Journal of Psychopharmacology

http://jop.sagepub.com

A community based investigation of the association between cannabis use, injuries and accidents
E. J. K. Wadsworth, S. C. Moss, S. A. Simpson and A. P. Smith
J Psychopharmacol 2006; 20; 5 originally published online Oct 4, 2005;
DOI: 10.1177/0269881105056642

The online version of this article can be found at:

http://jop.sagepub.com/cgi/content/abstract/20/1/5

Published by:

http://www.sagepublications.com

On behalf of:

British Association for Psychopharmacology

Additional services and information for Journal of Psychopharmacology can be found at:

Email Alerts: http://jop.sagepub.com/cgi/alerts

Subscriptions: http://jop.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.co.uk/journalsPermissions.nav

Citations http://jop.sagepub.com/cgi/content/refs/20/1/5

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009

 J
Original Papers
Psychopharm

A community based investigation of the Journal of Psychopharmacology

20(1) (2006) 513
association between cannabis use, 2006 British Association
for Psychopharmacology

injuries and accidents ISSN 0269-8811

SAGE Publications Ltd,
London, Thousand Oaks,
CA and New Delhi
10.1177/0269881105056642

E. J. K. Wadsworth Centre for Occupational and Health Psychology, Cardiff University, 63 Park Place, Cardiff, CF10 3AS, UK.
S. C. Moss Centre for Occupational and Health Psychology, Cardiff University, 63 Park Place, Cardiff, CF10 3AS, UK.
S. A. Simpson Centre for Occupational and Health Psychology, Cardiff University, 63 Park Place, Cardiff, CF10 3AS, UK.
A. P. Smith Centre for Occupational and Health Psychology, Cardiff University, 63 Park Place, Cardiff, CF10 3AS, UK.

Abstract
There are well documented acute and chronic effects of cannabis use.
However, less is known about any effects on safety within the context of
work and everyday life.
The aim of the study was to examine any association between cannabis
use and injuries and accidents.
A postal questionnaire survey was conducted among people selected at
random from the electoral registers of Cardiff and Merthyr Tydl.
Cannabis use was associated with both minor injuries and accidents,
particularly among those with high levels of other associated risk factors.
Cannabis use was associated with a signicant detrimental impact on
safety. It is possible that this is linked to an amplication of other risk
factors associated with accidents and injuries. This has potentially wide
reaching implications particularly in the context of other work and
lifestyle characteristics.
Keywords
cannabis, injuries, accidents

Introduction use on cognitive performance. However, the evidence suggests

After alcohol and tobacco, cannabis is the most popular drug of
choice in Europe (Calafat, 1999). In the UK it has consistently
been the most commonly used illicit drug: 11% reported using it
in 2001/02 (Aust and Condon, 2003). British teenagers are more
likely to have tried illicit drugs than their counterparts elsewhere
in Europe (SCIAOD, 2000), and longitudinal work suggests that
drug use is increasingly limited to cannabis as people move into
adulthood (Williams and Parker, 2001). This suggests a pattern of
increasingly long cannabis use careers among substantial numbers
of people.
The sedative effects of cannabis use are well established, with
users typically reporting mental slowness, tiredness, anxiety and
paranoia as well as relaxation and euphoria (Parrott et al., 2004).
Acute effects on cognition and performance, limited to periods of
intoxication, have been well-documented (Parrott, 1987; Golding,
1992; Heishman et al., 1997; Solowij, 1998). Somewhat less
research has focused on the long-term effects of chronic cannabis
that long-term cannabis use leads to subtle and selective impair-
ments of specific higher cognitive functions (Solowij, 1998). The
results of a parallel part of this study investigating the impact of
cannabis use on cognitive performance among workers are consis-
tent with these findings and are reported elsewhere (Wadsworth et
al., 2005).
In the light of these results a possible impact on safety seems
intuitively likely. Laboratory studies show that cannabis impairs
driving skills (Parrott, 1987), perhaps at a similar level to a blood
alcohol concentration of between 0.07 and 0.10% (Hall et al.,
1994). Furthermore, recent drug use has been associated both with
traffic accidents (Ramaekers et al., 2004) and road accident culpa-
bility (Drummer et al., 2003). Accident and Emergency based
studies also show a high prevalence of drug use among attendees
(MacDonald et al., 1999). These studies (Seymour et al., 1999) and
driving accidents studies (Kelly et al., 2004) also indicate increas-
ingly prevalent polydrug use. Their focus is necessarily on more
serious accidents and on incidents occurring while intoxicated.

