Benchmarking Community-Based Couple Therapy: Considering Measurement Reactivity

Couple therapy has been shown to be effective in randomized clinical trials; however, results from naturalistic couple therapy have been less consistent. This study utilized a benchmarking approach to compare the effectiveness of couple therapy in a community-based setting with findings from efficacy treatments, such as treatment within randomized clinical trials. The current study is the largest couple therapy sample published to date (N = 3,347 couples). Clients in couple therapy were asked to provide initial and weekly ratings of symptomology on the Outcome Questionnaire (OQ-45.2). We found that treatment effect sizes found at community clinics were smaller than efficacy studies (i.e., the benchmark). However, when taking into account measurement reactivity, the effect sizes were comparable. This is the first benchmarking study for community-based couple therapy, allowing for meaningful comparisons and understanding of outcomes in real-world couple therapy. Implications for the field are offered in terms of evaluating community-based psychotherapy studies with benchmarking for couple therapy. Results of this study provide clinicians and researchers a way to meaningfully compare couple therapy outcomes, accounting for differences in community-based practices and randomized clinical trials. This benchmark also underscores the impact of measurement sensitivity, an issue commonly overlooked in psychotherapy research and practice.     

The Findings of a Cochrane Meta-Analysis of Couple Therapy in Adult Depression: Implications for Research and Clinical Practice

The association between depressive symptoms and distressed intimate relationships supported the assumption that couple therapy, by focusing on the interpersonal context of depression, might be more effective as a treatment for depression than individual psychotherapy or drug therapy. This issue was addressed by a Cochrane meta-analysis assessing the evidence from clinical trials of couple therapy for depression in comparison with individual psychotherapy, drug therapy, and no/minimal intervention, including fourteen studies with 651 participants. No study was found free of bias and the quality of the evidence was low, with major problems of small sample sizes, missing outcome data, selective reporting, lack of information on random sequence generation and allocation concealment, recruitment of people not representative of clinical practice, and allegiance bias. The meta-analysis showed that both couple therapy and individual psychotherapy improved depressive symptoms at end of treatment and after 6 months or longer, with moderate effect sizes, without any difference between the two treatments. Couple therapy was more effective in reducing couple distress. This effect was larger in studies with distressed couples only and should be considered as relevant in its own right. Couple therapy is a viable option for the treatment of a depressed partner, especially in discordant couples. Future research should address several issues left open to provide a sound empirical foundation for clinical practice.     

Couple therapy for military veterans: overall effectiveness and predictors of response

Despite the numerous challenges facing U.S. veterans and their relationships, there have been no examinations of the effectiveness of couple therapy for relationship distress provided to veterans. In the present study, 177 couples presenting for couple therapy at two Veteran Administration Medical Centers completed assessments of relationship satisfaction prior to therapy and weekly during therapy. Results revealed that the average couple showed significant gains in relationship satisfaction during treatment (d=0.44 for men; d=0.47 for women); gains were larger for couples beginning therapy in the distressed range (d=0.61 for men; d=0.58 for women) than for couples in the nondistressed range (d=0.19 for men; d=0.22 for women). Rates of premature termination were high, with 19% of couples completing fewer than three sessions and 62% rated as not completing a "full course" of therapy. Benchmarking analyses demonstrated that the average gains were larger than would be expected from natural remission and similar to previous effectiveness trials; however, average gains were smaller than those observed in couple therapy efficacy trials. Relationship, psychological, and demographic characteristics were generally unrelated to the amount of change in therapy after controlling for initial satisfaction. However, African American couples showed significantly larger gains than Caucasian, non-Hispanic couples. Thus, though yielding smaller effects than those shown in efficacy trials, the impact of couple therapy for veterans' relationship problems appears to generalize across various demographic, psychological, and relationship characteristics.     

