside the group (i.e., as opposed to a
focus on intragroup interactions). The thera-
pist makes clarifying rather than interpretive
 130
comments and attempts to model adaptive
coping strategies. Interventions in supportive
therapy thus focus on directly enhancing
patients’ self-esteem and coping abilities,
rather than exploring intrapsychic experi-
ence. The therapist offers positive comments
to reinforce patients’ adaptation to loss via
emotional expression, problem solving, and
social engagement.
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
Psychodynamic and Supportive Work
and Participation Variables
The therapists provided ratings for
each patient regarding the patient’s engage-
ment in psychodynamic work, supportive
work, and participation in the group.
Therapists were trained to identify these
activities based on the definitions below.
The definitions were also provided to thera-
pists in written form at the end of each
session, along with a percentage rating
question regarding the quantity of each
patient’s psychodynamic and supportive
work during the session (e.g., “please use
the following definition to rate the patient’s
supportive work”). Therapists issued a per-
centage rating (possible scores ranging
from 0–100) based on patients’ verbal
behavior during each session, indicating
the percentage of time the patient devoted
to psychodynamic work and supportive
work, as well as the patient’s overall parti-
cipation in the group. These ratings were
averaged across all attended group sessions
in order to obtain an overall average score
for each patient.
Psychodynamic Work
Psychodynamic work was defined as an
attempt by the patient to understand a pro-
blem that involves the patient, other patients
in the group, or the group as a whole in terms
of psychodynamic factors. Psychodynamic fac-
tors include wishes, fears, and defenses, as well
as any affects, cognitions, or behaviors—
including relational patterns—that have moti-
vational properties. Psychodynamic work thus
Psychodynamic Work in Group Psychotherapy
involves the effort to understand how one’s
mind functions in terms of motivation and/or
defense. Conflict is implicitly, if not explicitly,
identified. Excluded from this definition were
mere descriptions of problems, psychodynamic
factors concerning persons external to the
group, and supportive work. Psychodynamic
work scores thus reflect each patient’s overall
proportion of session time devoted to such
reflective activity.
Supportive Work
Supportive work was defined as an
attempt by the patient to adapt to or solve
a life problem that involves the patient or
other patients in the group. These attempts
to solve problems are practical, typically
involving decisions about taking a particular
action, rather than intrapsychic in nature.
Excluded from this definition were mere
descriptions of problems, problems concern-
ing persons external to the group, and psy-
chodynamic work. In contrast to ratings of
psychodynamic work, supportive work
scores reflect the overall proportion of ses-
sion time each patient devoted to practical
problem solving and coping activities.
Participation
Participation was defined as the
patient’s amount of verbal activity in the
group, irrespective of content. Participation
was included as a variable in the study due to
the possibility that therapists’ ratings of ther-
apeutic work could be confounded by indi-
vidual patients’ level of verbal participation
in the groups.
Treatment Outcome
Treatment outcome was represented by
three outcome factors: (1) general symptoms,
(2) grief symptoms, and (3) target objective
severity/life dissatisfaction. These were derived
from a principal component analysis with
orthogonal rotation of the outcome measures
used in previous analyses of the data used in this
 Kealy et al.
study (Piper et al., 2007). Residual change
scores (pre- to post-therapy) were calculated
for each outcome variable, representing
patients’ outcome at post-therapy after account-
ing for their status at pre-therapy. For all three
factors, higher scores represent less favorable
change in outcome. The measures composing
the general symptoms factor were the Beck
Depression Inventory (Beck, Steer, & Brown,
1996); the Trait Anxiety Scale (Spielberger,
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
1983); the Global Severity Index of the Brief
Symptom Inventory (Derogatis, 1993); the
Inventory of Interpersonal Problems (Horowitz,
Rosenberg, Baer, Ureno, & Villasenor, 1988)
overall score; the overall score from the SAS-SR
(Weissman & Bothwell, 1976); and the Self-
Esteem Scale (Rosenberg, 1979). The grief
symptoms factor comprised 7 pathological
grief items (Prigerson et al., 1995); the Intrusion
and Avoidance subscales of the IES (Horowitz
et al., 1979); and the 13-item Present Feelings
Subscale of the Texas Revised Inventory of
Grief (Faschingbauer, Zisook, & DeVaul,
1987)—each of these scales was completed for
the one or two most significant death losses in
the patient’s life. The third outcome factor, per-
taining to target objectives and life dissatisfac-
tion, consisted of the Quality of Life Inventory
(Frisch, Cornell, Villanueva, & Retzlaff, 1992)
as well as individualized target objectives for-
mulated by the patient with the assistance of an
independent assessor. The patient’s average rat-
ing and the independent assessor’s average rat-
ing of severity of disturbance for the objectives
were used as outcome scores. A rater reliability
determination for the assessor’s rating, using six
raters and 16 cases, yielded an intraclass corre-
lation coefficient (2, 1) of .99, indicating very
high reliability. A detailed description of the
individual outcome variables and the principal
components analysis can be found in Piper et al.
