The Therapeutic Alliance and Outcomes

Home > Articles > The Therapeutic Alliance and Outcomes
By John T. Brinkmann, MA, CPO/L, FAAOP(D)

In essays submitted as part of the application process for O&P master's degree programs, most prospective students passionately describe their motivations for choosing the profession. They usually include descriptions of times they have had strong emotional responses when witnessing patients' dramatic functional improvements. Experienced clinicians often describe similar situations when asked why they chose O&P as a career or why they continue to enjoy their jobs many years after making that decision. Prospective students and experienced clinicians alike recognize the impact of our services, and gain meaning and enjoyment from a personal connection with patients. Is it possible that, in addition to being a benefit of the work we do, a personal connection with patients can also have a positive effect on our patients' functional status? This article describes the concept of a therapeutic alliance, makes comparisons between the alliance formed in psychotherapy and in O&P care, describes research studies related to how this alliance affects health outcomes, describes a common instrument used to assess the alliance, and highlights the importance of communication strategies in establishing and maintaining this alliance.

What Is a Therapeutic Alliance?

Therapeutic alliance refers to the personal bond between a clinician and a patient, and a variety of terms, including working alliance and helping alliance, have been used to describe this phenomenon.1 Initially, the term was used in the context of psychotherapy. In 1912, Sigmund Freud referred to the "sense of collaboration, warmth, and support between the client and the therapist," and the word alliance was used to describe this connection in a 1934 paper published in the International Journal of Psychoanalysis.1 Psychotherapists recognize that patients may redirect emotions from childhood relationships onto their therapists— transference—and that therapists may become emotionally entangled with patients—countertransference. Contrasted with these potentially pathological emotional connections, therapists recognize that appropriate emotional connection between a professional and a patient is important to the success of their interventions.

The alliance is an aspect of the helping relationship—but not all of the therapy relationship.

·         Alliance refers to the achievement of a collaborative relationship between helper and the client, meaning that  here is a consensus and willingness on the part of both parties to engage in and do the work that leads to improvement.

·         Alliance is an achievement, a quality of the partnership that develops between helper and the person being  helped. It is dynamic in time and content, meaning it changes over time not only in quality but also in terms of the elements that gave rise to it.

·         There is a broad agreement that this collaborative engagement involves, but is not necessarily restricted to,  consensus over the goals of treatment, a sense of confidence and commitment to the kinds of activities that the helper and helpee engage in as part of the helping journey, and the relationship or engagement is in a context of mutual trust, confidence, and liking of one another. However, different therapies, different stages of the  change process, and different occasions would demand and call for diverse actions and interactions to achieve good alliance.


The concept of a working alliance between a therapist and patient was discussed at length in a 1979 article by University of Michigan psychologist Edward Bordin, PhD.2 Bordin points to the dozens of psychotherapy models involving many different therapy models, and suggests that "the strength, rather than the kind of working alliance, will provide the major factor in change achieved through psychotherapy." He goes on to propose "that the working alliance…is one of the keys, if not the key, to the change process." He also suggests that this concept may be applicable to other types of relationships, such as that between a student and teacher or a community group and leader.2 Bordin describes three features of a working alliance: agreement on goals, assignment of tasks, and the development of bonds. When discussing the therapist's role in establishing agreement with the patient on goals, Bordin emphasizes the importance of helping clients understand their own responsibility and how their thoughts and feelings determine actions and experiences. O&P practitioners must often help patients face uncomfortable realities related to their bodies and within their environments, and must encourage healthy decisions for patients to move forward in the face of specific emotional and physical challenges. The psychotherapy process also requires the patient to engage in specific tasks (e.g., attending to specific emotions, thoughts, and actions) as part of the change process. O&P practitioners frequently give information and directions regarding specific behaviors, which are intended to help patients develop healthy habits related to their overall health, function, and mobility. Bordin states that "…the effectiveness of such tasks in furthering movement toward the goal will depend upon the vividness with which the therapist can link the assigned task to the patient's sense of his difficulties and his wish to change." Similarly, O&P patient education will be most effective if descriptions of patient tasks are closely connected with specific functional and mobility goals on which the patient and practitioner have agreed. For example, patients are often more willing to don compression garments such as cotton elastic bandages or shrinkers when they realize how essential they are to achieve a comfortable socket fit.

