Forty-four studies (n not reported) were included in the review. Overall methodological quality was poor. In the great majority of studies the method of randomisation was unclear and allocation concealment and blinding were not reported. Sample sizes were poorly reported and attrition ranged from 0 to 55%, where reported. Most studies did not use an intention to treat analysis. Some studies reported outcomes selectively (for example, significant outcomes).
Group-based interventions (13 RCTs): Eight RCTs reported significant benefits for intervention groups compared to controls on some outcomes. Five RCTs reported no statistically significant differences between groups on any outcome. Pooled analysis found that group psychoeducational interventions were effective for caregiver depression (weighted mean difference was -071, 95% CI: -0.95 to -0.46, p<0.00001; five RCTs, n=292). There were no significant differences between psychoeducational intervention groups (three RCTs, n=231) or support intervention groups (two RCTs, n=119) and controls for assessment of carer burden. There was no evidence of statistical heterogeneity in the analysis.
Technology based interventions (four RCTs): There were no significant differences between computer-based interventions and control groups for depression in carers of people with dementia (three RCTs, n=229). Data were reported for other outcomes. There was no evidence of statistical heterogeneity in the analysis.
Individual-based interventions (27 RCTs): 10 RCTs found no statistically significant differences between intervention and control groups; 12 reported a significant effect for interventions groups for some assessed outcomes; and one trial reported a significant effect in favour of the control group. Pooled results showed no statistically significant differences between psychoeducational interventions and control groups for carer depression (seven RCTs, n=297) or for self-efficacy (two RCTs, n=190). There was evidence of statistical heterogeneity for outcomes of depression (I2=60.9%) and self efficacy (I2=50.4%).