Corresponding author: E. J. K. Wadsworth, Centre for Occupational and Health Psychology, Cardiff University, 63 Park Place, Cardiff, CF10 3AS, UK.
Email: wadsworthej@cardiff.ac.uk

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009

6 Association between cannabis use, injuries and accidents
The aim of this study was to examine any association between
cannabis use and accidents, injuries and cognitive failures within
the context of work and everyday life. Cognitive failures are prob-
lems of memory, attention or action (effectively human errors).
Although under most circumstances they do not result in accident
or injury, cognitive failures can lead to accidents and injuries
(OHare et al., 1994; Larson et al., 1997). Each of these incident
types is also associated with particular characteristics which may
change with the circumstances in which the incident occurs.
Within our recent work we have shown associations with demo-
graphic (age, gender, income, education), personality (neuroti-
cism, risk taking), mental and physical health and lifestyle
(smoking, alcohol use, stress) factors (see Wadsworth et al., 2003;
Simpson et al., 2005). Incidents at work may be associated with
particular occupational characteristics in addition to these other
factors (see Wadsworth et al., 2003; Simpson et al., 2005). Fur-
thermore, drug use itself also seems to be linked to many factors.
These overlap with some of the same areas, such as risk taking,
gender, age, smoking and alcohol use (Wadsworth et al., 2004b),
though not necessarily in the same direction: for example, drug
use is associated with males (Wadsworth et al., 2004b), as are
accidents at work (Wadsworth et al., 2003; Simpson et al., 2005),
but cognitive failures at work are associated with females
(Simpson et al., 2005).
This study is an attempt to consider cannabis use and safety
both at work and outside work within the context of these other
factors and associations. Drug use is increasingly part of normal
young adult life (Parker et al., 2002). There is also growing evid-
ence of subtle cognitive effects of cannabis use linked to duration
of canabis use (Block and Ghoneim, 1993; Wadsworth et al.,
2005) that may not recover with abstinence (Solowij, 1995). It is
therefore increasingly important that research on the impact of
cannabis on safety attempts to control for the potentially con-
founding influence of other factors associated with accidents and
injuries (MacDonald et al., 2003).
Methods and Materials
A postal questionnaire survey was conducted among people
selected at random from the electoral registers of Cardiff and
Merthyr Tydfil. The questionnaire was based on that used in the
Bristol Stress and Health Study (Smith et al., 2000), and included
sections on demographics, health, accidents and injuries, health
related behaviours and work. It has been described in detail else-
where (Smith et al., 2004; Wadsworth et al., 2004a, 2004b), and
details of the measures used in the analyses are given in Table 1.
Participants
Thirty thousand people were selected at random from the electoral
registers for Cardiff and Merthyr Tydfil (22 500 and 7500, respec-
tively). These areas were selected for their differing social and
economic conditions. Employment rates and deprivation scores
are different for the two areas, with Merthyr Tydfil experiencing,
in general, higher unemployment and greater social and economic
Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009
deprivation than Cardiff. Index of Multiple Deprivation scores,
which combine the income, employment, health, education,
housing and service access domains are 17.88 for Cardiff, 42.77
for Merthyr and 21.75 for Wales as a whole (National Assembly
for Wales, 2000). Townsend scores (Townsend et al., 1988),
which measure deprivation by area using census data about unem-
ployment, overcrowding, non-car ownership and non-home own-
ership, for Cardiff and Merthyr Tydfil were 3.25 and 1.58,
respectively (National Assembly for Wales, 2000). And Jarman
scores (Jarman, 1984), which are an area-based measure of addi-
tional GP workload based on unemployment, overcrowding, lone
pensioners, single parents, born in the New Commonwealth, chil-
dren aged under 5, low social class and one year migrants, were
14.25 and 10.01 (National Assembly for Wales, 2000).
Procedure
Questionnaires and covering letters were posted in early May
2001. In an attempt to maximize frank reporting of drug use, no
identifiers were attached, precluding any follow-up procedure. The
questionnaires content has been described previously (Smith et
al., 2004). The questions about recreational drug use were taken
from the British Crime Survey (Ramsay and Partridge, 1998).
Outcome measures
Participants were asked about four incident types:
accidents during the previous year which required medical
attention
road traffic accidents during the previous year where the
respondent was the driver
minor injuries during the previous year which did not require
medical attention (quite and very frequent minor injuries were
compared with none, rare or occasional minor injuries)
cognitive failures problems of memory, attention or action
(quite and very frequent cognitive failures were compared with
none, rare or occasional cognitive failures).
They were asked about each incident type both outside work and
at work (in the case of traffic accidents this included driving to and
from work as well as driving as part of work).
Analyses
Analyses were carried out in two stages. First, logistic regression
modelling was used to assess any association between drug use
and accidents, minor injuries and cognitive failures. Each model
also included the factors listed in Table 1. This allowed any
association between cannabis use and the outcome measures to be
assessed independent of these other factors. The other factors were
chosen because of their association with accidents, injuries or cog-
nitive failures established previously (see final column). Some of
them are also associated with drug use (Wadsworth et al., 2004b),
and these are also identified in the final column.
Second, those factors listed in Table 1 that were associated
 Association between cannabis use, injuries and accidents 7