(2007).
Follow-Up Outcome
Patients returned to the clinic 6 months
after the completion of therapy in order to
provide follow-up data. The same treatment
outcome factors were used to represent
131
follow-up outcome at 6 months post-termi-
nation. Residual change scores (post-therapy
to follow-up) were calculated for each out-
come variable, representing patients’ out-
come gains in the 6-month period following
completion of group therapy (higher scores
represent less favorable change in outcome).
Addressing Dependence
The risk of incorrectly rejecting the
null hypothesis due to dependencies in
group intervention data was mitigated by
an adjustment of individual group member
scores through a three-level growth model
shown to adequately adjust for dependence
in group therapy data (Tasca, Illing, Ogrod-
niczuk, & Joyce, 2009). This adjustment is
an example of multilevel modeling where at
level 1 outcomes are estimated as a function
in time, level 2 utilizes the parameters of
level 1 (i.e., slope and intercept) to nest
data at the individual patient’s level, and
level 3 nests individual patient’s data at the
group level. All analyses in this study were
conducted after adjusting all independent
and dependent variables for the possible
bias of dependence.
Approach to Analysis
Using the entire sample of treatment
completers (N = 110), we first conducted
preliminary analyses to determine whether
there were significant relationships between
outcome and patients’ age, gender, use of
medication, presence of DSM-IV diagnosis,
type of loss, time since loss, and number of
group sessions attended. Associations
between these variables and therapists’ rat-
ings of psychodynamic and supportive work
were also conducted. We used Pearson cor-
relations for continuous variables (age, time
since loss, sessions attended), point-biserial
correlations for dichotomous variables (gen-
der, medication use, presence of diagnosis),
and Kruskal-Wallis H test for type of loss.
Significant variables would then be entered
as covariates in regression analyses of
 132
therapeutic work and outcome. We then
used Pearson correlations to examine rela-
tionships among psychodynamic work, sup-
portive work, and participation, followed by
t-tests to examine whether these variables
differed between interpretive and supportive
group therapies. Next, with regard to our
first objective, we conducted hierarchical
regression analyses using each of the three
post-treatment outcome factors as the depen-
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
dent variable. Predictor variables were
entered in separate steps due to our hypoth-
esis that psychodynamic work would emerge
as a significant predictor of outcome while
accounting for the effects of participation
and supportive work. Given our previous
finding of significance regarding group com-
position (Piper et al., 2007), the QOR-based
group composition score was entered in the
first step in order to control for the effects of
group composition in our analysis of thera-
peutic work. Thus, for each of these regres-
sion analyses, the group composition score
was entered in the first step, patients’ parti-
cipation was entered in the second step, and
supportive work and psychodynamic work
were entered in the third and fourth steps,
respectively. With regard to our second
objective—examining gains made during the
follow-up period—we were limited to ana-
lyzing data from treatment completers who
provided follow-up information (N = 84).
Potential differences between those who did
and did not provide follow-up data were
examined using t-tests for continuous vari-
ables (e.g., outcome and work variables)
and chi-square tests for categorical variables
(e.g., demographic information). We then
employed hierarchical regression analyses
using the follow-up outcome factors as
dependent variables. To control for the
effects of pre-post symptom change, pre-
post outcome was entered in the first step
along with the group composition percentage
score. Participation was then entered in the
second step, followed by supportive work in
the third step and psychodynamic work in
the fourth step.
Psychodynamic Work in Group Psychotherapy
RESULTS
Preliminary Assessment of Potential
Confounding Variables
No significant associations were
found between treatment outcome and
patients’ age, gender, use of medications,
presence of DSM-IV diagnosis, type of
loss, or time since loss. As well, no signifi-
cant associations were found between these
variables and psychodynamic and suppor-
tive work. The number of group sessions
attended was also not significantly asso-
ciated with treatment outcome or with
therapeutic work variables.
Preliminary Assessment of
Psychodynamic and Supportive Work
Mean ratings of patients’ therapeutic
work were 22.03 (SD = 8.05) for psycho-
dynamic work and 34.36 (SD = 11.2) for
supportive work. The mean participation
rating was 17.84 (SD = 4.31). No signifi-
cant differences were found between inter-
pretive and supportive therapies on any of
these variables. There was also no signifi-
cant correlation found between psychody-
namic and supportive work, confirming
therapists’ attention to the exclusive con-
tent of these ratings. Patients’ participation,
however, was positively associated with
both supportive work, r = .50, p < .001,
and psychodynamic work, r = .52,
p < .001.
Relationship Between Therapeutic
Work Variables and Pre-Post
Treatment Outcome
Table 1 provides information regard-
ing the hierarchical regression models for
each of the three outcome factors from
pre-treatment to post-treatment. In the
first regression analysis using post-treat-
ment change in general symptoms (outcome
factor 1) as the dependent variable, none of
 Kealy et al. 133

