Psychotherapies like cognitive behavioural therapy (CBT) and interpersonal therapy (IPT) are core interventions of mental health services in the treatment of depression; it is estimated that in 2017 1.4 million people were referred to talking therapies for depression and anxiety via the NHS (NHS England, 2017). But how much therapy you get, in terms of the total number of sessions, and how often you get it, can be a bit of a lottery; there are national and evidence based guidelines, but service provision can vary according to local demand and resource.
A new paper published in the British Journal of Psychiatry asked the question, are more frequent sessions of psychotherapy more effective for people with depression? (Bruijniks et al, 2020). This is a critical issue in terms of best use of a limited resource: we know that CBT and IPT work, but how to optimise them?
There’s already some evidence that the frequency of sessions is an important factor. A recent meta-analysis focussing on depression showed that a higher frequency of therapy was associated with better outcomes than the total number of therapy sessions (Cuijpers et al., 2013). Some work has shown that higher frequency therapy leads to positive outcomes such as lower dropout rate and faster recovery when compared to lower frequency therapy in a range of anxiety disorders (Herbert et al.,2004; Foa et al., 2018) However, most such work was limited by observational or retrospective methodologies.
This new study aimed to further our understanding by conducting a prospective randomised controlled trial to investigate the effects of once a week vs twice a week CBT or IPT for people with depression (Bruijniks et al, 2020).
The researchers hypothesised that twice weekly sessions would give a greater reduction in depressive symptoms in the first 16 weeks of treatment: putatively, better retention of information between sessions from the patient, and/or a better working relationship between the patient and therapist, which might be the mechanism(s) for this.
- Outpatients aged 18-65 from Dutch mental health centres
- Primary diagnosis of major depressive disorder or persistent depressive disorder
- Pre-treatment Beck Depression Inventory (BDI) score of more than 20; sufficient knowledge of the Dutch language.
- Having started or changed antidepressants in the last 3 months
- Clinically determined to have a higher risk of suicide
- Diagnosis of drug or alcohol dependence alongside depression diagnosis
- Cluster A or B personality disorder diagnosis.
The Beck Depression Inventory (BDI) is a 21 question self-reported scale to assess the symptoms of depression; a high score indicates a large symptom burden. A score of more than 20 indicates moderate depression, more than 29 indicates severe depression; and less than 9 indicates minimal or no depression.
The participants were randomised into one of four treatment groups, with therapy delivered according to a treatment manualised approach (Beck and Klerman for CBT and IPT respectively):
- 16 sessions of CBT twice a week,
- 16 sessions of IPT twice a week,
- 16 sessions of CBT once a week
- 16 sessions of IPT once a week.
Of course this highlights an important issue: as all get the same number of sessions, the twice weekly group finish their treatment in half the time of the other group. In one sense, there is the same ‘total input’ of clinician effort and time, with potentially gains from the more intense version. There was a “cooling off” period where, once the initial 16 sessions were delivered, all participants received 4 further sessions over the period between the end of the treatment and the 6 month follow up.
The participants completed the BDI on registration, before every therapy session, and at the six-month follow up.
As well as following a manual-based approach, therapists attended a two day skills’ workshop to optimise adherence to a set treatment plan. With consent, sessions were videotaped and audited to ensure that the therapists were competently doing so.
The primary outcome of the study was the BDI score at the 6 month follow up; secondary outcomes were the individual component scores that make up the BDI such as life satisfaction and general sense of wellbeing.
In total, 200 participants were recruited and randomised between the four groups: mean age 37, 61% female, 68% in employment. In each group there was extensive attrition (drop out from treatment) with 48 of the 200 participants lost to follow up for a variety of reasons throughout the 6 month experimental period; intention-to-treat analysis meant that these were included in the final results.
Overall, there was a significant difference in the BDI scores for the weekly vs twice weekly groups at the 6 month follow up for CBT and IPT. By the intervention end-point, those participants who received twice weekly sessions reported lower scores than the weekly participants, indicating a lower symptom burden.
