Does therapy work?

Short answer: Yes.

Research consistently shows that therapy works. But, what do we mean by "works"? In research terms, it generally means that there is measurable improvement in the symptoms or concerns that brought someone to therapy. The measures vary, and the research is not perfect, but the results are pretty consistent that therapy helps people achieve meaningful changes in their lives.

What can be surprising is that the type of therapy (often called the “modality”) doesn't really matter. Specific techniques or theoretical orientations don’t really matter when it comes to creating changes. This finding, sometimes called the Dodo bird verdict, is one of the most replicated but also most debated findings in psychotherapy research. For some specific conditions, particularly OCD, eating disorders, and certain trauma presentations, modality does appear to matter. But across the broad range of concerns that bring people to therapy, no single approach consistently outperforms the others.

Therapy seems to work because the therapist and the client form a relationship, one devoted entirely to the client's objectives for therapy. It is not a facsimile for friendship, or a replacement for other relationships. It is a trusting, private, and personal space where a client can explore their thoughts, emotions, behaviors, desires, history, goals - really whatever they want - with an objective partner invested in their wellbeing.

Therapy has plenty of critics and skeptics. In 1952, psychologist Hans Eysenck published one of the first comprehensive reviews of psychotherapy research. His conclusion was damning: the data failed to prove that psychotherapy facilitated the recovery of neurotic patients. He argued that people got better at roughly the same rate whether they had therapy or not.

It took 25 years for research to offer a convincing counter. A landmark 1977 meta-analysis pooled several hundred controlled studies and found that the average person receiving therapy was better off than 75% of untreated controls. Eysenck’s 1952 conclusion rested on a small number of outcome studies and on naturalistic data about people who recovered without treatment (Cuijpers et al., 2018), and the larger body of controlled research has since pointed the other way.

Eysenck’s later record gives a second reason to check his claims instead of taking them on his reputation. In 2019, King’s College London conducted an inquiry into his work with Ronald Grossarth-Maticek on personality and disease, and judged that 26 of their papers could not be relied on as scientifically sound. The committee cited doubts about how the datasets were collected and results with no parallel in medicine, such as a cancer death risk more than 100 times higher for one personality type. That inquiry did not cover his review of psychotherapy outcomes, so it says nothing about whether the 1952 conclusion was right. Concerns about the Grossarth-Maticek research had been raised since the late 1980s, and psychiatrist Anthony Pelosi, one of its early critics, suggested in 2019 that the research programme had led to “one of the worst scientific scandals of all time” (Pelosi, 2019). Eysenck is not the only chapter in psychology’s research history where published results turned out to be unreliable (we’ll examine these more closely in a future post).

So if therapy works, what makes it work? This turns out to be a harder question than it sounds. Bruce Wampold, one of the foremost researchers on psychotherapy outcomes, has spent his career trying to answer it. His conclusions, laid out in The Great Psychotherapy Debate, challenge most assumptions about what therapy actually is.

His work supports my earlier statement that the type of therapy doesn’t matter. Cognitive behavioral therapy, psychodynamic therapy, humanistic therapy, EMDR — the research consistently shows that no single modality outperforms the others in any meaningful way. What predicts outcome, across every modality studied, is the quality of the relationship between therapist and client. The alliance, as researchers call it. The degree to which the client feels heard, understood, and that the therapist is genuinely invested in their wellbeing.

It is worth noting that the relationship between alliance and outcome is correlational. Some researchers argue that people who are already improving form better alliances, rather than the alliance itself driving improvement. The evidence suggests the relationship is likely reciprocal. The debate continues. Practically, this means if therapy does not feel like it is working, the first step is to say so in the room with your therapist. That conversation is itself part of the process. The relationship is built on honesty and communication. If your therapist doesn’t respond well to feedback or if you feel you can’t be honest with them, then you might want to find a different therapist.