Content uploaded by Stacey McElroy-Heltzel
Author content
All content in this area was uploaded by Stacey McElroy-Heltzel on Jan 13, 2020
Content may be subject to copyright.
DOI: 10.1002/jclp.22681
RESEARCH ARTICLE
Integrating clients’religion and spirituality
within psychotherapy: A comprehensive
meta‐analysis
Laura E. Captari
|
Joshua N. Hook
|
William Hoyt
|
Don E. Davis
|
Stacey E. McElroy-Heltzel
|
Everett L. Worthington Jr.
Department of Psychology, University of
North Texas, Denton, Texas
Correspondence
Laura E. Captari, Department of Psychology,
University of North Texas, 1155 Union Circle
#311280, Denton, TX 76203‐5017.
Email: lauracaptari@my.unt.edu
Abstract
Some religious or spiritual (R/S) clients seek psychotherapy that
integrates R/S values, while others may be reticent to disclose
R/S‐related aspects of struggles in a presumably secular setting.
We meta‐analyzed 97 outcome studies (N= 7,181) examining
the efficacy of tailoring treatment to patients’R/S beliefs
and values. We compared the effectiveness of R/S‐tailored
psychotherapy with no‐treatment controls, alternate secular
treatments, and additive secular treatments. R/S‐adapted
psychotherapy resulted in greater improvement in clients’
psychological (g= 0.74, p< 0.000) and spiritual (g= 0.74,
p< 0.000) functioning compared with no treatment and non
R/S psychotherapies (psychological: g= 0.33, p< 0.001; spiritual:
g= 0.43, p< 0.001). In more rigorous additive studies, R/S‐
accommodated psychotherapies were equally effective to
standard approaches in reducing psychological distress
(g= 0.13, p= 0.258), but resulted in greater spiritual well‐being
(g= 0.34, p< 0.000). We feature several clinical examples and
conclude with evidence‐based therapeutic practices.
KEYWORDS
adaptations, meta‐analysis, psychotherapy, psychotherapy relationships,
religion, spirituality
J. Clin. Psychol. 2018;74:1938–1951.wileyonlinelibrary.com/journal/jclp1938
|
© 2018 Wiley Periodicals, Inc.
This article is adapted, by special permission of Oxford University Press, by the same authors in J. C. Norcross & B. E. Wampold (Eds.) (2018),
Psychotherapy relationships that work. volume 2 (3
rd
ed.). New York: Oxford University Press. The Interdivisional APA Task Force on Evidence‐Based
Psychotherapy Relationships and Responsiveness was cosponsored by the APA divisions of Psychotherapy (29) and Counseling Psychology (17).
1
|
INTRODUCTION
Religious or spiritual (R/S) beliefs and practices are woven into the fabric of many people’s lives. About 68% of the
world’s population view religion as an important aspect of their day‐to‐day experience (Diener, Tay, & Myers,
2011). In the United States, 89% believe in “God or a universal spirit,”75% describe religion as either “somewhat”
or “very”important to them, 80% pray regularly, and 50% belong to a local house of worship (Pew Research, 2016).
An extensive body of research has documented a positive relation between R/S and physical and mental health
(Koenig, King, & Carson, 2012). Specifically, R/S can foster increased social belonging, connection, and support; adjustment
to stressors through meaning‐making, coping, and resilience; and grounding of one’s identity through salient beliefs and
values (e.g., Paloutzian & Park, 2014). However, for some individuals, R/S can be a source of struggle and confusion, or
serve as a defense against unresolved psychological conflicts (e.g., Exline & Rose, 2014). Incorporating clients’R/S identity
within psychotherapy has the potential to impact both change processes and treatment outcomes. “When people walk
into the therapist’soffice,”Pargament (2011), writes, “they do not leave their spirituality behind in the waiting room (p. 4).
They bring their spiritual beliefs, practices, experiences, values, relationships, and struggles along with them.”
Many patients who experience R/S as a salient part of their identity hope that their therapist will integrate
these beliefs and values within psychotherapy (Vieten et al., 2013). Although some individuals forthrightly state
this, others do not. Such clients may be reticent to disclose R/S‐related aspects of their struggles in a setting they
presume to be limited to secular considerations, potentially hampering therapeutic outcomes. A number of patients
stand to benefit through secular psychotherapy, but may experience additional gains if treatment were
contextualized within their R/S values. For others, R/S struggles play a significant role in their psychological and
emotional distress, making it vital to address such topics in therapy.
In addition to patients’unique needs, R/S adaptations in psychotherapy may be complicated by several clinician
characteristics. Notably, psychotherapists as a whole are less likely to identify as R/S compared with the general
population. In one survey, 35% of psychologists—in contrast with 75% of the public—described their approach to life as
significantly influenced by R/S (Delaney, Miller, & Bisonó, 2007). Furthermore, relatively few psychotherapists receive
explicit training and supervision in how to ethically and sensitively address patients’R/S beliefs in assessment and
treatment (Schafer, Handal, Brawer, & Ubinger, 2011). Thus, although R/S is increasingly recognized as an important
aspect of multicultural competency, psychotherapists may be unsure how and in what way to best facilitate integration
of such concerns, which can result in “spiritually avoidant care”(Saunders, Miller, & Bright, 2010, p. 355).
In this study, we examine the effectiveness of R/S accommodation in psychotherapy. We begin by defining R/S
and discussing common measures of these constructs. We also offer clinical examples illustrating treatment
accommodations in psychotherapy, considering the patient’s R/S beliefs and worldview. Next, we present the
results of a meta‐analysis examining the efficacy of R/S‐adapted psychotherapies and explore patient, study, and
treatment characteristics that may moderate therapeutic effects. We also consider patient contributions, research
limitations, and diversity considerations. We conclude with therapeutic practices based on the research evidence.