Treatment‐as‐Usual for Couples: Trajectories Before and After Beginning Couple Therapy

Couple therapy has been shown to be a meaningful way to improve couples’ relationships. However, less information is known about couples’ functioning prior to entering treatment in community settings, as well as how their relationship functioning changes from initiating therapy onward. This study examined 87 couples who began community‐based couple therapy during a longitudinal study of couples in the military. The couples were assessed six times over the course of 3 years, including time points before and after starting couple therapy. Using an interrupted‐time series design, we examined trajectories across the start of couple therapy in relationship satisfaction, divorce proneness, and negative communication. The results demonstrated that couples’ relationship satisfaction was declining and both divorce proneness and negative communication were increasing prior to entering couple therapy. After starting couple therapy, couples’ functioning on all three variables leveled off but did not show further change, but previous experience in relationship education moderated these effects. Specifically, those who were assigned to the relationship education program (vs. control) demonstrated greater reductions in divorce proneness and greater increases marital satisfaction after starting therapy; however, they also started more distressed.     

Relative Cost Effectiveness of In-Person and Internet Interventions for Relationship Distress

While the efficacy of couple therapy has been repeatedly demonstrated, most distressed couples do not seek treatment. To improve reach and overcome barriers to therapy, Integrative Behavioral Couple Therapy (IBCT) was translated into a web-based program, OurRelationship (OR). While both IBCT and OR have been shown to improve relationship and individual functioning, the goal of the present study was to compare the relative cost effectiveness of these two treatment modalities. In IBCT, 74% of couples experienced reliable improvement, compared to 55% of couples in OR. Within-group Cohen's d effect sizes during treatment for relationship satisfaction were d = 0.87 for IBCT and d = 0.96 for OR. Relative cost effectiveness analyses revealed that even at the highest expense estimates for OR and the lowest expense estimates for IBCT, OR was more cost effective in creating reliable improvement once at least 229 couples were served and was more cost effective in creating large-size effects once at least 153 couples were served. Cost-effectiveness increases for both IBCT and OR as more couples are served; however, this cost savings occurs at a much more rapid rate for OR. These findings demonstrate that despite higher initial development costs, Internet programs are a cost-effective option for dissemination either as a stand-alone service or as an initial intervention in a stepped care model with more intensive in-person services.     

The clinical representativeness of couple therapy outcome research

The clinical representativeness of outcome studies is defined as the generalizability of recruitment processes, assessment/diagnostic procedures, treatment protocols, and therapeutic results from research settings to naturalistic treatment settings. The main goal of the present study was to examine the clinical representativeness of couple therapy in outcome studies. The data set was formed by 50 published clinical trials, including 34 couple therapy outcome studies for marital distress (CTMD) and 16 couple therapy outcome studies for comorbid relational and mental disorders (CTMD + C). The present findings showed that, overall, the clinical representativeness of couple therapy outcome studies is only fair (i.e., the mean global score is slightly lower than the midpoint of the rating scale used to assess representativeness). CTMD + C studies fared better than CTMD studies on many dimensions of clinical relevance. Studies in which pretherapy training was less intensive (for CTMD studies only), treatment was less structured, and therapists were more experienced showed larger effect sizes than those in which such was not the case.     

Marital therapy outcome measured by therapist, client, and behavior change

The present research examined marital therapy outcome, in a sample of 88 couples and 22 therapists, through the use of multidimensional criteria from multiple perspectives. Three paper-and-pencil patient self-report instruments were used to measure different aspects of the couple relationship. In addition, therapist post-therapy ratings and objective observer ratings of the couple behavior during therapy were obtained. Past debates about what the "right" criteria might be for measuring therapeutic change prompted our strategy. Special attention was focused on those criteria that could be classified as "inside" or "outside" of the couple relationship. Data analysis revealed a significant relationship between the couples' and therapists' ratings of improvement in therapy. Additionally, husband and wife negative acts (behaviors) in therapy were also significantly correlated with therapists' ratings of improvement. Both statistically significant and clinically significant criteria were employed to evaluate the therapy outcome. Improvement due to therapy as measured by the three self-report instruments varied from 38% to 58%. These differences appeared to stem from the different kinds of information yielded by the specific tests.     

