ArticlePDF AvailableLiterature Review

Current Status and Future Directions in Couple Therapy



Couple therapy research affirms that various approaches to couple treatment produce statistically and clinically significant improvement for a substantial proportion of couples in reducing overall relationship distress. Recent studies have extended these findings in indicating the effectiveness of couple-based interventions for a broad range of coexisting emotional, behavioral, or physical health problems in one or both partners. In contrast to these encouraging results, research also indicates that a sizeable percentage of couples fail to achieve significant gains from couple therapy or show significant deterioration afterward. Research on processes of change and predictors of treatment outcome in couple therapy provides preliminary evidence regarding factors potentially contributing to variable treatment response. The chapter concludes with 12 recommendations regarding future directions in couple therapy research and clinical training.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Annu. Rev. Psychol. 2006. 57:317–44
doi: 10.1146/annurev.psych.56.091103.070154
2006 by Annual Reviews. All rights reserved
First published online as a Review in Advance on July 8, 2005
Douglas K. Snyder and Angela M. Castellani
Department of Psychology, Texas A&M University, College Station, Texas 77843-4235;
email:, angela
Mark A. Whisman
Department of Psychology, University of Colorado, Boulder, Colorado 80309-0345;
marital therapy, couple therapy, couple research, couple change
processes, couple treatment predictors
Abstract Couple therapy research affirms that various approaches to couple treat-
ment produce statistically and clinically significant improvement for a substantial
proportion of couples in reducing overall relationship distress. Recent studies have
extended these findings in indicating the effectiveness of couple-based interventions
for a broad range of coexisting emotional, behavioral, or physical health problems
in one or both partners. In contrast to these encouraging results, research also indi-
cates that a sizeable percentage of couples fail to achieve significant gains from couple
therapy or show significant deterioration afterward. Research on processes of change
and predictors of treatment outcome in couple therapy provides preliminary evidence
regarding factors potentially contributing to variable treatment response. The chap-
ter concludes with 12 recommendations regarding future directions in couple therapy
research and clinical training.
INTRODUCTION .................................................... 318
IN TREATING RELATIONSHIP DISTRESS .............................. 319
IN TREATING COMORBID DISORDERS ............................... 323
Couple-Based Treatment of Specific Relationship Problems ................. 324
Couple-Based Treatment of Mental and Physical Health Problems ............ 326
PROCESSES OF CHANGE IN COUPLE THERAPY ........................ 330
Methods of Investigating Change Processes .............................. 330
Empirical Findings Regarding Change Processes .......................... 331
PREDICTORS OF COUPLE THERAPY OUTCOME ........................ 333
Methods of Identifying Predictors of Treatment Outcome ................... 333
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Empirical Findings Regarding Predictors of Couple Therapy Outcome ......... 334
Directions for Research .............................................. 336
Directions for Clinical Training ........................................ 338
CONCLUSIONS ..................................................... 339
The fundamental challenges of psychotherapy research—whether evaluating in-
dividual, couple, or family interventions—are to identify effective treatments, un-
derstand their underlying mechanisms of change, and delineate aspects of the
therapist, client, or context that influence their outcome. In this chapter, we exam-
ine the effectiveness of couple-based interventions for treating general relationship
distress as well as coexisting emotional, behavioral, and physical health problems.
We discuss methods for evaluating processes of change in couple therapy and pre-
dictors of treatment outcome, along with empirical findings in these domains. We
conclude with recommendations for future research and clinical training in couple
Couple therapy continues to gain in stature as a vital component of mental health
services. Three factors contribute to this growing recognition: (a) the prevalence
of couple distress in both community and clinic samples; (b) the impact of couple
distress on both the emotional and physical well-being of adult partners and their
offspring; and (c) increased evidence of the effectiveness of couple therapy not
only in treating couple distress and related relationship problems but also as a
primary or adjunct treatment for a variety of individual emotional, behavioral, or
physical health disorders.
Couple distress is prevalent in both community epidemiological studies and in
research involving individual treatment samples. In the United States, the most
salient indicator of couple distress remains a divorce rate of approximately 50%
among married couples (Kreider & Fields 2002), with about half of these oc-
curring within the first seven years of marriage. Independent of divorce, the re-
search literature suggests that many, if not most, marriages experience periods
of significant turmoil that place them at risk for dissolution or symptom devel-
opment (e.g., depression or anxiety) in one or both partners at some point in
their lives. In a recent national survey, the most frequently cited causes of acute
emotional distress were relationship problems including divorce, separation, and
other marital strains (Swindle et al. 2000). Other recent studies indicate that mari-
tally discordant individuals are overrepresented among individuals seeking mental
health services, regardless of whether they report marital distress as their primary
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
complaint (Lin et al. 1996). In a study of 800 employee assistance program (EAP)
clients, 65% rated family problems as “considerable” or “extreme” (Shumway et al.
The linkage of relationship distress to disruption of individual emotional and
physical well-being emphasizes the importance of improving and extending empir-
ically based strategies for treating couple distress. Research indicates that couple
distress covaries with individual emotional and behavioral disorders beyond gen-
eral distress in other close relationships (Whisman et al. 2000). Moreover, couple
distress—particularly negative communication—has direct adverse effects on car-
diovascular, endocrine, immune, neurosensory, and other physiological systems
that, in turn, contribute to physical health problems (Kiecolt-Glaser & Newton
2001). Nor are the effects of couple distress confined to the adult partners. Gottman
(1999) cites evidence indicating that “marital distress, conflict, and disruption are
associated with a wide range of deleterious effects on children, including de-
pression, withdrawal, poor social competence, health problems, poor academic
performance, a variety of conduct-related difficulties, and markedly decreased
longevity” (p. 4). In brief, couple distress has a markedly high prevalence; has a
strong linkage to emotional, behavioral, and health problems in the adult partners
and their offspring; and is among the most frequent primary or secondary concerns
reported by individuals seeking assistance from mental health professionals.
How effective is couple therapy? Previous reviews affirm that various versions
of couple therapy produce moderate, statistically significant, and often clinically
significant effects in reducing relationship distress. In this section, we examine
current findings regarding the effectiveness of couple therapy for treating overall
relationship distress. In the subsequent section, we review evidence regarding
the effectiveness of couple therapy for co-occurring individual and relationship
Since Christensen & Heavy’s (1999) review of couple therapy in the Annual
Review of Psychology,several qualitative and quantitative (meta-analytic) reviews
of couple therapy have appeared. Shadish & Baldwin (2003) reviewed six previous
meta-analyses of studies comparing couple therapy versus no-treatment control
groups, including four published reviews (Dunn & Schwebel 1995, Hahlweg &
Markman 1988, Johnson et al. 1999, Shadish et al. 1993) and two unpublished
reviews (Dutcher 1999, Wilson 1986). The samples of couple therapy studies
included in each of these reviews ranged from 4 [Johnson et al.s (1999) review of
emotion-focused couple therapy (EFCT)] to 163 [Shadish et al.s (1993) review of
couple and family therapy, of which 62 studies emphasizing couple therapy were
reanalyzed by Wilson (1986)]. Mean effect sizes across these six meta-analyses
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
ranged from approximately 0.50 (Wilson 1986) to 1.30 (Johnson et al. 1999).
Based on their review of these studies, Shadish & Baldwin (2003) reported an
overall mean effect size of 0.84 for couple therapy, indicating that the average
person receiving treatment for couple distress was better off at termination than
were 80% of individuals in the no-treatment control group.
Shadish & Baldwin (2003) also noted that their mean effect size for couple
therapy was generally comparable to or larger than that obtained by alternative in-
terventions ranging from individual therapy to medical interventions. They found
little evidence of differential effectiveness across different theoretical orientations
to couple therapy, particularly once other covariates (e.g., reactivity of measures)
were controlled. Noting the small number of couple therapy approaches listed by
the American Psychological Association Division 12 Task Force (Chambless &
Hollon 1998) as either well established [behavioral couple therapy (BCT)] or prob-
ably efficacious [EFCT and insight-oriented couple therapy (IOCT)], Shadish &
Baldwin (2003) argued that clinicians should also consider “meta-analytically sup-
ported treatments” such as cognitive-behavioral, systemic, and eclectic approaches
to couple therapy as viable approaches to treating general couple distress. They
also noted that numerous studies of couple therapy (particularly those emphasiz-
ing behavioral treatments) raised unanswered questions regarding their clinical
representativeness in that they failed to use clients referred through usual routes
and experienced therapists in actual clinic settings. Finally, among those studies
reporting data from follow-up at six months or longer, treatment effects tended to
be reduced but still significant.
Findings from alternative viewpoints or more recent research provide comple-
mentary perspectives to conclusions reached by Shadish & Baldwin in their 2003
summary. One such addition involves a follow-up meta-analysis of 30 random-
ized experiments with distressed couples contrasting BCT with a no-treatment
control (Shadish & Baldwin 2005). Their more recent analysis included 13 stud-
ies (7 published in journals and 6 unpublished dissertations) not included in the
Shadish et al. (1993) review. Overall, these 30 studies of BCT yielded a mean
effect size of 0.59, which was smaller than the mean effect size of 0.84 for cou-
ple therapy pooled across theoretical approaches reported by Shadish & Baldwin
(2003), and which indicated that the average individual receiving BCT was better
off at the end of treatment than were 72% of individuals in the control condi-
tion. In accounting for the more recent, smaller effect size obtained for BCT,
the authors noted the consequence of including nonpublished dissertations with
smaller sample sizes and small or negative effect sizes. They also reported that
of the various components comprising behavioral couple interventions (e.g., com-
munication training, problem-solving training, contingency contracting, behavior
exchange, desensitization, cognitive restructuring, and emotional expressiveness
training), only communication and problem-solving training led to larger effects,
whereas use of cognitive restructuring actually led to smaller effects. Shadish &
Baldwin (2005) found that the effects of BCT were unrelated to “dose” (defined by
number and length of sessions), reactivity of the dependent variables considered
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
(e.g., self-report measures of affect or cognition versus observational ratings of
communication behavior), or clinical representativeness.
Other than BCT, the sole approach to couple therapy evaluated in multiple trials
is EFCT, which combines an experiential, intrapsychic focus on inner emotional
experience with an emphasis on cyclical, self-reinforcing interactions (Johnson
et al. 1999). In four randomized trials, EFCT was superior to a waiting-list control
condition in reducing relationship distress, yielding recovery rates of 70%–73%
and a weighted mean effect size of 1.31 (Johnson 2002).
In addition to the two couple therapy approaches evaluated in multiple clinical
trials, several approaches have demonstrated positive outcomes in treating couple
distress in only one trial. First, Snyder & Wills (1989) compared insight-oriented
approaches to couple therapy with behavioral approaches in a controlled clinical
trial involving 79 distressed couples. The insight-oriented condition emphasized
the interpretation and resolution of conflictual emotional processes related to de-
velopmental issues, collusive interactions, and maladaptive relationship patterns.
At termination after approximately 20 sessions, couples in both treatment modali-
ties showed statistically and clinically significant gains in relationship satisfaction
compared with a wait-list control group. Treatment effect sizes at termination for
behavioral and insight-oriented conditions were 1.01 and 0.96, respectively; treat-
ment gains were substantially maintained at six-month follow-up. However, at four
years following treatment, 38% of couples in the behavioral condition had expe-
rienced divorce, in contrast to only 3% of couples treated in the insight-oriented
condition (Snyder et al. 1991).
Second, Goldman & Greenberg (1992) compared integrated systemic couple
therapy (ISCT) and EFCT with each other and with a wait-list control condition in
a randomized clinical trial of 42 couples. ISCT sought to disrupt repetitive, self-
perpetuating negative interactional cycles by changing the meaning attributed to
these cycles; changes in meaning were promoted by restructuring interactions and
reframing the problems using positive connotation followed by prescribing of the
symptom, encouraging the couple to “go slow, and finally prescribing a relapse or
reenactment of previous negative interactions. At the end of 10 one-hour weekly
sessions, ISCT and EFCT were both found to be superior to the control condition
and to be equally effective in alleviating marital distress, facilitating conflict res-
olution and goal attainment, and reducing target complaints. Moreover, couples
who received ISCT showed greater maintenance of gains in marital satisfaction
and goal attainment at four-month follow-up.
