What makes a good therapist?

There is a solid research base to the key principles of change offered by good therapy. Is there a similarly robust research base to show us what makes a good therapist?

We all know what makes a good therapist, don’t we? It’s quite simple. They’re calm, they’re consistent, they’re non-judgmental, they’re empathetic. 

Yes and no. Those qualities are essential. But the empirical research suggests they’re insufficient to be a good therapist. 

A big risk is that we assume we know what makes a good therapist, and we assume that we’re one of those good therapists. After all, we’re human, and we’re all prone to cognitive biases.

Cognitive biases like unrealistic optimism, that leads more than 90% of us to think we’re better than average drivers. But that’s impossible: only 50% by definition can be above average at anything.

So, we’re also likely to think we’re better therapists than we actually are, and some of our current ways of working reinforce those biases. For example, most supervision consists of us describing our clinical work ourselves - often designed, consciously or unconsciously, to present ourselves in the best light, without objective evidence of our sessions (though some might argue that verbatim notes come close). Compare that to other professions, where you work with a coach and together use empirical evidence to implement the marginal gains that lead to continuous improvement in your performance and outcomes. The small tweaks to your golf swing; the small tweaks to the way you approach the ball when taking a free kick; the small tweaks to the way you perform heart surgery. British Cycling would not have won many Olympic golds if they left the cyclists to describe their own performances to their coaches. Team GB won 67 cycling medals at Rio 2016, and 65 medals at London 2012, up from 18 medals in 2008 and as a direct result of their focus on marginal gains.

We know that therapists do differ in the way that they practice. So it’s somewhat inevitable that those differences matter. But which ones? And how do we examine our own ways of working and seek to improve our work and thereby, our client outcomes?

Let’s start by defining what we mean when we talk of “therapist effects” on client outcomes. Therapist effects are reflected in changes in client mental health such as symptom relief, interpersonal functioning, quality of life improvements. These are things we can measure, even if those measurements are inevitably imperfect. 

‍

We often talk about being “good enough” in therapy, but too often apply perfectionism to client outcome data that risks throwing the baby out with the bathwater. Yes, client self-report data is imperfect. But who better to tell us whether we’re being helpful than our clients? If we’re concerned clients are trying to please us, or that they’re not consciously aware of their behaviour, we can look to significant others for information about, for example, our clients’ emotional regulation and expression and mood.

Fundamentally, we can glean a lot from the research on therapist effects, while acknowledging its limitations.

There are two findings consistently identified across studies. Firstly, that therapist effects contribute approximately 8 to 10% to client outcomes. Secondly, that the best therapists are much better than those who are lower than average. One study found that the 13% of therapists that were significantly more effective than average had recovery rates twice those of therapists identified as less effective than average. The best therapists continued to have a significant impact with their clients over time; the least effective had less impact the more sessions they had. The best therapists were most effective with the most distressed clients. 

So what do the best therapists do? Interestingly, some things that we are often told matter most seem not to matter much at all. Demographics don’t matter. Your age, your gender, your experience doesn’t matter. Let’s pause on the significance of that for a moment. Merely having more hours with clients under your belt doesn’t seem to impact on your client outcomes anywhere near as much as what you do with that experience.

The evidence suggests that four variables separate the best therapists from the least effective. The first, and best supported, is the therapist’s ability to establish a good working relationship with their clients: empathy, goal consensus, positive regard, genuineness and appropriate challenge, repairing relationship ruptures. The three supporting and overlapping variables are interpersonal skills, ‘self-doubt’ and deliberate practice.

Self-doubt doesn’t mean anxiety. It means humility. Understanding that we’re likely to be really helpful for some clients and not others, understanding that we’re likely to have self-development gaps. Deliberate practice is about those marginal gains: acting like Sir Chris Hoy, taking all the evidence we have available - be it videotaped sessions, client self-report data, the opinions of our supervisors - and focusing our Continuing Professional Development time well. Experience might not make as much of a difference as we supposed, but well-targeted practice certainly does. Are we missing opportunities to notice and repair relationship ruptures? Are we working with our clients to establish clear goals we can work together towards? Are we shying away from challenging our clients when they need us to, because we want to remain in our comfort zone as their supporter and cheerleader?

A lot of this is complex stuff, requiring humility, thought and nuanced discussion with our own support team, including our supervisors. We’ve put together a short but comprehensive training on therapist effects to get you started.