What Clients of Couple Therapy Model Developers and Their Former Students Say About Change, Part II: Model-Independent Common Factors and an Integrative Framework
Sean D. Davis, PhD, Marital and Family Therapy Program, Alliant International University–Sacramento Campus; Fred P. Piercy, PhD, Department of Human Development, Virginia Polytechnic Institute and State University.
The authors wish to thank Drs. Frank M. Dattilio, Susan M. Johnson, Richard C. Schwartz, Judy Makinen, Ms. Beth O’Neil, MSW, and their clients for their participation in this study. We also thank Drs. Scott W. Johnson, Douglas H. Sprenkle, Lenore M. McWey, Anna Beth Benningfield, and Margaret L. Keeling for their support throughout various stages of this project.
This project was completed while Sean Davis was a doctoral student at Virginia Tech. Portions of this project were presented at the 2005 annual meeting of the American Association of Marriage and Family Therapy, Kansas City, Missouri.
This study was supported with monies from the 2005 AAMFT Graduate Student Research Award and the 2006 AAMFT Dissertation Award.
Abstract
Proponents of the common factors movement in marriage and family therapy (MFT) suggest that, rather than specific models of therapy, elements common across models of therapy and common to the process of therapy itself are responsible for therapeutic change. This article—the second of two companion articles—reports on a study designed to further investigate common factors in couple therapy. We used grounded theory techniques to analyze data from interviews with MFT model developers Dr. Susan M. Johnson, Dr. Frank M. Dattilio, Dr. Richard C. Schwartz, former students of Dr. Johnson and Dr. Schwartz, and each of their clients who had been successful in couple therapy. This article reports model-independent variables, that is, general aspects of therapy that are not directly related to the therapist’s model. Model-independent categories include client variables, therapist variables, the therapeutic alliance, therapeutic process, and expectancy and motivational factors, each with several subcategories. We also present a conceptual framework that outlines how model-dependent and model-independent common factors may interact to produce change. We discuss our findings and proposed framework in relation to the current common factors literature in psychology and MFT. We also discuss clinical, training, and research implications.
What is it about therapy that helps couples change? This question has been the focus of countless books, presentations, and journal articles. There are likely as many opinions on the matter as there are couples therapists. Decades of comparative efficacy research has demonstrated that marriage and family therapy (MFT) works (Shadish & Baldwin, 2002). What we know less about, however, is why MFT works. The same comparative efficacy research has yielded another clear conclusion—when therapy works, the vast majority of the outcome is not due to the unique contributions of any one model (Shadish & Baldwin, 2002; Wampold, 2001). So, when two couples receiving different therapies both improve, what is it that helped them do so?
The common factors position in MFT (Sprenkle & Blow, 2004a, 2004b) posits that the variables responsible for change in therapy likely fall into two main categories—those that are largely shared among effective MFT models and those that are inherent in the nature of the therapy process itself. Those variables shared among therapy models, as we discussed in our companion article, are often referred to as narrow factors (Lambert, 1992), and include common therapeutic tasks, such as changing cognitions, that occur in all therapies to some degree or another. Most effective therapies, common factors researchers say (Sprenkle & Blow, 2004a), focus on similar tasks but use different language to describe them. The curative factors inherent in the process of therapy are often called broad factors (Hubble, Duncan, & Miller, 1999; Tallman & Bohart, 1999), and include such aspects of therapy as client and therapist variables, and the nature of the therapeutic alliance. The broad factors will be the focus of this article, and will be referred to as “model-independent” variables due to their presence in therapy regardless of the use of a model.
The current MFT common factors literature reflects early stages of theory development in that it consists primarily of lists of narrow and broad variables thought to be related to change (Davis & Butler, 2004). Critics and proponents of the movement alike have noted that for the common factors literature to be optimally useful as a clinical, research, and training tool, the next step will be to move from checklists of factors to an empirically derived model of what factors are common among effective therapies and how these factors interact to produce change (Sexton & Ridley, 2004; Sexton, Ridley, & Kleiner, 2004; Sprenkle & Blow, 2004a, 2004b). The purpose of this article, the second of two companion articles, is to discuss the results of a study aimed at providing such a model. In the first article we discuss the research methodology as well as outline the narrow or “model-dependent” variables we found (i.e., common practices across different models), and discuss their relevance for clinical practice, research, and training. In this article we discuss the broad or “model-independent” variables that we found, and present a clinical model outlining how model-dependent and model-independent common factors interact to produce change.
Methodology
The research methodology, including therapist and client demographic information and nomenclature, was discussed in detail in the first article. It will be briefly reviewed here to serve as a reminder. Three different MFT model developers or refiners (i.e., Dr. Susan Johnson, cofounder of emotionally focused therapy, EFT; Dr. Frank Dattilio, leader in and refiner of cognitive-behavioral marital therapy, CBMT; Dr. Richard Schwartz, founder of the internal family systems model, IFS), their former couple therapy clients, their former students (except for Dr. Dattilio), and the students’ former couple therapy clients were all interviewed via telephone. These interviews were taped, transcribed, and analyzed using a modified grounded theory methodology that employed analytic induction, constant comparison, and open and axial coding. A number of qualitative procedures (e.g., triangulation of coders, audit trails) were used to ensure credibility and trustworthiness of the findings.
Participants in the study are referred to using a nomenclature in which the first three letters refer to the model used (e.g., EFT), the second two letters delineate between model developer (DV) or student (ST), the third two letters specify “client” (CL), and the fourth letter specifies husband (H) or wife (W) where appropriate. For example, CBT∼DV∼CL∼W refers to the wife in Dr. Dattilio’s client couple.
Results
Model-Independent Themes
Model-independent themes are those elements of therapy that appear to be related to outcome but are not directly related to the model, such as the therapeutic relationship or therapist and client attributes. They are themes that are inherent in the nature of the therapy process itself, and would be present in therapy regardless of which model the therapist utilized. Model-independent themes fell into the following five categories: (a) therapist variables; (b) client variables; (c) the therapeutic alliance; (d) therapeutic process; and (e) expectancy and motivational factors. Each category had several subcategories. We will describe each of these themes and their subcategories below.
Therapist Variables
Each therapist had a unique personality and style. Still, several common themes emerged as the first author asked the clients in this study to describe their therapists.
