Online Courses and CE: We offer a series of online educational programs for professionals and the public. Visit us here for previews and discounts on our online programs.

Follow PsychologySalon on Facebook: Become a fan of the PsychologySalon page; updates will appear in your news feed.

Looking for a therapist? We have eleven registered psychologists in our clinic, and we are accepting new clients. For information, visit

Tuesday, 6 August 2013

The Therapeutic Alliance: A Benefit or a Liability?

When talking about self-help or internet-based approaches to psychological therapy or self-management, people routinely point out a problem with such therapist-absent approaches: The demonstrated value of the therapeutic alliance.

The alliance is a general term for the complex relationship between therapist and client, composed of mutual understanding, empathy, an agreement on the goals and methods of change, and so on. Studies routinely show that the quality of the alliance is a stronger predictor of therapeutic change than is the modality of psychotherapy being delivered.

If this is the case, then self-care focused strategies have a real problem. How can there be a genuine alliance if there is no real relationship with the clinician? We might sense that the author of a self-help book or the teacher in an online course is a nice person, but clearly they have no real relationship with the user.

If the potent factor in therapy is the relationship, and we have a therapy with no relationship, how could the therapy be effective? In nutritional terms, it would be missing the crucial vitamin.

Arguments like this presume that the effect of the alliance is, in a sense, additive. Sure, there may be some useful information in a less personal modality like a self-help guide, but the relationship adds something important.

Could the relevance of the relationship be subtractive, instead?

In order to find a relation between alliance quality and outcome, there has to be variability in the alliance. Some therapist-client pairs have to have a better alliance, others a worse one.

We have been presuming that the good alliance adds something extra to whatever is delivered in therapy. What if this is not true? What if, instead, a bad relationship subtracts from the impact of what is delivered? In other words, what if the alliance looks important not because a good alliance is so great, but because a bad one fouls things up so badly?

If the alliance is only subtractive (which I suspect is unlikely), then it would actually be better to have a form of therapy with no alliance at all. A self-help book, a therapy guide, an online lecture-format course, all of which have no off-putting relationship factors, would be the ideal form of therapy.

I raise the issue because of the profound effects clients sometimes report from purely receptive therapy efforts – like reading self-help books. Even my own: I often get notes of appreciation from readers of The Assertiveness Workbook, saying how much it has helped them in their lives. It would be nice to take these accolades as evidence of my abilities as a therapist. “Look how much they benefitted just from reading something I wrote – imagine what they’d get from actually meeting with me.”

It’s a nice thought, but glosses over the obvious anxiety. Perhaps I expressed myself clearly in the book but would be less concise, less organized, and less effective in person. As well, readers of the book can imagine me as any kind of teacher they like. In person they might discover that they don’t like my way of speaking, the shape of my nose, or my obnoxious habit of interrupting. Perhaps I am literally a better therapist on paper than I am in person.

I suspect this is often true. How often have we met our therapeutic gurus and been disappointed with their narcissism, or imprecision, or sloppy technique? Reading their words we can see and understand their points, reflect on how the ideas relate to our own lives, and calmly decide how to put changes into effect. Faced with a sneezing, ill-dressed, imperfect individual, the waters are sometimes only muddied.

And gurus or not, we have all seen individuals dispense undeniably good advice in such a ham-handed manner that no one in their right mind would accept it. Forget my name, act as though you don’t care, shout me down, and then tell me to exercise, be nice to myself, eat my vegetables, and make enjoyable activity a priority in my life. I’ll do the opposite just to spite you.

I imagine that in actuality the effect runs both ways. A good therapist can add to their content with a warm, inspiring nature, and a bad one can subtract from otherwise good material by alienating the client. But we only hear about the former, leading many to assume – almost certainly wrongly – that the alliance is crucial for people to get anything out of a psychological intervention. The evidence on therapist-absent approaches suggests otherwise.

Online Course

Can you perform therapy on yourself? Well, in a word, no. Psychotherapy is inherently an interaction between client and clinician. But you can use many self-care strategies on your own.

Our clinic has developed a cognitive behavioral guide to self-care for depression. Though not a substitute for professional face-to-face care, UnDoing Depression may be a useful adjunct to your efforts.  The preview is below. For 50% off the regular fee of $140 USD, use coupon code “changeways70” when you visit our host site, here.

We also have courses for professionals and for the public entitled What Is Depression, What Causes Depression, Diagnosing Depression, Cognitive Behavioral Group Treatment of Depression, How to Buy Happiness, and Breathing Made Easy. For the full list with previews and substantial discounts, visit us at the Courses page of the Changeways Clinic website.


  1. In response to a query on Reddit about my point:

    Most comparisons suggest that self-care approaches are a bit less effective than direct face-to-face therapy, but some show no real difference. And most self-care approaches are pretty weak entities (the largest body of work has to do with the impact of being told "Go read this self-help book"). We spend a lot of time worrying about face-to-face effectiveness, but our other approaches are generally fairly primitive.

    I suppose my point is that there is a logical fallacy in some of the reasoning around the importance of the alliance. Studies do show that in face to face therapy the quality of the alliance is a (actually 'the') major determinant of outcome.

    The fallacy is assuming that the presence of a relationship is therefore essential. This entails a hidden assumption that the relationship "adds" something important.

    But the research showing the importance of the alliance in face-to-face outcome doesn't address the importance of having a therapist in the room. It simply says IF there is a therapist in the room, the quality of the relationship is important.

    I'm raising the possibility that one reason the alliance is important is that a poor one (or a poor therapist) can degrade the experience below that of a non-face-to-face form of therapy. So one could as easily say that having a therapist in the room is a risk as a benefit.

    My guess is that a face to face encounter is generally a positive addition to therapy, but that the much of the variability in outcome has to do with decrements brought about by the alliance, not just benefits.

  2. What do you think about Yalom's statement that, "The relationship heals?"

  3. I agree that the relationship can do so, and often does. For example, having someone listen to you and take you seriously for an hour can be an unusual and profound experience for some folks, and the positive regard of the clinician can help create a sense of confidence to face the world. But I think even Yalom wouldn't say "ONLY the relationship heals" nor that "The relationship cannot harm."

  4. The human interaction/personal connection through face to face interaction certainly can enhance and benefit the experience. There can be significant benefit for the patient in them discovering that someone actually cares and takes an interest in their problems, that someone believes the problems can be overcome. There is a feeling of trust and security that is missing when simply trying to "self diagnose and heal".
    However, if the relationship between the clinician and patient is a poor one, it can certainly be detrimental to the patient, particulary if they are unlikely to follow up with a different clinician. They may feel that "if this is as good as it gets, why should I continue with this"? It can also be frustrating and unproductive for the clinician.
    Ultimately, I think the benefit of a positive relationship is immeasurable in addition to the directed care that is provided.