A team of researchers from Australia published two meta-analyses. The first analysis (external link) looked at 40 years of suicide risk assessment data research. It found that 95% of high-risk patients assessed for suicidality will not die by suicide. It also found that 50% of suicides happened in patients with a lower suicide risk. It states "They found no statistical method to identify patients at a high-risk of suicide in a way that would improve treatment."
The second analysis (external link) looked at how suicide risk could be predicted. It states "…the four strongest risk factors (previous episodes of self-harm, suicidal intent, physical health problems and male gender) were so common that they are of no help in assessing suicide risk. There was no evidence to support the use of risk assessment scales either."