Prevention of marital instability and distress. Results of an 11-year longitudinal follow-up study

Divorce and destructive couple conflict are major risk factors for many forms of dysfunction and psychopathology in the family. Couple relationship education has been offered in an attempt to reduce those risks and enhance couple outcomes. The current effectiveness study had two aims: to assess (a) the long-term effects of relationship education and (b) the ability of partners to remember the skills typically taught during the communication skills training. One hundred and one couples from two studies, in which the effectiveness of the EPL (Ein Partnerschaftliches Lernprogramm für Paare [A Learning Program for Married Couples]) was investigated, were followed-up 11 years after the training. In the first study, EPL-takers had a significantly lower divorce and separation rate (27.5%) than non-EPL-takers (52.6%). This finding was replicated in a second study, showing a 20% dissolution rate in EPL-takers. For those couples still together, the rate of happy relationships was 80% at the eleven-year follow-up - thus, there is optimism for longer married couples who are willing to focus on improving their relationships. Only about 55% of the partners could remember at least one speaker skill whereas 70% remembered at least one listening skill taught during EPL. Consequences for prevention programs in the couple domain are discussed.     

Linking questionnaire reports and observer ratings of young couples' hostility and support

Past studies have correlated observer ratings with questionnaire self- and partner-reports of behaviors in close relationships. However, few studies have actually proposed and tested longitudinal models that link observer ratings to past behaviors and to questionnaire self- and partner-reports of behaviors during an observational task. Using data from a panel of 324 young couples, we demonstrate that (a) observer ratings of hostility and support are significantly related to couple reports of the same behavior in the relationship two years earlier, and (b) respondent and partner questionnaire reports of hostility and support during the observational task converge with observer ratings of the same behavior even after controlling for earlier self- and partner-reports. These findings demonstrate that observer reports based on brief discussion tasks reflect the tenor of the relationship over a relatively long period of time. They also demonstrate that couple reports of interactions reflect observable behaviors beyond that attributed to earlier self- and partner-reports. Consistent with previous research, effect sizes are larger for hostility than support but there are few differences between men and women.     

Therapy for couples after an affair: A randomized-controlled trial

To examine the efficacy of a couples treatment approach for promoting recovery from a recently disclosed affair, 89 couples that disclosed an affair by one of the partners in the past 6 months were randomly assigned to treatment (n = 46) or to a control group that waited about 3 months for treatment (n = 43). The couples completed self-report measures (Beck Depression Inventory, Impact of Event Scale-Revised, Partnership Questionnaire) at pre- and post-treatment. Since about half of the couples dropped out for various reasons (e.g., ongoing affair, separation), we used multiple imputations to handle the missing data problem. We analyzed the dyadic data with hierarchical linear modeling in a two-level model. Significant improvements on scores of anxiety corresponded with large effect sizes for both partners. Yet significant improvements on depression scores were only found for the unfaithful partner with moderate effect size. Results suggest that the treatment can improve individual complaints, but not relationship satisfaction in a sufficient amount for both partners. Hence, future research should address how this intervention could encourage couples to maintain therapy, and how they might achieve more and sustained improvement in relationship satisfaction.     

How Can We Improve Preventive and Educational Interventions for Intimate Relationships?

Improving intimate relationships with preventive and educational interventions has proven to be more difficult than originally conceived, and earlier models and approaches may be reaching their limits. Basic concerns remain about the long-term effectiveness of these interventions, whether they are reaching and benefiting couples most likely to need them, and how they might be exerting their effects. We identify six problems that we believe are hindering progress in the field, and for each we outline research findings that point to new ways forward. These problems include (a) the incomplete understanding of couple communication and unwarranted translation of communication findings into interventions; (b) the surprising stability of relationship satisfaction; (c) the powerful roles that personal histories, personalities, and stress play in determining relationship outcomes; (d) the difficulties involved in recruiting and retaining high-risk couples in intervention programs; (e) the lack of attention given to specific stages of relationship development in interventions; and (f) the tendency to deliver preventive and educational interventions in the same format as therapies for distressed couples. We conclude by noting that a large body of basic research about intimate relationships, and large-scale outcome research with diverse populations, hold great promise for advancing the field.     

