The THREAT is for targeted persons who cam be identified and ought to be warned and protected. In Canada, we have had a tarasoff -like case where sex trade workers were the targeted group. It is of course, very dfficult to conceptualise the protection of an identified group. The scenario where individuals are targeted by race, gender, position etc, do indeed constitute THREATs. In our hospital we arrange for wide spread communication to all staff and will have ward specific additional safety requirements e.g., a targeted staff has to inform the ward of intended visit and provide an expected time of arrival, whereupon s/he will be met at the door and escorted at all times while on the ward, with the patient restricted to either his room or another part of the ward. We use HCR-20 for medium term risk opinion (6 months to a year), START for up to 3 months, and have developed a new measure (Get Shorty) for day to day and shift to shift use. Get Shorty is in the early piloting stages. Johann -----Original Message----- From: startgroup-bounce@xxxxxxxxxxxxx [mailto:startgroup-bounce@xxxxxxxxxxxxx] On Behalf Of Denis Zavodny Sent: April 5, 2010 3:45 PM To: startgroup@xxxxxxxxxxxxx Subject: [startgroup] Re: Use of T.H.R.E.A.T. interesting discussion. I think these victims are targeted, perhaps because they are close enough to assault/viewed as invading an expanded personal space perimeter. I see this as a case akin to Meehl's broken leg where you might want to defer scoring of the START or go outside of ("override") the totality of the items in the START when determining risk. fwiw, the manual does note that the threat may be specific "to a person or groups of persons." furthermore, soon after Tarasoff, there were several cases identifying the responsibility to warn or protect certain identifiable classes of potential victims (Lipari v. Sears, Roebuck, and Co; Jablonski v. US). we've had consumers threaten to assault various classes of victims (the next woman who walks onto the unit, the next administrator they see, women with a certain body type, individuals of a certain ethnic culture, etc.). in one case, the consumer will reliably attack anyone with a beverage (he has polydipsia and outside drinks are banned from the unit). it may be helpful to consider the interface of specific signs as someone may be fine but when they start to decompensate they engage in the signature/anamestic risk behavior (e.g., pacing the hall) and targeting a certain class of individuals (e.g., delusional preoccupation with a race or profession). the decompensation may be a setting event/establishing operation for aggressive behavior. At times like that it might be useful to supplement the START with the DASA:IV or some other instrument tapping imminent risk as well as anamestic risk factors. An issue is the dicey methodology of how the class of victims is informed (risking pre-emptive attack by peers or panic by staff). Our units at our maximum hospital have two separate doors to the unit (on either side of the nurses station) and we've recently established one door as the entrance and the other as the exit so that unit staff can more readily identify, warn, and protect potential targeted individuals when then enter. I'd like to hear how other people handle this issue. I would also love to hear how people are using the START in a system with tools assessing more acute and longer term risk. We've been using the HCR-20 for long-term consumers for a number of years, began using the START in our hospitals last Fall, and are contemplating adding the DASA:IV and, thanks to a consultant's recommendation and the need for something quick for use with short-term civil consumers, at discharge, the COVR. Denis