Early Interventions for Primary Care Patients With Stress-related Ill-health: a Non-inferiority RCT

Overview

Stress-related ill-health is rated one of the largest health challenges in the western world. The most empirically supported treatment for stress-related disorders is cognitive behaviour therapy (CBT), but accessibility is low and most patients in primary care do not receive this evidence-based treatment. Collaborative care has been shown to be an effective working model for primary care patients with mental health problems. This type of care intervention has however not been compared against CBT, which is arguably the gold standard treatment for this patient group. The overarching purpose of this project is to investigate if implementation of two treatment models - collaborative care and therapist-guided self-help CBT -can be effective as early interventions for primary care patients with stress-related ill-health.

Stress-related ill-health is rated one of the largest health challenges in the western world. The most empirically supported treatment for stress-related disorders is cognitive behaviour therapy (CBT), but accessibility is low and most patients in primary care do not receive this evidence-based treatment. Collaborative care has been shown to be an effective working model for primary care patients with mental health problems. In this study collaborative care is conducted by a nurse who provides the patient with self-management support. This intervention has been implemented at Gustavsbergs primary care clinic in Stockholm as a means to increase accessibility and quality of care for the large group of patients with stress-related ill-health. This type of care intervention has however not been compared against CBT, which is arguably the gold standard treatment for this patient group. Our research group has developed and tested a therapist-guided self-help CBT (SH-CBT) that can be delivered online or as a self-help book with support by a clinician for stress-related ill health, which has been shown to be effective and also potentially very suitable for the primary care context. The overarching purpose of this project is to investigate if implementation of two treatment models - collaborative care and SH-CBT - can be effective as early interventions for primary care patients with stress-related ill-health. Main research question: Is a brief collaborative care intervention provided by a nurse at least as effective as SH-CBT delivered by a psychologist for primary care patients with mild to moderate symptoms of stress? Note: In may 2022, the research group made revisions to the study procedures in terms of administration format of CBT and inclusion criteria. These specific changes were made: (1) Participants randomized to SH-CBT are allowed to choose if they want to receive the treatment as therapist-guided online CBT or as therapist-guided bibliotherapy CBT. Participants are exposed to the same treatment content, but in the former case the material is presented on a secure website with asynchronous written therapist support and in the latter case the content is provided through a self-help book and the participant has 2-5 face-to-face appointments with the therapist; (2) It is no longer required that participants has adjustment disorder, instead participants have to have (a) a score on the Perceived Stress Scale-10 of 20 or higher, (b) stress-related ill-health as their principal problem, and (c) a duration of their problems of at least 1 month; (3) The sick-leave criterion is changed so that participants to be included are allowed to have been on full time sick-leave for maximum 2 months (instead of 50% for maximum 1 month). The revisions described above were made in order to better operationalize the patient population "patients with mild to moderate stress-related ill-health" and to increase the recruitment of participants to the study. When the revisions were implemented approximately 20 participants had been included in the study. The revisions were approved by the Swedish Ethical Review Authority (ID 2022-00990-02). Note 2: In the autumn of 2023, a new power calculation was conducted where full use of the 5 assessment points between baseline and post-treatment (12 weeks) was taken into account. This new power calculation, based in linear mixed models analysis for longitudinal data, showed that in order to have 80% power to demonstrate non-inferiority (non-inferiority margin d=0.3; one-sided 95% confidence interval) given a true effect difference of 0, approximately 170 participants (85 per treatment arm) will be needed.