Common Factors: Evidence-Based Practice and Recovery

First, we have noted a curious omission in current understandings of certain evidence-based practices—namely, those that involve a relationship between a patient and another person functioning as a service provider (1). There is a long-standing, consistent, and robust evidence base—dating back to 1936 (2)—suggesting that a set of common factors is involved in practically all forms of psychotherapy (some would argue more broadly that these factors are common to most forms of patient-practitioner relationships) and that these factors account for more of the variance in the effectiveness and outcomes of these interventions than any of the more technical, theory-based or targeted components.

These common factors include such relational components as the instillation of hope (and its impact on patient expectations); empathic, nonjudgmental listening, acceptance, and understanding; and the provision of information along with the encouragement and support to use it. Most studies comparing psychotherapies have found that these common factors account for twice as much of the variance in outcomes as any particular technique. For example, Lambert and Barley (3) found that these factors accounted for 30% of the variance in outcome, compared with 15% for particular techniques, with the placebo effect accounting for an additional 15% and spontaneous remission or natural recovery and other unknown factors accounting for the remaining 40%.

Not only have these factors been found to account for more of the variance in outcome than the specific therapeutic techniques that are typically considered “evidence based,” but when the same factors are measured in pharmacology trials (which they seldom are), they also have been found to account for more of the variance in effect and outcome than the specific medication prescribed. For example, in a reanalysis of data from the National Institute of Mental Health’s 1985 Treatment of Depression Collaborative Research Program, McKay and colleagues (4) found that psychiatrist effects (that is, common factors) accounted for 9.1% of the variance in scores on the Beck Depression Inventory, whereas medication accounted for only 3.4%.

This body of research suggests that the most evidence-based components of evidence-based practices that involve patient-practitioner relationships—even perhaps relationships that involve the prescription of medications—are those that characterize the nature and quality of the relationship itself. Although the specific techniques being used may have amassed somewhat of an evidence base attesting to their relative contribution to patient outcomes (somewhere between 3% and 15%), the nature and quality of the relationship itself appear to have amassed a much more robust and consistent evidence base attesting to a contribution of roughly two to three times the magnitude (between 9% and 30%).

We suggest, therefore, that the nature of the evidence used to justify the funding and implementation of many evidence-based practices actually provides more proof for investing in the effectiveness of practitioners who develop trusting, empathic, and hope-filled relationships than for any specific therapeutic intervention per se. In addition, these data suggest that the development of a therapeutic alliance that incorporates the common factors described above represents a set of practices for which the evidence base is more than ample to qualify as being itself evidence based. Yet the development and maintenance of such a healing relationship (5) is seldom, if ever, included in lists of evidence-based practices and is seldom, if ever, considered to be an evidence-based practice by policy makers, system leaders, administrators, funders, or training directors. We wonder why this is the case and whether something should be done about it.

There was a time in community psychiatry, of course, when much attention was paid to training practitioners to develop the skills involved in cultivating therapeutic alliances. This was prior to the time when the use of psychodynamic approaches to the care of persons with serious mental illnesses fell into disfavor. Thus we may have witnessed the babies of trust, empathy, acceptance, and encouragement being thrown out with the bathwater of psychoanalytic theory. Or, as suggested by an anonymous reviewer of a draft of this Open Forum, perhaps the emphasis in randomized controlled trials on standardization of the delivery of highly structured interventions has squeezed out any focus on basic relationship-building skills. In times of resource constriction, it certainly appears to require less time and expense to train practitioners with fewer years of education to deliver these highly structured interventions than to employ more highly trained and skilled professionals. Yet the data suggest that training in evidence-based practices that does not build on a solid foundation of common relational factors is unlikely to be effective.

Perhaps the central role of common factors is considered so obvious that it does not deserve mention. There may be an assumption that all mental health providers, or at least all clinically trained practitioners, are adept in the art of developing therapeutic alliances with their patients and therefore this body of research does not need to be considered on its own merits. Or perhaps there is an assumption that being empathic and understanding are personal qualities that cannot be instilled or enhanced through training or supervision. Either the practitioner was born with a natural interpersonal intuitiveness and sensitivity or he or she will pick it up over time through the role modeling of inspiring teachers or through his or her life experience. But all of these approaches leave the matter largely up to chance, which hardly seems a sufficiently evidence-based approach to disseminating a set of practices that account for two to three times the magnitude of effectiveness of those techniques in which practitioners are routinely trained.

The discussion above brings us to a question about the emerging model of recovery-oriented practices. We can respond in several ways to the question that we are frequently asked about whether there is an evidence base for recovery-oriented practices or where it is to be found (6). In this case, we would argue that recovery-oriented care emphasizes the importance of common factors because they provide a foundation for a trusting relationship through which any and all other interventions may be offered. At our program and at an increasing number of programs across the country that are introducing curricula on recovery-oriented care (for example, through the Recovery to Practice initiative supported by the Substance Abuse and Mental Health Services Administration), practitioners are trained first and foremost in how to instill hope and raise expectations; how to empathically and nonjudgmentally listen, accept, and understand the people they serve on their own terms; and how to provide accessible and useful information about a range of topics (including, most prominently, strategies for self-care) and encourage and support the person in using this knowledge to pursue his or her own recovery.

Rather than assuming that this foundation of trust will already be in place by virtue of the practitioner’s role as a helping professional or his or her natural sensitivity, recovery-oriented practitioners are taught that they will have to earn the trust of a distressed and possibly skeptical person who may have become accustomed to being misunderstood by people because of his or her own anomalous or extreme experiences and because of the stigma and discrimination that continue to affect people in our societies who have such experiences. As a result, attention is paid in training to experiential learning and self-reflection (7), and persons in recovery are employed as trainers to provide practitioners with a window onto the experiences, challenges, and efforts of persons living with mental illnesses (8). Although this training differs substantially from the psychodynamic training that many professionals received in earlier eras, these approaches nonetheless take as their primary target the cultivation of trusting, empathic, and accepting relationships with persons in extreme states of distress.

Finally, an attraction and potential benefit of hiring persons in recovery to be peer supporters is that they instill hope and raise expectations by virtue of their presence and tangible role modeling of the reality of recovery. They also may be inclined to be empathic, accepting, and understanding on the basis of similar life experiences, and they may have derived from these experiences and their own recovery efforts valuable lessons about self-care and community inclusion that they can share with the people they encourage and support.

These two examples of experiential, reflective learning and peer support suggest that we do not need to return to the past to train practitioners in the skills needed to promote common factors. Obviously, more research is needed to determine the degree to which recovery-oriented practitioners and peer staff succeed in embodying these common factors in their practice—and the degree to which this improves outcomes. But as the United States moves toward the provision of patient-centered medicine under health care reform, similar preliminary efforts are already under way in internal medicine that have shown that medical residents can be taught to be empathic and that this can improve not only the quality of care that they provide but also patient satisfaction (9,10). Surely, we might expect to derive the same benefits from training psychiatric residents and other professional trainees in these skills.

While this kind of research is under way, we can point to a place where we might begin to discover the evidence base for recovery-oriented practices: over 75 years of research on the nature, quality, and effects of healing relationships based on respect, empathy, and understanding. Rather than relegate this research to the dustbins of history or assume that such skills are already being practiced, it is time to reclaim this robust body of evidence and to rebuild our practices on the foundation it provides for humane, person-centered, and effective care regardless of who provides it.

This work was made possible, in part, by funding provided by the State of Connecticut’s Department of Mental Health and Addiction Services to the Yale Program for Recovery and Community Health.

The authors report no competing interests.