2
|
DEFINITIONS AND MEASURES
Historically, the terms religion and spirituality were closely linked, and at times, used interchangeably. Religion can be
defined as adherence to common beliefs, behaviors, and practices associated with a particular faith tradition and
community, which provides guidance and oversight (Hill et al., 2000). In contrast, spirituality is a broader concept
describing the subjective, embodied, emotional experience of closeness and connection with what is viewed as
sacred or transcendent. This often constitutes either (a) a divine being or object or (b) a sense of ultimate reality or
truth, and can be understood within the framework of implicit relational knowledge.
Spirituality has been further defined within four main categories, based on the sacred/transcendent object
(Davis et al., 2015). First, religious spirituality involves a felt sense of closeness and connection with a higher power
CAPTARI ET AL.
|
1939
or worship tradition as described by a specific religion (e.g., Christianity, Islam, Judaism, and Buddhism). Second,
humanistic spirituality involves a sense of closeness and connection with other human beings, including feeling
compassion, lovingkindness, or altruism. Third, nature spirituality involves a sense of closeness and connection with
the environment or aspects of nature, such being awestruck at a beautiful sunset or the grandeur of a mountain
landscape. Fourth, cosmos spirituality involves a sense of closeness and connection with the universe, such as
contemplating the vastness of outer space or one’s sense of being within the cosmos.
Adapting psychotherapy to a patient’s R/S framework may influence the treatment in several ways:
conceptualization, treatment goals, intervention, and interpersonal process. First, understanding the patient’s
R/S aids in conceptualizing causes of psychological distress and identifying key risk and protective factors. This
might include exploring the role of R/S in the patient’s history, identity, and current functioning, as well as any
areas of difficulty (e.g., spiritual struggles, spiritual bypass). Second, R/S‐tailored psychotherapy provides a
broader context within which to understand the patient’s reasons for attending psychotherapy. In addition to
symptom reduction and self‐development, R/S clients may identify additional goals, like developing a closer
relationship with Jesus Christ, faithfulness to Allah, following the teachings of Buddha, or greater connection
with the transcendent. Psychotherapeutic outcomes such as increased spiritual well‐being and positive religious
coping may prove important considerations when patients’goals extend beyond the psychological to the R/S
aspects of their lives.
Third, the patient’s R/S can be integrated within traditional interventions (e.g., behavior activation, challenging
negative thoughts, distress tolerance, and mindfulness skills) aimed at reaching treatment goals. R/S interventions
might incorporate methods consistent with a client’s R/S culture (e.g., prayer, meditation, religious imagery, sacred
scriptures, religious rituals, or services) that may be positive coping resources, or could explore and address
underlying R/S issues (e.g., anger at God, existential doubts, or spiritual abuse) contributing to psychological
distress. Finally, R/S integration may occur implicitly in the “being with”process of psychotherapy, as a patient
experiences and internalizes the consistent, attuned, and caring presence of the psychotherapist. Helping R/S
patients reflect on the ways in which the psychotherapy relationship mirrors or challenges their perceived
relationship with the sacred may positively impact how they relate to God or their higher power.
For the purposes of the present meta‐analysis, we examined treatment outcome via two patient dimensions:
Psychological outcomes and spiritual outcomes. Nearly every treatment study in our meta‐analysis included at least
one psychological outcome measure. For example, studies evaluating R/S‐adapted psychotherapy for depression
often administered the Beck Depression Inventory‐II (BDI‐II; Beck, Steer, & Brown, 1996), whereas those
examining R/S treatments for anxiety often used the Hamilton Anxiety Rating Scale (HAM‐A; Hamilton, 1959).
Patients’R/S was typically measured as a single demographic question, but some studies also assessed R/S
outcomes of accommodated psychotherapy. The Multidimensional Measure Of Religiousness/Spirituality (MMRS;
88 items; Fetzer Institute, 1999) was often used, assessing 12 domains, including daily spiritual experiences, private
religious practices, and organizational religiousness. Other researchers used the Spiritual Well‐Being Questionnaire
(SWBQ; 20 items; Gomez & Fisher, 2003), which measures personal, communal, environmental, and transcendental
well‐being, and some studies used the Brief Religious Coping Scale (BRCOPE; 14 items; Pargament, Feuille, &
Burdzy, 2011), which assesses aspects of negative religious coping (e.g., spiritual discontent, reappraisals of God’s
punishment) and positive religious coping (e.g., spiritual surrender, reappraisals of God’s protection and care).
3
|
CLINICAL EXAMPLES
R/S adaptations of psychotherapy are as unique as each patient who walks through the door. The following case examples
illustrate several of the complex ways that R/S can intersect with other cultural identities and influence the course of
treatment. We focus especially on the systematic ways that researchers have formally integrated R/S within treatments.
1940
|
CAPTARI ET AL.
3.1
|
Case 1: Religiously integrated CBT for depression
Religiously integrated cognitive behavior therapy (RCBT) is founded on the cognitive model while contextualizing
interventions within the patient’s religious framework. The psychotherapist makes “explicit use of the client’s own
religious tradition as a major foundation to identify and replace unhelpful thoughts and behaviors to reduce
depressive symptoms”(Pearce et al., 2015, p. 58). Some of the major tools of RCBT include (a) renewing the mind
by replacing negative self‐talk with sacred scriptures; (b) meditating on sacred writings and engaging in
contemplative prayer; (c) considering religious beliefs and resources; (d) cultivating forgiveness, hope, gratitude,
and generosity through daily religious practices; (e) identifying and making use of R/S resources in line with one’s
faith tradition; and (f) altruistic involvement in one’s religious community. Treatment manuals and patient
workbooks have been developed to guide RCBT with Christian, Muslim, Jewish, Buddhist, and Hindu patients.
“Katina”presented to psychotherapy with depression. She identified as a 42‐year‐old African American,
cisgender woman, and her most salient identity was her Christian faith. Katina was somewhat reticent about how
traditional psychotherapy could help her. Picking up on this, her therapist explored Katina’s religious history and
the role that R/S played in her life. They discussed ways she desired to incorporate this into treatment. Katina
reported growing up in a strict religious family, where she felt she could never be good enough. She recounted
episodes of physical and emotional abuse by her father throughout childhood, and the untimely death of her
mother from cancer when she was a teenager. Katina identified her Christian faith as a source of coping and hope in
the midst of this loss, but as psychotherapy progressed, she also became more aware of her anger toward God
because she believed that God took her mother away or (at best) allowed bad things to happen.
