Treating Alcohol-Related Violence

• Author: Nicola Bowes,Siriol David,Mary McMurran,Bryn Williams,Ingrid Zammit
• DOI: 10.1177/0093854811433759
• Published date: 01 March 2012
• Date: 01 March 2012

TREATING ALCOHOL-RELATED VIOLENCE
Intermediate Outcomes in a Feasibility Study
for a Randomized Controlled Trial in Prisons
NICOLA BOWES
Cardiff Metropolitan University
MARY MCMURRAN
University of Nottingham
BRYN WILLIAMS
Cardiff Metropolitan University
SIRIOL DAVID
INGRID ZAMMIT
National Offender Management Service Cymru
There is a lack of outcome evidence for alcohol interventions for offenders whose crime is alcohol related. In this study, the
authors report the intermediate outcomes of a feasibility study for a randomized controlled trial of an alcohol-related violence
intervention. Control of Violence for Angry Impulsive Drinkers (COVAID) was tested with sentenced prisoners in the United
Kingdom. Participants were 115 adult men who were randomly allocated to COVAID or treatment as usual. Measures were
the Alcohol-Related Aggression Questionnaire (ARAQ), the State-Trait Anger Expression Inventory (STAXI-2), the Eysenck
Impulsivity, Venturesome, and Empathy Scale (IVE), and the Controlled Drinking Self-Efficacy Scale (CDSES). After the
intervention, participants allocated to COVAID reported significantly greater improvement on the ARAQ Alcohol-Aggression
subscale and all CDSES subscales. No significant differences were obtained for the STAXI-2 or the IVE. COVAID may have
the potential to fill a gap in treatment provision for offenders whose crimes of violence are alcohol related.
Keywords: alcohol; violence; treatment; prisoners; COVAID; randomized controlled trial
Alcohol-related violence is an issue of major concern to society. Around half of all
crimes of violence are related to alcohol (Flatley, Kershaw, Smith, Chaplin, & Moon,
2010), and it is estimated that alcohol-related crime costs the economy of England and
Wales £7.3 billion each year (Prime Minister’s Strategy Unit, 2004). Despite this knowl-
edge, alcohol interventions for offenders are seriously underprovided and underdeveloped.
Recent reports on alcohol services in U.K. prison and probation services present a picture
of significant lack of provision (Fitzpatrick & Thorne, 2010; HM Inspectorate of Prisons,
2010; McSweeney, Webster, Turnbull, & Duffy, 2009). In addition, both of these reports
comment on the lack of evidence-based alcohol interventions for offenders whose criminal
behavior is related to their use of alcohol. A recent rapid evidence assessment of interventions
AUTHORS’ NOTE: Bryn Williams is now at Warwick Medical School, University of Warwick, UK. The
research was supported by the National Offender Management Service Cymru (Wales). Our thanks to Samantha
James, Karen Grove, Gemma Worgan, and staff of NOMS Cymru. All correspondence should be addressed to
Nicola Bowes, Psychology Department, School of Health Sciences, Cardiff Metropolitan University, Cardiff
CF5 2YB, UK; email: nbowes@cardiffmet.ac.uk.
CRIMINAL JUSTICE AND BEHAVIOR, Vol. 39 No. 3, March 2012 333-344
DOI: 10.1177/0093854811433759
© 2012 International Association for Correctional and Forensic Psychology
333