the independent variables (participation, were conducted to examine the relationship

supportive work, psychodynamic work)
were found to be significant. When change
in grief symptoms (outcome factor 2) was
entered as the dependent variable, the over-
all regression model failed to achieve omni-
bus significance. Psychodynamic work,
however, contributed significantly to
change in R2, indicating a significant asso-
ciation with post-treatment changes in grief
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
between therapeutic work and gains made
from post-treatment to 6-month follow-up
on each of the three outcome factors (see
Table 2). Each of these analyses included
pre-post change for the respective outcome
domain as a predictor variable in the first
step of the regression model, as well as the
percentage score reflecting QOR-based
group composition. With regard to general

symptoms, β = −.25, t = −2.16, p = .03. No symptoms as the dependent variable, there

predictor variables emerged as significant in
the regression model using change in target
objective severity and life dissatisfaction
(outcome factor 3) as the dependent vari-
able. Thus, the only significant relationship
found between ratings of patients’ thera-
peutic work and post-treatment outcome
was an association between psychodynamic
work and reduction of grief symptoms.
Relationship Between Therapeutic
Work Variables and Post-Treatment
to Follow-Up Outcome
No significant differences were found
between treatment completers who provided
follow-up data and those who did not with
regard to demographic characteristics, treat-
ment outcome, or work variables. Using the
sample of treatment completers who pro-
vided follow-up data, regression analyses
was again no significant effect of patients’
participation or supportive work. However,
psychodynamic work was significantly asso-
ciated with general symptom change at 6-
month follow-up, β = −.33, t = −2.68,
p = .009. When follow-up change in grief
symptoms was examined as the dependent
variable, psychodynamic work again
emerged as significant, β = −.30, t = −2.34,
p = .02. Supportive work was not significant
and participation was inversely associated
with follow-up outcome, β = .30, t = 2.14,
p = .04, likely due to the strength of the
psychodynamic work–grief outcome rela-
tionship. The final regression analysis, using
the follow-up residual gain score for target
objective severity/life dissatisfaction as the
dependent variable, did not achieve omnibus
significance for the regression model. How-
ever, psychodynamic work contributed sig-
nificantly to change in R2, indicating a