Katina identified her most problematic core beliefs as (a) I am worthless and no one can ever love and accept
me as I am, and (b) bad things keep happening to me and God does not stop them, so I cannot fully trust him. As
Katina and her therapist modified these negative core beliefs in the context of her faith, she found comfort in
meditating on Scripture passages about God’s presence and unconditional love. Katina identified that listening to
Christian music, journaling, taking reflective prayer walks, and attending a small group at her church were all ways
she could incorporate daily spiritual practices, rather than being controlled by negative self‐talk.
3.2
|
Case 2: Spiritual self‐schema therapy for addiction
Spiritual self‐schema therapy integrates CBT with Buddhist psychological principles, guided by a 12‐session
treatment manual (Avants & Margolin, 2004). The goal is to decrease impulsive behavior through modifying a
client’s habitual self‐schema, described as a “highly automatized system of knowledge or beliefs about one’s
intentions or capacities, stored in long‐term memory, that mediates the environment and interpersonal behavior”
(Margolin et al., 2007, p. 982). When a habitual self‐schema is activated, negative beliefs about the self can
motivate self‐destructive behaviors, such as drug use. This psychotherapy attempts to facilitate a shift from an
addict self‐schema to a spiritual self‐schema by fostering mindfulness, self‐compassion, and commitment to do no
harm to self or others. Spiritual self‐affirmations, meditation, prayer, and self‐schema check‐ins are used to foster
spiritual awareness. Each session also focuses on the development of a spiritual quality, including generosity,
lovingkindness, and truth.
“Tom”sought treatment at an inpatient drug rehabilitation facility at the urging of his parents. He identified as a
28‐year‐old White, cisgender man for whom spirituality was a salient personal identity. He grew up in a family with
a Jewish cultural heritage but had explored other perspectives and worldviews. Tom recently lost his job after
testing positive for cocaine on a drug screening. He had begun experimenting with alcohol and drugs in middle
school, and soon began dealing to other athletes on his sports team. Tom had hopes of playing football in college,
but received drug charges as a high school senior that forced him to forfeit his scholarship. Though Tom tried to
“get clean”and start fresh, he did not maintain sobriety. He was shaken by losing several friends to drug overdoses
and expressed a desire to change, but was resistant to acknowledging the full impact of his addiction.
CAPTARI ET AL.
|
1941
During psychotherapy, Tom was taught about the wandering nature of the mind, what his psychotherapist
called “monkey mind,”and how this contributed to his addict self‐schema. If Tom did not work to control his mind,
he usually thought about using drugs. Tom’s therapist introduced him to a meditation technique called anapanasati,
which involves sitting silently with eyes closed and practicing nonjudgmental awareness of thoughts, feelings, and
sensations while breathing naturally. Tom practiced this coping strategy whenever he felt the overwhelming urge
to use drugs. As treatment progressed, Tom became more aware of the ways in which he turned anger and hatred
toward his parents inward, leading to impulsive behavior that put him in harm’s way. Tom began to notice moments
throughout his day when his cravings subsided, and he felt more inner peace and calm. Over time, he came to
understand this as his core spiritual self, which offered him wisdom, groundedness, and lovingkindness.
3.3
|
Case 3: Religious cultural psychotherapy for anxiety
Religious cultural psychotherapy retains Beck’s cognitive model and accommodates treatment for working with
Muslim clients (Razali, Hasanah, Aminah, & Subramaniam, 1998). This approach draws on passages from the Holy
Koran and Hadith (sayings and customs of the prophet Muhammad) to (a) examine the evidence for and modify
automatic negative thoughts, (b) facilitate the development of positive religious coping skills, including prayer,
acceptance, and adherence to Islamic customs, and (c) help clients understand symptoms within the context of their
cultural and religious beliefs to reduce mental health stigma. Clients are encouraged to cultivate feelings of
closeness to Allah, reflect on truths from the Koran, and express their worries and fears in prayer.
“Abdul”sought psychotherapy after being diagnosed with generalized anxiety disorder by his physician and
resisting a referral for medication. He identified as a 50‐year‐old Palestinian American, cisgender man. His Muslim
faith was his most salient identity. Within his religious tradition, taking medication for psychological difficulties was
discouraged, so he sought psychotherapy to learn to manage his symptoms. Although a successful businessman,
Abdul was constantly tense, consumed with racing thoughts and fears that made it difficult to concentrate at work
and home. Even during daily prayer, he could not focus. This compounded his fears that Allah would punish him for
his lack of faithfulness.
In psychotherapy, Abdul acknowledged that he did not believe the world was a safe place. He felt that he must
constantly prepare himself and his family for the worst‐case scenario. He also worried about how rising political
tensions and Islamophobia in the United States might impact his safety and that of his wife and three children. The
psychotherapist validated Abdul’s fears while working with him to identify how his faith could be a positive source
of support. Abdul found that it eased his worries to meditate on the beliefs that Allah was always in control and
that he could trust Allah to take care of him and his family. As Abdul’s shame over his symptoms subsided, he
attended the mosque more frequently and gained comfort from connection with others in his faith community.
4
|
META‐ANALYTIC REVIEW
4.1
|
Inclusion criteria
We included outcome studies of psychotherapy broadly defined (Norcross, 1990), published in the English
language, which explicitly integrated R/S throughout the psychotherapeutic process, either through incorporation
of R/S content within a standard technique (e.g., Christian cognitive therapy) or the addition of R/S practices (e.g.,
prayer, meditation, and reading sacred texts) as an adjunctive to sessions. Additionally, all studies that we
considered for inclusion compared an R/S‐accommodated treatment with either (a) a no‐treatment control
condition or (b) an alternate treatment. Although the vast majority of studies used an experimental (randomized)
research design, a small group of studies used a quasiexperimental (nonrandomized) design due to limitations
arising from the treatment setting.
