Good points. Suicide hotlines, crisis counseling, etc., often deal with people who are not in an immediately lethal situation (that doesn't at all mean the crisis isn't real, or that the counter-suicide counseling isn't very important, nor does it mean that inadequately-trained professionals should man those phones or crisis centers). A certain number of callers, perhaps most, are calling for some reason that doesn't reflect a potentially lethal situation, but it is often hard to differentiate those from the ones who are more likely to make an attempt or -- very important -- whose conditions or behaviors are unstable. Instability and unpredictability are among the most important suicide risk factors.

The above also means -- and this is one of my big areas of interest these days -- that the interviewer should be trained, experienced, and prepared to do either (1) as comprehensive an assessment as is feasible under the circumstances or (2) a very competent screening which liberally passes "positive" patients along to someone who can do a more comprehensive assessment or at least work with more complex situations. Such things are certainly more difficult when dealing with a telephone hotline, where one can't see the patient, doesn't know the accuracy of the communications, often can't get to collateral information sources, and has little control over whether or not the patient follows recommendations (such as "go to the emergency room"). Perhaps most telephone hotlines shouldn't even be called "screening."

That brings up an interesting conundrum: Some clinics and hotlines use their less experienced folks for the screening role. "Screening" implies casting a very wide net, using basic questions that will catch most or all of the important cases. The questions are often pretty rote, with simple decision trees. Good screening should result in lots of "false positives," which are caught at the next levels of assessment.

Some "screening" is actually more like "triage." In triage, an experienced evaluator makes rapid decisions, usually under pressure (such as during a disaster), about who gets what kind of treatment, or no treatment. Triage requires a very well trained & experienced person (since important decisions are being made for which there may be no appeal). If your "screening" setting sounds more like a "triage" setting, then the people who man it should not be entry-level or marginally-trained professionals.

As to the literature on safety contracts: The best place to get a large list is at PubMed (the National Library of Medicine's online service, which provides references, abstracts, help with searches, etc.) That's where I looked when doing an article on the topic last year.