Cambridge, UK-based company Ieso Digital Health, provider of an online cognitive behavioural therapy (CBT) service, has carried out a study that sought to determine psychotherapy features associated with improvements in patient symptoms after applying deep learning to large-scale clinical data from its platform, published in JAMA Psychiatry.
According to the researchers, findings show that improvements are more likely to be seen if sessions focused on CBT change methods and less “non-therapy” content, such as conversations that were not related to the patient’s treatment.
“Clinical outcomes were measured in terms of reliable improvement in patient symptoms and treatment-engagement. Findings support the principle that CBT change methods help produce improvements in patients’ presenting symptomology. The application of deep learning to large clinical datasets can provide valuable insights into psychotherapy; informing the development of new treatments and helping standardize clinical practice,” they wrote.
HOW IT WAS DONE
Based on 90,000 hours of anonymised transcripts recorded on its platform — which patients can use to talk to therapists via instant messaging — Ieso says that it has trained a deep learning model to recognise the content used by therapists during these sessions, and use it to identify the aspects that are “significantly associated” with improvements in clinical outcomes.
The data was collected from over 17,500 patients receiving online CBT for the treatment of a mental health disorder from June 2012 to March 2018. When signing up to use the app, the company says patients provide informed consent in order to allow their anonymised data to be used for audit and research purposes.
CBT is the foundation of a number of digital tools tackling mental health conditions. Omada Health, for instance, announced in January that it would be incorporating the CBT offerings of now-defunct Lantern Health into its chronic disease management platform, while a 2018 JAMA study found Big Health's CBT approach effective among patients with insomnia symptoms.
The technology seems to be doing well enough for Ieso as well. Alongside these recent results, the company raised $24 million from investors back in 2017.
Michael Ewbank, senior scientist at Ieso Digital Health and lead author of the paper, said: “With our deep learning model, we can extract knowledge accumulated across thousands of hours of CBT in a way that would be impossible for a human to do.
“What is exciting about this study is that it demonstrates the potential of Ieso’s data set, where we can understand more about what the active ingredients of therapy are, what works for whom, and develop new and more effective treatments for mental health disorders. Our work represents a first step towards a practicable approach for quality controlled behavioural health care with the goal of improving the efficacy of psychotherapy.”