Self-help mobile messaging intervention for depression among older adults in resource-limited settings: a randomized controlled trial

The global health priority of reducing depression's burden in older adults is particularly critical in low- and middle-income countries (LMICs), where a significant portion of the older population resides. Brazil, experiencing a substantial increase in its older adult population, faces a considerable treatment gap for depression. Traditional interventions, often relying heavily on trained health professionals, are not feasible in resource-constrained healthcare systems. This study addresses this challenge by evaluating the effectiveness of a low-cost, self-help mobile messaging intervention (Viva Vida) in improving depressive symptoms among older adults in socioeconomically deprived areas of Brazil. The Viva Vida intervention, adapted from a previously successful task-shared collaborative care program, leverages the widespread use of WhatsApp to deliver psychoeducation and behavioral activation techniques without requiring direct health professional support. The study's goal is to determine whether this scalable digital intervention can effectively reduce depressive symptoms and improve access to care for older adults in LMICs.

Existing literature highlights the effectiveness of task-shared and collaborative care models in treating depression across various age groups, including older adults. However, these approaches often require substantial involvement from healthcare professionals, a significant constraint in LMICs with limited resources and overworked professionals. The COVID-19 pandemic further exacerbated this issue by reducing access to in-person consultations. Self-help digital mental health interventions offer a promising alternative, providing low-cost, accessible treatment options with minimal health professional support. Nevertheless, evidence on the effectiveness of such interventions in older adults, especially within LMICs, remains limited. This study builds upon previous research demonstrating the success of the PROACTIVE collaborative care intervention in Brazil while adapting it into a self-help digital format for increased scalability and accessibility.

PRODIGITAL-D was a pragmatic, single-blind, two-arm, individually randomized controlled trial conducted in 24 primary care clinics (UBSs) in Guarulhos, Brazil. Participants (aged 60+) registered with the UBSs and exhibiting depressive symptomatology (PHQ-9 ≥ 10) were recruited via phone and WhatsApp. 603 participants were randomized (1:1) to either the 6-week Viva Vida intervention (298) or a single message control group (305). Viva Vida delivered 48 audio and visual messages via WhatsApp, employing psychoeducation and behavioral activation principles. The control group received a single message with general information on depression. Outcomes were assessed at 3 and 5 months via phone interviews, using measures such as the PHQ-9, GAD-7, EQ-5D-5L, ICECAP-O, and 3-item UCLA loneliness scale. Missing data were addressed using multiple imputation by chained equations (MICE). Analyses included intention-to-treat analyses with imputed data and complete case analyses. Subgroup analyses explored potential effect modification.

At the 3-month follow-up, the Viva Vida intervention group showed a statistically significant improvement in depressive symptomatology (PHQ-9 < 10) compared to the control group (adjusted odds ratio = 1.57; 95% CI = 1.07–2.29; P = 0.019). This represented a 10.2 percentage point absolute difference in improvement rates. A similar beneficial effect was observed for the secondary outcome of a ≥50% reduction in PHQ-9 scores from baseline at 3 months (adjusted OR = 1.58; 95% CI = 1.08–2.29; P = 0.016). However, these differences were not sustained at the 5-month follow-up. No significant differences were found between groups for anxiety, loneliness, quality of life, or capability well-being at either follow-up. Complier average causal effect (CACE) analysis provided marginal evidence suggesting that the intervention's effect increased with the number of messages opened. The study also demonstrated the feasibility and acceptability of the Viva Vida intervention, with high adherence rates among participants.

The study findings demonstrate the short-term effectiveness of a readily scalable, low-cost, self-help digital intervention for improving depressive symptomatology in older adults in a resource-limited setting. The significant improvement observed at 3 months, but not at 5 months, suggests a potential need for booster messages or ongoing support to maintain the benefits. This aligns with other studies showing short-term efficacy of similar digital interventions. The use of WhatsApp, a familiar platform, facilitated high adherence and accessibility, underscoring the potential to bridge treatment gaps in areas with limited access to mental healthcare professionals. However, the relatively modest effect size compared to some collaborative care models highlights the need for further research on optimal intervention design and duration.

This RCT provides evidence supporting the feasibility and short-term effectiveness of the Viva Vida self-help digital intervention for depressive symptomatology in older adults. Its accessibility and low cost make it a valuable tool for addressing depression treatment gaps in resource-constrained settings. Future research should explore the impact of booster sessions, personalization strategies, and long-term outcomes.

The study's limitations include potential for attrition bias, despite similar missing data across groups and consistency between imputed and complete case analyses. The reliance on phone-based assessments might have introduced limitations in evaluating the severity of depressive symptoms and other conditions. The exclusion of individuals with low digital literacy or lack of WhatsApp access limits generalizability. Finally, cost-effectiveness analyses and qualitative data from the study would further inform the interpretation and implementation of this intervention.