The third feature of a working alliance relates to the development of a bond between the practitioner and patient. Bordin points out that the nature of a relationship will vary based on the length of the relationship and the level of intimacy required. Therapists and patients will be "more concerned about liking or disliking each other" if a long-term relationship is anticipated, and a higher level of trust is required "when attention is directed toward the more protected recesses of inner experience."2 Similarly, we recognize that different types of relationships, with different levels of intimacy and trust, are required when providing different types of O&P care. A less intense bond may be required with a patient who is seen once for the delivery of an orthosis that will be worn for a short period of time. A different and deeper bond is required with patients who have sustained traumatic injuries resulting in significant loss of function and independence, especially when care will be provided over the course of many years. Bordin also discusses "the difference between a caretaker and a consultant" in various types of psychotherapy.2 In some cases, a therapist or O&P clinician may need to take a more directive and assertive role in the treatment process. In others, a more passive approach, allowing the patient more independence in making major decisions, will be more effective. An effective approach when working with a fully engaged adult will be different than when the patient is an adult with cognitive deficits or a child.

The importance of the practitioner/patient connection may be clearer in professional relationships (such as those in psychotherapy) that regularly involve prolonged interpersonal interactions related to intimate details of personal life and relationships. To what extent does the therapeutic alliance influence care in physical medicine and rehabilitation?

The Impact of the Therapeutic Alliance on Health Outcomes

A number of studies have been conducted to determine the impact of therapeutic alliance on health outcomes. Hall et al. performed a systematic review in 2010 of thirteen prospective studies related to the influence of the therapeutic alliance on physical rehabilitation outcomes.3 A variety of conditions (including brain, musculoskeletal, and cardiac) and outcomes (including those related to pain, disability, quality of life, depression, and satisfaction with treatment) were studied. The results published in those studies demonstrated that adherence to the treatment plan, clinical instructions, depressive symptoms, satisfaction with treatment, and physical function were all positively associated with the therapeutic alliance. In 2014, Kelley et al. performed a systematic review and meta-analysis of "randomized controlled trials (RCTs) in adult patients in which the patient-clinician relationship was systematically manipulated and healthcare outcomes were either objective (e.g., blood pressure) or validated subjective measures (e.g. pain scores)."4 Since the primary focus for these researchers was "to determine whether training clinicians to improve interactions with patients could improve outcomes," they did not review "studies that manipulated the patient-clinician relationship solely from the patient side." These researchers concluded that "the patient-clinician relationship has a small…but statistically significant…effect on healthcare outcomes." The authors include an interesting discussion of effect size, noting that the effect size of any one healthcare intervention is often small since there are "many factors that impact health outcomes." While the effect of the therapeutic alliance as reported in the reviewed studies may be small, it should still be considered an important aspect of healthcare delivery. Lakke and Meerman's 2016 systematic review of research on the influence of the working alliance "on pain and physical functioning in patients with chronic musculoskeletal pain" included five studies involving 1,041 patients.1 This research continued the work done by Hall et al. by evaluating the methodological quality of the studies and including studies published after 2009. The authors report "a significant effect of working alliance on the outcome of pain severity, pain interference, and physical functioning in all studies."

The measurement of the therapeutic alliance occurred at different times in the studies reviewed. For instance, Lakke and Meerman noted that in the studies they reviewed, the measurement point ranged from after the first treatment to after the rehabilitation program was completed.1 When investigating the relationship between the therapeutic alliance and low back pain in 2013, Ferreira et al. measured the therapeutic alliance after the second therapy session.5 The rationale was that this timing allowed an alliance to form but avoided confusion between the quality of the relationship and the effect of the treatment. The authors considered that a later measurement may have measured the patient's perceptions of and satisfaction with the treatment itself, and not the nature of the alliance with the therapist and how it factored into the treatment outcome. The issue of the timing of the measurement raises important questions about how long it takes to form an effective clinical alliance and whether the relationship is affecting the treatment outcome or vice versa. These are significant concerns for O&P clinicians, since the quality of the devices we provide may change the patients' perceptions of the quality of their relationship with us, and the reverse is likely true as well.