Patience: respecting client’s pace. Each therapist showed a respect for the client’s pace of change by being patient with the change process. The therapists moved therapy along, but they did not push clients to go faster than they were comfortable. For example, taking a long-term perspective on her client’s change process helped Ms. O’Neil (IFS∼ST) demonstrate patience with her client’s pace. She said:
[We did the same things] over and over and over again … As a therapist you … have to kind of have the perspective [that] it may not be in this lifetime …
IFS∼ST∼CL mentioned that as Ms. O’Neil modeled patience, she (the client) was more able to be patient with herself:
[Ms. O’Neil was] patient in the short run and in the long term. So she could be very patient if in a session [in which] I was just flooded with parts or extremely compartmentalized or something like that; she could so patiently work with that and not get frustrated.
Caring yet firm and boundaried. The clients knew that their therapist cared about them, although “caring” did not always look the same. When one thinks of a “caring” therapist, images of a soft, warm, empathetic, Rogerian type of therapist often come to mind. This was not always the case in this study. Once each client knew his or her therapist cared, each therapist asked him or her to take risks and make difficult changes. This was usually done in the context of a warm, soft approach, though at times it involved taking a deliberate firm and direct stance, as was the case with Dr. Dattilio. He said that he “was very hard on [his male client]. I was very direct; probably the way he wanted his father to be, but in a loving way.” Though to an outsider this deliberate stance may seem uncaring, it had its intended effect of joining therapeutically, as CBT∼DV∼CL∼H notes when I asked him to describe Dr. Dattilio:
He … says, “You need to knock it off and knock it off now.”… I find him more fatherly [and] stern. I [also] find him to be … understanding and empathetic. I’m not saying he’s some [jerk] or something like that. I like him and I’d go back to him, but he’s not soft by any means.
Another element of being caring was being well boundaried. Therapists were able to show their care for their clients while at the same time maintaining healthy boundaries. IFS∼ST∼CL mentions how her therapist balanced caring with boundaries:
She had amazing boundaries. On a few occasions over the six years I distinctly remember … times when tears came [to] her eyes. And yet it was never one of those things like, “I need to take care of my therapist,” because she had the most amazing boundaries and the most solidness.
Cultural and religious sensitivity. Therapists were aware of and responsive to cultural and religious diversity. This strengthened the therapeutic relationship, and allowed therapists to do work with— instead of against—these important areas of their client’s lives. Dr. Johnson said that her clients “grew up in an Eastern-European culture which I know quite well and where I don’t think they had a very supportive environment to grow up in.” Her familiarity with the culture allowed her to remain present with the couple when they would escalate. Her inclusion of culture in the therapy sessions was apparent as her clients often framed their difficulties with being emotionally vulnerable as having cultural roots.
Bridgette’s (IFS∼ST∼CL) husband, Mohammed, was from Jordan, which presented the couple with several cultural issues. For example, when he found out that she had been sexually abused by her brother, he could not understand why she still talked to him. In his culture, the brother would have been forever banned from the family. Similar issues arose around an abortion Bridgette had had, as well as her decision to cremate her mother’s remains. Bridgette summed up Beth’s competency in helping her navigate these issues simply by saying, “She’s very culturally aware.”
Client Variables
This category and its subcategories were one of the most pervasive, clear, and distinct themes of the entire study. However, as the clients were fairly homogenous (i.e., well educated, relatively high socioeconomic status, Caucasian), readers are urged to exercise caution when transferring the results of this study to a more diverse clientele.
Humility. Clients in this study were all able to acknowledge by the end of therapy a shared ownership of the problem and its resolution. Furthermore, they were willing to change once they became aware of their part in the relationship problems. Tiffany (CBT∼DV∼CL∼W) stated:
We had the desire as a couple to make improvements, and we were both open enough to listen to where we needed to improve. And that’s very difficult to do, because it’s easy to always say it’s the other person’s fault and that I’m perfect and you’re the one that’s flawed (laughing). … Both of us have been willing to take a good look at who we are and where our insecurities and immaturities brought problems to the marriage.
Clients showed their humility by being willing to take risks and make themselves vulnerable to their spouse and the therapist. Paul (EFT∼DV∼CL∼H) stated that he and his wife were willing to “take off our armor during the session and basically practice what [Dr. Johnson] teaches.” Similarly, when asked what it was about her that made therapy successful, Bridgette (IFS∼ST∼CL) said, “… I’m extremely authentic; I take risks.” William (IFS∼DV∼CL) said that
[I figured that] if you don’t let your guard down in your therapy session, what’s the point of being there? … I found that I had to make myself very vulnerable …
Commitment and hard work. Most clients were in therapy for well over a year; several were in therapy for several years. All appeared committed to the process. For example, EFT∼ST∼CL∼W stated that “there was never in my mind any question of leaving the marriage. This was never an option and I suppose that I was willing and interested and motivated to find a way to work this out and get through this …” Similarly, CBT∼DV∼CL∼H said “… [My wife] was [committed] and so was I. That’s a good word—‘committed’—we both were very committed to it. So when they told us what to do and told us the techniques, [then] that is what we did.”
The length of time the clients in this study spent in therapy may make them atypical. Faced with managed care constraints, many clients are not able to remain in therapy as long as the clients in this study. Although the perseverance of the clients in this study may affect the results in unforeseen ways, we believe that the client variables discussed are likely factors of change even for clients who remain in therapy for a shorter time.
Psychologically/systemically aware. Clients were able to grasp abstract concepts that represented their psychological and systemic difficulties and use those abstract concepts to guide their change. While humility and commitment may be fostered by the therapist, it would appear that psychological/systemic awareness may be a more static trait. It may also be related to the class and education level of the clients of the prominent therapists in this study, and thus more obvious in this sample than in clients in general. It may be that clients who are not psychologically and systemically aware may respond more positively to some approaches than others. For example, there is research suggesting that lower SES clients prefer more directive approaches than their higher SES counterparts (Cline, Mejia, Coles, Klein, & Cline, 1984).
The Therapeutic Alliance
Most of the therapists mentioned that establishing a caring relationship with their clients was a necessary precursor to working effectively together. When asked how he was able to help his clients change, the first thing Dr. Dattilio said was, “by establishing a rapport.” Similarly, Dr. Makinen (EFT∼ST) said, “I think [it] really [comes] down to the alliance.”
Isomorphism. In each case, the therapeutic alliance served as a model for the client’s relationship. Attributes that the clients needed in order to change were modeled by the therapist. For example, Dr. Makinen used the therapeutic alliance to model how to set firm limits (for the wife) and how to validate (for the husband):
… he’s very controlling. And when he requested we pray … I felt it was very important to take the bull by the horns and also model for his wife how to set some clear boundaries around his behavior. I think that was a really important thing; otherwise … he would run the sessions.