The self in the couple relationship: part 1

This paper discusses the importance of the ‘good-enough’ development of the core self in the couple relationship. It argues that difficulties in the development of the self lie behind a significant number of troubled couple relationships. Part 1 focuses on individual psychotherapy and how changes to the core self are often attended by shifts in the marital relationship. It traces how this occurs through clinical material and analyses what kinds of changes to the self are needed for the couple relationship to function well. Psychodynamic theory is used to understand the core changes over time and how these are related to significant second-order change in the couple relationship. The construct of the ‘transitional space’ is used to theorize the couple relationship. Implications for the practice of individual psychotherapy are discussed. In Part 2, the significance of change in the experience of the self of one or both partners for a successful outcome of couples therapy is discussed, along with the place of work with individual partner(s) in the course of conjoint couples therapy.     

Introducing the Expectations and Preference Scales for Couple Therapy (EPSCT): Development,Psychometric Evaluation,and Suggested Use in Practice and Research

While it is known that client factors account for the largest proportion of outcome variance across treatment modalities, little is known about how clients’ characteristics affect the process and effectiveness of couple therapy. To further knowledge in this area, we created a brief, practice‐friendly measure, the Expectation and Preference Scales for Couple Therapy (EPSCT). Three self‐report scales assess clients’ Outcome expectations (e.g., I expect our relationship to improve as a result of couple therapy) and role expectations for Self (e.g., I expect to listen to my partner's concerns) and Partner (e.g., I expect my partner to blame me). Three Cognitive‐Behavioral, Emotionally Focused, and Family Systems preference scales use a forced‐choice format to measure the comparative strength of respondents’ preferences for interventions broadly reflective of each approach. A large item pool was developed from relevant literature and clinical experience and refined based on face and content analyses with two panels of experienced couple therapists and researchers. Across four studies with 1,175 participants, the scales’ internal consistency reliabilities were similar and their construct validity was supported with confirmatory factor analyses and significant correlations with several established measures, including expectation measures developed for individual psychotherapy and measures of attitudes toward professional help seeking and valuing personal growth. Across all studies, participants had stronger role expectations for themselves than their partners, although gender effects differed by sample. We discuss how to use the 15‐item EPSCT in clinical practice and in future research as a predictor of couple therapy processes and outcomes.     

Couples treatment for major depression: critique of the literature and suggestions for some different directions

Current couples approaches to the treatment of depression show considerable promise for some couples, but have substantial room for improvement – in terms of efficacy as well as applicability and acceptability to a broader population. Although conjoint approaches have been shown to be efficacious in reducing couple distress and are possibly efficacious in ameliorating depression that co-occurs with couple distress, methodological problems hamper generalization from the published findings. Therefore, it is not clear that most depressed persons are likely to benefit from currently available conjoint formats. Most basically, there may often be obstacles that prevent or delay partner involvement, highlighting the need for treatments that can enhance relationship functioning without relying on a conjoint format. Existing treatment delivery systems may also fail to reach a considerable segment of the population. Developing a more flexible set of options for implementing couple-focused treatment, as well as rethinking dominant assumptions about depression, may be integral to extending the viability of couples therapy for depression.     

Low-level relationship aggression and couple therapy outcomes.

Increasing evidence supports the efficacy of conjoint therapies that focus on intimate partner violence for couples who engage in mild to moderate physical aggression but want to preserve the relationship and end the aggression. However, there has been no examination of how this population responds to couple therapy that does not have a specific focus on aggression. This lacuna in the research literature is of concern because couples with a history of low-level aggression often seek couple therapy, but couple therapy without a focus on violence is thought to potentially exacerbate aggression. In the current study, the authors examined the efficacy of non-aggression-focused behavioral couple therapy for couples with and without a history of mild physical aggression. One hundred thirty-four couples, 45% of whom had experienced low-level aggression in the year prior to therapy, completed up to 26 sessions of couple therapy and 2 years of follow-up assessments. Results demonstrated no significant differences in relationship and individual outcomes by history of aggression. In addition, couples maintained very low levels of physical aggression during and after treatment and showed reductions in psychological aggression when relationship and individual functioning improved.     

