The What and Why of the Therapeutic Alliance

The therapeutic alliance is central to outcomes, yet how that relationship is regarded varies between therapeutic traditions. In this article, the relationship is viewed through the lenses of CBT and Rogerian/humanistic therapy.

Related article: Essential Qualities in the Therapeutic Alliance.

Introduction

“In my early professional years, I was asking the question: How can I treat, or cure, or change this person? Now I would phrase the question in this way: How can I provide a relationship which this person may use for his own personal growth?” (Carl R. Rogers)

The therapeutic relationship is how a therapist and a client connect, behave, and engage with one another. It is about how they build their relationship together, and is sometimes referred to as the bond that develops in the therapy room. It can enable confidence, reassurance, openness and honesty, paving the way for clients to accept themselves for who they are.

Without this important alliance, there can be no effective or meaningful therapy. This is especially true for clients who may have experienced neglect, abuse, or other traumatic events (for more in this topic, read Assessing and Treating Trauma) in their early years, leading to difficulty establishing and maintaining relationships in adulthood. Therapy allows clients to explore their attachments and bonds through their relationship with their therapist (O’Neil, 2016; Counselling Tutor, 2021; Miles, 2017).

Why is the therapeutic alliance so important?

“I’m in therapy to learn how to deal with people who should be in therapy.”

Take a moment to reflect on relationships you have which you value, where you feel comfortable and can be yourself, “warts and all”, with the other person. Chances are that a sense of safety, acceptance, and trust are central aspects. Sadly, some clients seek therapy because they have never – or rarely – had those sorts of relationships. They do not have the lived experience of acceptance for who they are instead of what they do. Good relationships help us meet our needs, and we may seek to meet a different need from one person than what we seek from another. Here is an exercise on relationship that you can either do for yourself or offer to your client (related topic: Why Mental Health Literacy Matters).

Exercise: Two relationships reflection

For this exercise, focus on two relationships and how you see yourself in each of them.

First, focus on a relationship you have with someone that may be somewhat regular, but is not particularly close. This could be someone such as the shopkeeper where you regularly shop, a neighbour, or someone such as a fellow co-worker or member of a committee you serve on, or a member of a group you are involved with. Spend a few moments tuning into the relationship.

1. How do you perceive this relationship? Describe it in a few words.
2. How do you see yourself when you are with this person?
3. How do you feel around the person? What is your behaviour like?
4. What needs do they meet for you?
5. How might you act when you really want something from them?

Now, think of a second relationship: one that is closer (whether it is a conflictual relationship or a relatively harmonious one). Answer the same questions:

1. How do you perceive this relationship? Describe it in a few words.
2. How do you see yourself when you are with this person?
3. How do you feel around the person? What is your behaviour like?
4. What needs do they meet for you?
5. How might you act when you really want something from them?

Take a moment and reflect on how you are in each of these relationships. What do you learn about yourself and your needs?

What does the research say about therapeutic alliance?

The notion of a therapeutic alliance was first entertained by Sigmund Freud in 1913, when he hypothesised that the relationship between therapist and patient was a primary component of successful therapeutic treatment (although he originally thought that most of the relationship was about “transference” and only later revised his approach to include the understanding that there could be genuine relationship between patient and therapist). The first formulation of therapeutic alliance using that term was around 1956, and it has been evolving since then. Researchers have devised many scales and measures to examine it, and there is now a huge database of research linking quality of the alliance to therapeutic outcome (Ardito & Rabellino, 2011).

Fortunately, the picture is consistently clear. A 1991 meta-analysis evaluated 24 studies to demonstrate a “moderate but reliable association between good therapeutic alliance and positive therapeutic outcome” (Horvath & Symonds, 1991). More recent meta-analyses confirmed these results and also showed that the quality of the therapist-client relationship was “more predictive of positive outcome than the type of intervention” (Martin, et al, 2000; Shirk & Karver, 2003; Karver et al, 2006; all in Ardito & Rabellino, 2011). Finally, a meta-analysis published in 2020 studied 37 articles with significant variation in methodology. The analysis posed the question of whether the therapeutic alliance “mediated” the therapeutic outcome and found that, 70% of the time, it did (Baier, Kline, and Feeny, 2020).