1942
|
CAPTARI ET AL.
We excluded studies of peer‐led support groups (e.g., Alcoholics Anonymous, Celebrate Recovery) as well as
standalone self‐help interventions (e.g., meditation, intercessory prayer) that were not psychological treatment.
Because we were interested primarily in psychological and spiritual outcomes of psychotherapy, we did not
consider studies that examined physical health as the primary outcome measure. However, we did include outcome
studies in which psychological intervention was provided to individuals with a medical problem (e.g., cancer,
hypertension) who sought psychotherapy for associated psychological (e.g., depression, anxiety) or spiritual (e.g.,
meaningless, feeling far from God) difficulties.
4.2
|
Literature search
We identified studies for analysis using a comprehensive approach of both backward and forward search. First, we
identified relevant psychotherapy outcome studies through database searches of PsycINFO, PsycArticles,
Psychology and Behavioral Sciences Collection, SocINDEX, and Dissertation Abstracts International, as of May
15, 2017, using the key terms (counseling OR therapy) AND (religio* OR spiritu*) to define our search criteria.
Second, we used previous meta‐analyses and systematic reviews of the literature to identify additional outcome
studies that met the aforementioned criteria. Finally, we reviewed all issues to date of Spirituality in Clinical Practice
because this journal focuses on R/S‐oriented interventions.
Subsequently, we contacted the corresponding author for each study identified through the previous methods
to inquire about additional investigations they had conducted or were aware of, including unpublished file‐drawer
studies. Compared with findings supporting the null hypothesis, findings that support differences in treatments
have been found to be several times more likely to be published. Furthermore, effect sizes tend to be significantly
larger in published compared with unpublished studies. To minimize the risk of overestimating population effects
and account for publication bias, we included both published and unpublished findings.
Overall, we identified 102 independent samples from 97 studies with data available for inclusion. Among these,
45 samples (44 studies) compared R/S psychotherapy to a no‐treatment control condition only, 43 samples (40
studies) compared R/S psychotherapy to a comparison treatment only, and 14 samples (13 studies) compared R/S
psychotherapy to both no‐treatment control and other treatments. Among the 57 samples (53 studies) with a
comparison condition, 24 samples (23 studies) used an additive design, in which R/S was added to a standard
treatment and then compared with the standard treatment. Considered together, we examined a total of 116
comparisons in our analyses.
4.3
|
Effect size
For each study, we report g, which is a statistic that reflects the posttest difference between the R/S condition and
the comparison (or no‐treatment control) condition in standard deviation (SD) units. Mean differences were
reversed for negatively valenced outcome measures (e.g., depression, anxiety) so that positive values of effect sizes
indicate more favorable outcomes for the R/S condition relative to the comparison or control condition. Some
published studies did not include sufficient data to classify the study and calculate effect sizes. In these situations,
we contacted the corresponding author to request the missing data. In situations where the data were not
available, we excluded the studies from the meta‐analysis.
4.4
|
Study coding
We extracted the sample size and associated statistical information (e.g., means, SDs) necessary to determine the effect
direction and to calculate effect sizes. For potential moderators, we coded a number of study, treatment, and patient
characteristics. Study characteristics involved data source (e.g., published or unpublished), use of randomization, time
lapse to follow‐up data collection point, outcomes measured (coded as either religious or spiritual), and whether or not
CAPTARI ET AL.
|
1943
the study used an additive design. Treatment characteristics included therapeutic approach (e.g., CBT, psychospiritual),
intervention format (e.g., group, individual), accommodative focus (e.g., Christian, Muslim), number of sessions, use of a
treatment manual, and psychotherapy fidelity checks. Patient characteristics included age, gender, race or ethnicity,
presenting problem, use of psychotropic medication, and religious affiliation.
4.5
|
Data analysis
We used Comprehensive Meta‐Analysis Version 3.0 (Borenstein, Hedges, Higgins, & Rothstein, 2012). Random
effects models were used because effects were considered to be sampled from a population of effects. Consistent
with random effects models, studies were weighted by the sum of the inverse sampling variance plus tau‐squared.
Some studies reported more than one outcome measure. We calculated the effect size for the one psychological
outcome and one spiritual outcome that best assessed the goal of the specific psychotherapy. For example, if a
study purported to examine R/S cognitive behavioral therapy for depression, a measure such as the BDI‐II (Beck
et al., 1996) was chosen to account for psychological outcomes and a measure such as the SWBQ (Gomez & Fisher,
2003) was chosen to represent spiritual outcomes.
4.6
|
The studies and patients
We analyzed data from 7,181 patients (3,495 from R/S interventions, 1,634 from alternate interventions, and 2,052
from no treatment or control conditions), which was gathered from 102 independent samples. Most participants were
diagnosed with a primary mental health disorder (k= 50), such as depression, anxiety, or posttraumatic stress
disorder. Other participants received psychotherapy targeting psychological symptom distress and/or spiritual well‐
being following the diagnosis of a medication condition (k= 29), such as cancer or human immunodeficiency virus. Still
other participants reported couple conflicts, spiritual problems, unforgiveness, or similar life challenges (k=23).
Across studies, a number of diverse R/S perspectives were integrated in psychotherapy, with the majority being
Christianity (k= 28), Islam (k= 18), and general spirituality (k= 51). Treatment was provided in individual (k= 38),
group (k= 57), individual + group (k= 2), and couple or family (k= 4) formats. Psychotherapists utilized a variety of
approaches, which we categorized into six broad areas: Cognitive and/or behavioral (k= 33), existential and/or
narrative (k= 7), general psychospiritual (k= 33), mind‐body (k= 17), REACH forgiveness (k= 4), and supportive
and/or pastoral (k= 8). Follow‐up time ranged from 1–6 months for R/S treatment‐control studies (M= 3.10;
SD = 2.28), and from 1–12 months for R/S treatment‐alternate studies (M= 3.10; SD = 2.28).
