Therapy sessions and fit statistics

People starting therapy often ask the same things. How many sessions will it take? Does it matter which therapist I see? And how many people stop before the end?

The figures here come from national services, clinic records, trials and reviews that pool many studies. The relationship figures are correlations. The preference and feedback results are effect sizes or odds ratios, and those are not shares of people.

Each figure names the group it counts, and most review figures say how many studies they pooled.

How many sessions, and when people improve

NHS Talking Therapies is the national service for anxiety and depression in England. Its figures count referrals that finished a course of treatment. The Canadian figure comes from the records of one specialised clinic. There, a response meant a reliable and clinically significant improvement on a depression scale. The clinic's share counts anyone who improved at some point, while its session count covers only patients still improved at their last session. NHS Talking Therapies treats people in courses of treatment, and the clinic gave a course of low-intensity CBT. Neither count describes open-ended or psychodynamic therapy. Recovery means what each service defines it to mean.

• In England in 2024-25, referrals that finished a course of NHS Talking Therapies treatment had 8.4 sessions on average. (administrative records, 2024-25 data, checked 10 Oct 2026) ³
• In England in 2024-25, 670,419 referrals to NHS Talking Therapies finished a course of treatment. (administrative records, 2024-25 data, checked 10 Oct 2026) ³
• In England in 2024-25, 50.5% of referrals that finished a course of treatment in NHS Talking Therapies moved to recovery. (administrative records, 2024-25 data, checked 10 Oct 2026) ³
• At one specialised outpatient mental health centre in Ontario, Canada, 29.4% of the adults given low-intensity CBT showed a reliable and clinically significant improvement on a depression scale. (administrative records, 2017-2024 data, checked 10 Oct 2026) ⁵
• At one specialised outpatient mental health centre in Ontario, Canada, half of the adults whose depression scores reliably improved with low-intensity CBT had improved by 7 sessions. (administrative records, 2017-2024 data, checked 10 Oct 2026) ⁵

The relationship between client and therapist

The alliance is the working bond between client and therapist. These reviews pool studies that measured the relationship and the outcome in the same clients. A correlation shows that the two rise together. It does not show that the relationship causes the improvement.

• Across 295 studies covering more than 30,000 patients, the correlation between the strength of the therapeutic alliance and the outcome of face-to-face therapy was r = 0.278. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ¹
• Across 23 studies of internet-based psychotherapy, the correlation between the therapeutic alliance and outcome was r = 0.275, about the same as face to face. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ¹
• Across 82 samples covering 6,138 clients, the correlation between therapist empathy and client outcome was r = 0.28. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ⁶
• Across 54 studies covering 7,278 adult patients, the correlation between patient and therapist agreeing on the goals of therapy and its outcome was r = 0.24. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ⁷
• Across 53 studies covering 5,286 adult patients, the correlation between patient and therapist working together and the outcome of therapy was r = 0.29. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ⁷

Preferences, and who stops early

Dropout means stopping therapy before the therapist or the plan expected. The preference review compares clients who got the kind of treatment they wanted with clients who did not. The figures from India and Kenya come from trials of brief programmes run by lay counsellors and community health workers. The Colombian figure comes from a national psychosocial support service for victims of armed conflict.

• Across 669 studies covering 83,834 adult clients, the average dropout rate from therapy, weighted across studies, was 19.7%. (meta-analysis of studies, published 2012, checked 10 Oct 2026) ²
• In a national US household survey in 2001-2003, 22.4% of adults who had mental health treatment in the past year, from any kind of provider, quit before their provider wanted them to stop. (survey, past 12 months, 2001-2003 data, checked 10 Oct 2026) ⁸
• In a national household survey in the United States in 2001-2003, over 70% of the adults who quit outpatient mental health treatment early did so after the first or second visit. (survey, past 12 months, 2001-2003 data, checked 10 Oct 2026) ⁸
• Across 53 studies covering more than 16,000 clients, dropping out was less common when clients got the kind of treatment they preferred, an odds ratio of 1.79. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ⁴
• Across 53 studies covering more than 16,000 clients, clients who got the kind of treatment they preferred had better outcomes than clients who did not, an effect size of d = 0.28. (meta-analysis of studies, published 2018, checked 10 Oct 2026) ⁴
• In a trial in Goa, India, 69% of adults with moderately severe to severe depression offered a brief behavioural therapy by lay counsellors ended it as planned rather than stopping early. (randomised trial, 2013-2015 data, checked 10 Oct 2026) ⁹
• In a trial in Nairobi, Kenya, 21.5% of women with a history of gender-based violence and psychological distress, offered a brief psychological programme from lay community health workers, attended none of its sessions. (randomised trial, 2015 data, checked 10 Oct 2026) ¹⁰
• In Colombia, 29% of people attending individual sessions in the national psychosocial support programme for victims of armed conflict received only one session. (administrative records, 2013-2021 data, checked 10 Oct 2026) ¹¹

Does the therapist matter?

A therapist effect is the part of the variation in outcomes that lies between therapists rather than between clients. The UK figures come from one service, where the authors identified two groups of therapists: eight above average and ten below, out of sixty-one. Each UK figure is the average of those therapists' own recovery rates. In the pooled review, the therapist share of variation ranged across studies from under one percent to about three in ten.