More recently, findings have emerged for an integrative behavioral approach
to couple therapy (IBCT; Jacobson & Christensen 1996) that combines tradi-
tional behavioral techniques for promoting change (specifically, communication
and behavior-exchange skills training) with strategies aimed at fostering emotional
acceptance. Interventions aimed at increasing acceptance include promoting tol-
erance and encouraging partners to appreciate differences and to use them to
enhance their marriage. In the largest randomized clinical trial of couple therapy
ever conducted, Christensen and colleagues (2004) compared the expanded IBCT
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
with traditional BCT by assigning 134 distressed couples to the two conditions,
stratified into moderately and severely distressed groups. Couples in IBCT made
steady improvements in satisfaction throughout the course of treatment, whereas
BCT couples improved more quickly than IBCT couples early in treatment but
then plateaued later in treatment. Both treatments produced similar levels of clini-
cally significant improvement by the end of treatment (71% of IBCT couples and
59% of BCT couples were reliably improved or recovered, based on self-reports
of overall relationship satisfaction).
Although various specific approaches to couple therapy have now demonstrated
effectiveness in reducing relationship distress in controlled trials, a substantial per-
centage of individuals fail to show significant improvement from these treatments
and an even greater percentage of individuals show deterioration in gains at follow-
up. Specifically, previous research has shown that about one third of couples fail
to achieve significant gains from treatment, and in only half of treated couples do
both partners show significant improvement in marital satisfaction at termination.
Moreover, assessment at two years or longer after termination indicates significant
deterioration among 30%–60% of treated couples (Cookerly 1980, Jacobson et al.
1987, Snyder et al. 1991). Such findings have fostered two alternative lines of
attack for treating couple distress: (a) distillation and emphasis of common factors
hypothesized to contribute to beneficial effects across “singular” treatment ap-
proaches, and (b) pluralistic models incorporating multiple components of diverse
treatment approaches.
Adopting the former strategy, Sprenkle & Blow (2004) argued that common
mechanisms of change cutting across the diverse couple therapies account for
the absence of significant differences in their overall effectiveness. They cited
five types or classes of common factors characterizing psychotherapy in general,
and three factors specific to couple or family therapy. Common factors viewed
as generic to psychotherapy include (a) client characteristics (e.g., learning style,
perseverance, and compliance with instructions or assignments), (b) therapist char-
acteristics (e.g., abilities to foster a therapeutic alliance and to match activity level
to clients’ expectations or preferences), (c) characteristics of the therapeutic re-
lationship (e.g., emotional connectedness and congruence between the therapist’s
and client’s specific expectations or goals), (d )expectancy or placebo effects, and
(e) nonspecific interventions promoting emotional experiencing, cognitive mas-
tery, and behavioral regulation. Those common factors viewed by Sprenkle &
Blow (2004) as specific to couple or family therapies include (a) emphasis on the
interpersonal context in which specific problems occur, (b) inclusion of multiple
members of the extended family system in direct treatment, and (c) fostering an
expanded therapeutic alliance across partners or multiple members of the family
as a whole. To date, there has been little research documenting specific treatment
effects attributable to proposed common factors—and no efforts in designing cou-
ple treatment approaches explicitly intended to maximize the therapeutic impact
of common factors (Sexton et al. 2004).
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
An alternative to the common factors approach involves efforts to incorpo-
rate active, specific treatment components from diverse approaches into multi-
component interventions in a systematic manner. Such approaches have variously
been described as “integrative” (e.g., Gurman 1981, 2002) or “pluralistic” (Snyder
1999), and are distinguished from eclecticism by their systematic selection or syn-
thesis within a conceptually coherent model. Gurman (2002) described a “depth-
behavioral” integrative approach to couple therapy that emphasizes the critical
interrelation of intrapsychic and interpersonal factors in couples’ interactions and
defines the goal of couple therapy as the loosening and broadening of each spouse’s
implicit matrix of assumptions, expectations, and requirements of intimate inter-
personal contact. This is accomplished through interpretation, cognitive restructur-
ing, and creation of therapeutic tasks to promote each spouse’s exposure to those
aspects of him- or herself and his or her partner that are blocked from awareness.
Snyder (1999) proposed a hierarchical approach to couple therapy incorporating
structural, behavioral, and cognitive techniques earlier in the therapeutic sequence
and drawing on insight-oriented techniques termed “affective reconstruction” later
in treatment primarily if relationship problems prove resistant to the earlier inter-
ventions. In affective reconstruction, previous relationships, their affective com-
ponents, and strategies for emotional gratification and anxiety containment are
reconstructed, with a focus on identifying for each partner recurring maladaptive
patterns in their interpersonal conflicts and coping styles across relationships. In
addition, interventions examine ways in which previous coping strategies vital to
prior relationships represent distortions or inappropriate solutions for emotional
intimacy and satisfaction in the current relationship. Neither the integrative depth-
behavioral approach proposed by Gurman (2002) nor the pluralistic approach
advocated by Snyder (1999) has been subjected to empirical evaluation, although
both approaches build on couple treatment approaches previously supported in
randomized clinical trials.
The co-occurrence between overall couple distress and specific relationship prob-
lems, as well as individual emotional or behavioral disorders, has been well estab-
lished in the research literature over the past decade. Based on these findings, new
couple-based treatments have emerged for treating distressed couples for whom
one or both partners have coexisting emotional, behavioral, or physical health
problems (Snyder & Whisman 2003). Snyder & Whisman (2004a) examined the
covariation between general relationship distress and problems in specific areas of
the couple’s relationship in a sample of 1020 community couples and 50 couples in
therapy, based on partners’ scores on the Marital Satisfaction Inventory—Revised
(Snyder 1997). Results indicated that individuals reporting moderate or higher
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
global relationship distress were five to six times more likely than nondistressed
persons to report specific relationship problems in the areas of physical aggression,
the sexual relationship, finances, and child rearing. Whisman (1999; Whisman &
Uebelacker 2005) evaluated the association between marital distress and 12-month
prevalence rates of 13 psychiatric disorders in 2538 married persons comprising the
National Comorbidity Survey. Results indicated that maritally distressed persons
were two to three times more likely than were nondistressed persons to experience
disorders involving mood, anxiety, or substance abuse.
The co-occurrence between overall couple distress and specific individual or
relationship problems has led to three couple-based treatment strategies for ad-
dressing these comorbid difficulties (Baucom et al. 1998). The first uses general
couple therapy to reduce overall relationship distress based on the premise that
marital conflict serves as a broad stressor that contributes to the development,
exacerbation, or maintenance of specific individual or relationship problems. The
second strategy involves developing disorder-specific couple interventions that
focus on particular partner interactions presumed to directly influence either the
co-occurring problems or their treatment. The third couple-based strategy involves
partner-assisted interventions in which one partner serves as a “surrogate therapist”
or coach in assisting the other partner with individual problems.
Couple-Based Treatment of Specific Relationship Problems
Couple-based interventions have been well established as effective in treating
two specific components of relationship functioning: difficulties in the sexual re-
lationship and problems of physical aggression. Recent findings also suggest the
effectiveness of couple therapy in treating couples dealing with issues of infidelity.
Epidemiological data indicate that 43% of women and 31%
of men will experience sexual dysfunction during their lifetime (Laumann et al.
1999). A recent review of couple-based treatments for sexual dysfunctions (Regev
et al. 2003) concluded that (a)sex therapy, primarily consisting of sensate focus,
is comparable to communication therapy in primary and secondary anorgasmic
women; (b) couples receiving couple therapy in addition to sex therapy demon-
strate more pronounced and comprehensive treatment gains, including significantly
more intense experiences of sex and sexual desire; and (c)sex therapy positively in-
fluences both sexual and marital problems, whereas general couple therapy appears
to facilitate resolution of marital problems only. Baucom et al. (1998) identified
several couple-based interventions with documented effectiveness in treating fe-
male sexual dysfunctions related to lifelong or situational orgasmic disorders or
hypoactive sexual desire. In partner-assisted treatment of orgasmic disorders, male
partners participate with their female partner in techniques of sensate focus; to-
ward the end of treatment, women are coached in sharing effective techniques of
masturbation with their partners. Couple-based interventions also assist couples
in discussing and resolving specific difficulties they experience in their sexual
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
interactions. Findings have affirmed the efficacy of couple interventions in treat-
ing women with primary or secondary orgasmic disorders, with improvement
rates ranging from 65% to 90%. Additional evidence supports combining general
behaviorally oriented couple therapy with orgasm-consistency training in the treat-
ment of women reporting hypoactive sexual desire (Hurlbert et al. 1993). Baucom
et al. (1998) noted that few studies of couple-based interventions have targeted
male sexual disorders despite evidence that sexual and marital problems are more
closely linked in men than in women and despite reports of recent increases in
male complaints of low sexual desire.
PHYSICAL AGGRESSION Mild to moderate physical aggression (e.g., pushing, grab-
bing, shoving, or slapping) occurs in more than half of couples seeking couple ther-
apy (Holtzworth-Munroe et al. 2003). Among couple therapy samples, as much as
85% of partner aggression is reciprocal, with both partners engaging in primarily
low levels of aggression. Moreover, psychological aggression (e.g., verbal abuse
or threats of violence) predicts physical abuse a year later (Murphy & O’Leary
1989). Although most therapists agree that couple therapy is inappropriate for
couples characterized by severe physical aggression (primarily violence by male
partners resulting in female partners’ injuries), couple-based interventions have
been effective in treating mild to moderate levels of aggression. Such interven-
tions emphasize anger management (e.g., recognition of anger, time-outs, and
self-regulation techniques) and communication skills (e.g., emotional expressive-
ness and problem solving). In their review of randomized trials comparing conjoint
treatment to gender-specific treatment, Holtzworth-Munroe et al. (2003) concluded
that conjoint couple therapy that has a direct and specific focus on eliminating vio-
lence “may be as effective as the more widely utilized gender specific treatments”
(p. 227). A more recent study compared the outcomes of a domestic violence-
focused treatment for 51 couples randomly assigned to either individual couple
therapy, a multicouple group treatment, or a no-treatment comparison group (Stith
et al. 2004). Male partner rates of physical aggression at six-month follow-up
were highest in the comparison group (66%) and lower in the multicouple group
(25%) than in the individual-couple therapy group (43%). Moreover, both marital
aggression and acceptance of physical aggression decreased significantly among
participants in the multicouple group therapy but not among participants in either
the individual-couple therapy or no-treatment comparison conditions, a finding
that suggests the multicouple group format has an incremental impact in changing
underlying attitudes toward relationship aggression.
EXTRAMARITAL AFFAIRS Research suggests that, on average, between 1.5% and
4% of married individuals will engage in extramarital sex in any given year (Allen
et al. 2005), and approximately one in three men aged 60–69 and one in five
women aged 40–49 report engaging in extramarital sex at some point in their lives
(Wiederman 1997). Although couples report extramarital affairs as a leading cause
of divorce and couple therapists describe infidelity as among the most difficult
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
problems to treat (Whisman et al. 1997), until recently there has been almost no
empirical study of interventions for couples dealing with affairs. Atkins and col-
leagues (2005b) examined treatment outcomes for 19 couples reporting an affair
by one of the partners participating in the randomized trial of IBCT versus BCT by
Christensen and colleagues (2004). Results showed that couples who reported infi-
delity were more distressed when they began treatment than were couples who did
not report infidelity, but couples for whom there had been an affair also improved
at a greater rate during the course of therapy than did couples not dealing with
infidelity. Gordon et al. (2004) reported findings from a replicated case study of an
integrative approach designed specifically to assist couples recovering from an ex-
tramarital affair. The six-month intervention comprised three phases that targeted
(a) coping with initial emotional and behavioral disruption of individual and rela-
tionship functioning following discovery or disclosure of the affair; (b)exploring
individual, relationship, and outside contextual factors contributing to the initial
onset or maintenance of the affair; and (c) reaching an informed decision about
how to move on, either individually or as a couple. At termination, the majority
of participants in the study reported less emotional and marital distress, and indi-
viduals whose partner had participated in the affair reported greater forgiveness
toward their partner.
