Effectiveness of digital interventions for eight mental disorders: A meta-analytic synthesis

In this unified series of meta-analyses, we integrate the effects of digital interventions in adults with mental disorders compared to inactive controls. We cover eight indications: depressive disorder, insomnia, specific phobias, generalized anxiety, panic, social anxiety, obsessive-compulsive, and posttraumatic stress disorder.

Digital intervention trials in patients with a diagnosed mental disorder (confirmed by clinical interviews) were extracted from the Metapsy living databases for psychological treatments. Standardized meta-analyses were conducted to pool effects for each disorder, as well as separately for guided and unguided treatments. We also examined study dropout rates, conducted meta-regression analyses stratified by disorder, and identified treatments that have since become available as prescribable digital therapeutics in routine care.

In total, 168 studies (22,144 patients) were included. Moderate effect sizes were observed for PTSD (g = 0.57), depression (g = 0.62), and obsessive-compulsive disorder (g = 0.68). Large effects emerged for generalized anxiety (g = 0.80), social anxiety (g = 0.84), insomnia (g = 0.94), panic disorder (g = 1.05), and specific phobias (g = 1.18). Pooled study dropout rates were generally moderate (≤20 %), but higher in intervention arms (RR = 1.13-2.66). Trials with low risk of bias and care-as-usual comparisons were limited across indications. We found 16 trials evaluating a prescribable digital therapeutic (g = 0.33-1.60).

Digital interventions can be effective across a wide range of diagnosed mental disorders. For some indications, more high-quality trials and comparisons against care-as-usual are needed to confirm the robustness of the effect, particularly for unguided treatments. Digital interventions are increasingly commercialized as prescribable digital therapeutics. Rising industry involvement may present both opportunities and new challenges for the field.