Digital training for building resilience: Systematic review, meta-analysis, and meta-regression.

Resilience is learnable and broadly described as an individual's adaptive coping ability, its potential value for stress reduction must be explored. With a global coronavirus pandemic, innovative ways to deliver resilience training amidst heightened mental health concerns must be urgently examined. This systematic review aimed to (1) evaluate the effectiveness of digital training for building resilience and reducing anxiety, depressive and stress symptoms and (2) to identify essential features for designing future digital training. A three-step search was conducted in eight electronic databases, trial registries and grey literature to locate eligible studies. Randomised controlled trials examining the effects of digital training aimed at enhancing resilience were included. Data analysis was conducted using the Stata version 17. Twenty-two randomised controlled trials involving 2876 participants were included. Meta-analysis revealed that digital training significantly enhanced the participants' resilience with moderate to large effect (g = 0.54-1.09) at post-intervention and follow-up. Subgroup analyses suggested that training delivered via the Internet with a flexible programme schedule was more effective than its counterparts. This review supports the use of digital training in improving resilience. Further high-quality randomised controlled trials with large sample size are needed.