Couple-Based Treatment of Mental and
Physical Health Problems
Research has documented the effectiveness of couple-based interventions for a
broad range of emotional and behavioral dysfunctions, including alcohol and re-
lated substance-use disorders, mood and anxiety disorders, and chronic pain and
related health problems. Promising couple-based interventions have also recently
emerged for a variety of other difficulties; interventions that have received at least
preliminary empirical evidence of their effectiveness are described here.
Alcohol- and drug-use disorders comprise the most
common psychiatric disorders in the general population, with lifetime prevalence
rates of 23.5% and 11.9%, respectively (Kessler et al. 1994). BCT for alcoholism
and drug abuse aims to alter couple and family interaction patterns to promote a
family environment more conducive to abstinence and sobriety (e.g., by reducing
the partner’s recurring complaints about past drinking and promoting attention to
positive aspects of current sober behavior), as well as to improve communication
and positive activities (Fals-Stewart et al. 2003). BCT for alcohol and drug abuse
typically involves 15–20 outpatient couple sessions over five to six months. In their
review of the literature regarding BCT, O’Farrell & Fals-Stewart (2000) concluded:
First, BCT for both alcoholism and drug abuse produces more abstinences and
fewer substance-related problems, happier relationships, fewer couple sepa-
rations and lower risk for divorce than does individual-based treatment. Sec-
ond, domestic violence is substantially reduced after BCT for both alcoholism
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
and drug abuse. Third, cost outcomes after BCT are very favorable for both
alcoholism and drug abuse, and are superior to individual-based treatment for
drug abuse (p. 51).
Recently, Fals-Stewart and colleagues (2005a) examined the clinical efficacy
and cost-effectiveness of a shortened version of BCT with 100 alcoholic male
patients and their partners. In the shortened version, couples participated in only
6 rather than 12 conjoint sessions; alcoholic clients participated in an additional
12 weekly individual sessions. Results indicated that those assigned to the brief
version had posttreatment and 12-month outcomes equivalent to clients receiv-
ing standard BCT, thereby supporting the cost-effectiveness of this shortened
MOOD DISORDERS Lifetime prevalence rates for major depressive episode and
dysthymia are estimated at 17.1% and 6.4%, respectively (Kessler et al. 1994).
Three clinical trials have shown that behavioral couple interventions for depres-
sion emphasizing behavior exchange, communication and problem-solving skills,
and cognitive interventions (e.g., cognitive reframing and directing attention to
positive change) are effective in relieving depression when provided to maritally
distressed couples with a depressed partner (Gupta et al. 2003). Furthermore,
compared with individual-based therapies, BCT has the incremental benefit of
improving overall relationship satisfaction. Clinical guidelines for providing BCT
for depressed individuals are provided by Beach & Gupta (2003). Similarly, con-
joint couple therapy incorporating components of interpersonal psychotherapy for
depression aimed at helping depressed individuals better understand and nego-
tiate their interpersonal relationships has been shown to be effective in treating
depression (Foley et al. 1989). More recently, Leff et al. (2000) reported promis-
ing results for systemic couple therapy, using interventions designed to reduce
problematic patterns of interacting, for depressed married individuals with a crit-
ical spouse. Finally, Dessaulles et al. (2003) compared 14 sessions of EFCT with
pharmacotherapy in treating wives’ major depression in 18 couples randomly
assigned to treatment condition. Both interventions were equally effective in reduc-
ing depressive symptoms, although there was some evidence that women receiv-
ing EFCT made greater improvement following termination than those receiving
ANXIETY DISORDERS Excessive anxiety is one of the most frequent mental health
problems in the United States, with a lifetime prevalence rate of developing any
anxiety disorder at 24.9% (Kessler et al. 1994). In their review of couple-based in-
terventions for anxiety disorders, Baucom et al. (2003) noted that anxiety disorders
may negatively impact couple functioning by disrupting interaction patterns, in-
creasing tension and arguments, restricting relationship activities, or decreasing at-
tention to the needs of the nonanxious partner. Baucom et al. (2003) described ways
in which efficacious treatments for anxiety disorders (e.g., exposure and response
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
prevention, cognitive restructuring, and relaxation training) can be incorporated
into either (a) partner-assisted interventions using the partner to assist with expo-
sure exercises and provide support or (b) couple-based interventions focusing on
ways in which couple functioning maintains anxiety symptoms, as well as ways
in which the anxiety influences couple functioning. In their review of specific
couple-based interventions for various anxiety disorders, Baucom et al. (1998)
concluded that partner-assisted exposure treatment of obsessive-compulsive dis-
order is at least as effective as treating the patient without such assistance; they
also determined that exposure interventions for agoraphobia may show enhanced
benefit from involving the partner even when there is no overt relationship distress,
a conclusion affirmed in a recent review by Byrne et al. (2004).
Over the past 15 years, psychosocial pain research-
ers have become increasingly interested in the role that partners play in how pa-
tients adjust to pain and in involving partners in psychosocial pain-management
efforts (Keefe et al. 2006). For example, partners can encourage patients in ac-
quiring more effective pain-control strategies, and can be discouraged themselves
from criticizing appropriate coping skills, enforcing the patient’s rest, or insisting
that pain medication is the only way to manage pain. Moreover, couple-based in-
terventions for chronic pain—and for physical illness more generally—can help
partners to cope with their own emotional struggles with caretaking, promote more
effective communication around pain and emotional distress, and facilitate couple
processes for providing emotional and tangible support, dealing with conflict, and
expressing affection and intimacy (Keefe et al. 2006).
Keefe and colleagues (1996) randomly assigned 88 patients with osteoarthritic
knee pain to spouse-assisted coping skills training, a conventional coping skills
training condition alone, or an arthritis education and partner-support control con-
dition. Patients in the spouse-assisted coping skills training condition received
training in a variety of cognitive and behavioral pain-coping skills (e.g., relax-
ation, imagery, distraction techniques, activity pacing, goal setting, and cognitive
restructuring) and they and their partners received training in various couples
skills (e.g., joint practice, communication skills, behavioral rehearsal, problem
solving, and maintenance training). Patients in the partner-assisted coping skills
training had the best outcomes across multiple criteria, whereas those in the arthri-
tis education–social support control condition had the worst outcomes. Moreover,
patients in the partner-assisted coping skills training who showed increases in mar-
ital adjustment were more likely to show lower levels of psychological disability,
physical disability, and pain behavior at 12-month follow-up.
Couple-based interventions for patients dealing with cancer have resulted in
similar findings. Specifically, Keefe and colleagues (2005) recently completed
a study that tested the effects of a partner-guided pain management interven-
tion for 78 patients with end-of-life cancer pain. The intervention, delivered by a
nurse, integrated information about cancer pain with training in three pain-coping
skills and emphasized the role these skills could play in controlling patient and
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
partner emotional responses and relational exchanges. Results indicated that pa-
tients receiving this intervention tended to report reduced levels of pain, and that
their spouses improved in their sense of efficacy for helping the patient control
pain and tended to report reduced levels of caregiver strain.
Finally, a recent study examined the impact of an eight-session couple therapy
for nine couples in which one partner was diagnosed with a terminal illness (Mohr
et al. 2003). Conjoint sessions emphasized (a) helping patients and their partners to
find meaning together through examining beliefs, goals, and values; (b) increasing
intimacy, emotional support, and reciprocity; and (c)facilitating conversations
about death and dying. Results indicated improvements in couples’ relationship
quality and significant decreases in patients’ distress about dying and the frequency
of partners’ worry about their partner dying.
Given the direct adverse impact of couple distress on cardiovascular and im-
munological functioning (Kiecolt-Glaser & Newton 2001)—and the indirect neg-
ative effects on health through couple influences on nutrition, substance use, and
exercise—it is not surprising that couple-based interventions have been developed
for a range of health-related concerns including obesity and nicotine use as well as
alcohol- and drug-use disorders as described above. Common to such treatments
are specific interventions aimed at changing patients’ interpersonal environments
linked to health-risk behaviors, encouraging healthy alternatives through partners’
use of social reinforcement, and generally decreasing relationship stress (see also
Schmaling & Sher 2000). Although evidence for some of these treatments emerged
almost 25 years ago (e.g., spouse involvement in the behavioral treatment of over-
weight women; Pearce et al. 1981), other promising developments (e.g., couple-
based interventions for patients diagnosed with coronary artery disease; Sher et al.
2002) await further empirical evaluation.
Couple-based treatments conti-
nue to be developed at an increasing pace for a broad spectrum of individual
emotional, behavioral, and health-related difficulties. Although evidence for these
treatments’ effectiveness remains primarily anecdotal, several noteworthy ex-
ceptions exist. Monson et al. (2004) described recent findings from cognitive-
behavioral couple treatment of posttraumatic stress disorder (PTSD) in a pilot
study of seven couples in which the husband was diagnosed with PTSD secondary
to Vietnam combat experiences. The treatment emphasized three components in-
volving psychoeducation about PTSD and relationship problems, communication
skills training, and interventions targeting cognitions contributing to the associa-
tion between PTSD and relationship problems. Following the 15-session treatment,
clinicians’ and partners’ ratings of the veterans’ PTSD symptoms showed signif-
icant improvement, with effect sizes exceeding 1.00; the veterans’ self-reported
reductions in PTSD symptoms were not statistically significant (in part due to the
small sample), but still yielded a moderate effect size of 0.64. Wives’ own ratings
of anxiety also improved with this couple-based intervention, and both partners
reported improved social functioning in the household.
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Finally, two couple-based treatments have been developed for couples in which
one partner has been diagnosed with borderline personality disorder (BPD). The
first (Fruzzetti & Iverson 2006) builds on dialectical behavior therapy (DBT) for
individuals with BPD. In a sample of 22 couples participating in a six-session
couples group, decreases in invalidating behaviors (e.g., dismissive, minimizing,
or rejecting statements) and increases in validating responses (e.g., acceptance
and understanding) pre- to posttreatment predicted decreased levels of individual
and relationship distress in a moderately distressed sample (Lillis & Fruzzetti
2004). More recently, Kirby & Baucom (2004) reported results of a couple-based
group intervention combining elements of DBT with cognitive behavioral couple
therapy (CBCT) for 10 couples in which one partner had been diagnosed with BPD
and had already received individual DBT. Following 16 two-hour sessions (with
five couples per group), the women partners showed less depression and related
negative affect, increases in positive affect, and improved ability to regulate their
own emotions; effect sizes ranged from 0.72 to 1.00.
