Registration of aggressive incidents: incidentbased versus periodbased methods

1
Registration of aggressive incidents
incident-based versus period-based methods

• Klaus Drieschner
• Trajectum Hoeve Boschoord

Preston, 27 March 2008
2
Setting Trajectum Hoeve Boschoord

• One of the five De Borg treatment centers
• Treatment of people with mild intellectual
  disability and (mostly aggressive) challenging
  behaviour (SGLVG)
• ? 160 beds, of which ? 90 in a medium security
  setting
• gt 75 court mandated treatments

3
Relevance of aggressive behaviour

• Aggressive behaviour during the treatment is
• the most frequent kind of problem behaviour in
  de Dutch SGLVG population (73 of the clients)
• a common reason for admission to Dutch SGLVG
  treatment
• sometimes an obstacle for proper treatment
• a major cause of stress and burnout among
  treatment staff
• an important determinant of the type of
  follow-up setting (e.g. level of restrictivity
  participation)
• and consequently, an important outcome variable
  for SGLVG treatment.

4
Types of aggressive behaviour during treatment

• Verbal aggression
• e.g. insulting, threatening
• Aggression towards objects
• e.g. damaging or breaking things, setting fire
• Physical aggression
• e.g. threatening gestures, pushing, hitting,
  attacking
• Aggression against self
• e.g. scratching, cutting, burning, (attempt)
  suicide
• Sexual aggression
• e.g. sexually inappropriate or threatening
  statements, exhibitionism, touching, forcing sex

5
Registration of aggressive incidents two types
of instruments
6
Incident-based registration

• Characteristics
• A separate registration form is completed after
  each incident
• Detailed information about each incident (e.g.
  time, place, trigger, consequence, measures
  taken)
• Examples
• Staff Observation Aggression Scale-Revised
  (SOAS-R Nijman et al. 1999)
• Attempted and Actual Assault Scale (ATTACKS
  Bowers et al. 2002)
• Advantages and disadvantages
• Can enhance awareness for causes and patterns
  of aggression
• Can yield information for effective aggression
  management
• Prone to underreporting and form-filling-fatigue
  
• Not easily embedded into daily routine

7
Period-based registration

• Characteristics
• Registration form is completed in fixed intervals
  (e.g. daily, weekly, two-weekly), whether or not
  incidents occurred
• Limited information about incidents type,
  severity, (frequency)
• Examples
• Modified Overt Aggression Scale (MOAS Kay et al.
  1988)
• Social Dysfunction and Aggression Scale (SDAS
  Wistedt et al. 1990)
• Advantages and disadvantages
• Quick (not reporting an incident saves little
  time)
• Becomes part of routine
• Allows for estimation of underreporting
• - Yields little information about causes and
  patterns of aggressive incidents

8
One column for each client ? ? ? ? ? ? ?
e.g. Verbal - L1 MILD Shouts angrily, curses
mildly or makes personal insults
III II I I
I I I
I I I
I I

I I II I
IIII I II
I I
? ? 5 types of aggression ? 4 levels for each
type ? ?
e.g. Against property - L3 STRONG Breaks
objects, smashes windows
e.g. Physical - L2 MODERATE Strikes, kicks,
pushes scratches, pulls hair out of others
(without injury)
e.g. Self-aggression - L4 EXTREME Inflicts
major injury on self or makes suicide attempt
e.g. Sexual 3 STRONG Sexually touches or
fondles others non-consensually
X X X
X X X
X X X X X
Signature unit-manager - - 20...
...
9
Number of MOAS incidents at Hoeve Boschoord(160
beds 178 patients 24 weeks)
In average 1,4 incidents per patient per week
10
MOAS incidents weighted for severity(160 beds
178 patients 24 weeks)
11
Percentages of patients causing MOAS
incidents(160 beds 178 patients 24 weeks)
12
Incidents of aggressive behaviour reported in 24
weeks SOAS-R versus MOAS
13
Expected effect of staff-reminding intervention
if form-filling fatigue applies
14
Observed effect of staff-reminding intervention
15
however, we must always be alert for
under-registration
16
MOAS as outcome measure

• ARIMA time series analysis
• Either for groups or for N1
• Estimation of form-filling-fatigue bias is pivotal

17
Conclusions

• Period-based registration of aggression is the
  method of choice if one is primarily interested
  in
• The frequency, type and severity of aggressive
  incidents
• Changes over long periods of time
• Incident based registration may be adequate
  for
• The registration of (infrequent) serious
  incidents
• Detailed analyses of antecedents and consequences
  of incidents
• The MOAS seems suitable for measuring change,
  but
• Attention has to be paid to the quality of the
  registration
• The influence of form filling fatigue bias must
  be estimated
• Thank you k.drieschner_at_hoeveboschoord.dji.m
  injus.nl