No easy fix for adherence in internet interventions: two randomized studies targeting adherence self-efficacy.

Low adherence remains a persistent challenge in self-guided internet interventions, and universal strategies are needed to address it. In this study, we designed and evaluated a brief self-guided add-on module aimed at strengthening self-efficacy specific to adhering to internet interventions. The module included a testimonial video and two text-based exercises grounded in self-efficacy theory. It targeted self-efficacy specific to integrating and maintaining use of internet interventions. In Phase 1, university students (N = 217) were randomized to the module or a waitlist control. The aim was to test the immediate effect of the module on internet intervention adherence self-efficacy. In Phase 2, final-year medical students and medical interns (N = 214) were randomized to receive either the module before a two-week self-guided intervention (Med-Stress Student) or the intervention alone. The goal was to test the module in a naturalistic setting. In Phase 1, an independent samples t-test did not detect a between-group difference in adherence self-efficacy. In Phase 2, linear mixed effects models indicated that the groups did not differ in adherence self-efficacy immediately after the module or at posttest. Wilcoxon tests detected no between-group differences in actual adherence. Mixed ANOVAs found no group by time interactions for work engagement or job stress. Job stress decreased from baseline to posttest across both groups, but this uncontrolled change cannot be attributed to the intervention. In conclusion, we found no evidence that the brief, one-time pre-intervention module improved adherence to internet interventions. High attrition in Phase 2 limits this conclusion. Future studies should test whether continuous support improves adherence.