How does couple therapy work? Although each of the empirically supported ap-
proaches to couple therapy posits specific processes or mechanisms of change,
there has been little research explicitly indicating these proposed mechanisms as
responsible for observed therapeutic effects. In this section, we briefly describe
three approaches to examining change processes in couple therapy. We then sum-
marize the few available empirical findings related to change mechanisms in the
context of theoretical formulations underlying the major approaches to couple
Methods of Investigating Change Processes
REGRESSION ANALYSIS OF MEDIATION Probably the best-known and most widely
used approach for examining change processes in therapy involves use of regres-
sion analysis, following procedures outlined by Baron & Kenny (1986), for estab-
lishing mediation effects. Using these guidelines, a proposed mediating variable
(e.g., communication processes) is shown to account (either entirely or partially)
for the relation between some predictor variable (e.g., treatment condition status)
and some outcome variable (e.g., relationship satisfaction) when the following
four conditions are met: (a) The predictor affects the criterion (e.g., treatment
leads to increased relationship satisfaction); (b) the predictor affects the mediator
(e.g., treatment leads to gains in communication skills); (c) the mediator affects
the criterion (e.g., gains in communication skills lead to increased relationship sat-
isfaction), controlling for the predictor; and (d ) the relation between the predictor
and the criterion is reduced (partial mediation) or eliminated (complete media-
tion) after controlling for the relation between the mediator and the criterion. In
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
“moderated mediation” (see Whisman & Snyder 1997), the mediating or change
mechanisms are demonstrated to have a stronger effect for one group than for
another (e.g., mediating effects of therapeutic alliance for couples receiving BCT
versus EFCT, or effects of behavior-exchange skills training for younger couples
relative to older ones).
More recent-
ly, couple researchers (e.g., Doss et al. 2005) have examined the relation between
proposed mechanisms of change and outcome variables through the use of hierar-
chical linear modeling (HLM; Raudenbush & Bryk 2001), also known as growth
curve analysis. This approach to identifying change mechanisms involves two
steps. First, multiple assessments of some criterion variable are used to estimate
a trajectory, or growth curve, which allows investigators to describe the nature
of change for a given criterion or outcome within a sample. In the second step
of growth curve analysis, the parameters summarizing change of each person are
treated as new dependent variables that are then predicted from other within- or
between-subject variables proposed as mechanisms of change.
A third approach to investigating change
processes involves task analysis of proximal outcomes that occur within or be-
tween sessions by focusing on specific therapeutic events (Heatherington et al.
2005, Rice & Greenberg 1984). Such task analysis involves disassembling the
therapeutic process into smaller, measurable in-session units or events to capture
the actual sequence of therapist-client interactions and then delineating the link-
age of these events to proximal or “mini” outcomes that presumably build on each
other and contribute to molar, more distal outcomes (e.g., relationship satisfaction
at termination). For example, an investigator examining presumed mechanisms
of change in EFCT might examine events delimited by therapist clarification of
underlying primary attachment-related fears and the client’s owning and express-
ing those fears in a “softened” manner, and the linkage of this sequence to the
other partner’s likelihood of shifting from antagonistic or defensive responses to
empathic or nurturing ones.
Empirical Findings Regarding Change Processes
Previous reviews (e.g., Gottman 1998, Lebow 2000) have generally concurred in
their conclusion that little empirical evidence exists regarding presumed mecha-
nisms of change in couple therapy. Such conclusions rest in part from disappointing
findings from mediation analyses adopting the traditional regression approach. For
example, whereas BCT emphasizes the importance of improving communication
skills as a means of reducing relationship distress—and although such skills do
typically increase among couples receiving BCT—several studies have failed to
find an association between the magnitude of changes in communication behav-
iors and gains in relationship satisfaction. Similarly, although CBCT has been
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
shown to produce positive change in targeted cognitions (e.g., expectancies and
attributions), changes in these cognitions have not been linked to couples’ gains
in satisfaction following CBCT (see Whisman & Snyder 1997 for a summary of
relevant studies).
However, recent findings drawing on HLM and task analysis offer encourage-
ment regarding significant developments in identifying change mechanisms in
couple therapy. For example, Doss et al. (2005) used hierarchical growth curve
analysis to examine mechanisms of change in 134 couples randomly assigned to
either traditional BCT or ICBT. Both therapies were effective in increasing emo-
tional acceptance and improving communication behaviors across the course of
therapy; however, these changes differed by treatment modality in a manner consis-
tent with their respective presumed change mechanisms. Specifically, acceptance
increased significantly more for couples in IBCT than for couples in BCT, whereas
couples in BCT showed larger gains in positive communication. Moreover, exam-
ination of change separately in the first and second halves of therapy indicated
that change in targeted behaviors was a powerful mechanism of change early in
therapy, whereas in the second half of therapy, emotional acceptance was more
strongly related to changes in relationship satisfaction.
Task analysis has been used successfully to examine change processes in both
IBCT and EFCT. In a pilot study with 12 distressed couples randomly assigned to
either IBCT or traditional BCT (Cordova et al. 1998), couples in IBCT showed rela-
tively more constructive detachment (i.e., talking about problems without blaming
or being compelled to solve them—both indicators of acceptance) over the course
of therapy and more “soft expressions” of emotion in late sessions relative to ear-
lier ones. Changes in husbands’ and wives’ constructive detachment from early to
late sessions predicted couples’ gains in relationship satisfaction.
The most compelling findings regarding specific change processes in couple
therapy have emerged using a task-analysis approach to investigating change in
EFCT. An early study by Johnson & Greenberg (1988) comparing partner ex-
changes in “best” sessions of three successfully treated couples versus those for
three couples with poor outcome showed that high-change couples exhibited more
frequent “softening” events in which a previously critical partner expressed vul-
nerability and asked for comfort and connection from his or her spouse. A second
report regarding three task analytic studies of EFCT (Greenberg et al. 1993) showed
that (a) couples receiving EFCT demonstrated more shifts from hostility to affilia-
tive behaviors than did wait-list couples; (b) best sessions as identified by couples
were characterized by more depth of experiencing and affiliative and autonomous
statements than were sessions identified as poor; and (c) intimate, emotionally
laden self-disclosure by one partner was more likely to lead to affiliative state-
ments by the other partner than were other randomly selected responses. Finally,
a recent task analysis of four EFCT sessions by Bradley & Furrow (2004) found
that emotional experiencing and the disclosure of attachment-related affect and
fears were the key client features of successful softening events; consistent with
proposed mechanisms of change in EFCT, specific therapist interventions linked
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
to softening events involved intensifying a couple’s emotional experience and
promoting intrapsychic awareness and interpersonal shifts in attachment-related
For whom does couple therapy work? As documented in the preceding sections,
there is considerable variability in individuals’ response to couple therapy. Hence,
investigators have been interested in predicting outcome to treatment. In this sec-
tion, we briefly describe methods for identifying predictors of treatment response,
distinguishing between methods emphasizing prognostic versus prescriptive indi-
cators. We then review empirical findings regarding predictors of couple therapy
Methods of Identifying Predictors of Treatment Outcome
evaluating predictors of treatment outcome, investigators have made a distinction
between prognostic indicators, which predict response to a particular treatment
(or response across treatments, irrespective of specific approach), and prescrip-
tive indicators, which predict response to one versus another treatment (Hollon
& Najavits 1988). Identifying a prognostic indicator requires only evaluating the
association between the predictor variable and some outcome measure. This can
be done by regressing posttreatment outcome scores on the predictor, controlling
for pretreatment scores on the outcome variable; similar analyses can be done
with dichotomous outcomes (e.g., clinical significance outcomes) using logistic
regression analyses. The overwhelming preponderance of research on predictors
of response to couple therapy has emphasized the delineation of these more general
prognostic indicators.
uating whether some variable predicts outcome to one specific treatment versus
another treatment—that is, identifying prescriptive indicators—requires testing for
an interaction or moderator effect. Prescriptive indicators are examined using the
aptitude-treatment interaction (ATI) paradigm (Cronbach & Snow 1977, Dance
& Neufeld 1988). In using regression to identify prescriptive indicators, one tests
for a significant treatment × predictor interaction indicating that the association
between the predictor and treatment outcome varies as a function of the type of
treatment (i.e., that the type of treatment moderates the association between the
predictor and outcome). Alternative approaches for identifying prescriptive indi-
cators exist when using analysis of individual growth curves (Rogosa 1991).
Research design requirements for identifying prescriptive indicators are more
rigorous than those for identifying more general prognostic indicators, particularly
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
as these relate to adequate power for detecting effects (Whisman & McClelland
2005). For example, using Cohen’s (1987) power tables, an investigator would
need to obtain minimum sample sizes of 26, 55, or 392 participants in order to
have adequate power (i.e., power of .80, at α = .05) for detecting large, medium,
or small effect sizes, respectively. However, prescriptive indicators generally of-
fer far greater usefulness for clinicians than do prognostic indicators in that the
former go beyond the question of who responds well to therapy to address is-
sues of treatment selection in evaluating who responds well to which kinds of
Empirical Findings Regarding Predictors
of Couple Therapy Outcome
cades, a substantial body of research has identified general prognostic indicators
of response to couple therapy including demographic, relationship, and individual
characteristics. Most of these findings have been derived from controlled clini-
cal trials of BCT and are reviewed in greater detail by Whisman et al. (2005);
exceptions are noted where applicable.
Several studies have found that younger couples respond more favorably to
BCT (Baucom 1984, Hahlweg et al. 1984, O’Leary & Turkewitz 1981), whereas
others have found no association between age and treatment outcome (Crowe
1978, Jacobson et al. 1986). In addition, Crowe (1978) found that less-educated
couples had better response to BCT than those with higher education. A predic-
tion study collapsing across behavioral and insight-oriented treatment conditions
(Snyder et al. 1993) found that initial status of being unemployed or employed
in a position of unskilled labor predicted poor treatment outcome four years
after termination. In a controlled trial of IBCT versus traditional BCT, couples
who were married longer showed greater treatment gains, regardless of condition
(Atkins 2005a).
Results from various studies indicate that couples having the greatest difficul-
ties in their relationship are less likely to benefit from treatment, with initial levels
of relationship distress accounting for up to 46% of the variance in treatment out-
come (Johnson 2002). Lack of commitment and behavioral steps taken toward
divorce have been associated with poor treatment outcome to BCT in two studies
(Beach & Broderick 1983, Hahlweg et al. 1984) but not in another (Jacobson et al.
1986). Hahlweg et al. (1984) found that BCT outcome was predicted by nega-
tive communication behavior. Snyder et al. (1993) found that poorer outcome to
couple therapy was predicted by lower relationship quality, greater negative rela-
tionship affect and disengagement, and greater desired change in the relationship.
By contrast, initial levels of relationship distress were not significantly related
to treatment outcome in a study of EFCT (Johnson & Talitman 1997), although
partners’ therapeutic alliance accounted for 22% of the variance in response to
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Whisman & Jacobson (1990) operationalized inequality of partners’ power in
their marriage in terms of asymmetry in the relative frequencies of verbal commu-
nication content patterns, and found that power inequality prior to therapy predicted
positive treatment outcome at posttest and at six-month follow-up. Gray-Little et al.
(1996) operationalized power in terms of which partner had more influence in a
problem-solving interaction, and found that wife-dominant couples improved the
most in response to couple therapy in terms of increased satisfaction and improved
communication. In a study of EFCT, Johnson & Talitman (1997) found that the
best predictor of outcome was the wife’s belief that her partner still cared for her.
Greater interpersonal sensitivity and emotional expressiveness—as determined
by measures of “femininity”—have been found to predict better outcome at ter-
mination (Baucom & Aiken 1984) and long-term follow-up (Snyder et al. 1993)
but were not predictive in a third study (Jacobson et al. 1986). Couples in which
partners exhibit a higher degree of traditionality (i.e., higher affiliation needs in
the wife and higher independence needs in the husband) have been shown to have
poorer response to BCT (Jacobson et al. 1986). Partners’ higher levels of depressed
affect have been linked to poorer outcome in one study (Snyder et al. 1993) but
not in another (Jacobson et al. 1986).