TABLE 1. Hierarchical Regression Analysis Predicting Change in Symptoms From Pre-Treatment to Post-Treatment
Among Treatment Completers (N = 110)
General Symptoms (OF-1) Grief Symptoms (OF-2) Life Dissatisfaction (OF-3)

2 2
F ΔR ΔF F ΔR ΔF F ΔR2 ΔF

Model 1a 0.01 0.00 0.01 0.01 0.00 0.01 0.00 0.00 0.00
Model 2b 0.02 0.00 0.02 2.14 0.04 4.26* 0.19 0.00 0.38
Model 3c 0.09 0.00 0.24 1.41 0.00 0.00 0.35 0.01 0.66
Model 4d 0.35 0.01 1.13 2.27 0.04 4.68* 0.44 0.01 0.72

Note. OF-1 = Outcome Factor 1; OF-2 = Outcome Factor 2; OF-3 = Outcome Factor 3; QOR = Quality of Object Relations.
*p < .05.
a
Predictor variable: QOR composition score.
b
Predictor variables: QOR composition score, Participation.
c
Predictor variables: QOR composition score, Participation, Supportive Work.
d
Predictor variables: QOR composition score, Participation, Supportive Work, Psychodynamic Work.
 134 Psychodynamic Work in Group Psychotherapy

TABLE 2. Hierarchical Regression Analysis Predicting Change in Symptoms From Post-Treatment to 6-Month
Follow-Up Among Treatment Completers (N = 84)
General Symptoms (OF-1) Grief Symptoms (OF-2) Life Dissatisfaction (OF-3)

F ΔR2 ΔF F ΔR2 ΔF F ΔR2 ΔF

a
Model 1 0.94 0.02 0.94 0.25 0.01 0.25 0.51 0.01 0.51
Model 2b 1.06 0.02 1.30 1.67 0.05 4.51* 0.81 0.02 1.39
Model 3c 1.51 0.03 2.78 1.89 0.03 2.45 0.65 0.00 0.19
Model 4d 2.73* 0.08 7.16** 2.69* 0.06 5.47* 1.76 0.07 6.08*

Note. OF-1 = Outcome Factor 1; OF-2 = Outcome Factor 2; OF-3 = Outcome Factor 3; QOR = Quality of Object Relations.
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017

*p < .05. **p < .01.

a
Predictor variables: QOR composition score, pre-post outcome.
b
Predictor variables: QOR composition score, pre-post outcome, Participation.
c
Predictor variables: QOR composition score, pre-post outcome, Participation, Supportive Work.
d
Predictor variables: QOR composition score, pre-post outcome, Participation, Supportive Work, Psychodynamic Work.

