Fig. 3: Forest plot for outcome cognition.
Plot representing comparison of self-guided digital interventions versus controls (divided by therapeutic framework and active versus non-active controls) for outcome cognition at the end of the intervention. Data was pooled from seventy-one comparisons (n = 4345). The random-effects meta-analysis found a small-to-moderate treatment effect of all digital self-guided interventions compared to controls (g = –0.51 (95%CI –0.64 to –0.37, p < 0.00001)). Heterogeneity between trials was moderate (I2 = 77%). Analysed independently, a moderate significant effect size was found for both cognitive rehabilitation (k = 21, g = –0.67, 95% CI –0.93 to –0.41; Z = 5.13, p < 0.00001) and virtual reality (k = 13, g = –0.55, 95%CI –0.94 to –0.17; Z = 2.83, p = 0.005) compared to controls. I2 statistic identified significant heterogeneity (I2 = 81%). A small-to-moderate significant effect was found for cognitive training (k = 32, g = –0.36, 95% CI –0.55 to –0.17; Z = 3.69, p = 0.0002) and videogames (k = 3, g = –0.52, 95% CI –-0.86 to –0.18; Z = 3.03, p = 0.002). The effect of internet-delivered courses on cognition was not significant (k = 2, p = 0.13). For simplification only cognitive training, cognitive rehabilitation and virtual reality interventions are displayed.
