Health
What Does "Evidence-Based" Mean in Therapy?
Part I: "Evidence-based" doesn't always include evidence for 'hows' and 'whys.'
Posted April 7, 2022 Reviewed by Vanessa Lancaster
Key points
- "Evidence-based" means evidence for whether a therapy works not how.
- Efficacy of therapy is determined by clinical trials like in medicine, but this doesn't tell you much about how the treatment works.
- Understanding how a therapy works, requires "primary research" from things like psychological studies, not just clinical trials.
Nowadays, therapies are very keen to describe themselves as "evidence-based." Cognitive behavioural therapy (CBT), one of the most popular schools of therapy, is often advertised in these terms, as is mindfulness.
But "evidence-based" actually means something much narrower than most people think. The evidence we have for therapies tells us something about whether therapies work (and even then, this evidence is complicated) but very little about how or why they work.
For example, in an earlier blog, I showed that though we have evidence showing that CBT works, there's far less evidence for its understanding of mental illness and its therapeutic approach. We might wonder how we ended up in a situation where we know that something works without quite knowing how.
The term "evidence-based" comes from the evidence-based medicine (EBM) movement, which started in the 1960s in both the US and the UK (though, as with most things in history, it has much earlier roots). EBM arose out of a desire to develop a systematic way of understanding which treatments work and when to use them.
Before EBM, medicine relied on the expertise of doctors and other medical professionals, who 'vouched' for particular treatments based on their own experience and expertise. As you might expect, there were a lot of problems with this method, extending from not having reliable guidance for when to use a particular treatment to fraud.
Evidence-based medicine works mainly through clinical trials, more specifically randomised control trials. These trials involve randomly allocating (to prevent bias) patients with a particular condition to two groups. One group gets the treatment tested, and the other 'control' group is usually given a placebo, an inactive version of the treatment.
If the non-control group (the one who gets the active treatment) gets better, we have reason to think it is the treatment that is making them better – provided the study has been designed properly, the main meaningful difference between the two groups is whether they had the treatment or not.
But this is all clinical trials tell you. They tell you that a treatment works, but they don't say much about how the treatment is working. Research into those questions – sometimes referred to as "primary research" or research into the "mechanism" or "process" – is done differently from research into efficacy using clinical trials.
Primary research may involve everything from biological experiments using animal models and psychological studies to looking at genetics or brain imaging, depending on the treatment being studied. This means that we actually don't have a full picture of how even very familiar medicines, like aspirin, work – we know what they do and what kinds of things they can treat, but not always how.
Research into therapy is in the same situation. So when someone tells you that a particular school of therapy is "evidence-based," they usually talk exclusively about evidence from clinical trials. This is not to say that there is no primary research into therapy, but the vast majority of the evidence for therapy is from clinical trials.
Additionally, evidence from clinical trials is sifted for quality and pooled into bigger reviews or analyses to give a richer picture of whether a therapy is efficacious. In contrast, though there are individual primary research studies into the question of how or why these therapies work, there are comparatively far fewer of these bigger analyses and reviews.
The fact that we have a hazy picture of how treatments work is not necessarily a bad thing. After all, when there's a problem, it makes sense to focus on figuring out a reliable way to solve it, then look into the hows later. But therapy is different from other areas of medicine in a way that makes the 'how' question quite a bit more important.
In explaining how many schools of therapy often put forward a theory of why people are mentally ill or suffer from psychological distress in order to explain their therapeutic approach. For example, CBT suggests mental illness and psychological distress are related to problems with reasoning ("cognitive distortions") and claims that CBT can help fix those reasoning errors.
Because therapies put forward a story about mental health and illness, it affects how people think about themselves and what might be going wrong when they are having a difficult time with their mental health. This story is likely to be taken very seriously, when the people putting it forward are experts who describe their therapy as evidence-based.
We can already see how much of an impact CBT's understanding of mental health has had when looking at how widespread terms like "catastrophizing" and "reality-testing" have become. Unfortunately, CBT's story about mental illness is not as well-evidenced as its efficacy.
Given that theories about mental health have such a profound impact on how people think about and understand themselves, both clients and clinicians should be clear on what "evidence-based" means when talking about therapy. Though we have evidence from clinical trials that these therapies are efficacious in treating a host of conditions, the same does not go for their theories about mental health.
This post is based on academic research in a journal article titled "Why Theoretical Adequacy (not just Therapeutic Efficacy) Matters," forthcoming in Philosophy, Psychology, and Psychiatry.
