JMIR Publications

ABSTRACT

Background:

Digital transformation has reformed how mental health interventions can be delivered, ranging from traditional human delivery to fully digital modes. Although digital delivery offers advantages, previous research suggests that human involvement may be crucial for intervention effects. Evidence concerning how mode of delivery influences the effects of interventions aimed specifically at promoting mental well-being among higher education students remains limited.

Objective:

This systematic review and meta-analysis examined the effects of individually delivered interventions on positive affect and life satisfaction among emerging adults in higher education and whether effects differed between digital, human, and blended modes of delivery. Additionally, intervention delivery was classified at the level of interaction within generic phases of the intervention process.

Methods:

Nine databases and trial registries were searched, with the final search conducted on August 5, 2026. Randomized controlled trials of individually delivered mental well-being interventions among higher education students aged 18-29 years without a reported diagnosed mental disorder were eligible. Intervention delivery was independently coded across 9 phases of the intervention process and classified overall as digital, human, or blended. Postintervention effects were synthesized as standardized mean differences using random-effects meta-analyses. Risk of bias was assessed using RoB 2, and certainty of evidence was assessed using GRADE

Results:

A total of 43 studies involving 7646 participants were included, out of which 32 studies contributed to the positive affect meta-analysis and 18 to the life satisfaction meta-analysis. Interventions showed a small positive overall effect on positive affect (SMD 0.24, 95% CI 0.13-0.35; I²=63%) and life satisfaction (SMD 0.34, 95% CI 0.07-0.62; I²=90%). There was no evidence that effects differed according to mode of delivery for positive affect (χ²₂=0.47, P=.79) or life satisfaction (χ²₂=0.49, P=.78). Phase-level coding indicated that interventions classified overall as blended consisted of human and digital delivery across phases rather than blended delivery within phases. Several phases of the intervention process were infrequently reported, particularly goal-setting and intervention selection. Sensitivity analyses excluding studies at high overall risk of bias yielded similar pooled effect estimates. Certainty of evidence was moderate for positive affect and low for life satisfaction.

Conclusions:

Individually delivered interventions showed small positive effects on mental well-being among higher education students, with greater certainty of evidence for positive affect than for life satisfaction. However, in contrast to previous research, the findings did not indicate that the effects differed between digital, human, and blended modes of delivery. The phase-based findings suggest that broad labels on delivery mode may conceal variation in how delivery occurs across phases of the intervention process. Greater attention to theoretically relevant phases such as goal-setting and intervention selection may provide opportunities for individualizing the intervention process. Further research should examine during which phases, and for whom, different modes of delivery provide particular benefits. Clinical Trial: Registered in PROSPERO International Prospective Register of Systematic Reviews on November 25, 2025. Last updated on May 15, 2026: 2025 CRD420251131950.