Although findings regarding prognostic indicators of couple treatment response
are mixed and the predictive utility of any single predictor appears modest, incre-
mental prediction from multiple indicators pooled across predictor domains can
be substantial. For example, in the study comparing BCT with IOCT (Snyder
et al. 1991), the unconditional probability (base rate) of divorce or relationship
distress four years after completing couple therapy was .35. In their analyses
pooling replicated predictors across partners and across demographic, individual,
and relationship domains, Snyder et al. (1993) were able to double the accu-
racy in predicting four-year follow-up status from prognostic indicators obtained
either at intake or termination (with conditional probabilities of .71 and .86,
garding general prognostic indicators of response to couple therapy, research iden-
tifying prescriptive indicators of couple treatment response has been rare. An early
study by O’Leary & Turkewitz (1981) suggested that younger couples responded
better to behavioral interventions emphasizing behavior-exchange skills, whereas
older couples showed more favorable response to general communication skills
training. More recently, research comparing IBCT with traditional BCT suggests
that severely distressed couples may respond more favorably to BCT than to IBCT
during the initial stages of treatment, although both treatments produce equivalent
gains at outcome and preliminary findings indicate that IBCT may produce more
enduring gains at extended follow-up (Atkins & Christensen 2004). Moreover,
exploratory analyses from this clinical trial reported by Atkins et al. (2005a) sug-
gested that sexually dissatisfied couples showed slower initial response but more
consistent gains overall in IBCT versus BCT.
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Previous reviews of couple therapy have identified a variety of directions for further
research and training in couple-based treatments. Based on these reviews and our
own evaluation of the literature, we have extracted 12 essential directions for future
research and training.
Directions for Research
Couple therapy outcome research will benefit from smaller-level studies such
as single- or replicated-case designs, analysis of treatment components, and
open clinical trials. We adopt this conclusion without alteration from a review of
methodological issues in couple research by Christensen et al. (2005). Christensen
and colleagues noted limitations in funding for large clinical trials of couple
therapy, and emphasized that “working with a smaller number of couples in a
more detailed manner often can provide the understanding and insight needed be-
fore launching a more time-consuming and expensive randomized clinical trial”
(Christensen et al. 2005, p. 13). Such smaller-scale investigations also promote
outcome research in community agency and private practice settings potentially
addressing issues of representativeness remaining from studies conducted exclu-
sively within the university research context.
Couple therapy research needs to extend beyond initial treatment impact to
identify individual, relationship, and treatment factors contributing to deterio-
ration or relapse and effective means of reducing or eliminating these effects.
Among those individuals who initially respond favorably to couple therapy, ap-
proximately 30%–60% subsequently evidence significant deterioration. Numerous
reviews (e.g., Johnson 2002, Lebow & Gurman 1995) have noted the need to de-
velop specific interventions targeting relapse as well as guidelines regarding their
delivery (e.g., timing and format of their delivery as well as criteria for targeting
couples at greatest risk) to reduce deterioration effects.
Research on couple therapy needs to move beyond existing therapies to examine
integrative approaches—including indicators for selecting, sequencing, and pac-
ing specific treatment components, alternative integrative models, and moderators
of therapeutic effectiveness. This conclusion was asserted by Snyder & Whisman
(2003, 2004b) specifically as it relates to treating couples with coexisting mental
and physical disorders, but has also been voiced by others (e.g., Lebow & Gurman
1995) as it applies to couple therapy more generally. Efforts to decompose couple-
based interventions into their smallest transportable components should lead to
research on the most effective ways of reassembling these in a manner uniquely
tailored to couples’ variation in individual and relationship functioning (Snyder
et al. 2003). Each intervention incorporated into an integrative approach needs to
be considered with respect to its necessity, sufficiency, and interactive effects.
Couple-based interventions for specific individual and relationship problems
need to be developed and examined for both their intermediate and long-term
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
effectiveness. Significant progress in this regard has been achieved over the past
decade; however, much of the literature espousing couple-based interventions for
emotional and behavioral disorders relies on qualitative analyses or anecdotal evi-
dence. Evaluations of existing or new couple therapies for specific disorders need
to be complemented by studies examining the adaptation of existing approaches to
couple distress when specific dysfunctions (e.g., personality disorders)—although
not an explicit target of treatment—moderate both treatment process and outcome.
Greater attention needs to be focused on the generalizability of research find-
ings across such potential moderators as age, family life stage, gender, culture
and ethnicity (including interethnic couples), family structure (including composi-
tion of stepfamily and extended family systems), and nontraditional relationships
(including cohabiting and same-gender couples). Virtually every recent review
of couple therapy research has decried the lack of findings regarding the gener-
alizability of treatment outcome and processes across these or similar potential
moderators. Among the clinical trials of couple therapy reviewed in the preceding
sections, only the comparison of IBCT with traditional BCT (Christensen et al.
2004) included significant representation (22%) from ethnically diverse groups.
Cross-cultural comparisons of couple therapy—particularly couple-based treat-
ments of individual emotional or health problems—are rare, despite documented
differences in how couples of diverse cultural backgrounds contend with mental
and physical illness (Osterman et al. 2003). Research on couple therapy for specific
disorders tends to focus on one gender to the exclusion of the other—e.g., sampling
men with substance use disorders or women with affective disorders. Empirical
studies of couple therapy with same-gender couples are virtually nonexistent.
Because of the growing concerns about spiraling health care costs, research
needs to assess the costs, benefits, cost-benefit ratio, and cost-effectiveness of
couple-based interventions. We embrace this conclusion verbatim as asserted by
Fals-Stewart et al. (2005b) in a recent article defining various components of cost
analysis and methods for calculating these indices. Fals-Stewart and colleagues de-
termined that the few evaluations of cost-benefit and cost-effectiveness of couple-
and family-based interventions completed to date have yielded favorable results;
nevertheless, they noted that evaluations of this sort for couple and family therapies
have lagged behind those for other psychosocial interventions.
Studies of couple therapy outcome need to be complemented by research on
change processes. Again, this recommendation regarding future research has been
articulated in virtually every review of couple therapy appearing in the past decade.
Further bolstering the appeal for research on mechanisms of change are encour-
aging findings from recent studies drawing on HLM and especially those using an
events-based or task analysis approach. In their recent review of change process
research in couple and family therapy, Heatherington et al. (2005) identify five crit-
ical foci toward which future process research should be oriented: (a) midrange
theories about systemic change processes; (b) client change processes (e.g., partner
behaviors facilitating proximal outcomes); (c) intrapersonal processes (e.g., emo-
tion and cognition); (d ) strategies for analyzing data from multiple participants;
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
and (e) consistent with a recommendation noted above, investigation of the degree
to which various change processes generalize across diverse populations.
Research regarding mediators and moderators of treatment outcome requires
attention to critical design issues to ensure the potential for identifying relevant ef-
fects. Recent papers have emphasized methodological issues in examining change
processes in psychotherapy generally (e.g., Doss 2004) and moderators in couple
and family research specifically (Whisman & McClelland 2005). For example, the
power to detect small or moderate effects may be enhanced not only by increasing
sample size but also by using more reliable and accurate measures and by increas-
ing variance of the predictor or moderator variable. The latter point is particularly
important because inclusion and exclusion criteria for clinical trials often result
in restricted range of predictor variables. For example, one is prohibited from ex-
amining the effects of individual psychopathology on couple therapy process or
outcome if individuals with emotional or behavioral disorders are excluded from
the clinical trial.
Couple-based interventions likely will be enhanced by incorporating basic re-
search on emotion regulation processes. There has been growing interest over the
past decade in the role of emotion regulation processes in couples and families
(Snyder et al. 2006). Individuals’ ability to regulate their emotions effectively—
especially in interpersonal contexts that involve potentially caustic exchanges—
plays a pivotal role in keeping individuals and their significant relationships
functioning well. Poor or inadequate emotion regulation at either the intrapersonal
or the interpersonal levels may be a major contributing factor in relationship dissat-
isfaction and dissolution. Moreover, research on emotion regulation may facilitate
better understanding of the role of individual psychopathology on couple therapy
processes and outcomes.
Directions for Clinical Training
Couple therapists should be trained in common factors and mechanisms of change
that potentially undergird most forms of successful treatment. We concur with this
conclusion as asserted by Sprenkle & Blow (2004) while also recognizing caveats
noted by Sexton et al. (2004) that the common factors perspective may “over-
look the multilevel nature of practice, the diversity of clients and settings, and the
complexity of therapeutic change” (p. 131). Recognizing the therapeutic effects
of nonspecific (common) treatment components does not obviate attending to the
critical role of specific treatment counterparts (Snyder et al. 1988). For example,
fundamental but nonspecific interventions facilitating the therapeutic alliance may
be necessary but insufficient for treating relationship distress unless followed by
specific interventions such as challenging dysfunctional attributions, emphasizing
and heightening primary emotions, or interpreting recurring maladaptive relation-
ship patterns.
Couple therapists need to be trained to conceptualize and practice inte-
gratively across diverse theoretical orientations. The complexity of individual,
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
interpersonal, and situational factors contributing to the development, exacer-
bation, or maintenance of couple distress often requires selecting, sequencing,
and pacing multiple interventions falling outside of any one theoretical approach
(Snyder et al. 2003). Therapists are often vulnerable to viewing presenting com-
plaints through the filtering lens of their own preferred theoretical or treatment
modality. Treating difficult couples with coexisting mental or physical health prob-
lems, in particular, may be hindered if interventions are restricted to one particular
theoretical tradition. Even when practicing within a given treatment approach,
couple therapists need to be trained in how to “make their next move” (Sprenkle
2002a) in terms of selecting among either specific or nonspecific interventions
consistent with that approach.
Couple therapists need to be competent in recognizing and treating the recur-
sive influences of individual and relationship difficulties. Given robust findings
regarding the comorbidity of relationship distress with individual emotional and
behavioral problems in both community and clinical samples, couple therapists
must be schooled in psychopathology and principles of individual assessment and
treatment, including familiarity with biological interventions for relationship dif-
ficulties rooted at least in part in physical or mental illness of one or both partners.
Similarly, couple therapists need to be familiar with both existing and emerging
couple-based interventions for individual emotional and physical health problems,
as well as adaptations of existing treatments made necessary by such difficulties.
As evidence of the disconnection between research findings and clinical prac-
tice, Fals-Stewart & Birchler (2001) surveyed program administrators from 398
community-based outpatient substance abuse treatment programs in the United
States regarding use of different family- and couple-based therapies in their pro-
grams. Whereas 27% of the programs provided some type of couple-based treat-
ment, less than 5% of the agencies used behaviorally oriented couple therapy and
none used BCT specifically. Consequently, greater efforts are needed to identify
possible barriers impeding the transfer of couple-based interventions from research
to practice settings and to develop strategies aimed at reducing or eliminating these
Couple therapy comprises an essential component of mental health services. Re-
search demonstrates its effectiveness in treating generalized relationship distress
as well as comorbid relationship problems and individual emotional and behav-
ioral difficulties. Systematic investigations delineating processes of change and
prescriptive indicators of treatment response will be critical to narrow the oft-cited
gap between clinical research and practice. Recent findings offer considerable en-
couragement for translating the results of couple therapy research into improved
training of couple therapists and more effective interventions in community agency
and practice settings.
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
The Annual Review of Psychology is online at
Allen ES, Atkins DC, Baucom DH, Snyder DK,
Gordon KC, et al. 2005. Intrapersonal, inter-
personal, and contextual factors in engaging
in and responding to extramarital involve-
ment. Clin. Psychol. Sci. Pract. 12:101–30
Atkins DC, Berns SB, George W, Doss B, Gat-
tis K, et al. 2005a. Predicting success in a
randomized clinical trial of marital therapy.
J. Consult. Clin. Psychol. In press
Atkins DC, Christensen A. 2004. Marital Sat-
isfaction and Stability for Two Years Post-
Treatment: Comparing the Trajectories of
TBCT and IBCT. Presented at Assoc. Adv.
Behav. Ther., New Orleans, LA
Atkins DC, Eldridge KE, Baucom DH, Chris-
tensen A. 2005b. Behavioral marital therapy
and infidelity: optimism in the face of be-
trayal. J. Consult. Clin. Psychol. 73:144–50
Baron RM, Kenny DA. 1986. The moderator-
mediator variable distinction in social psy-
chological research: conceptual, strategic,
and statistical considerations. J. Personal.