Mutual trust and respect. Each therapeutic alliance was characterized by mutual trust and respect. Clients trusted both the therapist and the process of therapy in general. For example, Dr. Johnson’s clients began to trust her because, as EFT∼DV∼CL∼H said, “When we had a tendency to flare up, even during the session, she had [a] magical power to lead us out of that.” EFT∼DV∼CL∼W said that “… [We both] trusted Sue from the very beginning.” Dr. Johnson mentioned that “… they trusted me a lot, so they let me work with them on a very refined level at times.”
IFS∼ST∼CL began to trust her therapist when she saw that her therapist could competently deal with the client’s “parts” that she was so afraid of. This trust born of competence was isomorphic—the client also began to trust in herself as her therapist helped her experience those parts and effectively deal with them.
Therapists and clients also had a mutual respect for each other. For clients, the development of respect for their therapist seemed to closely follow the development of trust—it largely hinged on successful experiences in therapy. Therapists seemed to develop respect for their clients as they saw them working hard and taking risks in therapy. When I (first author) asked Dr. Johnson to choose four or five words to describe her relationship with her clients, she said, among other things, “Respect … I knew the culture they came from and I knew what it took to take those emotional risks.”
Other times respect developed naturally over a long period of working together. When I asked Ms. O’Neil (IFS∼ST) the same question, she said, “I would say ‘respect.’ We really liked each other. I think there was … some level of [a peer-to-peer relationship].”
Therapeutic Process
Structure/flow balance. In general, therapists were careful to strike a balance between structuring the session and allowing the session to “flow.” The therapists were like a coach or facilitator—most of the work in therapy was done by the clients, but within the structure that the therapist set up. IFS∼DV∼CL illustrates this principle well as he discussed what Dr. Schwartz did that helped therapy be productive:
He actually did very little, but what he did was remarkable. My job in a parts therapy session as I see it was to go in there, tell him a little bit about the external circumstances of my life, get in self, find a part that needed to be talked to, and have self talk to the part … But basically what Dick would say, especially at the beginning, was, “No, you’re not in self yet. That’s a part.” Or, he would get me … to get that part to step back … So, just very gentle guides like that.
The degree of the client’s emotional reactivity appeared to be a guide in how much the therapist structured the session. For example, as her client, Charles, became less emotionally reactive (primarily through hearing himself talk on the tape recorder), Dr. Makinen structured the session less and allowed them to interact on their own more.
Neutrality. Therapists maintained a sense of neutrality over the course of therapy, refusing to enter into triangles with their client(s). This balance held true for the overall course of therapy more than it did session-by-session. For example, Ms. O’Neil’s (IFS∼ST) client mentioned that “When I would be working with relationship stuff with [Ms. O’Neil] she never took sides. … she always had … empathy with Mohammed and I simultaneously.”
Repetition. Repetition was another factor common to the therapeutic process. Both clients and therapists mentioned that they did very similar things over and over again in therapy. If the clients saw what was being repeated as relevant to their problems (a common factor discussed later), the repetition seemed to give them a sense that the practicing helped them move forward. CBT∼DV∼CL∼W, in referring to the communication skills that she repeated with Dr. Dattilio over the course of 15 months, said that “[the skills are] coming more naturally now because I’m doing [them] more frequently. We’ve noticed a tremendous decrease in the amount of turbulence.”
Repetition also appeared to help clients continue the work of therapy in their daily lives. Bridgette (IFS∼ST∼CL) stated that
… as time went on, I would be able to ask myself the very same questions that [the therapist] would ask me in therapy when working around parts. It was almost like I could hear her voice in my head, saying things like, “Okay, who is that? How old is that part? When did I first remember having this feeling?” So I was able to internalize the process and employ it on myself just from years of doing it week after week.
Collaboration and accommodation. Therapists fostered an atmosphere of collaboration and accommodation. Instead of refusing her client’s request, Dr. Makinen (EFT∼ST) accommodated her client’s desire to pray to begin each session as long as it was done on her terms. She described this as “meeting the clients where they are at.” She further accommodated her client’s religious beliefs by framing her theory-specific interventions (e.g., identifying and processing primary emotions) in spiritual terms. She said that the effect of this was “validating, because he would talk about how God works in his life. And I was able to understand that and validate his fears and frame it in terms … less threatening [to him]. …”
Safety. Most clients mentioned feeling safe when they were with the therapist. This seemed to be largely a product of knowing what to expect in therapy and trusting that the therapist was competent enough to not let things get too overwhelming. CBT∼DV∼CL∼H attributed much of their success to Dr. Dattilio’s ability to “[create] an atmosphere with two willing partners that were going [to] listen. …”
Safety was also a product of the client’s willingness to take risks to be vulnerable. A safe environment resulted as clients risked their emotional safety by making themselves vulnerable and the therapist structured the session in a way that fostered this expression (usually by focusing on having the spouse hear his or her vulnerable partner differently). EFT∼DV∼CL∼H describes the safety he felt this way: “Sue … created a nonthreatening atmosphere; she encouraged … sharing our vulnerability.”
To Ms. O’Neil (IFS∼ST), establishing a safe relationship was a top priority. When I asked her what she thought her client would say helped her change, she said, “I think she’d probably say safety, … the sense of feeling safe and trusting. I tend to believe … people heal more by the environment we create for them to work with their parts in.”
Expectancy and Motivational Factors
Faith in the referral source. All clients in the study were referred to their therapists by someone whom they trusted. This had an influence on their expectation that therapy would work. William’s (IFS∼DV∼CL) girlfriend, who was a therapist, knew of Dr. Schwartz’s work and suggested that William see Dr. Schwartz. William said, “I didn’t ever have any doubts that [therapy] was going to work [because] I totally trusted this person that I was having this relationship with, and she totally trusted Dick. Because of those two levels of trust I trusted that the therapy was eventually going to take me where I wanted to go.” All of the clients mentioned being referred by a trusted friend or health professional, which led them to presuppose that their therapist was competent before actually meeting them.