Most of the scales used to measure the alliance have examined personal attachments between therapist and client, and collaboration and desire to invest in the therapeutic process. Some researchers have suggested that future research should pay special attention to the comparison between clients’ and therapists’ assessments of the alliance; these have often been found to differ, and evidence suggests that the client’s assessment is a better predictor of the outcome of therapy (Castonguay et al, 2006).

The alliance through several lenses

To some extent, the therapeutic relationship is viewed as a “common factor” in therapy: something that all therapies have to deal with, no matter what school of philosophy or theoretical stance they hail from. In delineating the elements that lead to change, Howard Kassinove and Raymond Tafrate, in The practitioner’s guide to anger management (2019), list client, therapist, technique-specific, and “common” factors. The last category refers to aspects that occur across all models of psychotherapy, such as the context of therapy, the presence of an acceptable rationale for the interventions to be used, the client-therapist relationship, expectations for improvement, and perceived social support. These, they assert (citing work by Johnsen and Friborg, 2015, in Kassinove & Tafrate, 2019), account for fully 40% of the change (with technique-specific factors only accounting for 15%).

Thus, the alliance is acknowledged as important, and yet the various strands of therapy view it quite differently. Here we summarise how the alliance tends to be viewed in the cognitive behavioural therapies and the Rogerian/humanistic stance.

CBT and the therapeutic relationship

A long debate on a nonspecific variable

Castonguay, Constantino, McAleavey, and Goldfried, in their article in The therapeutic alliance: An evidence-based guide to practice (Castonguay et al, in Muran & Barber, 2010) acknowledge the long discussion/debate that has surrounded CBT from its inception. Even though CBT has “from the beginning recognised the importance of the therapeutic relationship in the change process” (p 150), it has only recently given considerable attention to it in the research literature. Castonguay and colleagues make these points in explaining the evolution of how the relationship has been perceived within the cognitive behavioural therapies:

• CBT historically considered the alliance to be a “nonspecific” variable, essentially an interpersonal factor auxiliary to the specific variables that actually produce change (and its therapists acknowledge that, in the 1950s and 1960s when CBT therapies were beginning to develop, there was no empirical support for the importance of the relationship outside of the Rogerian variables of empathy, warmth, and genuineness).

• Wolpe and Lazarus famously concluded that a client’s positive emotional reaction toward a therapist would engender “nonspecific reciprocal inhibition”. Translated, that means that the presence of the therapist reduces anxiety and therefore facilitates the aim of desensitisation through specific behavioural techniques.

• The alliance in CBT is different (in theory) to other orientations in that CBT emphasises collaboration and teamwork more than many other therapies. The model of “collaborative empiricism” has emerged, meaning that client and therapist work together to identify the central problems faced by the client and to identify possible solutions; this is as opposed to Gestalt therapy, where the client has a less dominant role, or client-centred (Rogerian) therapy, where the power for change is deemed to be in the client’s hands, but there is not as direct a method for moving past distorted perceptions to verifiable data.

• In the CBT model, two “scientists” are working together, one – the client – providing the “raw data” and the other – the therapist – guiding the research questions.

• A sound alliance is built through the implementation of CBT techniques in a clear, structured and supportive manner. The corollary understanding to this is that the alliance has been treated as a factor that facilitates the use of and adherence to specific techniques, but not as a change mechanism in itself, so it becomes a “necessary but not sufficient” change factor (p 154).

• Challenges to the alliance are met by working on the client characteristics that are negatively impacting the relationship through CBT methods, such as maladaptive thoughts or avoidance patterns, and also by manipulating client expectancies for treatment success.

Techniques to resolve therapeutic rupture

Techniques of resolving rupture include “listening skills” and the therapist recognising and expressing their contribution to the problem, which functions as a disarming technique, encouraging the client to express their feelings about treatment. Discussion of the client’s contribution to the therapeutic impasse may lead by extension to examination of the client’s interpersonal problems outside of therapy. Therapists are also encouraged to reconsider their case conceptualisations and assumptions about the client in the case of therapeutic breaks. These options constitute the longstanding CBT techniques for addressing relationship (Castonguay et al, in Muran and Barber, 2010). In summary, the therapeutic alliance in cognitive behavioural therapies has a role as facilitator of the all-important CBT techniques and as a here-and-now bond between therapist and client which allows discovery of the client’s maladaptive patterns.

