Affecting an estimated 332 million worldwide, depression represents a major public health challenge of our era (World Health Organisation, 2025). Despite widely implemented first-line treatments such as cognitive behavioural therapy (CBT) and antidepressant medication (NICE, 2022), only 30-40% of patients fully recover. This burdensome reality leaves the majority cycling through care services with ongoing impairment, repeated referrals, little sense of real progress or even deterioration. Depression’s chronicity, relapse rates, and treatment resistance make it one of the most resource-intensive and complex mental health conditions to manage in modern mental health practice and suggest potential limitations in current guideline-driven models of care.

The literature underpinning these guidelines is dominated by CBT, which has the largest share of randomised controlled trials (RCTs) in depression (Cuijpers, 2020). While this reflects CBT’s strong empirical foundations, it raises concerns of a structural bias during guideline development where treatment recommendations reflect the volume of available trials rather than evidence supporting other interventions. For example, an effective yet comparatively underrepresented therapy is Short-term Psychodynamic Psychotherapy (STPP), which focuses on resolving unconscious conflict and maladaptive relational patterns. Malkomsen et al. (2025) compare CBT and STPP in real-life practice, helping to reduce the paucity of representative, low-risk-of-bias comparative effectiveness evidence.

Methods

The study was an RCT involving 100 adults with major depressive disorder (MDD) receiving care in two public outpatient clinics in Oslo. The sample was limited to participants who spoke a Scandinavian language, excluding those with certain intellectual, neurological or psychiatric conditions.

Participants were randomly allocated to cognitive behavioural therapy (16 weekly sessions plus three-monthly boosters; n=50) or short-term psychodynamic psychotherapy (28 weekly sessions; n=50). Treatment fidelity was monitored through the recording of sessions, which ensured that clinicians who received regular supervision and training delivered the therapies consistently.

At baseline, participants underwent assessments to assess comorbid ‘personality disorders’ and ensure they met the inclusion criteria. Depression severity was measured using the clinician-rated Hamilton Depression Rating Scale and self-report Beck Depression Inventory-II, alongside secondary measures of anxiety, functioning, and quality of life.

To explore changes across therapies, clinics, clinicians, and session dosage, participants were reassessed at weeks 8, 16 and 28. Once complete, the data were analysed using linear mixed models in the groups they were originally assigned, even if they did not complete therapy, as it provides a more realistic estimate of how the treatments work, where patients often miss a session or discontinue therapy.

Results  

Over almost five years, 242 individuals were screened, with 100 participants randomly assigned to CBT or STPP. After participant attrition, 81 remained in the study. The two treatment groups were similar at the start of the study in terms of age, gender, ethnicity, and clinical history, including rates of recurrent depression and earlier psychotherapy.

Both CBT and STPP led to large improvements in depressive symptoms over the course of the treatment, with around three-quarters of the patients no longer meeting the diagnostic criteria for depression at 28 weeks. A minority of participants deteriorated, suggesting that overall, both treatments were generally safe and effective.

There was little evidence that one therapy outperformed the other, as differences in effectiveness between CBT and STPP were small and not statistically meaningful across the main outcome measures. A similar pattern was observed with the secondary outcomes; anxiety, daily functioning, work and social adjustment, and mental health–related quality of life all improved over time, regardless of treatment type, with no observable differences between the therapies. Physical health did not change meaningfully in either group. Eleven participants stopped treatment (defined as attending fewer than eight sessions), with the data providing no meaningful evidence of a between-group difference in dropout rates.

Moreover, there was little evidence that results differed by clinic or therapist, with only one small site-related effect appearing on the clinician-rated HDRS. The analysis of therapist effects on the HDRS did not converge, meaning it was not possible to reliably assess whether individual therapists influenced outcomes. However, these issues were not strong enough to meaningfully challenge the overall pattern of equivalence between the therapies.

Conclusions

Based on the evidence, both CBT and STPP are effective short-term treatments for depression, with neither therapy outrightly outperforming the other. However, not all individuals respond to short-term care, highlighting the need for flexible pathways and access to alternative or longer-term services.

Strengths and limitations

Overall, the study contributes meaningfully to a limited comparative evidence base, with its key strength being its real-world outpatient setting, which enabled the inclusion of typically excluded high-risk participants (e.g., suicidal ideation) and those taking antidepressant medication. This makes the findings more relatable to everyday clinical practice and did not come at the cost of scientific rigour, as the use of randomisation, blinded clinician assessment, high inter-rater reliability, and ongoing clinical supervision ensured the study was methodologically sound.

