Cognitive behavioural therapy (CBT) and interpersonal psychotherapy (IPT) produce comparable average outcomes for depression. That raises an obvious question: could some people do better with one than the other? Researchers call these patient-by-treatment patterns aptitude-treatment interactions, and few have been found or replicated.
Coyne and colleagues set out to replicate earlier findings from a systematic review of trials comparing CBT and IPT. Adult outpatients with major depression were randomly assigned to 16 weeks of CBT (41 people) or IPT (39 people). Characteristics were measured at the start, and patients rated their depression throughout treatment.
One finding replicated. Patients with more self-sacrificing interpersonal problems did better with CBT than with IPT, while the reverse was true for patients with fewer such problems. Other moderators either failed to replicate or pointed in the opposite direction to earlier research.
So the results offer some guidance on matching treatment for some patients. They also show the limits of traditional methods for this question, and the authors suggest that different methods are needed to personalise treatment more broadly and reliably. Some patterns held. Many did not.
If low mood has been with you for a while, please reach out to a psychologist or your GP. Both of these therapies are effective options, and a good clinician will help you choose.
What the study did
Cognitive behavioural therapy (CBT) and interpersonal psychotherapy (IPT) are both evidence-based treatments for depression, and on average they produce comparable outcomes. It is plausible, though, that patients with particular characteristics respond better to one than the other. Researchers test this by looking at whether patient characteristics change the comparative effect of two treatments, known as aptitude-treatment interactions (ATIs).
The difficulty is that few ATIs have emerged or been replicated, which leaves little guidance for choosing a treatment. Coyne and colleagues drew on a systematic review of patient ATIs in trials comparing CBT with IPT for depression (Bernecker et al., 2017). They aimed to replicate significant ATIs previously established in a single study, and significant ATIs examined twice where only one study showed a moderating effect.
The data came from a trial in which adult outpatients with major depression were randomly assigned to 16 weeks of CBT (n = 41) or IPT (n = 39). Patient characteristics were measured at baseline, and patients rated their depression throughout treatment.
What it found
Multilevel models revealed one ATI replication. For patients with more self-sacrificing interpersonal problems, CBT outperformed IPT. For patients with fewer such problems, the reverse was true. Other moderators either failed to replicate or directionally contradicted prior research.
Why it matters
One replicated finding is a modest but real step. The authors say the results help inform optimal treatment matching for some patients, which reflects a type of psychotherapy personalisation. Knowing that interpersonal style may relate to which therapy helps more offers a starting point for conversations between clinicians and clients.
Limits
Most of the earlier findings did not hold up here, and some pointed the other way. That is informative in itself. Treat no single moderator finding as settled. The study was also a secondary look at one trial with a modest number of participants, so the result should be read with care.
The authors draw a broader lesson. The results highlight limitations of traditional ATI research and suggest that different methods are needed to inform responsive personalisation efforts more expansively and reliably. The question of what works best for whom is still being worked out.
Reading the results in context
It helps to be clear about what "self-sacrificing interpersonal problems" suggests in everyday terms. It points to patterns in how a person relates to others, and the finding was that these patterns were linked to which therapy performed better. For those patients CBT outperformed IPT, and for patients with fewer such problems IPT came out ahead.
This should not be read as a rule for choosing therapy. The abstract describes it as one ATI replication from a set of moderators that were tested, and the others either failed to replicate or contradicted earlier research directionally. A single replicated pattern is encouraging, though it is only a pattern in this one dataset.
The authors place the finding in a wider conversation. Personalisation of psychotherapy, meaning fitting treatment to the person, is an appealing goal. Their conclusion is that traditional ATI research has limitations and that new methods are needed. The researchers say so themselves. Treatment-matching claims deserve some caution.
Seeking help
Both CBT and IPT are established treatments, and neither is the wrong choice on average. If you are living with depression, please speak with a psychologist or your GP. Together you can work out an approach that fits you, and adjust it if it is not helping.