Perception of the therapist as competent. Being referred from a trusted referral source seemed—initially at least—to influence the client’s perception of the therapist as competent. Clients began therapy assuming that it would work, and their therapist’s competence bolstered that assumption. CBT∼DV∼CL∼H described Dr. Dattilio as “fair, smart … pinpoint accurate, and … knowledgeable.” EFT∼DV∼CL∼W paid Dr. Johnson the following compliment: “It’s just that she is just so gifted. She is … able to intuitively guide you through difficult situations.” When I asked her to describe her therapist, IFS∼ST∼CL simply said, “I could go on and on about things Beth did to help therapy be productive.” Each client had similar things to say about his or her therapist’s competence.
Fit of the model. Each therapist provided a viable explanation of the client’s problems (i.e., the model) and—through his or her nonanxious presence and the fit of the model to the client’s problems—gave the client hope that the therapist had seen this before and could provide a way out. The therapist’s model provided structure and order to the client’s chaos, which gave the client hope. The model helped them make sense of their experience. CBT∼DV∼CL∼W, for example, said, “I didn’t even realize [that the marriage had become polarized] until we started in therapy and it was pointed out that [the polarization was] exactly what was happening. And then once it was pointed out to us … it was like, ‘Wow, that’s exactly what’s happening and has been for years.’ ”
Dr. Dattilio oriented his clients to his model, but the clients did not know that was what was happening—they did not have a session where he sat down and explained the model to them (at least not that they could remember). Instead, the model was explained through what was focused on in the dialogue between the therapist and clients. This was true of EFT as well. None of the CBT or EFT therapists didactically “taught” their model per se, but rather taught it naturally through the conversations and experiences they had with their clients. Ms. O’Neil (IFS∼ST) and, to a lesser extent, Dr. Richard Schwartz, were the only therapists who explained the theory to their clients, presumably because an understanding of the IFS framework is necessary for clients to be able to use the model as a guide through their problems. Regardless of how they learned the model, it does seem important that clients become oriented to a model that gives them a credible explanation for their problems and provides a way out.
Motivational beliefs and experiences. Each of the clients mentioned having certain motivational beliefs or thoughts that helped them continue working in therapy even when things were difficult. Some of these beliefs were presented by the therapist, often as a reframe to their struggles. Others were beliefs that the client already possessed. For example, both Dr. Makinen and Dr. Dattilio’s clients mentioned that because they had children and had been married for so long they wanted to continue the marriage. Regardless of where the beliefs originated, the important factor appears to be that they gave the client a credible reason to resolve their problems in their current relationship no matter how difficult the work became.
One common belief was that going through difficult times successfully would increase their love for each other. IFS∼ST∼CL said that “with each setback that Mohammed and I had it eventually brought us greater trust and deeper love, even if that didn’t take place right off the bat.”
Dr. Johnson proposed a similar belief about working through difficulties in a relationship:
You have to [take emotional risks] if you’re going to build a real deep basis of trust. You also have to do it if you’re going to repair rifts because … rifts are inevitable in any relationship. It’s actually when you have a fight or a rift when the repair is often the place where the relationship really starts to grow. It’s the repair process where the relationship starts to grow.
Though Dr. Johnson’s clients did not mention this belief verbatim, they did echo the general theme of it several times.
Dr. Dattilio proposed a similar belief in an attempt to prepare his clients for the difficult work that lay ahead. CBT∼DV∼CL∼W recalls, “They were pretty up-front about it and said, ‘You’re going to enter some periods where you’re not going to like this.’ ” Her husband said, “They said to us, ‘it’s going to get worse before it gets better.’ ” This helped them continue on when the work was difficult.
Discussion
How Change Occurs: A Synthesis of Model-Dependent and Model-Independent Themes
We have used the final data analysis results—the creation of relational statements—to develop an integrated common factor framework for understanding the therapeutic process from a common factor lens. That is, we will speculate on how each of the common factor categories and subcategories may relate to each other to bring about change. Since this final theory-building stage of data analysis involves moving from description to interpretation, it is often the most complex and—inherent in its originality—potentially subjective portion of the data analysis in modified grounded theory.
The model-dependent variables (i.e., common conceptualizations and common interventions) and common outcomes appear to be in a sequential order. This is, in part, because such a linear explanation is unavoidable when explaining separate phenomena in writing (i.e., one has to come first, second, and third). It is also in part because, in general, that is how therapy naturally progresses, and that is how the participants in our study described it. Keep in mind, however, that the progression through each of these stages is more circular—one stage informs and is informed by the other. For example, the conceptualization phase is interventive in that it gives the clients hope. In addition, there is no distinct line between where the intervention phase ends and the outcomes start; they often co-occur. As we discuss the interaction between elements of each stage and the model-independent themes we will italicize the common factor being discussed, where relevant. If the reader wishes to know more about the italicized common factor, he or she can refer to where we discussed that common factor earlier or to Davis (2005). With this in mind, we present in Figure 1 a diagram outlining the model derived from the data.
How model-dependent and model-independent themes combine to create change.
The Yellow Brick Road Map Phase
At least two phenomena seemed to characterize the very beginnings of therapy. First, at some level, the clients seemed confused as to how to help their relationship (client conceptualization). Efforts to change on their own had failed. Second, the therapists in this study had a clear idea of what was “dysfunction” in a couple, what was “health,” and how they could get their clients from the former to the latter. They knew how to help (expectancy and motivational factors).
Regardless of the model utilized, the therapist begins therapy by searching for cues that signal dysfunction in the couple. For the EFT therapists, these cues could be attachment injuries, secondary emotions, and so forth. For the CBT therapist, these could be irrational thoughts and self-defeating behaviors. For an IFS therapist, these could be extreme emotional reactions to each partner or one’s self or parts. The list could go on endlessly for every MFT model in existence. However, rather than detailing model-specific elements of each model, the purpose of this study is to provide a framework within which to make sense of change happening across specific treatments. Though there were several model-specific differences, two common themes seemed to characterize the cues therapists in this study searched for: (a) family of origin or previous relationship influences on current behavior; and (b) interactional cycles in which those early influences are played out and current problems are perpetuated. The cues they pay attention to appear to be in the context of at least one of these two categories.
Once a therapist starts to notice cues that signal dysfunction, he or she begins to present these—and/or healthier alternatives—to the couple. The couple may begin to adopt the therapist’s explanations as an adequate replacement for the chaos that characterizes their current conceptualizations provided that the following conditions exist: (a) the therapist is viewed as credible (expectancy and motivational factors); (b) the proposed explanations (i.e., the therapist’s model) fit the couple’s experience and address issues realistically related to relational health and dysfunction (expectancy and motivational factors); (c) the therapist proposes his or her explanations in a way that minimizes client resistance (therapeutic process); and (d) the clients are at least somewhat willing to take personal responsibility for their part in the relationship (client variable).
