I'm jumping into this thread late, but I couldn't help but respond to this based on my experience that I'm having in the psych ED right now.
Our medical ED has incorporated a suicide risk screening questionnaire that gets asked of all patients regardless of complaint. It's a 5- or 6-question survey where each positive (i.e., "risk-increasing") answer is +1 point. Our psych ED includes what is effectively an ED-focused consult team that decides which patients are sent for further evaluation to the psych ER, which can be handled by the consult team, and which don't need to be seen.
The responses to various score ranges are algorithmic, such that low scores (say, 1-2) are treated with a "psych activation" - i.e., the patient shows up on the list of patients to be seen by the triage team. Roughly 10-20% of patients coming through the ED give at least one positive answer, resulting in a "psych activation" list of 15-20 patients, if not more, at any one time to be seen. The highest scores automatically get a patient placed on 1:1, which requires a psychiatrist to evaluate the patient in order to clear. In many cases, the nursing staff use high suicide screens to justify calling the local hospital police to detail the patient without a warrant until they can be seen by a psychiatrist (which are granted by the officer 95% of the time), again requiring a psychiatrist to fully evaluate the patient - a process which is frighteningly routine given how easily both the nursing staff and police officers disregard the rights of our patients.
Great intentions, but I have found this screening program to be nothing but a headache. The knee-jerk responses place a huge burden on the consult team when simple commonsense would dictate that this patient does not, in fact, require a psychiatrist, a 1:1 sitter, or a warrantless police detention. It results in many patients that are clearly malingering being sent to the psych ER (rather than being evaluated and subsequently discharged by the triage team) thanks to asinine policies that remove medical decision-making and clinical judgment and replace them with a flowchart of "how do I treat the patient with a psychiatric history" in addition to having people that have little to no psych training effectively determining the disposition of a patient based on said flowchart.
Sadly, I find that most nights I've spent working on this consult team are spent undoing ridiculous journeys down the "how to treat the patient with a psychiatric history" yellow brick road than focusing limited resources on patients that could truly benefit from the help. I'm not sure that the "powers that be" could have executed a suicide screening program more poorly.