Soc. Psychol. 51:1173–82
Baucom DH. 1984. The active ingredients of
behavioral marital therapy: the effectiveness
of problem-solving/communication training,
contingency contracting, and their combina-
tion. See Hahlweg & Jacobson 1984, pp. 73–
Baucom DH, Aiken PA. 1984. Sex role identity,
marital satisfaction, and response to behav-
ioral marital therapy. J. Consult. Clin. Psy-
chol. 52:438–44
Baucom DH, Shoham V, Mueser KT, Daiuto
AD, Stickle TR. 1998. Empirically supported
couple and family interventions for marital
distress and adult mental health problems. J.
Consult. Clin. Psychol. 66:53–88
Baucom DH, Stanton S, Epstein NB. 2003.
Anxiety disorders. See Snyder & Whisman
2003, pp. 57–87
Beach SRH, Broderick JE. 1983. Commitment:
avariable in women’s response to marital
therapy. Am. J. Fam. Ther. 11:16–24
Beach SRH, Gupta M. 2003. Depression. See
Snyder & Whisman 2003, pp. 88–113
Bradley B, Furrow JL. 2004. Toward a mini-
theory of the blamer softening event: tracking
the moment-by-moment process. J. Marital
Fam. Ther. 30:1–12
Byrne M, Carr A, Clark M. 2004. The efficacy
of couples-based interventions for panic dis-
order with agoraphobia. J. Fam. Ther. 26:
Chambless DL, Hollon SD. 1998. Defining
empirically supported therapies. J. Consult.
Clin. Psychol. 66:7–18
Christensen A, Atkins DC, Berns S, Wheeler J,
Baucom DH, et al. 2004. Traditional versus
integrative behavioral couple therapy for sig-
nificantly and chronically distressed married
couples. J. Consult. Clin. Psychol. 72:176–
Christensen A, Baucom DH, Vu CA, Stan-
ton S. 2005. Methodologically sound, cost-
effective research on the outcome of couple
therapy. J. Fam. Psychol. 19:6–17
Christensen A, Heavy CL. 1999. Interventions
for couples. Annu. Rev. Psychol. 50:165–
Cohen J. 1987. Statistical Power Analysis for
the Behavioral Sciences. Hillsdale, NJ: Erl-
baum. 474 pp. 2nd ed.
Cookerly JR. 1980. Does marital therapy do any
lasting good? J. Marital Fam. Ther. 6:393–
Cordova JV, Jacobson NS, Christensen A.
1998. Acceptance versus change interven-
tions in behavioral couple therapy: impact on
couples’ in-session communication. J. Mar-
ital Fam. Ther. 24:437–55
Cronbach LJ, Snow RE. 1977. Aptitudes and
Instructional Methods: A Handbook for Re-
search on Interactions. Oxford, UK: Irving-
ton. 574 pp.
Crowe MJ. 1978. Conjoint marital therapy:
a controlled outcome study. Psychol. Med.
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Dance KA, Neufeld RWJ. 1988. Aptitude-
treatment interaction research in the clini-
cal setting: a review of attempts to dispel
the “patient uniformity” myth. Psychol. Bull.
Dessaulles A, Johnson SM, Denton WH. 2003.
Emotion-focused therapy for couples in the
treatment of depression: a pilot study. Am. J.
Fam. Ther. 31:345–53
Doss BD. 2004. Changing the way we study
change in psychotherapy. Clin. Psychol. Sci.
Pract. 11:368–86
Doss BD, Thum YM, Sevier M, Atkins DC,
Christensen A. 2005. Improving relation-
ships: mechanisms of change in couple ther-
apy. J. Consult. Clin. Psychol. In press
Dunn RL, Schwebel AI. 1995. Meta-analytic
review of marital therapy outcome research.
J. Fam. Psychol. 9:58–68
Dutcher TD. 1999. A meta-analytic study of
marital therapy modalities and clients’ pre-
senting problems. PhD thesis. Union Inst. 73
Fals-Stewart W, Birchler GR. 2001. A national
survey of the use of couples therapy in sub-
stance abuse treatment. J. Subst. Abuse Treat.
Fals-Stewart W, Birchler GR, O’Farrell TJ.
2003. Alcohol and other substance abuse. See
Snyder & Whisman 2003, pp. 159–80
Fals-Stewart W, Klostermann K, Yates BT,
O’Farrell TJ, Birchler GR. 2005a. Brief re-
lationship therapy for alcoholism: a random-
ized clinical trial examining clinical efficacy
and cost-effectiveness. Psychol. Addict. Be-
hav. In press
Fals-Stewart W, Yates BT, Klostermann K.
2005b. Assessing the costs, benefits, cost-
benefit ratio, and cost-effectiveness of mari-
tal and family treatments: why we should and
how we can. J. Fam. Psychol. 19:28–39
Foley SH, Rounsaville BJ, Weissman MM,
Sholomaskas D, Chevron E. 1989. Individ-
ual versus conjoint interpersonal psychother-
apy for depressed patients with marital
disputes. Int. J. Fam. Psychiatry 10:29–
Fruzzetti AE, Iverson KM. 2006. Interven-
ing with couples and families to treat emo-
tion dysregulation and psychopathology. See
Snyder et al. 2006. In press
Goldman A, Greenberg L. 1992. Comparison
of integrated systemic and emotionally fo-
cused approaches to couples therapy. J. Con-
sult. Clin. Psychol. 60:962–69
Gordon KC, Baucom DH, Snyder DK. 2004.
An integrative intervention for promoting re-
covery from extramarital affairs. J. Marital
Fam. Ther. 30:213–31
Gottman JM. 1998. Psychology and the study
of marital processes. Annu. Rev. Psychol. 49:
Gottman JM. 1999. The Marriage Clinic: A
Scientifically Based Marital Therapy.New
York: Norton. 456 pp.
Gray-Little B, Baucom DH, Hamby SL. 1996.
Marital power, marital adjustment and ther-
apy outcome. J. Fam. Psychol. 10:292–303
Greenberg LS, Ford CL, Alden L, Johnson SM.
1993. In-session change in emotionally fo-
cused therapy. J. Consult. Clin. Psychol. 61:
Gupta M, Coyne JC, Beach SRH. 2003. Cou-
ples treatment for major depression: critique
of the literature and suggestions for some dif-
ferent directions. J. Fam. Ther. 25:317–46
Gurman AS. 1981. Integrative marital therapy:
toward the development of an interpersonal
approach. In Forms of Brief Therapy, ed. SH
Budman, pp. 415–57. New York: Guilford.
482 pp.
Gurman AS. 2002. Brief integrative marital
therapy: a depth-behavioral approach. In
Clinical Handbook of Couple Therapy, ed.
AS Gurman, NS Jacobson, pp. 180–220.
New York: Guilford. 731 pp. 3rd ed.
Hahlweg K, Jacobson NS, ed. 1984. Marital
Interaction: Analysis and Modification.New
York: Guilford. 450 pp.
Hahlweg K, Markman HJ. 1988. Effectiveness
of behavioral marital therapy: empirical sta-
tus of behavioral techniques in preventing
and alleviating marital distress. J. Consult.
Clin. Psychol. 56:440–47
Hahlweg K, Schindler L, Revenstorf D, Bren-
gelmann JC. 1984. The Munich marital
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
therapy study. See Hahlweg & Jacobson
1984, pp. 3–26
Heatherington L, Friedlander ML, Greenberg
L. 2005. Change process research in cou-
ple and family therapy: methodological chal-
lenges and opportunities. J. Fam. Psychol.
Hollon SD, Najavits L. 1988. Review of empiri-
cal studies on cognitive therapy. In American
Psychiatric Press Review of Psychiatry, ed.
AJ Frances, RE Hales, 7:643–66. Washing-
ton, DC: Am. Psychiatr. Press. 720 pp.
Holtzworth-Munroe A, Marshall AD, Meehan
JC, Rehman U. 2003. Physical aggression.
See Snyder & Whisman 2003, pp. 201–30
Hurlbert DF, White LC, Powell RD, Apt C.
1993. Orgasm consistency training in the
treatment of women reporting hypoactive
sexual desire: an outcome comparison of
women-only groups and couples-only
groups. J. Behav. Ther. Exp. Psychiatry. 24:
Jacobson NS, Christensen A. 1996. Integrative
Couple Therapy: Promoting Acceptance and
Change.New York: Norton. 283 pp.
Jacobson NS, Follette WC, Pagel M. 1986. Pre-
dicting who will benefit from behavioral mar-
ital therapy. J. Consult. Clin. Psychol. 54:
Jacobson NS, Schmaling KB, Holtzworth-
Munroe A. 1987. Component analysis of be-
havioral marital therapy: 2-year follow-up
and prediction of relapse. J. Marital Fam.
Ther. 13:187–95
Johnson SM. 2002. Marital problems. See
Sprenkle 2002b, pp. 163–90
Johnson SM, Greenberg LS. 1988. Relating
process to outcome in marital therapy. J.
Marital Fam. Ther. 14:175–83
Johnson SM, Hunsley J, Greenberg L, Schindler
D. 1999. Emotionally focused couples ther-
apy: status and challenges. Clin. Psychol. Sci.
Pract. 6:67–79
Johnson SM, Talitman E. 1997. Predictors of
success in emotionally focused marital ther-
apy. J. Marital Fam. Ther. 23:135–52
Keefe FJ, Ahles TA, Sutton L, Dalton J, Bau-
com DH, et al. 2005. Partner-guided cancer
pain management at end of life: a preliminary
study. J. Pain Symptom Manage. 29:263–
Keefe FJ, Caldwell DS, Baucom DH, Salley
A, Robinson E, et al. 1996. Spouse-assisted
coping skills training in the management of
osteoarthritis knee pain. Arthritis Care Res.
Keefe FJ, Porter LS, Labban J. 2006. Emotion
regulation processes in disease-related pain:
a couples-based perspective. See Snyder et al.
2006. In press
Kessler RC, McGonagle KA, Zhao S, Nelson
CB, Hughes M, et al. 1994. Lifetime and 12-
month prevalence of DSM-III-R psychiatric
disorders in the United States: results from
the National Comorbidity Survey. Arch. Gen.
Psychiatry 51:8–19
Kiecolt-Glaser JK, Newton TL. 2001. Mar-
riage and health: his and hers. Psychol. Bull.
Kirby JS, Baucom DH. 2004. Treating Emo-
tional Dysregulation in a Couples Context:
A Pilot Study of a Couples Skills Group In-
tervention. Presented at Assoc. Adv. Behav.
Ther., New Orleans, LA
Kreider RM, Fields JM. 2002. Number, tim-
ing, and duration of marriages and divorces:
1996. Current Pop. Rep. P70–80.Washing-
ton, DC: US Census Bur.
Laumann EO, Paik A, Rosen RC. 1999. Sexual
dysfunction in the United States: prevalence
and predictors. J. Am. Med. Assoc. 281:537–
Lebow JL. 2000. What does the research tell us
about couple and family therapies? J. Clin.
Psychol. 56:1083–94
Lebow JL, Gurman AS. 1995. Research assess-
ing couple and family therapy. Annu. Rev.
Psychol. 46:27–57
LeffJ, Vearnals S, Brewin CR, Wolff G, Alexan-
der B, et al. 2000. The London Depression In-
tervention Trial: randomised controlled trial
of antidepressants v. couple therapy in the
treatment and maintenance of people with
depression living with a partner: clinical out-
come and costs. Br. J. Psychiatry 177:95–
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
Lillis EA, Fruzzetti AE. 2004. DBT for Cou-
ples: A Brief Intervention for the Treatment
of Couple and Individual Distress. Presented
at Assoc. Adv. Behav. Ther., New Orleans,
Lin E, Goering P, Offord DR, Campbell D,
Boyle MH. 1996. The use of mental health
services in Ontario: epidemiologic findings.