5
|
RESULTS
Meta‐analytic results for psychological and spiritual outcomes are summarized in Table 1, which reports separate
effect sizes by outcome type (psychological or spiritual) and comparison type (no‐treatment controls, alternative
treatment comparisons, or additive‐treatment comparisons). The statistics reported are for the 90 studies that
used an experimental design.
Our first analysis examined whether or not patients who received R/S‐integrated treatment showed greater
improvement compared with patients in a no‐treatment control condition. R/S‐adapted psychotherapy out‐
performed no‐treatment control conditions on both psychological (g= 0.74, p< 0.000) and spiritual (g= 0.74,
p< 0.000) outcomes. These gains were similar at follow‐up (psychological: g= 0.81, p= 0.002; spiritual: g= 0.71,
p= 0.006). Treated participants were better off than no‐treatment control patients by about 0.7 SDs on average for
both sets of outcomes—a large effect that is typical of treatment‐control effect sizes for many forms of
psychotherapy (Wampold & Imel, 2015). We encourage caution when interpreting this result—something almost
always works better than nothing.
1944
|
CAPTARI ET AL.
Second, we examined whether patients who received R/S‐integrated treatment displayed greater improvement
on psychological and spiritual outcomes compared with patients receiving any alternate (e.g., secular) form of
psychotherapy. Studies that used an identical theoretical orientation and therapy duration to isolate the impact of R/S
accommodation were examined separately (see next analysis). R/S‐adapted psychotherapy out‐performed alternate
treatments on both psychological (g=0.33,p< 0.001) and spiritual (g=0.43,p< 0.001) measures. Psychological gains
were maintained at follow‐up (g=0.33, p= 0.007), but spiritual gains were not (g=0.21, p= 0.245). At termination,
participants who received R/S accommodative psychotherapy were better off than those who received an alternate
secular treatment by about 0.3 SDs on average for both sets of outcomes, which is a small–medium effect. At follow‐
up, these participants continued to report less psychological symptoms by about 0.3 SDs.
In our third and most important analysis, we examined studies in which the R/S and alternate (e.g., secular)
psychotherapy conditions used the same theoretical approach and treatment duration (thus constituting an
additive design). The goal here was to isolate the additive effects of R/S‐specific intervention elements. There was
not a significant effect of R/S integration on psychological outcomes directly following treatment (g= 0.13,
p= 0.258) or at follow‐up (g= 0.22, p= 0.062). This means that R/S‐accommodated treatments were as effective,
but not more effective, than standard psychotherapy. However, R/S‐adapted psychotherapy did outperform
standard psychotherapy on spiritual outcomes, both directly following treatment (g= 0.34, p< 0.000) and at follow‐
up (g= 0.33, p= 0.037). Participants who received R/S‐accommodative psychotherapy reported greater spiritual
well‐being by about 0.3 SDs on average, which is a small‐medium effect.
These meta‐analytic results provide substantial empirical support for incorporating R/S into psychological
treatment. Consistent with previous meta‐analyses, R/S‐adapted psychotherapy resulted in greater improvement in
patients’psychological and spiritual functioning compared with alternative non‐R/S psychotherapies. With more
stringent criteria, R/S treatments were equivalent to secular treatments on psychological outcomes and were
superior to secular treatments on spiritual outcomes, both at posttest and follow‐up.
5.1
|
Publication bias
The file‐drawer problem refers to the possibility that there may be unpublished studies that we were unable to
retrieve and include in the meta‐analysis. Thus, we conducted a fail‐safe Nanalysis to estimate the number of
TABLE 1 Results for psychological and spiritual outcomes of all randomized studies
Posttest Follow‐up
Comparison Nkg 95% CI I
2
Nkg 95% CI I
2
Psychological outcomes
Control 3664 50 0.74*** 0.52, 0.96 89.92 1522 17 0.81** 0.30, 1.31 95.10
Alternate 2283 31 0.33*** 0.20, 0.47 60.49 896 14 0.33** 0.09, 0.57 66.92
Additive 805 19 0.13 −0.09, 0.34 53.79 465 14 0.22 −0.01, 0.44 30.33
Spiritual outcomes
Control 2373 29 0.74*** 0.48, 0.99 88.34 1112 11 0.71** 0.20, 1.21 93.56
Alternate 817 13 0.43*** 0.19, 0.66 63.45 404 7 0.21 −0.14, 0.56 66.98
Additive 601 13 0.34*** 0.18, 0.50 0 268 8 0.32* 0.02, 0.62 30.70
Note. CI: confidence interval for g;g: Hedge’sg, a measure of effect size, which corrects for potential bias in Cohen’sd;
I
2
: percentage of the observed variance that reflects real differences in effect sizes; k: number of effect sizes summarized;
N: sample size summed across studies.
*p<0.05.
**p<0.01.
***p<0.001.
CAPTARI ET AL.
|
1945
additional studies with nonsignificant results that would need to be added to the meta‐analysis to change the
overall conclusions. Results (see Table 2) suggest minimal impact of the file‐drawer problem, with the exception
of follow‐up analyses, which may be more modestly prone to bias because of small sample sizes. If all existent
file‐drawer studies were retrievable, effect sizes for follow‐up analyses of spiritual outcomes, in particular, could be
weakened or become nonsignificant. We also conducted the trim and fill procedure to estimate the number of
missing studies due to publication bias and statistically impute these studies, recalculating the overall effect size.
Results suggested minor over or underestimation of some effects, but overall conclusions remain unchanged.
5.2
|
Moderators
We categorized moderator variables into three groups: (a) Patient characteristics, (b) study characteristics, and (c)
treatment characteristics. Potential moderators were tested on the within‐group effect sizes, because we were
interested in assessing how these variables might impact response to treatment. We conducted metaregression to
control for potential confounding among variables and allow examination of unique effects. We dummy coded
polychotomous categorical moderators, indicating our reference groups as Christian (accommodative focus),
cognitive behavioral (treatment approach), and individual therapy (modality).