relationship between psychodynamic work
and changes in life dissatisfaction during the
6-month follow-up period, β = −.31,
t = −2.47, p = .02.
DISCUSSION
This study provides evidence for the
relevance of the type of patients’ therapeutic
work in psychotherapy sessions. For patients
participating in two forms of short-term
group therapy for CG, engagement in psy-
chodynamic work was associated with ther-
apeutic progress. We defined psychodynamic
work as the patient’s effort to understand
and reflect upon problems and relational pat-
terns in terms of inner motivation and/or
conflict. Psychodynamic work was asso-
ciated with pre-post improvement in grief
symptoms, but not with pre-post improve-
ment in general symptoms or life satisfaction.
By contrast, supportive work—defined as the
patient’s attempt to adapt to or solve pro-
blems in terms of practical activities—was
not associated with any pre-post treatment
change. Supportive work was also not asso-
ciated with symptom change during the 6-
month follow-up period. Psychodynamic
work, however, predicted further improve-
ment in grief symptoms during the 6-month
follow-up period, even while controlling for
the effects of participation (the patient’s
overall amount of verbal activity) and pre-
post change in grief symptoms. Furthermore,
psychodynamic work was related to
improvement in general symptoms and life
dissatisfaction during the 6 months following
treatment—also beyond the effects of pre-
post change and participation. Previously
we found group composition, in terms of
the percentage of high-QOR group members,
to influence outcome in this trial of therapy
for CG. The present study, however, found
psychodynamic work to be significantly
related to symptom relief and further post-
treatment improvement beyond the effects of
group composition. These results occurred
across both interpretive and supportive treat-
ments, indicating the importance of patients’
psychodynamic work over the type of inter-
vention promoted by group therapists.
Although we had hypothesized a link
between psychodynamic work and improve-
ment in each of the three outcome factors,
only pre-post changes in grief symptoms
were significantly related to psychodynamic
work. This likely reflects the clinical focus on
CG in the therapies we studied. For patients
struggling with CG, the attachment and iden-
tity-related conflicts that accompany loss
through death are primary targets for self-
reflection. Group therapists also encouraged
patients to discuss experiences pertaining to
their loss(es), giving stronger emphasis to
bereavement-related issues than to other
 Kealy et al.
concerns. This emphasis likely facilitated
relief from grief symptoms ahead of other
forms of distress. Further post-treatment
improvement in grief symptoms may be
related to patients’ application of the skills
developed in therapy in the months following
termination. Patients may well have lever-
aged the momentum gained in group therapy
to further reflect on bereavement-related
yearnings and conflicts after the group had
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
ended.
In-session psychodynamic work was
also predictive of broader improvement in
the period between termination and the 6-
month follow-up assessment. Having
achieved some resolution of bereavement-
fueled distress, patients may have been better
able to extend their reflective abilities to
other problem areas such as mood distur-
bance, self-esteem, and social/interpersonal
dysfunction. Although patients were free to
discuss such difficulties in both forms of
group therapies for CG, addressing these
issues was not the primary objective of the
groups. Insight and action regarding broader
symptom and functional problems were pos-
sibly developed later on, after patients had
completed therapy and benefitted from
reflection regarding loss-related concerns.
It is interesting to speculate on possible
mechanisms by which in-session psychody-
namic work contributed to post-therapy
improvement. One possibility is that such
work involved patients’ search for under-
standing regarding aspects of their psycholo-
gical functioning that had inhibited a normal
mourning process (Shaw, 2015). Such reflec-