The alliance in Rogerian/humanistic therapy

In contrast to the cognitive behavioural therapies, the therapeutic relationship has long been central to humanistic practitioners’ paradigm for therapy, with Carl Rogers’ writings focusing primarily on the therapist attitudes and behaviours that would be optimal for facilitating client change in psychotherapy. Rogers acknowledged that the therapist-client relationship was asymmetrical – with the therapist holding the greater balance of power – even when the therapist fully respected the client’s autonomy, and acted ideally in a compassionate, authentic, and egalitarian manner. As well as the bond or positive feelings that would develop between client and therapist if the therapist conveyed these optimal qualities, the humanistic construct of the working alliance deemed that therapist and client should jointly generate the goals and tasks of therapy, collaborating on decision-making.

Six conditions for personality change overlapping with therapeutic alliance

Finally, we note that Rogers deemed six conditions essential for personality change to occur; these overlap greatly with the notion of the therapeutic alliance and include:

• That two people (here, therapist and client) be in psychological contact.
  
• That the client be in a state of incongruence, being vulnerable (why would they turn up to therapy otherwise, unless mandated?)
  
• That the therapist be congruent (as above, the client needs to see that modelled).
  
• That the therapist experience unconditional positive regard for the client.
  
• That the therapist experience an empathic understanding of the client’s internal frame of reference and endeavour to communicate this understanding to the client.
  
• That the client perceive, at least to some extent, the therapist’s unconditional positive regard and empathic understanding (Watson & Kalogerakos, in Muran & Barber, 2010).

A central role for the therapeutic alliance in Rogerian/humanistic counselling

The Rogerian/humanistic approach to the therapeutic alliance is, in essence, the approach of that therapy to the whole of counselling; it is that central. Within the mostly non-directive modus operandi, therapists strive to balance guidance with acceptance, remaining in the role of the client’s “handmaiden”, and continuing to see the client as the expert on their own experience.

Psychodynamic therapies look into relational impairment from early years

Both the CBT and humanistic strands of therapy, heavily present-focused, have relatively less inclination to look deeply into causes of impairment resulting from relational patterns originating in the early years. For that, the psychodynamic therapies are most useful; the roles of attachment and transference are primary. Though those approaches are outside of the scope of this article, we heartily recommend looking at the therapeutic alliance through those lenses as well.

Key takeaways

• Research consistently finds that, regardless of type of therapeutic intervention or method, the therapeutic relationship – the bond between therapist and client – is crucial for effective therapy.
  
• In CBT’s collaboration between therapist and client, a sound alliance is built through the implementation of CBT techniques in a structured and supportive manner; the alliance is not seen as a change factor in itself.
  
• In Rogerian/humanistic therapy, the therapeutic relationship is deemed central to client change, with Rogers elaborating six conditions necessary for change to occur.

References

• Ardito, R.B., & Rabellino, D. (2011). Therapeutic alliance and outcome of psychotherapy: Historical excursus, measurements, and prospects for research. Frontiers in Psychology. Retrieved on 1 March, 2021, from: https://www.frontiersin.org/articles/10.3389/fpsyg.2011.00270/full%C2%A0
• Bair, A.L., Kline, A.C., & Feeny, N.C. (2020). Therapeutic alliance as a mediator of change: A systematic review and evaluation of research. Clinical Psychological Review, Vol 82, Dec 2020, 101921. Retrieved on 1 March, 2021, from: https://www.sciencedirect.com/science/article/abs/pii/S0272735820301094
• Castonguay, L.G., Constantino, M.J., & Grosse Holtforth, M. (2006). The working alliance: Where are we and where should we go? Psychotherapy (Chic) 43, 271-27910.1037/0033-3204.43.271 [Google Scholar]. Retrieved on 2 March, 2021, from: (PDF) The working alliance: Where are we and where should we go? (researchgate.net)
• Horvath, A.O., & Symonds, B.D. (1991). Relation between working alliance and outcome in psychotherapy: a meta-analysis. J Couns. Psychol.: 38, 139-14910.1037/0022-0167.38.2.139 [Google Scholar]
• Kassinove, H., & Tafrate, R.C. (2019). The practitioner’s guide to anger management: Customizable interventions, treatments and tools for clients with problem anger. Oakland, California: New Harbinger Publications.