JMIR Publications

ABSTRACT

Background:

Internet-delivered cognitive behavioural therapy (iCBT) reduces psychological distress in patients with cardiovascular disease (CVD), but whether formats with less clinical support achieve outcomes comparable to structured therapist-directed approaches remains unclear.

Objective:

To evaluate the non-inferiority of self-selected versus therapist-selected content and feedback on demand versus regular scheduled therapist feedback in individualised iCBT for patients with CVD and psychological distress.

Methods:

This 2×2 factorial non-inferiority randomised controlled trial compared therapist-selected programs with regular scheduled feedback (TS-RF) or feedback on demand (TS-OD), and self-selected programs with regular scheduled feedback (SS-RF) or feedback on demand (SS-OD). Participants with CVD and symptoms of depression, anxiety, or stress were recruited nationwide in Sweden. Primary analyses compared therapist-selected (TS-RF and TS-OD) with self-selected programs (SS-RF and SS-OD), and regular scheduled feedback (TS-RF and SS-RF) with feedback on demand (TS-OD and SS-OD). Exploratory analyses compared each alternative format (TS-OD, SS-RF, and SS-OD) with TS-RF. Primary outcomes were depressive symptoms (Patient Health Questionnaire-9 [PHQ-9]), anxiety (Generalized Anxiety Disorder-7 [GAD-7]), and perceived stress (Perceived Stress Scale-10 [PSS-10]); cardiac anxiety (Cardiac Anxiety Questionnaire [CAQ]) was secondary. Outcomes were assessed at baseline, post-treatment (9 weeks), and 6-month follow-up. Non-inferiority used a predefined margin of d=0.30.

Results:

Among 237 randomised participants, all groups showed large improvements (Cohen's d >0.90) in depressive symptoms, anxiety, and stress, largely maintained at 6 months. Differences between therapist-selected and self-selected programs and between regular scheduled and on-demand feedback were generally small, but confidence intervals frequently exceeded the predefined non-inferiority margin, precluding firm conclusions. High adherence was more common in therapist-selected than self-selected programs (71.2% vs 45.4%) and with regular scheduled than on-demand feedback (64.4% vs 52.1%). On-demand feedback reduced therapist time (approximately 2 vs 6 minutes per participant per week). Exploratory comparisons of the original trial arms likewise failed to demonstrate non-inferiority of TS-OD, SS-RF, or SS-OD relative to TS-RF.

Conclusions:

iCBT formats with reduced clinician input may yield similar outcomes for some patients with CVD, although non-inferiority was not consistently demonstrated. Therapist-selected programs and regular scheduled feedback were associated with higher adherence. Thus, greater self-direction may increase flexibility and reduce therapist time, while regular scheduled support may remain important for engagement. Future research should identify patients who can successfully use lower-support formats. Clinical Trial: Clinical Trials.gov (registration number NCT04726722)