Table 1 Factors included in the logistic regression analyses

Factor Categorization Included because of previous association with

Demographics
Age 25, 25  40, 40  60, 60 Injuries1,2, cognitive1,2, drugs3
Sex Male, female Accidents1,2, cognitive2, drugs3
Income 10k, 10k  20k, 20k  30k, 30k Injuries1,2, cognitive2
Education No qualications, O level or equivalent, A level or equivalent,
City & Guilds, degree or high degree, professional qualication Drugs3

Personality
Neuroticism Quartiles of EPI-N Cognitive2, drugs3
Risk taking None, rare, occasional compared with quite or very frequent Injuries2, cognitive2, drugs3

Mental health
Anxiety Clinical cut-point of HADSa Cognitive1, 2
Depression Clinical cut-point of HADSd Cognitive2
Sleep problems Difculty sleeping in the last 14 days Injuries1,2, cognitive1

Physical health
14-day symptoms 3 or more compared with fewer from checklist Injuries2, cognitive1,2
12-month symptoms 2 or more compared with fewer from checklist Accidents2
Chronic symptoms 1 or more compared with fewer from checklist Accidents2, injuries2
General health Very good, good, moderate compared with bad or very bad Accidents2

Lifestyle
Smoking Current smokers compared with non-smokers Accidents1, injuries1, cognitive2, drugs3
Alcohol Those in the top 10% (calculated separately for men and women) Accidents2, drugs3
compared with those below

Other
Work status* Workers compared with non-workers Drugs3
Combined incidents Combinations of other accidents, injuries and cognitive failures Accidents2, injuries2, cognitive2
reported at work or outside work as appropriate

Occupational**
Risk taking at work None, rare, occasional compared with quite or very frequent Accidents2
Work stress None, mild, moderate compared with very or extreme Injuries2, cognitive1,2
Total negative score Quartiles of negative occupational characteristics Accidents1,2, injuries1,2, cognitive1
Social class Manual compared with non-manual Accidents2, injuries2
Employment Self-employed compared with others Injuries2
Experience Less than 6 months in post compared with longer Cognitive2

*Excluded from models for work-related outcomes. **Included only in models for work-related outcomes. 1Wadsworth et al., 2003, 2Simpson et al.
(2005), 3Wadsworth et al., 2004b.

with each incident type were established using backward stepwise
logistic regression. Participants were then categorized according to
whether they had high or low levels of these associated risk
factors. Within each of these categories, participants were further
divided into those who had used cannabis in the last year and
those who had not used any drugs. The resulting four level vari-
ables (i.e., low level of other risk factors and no cannabis use; low
level of other risk factors and cannabis use; high level of other risk
factors and no cannabis use; and high level of other risk factors
and cannabis use) were then included in logistic regression analy-
ses. This allowed the assessment of any association with cannabis
use over and above that of the other associated risk factors among
two groups of respondents: those with lower levels of other associ-
ated risk factors and those with higher levels of other associated
risk factors.