The findings of the moderator analyses on the within‐group effect sizes are reported in Table 3. Unpublished
studies trended toward smaller effects in symptom reduction compared with those published (p= 0.093).
Treatments including psychotropic medication predicted larger effects than those not including medication on
both psychological (p< 0.001) and spiritual (p= 0.001) outcomes. General spiritual accommodated psychotherapy
was significantly less effective than Christian accommodated psychotherapy on spiritual outcomes (p= 0.050),
but not on psychological outcomes. In terms of treatment approach, supportive and pastoral therapy showed
weaker effects than CBT on symptom reduction (p= 0.016) and spiritual outcomes (p= 0.012). Mind‐body
psychotherapies predicted slightly smaller reductions in psychological distress compared with CBT (p= 0.051).
Finally, group formats trended toward weaker effects than individual psychotherapy on psychological outcomes
(p=0.067).
TABLE 2 Results for fail‐safe Nanalyses
Posttest Follow‐up
Comparison Rosenthal’sK+
zfor observed
studies Orwin’sK+ Rosenthal’sK+
zfor observed
studies Orwin’sK+
Psychological outcomes
Control 4133 17.26*** 224 486 10.36*** 64
Alternate 460 7.47*** 69 67 4.71*** 32
Additive 122 4.91*** 57 32 3.45*** 30
Spiritual outcomes
Control 1851 15.52*** 166 224 9.05*** 46
Alternate 136 6.02*** 47 0 1.95* 4
Additive 71 4.81*** 39 7 2.64** 17
Note. Orwin’sK+: the number of missing studies with a non‐existent or trivial g(in this case <0.10) that would need to be
added to the analyses to bring the overall gunder 0.10; Rosenthal’sK+: the number of missing studies with a mean effect of
zero that would need to be added to the analyses to bring the p< 0.05; z= the overall z‐score for observed studies.
*p<0.05.
**p<0.01.
***
p<0.001.
1946
|
CAPTARI ET AL.
TABLE 3 Multiple moderator analyses prepost R/S psychotherapy
Comparisons kB 95% CI z(B)Qdfp
Psychological outcomes
Overall model: 64 120.03*** 14 0.000
Intercept 2.89 1.85, 3.93 5.43*** 0.000
Race/ethnicity 0.00 −0.00, 0.01 0.69 0.487
Published −0.51 −1.10, 0.09 −1.68
+
0.093
Accommodative focus 1.01 2 0.603
General spiritual 0.02 −0.61, 0.64 0.06 0.956
Muslim 0.40 −0.55, 1.36 0.83 0.407
Treatment approach 14.77* 5 0.011
Existential/narrative −0.70 −1.58, 0.17 −1.58 0.114
General psychospiritual 0.11 −0.57, 0.80 0.33 0.742
Mind‐body −0.79 −1.58, 0.00 −1.95
+
0.051
REACH forgiveness −0.12 −1.11, 0.88 −0.23 0.821
Supportive/pastoral −1.11 −2.00, −0.21 −2.42* 0.016
Modality 3.52 2 0.172
Group −0.57 −1.18, 0.04 −1.83
+
0.067
Couple or family −0.64 −1.82, 0.54 −1.07 0.286
Medication included in Tx −1.55 −2.32, −0.77 −3.93*** 0.000
Treatment manual 0.22 −0.37, 0.82 0.73 0.467
Treatment fidelity check 0.22 −0.40, 0.83 0.69 0.491
Spiritual outcomes
Overall model: 38 41.06*** 13 0.000
Intercept 0.64 0.36, 0.92 4.50*** 0.000
Race/ethnicity −0.00 −0.01, 0.00 −1.08 0.278
Published −0.14 −0.44, 0.16 −0.91 0.365
Accommodative focus 8.33* 2 0.016
General spiritual −0.26 −0.52, 0.00 −1.96* 0.050
Muslim 0.26 −0.23, 0.75 1.05 0.296
Treatment approach 10.39
+
5 0.065
Existential/Narrative 0.07 −0.31, 0.45 0.37 0.713
General psychospiritual 0.30 −0.12, 0.71 1.40 0.162
Mind‐body 0.21 −0.21, 0.64 0.98 0.330
REACH forgiveness 0.05 −0.70, 0.80 0.14 0.893
Supportive/pastoral −0.51 −0.91, −0.11 −2.50* 0.012
Medication included in Tx −1.63 −2.56, −0.71 −3.46*** 0.001
Modality −0.30 −0.70, 0.11 −1.44 0.150
(Continues)
CAPTARI ET AL.
|
1947
6
|
PATIENT CONTRIBUTIONS
It has been theorized that a client’s level of R/S affiliation, practice, and/or commitment may impact the
effectiveness of R/S psychotherapeutic accommodations. The majority of outcome studies assessed patient R/S
solely as a demographic variable (participants self‐identified as Christian, Muslim, Jewish, spiritual, etc.). Thus, in
most cases, patient R/S was viewed as a dichotomous rather than continuous variable and used for informational
purposes only. We tested religious affiliation (percentage of the sample religiously affiliated) as a potential
moderator, but did not find a significant effect. Much is lost by analyzing treatment outcomes of all R/S individuals
based on identification alone, because there is tremendous variance in the extent to which people are influenced by
their R/S beliefs. More specific measurements of the strength of R/S are needed to better understand the relation
between patient R/S and psychotherapy outcome.
Regrettably, we were not able test moderation of the strength of R/S commitment because few studies
measured or reported this information. Only a few investigations used a measure of R/S beliefs or commitment
in the pretreatment screening process, identifying a minimum cutoff score for inclusion to ensure that all patients in
the study were at least moderately R/S. This lack of specificity in research and measurement represents a
significant gap for future exploration.