That said, there are key limitations that need to be considered:

1. Unequal therapy sessions: The authors justified differences in the number of sessions between the two groups on the basis that psychodynamic therapies usually require a longer treatment duration. While this is intuitive, the authors did not provide sufficient evidence to justify the numbers chosen. Although their analyses found that treatment dosage did not affect outcomes, the study was only powered to detect medium effects (d=0.6), meaning that smaller clinically significant differences between the treatments due to dosage could have been missed.
2. Antidepressant Usage: Some participants started taking antidepressants during the course of treatment, which may have worked independently or synergistically with therapy to reduce symptoms, introducing a risk of co-intervention bias. Without knowing who took the medication and when, it remains unclear whether the results reflect therapeutic change, pharmacological effects, or their interaction.
3. Vague reporting of Negative Effects: There was no formal reporting of harms, with “deterioration” used as the only descriptor. This lacks qualitative richness, overlooks participants’ lived experiences of the therapies and limits transparency. As a result, the reader’s ability to weigh the benefits against the risks of each therapy is weakened, which is particularly important given the growing emphasis on treatment-related harm in psychotherapy research.
4. Artificial therapy conditions: Although the study aimed to reflect routine practice, the sessions were recorded, which is typically not the case. This may have changed how openly patients spoke and how rigidly clinicians adhered to therapy techniques. For example, STPP relies more on emotional depth and spontaneity, which are more sensitive to observation, limiting how closely the study reflects everyday practice.
5. Small sample and no control group: The authors themselves note that the relatively small sample size means non-inferiority of STPP to CBT could not be formally established, and that no untreated control group was included, as the study was primarily designed to explore moderators and mediators of change rather than overall outcome differences.

Implications for practice

It is tempting to consider this study as evidence that CBT and STPP are broadly equal and interchangeable therapies for depression. However, the study’s most important implication is not about finding which therapy works ‘best’ but what the findings mean for the people who use, deliver, research, and regulate mental health care.

Policy makers and clinicians need to consider this study in light of its sociocultural context. The sample included was predominantly European and limited to only those who spoke a Scandinavian language, reflecting a context of relatively high independence, social equality, and low power distance (The Culture Factor Group, 2025). These factors are crucial when interpreting treatment outcomes, as they influence perception, symptom expression, and communication, ultimately affecting the therapeutic alliance (Dai, 2022; Kyynärsalmi, 2025). For example, in the present study, patients may have seen clinicians more like collaborative partners rather than authority figures, making them more likely to voice disagreement, ask questions and actively shape the course of their treatments. Consequently, the positive outcomes of the therapies may partly be due to these cultural patterns of active participation and open dialogue, which may not generalise to settings with greater power asymmetries where speaking up to perceived authority figures is less openly encouraged. Seen through an intersectional lens, service users from ethnic minorities or migrant backgrounds may meet linguistic barriers, cultural mismatch with therapeutic models, or prior experiences of institutional mistrust that influence how therapy is experienced and taken up, regardless of its demonstrated effectiveness in more socially advantaged populations (Bhui, 2012). As a result, these findings provide minimal generalisation as to how CBT or STPP may compare for more diverse populations, such as patient groups typically seen in the NHS (NHS England, 2023). Therefore, it is unlikely that these results can immediately translate into practice, and researchers need to replicate this across more varied social and cultural contexts before guideline changes can be made.

Service users and carers should take this evidence as a starting point rather than a definitive answer that both therapies are interchangeably effective, because equivalence seen at the group level does not always translate on an individual level. The lived experience of therapy and individual needs should always be at the forefront of care planning. What works for one may not work for another, and a lack of benefit from a particular approach should not be seen as a sign of failure, but rather a cue to adjust care.

Statement of Interest

Bianca Kewlani has no known competing interests declared. The study was publicly funded by Norwegian universities and hospitals, which did not influence the results or conclusions of the study.

Edited by

Dr Dafni Katsampa.

Links

Primary paper

Anders Malkomsen, Theresa Wilberg, Bente Bull-Hansen, Toril Dammen, Julie Horgen Evensen, Benjamin Hummelen, André Løvgren, Kåre Osnes, Randi Ulberg & Jan Ivar Røssberg (2025). Comparative effectiveness of short-term psychodynamic psychotherapy and cognitive behavioral therapy for major depression in psychiatric outpatient clinics: A randomized controlled trial. BMC Psychiatry, 25, 113.

Other references

Bhui, K., Ascoli, M., & Nuamh, O. (2012). The place of race and racism in cultural competence: What can we learn from the English experience about the narratives of evidence and argument? Transcultural Psychiatry, 49(2), 185–205.

Cuijpers, P., Karyotaki, E., Reijnders, M., Purgato, M., Barbui, C., & Cipriani, A. (2020). A network meta-analysis of the effects of psychotherapies, pharmacotherapies, and their combination in the treatment of adult depression. World Psychiatry, 19(1), 92–107.

Dai, Y., Li, H., Xie, W., & Deng, T. (2022). Power distance belief and workplace communication: The mediating role of fear of authority. International Journal of Environmental Research and Public Health, 19(5), 2932.

Kyynärsalmi, A., Hammarén, M., & Kanste, O. (2025). Understanding power distance in healthcare: A scoping review. Nursing Inquiry. Advance online publication.

National Health Service (NHS) England. (2023). Patient and carer race equality framework. 

National Institute for Health and Care Excellence (NICE). (2022, June 29). Depression in adults: Treatment and management (NICE guideline NG222).

The Culture Factor. (2025). Country comparison tool: Norway.

World Health Organization. (2025, August 29). Depressive disorder (depression) – Fact sheet.