Specifically looking at CBT:
- At the 6-month end-point those receiving twice weekly sessions had a mean BDI score of 20.69 (95% CI 17.18 to 24.21) compared to a mean score of 23.01 (95% CI 19.45 to 26.57) for the once-weekly group. Although statistically significant, it is worth noting that there is quite a large overlap of the 95% confidence intervals
- Those who undertook twice weekly sessions were less likely to drop-out of the intervention
- Interestingly, the presence of an anxiety component, or the use of antidepressant medication, did not alter the results
- Cohen’s d for effect size was calculated as d = 0.55 indicating a moderate effect size of the twice weekly treatment. A Cohen’s d of more than 0.8 is typically considered a large effect size, so the effect we’re seeing is not huge.
A similar pattern of BDI difference between weekly and twice weekly IPT participants was also seen.
There were no differences between IPT or CBT when same frequencies were compared.
The authors concluded that IPT and CBT give larger reductions self-reported depression symptoms when provided twice weekly, rather than weekly. The also attained these benefits more quickly. Therapy frequency appears to be a modulator of effectiveness, albeit that the actual size of the difference was moderate, and a small percentage overall actually attained a clinically significant improvement.
Strengths and limitations
- This study addressed an identified gap in the literature with a relatively straightforward study design
- The noted variation seen between therapists, despite manualised delivery, and the drop-out rate, felt true to life making this study a useful model of real life clinical behaviour.
- Blinding is always difficult in a study like this. It was clear to participants, researchers and therapists which arm of the trial they were in. What’s not made clear is the extent to which the participants or therapists were aware of the hypothesis of the study: did any of them know they were in a group expected to perform better or worse than the other? In a study with a self-reported assessment as a primary outcome this feels like an important factor
- In all four groups the score for BDI reduced over time, however very few participants reported a BDI score indicating remission at the 6 month follow up
- Perhaps a future study could investigate the frequency effect over a longer period. Looking at the graphs for BDI, a longer treatment period might have led to even larger differences between the twice weekly and weekly groups
- Participants took the BDI before every therapy session meaning the twice weekly group completed the BDI twice as often as the weekly group. The researchers identified that this could contribute to the difference of scores, potentially the BDI acted as an element of self-reflection that was therapeutic. What’s not clear is why the researchers chose to ask the participants to fill this out twice; surely a once weekly BDI was appropriate for both groups and could have avoided this issue?
Implications for practice
- The researchers conclude from their study that twice weekly, rather than weekly CBT or IPT is the appropriate more effective way to deliver therapy for people with depression. Interestingly, this is what Beck first envisioned for CBT, whereas IPT was always seen as a weekly intervention
- This finding is potentially important in maximising the benefits of a limited resource, especially as it comes at no ‘cost’ in terms of additional therapist time
- However, one might argue that attending twice-weekly might be less convenient or practical for some people, and further, the actual difference in outcomes (although statistically significant) was not very large
- The impact, if any, on the psychologists undertaking this more intensive (though shorter-lasting) therapy would be important to discover
- We would be very interested in, and particularly welcome the feedback of service users receiving such therapy, and psychologists delivering it, on how they think such changes might feel and impact on therapy beyond ‘just outcomes’.
Statement of interests
Jack Kerwin doesn’t have any conflicts of interest other than being a medical student with an interest in psychiatry and psychotherapy. Derek Tracy has no conflict.
Bruijniks, S., Lemmens, L., Hollon, S., Peeters, F., Cuijpers, P., Arntz, A., . . . Huibers, M. (2020) The effects of once- versus twice-weekly sessions on psychotherapy outcomes in depressed patients. The British Journal of Psychiatry, 1-9. doi:10.1192/bjp.2019.265 https://doi.org/10.1192/bjp.2019.265
1.4 million people referred to NHS mental health therapy in the past year, NHS England (2017), Accessed 20th Feb 2020 (www.england.nhs.uk/2017/12/1-4-million-people-referred-to-nhs-mental-health-therapy-in-the-past-year/)
Cuijpers P, Huibers M, Daniel Ebert D, Koole SL, Andersson G. (2013) How much psychotherapy is needed to treat depression? A metaregression analysis. Journal of Affective Disorders
Foa EB, McLean CP, Zang Y, Rosenfield D, Yadin E, Yarvis JS, et al. (2018) Effect of prolonged exposure therapy delivered over 2 weeks vs 8 weeks vs present centred therapy on PTSD symptom severity in military personnel a randomized clinical trial. JAMA
Herbert JD, Rheingold AA, Gaudiano BA, Myers VH. (2004) Standard vs extended cognitive behaviour therapy for social anxiety disorder. Behavioural and Cognitive Psychotherapy.