---

334 CRIMINAL JUSTICE AND BEHAVIOR
specifically for alcohol-related violence has shown a dearth of evaluated interventions
(McMurran, in press). Offenders in custody form one important target group for interven-
tions addressing alcohol-related violence. Using the Alcohol Use Disorders Identification
Test (AUDIT; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001), 63% of male sen-
tenced prisoners in England and Wales have been identified as hazardous drinkers
(Singleton, Farrell, & Meltzer, 1999). Around half of prisoners who drink heavily admit to
being violent when intoxicated (Bowes, Sutton, Jenkins, & McMurran, 2009).
In England and Wales, the National Offender Management Service (NOMS) has a strat-
egy for working with alcohol misusing offenders (NOMS, 2006), which endorses the treat-
ment of offenders, and the Ministry of Justice has issued guidance to clinicians on what
works to reduce alcohol-related offending (Ministry of Justice, 2010). This guideline sug-
gests that cognitive-behavioral based interventions with motivational approaches are prom-
ising in treating both alcohol problems and violence. However, it is noted that, to date, there
have been no outcome studies of the impact of these programs delivered in either prison or
community settings in England and Wales (Ministry of Justice, 2010).
The treatment program evaluated here is Control of Violence for Angry Impulsive
Drinkers (COVAID; McMurran, 2007b; McMurran & Delight, 2010). COVAID is a struc-
tured cognitive-behavioral treatment program that aims to reduce the likelihood of alcohol-
related aggression and violence by targeting cognitions, emotions, and behavior known to
be related to aggression and violence. This includes addressing alcohol intoxication, which
may be considered a causal risk factor in relation to aggression, that is, “a risk factor that
can change, and, when changed, cause(s) a change in risk for the outcome” (Murray,
Farrington, & Eisner, 2009, p. 4). Alcohol intoxication has been shown to relate to violence
in longitudinal studies. In a longitudinal study of a birth cohort in New Zealand, Fergusson
and colleagues (Fergusson & Horwood, 2000; Fergusson, Lynskey, & Horwood, 1996)
have shown that a substantial amount of the relationship between alcohol use and crime is
related to shared factors, such as social disadvantage and deviant peer affiliations.
Nevertheless, when these confounding variables are controlled for, a significant relation-
ship remains between alcohol misuse and crime, particularly violent crime, with heavy
drinkers being three times more likely to be violent than light drinkers. That is, the anteced-
ent risk factors for alcohol misuse and violent offending are highly similar, but there is also
a direct relationship between alcohol misuse and violent offending. In addition, alcohol
intoxication has been shown to temporally precede violence. For example, in a case-
crossover study of violent offenders, Hǻggard-Grann, Hallqvist, Lǻngström, and Möller
(2006) showed that there was a thirteenfold increase in the risk of committing violence for
those who had been drinking alcohol in the 24 hours before the event.
Alcohol intoxication on its own does not explain violence. In a review of meta-analyses
of experimental studies of alcohol on aggression, Exum (2006) reported overall effect sizes
of around 0.50, showing that alcohol accounts for 25% of the variance in aggressive behav-
ior. Therefore, the COVAID intervention addresses other contributory risk factors. These
include context factors, such as drinking in high-risk venues (Graham, Bernards, Osgood,
& Wells, 2006), and individual factors, such as hostile attributions (Dodge, Price,
Bachorowski, & Newman, 1990), anger arousal (Novaco, 2011), alcohol outcome expec-
tancies of both aggression (McMurran et al., 2006) and social confidence (McMurran,
2007c), and impulsive social problem solving (McMurran, Blair, & Egan, 2002; Ramadan
& McMurran, 2005). To integrate the treatment targets into a coherent model, the COVAID

---

Bowes et al. / ALCOHOL-RELATED VIOLENCE 335
intervention is based on Novaco’s angry aggression “system,” in which aggression is
explained by the interaction of provocations, cognitive appraisals, physiological arousal,
and learned behavioral responses (Robins & Novaco, 1999). Reducing angry aggression
requires change in all parts of this system. In COVAID, the added risks of alcohol intoxica-
tion and drinking contexts are added to this system to identify the need to change drinking
and related behaviors, and so reduce the risk of alcohol-related violence (McMurran, 2007b).
The COVAID program consists of 10 group sessions, with individual support work as
required. It is a fully manualized program, which adheres to the principles of what works
in offender treatment. COVAID was accredited by the Ministry of Justice’s Correctional
Services Accreditation Panel (CSAP) in 2010 for use in prisons in England and Wales.
Small-scale evaluations have indicated that COVAID may be beneficial to offenders
(McCulloch & McMurran, 2008; McMurran & Cusens, 2003), with positive changes on
measures of alcohol-related violence expectancies and controlled drinking self-efficacy.
However, there is a need to evaluate COVAID using more robust research methods with
larger samples.
Randomized controlled trials (RCTs) are widely viewed as the most robust method of
outcome research (Farrington, 2003; Harper & Chitty, 2005). The current study is a feasi-
bility study for an RCT, one purpose of...