Can. J. Psychiatry 41:572–77
Mohr DC, Moran PJ, Kohn C, Hart S, Arm-
strong K, et al. 2003. Couple therapy at the
end of life. Psychooncology 12:620–27
Monson CM, Stevens SP, Schnur PP. 2004.
Cognitive-behavioral couple’s treatment for
posttraumatic stress disorder. In Focus on
Posttraumatic Stress Disorder Research, ed.
TA Corales. pp. 251–80. Hauppauge, NY:
Nova Sci. 290 pp.
Murphy CM, O’Leary KD. 1989. Psycholog-
ical aggression predicts physical aggression
in early marriage. J. Consult. Clin. Psychol.
O’Farrell TJ, Fals-Stewart W. 2000. Behav-
ioral couples therapy for alcoholism and drug
abuse. J. Subst. Abuse Treat. 18:51–54
O’Leary KD, Turkewitz H. 1981. A compara-
tive outcome study of behavioral marital ther-
apy and communication therapy. J. Marital
Fam. Ther. 7:159–69
Osterman GP, Sher TG, Hales G, Canar WJ,
Singla R, et al. 2003. Physical illness. See
Snyder & Whisman 2003, pp. 350–69
Pearce JW, LeBow MD, Orchard J. 1981. Role
of spouse involvement in the behavioral treat-
ment of overweight women. J. Consult. Clin.
Psychol. 49:236–44
Raudenbush SW, Bryk AS. 2001. Hierarchical
Linear Models: Applications and Data Anal-
ysis Methods.Newbury Park, CA: Sage. 485
pp. 2nd ed.
Regev LG, O’Donohue W, Avina C. 2003.
Sexual dysfunction. See Snyder & Whisman
2003, pp. 181–200
Rice LN, Greenberg LS. 1984. Patterns of
Change: Intensive Analysis of Psychother-
apy Process. New York: Guilford. 308 pp.
Rogosa D. 1991. A longitudinal approach to
ATI research: models for individual growth
and models for individual differences in re-
sponse to intervention. In Improving Inquiry
in Social Science: A Volume in Honor of
Lee J. Cronbach, ed. RE Snow, DE Wi-
ley, pp. 221–48. Hillsdale, NJ: Erlbaum. 423
Schmaling KB, Sher TG, ed. 2000. The Psy-
chology of Couples and Illness: Theory, Re-
search, and Practice.Washington, DC: Am.
Psychol. Assoc. 407 pp.
Sexton TL, Ridley CR, Kleiner AJ. 2004.
Beyond common factors: multilevel-process
models of therapeutic change in marriage
and family therapy. J. Marital Fam. Ther.
Shadish WR, Baldwin SA. 2003. Meta-analysis
of MFT interventions. J. Marital Fam. Ther.
Shadish WR, Baldwin SA. 2005. Effects of
behavioral marital therapy: a meta-analysis
of randomized controlled trials. J. Consult.
Clin. Psychol. 73:6–14
Shadish WR, Montgomery LM, Wilson P, Wil-
son MR, Bright I, et al. 1993. Effects of
family and marital psychotherapies: a meta-
analysis. J. Consult. Clin. Psychol. 61:992–
Sher TG, Bellg AJ, Braun L, Domas A, Rosen-
son R, et al. 2002. Partners for life: a theoret-
ical approach to developing an intervention
for cardiac risk reduction. Health Educ. Res.
Shumway ST, Wampler RS, Dersch C,
Arredondo R. 2004. A place for marriage and
family services in employee assistance pro-
grams (EAPs): a survey of EAP client prob-
lems and needs. J. Marital Fam. Ther. 30:71–
Snyder DK. 1997. Manual for the Marital Sat-
isfaction Inventory—Revised. Los Angeles:
West. Psychol. Serv. 126 pp.
Snyder DK. 1999. Affective reconstruction in
the context of a pluralistic approach to cou-
ples therapy. Clin. Psychol. Sci. Pract. 6:
Snyder DK, Mangrum LF, Wills RM. 1993.
Predicting couples’ response to marital ther-
apy: a comparison of short- and long-term
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
25 Oct 2005 16:0 AR ANRV264-PS57-12.tex XMLPublish
(2004/02/24) P1: KUV
predictors. J. Consult. Clin. Psychol. 61:61–
Snyder DK, Schneider WJ, Castellani AM.
2003. Tailoring couple therapy to individual
differences: a conceptual approach. See Sny-
der & Whisman 2003, pp. 27–51
Snyder DK, Simpson JA, Hughes JN, eds. 2006.
Emotion Regulation in Families: Pathways to
Dysfunction and Health.Washington, DC:
Am. Psychol. Assoc. In press
Snyder DK, Whisman MA, eds. 2003. Treating
Difficult Couples: Helping Clients with Co-
existing Mental and Relationship Disorders.
New York: Guilford. 448 pp.
Snyder DK, Whisman MA. 2004a. Comorbid
Relationship and Individual Distress: Chal-
lenges for Intervention and Research. Pre-
sented at Assoc. Adv. Behav. Ther., New Or-
leans, LA
Snyder DK, Whisman MA. 2004b. Treating
distressed couples with coexisting mental
and physical disorders: directions for clinical
training and practice. J. Marital Fam. Ther.
Snyder DK, Wills RM. 1989. Behavioral ver-
sus insight-oriented marital therapy: effects
on individual and interspousal functioning.
J. Consult. Clin. Psychol. 57:39–46
Snyder DK, Wills RM, Faitler SL. 1988. Distin-
guishing specific from nonspecific interven-
tions in comparative outcome studies: reply
to Collins and Thompson. J. Consult. Clin.
Psychol. 56:934–35
Snyder DK, Wills RM, Grady-Fletcher A. 1991.
Long-term effectiveness of behavioral versus
insight-oriented marital therapy: a four-year
follow-up study. J. Consult. Clin. Psychol.
Sprenkle DH. 2002a. Editor’s introduction. See
Sprenkle 2002b, pp. 9–25
Sprenkle DH, ed. 2002b. Effectiveness Re-
search in Marriage and Family Therapy.
Alexandria VA: Am. Assoc. Marriage Fam.
Ther. 370 pp.
Sprenkle DH, Blow AJ. 2004. Common factors
and our sacred models. J. Marital Fam. Ther.
Stith SM, Rosen KH, McCollum EE, Thom-
sen CJ. 2004. Treating intimate partner vio-
lence within intact couple relationships: out-
comes of multi-couple versus individual cou-
ple therapy. J. Marital Fam. Ther. 30:305–18
Swindle R, Heller K, Pescosolido B, Kikuzawa
S. 2000. Responses to nervous breakdowns in
America over a 40-year period: mental health
policy implications. Am. Psychol. 55:740–49
Whisman MA. 1999. Marital dissatisfaction
and psychiatric disorders: results from the
National Comorbidity Survey. J. Abnorm.
Psychol. 108:701–6
Whisman MA, Dixon AE, Johnson B. 1997.
Therapists’ perspectives of couple problems
and treatment issues in couple therapy. J.
Fam. Psychol. 11:361–66
Whisman MA, Jacobson NS. 1990. Power, mar-
ital satisfaction, and response to marital ther-
apy. J. Fam. Psychol. 4:202–12
Whisman MA, McClelland GH. 2005. Design-
ing, testing, and interpreting interactions and
moderator effects in family research. J. Fam.
Psychol. 19:111–20
Whisman MA, McKelvie ML, Chatav Y. 2005.
Couple distress. In Improving Outcomes and
Preventing Relapse in Cognitive Behavioral
Therapy, ed. MM Anthony, DR Ledley, RG
Heimberg. New York: Guilford. In press
Whisman MA, Sheldon CT, Goering P. 2000.
Psychiatric disorders and dissatisfaction with
social relationships: Does type of relation-
ship matter? J. Abnorm. Psychol. 109:803–
Whisman MA, Snyder DK. 1997. Evaluating
and improving the efficacy of conjoint couple
therapy. In Clinical Handbook of Marriage
and Couples Interventions, ed. WK Halford,
HJ Markman, pp. 679–93. New York: Wiley.
720 pp.
Whisman MA, Uebelacker LA. 2005. Distress
and impairment associated with relationship
discord in a national sample of married or
cohabiting adults. J. Fam. Psychol. In press
Wiederman MW. 1997. Extramarital sex:
prevalence and correlates in a national sur-
vey. J. Sex Res. 34:167–74
Wilson GL. 1986. The comparative efficacy
of alternative treatments in the resolution of
marital dysfunction. PhD thesis. Univ. Mon-
tana. 108 pp.
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
November 8, 2005 22:20 Annual Reviews AR264-FM
Annual Review of Psychology
Volume 57, 2006
Frontispiece—Herbert C. Kelman xvi
Interests, Relationships, Identities: Three Central Issues for Individuals and
Groups in Negotiating Their Social Environment, Herbert C. Kelman 1
Emotion and Cognition: Insights from Studies of the Human Amygdala,
Elizabeth A. Phelps 27
Stressful Experience and Learning Across the Lifespan, Tracey J. Shors 55
Behavioral Theories and the Neurophysiology of Reward, Wolfram Schultz 87
Genetics of Affective and Anxiety Disorders,
E.D. Leonardo and Ren
e Hen 117
Sleep, Memory, and Plasticity, Matthew P. Walker and Robert Stickgold 139
Neuroecology, David F. Sherry 167
The Evolutionary Psychology of Facial Beauty, Gillian Rhodes 199
Explanation and Understanding, Frank C. Keil 227
Adolescent Development in Interpersonal and Societal Contexts,
Judith G. Smetana, Nicole Campione-Barr, and Aaron Metzger 255
Enduring Effects for Cognitive Therapy in the Treatment of Depression
and Anxiety, Steven D. Hollon, Michael O. Stewart, and Daniel Strunk 285
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
November 8, 2005 22:20 Annual Reviews AR264-FM
Current Status and Future Directions in Couple Therapy,
Douglas K. Snyder, Angela M. Castellani, and Mark A. Whisman 317
Attitudes and Persuasion, William D. Crano and Radmila Prislin 345
Psychological Perspectives on Legitimacy and Legitimation, TomR.Tyler 375
Personality and the Prediction of Consequential Outcomes, Daniel J. Ozer
and Ver
onica Benet-Mart
ınez 401
Child Development and the Physical Environment, Gary W. Evans 423
Consumer Psychology: Categorization, Inferences, Affect, and Persuasion,
Barbara Loken 453
Classroom Goal Structure, Student Motivation, and Academic
Achievement, Judith L. Meece, Eric M. Anderman,
and Lynley H. Anderman 487
Analysis of Longitudinal Data: The Integration of Theoretical Model,
Temporal Design, and Statistical Model, Linda M. Collins 505
The Internet as Psychological Laboratory, Linda J. Skitka
and Edward G. Sargis 529
Family Violence, Patrick Tolan, Deborah Gorman-Smith, and David Henry 557
Understanding Affirmative Action, Faye J. Crosby, Aarti Iyer,
and Sirinda Sincharoen 585
Subject Index 613
Cumulative Index of Contributing Authors, Volumes 47–57 637
Cumulative Index of Chapter Titles, Volumes 47–57 642
An online log of corrections to Annual Review of Psychology chapters
may be found at
Annu. Rev. Psychol. 2006.57:317-344. Downloaded from
by University of Delaware on 11/30/05. For personal use only.
... Effective treatment involves understanding human nature developed through therapeutic theories and research and choosing the appropriate techniques to facilitate couples' growth (Leibert, 2011). Research in couple treatment theories establishes the effectiveness of multiple interventions in improving a wide range of cognitive and affective behavioral patterns and the overall relationship (Snyder et al., 2006). However, choosing, assimilating, and applying the correct technique in practice from the dispersed empirical data is overwhelming (Gurman, 2008), more so for a beginning practitioner. ...