• In a systematic review of 20 studies of therapist effects, differences between therapists accounted for a weighted average of 5% of the variation in patient outcomes. (2018 data, checked 10 Oct 2026) ¹²
• In one UK primary care therapy service, part of the government IAPT programme, the eight therapists identified as above average had a mean recovery rate of 63.7%: the average of each therapist's own rate. (administrative records, 2010-2013 data, checked 10 Oct 2026) ¹³
• In one UK primary care therapy service, part of the government IAPT programme, the ten therapists identified as below average had a mean recovery rate of 25.6%: the average of each therapist's own rate. (administrative records, 2010-2013 data, checked 10 Oct 2026) ¹³

Feedback and outcome monitoring

Progress feedback, also called routine outcome monitoring, means the client fills in a short questionnaire during therapy and the therapist sees the scores. A client who is not on track is one whose scores are not improving as expected. The review pools randomised and non-randomised studies.

• Across 58 studies covering 21,699 patients, therapy with progress feedback reduced symptoms slightly more than therapy without it, an effect size of d = 0.15. (meta-analysis of studies, published 2021, checked 10 Oct 2026) ¹⁴
• For the not-on-track patients in a review of progress feedback studies, therapy with progress feedback reduced symptoms slightly more than therapy without it, an effect size of d = 0.17. (meta-analysis of studies, published 2021, checked 10 Oct 2026) ¹⁴
• Across the progress feedback studies that reported dropout, feedback had a small effect in favour of fewer dropouts, an odds ratio of 1.19. (meta-analysis of studies, published 2021, checked 10 Oct 2026) ¹⁴

The relationship and the outcome of therapy

Each row is a correlation from a review that pooled many studies. A correlation of zero would mean no link at all. None of these is a share of people.

How to read these figures

A correlation (r) shows how closely two things rise and fall together. It is not a share of people, and it does not prove cause.

An effect size (d) shows how far apart two groups ended up, in standard units. An odds ratio compares the odds of an event, such as dropping out, between two groups.

Session counts come from a national service in England, one clinic in Canada, a national programme in Colombia and two trials of brief programmes. They are not counts for open-ended therapy, and they are averages for particular groups, not a target for any one person.

Figures marked Worldwide pool studies from several countries. They are not national data.

Across these studies, a good working bond and shared goals go with better outcomes, and getting the kind of therapy you prefer goes with fewer dropouts. If the fit feels wrong, you can say so to your therapist, or look for a different one. Our guide on how to find the right therapist explains what to look for. If you are in distress right now, or thinking about ending your life, see our helplines and suicide hotlines worldwide.

Cite this page: TherapyRoute. Therapy sessions and fit statistics. https://www.therapyroute.com/article/therapy-sessions-and-fit-statistics-by-therapyroute. Last updated 10 October 2026.

References

1. [1] The alliance in adult psychotherapy: A meta-analytic synthesis. Psychotherapy (Flückiger, Del Re, Wampold and Horvath), 2018 pubmed.ncbi.nlm.nih.gov.
2. [2] Premature discontinuation in adult psychotherapy: a meta-analysis. Journal of Consulting and Clinical Psychology (Swift and Greenberg), 2012 pubmed.ncbi.nlm.nih.gov.
3. [3] NHS Talking Therapies, for anxiety and depression, Annual reports, 2024-25. NHS England, 25 September 2025 digital.nhs.uk.
4. [4] The impact of accommodating client preference in psychotherapy: A meta-analysis. Journal of Clinical Psychology (Swift, Callahan, Cooper and Parkin), 2018 europepmc.org.
5. [5] Evaluating Dose Response of Cognitive Behavioural Therapy Using Outpatient Electronic Medical Record Data: An Observational Study. Canadian Journal of Psychiatry (Hollingbery, Bryan, Giovannetti and others), 2026 pmc.ncbi.nlm.nih.gov.
6. [6] Therapist empathy and client outcome: An updated meta-analysis. Psychotherapy (Elliott, Bohart, Watson and Murphy), 2018 pubmed.ncbi.nlm.nih.gov.
7. [7] Meta-analyses of the relation of goal consensus and collaboration to psychotherapy outcome. Psychotherapy (Tryon, Birch and Verkuilen), 2018 europepmc.org.
8. [8] Dropout from outpatient mental health care in the United States. Psychiatric Services (Olfson, Mojtabai, Sampson and others), 2009 pmc.ncbi.nlm.nih.gov.
9. [9] The Healthy Activity Program (HAP), a lay counsellor-delivered brief psychological treatment for severe depression, in primary care in India: a randomised controlled trial. The Lancet (Patel, Weobong, Weiss and others), 2017 pmc.ncbi.nlm.nih.gov.
10. [10] Effectiveness of a brief behavioural intervention on psychological distress among women with a history of gender-based violence in urban Kenya: A randomised clinical trial. PLOS Medicine (Bryant, Schafer, Dawson and others), 2017 pmc.ncbi.nlm.nih.gov.
11. [11] Intervention provision and engagement in Colombia's PAPSIVI - a national psychosocial support service for over half a million victims of armed conflict. Conflict and Health (Fernandez, Acosta-Ortiz, García Durán and others), 2026 pmc.ncbi.nlm.nih.gov.
12. [12] A systematic review of therapist effects: A critical narrative update and refinement to review. Clinical Psychology Review (Johns, Barkham, Kellett and Saxon), February 2019 pubmed.ncbi.nlm.nih.gov.
13. [13] The relationship between therapist effects and therapy delivery factors: therapy modality, dosage, and non-completion. Administration and Policy in Mental Health and Mental Health Services Research (Saxon, Firth and Barkham), 2017 pmc.ncbi.nlm.nih.gov.
14. [14] Using progress feedback to improve outcomes and reduce drop-out, treatment duration, and deterioration: A multilevel meta-analysis. Clinical Psychology Review (de Jong, Conijn, Gallagher, Reshetnikova, Heij and Lutz), 2021 europepmc.org.