Publikasjoner

Finn publikasjon

John Olav Roaldset Stål Bjørkly O. M. Linaker

A multifaceted model for risk assessment of violernt behaviour in acutely admitted psychiatric patients

Psychiatry Research (epub 2012). | 2012

Sammendrag

Violenceriskassessmentinstrumentsaremainlybasedonhistoricalandclinicalriskfactors.Biological
factorshavebeenrelatedtoviolentbehaviourbuthardlyusedinriskassessment.Arecentstudy
indicatedthatbothlowtotalcholesterol(TC)andthepatients’ownestimatesabouttheirriskoffuture
violentbehaviour,theSelf-reportRiskScale(SRS),significantlypredictedviolentbehaviourduring
hospitalstayandthefirst3monthsafterdischargefromanacutepsychiatricdepartment.Ouraimwas
to investigatewhethercombiningthreemethods,ascreeninginstrumentforriskofviolence(V-RISK-
10), lowTCandtheSRS,intoonemultifacetedriskassessmentmodelmayenhancepredictivevalidity.
This prospectivenaturalisticstudyinvolved134ofall(489)acutelyadmittedpatientstoageneral
psychiatrichospitalduring1year.LowTC,SRSandV-RISK-10atadmissionwereprospectively
comparedwithrecordedviolenceduringhospitalstayandthefirst3monthsafterdischarge.The
multifacetedriskassessmentmodelyieldedasignificantincreaseinexplainedvariancebeyondthatof
the V-RISK-10.AUCvalueswerehigher,butdifferenceswerenotsignificant.Thebiopsychosocial
approachtoviolenceriskassessmentseemspromising,butfurtherstudiesareneededtotestthe
feasibilityandpredictivevalidityofmultifacetedmodels.