Once the couple adopts the therapist’s conceptualization, the clients seemed to begin to feel hope (expectancy and motivational factors). They seemed to be encouraged that there was a way out of their current situation that was achievable because their therapist has seen it before and was calm about the situation (expectancy and motivational factors). The clients may begin to accept the therapist’s conceptualization of the problem (yellow brick road map).
The Intervention Stage
The client’s adoption of a model is an intervention in and of itself, as it seems to provide clients with hope and the beginnings of a way to make sense of the chaos that has become their relationship. However, more formal interventions follow the initial yellow brick road map phase.
A therapist intervenes by using model-specific and common methods and interventions. Most of the interventions seemed to be aimed at altering affective, cognitive, and behavioral elements of the interactional cycle between the two partners (interactional cycle; common interventions), though each model seemed to emphasize a different aspect of the cycle. For example, EFT, CBT, and IFS therapists all may notice when a client reacts with anger when his spouse rolls her eyes. They may also notice how these two behaviors likely perpetuate one another, or are at least a small part of a larger cycle of behaviors that perpetuate one another. Each therapist might inquire about these behaviors in some fashion, and may expect to find that there was some historical root to the related affect, behavior, and cognition whether it is in the family of origin or a previous relationship. Perhaps when the husband’s mother used to roll her eyes the husband felt dismissed and devalued. Now, when he sees the same behavior, he reacts with anger and tries to get the wife to stop. Perhaps when the wife’s father would get angry, her mother would turn her head and roll her eyes in disdain, then later talk to the daughter about how weak her father was for getting so angry. In this couple, the behaviors, cognitions, and affect of each partner perpetuate those of the other. So, the cycle would be like the cycle in Figure 2 (with an arbitrary starting point).
Illustrative dysfunctional interactional cycle.
A therapist from each model may intervene at different levels of the cycle. The EFT therapist would likely focus on how the display of secondary emotion perpetuates the cycle; the husband’s anger invites the wife’s contempt, and vice versa. The EFT therapist will fashion interventions aimed at helping each partner identify, own, and express his or her primary (e.g., hurt and fear) rather than secondary (e.g., anger and contempt) emotions, with the assumption that the expressed primary emotions will evoke the same from his or her partner, thus altering the cycle and allowing them to heal their attachment injuries (Johnson, 2004).
The CBT therapist may be more likely to focus on the automatic thoughts, schemas (i.e., in this case, deep-seated cognitions about relationships) and behaviors that perpetuate the cycle (Dattilio, 2005). He or she may help the clients explore alternative explanations to their partner’s behavior, and help each partner explore different ways of responding to each other, which would thus alter the cycle. The IFS therapist (Breunlin, Schwartz, & Mac Kune-Karrer, 2001) may pay more attention to cognitive and emotional aspects of the cycle. He or she may help partners explore the beliefs associated with each “part” that was reacting so strongly to their partner’s “part.” This could help them explore different ways of interacting with each other.
So, while the entry into the cycle varies across models, they all appear to be focusing on altering an aspect of the cycle. A change in one aspect of the cycle seemed to be associated with changes in the others. Therefore, the curative element may not be which aspect the therapists focus on or how they intervene; rather, the curative element may be that they systematically focus on altering the cycle.
Regardless of the therapists’ entry point into the cycle, most of their interventions seemed to serve the following purposes (see Figure 1): (a) to slow down the process (common intervention); (b) to help the couples stand meta to themselves in the cycle, thus experiencing themselves and their partners differently (common intervention); and (c) to encourage personal responsibility by changing their stance in the cycle (common intervention). Therapists seemed to use a myriad of model-specific interventions to facilitate these goals. Therapists seem to help their clients stand back from themselves by slowing down the process. As couples begin to slow down, they seem to be better able to stand meta to themselves in the cycle—to see their own role in the relationship problems, as well as their spouse’s role. Once they can acknowledge their own role in the cycle they seem to be encouraged to take personal responsibility for changing their stance in the cycle, whether it is via altering affect, behavior, cognition, or some combination of the three.
The success of the above-mentioned tasks seems to be determined largely by the model-independent variables discussed in this article. For example, important factors may include the degree to which the clients trust that the therapist is acting in their best interest (therapeutic alliance), whether or not the therapeutic relationship is isomorphic to the goals of therapy (therapeutic process), the level of safety of the therapeutic environment (therapeutic process), the degree to which the process is repeated (therapeutic process), the therapist’s ability to present interventions in a way that is direct but does not elicit resistance (therapeutic process), a client’s willingness to accept personal responsibility for his or her role in the relationship problems and commitment to work on the relationship (client variables), and the client’s ability to grasp psychological and systemic concepts (client variable).
The Outcome Stage
Not surprisingly, the categories that were the target of intervention also seemed to be categories in which clients stated they changed. Therapists seemed to focus on shifting affect, behavior, and cognitions in the cycle by slowing down the process, helping their clients stand meta to themselves and their partners, and helping them take personal responsibility for changing their stance in the cycle. As therapists did this, and clients were open to the process, their interactional cycle seemed to shift from being destructive to healing. As clients made these changes, they seemed to experience a shift in how they thought, felt, and acted toward their partners (softening; see Figure 3 for an illustration of a healthy interactional cycle). They began to be kinder to each other. They seemed to replace attempts to control or withdraw from their partners with nurturing their partners’ autonomy. They seemed to increase in self-confidence, and their rushed, anxious approach to life slowed down. As they stood meta to themselves and saw themselves and their partners differently, they seemed to be able to slow down and stop trying to control each other.
Healthy interactional cycle.
Clinical Implications
The major clinical implications from this study come from the change in lens that the proposed framework applies. Instead of exclusive focus on model-specific tenets, the therapist has a supplemental road map, one that can be superimposed on a wide range of existing models. In this supplemental framework, the therapist focuses on the combination of common factors mentioned above. For example, does the clinician provide a credible model that points toward healthy functioning and how to achieve it? Does the therapist reflect competence and provide tools by which clients can break repetitive cycles of behavior? Is the therapeutic alliance strong? Can the therapist help the clients stand meta to their process? Can the therapist enhance the expectancy and other motivational factors clients need to move toward change and a more softened state in the interactional cycle? Such questions can guide therapists, regardless of their theoretical orientation.