7
|
LIMITATIONS OF THE RESEARCH
Despite growing empirical support for R/S‐adapted treatments, some shortcomings must be acknowledged. Our results
at follow‐up should be interpreted more cautiously in light of fewer studies including this data, resulting in a smaller
analyzable sample. Generalizability of findings has been limited by the use of relatively homogeneous samples that fail
to capture many aspects of diversity evident in “real world”clients. Furthermore, small samples make it difficult to
capture a true treatment effect. Many studies did not use a bona fide comparative secular and R/S treatment (e.g., same
theoretical orientation and duration). Without isolating the R/S component, it is difficult to tease apart the relative
impact of R/S from other features of psychotherapy. Future research can ensure the alternative treatment condition is
identical, with the exception of the R/S component. Investigators would also be wise to recognize and control for
potential allegiance effects, since many researchers in this area identify as R/S themselves.
Researchers have varying perspectives about what constitutes R/S integration. We have outlined four broad
ways treatment can be tailored (conceptualization, treatment goals, intervention, and interpersonal process). We
encourage researchers to be specific about how and how much they integrate R/S, and to make treatment manuals
available for cross‐validation studies. More psychotherapy outcome studies are needed examining therapeutic
approaches aside from CBT, as well as adaptations to other major world religions and spiritual traditions besides
Christianity.
TABLE 3 (Continued)
Comparisons kB 95% CI z(B)Qdfp
Treatment manual 0.12 −0.14, 0.39 0.90 0.366
Treatment fidelity check −0.18 −0.46, 0.11 −1.24 0.217
Note. Reference groups are as follows: Accommodative focus—Christian, treatment approach—Cognitive Behavioral, modality—
individual, medication incorporated in treatment—yes, treatment manual—yes, and treatment fidelity check—yes.
B: slope; CI: confidence interval; k: number of studies; p: two‐sided p‐value indicating statistical significance for each level
of the model; Q: homogeneity test; z(B): zstatistic for the slope.
+
p< 0.10.
*p<0.05.
**p<0.01.
***p<0.001.
1948
|
CAPTARI ET AL.
Particularly in light of the differences noted between psychological and spiritual change trajectories, future
research could include measures of spiritual well‐being and assess spiritual outcomes. While R/S composition is
commonly included as a patient demographic, future studies can gather more extensive data about the client’s R/S
and evaluate treatment effectiveness based on strength of religious commitment, as well as daily spiritual
experiences and practices, such as attendance at religious services, prayer, and reading of sacred texts. Finally, in
most cases, only patient R/S has been assessed. Further exploration of the relative importance of value similarity
between psychotherapist and client is warranted.
8
|
DIVERSITY CONSIDERATIONS
While the majority of studies reported demographic information, these data were rarely integrated in
posttreatment analyses. Little research has explored R/S diversity, including religious commitment and daily
spiritual practices. As the research base grows on incorporating patients’R/S beliefs and values in treatment, it is
vital that psychotherapists better understand what sorts of individuals are most likely to benefit. For example, it
may be that older individuals, women, those of lower socioeconomic status, or clients of certain racial or ethnic
backgrounds experience a greater or lesser benefit of religious tailoring. Many questions remain to be explored,
including diversity aspects of not only the patient, but also the psychotherapist offering treatment. We encourage
researchers to examine these diversity variables as moderators when analyzing treatment outcomes.
9
|
THERAPEUTIC PRACTICES
The APA Ethics Code (2017, Principle E) affirms religion as a key consideration within diversity, and research
has shown that attending to clients’R/S values and beliefs can positively influence treatment outcomes.
“To ignore religion as a cultural issue may not only be unethical, but also lead to malpractice”(Plante, 2014,
p. 289). R/S‐specific ethical competencies have been proposed (Vieten et al., 2013), and can be used to guide
therapists’development of relevant attitudes, knowledge, and skills that promote spiritually conscious
psychological care.
Taking time to explore and understand a client’s R/S values and experiences communicates that this aspect of
their identity is welcome and an asset in the therapeutic process. R/S is an important cultural lens that can be
creatively applied in conceptualization, treatment goals, intervention, and interpersonal process, remembering that
R/S adaptations of psychotherapy are as unique as each patient who walks through the door.
Avoid making assumptions based on religious identification, and instead, explore patients’unique desires,
needs, and expectations. What does R/S mean to them? What role does it play in their day‐to‐day life? How do they
perceive and interact with the divine? In what ways do they perceive R/S to be a source of strength? What practices
or activities might be powerful coping resources? Conversely, how might R/S be a source of struggle or inner
turmoil? Are they experiencing difficulty reconciling previously‐held R/S beliefs with their lived experience? Have
they been the target of R/S oppression or abuse? How might this have impacted their relationship with the sacred?
Throughout treatment, these are just a few of many potentially relevant R/S areas to explore.
To conclude, we offer several clinical applications based on the findings of this meta‐analysis. We also
recommend several texts detailing evidence‐based strategies for R/S integration within clinical practice (Aten,
O’Grady, & Worthington, 2012; Pargament, 2011).
•Treat religion and spirituality as a potentially important aspect of the client’s identity. Express curiosity about
each client’s lived experience. Explore R/S history, values, and commitment as part of the intake process, and
consider intersectionality with other dimensions of diversity.
CAPTARI ET AL.
|
1949
•Incorporate a patient’s R/S values and worldview in psychotherapy as requested and when clinically indicated.
Research shows that accommodative psychotherapies are at least as effective as secular approaches in reducing
psychological symptoms, and can be provided by therapists regardless of their personal R/S background.
•Consider R/S adaptations in psychotherapy for their unique benefits to clients’spiritual lives, including greater
spiritual well‐being and increased connection with the sacred. When a client’s treatment goals include not only
symptom remission, but also spiritual development, integration of R/S within psychotherapy is a treatment of choice.
•Tailor treatment especially when working with patients whose R/S is an influential force in their day‐to‐day life.
Preliminary findings suggest that accommodative psychotherapies may result in the greatest symptom reduction
among clients with a high level of R/S commitment.