tion—perhaps concerning frustrated yearn-
ings or ambivalent feelings toward a lost
loved one—may have been developed in par-
tial form at the point of termination. Further
reflection may have subsequently been
applied to nascent insights during the fol-
low-up period, yielding additional and per-
haps deeper self-understanding. In this way,
psychodynamic work may be regarded as
facilitating an acquired skill. Skills require
practice in order to become integrated into
an individual’s repertoire of abilities. As with
135
any skill, the ability to search for psychody-
namic understanding produces further
returns as one becomes more proficient.
Future research exploring patients’ post-ther-
apy work could examine how these reflective
abilities are applied following termination.
Interestingly, psychodynamic work
was equally prominent in supportive groups
as in interpretive therapy. Although thera-
pists had been carefully supervised to avoid
interpretive interventions in supportive ther-
apy groups, their own ratings indicated that
members of these groups were nevertheless
working to understand intrapsychic conflict
to a similar extent as patients who received
interpretive interventions. Moreover, such
effort among the members of supportive
groups was associated with outcome. This
suggests that, for patients with CG, interpre-
tive interventions—long thought to be cru-
cial for progress in psychodynamic
psychotherapy—may not be necessary for
the development of reflective abilities that
promote sustained outcome benefits. This
accords with the view of the patient as an
“active self-healer” (Bohart, 2007, p. 257)
whose implicit plan strongly determines ther-
apeutic progress (Weiss, 1993). There may,
however, be inherent qualities of group treat-
ments that nurture productive therapeutic
work. Group psychotherapy involves multi-
ple therapeutic factors involving group mem-
bers’ sharing of experiences, mutual
acceptance, and exchange of help (Rice,
2015; Yalom & Leszcz, 2005). This commu-
nal milieu—reinforced by the therapist’s gui-
dance—is regarded as a sufficient “secure
base” (Marmarosh, Markin, & Spiegel,
2013) from which to explore. From the per-
spective of mentalization theory (e.g. Fonagy
& Allison, 2014), the relational security and
reciprocity of the group may contribute to
the restoration of the patient’s capacity to
trust in the utilization of other minds for
self-reflective purposes. Indeed, group ther-
apy affords a unique opportunity for inter-
personal learning (Yalom & Leszcz, 2005).
Patients learn about their own difficulties by
seeing similar problems worked on by other
 136
group members, and through co-members’
questions, feedback, and suggestions. More-
over, a patient may become intrigued by the
reflective efforts pursued by others in the
group and “catch on” to a method of
approaching their difficulties from the per-
spective of intrapsychic dynamics.
The present study has a number of lim-
itations that must be noted. First of all, the
homogeneity of the sample—adults with CG
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
—may limit generalizability across a broader
array of patients and disorders. Secondly, the
degree to which our findings might generalize
to other therapies, such as individual therapy,
psychoeducational groups, or longer-term
group therapy, is unknown. Another limita-
tion pertains to the therapeutic work variables
assessed in our study. Although psychody-
namic work refers to insight-seeking, reflective
activity, patients’ actual acquisition of insight
was not assessed. It is thus not possible to
conclude that psychodynamic work led to spe-
cific insights, which in turn influenced symp-
tom improvement. Future research could
explore this question with a study design inclu-
sive of insight measurement. Patients’ in-ses-
sion work in our study was also assessed
using a simple percentage score rating, with
some risk of bias due to having been provided
by the group therapists. The lack of rater relia-
bility information regarding these therapeutic
work variables is also a limitation. Yet con-
cerns about rater bias are mitigated by the fact
that outcome ratings were made by patient
self-report. Ideally, assessment of patients’ in-