Effectiveness of the ‘Hold me Tight’ Relationship Enhancement Program in a Self‐referred and a Clinician‐referred Sample: An Emotionally Focused Couples Therapy‐Based Approach

While evidence‐based couple therapies are available, only a minority of troubled couples seek help and they often do this too late. To reach more couples earlier, the couple relationship education (CRE) group program “Hold me Tight” (HmT) based on Emotionally Focused Couples Therapy (EFCT) was developed. This study is the first to examine the effectiveness of HmT. Using a three‐wave (waiting period, treatment, and follow‐up) within‐subject design, HmT was delivered to 79 self‐referred couples and 50 clinician‐referred couples. We applied a comprehensive outcome measure battery. Our main findings were that (1) self‐referred couples significantly improved during HmT on all measures, that is relationship satisfaction, security of partner‐bond, forgiveness, daily coordination, maintenance behavior, and psychological complaints, with a moderate‐to‐large mean effect size (d = .63), which was maintained (d = .57) during the 3.5 month follow‐up; (2) in clinician‐referred couples, who were vulnerable in terms of insecure attachment status and psychopathology, the improvement during HmT was moderate (d = .42), but this was reduced during the 3.5‐month follow‐up to a small effect (d = .22); (3) emotional functioning (typical HmT target) as well as behavioral functioning (typical Behavioral Couples Therapy‐based CRE target) improved during HmT; and (4) individual psychological complaints, although not specifically targeted, were reduced during HmT. These findings suggest that HmT is a promising intervention for enhancement of relationship functioning. Clinical implications are discussed.     

Cognitive behavioral therapy in practice: treatment delivered by trainees at an outpatient clinic is clinically effective

Cognitive–behavioral therapy (CBT) is known to be effective for a number of disorders, and can be delivered effectively by trainees in controlled settings. However, the effectiveness of trainee therapists in general practice compared to that of more experienced therapists is unknown. In this study, the authors used a benchmarking strategy to compare the outcomes of naturalistic CBT delivered by trainee therapists to those of efficacy and effectiveness studies using primarily professional therapists. Patients (N = 249) with mood and anxiety disorders were treated by trainees (primarily by interns and postdocs) using nonstandardized nonmanualized CBT at an outpatient clinic in an urban academic medical center. Changes in anxiety and depression were assessed using effect sizes, reliable and clinically significant change, and benchmarked to efficacy and effectiveness studies. Symptoms of anxiety and depression significantly improved from start to end of treatment. Rates of improvement and recovery compared favorably to those achieved in other studies, with the exception of recovery rates in severe depression. Effect sizes were in the medium to large range, but generally lower than those achieved in other studies. Results suggest that CBT can be delivered effectively by trainees in an outpatient setting.     

Individual distress and dyadic adjustment over a three‐year follow‐up period in couple therapy: a bi‐directional relationship?

This study examined the relationship between depressive symptoms and dyadic adjustment, as well as between interpersonal problems and dyadic adjustment, during residential couple therapy and at a three‐year follow‐up (N=106). Mixed models were used in the analyses. Significant improvement (p<.001) occurred on all measures from admission to discharge (effect sizes .27?.83) and from admission to three‐year follow‐up (effect sizes .52?.79). During the observation period, improvement in depressive symptoms at the first time point predicted improvement in dyadic adjustment at the subsequent time point. Furthermore, the dyadic adjustment level at discharge predicted improvement in depressive symptoms in the follow‐up period. There were only modest associations between personality variables and dyadic adjustment. The clinical implication is that in couples suffering from co‐existing relational and symptomatic distress, couple therapy should include the aim of lowering depressive symptoms.     

Optimizing the alliance in couple therapy

The complexity of the relation between alliance and outcome in couple therapy was investigated in a study of 47 couples in brief therapy. Self-rated alliance was measured after the first and third sessions using the couple version of the Working Alliance Inventory. The results indicated that the correlation between alliance and outcome was significantly stronger when the partners agreed about the strength of the alliance, when the male partner's alliance was stronger than the female's, and when the strength of both partners' alliance increased as therapy progressed. The authors suggest that a unique feature of couple therapy is that the partners have both a preexisting relationship with each other (allegiance) and an alliance with the therapist to balance.