•Follow the client’s lead when incorporating R/S beliefs and practices into psychotherapy. Avoid making
assumptions based on religious identification, and instead, explore their unique desires, needs, and expectations.
•Practice respect and cultural humility when discussing patients’religious worldviews and practices. Be especially
sensitive to one’s own potential biases about organized religion and to clients’experiences of the sacred.
REFERENCES
Aten, J. D., O’Grady, K., & Worthington, E., Jr (2012). The psychology of religion and spirituality for clinicians: Using research in
your practice. New York: Routledge.
Avants,S.K.,&Margolin,A.(2004).DevelopmentofSpiritualSelf‐Schema (3‐S) therapy for the treatment of addictive and HIV
risk behavior: A convergence of cognitive and Buddhist psychology. Journal of Psychotherapy Integration,14(3), 253–289.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression Inventory manual, San Antonio, TX: The Psychological
Corporation.
Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. R. (2012). Comprehensive meta‐analysis. Version 3.0, Englewood, NJ:
Biostat, Inc.
Davis, D. E., Rice, K., Hook, J. N., Van Tongeren, D. R., DeBlaere, C., Choe, E., & Worthington, E. L., Jr. (2015). Development
of the sources of spirituality scale. Journal of Counseling Psychology,62(3), 503–513.
Delaney, H. D., Miller, W. R., & Bisonó, A. M. (2007). Religiosity and spirituality among psychologists: A survey of clinician
members of the American Psychological Association. Professional Psychology: Research and Practice,38, 538–546.
Diener, E., Tay, L., & Myers, D. G. (2011). The religion paradox: If religion makes people happy, why are so many dropping
out? Journal of Personality and Social Psychology,101(6), 1278–1290.
Exline, J. J., & Rose, E. D. (2014). Religious and spiritual struggles. In Paloutzian, R. F., & Park, C. L. (Eds.), Handbook of the
psychology of religion and spirituality (pp. 380–398). New York: Guilford Publications.
Fetzer Institute (1999). Multidimensional measure of religiousness/spirituality for use in health research, Kalamazoo, MI:
Fetzer Institute. Retrieved from. http://fetzer.org/resources/multidimensional‐measurement‐religiousnessspirituality‐
use‐health‐research.
Gomez, R., & Fisher, J. W. (2003). Domains of spiritual well‐being and development and validation of the Spiritual Well‐
Being Questionnaire. Personality and Individual Differences,35(8), 1975–1991.
Hamilton, M. A. X. (1959). The assessment of anxiety states by rating. British Journal of Medical Psychology,32(1), 50–55.
Hill, P. C., Pargament, K. I., Hood, R. W., Jr., McCullough, J. M. E., Swyers, J. P., Larson, D. B., & Zinnbauer, B. J. (2000).
Conceptualizing religion and spirituality: Points of commonality, points of departure. Journal for the Theory of Social
Behavior,30,51–77.
Koenig, H. G., King, D., & Carson, V. B. (2012). Handbook of religion and health, New York: Oxford University Press.
Margolin, A., Schuman‐Olivier, Z., Beitel, M., Arnold, R. M., Fulwiler, C. E., & Avants, S. K. (2007). A preliminary study of spiritual
self‐schema (3‐S+) therapy for reducing impulsivity in HIV‐positive drug users. Journal of Clinical Psychology,63, 979–999.
Norcross, J. C. (1990). An eclectic definition of psychotherapy. In Zeig, J. K., & Munion, W. M. (Eds.), What is psychotherapy?
Contemporary perspectives (pp. 218–220). San Francisco, CA: Jossey‐Bass.
Paloutzian, R. F., & Park, C. L. (2014). Handbook of the psychology of religion and spirituality. New York: Guilford Publications.
Pargament, K. I. (2011). Spiritually integrated psychotherapy: Understanding and addressing the sacred. New York: Guilford Press
Pargament, K., Feuille, M., & Burdzy, D. (2011). The Brief RCOPE: Current psychometric status of a short measure of
religious coping. Religions,2(1), 51–76.
Pearce, M. J., Koenig, H. G., Robins, C. J., Nelson, B., Shaw, S. F., Cohen, H. J., & King, M. B. (2015). Religiously integrated
cognitive behavioral therapy: A new method of treatment for major depression in patients with chronic medical illness.
Psychotherapy,52(1), 56–66.
Pew Research (2016). Religious landscape study, http://www.pewforum.org/religious‐landscape‐study.
1950
|
CAPTARI ET AL.
Plante, T. G. (2014). Four steps to improve religious/spiritual cultural competence in professional psychology. Spirituality in
Clinical Practice,1(4), 288–292.
Razali, S. M., Hasanah, C. I., Aminah, K., & Subramaniam, M. (1998). Religious‐sociocultural psychotherapy in patients with
anxiety and depression. Australian and New Zealand Journal of Psychiatry,32, 867–872.
Saunders, S. M., Miller, M. L., & Bright, M. M. (2010). Spiritually conscious psychological care. Professional Psychology:
Research and Practice,41(5), 355–362.
Schafer, R. M., Handal, P. J., Brawer, P. A., & Ubinger, M. (2011). Training and education in religion/spirituality within APA‐
Accredited clinical psychology programs: 8 years later. Journal of Religion and Health,50, 232–239.
Vieten, C., Scammell, S., Pilato, R., Ammondson, I., Pargament, K. I., & Lukoff, D. (2013). Spiritual and religious competencies
for psychologists. Psychology of Religion and Spirituality,5(3), 129–144.
Wampold, B. E., & Imel, Z. E. (2015). The great psychotherapy debate: The evidence for what makes psychotherapy work.
New York: Routledge.
How to cite this article: Captari LE, Hook JN, Hoyt W, Davis DE, McElroy-Heltzel SE, Worthington
EL Jr. Integrating clients’religion and spirituality within psychotherapy: A comprehensive meta‐analysis.
J Clin Psychol. 2018;74:1938–1951. https://doi.org/10.1002/jclp.22681
CAPTARI ET AL.
|
1951
