JMIR Publications

ABSTRACT

Background:

While numerous internet-based cognitive behavioral therapy (iCBT) programs have been developed and tested through randomized controlled trials (RCTs) for various mental health conditions and specific populations, research on their effectiveness and real-world application remains limited.

Objective:

This study aimed to examine the effectiveness of a previously developed iCBT program implemented in an existing app for improving depressive symptoms among pregnant women in a real-world setting.

Methods:

The previously developed iCBT program for preventing perinatal depression was already implemented in an existing app that aims to provide information to pregnant women about pregnancy and babies. The log data stored in the app identified iCBT program users and non-users, allowing us to conduct this retrospective cohort study. The data from September 2022 to September 2024 was extracted from the app after anonymous processing. The primary outcome was the score on the self-reported Edinburgh Postnatal Depression Scale (EPDS), which participants input by themselves on the app. The exposure group was defined as completers of all 6 modules of the iCBT program. The non-exposure group was defined as non-users of the program who matched the baseline characteristics of the exposure group. The change in EPDS score before and after using the program was compared using effect size, and repeated two-way analysis of variance (repeated two-way ANOVA) was conducted to test the difference between the exposure and non-exposure groups.

Results:

119 women who completed the iCBT program and pair-matched 448 controls were selected. The average EPDS scores at the baseline were 7.24 in the exposure group and 7.25 in the non-exposure group. After using the iCBT program, the EPDS scores changed by -0.69 and +0.99 in the exposure and non-exposure groups, respectively (Cohen’s d = 0.31, 95%CI 0.11 - 0.51). The repeated two-way ANOVA showed statistical significance in interaction terms between the groups and the measurement time points (P=.04).

Conclusions:

The previously developed iCBT program showed a significant modest effect on decreased depressive symptoms among pregnant women in the real-world setting. Future research should attempt to minimize dropouts and increase participation in the program.