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009

8 Association between cannabis use, injuries and accidents

Ethical approval Table 2 Associations between cannabis use and reporting any type of
incident
The study was approved by the Cardiff University School of Psy-
chology Ethics Committee. OR CI

n  4676 Any incident

Results No drug use 1.00
Cannabis use 1.41 1.121.77
Participants Low levels of other risks and no drug use 1.00
Low levels of other risks and cannabis 1.52* 1.042.21
In total 7979 people completed and returned a questionnaire.
High levels of other risks and no drug use 2.60 2.282.96
Respondents were predominantly female (n  4601, 58%), their
High levels of other risks and cannabis 3.84** 3.004.92
mean age was 45.61 (SD  18.00, range 1697), most were
working (n  4620, 58%), most were white (n  7584, 97%) and
n  4895 Any non-work incident
26% had a degree or higher educational qualification, while 18%
had no educational qualifications.
No drug use 1.00
Twelve per cent reported having used drugs in the last year,
Cannabis use 1.37 1.071.74
and 7% in the last month. Among those who had used drugs in the
last year, 59% reported using only cannabis. Those who reported Low levels of other risks and no drug use 1.00
using other illict drugs as well as cannabis in the previous year Low levels of other risks and cannabis 1.34 0.872.06
were excluded from the following analyses. High levels of other risks and no drug use 2.74 2.393.14
Overall 11% of respondents reported an accident, 2% a road High levels of other risks and cannabis 4.28*** 3.365.46
traffic accident, 14% quite or very frequent minor injuries and
18% quite or very frequent cognitive failures. n  2859 Any work incident
Modelling all incidents together (i.e., any accident at work,
outside work or rta) in the last year, or quite or very frequent No drug use 1.00
minor injuries (at work or outside work) in the last year, or quite Cannabis use 1.34 0.951.88
or very frequent cognitive failures (at work or outside work) 2661
(38%) respondents reported an incident. Low levels of other risks and no drug use 1.00
Table 2 shows first the adjusted odds ratio (i.e., the odds ratio Low levels of other risks and cannabis 1.15 0.652.02
after adjusting for all the factors listed in Table 1) for reporting an High levels of other risks and no drug use 3.42 2.824.15
incident among cannabis users (OR  1.41). This indicates that High levels of other risks and cannabis 4.89**** 3.426.99
cannabis use was associated with reporting an incident independ-
ent of the other factors included in the model. Second, Table 2 Repeated measures contrasts p values: *0.03, **0.002, ***0.0001,
shows the odds ratios for the grouped variable categorizing ****0.04.
respondents according to both their level of other associated risk
factors and their cannabis use. Comparing levels of risk among
non-cannabis users (i.e., the non-bold lines) shows, as expected, Cognitive failures
an association between higher levels of associated risk factors and
reporting an incident (OR  2.60). Comparing cannabis use status There were no significant association between cannabis use and
within each level of risk (i.e., each bold line with the non-bold line cognitive failure (see Table 3).
below it) shows associations between cannabis use and reporting
an incident both for those with lower levels of other associated Minor injuries
risk factors (OR  1.52) and for those with higher levels of other
associated risk factors (OR  3.84). Cannabis use was associated with quite or very frequent minor
Similarly, analysing work and non-work incidents separately injuries. This was significant among those with high levels of
showed that for non-work-incidents there was an association with other risk factors. Restricting the analyses to minor injuries
cannabis use, and this was also significant among those with high outside work and then at work also showed this pattern: cannabis
levels of other risk factors. For work incidents there was an associ- use was associated with quite or very frequent minor injuries
ation with cannabis use among those with high levels of other risk outside work this was significant among those with high levels
factors (the overall adjusted odds ratio also approached signifi- of other risk factors; cannabis use was also associated with minor
cance) (see Table 2). injuries at work among those with high levels of other risk factors
Next each incident type was considered separately. only (see Table 4).

Accidents
Considering all accidents together there was an association with

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009

 Association between cannabis use, injuries and accidents 9

Table 3 Associations between cannabis use and cognitive failures Table 4 Associations between cannabis use and minor injuries

OR CI OR CI

n  4424 Cognitive failures n  4424 Minor injuries

No drug use 1.00 No drug use 1.00
Cannabis use 1.19 0.861.64 Cannabis use 1.34 0.991.80
Low levels of other risks and no drug use 1.00 Low levels of other risks and no drug use 1.00
Low levels of other risks and cannabis 1.15 0.671.98 Low levels of other risks and cannabis 1.51 0.882.60
High levels of other risks and no drug use 3.63 2.984.43 High levels of other risks and no drug use 3.14 2.583.84
High levels of other risks and cannabis 4.39 3.106.21 High levels of other risks and cannabis 6.39* 4.688.72