This study also highlights client resourcefulness (Duncan, Solovey, & Rusk, 1992). Perhaps clients are more flexible than we give them credit for given that we sometimes insist that we should use one model with all clients. Despite receiving different forms of therapy in different formats, the clients in this study tell us that they achieved their goals of establishing a healthier relationship. While this certainly reflects our selection criteria, it also points to a more general truth—clients benefit from a wide range of therapies, particularly if they view their therapist and his or her model as offering a credible explanation for how they got into their problem and how to get out of it.
Findings from a study of EFT (Johnson & Tallitman, 1997) support this assertion. In searching for mediators of successful therapy in EFT, they found that “couples most likely to be satisfied after 12 sessions of EFT and at follow-up were couples who made a positive alliance with the therapist and, more specifically, who saw the tasks of EFT, which promote emotional engagement, as relevant to their problem’’ (p. 146). Outcome was mediated by the therapeutic relationship, and the strength of the relationship was largely determined by the credibility the clients lent to the therapist’s problem conceptualizations and interventions. Clients in our study mentioned a similar phenomenon.
Repetition is another important concept in helping people change (Helmeke & Sprenkle, 2000). Many clients in this study mentioned that it was after doing the same things over and over again that they started to change. This study suggests that patience may be needed more than a new model when change is not happening as quickly as hoped.
This study supports other research that highlights the importance of the beginning stages of therapy. Studies suggest that between 56% and 71% of the outcome variance is attributed to changes made in the early stages of treatment (Fennell & Teasdale, 1987; Howard, Lueger, Maling, & Martinovich, 1993). In this study, many clients’ hope that change could happen was bolstered in the initial sessions of therapy as their therapist shared his or her conceptualization of the problem with them. The clients believed that their therapist had seen their problem before and was still calm about it, so surely there must be hope. This appeared to be related to whether or not the therapist’s conceptualization fit the clients’ experience of their problem, and their perception of the therapist’s competence. Their perception of the therapist’s competence was enhanced by the positive comments made by the referral source.
Butler and Gardner (2003) mention that therapists should provide more structure in therapy when clients are emotionally reactive, then lessen the structure as clients become better able to regulate their emotions and sustain conversations on their own. Similarly, in this study we saw the therapists match their level of directiveness to the clients’ emotional reactivity. As therapy progressed with most of the couples in this study, the therapist became less and less involved in structuring therapy as the clients grew in their ability to regulate their emotions.
Research Implications
Much of the research on couple therapy has focused on which model works the best. The common factors model presented here suggests that our research be more comprehensive. We need to look more closely at the role of common factors in the change process. In this study we called attention to the importance of understanding common aspects of both distressed and healthy couples so a clinician can know where to target treatment and to what effects treatment should be aimed. Future research could explore the extent to which an MFT model’s conceptualizations of dysfunction and health meshes with the literature on distressed and healthy couples, and whether the mechanisms the model proposes as taking a couple from dysfunction to health really work as intended.
There are a number of other common factors and processes that deserve closer study. For example, does the strength of a referral's recommendation influence the process and outcome of therapy, and if so, how? Are certain clients better matches with certain therapies than others? What roles do expectancy and motivation play in predicting outcome? These are examples of only a few of the empirical questions suggested by the results of the present study. In short, any of the model-dependent or model-independent themes in this study could be independent variables in a quantitative study measuring clinical outcome.
It would also be interesting to study common factors of treatment failure. Knowing what happens when things go wrong is arguably as useful as knowing what happens when things go right. Finding areas of overlap in both conditions could shed further light on which variables are truly related to outcome and which are ‘‘little bits of superstitious behavior here and there’’ (Kazdin, 2001, p. 147).
Since all of the models in this study are arguably modernist, similar qualitative research investigating postmodern therapies would be interesting. Such research may confirm certain aspects of our model (e.g., interactional cycles) and disconfirm others (e.g., family of origin).
Future research that identifies and links relational processes to health and change will facilitate a broader, more comprehensive, and potentially more useful understanding of the change process.
Training Implications
Current COAMFTE guidelines place a strong emphasis on teaching models of therapy. Several authors, including the first author, (Davis & Butler, 2004; Nichols & Fellenberg, 2000; Sprenkle & Blow, 2004a) have proposed that the teaching of models should be de-emphasized, and be replaced with more instruction on common factors. Based on the results of this study, we now believe that the teaching of models in MFT training programs should be supplemented rather than de-emphasized (Blow, Sprenkle, & Davis, 2007). The clients in this model were helped largely because the therapists had a model that provided a road map for how to help. Taking away that road map and replacing it with only the model-independent common factors does not seem warranted. However, the role of model-independent factors should not be ignored in training. Instead, we believe it would be helpful to alter the way we teach models to more accurately reflect the way they seem to work; namely, as a vehicle of change rather than the sole contributor to change. Currently, individual models are often taught as mutually exclusive, each with their own proponents. (Peggy Papp used to refer to this in her workshops as the “battle of the name brands.”) The results of this study throw this practice into question. Instead, this study supports a training approach in which students have a thorough grasp of several models, but a humility that goes with the understanding that a variety of common factors are also at work, common factors that they can and should learn.
Several themes emerged from the data relative to common therapist characteristics, and the effect those characteristics had on the therapeutic alliance and in-session process. These results suggest that therapists’ training could be enhanced by an increased focus on the self of the therapist (Asay & Lambert, 1999; Davis, 2005). Educators could focus, for example, on helping therapists become more caring, warm, direct, and boundaried. We agree with Asay and Lambert’s (1999) assertion that “changing the emphasis in graduate training toward the development of the therapist as a person who prizes others can only make the enterprise of therapy more valuable, meaningful, and effective” (p. 49).
There is an increasing focus on therapist competencies in the training literature (Nelson & Smock, 2005). Many of the model-independent variables found in this study are in the therapist’s control (e.g., therapeutic alliance, building expectancy for change) and could easily be integrated into a competency-based training approach.
Integration of Findings With the Current Common Factors Literature
The findings of this study confirm, challenge, and expand current common factors theory. We will discuss how this study meshes with common factors literature in psychology and in MFT.
