INteresting points, and yes, our emergency department makes heavy use of safety contracts. I understand your point that they don't really reduce suicide contracts, but I like them for a number of reasons. One, it is a good way to define the role of both therapist and clinet, and I have to think there's something helpful about writing up a concrete plan toward helping a client to not feel so overwhelmed. Another is that it gives us more good information to evaluate a client's abilty to follow through. The more specific, the more support sources, the more available the resources, the more likely I am to believe a client can go home. I think from a liability standpoint, a safety contract must at least demosntrate that there has been some discussion about the suicidal ideas and some attention paid to how at risk the client actually is.

As for clients feeling they've been brushed off, we have a recurrent problem of a few repeat clients that have come to know the right things to say to get themselves hospitalized. I get a littlle dumbfounded as to why anyone would want to work so hard to get INTO a hospital-I'd much prefer a week to myself in Puerto Back Yard, but that's beside the point. Nevertheless, for us to hospitalize people in this group can be detrimental in the long term. I'm sure some feel brushed off, but to give them what they want teaches them that going to the emergency room and lying about their situaiton works, and reinforces that going to the hospital is a good way to not deal iwth problems. Too often these are people that have not followed through with recommended treatments and use the hospital as a first, rather than as a last, resort.

Evaluating for hospitalization is to a large degree a gamble. WE try to explain this to emergency room doctors all the time, when they get upset with our decision to not hospitalize a client. It's not like blood levels, or temperature. It can't be measured in numbers-although we are about to get a new form that is a sort of a checklist that may very well put the whole thing to numbers.