What is it about?

Internet delivered Cognitive Behavior Therapy works for a wide variety of problems, but not every patient get a good enough effect of the treatment. Predicting who will benefit beforehand is still largely impossible. Instead, we started treatment for all patients and then made a structured prediction after 4 weeks of treatment. Those who we believed would have poor outcomes were indeed much more likely to have poor outcomes. More importantly, those who we believed would not benefit who then recieved extra care and attention did much better, showing that unsuccessful treatments can both be identified after about 4 weeks and avoided if the therapist is informed and has a strategy for helping the patient more with the treatment.

Featured Image

Why is it important?

This study shows a new way of working with outcome predictions for treatments, where allowing the patient to actually start treatment and progress a bit gives us a lot more precision in our predictions. It also shows that those who are likely to fail treatment are not lost causes, but that they can in fact have good outcomes of treatment if we intervene while there is still time left in treatment. This can help us decrease the number of failed treatment attempts while still offering this low intensity and easily accessible treatment as a first choice.

Perspectives

This study also shows a very early example of a prediction algorithm that is precise enough to lead to a clinically relevant increase in treatment successes if implemented in a treatment package.

Erik Forsell
Karolinska Institutet

Read the Original

This page is a summary of: Proof of Concept for an Adaptive Treatment Strategy to Prevent Failures in Internet-Delivered CBT: A Single-Blind Randomized Clinical Trial With Insomnia Patients, American Journal of Psychiatry, April 2019, American Psychiatric Association,
DOI: 10.1176/appi.ajp.2018.18060699.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page