n  4754 Cognitive failures outside n  4754 Minor injuries outside

work work
No drug use 1.00 No drug use 1.00
Cannabis use 1.15 0.831.62 Cannabis use 1.51 1.072.14
Low levels of other risks and no drug use 1.00 Low levels of other risks and no drug use 1.00
Low levels of other risks and cannabis 1.12 0.691.81 Low levels of other risks and cannabis 1.52 0.802.89
High levels of other risks and no drug use 3.53 2.924.27 High levels of other risks and no drug use 3.21 2.524.09
High levels of other risks and cannabis 3.90 2.476.18 High levels of other risks and cannabis 7.84* 5.5511.07

n  2801 Cognitive failures at work n  2801 Minor injuries at work

No drug use 1.00 No drug use 1.00
Cannabis use 1.35 0.822.21 Cannabis use 1.17 0.741.86
Low levels of other risks and no drug use 1.00 Low levels of other risks and no drug use 1.00
Low levels of other risks and cannabis 1.47 0.772.80 Low levels of other risks and cannabis 1.10 0.472.57
High levels of other risks and no drug use 3.93 3.035.09 High levels of other risks and no drug use 4.13 3.095.51
High levels of other risks and cannabis 4.47 2.667.52 High levels of other risks and cannabis 8.49** 5.3713.42

Repeated measures contrasts p values: *0.0001, **0.001.

cannabis use among those with high levels of other risk factors
only. This pattern was repeated when just accidents outside work
were considered. There was no association between cannabis use
and work accidents (see Table 5).
Road trafc accidents
Cannabis use was associated with all road traffic accidents (see
Table 6). Those with low levels of other risk factors who used
cannabis were more likely to report a traffic accident (see Table
6). Similarly, those with low levels of other risks who also used
cannabis were more likely to report a work-related road traffic
accident in the last year (the overall odds ratio for work-related
road traffic accidents also approached significance) (see Table 6).
There was no significant association between cannabis use and
non-work-related road traffic accidents.
Discussion
Cannabis use was associated with self-reported accidents, road
traffic accidents and minor injuries. Specifically, those who had
higher levels of other risk factors associated with accidents and
who also used cannabis were more likely to report an accident in
the previous year. This pattern was repeated when accidents
outside work were analysed separately, but was not apparent for
accidents at work. Those who used cannabis were also more likely
to report a road traffic accident in the previous year. This was also
apparent among those with lower levels of other associated risk
factors. Analysing work- and non-work related road accidents sep-
arately showed an association only for work-related accidents (i.e.,
while driving to or from work, or as part of work) among those
who also reported lower levels of other associated risk factors.
Cannabis users were also more likely to report quite or very fre-
quent minor injuries. This too was apparent among those who had
higher levels of other associated risk factors. The same pattern
emerged when non-work related minor injuries were analysed sep-
arately, while for work-related minor injuries the only association
was among those with higher levels of other associated risk factors.
However, there was no association between cannabis use and
cognitive failure. This suggests that cannabis users may be
unaware of any impact on their cognitive performance. This is
consistent with recent parallel work which suggests measurable
cognitive performance deficits among cannabis users but relatively
little awareness by users of any detrimental performance effects at
work (Wadsworth et al., 2005).

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009

10 Association between cannabis use, injuries and accidents

Table 5 Associations between cannabis use and accidents Table 6 Associations between cannabis use and road trafc accidents

OR CI OR CI

n  4424 Accidents n  4429 All road trafc accidents

No drug use 1.00 No drug use 1.00
Cannabis use 0.98 0.701.37 Cannabis use 1.92 1.043.54
Low levels of other risks and no drug use 1.00 Low levels of other risks and no drug use 1.00
Low levels of other risks and cannabis 1.02 0.492.12 Low levels of other risks and cannabis 2.91* 1.008.47
High levels of other risks and no drug use 2.22 1.812.71 High levels of other risks and no drug use 3.21 2.055.03
High levels of other risks and cannabis 3.02* 2.184.19 High levels of other risks and cannabis 5.03 2.669.52