Measurement Instruments

Hall et al. reported that the studies they reviewed used measures of alliance that were "developed for use in psychotherapy," and the reliability and validity in the context of physical rehabilitation had not been confirmed.3 The most common method for assessing the therapeutic alliance in the reviews by Hall et al., and Lakke and Meerman was the Working Alliance Inventory (WAI). This instrument, developed by Adam Horvath, MSW, EdD, professor emeritus of counselling psychology at Simon Fraser University, Burnaby, Canada, and a practicing psychotherapist, is based on the three features of the working alliance Bordin described.6 One long (36 item) and two short (12 item) versions of this instrument are available. When completing the instrument, the patient and therapist describe "some of the different ways you might have thought or felt about" the other individual in the context of their professional relationship. Horvath cautions that "the WAI is not a standardized measure…. Each source of assessment…applies his or her own ‘yardstick' in answering questions….

Trusting, liking, working together, etc., do not have an objective metric.... Comparing means of alliance scores obtained from groups of individuals in different circumstances and contexts is fraught with risks...."6

The WAI is a simple instrument that, with some modification, could be used by O&P practitioners to evaluate therapeutic alliances with their patients. Results must be considered judiciously since the instrument has not been validated for this purpose. It may be best applied as a basis for conversations with patients about the nature of the professional relationship.

Building a Therapeutic Alliance

Each of the features of the alliance identified by Bordin (goals, tasks, and bonds) is important in O&P care. Discussion of the patient's and provider's goals is an important aspect of a thorough initial patient assessment and continues throughout the treatment process. Practitioners work closely with patients to describe specific behaviors related to donning and doffing procedures, wear schedules, and device adjustments, and determine adherence to these recommended procedures during follow-up appointments. Research has been conducted to identify essential aspects of developing an effective therapeutic alliance. Hall et al. reported that "although a few studies have attempted to identify the factors that influence the alliance, there is no conclusive evidence as to which factors are most important. The limited data would suggest that providing positive feedback, answering the patient's questions, and providing clear instructions for home practice are positively correlated with a good working alliance and satisfaction with treatment."3 Kelley et al. reported that "…any intervention designed to improve communication— if effectively employed—is also likely to improve the quality of interpersonal relationship."4


Prosthetic and orthotic practice will always involve a special combination of clinical, technical, and interpersonal skills. Working purposefully to improve our relational and communication skills will not only allow us to enjoy our interactions with patients, but it may also make our treatment more effective and improve their experience and function in practical ways.

John T. Brinkmann, MA, CPO/L, FAAOP(D), is an assistant professor at Northwestern University Prosthetics-Orthotics Center. He has more than 20 years of experience treating a wide variety of patients.


1. Lakke S. E., and S. Meerman. 2016. Does working alliance have an influence on pain and physical functioning in patients with chronic musculoskeletal pain; a systematic review. Journal of Compassionate Health Care 3(1):1.

2.Bordin E. S. 1979. The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, research & practice 16(3):252.

3.Hall A. M., P. H. Ferreira, C. G. Maher, J. Latimer, and M. L. Ferreira. 2016 The influence of the therapist-patient relationship on treatment outcome in physical rehabilitation: a systematic review. Physical therapy 90(8):1099-110.

4.Kelley, J. M., G. Kraft-Todd, L. Schapira, J. Kossowsky, and H. Riess. 2014. The influence of the patient-clinician relationship on healthcare outcomes: a systematic review and meta-analysis of randomized controlled trials. PloS ONE 9(4):e94207.

5.Ferreira, P. H., M. L. Ferreira, C. G. Maher, K. M. Refshauge, J. Latimer, and R. D. Adams. 2013. The therapeutic alliance between clinicians and patients predicts outcome in chronic low back pain. Physical therapy 93(4):470-8.

6. Horvath, A. "Working Alliance Inventory."