Risk Rating Scale (IRRS):
An effective tool for identifying inpatients
at increased risk for violence
Dr. Andrew Starzomski
East Coast Forensic Hospital
Capital Health, Halifax, Nova Scotia
• 1. The problem of inpatient violence
• 2. The challenge of ‘short term’ risk assessment
• 3. What is the IRRS, and what is ‘in’ it?
• 4. A quick case study
• 5. What’s to like?
• 6. Where are we heading with the tool?
• Though there is abundant attention / procedure
for addressing clinical issues (symptoms,
treatment) on psychiatric units, not the same for
• East Coast Forensic’s Mentally Ill Offender
Unit (court-ordered assessments) in ‘01/’02:
– 55 violent incidents of 169 admissions; staff victim
56% of time, co-patient victim 44% of time
– Injury to co-clients/staff, related costs
– Staff focus: clinical/treatment vs ?
– Nature / quality / rates of recovery?
– Public perception / unease about seeking care for
self / family?
– Shortage of trained heath care professionals due to
Why Create a New Tool?
• Existing tools solely addressing individual-
oriented variables (e.g., symptoms, mental state)
are missing a great deal… therefore incomplete
as an approach or orientation to the issue
• Steinert: necessity of evaluating person-
environment fit and interaction
• Need for a brief measure capable of detecting
change in key variables that can occur quickly
– Other existing measures are lengthier and without a
Psychologists, and others, love a
good measure of bad behaviour
What is IN the IRRS?
• Seven domains are evaluated:
– History of Violence
– Subtypes (re: personality & psychosis)
– Acuity of hostility / anger
– Communication impairment
– Social status / relationship problems
– Milieu strain
What is the IRRS?
• Checklist format/ info rated by trained clinician
based on file, observation info
• Quickly addresses a set of historical and
dynamic variables associated with short-term
• Developed with breadth in mind: there are
MANY pathways / causes / forms of
• 7 items get rated on a 0-1-2 scale
How is the IRRS used?
– After training, nurses rate new inpatients based on
file review, info emerging from observations /
interview… inter-rating evaluation
– Information is shared with staff, areas of uncertainty
flagged / more info needed, intervention strategies
– Ratings every 4-7 days as-needed…
– Focus on inpatient environment
IRRS: A Case Example
– 41 year old single male from rural NS
– NCR on charges Mischief, Assault Peace Officer
– Dx: features of psychosis, mania, OCD, behavioural /
– Tx: atypicals, mood stabilizers, anxiolytics, ECT
• some lengthy periods of stability and decent community
functioning over last several years
– Aggression as tactic for attention, confinement
– Back at ECFH since summer 08 / group home aggression
IRRS: A Case Example
– As of late winter 2009:
• Ongoing uncertainty about discharge
• Implementation of a new rehabilitation agenda via the
“Treatment Mall”: social / lifestyle stress
• Uncertainty that maintenance ECT doing much
• Recurrence of aggressive thoughts, tangentiality,
communication impairments, lability
• Aggression – numerous times, toward pts & staff
• How does this translate to IRRS ratings?
IRRS: A Case Example
• Gordon: IRRS Scores / Behaviour
1/15 3/11 3/23 3/27 3/30 4/8 4/13 4/25 5/7 5/11
- 60 patients at MIOU tracked through 2006-2007
- Never-violent patients score sig lower on
average during admission than those who were
aggressive once or more
- Scores vary meaningfully over course of
admission for those who are violent (IRRS
reductions linked to better adjustment on the
- IRRS scores hold up for predicting verbal and
physical aggression for about 10 days (not just
24-72 hours), after which time ratings are no
longer able to predict aggressive behaviour
Other Evidence of Utility:
• From the beginning I thought it was an interesting idea.
• Easy to do
• Anticipate it becoming a helpful tool to help predict possible
• Helps with awareness for aggression
• Has been quite accurate in some cases
• Should be used in all mental health facilities.
Benefits of the IRRS
• Common language
• Staff know their patients better and as a result
are able to recognize / identify risk
• Early recognition and appropriate response
• Interdisciplinary work
of the IRRS… pending evaluation
• Hopefully: decreased incidents of aggression… needs
• Hopefully: Increased staff satisfaction and retention…
needs more formal evaluation
• Hopefully: Increased quality of patient care… needs
Where do we go from here?
• More data, more evaluation
• Other settings: short term / crisis unit,
mainstream Emergency suites, youth,
correctional health units
• A tool to help create safer health settings:
building awareness, promotion, etc.