Full title is "How accurate are suicide risk prediction models? Asking the right questions for clinical practice".
Suicide prediction tools do not work and current advice is that they should not be used to predict a person's risk.
This is for a few reasons.
Reliance on risk predictors means you don't provide services to people who go on to take their life. It also means you "section" (detain under the mental health act) people who are depressed but who are not going to kill themselves. This causes harm.
First, findings were not compared with current clinical practice, where risk assessment is routine
This undermines the conclusion of the paper regarding allocation of resources. Actual services are exclusively allocated within current practice. It advocates a policy based on logic, not field experience.
More troubling is that it misses the point that suicide risk assessment in clinical settings is an intervention. Suicide risk assessment is clinicians talking with patients about suicide. Assessment is part of treatment.
In clinical practice, the structured tools typically are based on patient response...a set of standard questions like a medical history or an ordinary diagnosis by a physician. In the US a typical assessment includes "Do you have guns in your house?" with follow up about their status, storage and use when the answer is affirmative. Or in terms of intervention, the clinician and patient discuss guns because firearms are a common means of suicide in the US.
Suicide prediction tools do not work and current advice is that they should not be used to predict a person's risk.
This is for a few reasons.
Reliance on risk predictors means you don't provide services to people who go on to take their life. It also means you "section" (detain under the mental health act) people who are depressed but who are not going to kill themselves. This causes harm.