Five quasi-experimental studies (n=23,995 participants) were included in the review. Sample sizes ranged from 1,685 to 7,010. The proportion of elderly patients ranged from 13 to 25% of the samples. Post-intervention follow-up ranged from six years to 10 years.
In two studies where the follow-up was undertaken by a psychiatrist, significant reductions in completed suicide were observed for men (fixed-effect model pooled IRR 0.30, 95% CI 0.13 to 0.68; random-effects model IRR 0.33, 95% CI 0.14 to 0.80). Similar reductions were also observed for women (fixed-effect model pooled IRR 0.33, 95% CI 0.19 to 0.58; random-effects model IRR 0.33, 95% CI 0.19 to 0.60).
Three studies reported follow-up conducted by a general practitioner. There were statistically significant reductions in completed suicides in women found (fixed-effect model IRR 0.36, 95%Ci 0.21 to 0.60; random-effects model IRR 0.39, 95% CI 0.22 to 0.66). There were no statistically significant reductions in completed suicides in men (fixed effect model IRR 0.73, 95% CI 0.45 to 1.18; random-effects model IRR 0.74, 95% CI 0.45 to 1.23).
There was no statistically significant heterogeneity observed across the studies for any of the results. Visual appraisal of the funnel plots and the Begg's tests showed no evidence of publication bias.