Fourteen RCTs were included (n=1,655 participants): nine of preventive interventions and five of treatment interventions. Full details of the quality assessment for each study were provided in the paper. Six of the 14 studies reported the method of randomisation. Three studies reported concealment of allocation. Three studies used an intention-to-treat analysis. The proportion of people who did not complete the studies ranged from 5% to 35%. Most studies reported use of a treatment manual. All studies used an outcome measure known to be reliable and valid.
Preventive interventions: There was no statistically significant difference between preventive grief interventions and control post intervention (SMD -0.03, 95% CI -0.18 to 0.11) and at follow-up (SMD 0.13, 95% CI -0.08 to 0.33). Statistically significant heterogeneity was present in the analysis of follow-up data (p=0.07).
Treatment interventions: There was a statistically significant benefit with a complicated grief treatment intervention compared to control post-intervention (SMD -0.53, 95% CI -1.00 to -0.07) and at follow-up (SMD -1.38, 95% CI -2.08 to -0.68). Statistically significant heterogeneity was present in the analysis of post-intervention data (p=0.009).
The difference between the pooled standardised mean differences of the preventive and treatment interventions was statistically significant and in favour of treatment interventions at post intervention and at follow-up.