Last updated on 1st May 2015
Is this one question or many? If you see a therapist, how many treatment sessions are you likely to need? Sometimes that's a little like asking "If I go on a journey, how long should I travel for?" Happily though, we do now have enough research evidence to be able to respond fairly helpfully to this "how many treatment sessions?" question. To give useful answers though, it’s probably sensible to break the very general “how many sessions?” query into a number of more targeted sub-questions.
These sub-questions could include “On average, how quickly am I likely to feel appreciable benefit?”, “How many sessions will it take to achieve full recovery?”, “Are there some kinds of problem that are likely to need more treatment sessions than others?”, “Is there any point in getting ‘top-up’ sessions after the main therapy has finished?”, “How would I know if the therapy isn’t working?” and “If the therapy isn’t working, what’s the best thing to do next?” Let's look at these sub-questions in a bit more detail.
How quickly am I likely to feel appreciable benefit? A good place to look for answers to this question is in the 2013 sixth edition of the excellent "Handbook of psychotherapy and behavior change". In his chapter "The efficacy and effectiveness of psychotherapy", Michael Lambert reviews relevant publications and concludes (p.204) "Therapy is highly efficient for a large minority of clients, perhaps 30% of whom attain a lasting benefit after only three sessions." and when monitoring for "reliable improvement ... it appears 50% of patients respond by the 8th session and 75% are predicted to need at least 14 sessions to experience this degree of relief."
"Reliable improvement" is a formal way of measuring "appreciable benefit"; it is not however "recovery". In the "Handbook of psychotherapy and behavior change" chapter “Measuring change in psychotherapy” (Ogles 2013) it’s highlighted that two of the “most prominent definitions of clinically significant change include: (1) treated clients make statistically reliable improvements as a result of treatment (improvement), (2) treated clients are empirically indistinguishable from ‘normal’ peers following treatment (recovery).” The most widely used assessment measure for "improvement" is called the “reliable change index (RCI)”. The RCI was developed over twenty years ago (Jacobson and Truax 1991) and it compares pre-treatment and post-treatment assessment scores while taking into account the precision of the assessment measure being used. “The change is considered reliable, or unlikely to be the product of measurement error, if the change index (RCI) is greater than 1.96.” (p.155). If you've continued to read this far without developing a headache, it's probably helpful to underline that "reliable improvement" is the most widely used measure of "appreciable benefit" and Michael Lambert's comments about "reliable improvement" in the previous paragraph largely refer to the "reliable change index (RCI)". Now what about "recovery"?
How many sessions will it take to achieve full recovery? By "full recovery", we mean here "indistinguishable from 'normal' peers". Trying to attain "full recovery" rather than just "reliable improvement" is important for the obvious reasons that we feel better and function better after full recovery than we do after reliable improvement. It's also important because the chance of relapsing is considerably less after full recovery than it is after reliable improvement - see, for example, "Recovery and subsequent recurrence in patients with recurrent major depressive disorder". Again Michael Lambert's chapter is an excellent source of evidence-based information and he writes "For patients who begin therapy in the dysfunctional range, 50% can be expected to achieve ... recovery after about 20 sessions of psychotherapy. More than 50 sessions are needed for 75% of patients to meet this criterion." Now these are average findings. Some therapists achieve better results than this and others worse. Commenting on one of the relevant major studies, Lambert (p.197) writes "Outcome by therapist showed considerable variability, with the most effective therapist's patients showing rapid and substantial treatment response, while the least effective therapist's patients showed an average worsening in functioning." This variation in therapist helpfulness is such an important issue and I have written extensively about it - for example, in the recent blog post "Fascinatingly, therapists themselves vary considerably in their effectiveness".
Unsurprisingly clients also make a big difference to the number of therapy sessions they need to make progress. As in most areas of our lives, commitment and hard work help hugely. With cognitive & behavioural therapies, encouraging clients to actively work on relevant challenges between therapy sessions adds considerably to effectiveness - see, for example the paper "Meta-analysis of homework effects in cognitive and behavioral therapy: A replication and extension". As the paper "Homework compliance counts in cognitive-behavioral therapy" shows, committed clients who really tackle these between-session challenges make better progress. You see the same pattern in mindfulness training, with the recent paper "The effects of amount of home meditation practice in mindfulness based cognitive therapy on hazard of relapse to depression" showing that clients, who work to do the formal daily meditation practice, achieve better results than those who don't.
Are there some kinds of problem that are likely to need more treatment sessions than others? Yes! In general clients who want help with more severe problems will probably need more treatment sessions to achieve recovery. By more severe, I'm referring to variables like the extent & intensity of symptoms at presentation, how long the problem has been going on for, and how resistant it has proved to previous attempts at therapy. Relevant too is the overall quality of the client's life ... so it's not just the severity of the presenting problems that govern speed of response, but also the strengths & resources the client has more generally in other areas of their life. “Is there any point in getting ‘top-up’ sessions after the main therapy has finished?” Often this is the case, especially for problems that have gone on longer before therapy was started. “How would I know if the therapy isn’t working?” and “If the therapy isn’t working, what’s the best thing to do next?” These are very important questions. Again I have written extensively about these issues - see the two recent blog posts "Routine outcome monitoring can really help therapists clarify where they need to try harder" and "Practice-based evidence can complement evidence-based practice so very well".
So what are the overall messages? "If you see a therapist, how many sessions are you likely to need?" is a really relevant question, not least because it's likely to help having some idea of the size of the journey before embarking on it. Reiterating Lambert's review findings, quite a few clients (possibly about 30%), may only need three or so sessions to achieve very worthwhile benefits. About 50% are likely to have achieved worthwhile benefit after eight sessions and approximately 75% after fourteen. There is a problem here though. Aiming for formal "full recovery" makes better sense if possible. Here we may well need twenty sessions of therapy to help about 50% of clients to full recovery. To help 75% reach this very desirable goal may well take at least fifty sessions, so for example Arntz & colleagues' recent successful treatment of clients suffering with diverse personality disorders - "Results of a multicenter randomized controlled trial of the clinical effectiveness of schema therapy for personality disorders" - involved forty sessions in the first year of treatment and ten in the second. As Richard Layard & David Clark have so cogently argued in their tremendous 2014 book "Thrive: the power of evidence-based psychological therapies", there is a compelling case - in terms of both relief of suffering and economic saving - for more major investment in the treatment and prevention of psychological disorders. Research strongly suggests that, in general, clients are more often under-treated than over-treated. This comment however crucially assumes that worthwhile progress is being made. It is important to consider careful monitoring of treatment response using well-validated questionnaires to clarify that this is the case. Lack of response over the first three to six sessions should be a red flag. Is this therapy practised in this way by this therapist really right for this client? I have looked here at the "how many sessions?" question and I hope this post will help those involved in therapy be more informed about what is likely to be an adequate dose.