session activities would be done using a more
sophisticated instrument and from multiple
perspectives: therapists, patients themselves,
and blinded external raters.
Despite the aforementioned limitations,
this study provides evidence for the connection
between patients’ insight-seeking, reflective
work and their benefit from short-term group
psychotherapy. In particular, the study demon-
strates that efforts to understand psychody-
namic issues may contribute to immediate
post-therapy improvement in focal symptoms,
as well as further changes across other symptom
domains in the months following termination.
Psychodynamic Work in Group Psychotherapy
This finding cut across two approaches to short-
term group therapy for CG, suggesting that
patients’ reflective, psychodynamic work may
be a general curative mechanism of group treat-
ments.
ORCID
Carlos A. Sierra-Hernandez http://orcid.
org/0000-0002-9530-2049
REFERENCES
Allen, J. G. (2013). Restoring mentalizing in
attachment relationships: Treating trauma with
plain old therapy. Washington, DC: American
Psychiatric Association.
American Psychiatric Association. (2000). Diag-
nostic and statistical manual of mental disorders
(4th ed., text revision). Washington, DC: Author.
Azim, H. F., Piper, W. E., Segal, P. M., Nixon, G.
W., & Duncan, S. C. (1991). The quality of object
relations scale. Bulletin of the Menninger Clinic,
55, 323–343.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996).
Beck Depression Inventory manual (2nd ed.). San
Antonio, TX: Psychological Corporation.
Bohart, A. C. (2007). Insight and the active client.
In L. G. Castonguay, & C. E. Hill (eds.), Insight in
psychotherapy (pp. 257–277). Washington, DC:
American Psychological Association.
Bowlby, J. (1980). Attachment and loss, volume
III. Loss: Sadness and depression. London, UK:
Hogarth Press.
Connolly Gibbons, M. B., Crits-Christoph, P.,
Barber, J. P., & Schamberger, M. (2007). Insight
in psychotherapy: A review of empirical literature.
In L. G. Castonguay, & C. E. Hill (eds.), Insight in
psychotherapy (pp. 143–165). Washington, DC:
American Psychological Association.
Derogatis, L. R. (1993). Brief Symptom Inventory:
Administration, scoring, and procedures manual (3rd
ed.). Minneapolis, MN: National Computer Systems.
Einstein, A. (1954/1982). Ideas and Opinions.
New York, NY: Three Rivers Press.
 Kealy et al.
Faschingbauer, T. R., Zisook, S., & DeVaul, R. A.
(1987). The Texas revised inventory of grief. In S.
Zisook (Ed.), Biopsychosocial aspects of bereavement
(pp. 109–123). Washington, DC: American Psychia-
tric Press.
First, M. B., Gibbon, M., Spitzer, R. L., Williams, J.
B. W., & Benjamin, L. S. & Multi-Health Systems
Staff. (2000). Structured clinical interview for DSM-
IV AXIS II personality disorders: SCID II patient
questionnaire & computer-assisted SCID II expert
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
system. Toronto, Canada: Multi-Health Systems.
First, M. B., Gibbon, M., Williams, J. B. W., &
Spitzer, R. L. &Multi-Health Systems Staff. (2001).
SCID SCREEN patient questionnaire (SSPQ) &
SCID SCREEN patient questionnaire extended
(SSPQ-X): computer program for windows software
manual. Toronto, Canada: Multi-Health Systems.
Fonagy, P., & Allison, E. (2014). The role of
mentalizing and epistemic trust in the therapeutic
relationship. Psychotherapy, 51, 372–380.
doi:10.1037/a0036505
Frisch, M. B., Cornell, J., Villanueva, M., &
Retzlaff, P. J. (1992). Clinical validation of the
quality of life inventory. A measure of life satisfac-
tion for use in treatment planning and outcome
assessment. Psychological Assessment, 4, 92–101.
doi:10.1037/1040-3590.4.1.92
Gabbard, G. O. (2005). Major modalities: Psycho-
analytic / psychodynamic. In G. O. Gabbard, J. S.
Beck, & J. Holmes (Eds.) Oxford Textbook of Psy-
chotherapy. Oxford, UK: Oxford University Press.
Horowitz, L., Rosenberg, S. E., Baer, B. A., Ureno,
G., & Villasenor, V. S. (1988). Inventory of inter-
personal problems: Psychometric properties and
clinical applications. Journal of Consulting and
Clinical Psychology, 56, 885–892. doi:10.1037/
0022-006X.56.6.885
Horowitz, M. J., Wilner, N., & Alvarez, W.
(1979). Impact of event scale: A measure of sub-