... We searched for couple therapy and marital intervention keywords in Google Scholar and Microsoft Academic search engines and conducted an advanced search in ProQuest, EBSCO, JSTOR, Wiley, Springer, Sage, and Taylor and Francis databases (Fig. 1). We started with a set of studies based on prior research (Baucom & Fischer, 2019;Christensen et al., 2010;Epstein, 2001;Gottman & Krokoff, 1989;Gottman & Levenson, 1999;Gurman, 2008;Jacobson & Addis, 1993;Johnson & Wittenborn, 2012;Shadish & Baldwin, 2003;Snyder et al., 2006). The review of initial manuscripts, including handbooks and encyclopedias, empirical reviews, comparison and longitudinal analysis, enabled a basic understanding of different modalities and couple therapy's current status and direction. ...
... The emotional dimension is a crucial change area and would include techniques that focus on a deeper understanding of affect for self and partner. However, past developmental experiences can influence the present affect patterns (Snyder et al., 2006). ...
Full-text available
One can find a rich set of empirically evaluated techniques across different schools in couple therapy over its evolution of five decades. Though there are multiple systematic reviews and analyses of couple intervention studies, none focus on reviewing the universal dimensions of change across therapeutic techniques. Understanding the common areas of change would enable integrated learning across therapy modalities for novice therapists. Therefore, the aim is to identify the techniques employed in couple intervention research and categorize their change dimensions. We examined 40 articles on couple interventions published across 16 journals and identified 111 techniques. The five therapeutic change dimensions, namely behavior, cognition, emotion, attachment, and holistic, were categorized based on the common factor integration of techniques. The identified techniques were further classified under the five dimensions using the voting procedure to validate the universality of change dimensions.
... Over the years, couple therapy researchers have been able to support the efficacy of couple therapy. Reviews and meta-analyses have reported the relevance of couple therapy with effect sizes similar-perhaps even larger in some cases-to effect sizes obtained in individual therapy (Gurman, 2011;Snyder et al., 2006). In their meta-analysis, Shadish and Baldwin (2003) revealed a large overall mean effect size (d = 0.84) for couple therapy compared to no intervention. ...
... In their meta-analysis, Shadish and Baldwin (2003) revealed a large overall mean effect size (d = 0.84) for couple therapy compared to no intervention. Researchers have also highlighted that two thirds of couples completing couple therapy reported positive outcomes (Lebow & Gurman, 1995;Snyder et al., 2006). However, the efficacy of couple therapy is not absolute, and not all couples may benefit from it. ...
... However, the efficacy of couple therapy is not absolute, and not all couples may benefit from it. For example, efficacy trials have demonstrated that in only half of the treated couples did both partners report improvements (Snyder et al., 2006) and that 30% of the couples improving after therapy remained romantically distressed (Snyder & Halford, 2012). ...
Full-text available
The literature has supported the broad importance of the coparenting relationship for the couple and for family. However, couple therapy research has overlooked couples of parents and coparenting as a target and an outcome of therapy. Therefore, this thesis aimed to explore the relevance of working on the coparenting relationship within couple therapy and observed the nature of changes related to this specific therapeutic work. As a preliminary step, we reviewed and evaluated the efficacy of coparenting interventions within the broader framework of intervention programs for couples. Then, we focused on a couple therapy, and presented the Integrative Brief Systemic Intervention (IBSI) that systematically integrates therapeutic work on both the romantic and coparenting relationships. Finally, we conducted two process studies to identify the nature of the changes related to integrating coparenting work into the IBSI. This thesis offers the first evidence of the relevance of working on coparenting within both prevention programs and IBSI. The process studies allowed us to describe the process unfolding when parents improved their coparenting satisfaction within the IBSI. This process comprised six steps and discriminated couples whose coparenting outcomes improved from couples who did not improve after couple therapy. We finally discussed the implications of this research for both empirical literature and clinical practice.
... Dieser positive Zustand hält nach der Eheschließung an, wonach auch in den ersten Ehejahren ähnlich hohe Zufriedenheitswerte (80-85 %) berichtet werden (vgl. Gallup, 1990 (Christensen et al., 2010;Snyder et al., 2006). In der täglichen Praxis der Eheberatung im deutschsprachigen Raum sind die Erfolgsquoten mit ca. 25 % noch niedriger (vgl. ...
... Overall, most couple treatments reduce relational distress by this standard . There is considerable evidence that couple therapies reduce individual psychopathologies as well, especially substance abuse (O'Farrell & Schein, 2011), mood disorders (Whisman & Beach, 2015), and general distress (Snyder et al., 2006). ...
Full-text available
Hope focused couple therapy (HFCT) is often spiritually integrated treatment using forgiveness, prayer, sanctification themes, and relational virtues as mechanisms for change. Building on earlier work examining the effects of HFCT, this clinical field study of couple therapy examined trends and predictors of relational outcomes. Couples (N = 236; N = 472 individuals) seeking HFCT were assessed at three time points: pre- and post-treatment and 6 months after treatment. A subsample was assessed at a fourth time point: 2–10 years after treatment (n = 54 couples; n = 122 individuals). Couples showed improvements in relational outcomes, whether spiritual integration was included or not. Cohen’s d effect sizes in relationship satisfaction ranged from 0.44 for long-term change to 1.08 for pre- and post-change. Participants in the clinical range at baseline improved the most, while non-clinical couples (at baseline) maintained relationship adjustment. Multilevel growth modeling indicated that couples increased in dyadic adjustment from pre-intervention through 6-month follow-up. For the subsample of individuals with an additional long-term follow-up assessment, multilevel growth modeling indicated these couples also showed a significant rate of increase in dyadic adjustment from pre-intervention to long-term follow-up. Though these results need to be replicated with a comparison or control group, these findings suggest that HFCT may assist couples during times of relational stress with expectation of maintenance or gains in dyadic adjustment over the long term.
... Impairment of emotional regulation and marital turmoil at the individual and couple level have an interactive relationship with each other. In fact, how couples regulate emotion in the etiology of confusion and individual pathology is effective in the continuation of confusion and disorder, recurrence or maintenance of therapeutic progress and prognosis of treatment (Snyder & et al., 2006). Thus, marital infidelity can lead to dysfunction of emotion regulation in individuals. ...
Full-text available
This article explores professionals' understanding and experiences of parental high conflicts in Norwegian family counsellor and child welfare services. The data were analysed using reflexive thematic analysis, examining four focus group interviews with a total of 24 professionals. We used tame and wicked problems as a theoretical frame of reference in order to discuss how high conflict cases can be understood. The analysis shows that the complexity and experiences of high conflicts challenge professionals in their assessments and development of solutions. Our conclusion is that the nature of the complexity, unpredictability, and instability of high conflicts fits within the framework of wicked problems.
Integrative psychotherapeutic practice is the process of creating a broad, overarching framework as a guide to selecting and combining concepts and interventions. Attracting attention over the past 20 years, it has been identified as the dominant trend in family therapy moving into the 21st century. It is no surprise that clinicians necessarily combine several theories, techniques, and factors in their work given the multidimensional nature of work with clients and families. Integrative practice has evolved over time with different models and frameworks proposed, and yet in the field, there are concerns that it is not widely taught despite many identifying as eclectic or integrative. The purpose of this paper is twofold. Firstly, it reviews the topic of integrative practice, its history, evolution, strengths and pitfalls, and models. Secondly, a systemic meta‐framework is proposed as a guide to clinical practice and teaching. The review found a move from eclecticism to integrative practice, with the evolution of integration leading to five primary methods. The proposed QIFT Systemic Meta‐Framework for Integrative Practice both fits within and extends these methods, with the following key elements: way of being, therapeutic alliance with a focus on neurobiology, systemic assessment and formulation, and decision‐making highlighting theory of change.
Full-text available
In this article, we attempt to distinguish between the properties of moderator and mediator variables at a number of levels. First, we seek to make theorists and researchers aware of the importance of not using the terms moderator and mediator interchangeably by carefully elaborating, both conceptually and strategically, the many ways in which moderators and mediators differ. We then go beyond this largely pedagogical function and delineate the conceptual and strategic implications of making use of such distinctions with regard to a wide range of phenomena, including control and stress, attitudes, and personality traits. We also provide a specific compendium of analytic procedures appropriate for making the most effective use of the moderator and mediator distinction, both separately and in terms of a broader causal system that includes both moderators and mediators. (46 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Collins and Thompson (1988) challenged the utility of such labels as behavioral and insight-oriented in comparative therapy research and questioned the manner in which these theoretical approaches were operationalized in a recent study by Wills, Faitler, and Snyder (1987). We argue that such labels, while imperfect, identify the context in which theory-driven models of clinical intervention are tested. Psychotherapy research requires that specific and nonspecific components of treatment approaches be distinguished. The study by Wills et al provides a model for verifying therapist fidelity in manual-guided treatment research.
Examined a measure of marital power based on the clinical description of the "dominance through talking" and the "dominance through listening" patterns (N. S. Jacobson and A. Holtzworth-Munroe, 1986) and defined in terms of asymmetry in the relative frequencies of verbal communication content patterns. Ss included 23 happily married couples and 31 distressed couples seeking marital therapy. Marital power was assessed from transcripts made from the couples engaged in discussing the events of their day. The Dyadic Adjustment Scale was used to measure marital satisfaction. Results indicate an inverse relationship between marital satisfaction and power inequality. Power inequality prior to therapy predicted positive treatment outcome at posttest and at 6-mo follow-up for couples receiving a regimen of social-learning-based marital therapy. (PsycINFO Database Record (c) 2016 APA, all rights reserved)
Objective. To evaluate the effects of a spouse-assisted pain-coping skills training intervention on pain, psychological disability, physical disability, pain-coping, and pain behavior in patients with osteoarthritis (OA) of the knees. Methods. Eighty-eight OA patients with persistent knee pain were randomly assigned to 1 of 3 conditions: 1) spouse-assisted pain-coping skills training, (spouse-assisted CST), 2) a conventional CST intervention with no spouse involvement (CST), or 3) an arthritis education-spousal support (AE-SS) control condition. All treatment was carried out in 10 weekly, 2-hour group sessions. Results. Data analysis revealed that at the completion of treatment, patients in the spouse-assisted CST condition had significantly lower levels of pain, psychological disability, and pain behavior, and higher scores on measures of coping attempts, marital adjustment, and self-efficacy than patients in the AE-SS control condition. Compared to patients in the AE-SS control condition, patients who received CST without spouse involvement had significantly higher post-treatment levels of self-efficacy and marital adjustment and showed a tendency toward lower levels of pain and psychological disability and higher scores on measures of coping attempts and ratings of the perceived effectiveness of pain-coping strategies. Conclusion. These findings suggest that spouse-assisted CST has potential as a method for reducing pain and disability in OA patients.
Randomly selected samples of practicing couple therapists who were members of the American Psychological Association's Division 43 or the Association for Marriage and Family Therapy completed a survey of couple problem areas and therapeutic issues encountered in couple therapy. Therapists rated problem areas in terms of occurrence, treatment difficulty, and damaging impact. A composite of these 3 dimensions suggested that the most important problems were lack of loving feelings, power struggles, communication, extramarital affairs, and unrealistic: expectations. Comparison of the findings with therapist ratings obtained by S. K. Geiss and K. D. O'Leary (1981) suggests considerable stability in presenting problems in couple therapy over the past 15 years. Therapist-generated characteris tics associated with negative outcome were also identified, the most common being partners' inability or unwillingness to change and lack of commitment.
Despite a recent surge of interest in the mechanisms and processes of change during psychotherapy, investigations to date have yielded lamentably few interpretable results. The present article highlights previous barriers to the study of change in psychotherapy and offers a conceptual and methodological framework to increase the interpretability of future studies. A frequently overlooked distinction between change mechanisms, or intermediate changes in the client over the course of treatment, and change processes, or the active ingredients of the therapeutic process, is presented and developed into a multiphase model of programmatic change research. It is argued that investigators should first develop an understanding of change mechanisms and only subsequently conduct targeted process research to identify important change processes.