Integration with common factors in psychology. This study generally supports and expands Lambert’s (1992) often-cited common factors conceptualization in psychology. Lambert attributes the largest part of the outcome variance, 40%, to “extratherapeutic variables.” These are events completely outside of the therapist’s control. Such events could include job changes, moving homes, formations of new friendships, and so on. In fact, when considering that Lambert also ascribes 30% of outcome variance to the therapeutic relationship, 15% to expectancy or placebo effects, and 15% to model-specific techniques, only 30% of the outcome of therapy can be influenced by the therapist (i.e., 15% for the therapist’s half of the therapeutic relationship and 15% for model-specific techniques). Seventy percent of the outcome of therapy may be outside of the therapist’s control!1
This study suggests that Lambert’s (1992) earlier estimates of the therapist’s influence on therapy may be conservative. Though we could not comfortably assign a percentage, all of the clients made it clear that they could not have achieved the changes that they did without the help of their therapist. For example, IFS∼DV∼CL said, “There’s absolutely no way that I would have been in a place of this kind of internal clarity … if I hadn’t been working with this kind of model or something like it as [fervently] as I had.” (Note that the client’s motivation, a common factor, is simultaneously reflected in this statement).
This study also suggests that therapists have some control over what Lambert (1992) referred to as expectancy and placebo effects. The referral source, which is outside of the therapist’s control, did seem to influence whether or not the clients initially expected to improve. Once clients began therapy, however, their hope increased as the therapist provided them with a way of thinking about their problems that provided an explanation for how they got where they were and what they could do to change. As previously discussed, clients reported that seeing their therapist as calm when they told him or her about their problems gave them hope. In short, therapists have the ability to enhance initial expectations, a point seldom discussed in the current common factors literature.
The strength of client variables over therapist variables suggested in the common factors literature may not be that clear-cut. For example, a poor therapist may thwart even the most motivated client, and a good therapist may be able to motivate an unmotivated client. Having said that, the results of this study generally confirm the viewpoint of Miller, Duncan, and Hubble (1997) that “the research literature makes it clear that the client is actually the single, most potent contributor to outcome in psychotherapy” (pp. 25–26). Dr. Dattilio captured this point with an analogy:
It’s no different than physical therapy. I was paralyzed … five or six years ago. I had [an operation] and I learned a lot about physical therapy. … My recovery came as a result of busting my ass. The harder I worked, [the more I overcame] … everything. I’m back to where I was. I mean you really have to work your ass off. And if you do, you make head way and if you don’t then you live with what you got.
We suggest that a model will likely be effective if it (a) orients the therapist to credible aspects of dysfunction; (b) provides a clear definition of a healthy relationship; and (c) provides a clear operational map for how to help a client from dysfunction to health. Frank and Frank (1991) believed that a credible ritual or procedure is one element of all effective therapies, and have proposed a similar explanation of how a model contributes to therapy. They suggest that a credible ritual consists of the following six elements in which the therapist (a) combats the client’s demoralization and alienation by establishing a strong relationship; (b) links hope for improvement to the process of therapy, which heightens the patient’s expectation; (c) offers new learning experiences; (d) facilitates emotional arousal and reprocessing; (e) facilitates a sense of mastery or self-efficacy; and (f) offers opportunities for the client to practice new behaviors.
Integration with common factors in MFT. This study further confirms, refines, and expands contemporary frameworks of common factors in MFT (Sprenkle & Blow, 2004a). Sprenkle and Blow confirm Lambert’s (1992) four categories of common factors as being applicable to MFT as well, and then propose the following three factors as common factors unique to MFT: (a) relational conceptualization; (b) the expanded direct treatment system; and (c) the expanded therapeutic alliance. This study supports their “relational conceptualization” category, and further refines this category by proposing a framework that incorporates a relational conceptualization.
This study also provides support for Sprenkle and Blow’s (2004a)“expanded direct treatment system” and “expanded therapeutic alliance” categories. Even when working with individual clients, the work by the therapists and clients in this study was always done in the context of the client’s relationship with his or her partner, and was characterized by familiar couple therapy processes (e.g., therapists’ refusal to triangulate). This study expands the current MFT common factors literature by providing a conceptual model that proposes how model-dependent factors interact with each other sequentially to bring about change in MFT, and how those factors are mediated by model-independent factors.
This study also sheds light on the interaction between affective, behavioral, and cognitive factors in MFT (Sprenkle & Blow, 2004a). Proponents of EFT and CBT differ over whether it is more important to focus primarily on cognitive or affective elements of the couple's experience in an attempt to shift their interactional cycle. Cognitive therapists propose that working on cognitions gives clients something concrete to focus on that they can change, which will in turn change emotions. EFT therapists propose that a client’s emotional experience is a more powerful organizing factor in relationships, and should therefore be the focus of intervention. Neither needs to try to change the other’s viewpoint. In this study, changes in one element (i.e., affect, behavior, or cognition) seemed to either co-occur with or be closely followed by changes in the other two elements. Perhaps the choice over whether to emphasize affect or cognition is a “both/and” rather than an “either/or.”
Sprenkle and Blow (2004a, 2004b) propose what they call a “moderated common factors approach.” This approach differs from traditional common factors approaches (Duncan & Miller, 2000; Wampold, 2001) in that it does not disparage the use of treatment models and comparative efficacy research. Rather, Sprenkle and Blow propose that treatment models are the vehicle through which common factors operate, and as such are necessary components of effective therapy. This study supports their assertion, and provides additional detail into how having a model facilitates the therapy process (e.g., by increasing client’s hope as the therapist’s model-informed conceptualization fits the client’s experience).
Common factors researchers and proponents of specific models have long debated the usefulness of treatment manuals (Hubble et al., 1999). Proponents of treatment manuals claim that the manuals are necessary to provide enough detail about the model to allow clinicians to be able to use it competently and to perform research on the model that is reliable and generalizable (Crits-Christoph & Mints, 1991). Common factors researchers claim that treatment manuals inhibit creative processes, and that the change process is too unique from client to client to be put into “cookbook” form. We believe that the better a clinician knows different models, the more freedom, flexibility, and potential creativity he or she will have in therapy (cf. Blow, Sprenkle, & Davis, 2007). Well-written treatment manuals could only deepen this knowledge, as long as they are not rigidly adhered to in a “cookbook” manner.
Strengths and Limitations of the Study
As with all studies, this study has several strengths and limitations inherent in its design. One of the main limitations in this study is the client sample. In general, the clients in this study were Caucasian, had a relatively high socioeconomic status, remained in therapy for a long time, and were well educated. Although the variables that resulted from the study are likely applicable across a diverse clientele, such a homogeneous sample may limit the transferability of the results in unforeseen ways. Therapists, educators, and researchers using the results of this study to guide their work are encouraged to remain open to new data that may come from a more diverse clientele.