jective stress. Psychosomatic Medicine, 41, 209–
218. doi:10.1097/00006842-197905000-00004
Jones, E. E. (2000). Therapeutic action: A
guide to psychoanalytic therapy. Northvale,
NJ: Aronson.
Lacewing, M. (2014). Psychodynamic psychother-
apy, insight, and therapeutic action. Clinical Psy-
chology: Science and Practice, 21, 154–171.
137
Lobb, E. A., Kristjanson, L. J., Aoun, S. M., Mon-
terosso, L., Halkett, G. K., & Davies, A. (2010).
Predictors of complicated grief: A systematic
review of empirical studies. Death Studies, 34,
673–698. doi:10.1080/07481187.2010.496686
Maccallum, F., & Bryant, R. A. (2013). A cogni-
tive attachment model of prolonged grief: Integrat-
ing attachments, memory, and identity. Clinical
Psychology Review, 33, 713–727. doi:10.1016/j.
cpr.2013.05.001
Marmarosh, C. L., Markin, R. D., & Spiegel, E.
B. (2013). Attachment in group psychotherapy.
Washington, DC: American Psychological Asso-
ciation.
Messer, S. B., & McWilliams, N. (2007).
Insight in psychodynamic therapy: Theory and
assessment. In L. G. Castonguay, & C. E. Hill
(eds.), Insight in psychotherapy (pp. 10–29).
Washington, DC: American Psychological
Association.
Ogden, T. H., & Gabbard, G. O. (2010). The lure
of the symptom in psychoanalytic treatment. Jour-
nal of the American Psychoanalytic Association,
58, 533–544. doi:10.1177/0003065110376080
Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S., &
Weideman, R. (2009). Follow-up outcome in
short-term group therapy for complicated grief.
Group Dynamics: Theory, Research, and Practice,
13, 46–58. doi:10.1037/a0013830
Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S., &
Weideman, R. (2011). Short-term group therapies
for complicated grief: Two research-based models.
Washington, DC: American Psychological Asso-
ciation.
Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S.,
Weideman, R., & Rosie, J. S. (2002). Relation-
ships among affect, work, and outcome in group
therapy for patients with complicated grief. Amer-
ican Journal of Psychotherapy, 56, 347–361.
Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S.,
Weideman, R., & Rosie, J. S. (2007). Group com-
position and group therapy for complicated grief.
Journal of Consulting and Clinical Psychology,
75, 116–125. doi:10.1037/0022-006X.75.1.116
Prigerson, H. G., Frank, E., Kasl, S. V., Reynolds,
C. F., Anderson, B., Zubenko, G. S., … Kupfer, D.
J. (1995). Complicated grief and bereavement-
 138
related depression as distinct disorders: Prelimin-
ary empirical validation in elderly bereaved
spouses. American Journal of Psychiatry, 152,
22–30. doi:10.1176/ajp.152.1.22
Rice, A. (2015). Common therapeutic factors in
bereavement groups. Death Studies, 39, 165–172.
doi:10.1080/07481187.2014.946627
Rosenberg, M. (1979). Conceiving the self. New
York, NY: Basic Books.
Downloaded by [The University of British Columbia Library] at 13:37 02 August 2017
Shaw, J. A. (2015). The capacity to mourn:
Thoughts for a developmental line. Psychiatry,
78, 55–64. doi:10.1080/00332747.2015.1015876
Shear, K., & Shair, H. (2005). Attachment, loss,
and complicated grief. Developmental Psychobiol-
ogy, 47, 253–267. doi:10.1002/dev.v47:3
Spielberger, C. D. (1983). Manual for the state-
trait anxiety inventory. Palo Alto, CA: Consulting
Psychologists Press.
View publication stats
Psychodynamic Work in Group Psychotherapy
Tasca, G. A., Illing, V., Ogrodniczuk, J. S., & Joyce,
A. S. (2009). Assessing and adjusting for dependent
observations in group treatment research using mul-
tilevel models. Group Dynamics: Theory, Research,
and Practice, 13, 151–162. doi:10.1037/a0014837
Weiss, J. (1993). How psychotherapy works: Pro-
cess and technique. New York, NY: Guilford Press.
Weissman, M. M., & Bothwell, S. (1976). Assess-
ment of social adjustment by patient self-report.
Archives of General Psychiatry, 33, 1111–1115.
doi:10.1001/archpsyc.1976.01770090101010
Wittouck, C., Van Autreve, S., De Jaegere, E.,
Portzky, G., & van Heeringen, K. (2011). The pre-
vention and treatment of complicated grief: A meta-
analysis. Clinical Psychology Review, 31, 69–78.
doi:10.1016/j.cpr.2010.09.005
Yalom, I. D., & Leszcz, M. (2005). The theory and
practice of group psychotherapy (5th ed.). New
York, NY: Basic Books.