Nineteen studies met the inclusion criteria and 10 studies provided sufficient data to be included in the meta-analysis (n=3,775 participants, range 61 to 1,958). Results of the quality assessment were not reported.
Suicide (four cohorts; n=2,559): Three studies reported rates of suicide of 6% to 8%; these recruited particularly high-risk groups for suicide. The fourth study reported a suicide rate of 0.9%. Pooled estimates for Beck hopelessness scale were: sensitivity 80% (95% CI 68 to 90, I2=57%), specificity 42% (95% CI 41 to 44, I2=76%), positive likelihood ratio 1.55 (95% CI 1.31 to 1.83, I2=44%), negative likelihood ratio 0.45 (95% CI 0.20 to 1.03, I2=49%), diagnostic odds ratio 3.39 (95% CI 1.29 to 8.88, I2=37%) and the area under the curve was 0.70 (95% CI 0.59 to 0.85).
Self-harm (six cohorts; n=1,216): Pooled estimates for Beck hopelessness scale were: sensitivity 78% (95% CI 74 to 82, I2=0%), specificity 42% (95% CI 38 to 45, I2=90%), positive likelihood ratio 1.29 (95% CI 1.09 to 1.52, I2=74%), negative likelihood ratio 0.58 (95% CI 0.47 to 0.71, I2=0%), diagnostic odds ratio 2.27 (95% CI 1.53 to 3.37, I2=35%) and area under the curve was 0.63 (95% CI 0.57 to 0.70). Three studies reported two to seven people had died by suicide during the follow-up period.
Results of sensitivity analyses and the meta-regression were reported. There was no evidence of publication bias.