Additionally, the therapists chose the clients interviewed in this study because they did well in therapy. Though this could be interpreted as “these are people so motivated that they would have gotten better going to anyone,” they are also clients we can—and did—learn a lot from.
A challenge facing a qualitative grounded theory researcher is to minimize the effects of his or her preferences on the reporting of the data. The data need to reflect as closely as possible what the participants say. The patterns and themes that I (SD) see in the data are inevitably colored by my preferences, beliefs, and past experiences. Namely, my interest in common factors may have led me to ask questions that confirmed my preexisting beliefs and preferences. This is an inescapable aspect of qualitative inquiry, and although I took standard precautions to minimize this effect (e.g., triangulation of data, constant-comparative method of data analysis, and so forth), it is quite possible that another researcher would find different themes simply because of our varying backgrounds.
We included former students of model developers in order to examine the practice of a “second generation” of the clinical models we studied. Of course, it is likely that the model developers recruited their “star” former students. Therefore, while the student’s practice may be representative of his or her mentor, it may not be as representative of the average clinician. We did note that the former student’s practice seemed less “textbook” than the model developer’s practice, yet that in no way diminished the former student's effectiveness. Dr. Makinen’s use of a tape recorder to help her client stand meta to himself is an example of this—such an intervention would not be found in any EFT training manual, but was consistent with the goals of EFT.
Another limitation of this study is the inability to be certain about causal links between any of the variables in the study and outcome. Simply because we found common variables across theories does not mean that all of these variables are necessarily causally linked to outcome. Clearly, controlled experimental studies are needed to answer this question.
We sought to establish credibility and trustworthiness with a long and careful data analysis, thick descriptions, triangulation with multiple researchers, and illustrative quotes. Nevertheless, we did not conduct member checks with the participants once the study was completed. Although we hope this is not the case, it is possible that participants will disagree with the way their ideas were construed.
Our study focused on therapists’ reflections about the therapy rather than behavioral observations of the actual therapy itself. Therefore, our data represent what participants remember and not necessarily what actually happened. Therefore, it is important to remember when we refer to therapy, we are really referring to memories of therapy. Of course, perceived reality is the reality we all live in, and is consequently a rich resource for qualitative theory building. Future quantitative research that observes the actual therapy may expand and refine our model. We hope that someone will join us in undertaking that research.
In addition to its limitations, this study also has several strengths. Specifically, the majority of the common factors literature—both in psychology and MFT—has been deductively derived. This is the first study that we are aware of that specifically investigates common factors from an inductive, qualitative perspective. As such, the variables in the study include several subcategories not previously detailed in the common factors literature. Additionally, the model derived from the study builds on contemporary models of common factors in MFT (Sprenkle & Blow, 2004a) by adding, expanding, and refining current factors, as well as providing a framework within which to understand them.
Reflections on Interviewing the Founders
I (SD) was honored that each therapist and client who participated in this study took time out of their busy schedules to talk with me. The model developers in particular took a risk participating in this study, as they each understood that this was a study on common factors. They did not know if my analysis of their interviews would reflect favorably on them or their model. They did not know if I would be respectful of their model, or if I would twist their interviews to further my own agenda. I appreciate the fact that they trusted me, and I attempted to honor that trust by being respectful of their models. Model developers are often criticized by common factors researchers as being solely devoted to furthering their model at the expense of a more comprehensive understanding of change. Regardless of the veracity of those critiques, the model developers' participation in this study on common factors implies their commitment to a broader study of change rather than a fervent promotion of their models.
Several themes stuck out to me as I interviewed the model developers. First, I was impressed with the model-specific clarity with which they articulated their practice. It was very interesting to hear them describe how their clients changed as their theory says they should, and then to hear their clients say the same thing. From the interviews, they really do practice exactly as they say they do. It made the practice of therapy as it is portrayed in textbooks seem much more real to me. It was also interesting to interview their former students, who each had a style of practice distinct from the developer's style. Most of them were not as ‘‘textbook’’ in their approach as were their trainers, yet they were effective nonetheless. Hearing both styles work bolstered my faith in the power of a well-trained therapist to effect change in people's lives.
I was also impressed with the passion the therapists had in their work with their clients. It was obvious that they were each invested in helping their clients overcome their problems. Their clients all noted this passion; it seemed to spread throughout the client system. Their clients seemed to feel comfortable with them in large part because the therapists believed in what they did so fervently. This passion for their clinical work was evident in the interviews as they would often become excited when they were explaining a concept to me. In some instances it felt as if I were being taken on a guided tour of something they had just discovered and were still excited about. I found their excitement for ideas invigorating. It was fun to feel their passion for their clinical work. I believe that their passion is a large part of the reason that their clients in this study improved.
I was struck by the fact that each of the therapists mentioned how important they perceived the therapeutic relationship to be. Most of them said that they believed the relationship they establish with their clients is vital to the success of therapy. They mentioned that establishing a therapeutic relationship was on the top of their list of priorities with their clients. They also mentioned how important it was that the clients are motivated and willing to engage in therapy. They willingly acknowledged that successful therapy was more than having a good model.
Conclusion
The current state of the common factors literature in MFT is characteristic of an emergent school of thought, as there are less than a dozen published articles pertaining specifically to common factors in MFT (Davis & Butler, 2004). It is no surprise, then, that the current literature does not represent a comprehensive theory. Sexton et al. (2004) state:
Without question, finding a common core of factors to explain successful therapy would be a major breakthrough. This finding would simplify practice, training, and research. It would unify the theoretical schools of MFT, which often compete against one another and find themselves in contentious struggles. In essence, it would serve as a shorthand explanation for the complexity of practice and the diversity of clients, settings, and the sometimes disparate research findings. (p. 131)
These are indeed grand and worthy goals. We hope that this study moves the field one step closer to Sexton et al's (2004) ideal.
Therapeutic frameworks are not helpful if they are either so specific that they allow no room for flexibility or are so vague that they provide no direction for complex situations. The framework based on our results is somewhere in the middle. We hope that it might generate meaningful research and training. We also hope that it is specific enough to guide practice, yet broad enough to capture commonalities among different MFT theories and model-independent aspects of the therapy process. Finally, we hope that others will join our efforts to further refine common factors theory, research, and training.
