Effectiveness & Implementation of a Text Messaging Intervention for Depression & Anxiety in COVID-19

Effectiveness and Implementation of a Text Messaging Intervention for Depression and Anxiety During COVID-19 among Latinx and Non-Hispanic White Subscribers: a hybrid design study

Alein Y. Haro-Ramos, Hector P. Rodríguez, Adrian Aguilera

Background: Text messaging interventions are increasingly being used to address mental health symptoms due to the COVID-19 pandemic. Despite the widespread adoption of text interventions, little is known about racial and ethnic subgroup differences in effectiveness, feasibility, and implementation factors.

Objectives: A hybrid type 1 mixed-methods study was conducted to compare the effectiveness and implementation of StayWell for Latinx/Hispanic and non-Hispanic White (White) adult users using elements of the RE-AIM framework. We examine differences in the reach, effectiveness, adoption, implementation, and maintenance.

Methods: Adults using StayWell received daily supportive text messages for 60 days. Data were collected via online surveys and through the HealthySMS portal. Reach was assessed by the share of adults reached via distinct recruitment methods, effectiveness was evaluated using depression (PHQ-8) and anxiety (GAD-7) scores, adoption was assessed using response rates, implementation was evaluated based on feedback from users about the number of messages, the timing of messages, the difficulty of using the program, and usability of StayWell, and maintenance was assessed by user reports of the likelihood of continuing and recommending the program. Categorical variables were compared using the chi-square test, and continuous variables were analyzed using t-tests. Mixed effects regression analysis modeled the change in PHQ-8 and GAD-7 scores between Latinx and White users after completing StayWell using an interaction. We conduct thematic text analysis of responses to an open-ended item inquiring about experiences in the program to help contextualize differences in the effectiveness and implementation of StayWell between Latinx and White users.

Results: Anxiety and depressive symptoms decreased among StayWell users over time. Compared to White adults, Latinx adults had significantly greater reductions in PHQ-8 (25.8% vs. 13.8% reduction, p=0.02) but not in GAD-7 scores (21.2% vs. 15.9% reduction, p=0.23). Despite Latinxs reporting lower usability (76.8 vs. 83.9, p=0.001) compared to Whites, Latinxs were more likely to report interest in continuing the program (7.5 vs. 6.2, p=0.001) and recommending StayWell to a family member or friend (7.8 vs. 7.0, p=0.013).

Conclusions: Latinx adult users of StayWell reported a greater reduction in depression symptoms than White adult users. Latinxs were more likely to accept and maintain the use of StayWell than Whites; however, lower perceived usability among Latinxs suggests that future iterations of StayWell should conduct user experience research to improve usability among Latinxs. Among Latinxs, repetitive or impersonal content was a critical barrier to implementing StayWell; among the group, an essential suggestion was the availability of support groups or engagement with a behavioral health expert via text.