n  4754 Accidents outside work n  4754 Non-work road trafc

No drug use 1.00 accidents
Cannabis use 1.01 0.701.46 No drug use 1.00
Low levels of other risks and no drug use 1.00 Cannabis use 1.73 0.813.69
Low levels of other risks and cannabis 1.32 0.682.55 Low levels of other risks and no drug use 1.00
High levels of other risks and no drug use 2.10 1.692.61 Low levels of other risks and cannabis 1.83 0.556.11
High levels of other risks and cannabis 3.16** 2.224.49 High levels of other risks and no drug use 4.52 2.777.38
High levels of other risks and cannabis 6.75 3.1114.63
n  2801 Accidents at work
No drug use 1.00 n  2801 Work related road trafc
Cannabis use 0.91 0.431.89 accidents
Low levels of other risks and no drug use 1.00 No drug use 1.00
Low levels of other risks and cannabis 0.92 0.223.92 Cannabis use 3.01 0.8910.17
High levels of other risks and no drug use 3.11 2.004.84 Low levels of other risks and no drug use 1.00
High levels of other risks and cannabis 3.85 1.897.82 Low levels of other risks and cannabis 3.24** 1.198.79
High levels of other risks and no drug use 3.03 1.456.33
Repeated measures contrasts p values: *0.04, **0.02. High levels of other risks and cannabis 6.06 1.3726.77

Repeated measures contrasts p values: *0.05, **0.02.

Some caution is needed interpreting these data. Timings of
cannabis use in relation to accidents and injuries were not avail-
able, the data were entirely self-reported, and there was no
information about amount, frequency or duration of cannabis use.
However, they do suggest a possible impact of cannabis use on
safety both in and out of the workplace. They are also consistent
with laboratory findings suggesting that, for example, acute
cannabis use affects the attention, tracking and psychomotor skills
used in driving (Coambs and McAndrews, 1994). Similarly, they
are consistent with the effects on psychomotor speed, information
processing (Wadsworth et al., 2005), and the attentional/executive
system (Pope and Yurgelun-Todd, 1996) associated with longer-
term use, and the suggestion that longer-term users process some
kinds of information more slowly, an effect which may be more
apparent under heavier cognitive load (Leavitt et al., 1992, 1993).
The associations described may represent acute effects (i.e.,
accidents or injuries occurring while intoxicated). Drugs have
been found in up to 25% of those involved in road accidents with
cannabis generally the most commonly detected (Kelly et al.,
2004). There is growing evidence that recent cannabis use (deter-
mined by detection of THC in the blood rather than past use
determined by detection of THC in the urine), particularly at
higher concentrations, does increase traffic accident risk (Ramaek-
ers et al., 2004).
In addition, recent studies of road traffic accidents have sug-
gested that culpability increased with higher concentrations of
blood THC (Drummer et al., 2003). However, it is also possible
that they include an association with accidents or injuries while
not intoxicated resulting from cognitive performance deficits asso-
ciated with longer-term cannabis use (Solowij, 1998). This would
be consistent with parallel work suggesting impairments of psy-
chomotor speed, information processing and memory among those
using cannabis for longer (Wadsworth et al., 2005).
Epidemiological studies, despite showing a correlation
between cannabis use and injuries, cannot rule out alternative
explanations, and have shown inconsistent results (MacDonald et
al., 2003). For example, the association between cannabis use and
traffic accidents in a population-based study in Canada may reflect
risk-taking behaviour in general (Chipman, 1995). Risky driving
behaviour is common among younger drivers, particularly males
prone to externalizing behaviour (including substance abuse) (Fer-
gusson et al., 2003b). Alternatively, the findings may represent
injury as a result of detrimental effects of withdrawal such as rest-
lessness and anxiety (Ashton, 2001). Indeed, there is some evid-

Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009


ence that information processing speeds among heavy cannabis
users may be normalized when the drug is used (Kelleher et al.,
2004).
This work represents an attempt to control for the potentially
confounding influence of many other factors known to be associ-
ated with accidents and injuries as called for in a recent compre-
hensive review of the literature (MacDonald et al., 2003).
However, the extent to which statistical adjustment overcomes
potential confounding is dependent on the identification and meas-
urement of those factors. Alternate explanations for the associ-
ations between cannabis use and accident and injury cannot be
ruled out by this study. Despite this, the study suggests that there
is an association between cannabis use and both accidents and
injuries over and above that of the combined effect of higher
levels of other factors associated with accident and injury. This
points to the possibility of an independent association with
cannabis use. It also implies that either cannabis use may amplifiy
the risks associated with other influential factors, or that these
other influential factors may amplify the risk associated with
cannabis use. This has been suggested previously with reference to
road traffic accidents where cannabis use may amplify the effects
of alcohol (Hall et al., 1994). Similarly, the combination of prob-
lematic alcohol use and cannabis use in adolescence is more
strongly associated with both adult alcohol and drug abuse than
adolescent use of cannabis or alcohol alone (Stenbacka, 2003).
Parallels with many other areas also exist. For example work
characteristics, where the imbalance of effort and reward is a
powerful predictor of coronary heart disease (Siegrist, 1996;
Bosma et al., 1998), and hypertension where control is improved
by the use of drug combinations (Borghi et al., 2004).
These results highlight the fact that cannabis use takes place
within a wide range of other daily behaviours, influences, factors
and contexts. Some of these other characteristics are also associ-
ated with an impact on safety. There may be some interaction
between one or more of these factors and cannabis use that modi-
fies their relationship with safety. Altering accident, injury and
error rates is dependent on the identification of factors that are
both associated with the outcomes and are modifiable. Research
must therefore include the context and other associated factors to
assess not only any independent effects of cannabis use, but also
any interactive influence between cannabis use and other factors.
A further recent review has identified a consistent association
between cannabis use and lower educational attainment (Macleod
et al., 2004). Respondents in this study were more likely to have a
higher educational qualification, as were those who had used other
drugs as well as cannabis (Wadsworth et al., 2004b). Among those
who took part in parallel laboratory-based work, however, fewer
cannabis users than controls had a higher educational qualification
(61% vs. 74%), though this difference was not significant
(Wadsworth et al., 2005). These findings probably reflect the
population sampled (Cardiff is a university city with a relatively
young population), and some selection bias. As a cross-sectional
study with a self-selected sample these data cannot infer causality.
However, the implication is that some of the cognitive perform-
ance (Wadsworth et al., 2005) and safety problems associated
with cannabis use are also apparent among cannabis users with
Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009
Association between cannabis use, injuries and accidents 11
relatively high educational attainment and were apparent when
education was included in the analyses. This is consistent with the
suggestion that the association between cannabis use and lower
educational achievement may reflect the effects of the social
context of cannabis use rather than any direct effect on ability or
motivation (Fergusson et al., 2003a).
Similarly, alcohol use was controlled for in the analyses, sug-
gesting that even though drug use and heavy drinking are strongly
linked (Leitner et al., 1993; Goddard and Higgins, 2000;
Wadsworth et al., 2004b), these findings may represent a cannabis
effect rather than a more general substance use effect.
There is little work focusing on the use of only cannabis and
accidents or injuries. However, research considering more general
drug use suggests links with accidents and injuries outside the
workplace (MacDonald et al., 1999; Seymour et al., 1999), which
may well represent the more acute effects of drug use. There is
little evidence, though, for any association between drug use and
accidents at work (Normand et al., 1990; Dell and Berkhout,
1998), where being under the influence of drugs at the time is
likely to be much less prevalent (Alleyne et al., 1991). Rather,
research in this area often suggests that drug use is associated with
higher absenteeism and employee turnover (Normand et al., 1990;
Peat, 1995). There is also little research into those injuries that are
much less serious (i.e., do not require medical attention). These
are not reported and recorded in work, and by definition do not
come into contact with the medical system. They are, however,
more prevalent. It is also possible that epidemiological studies do
not find a link between drug use and workplace accidents and
injuries because there is an element of self-selection. Those who
use cannabis may choose to do less inherently dangerous jobs, and
those who do riskier jobs may choose not to use cannabis. Further-
more, cannabis use seems to produce subtle and selective cogni-
tive impairments that may only become apparent after several
years of use (Solowij, 1998) and may be linked to duration of use
(Hall and Solowij, 1998; Solowij, 1998; Wadsworth et al., 2005).
Studies of road traffic accidents and drug use are increasingly
focusing on culpability (e.g., Drummer et al., 2003). Research into
other kinds of accidents and injuries also needs to find a way to
include this concept. If cannabis use is involved in accidents or
injuries any effect is due to the cognitive impairment associated
with either acute or chronic use. As such research should focus on
accidents and injuries that result from human error and on those
who make such errors regardless of whether they are the ones
injured as a result. To this end it could be argued that minor
injuries (such as cuts and bruises), which were more strongly asso-
ciated with cannabis use here than accidents, may be more likely
to be sustained by an individual as a result of their own error.
As described above, there are limitations to the methodology
used in this study. The results, however, suggest a detrimental
impact of cannabis use on safety that is apparent both in and out of
the workplace. In particular the associations were significant
among those who also had higher levels of other risk factors. This
not only emphasizes the importance of studying cannabis use in
context, but also raises the possibility that, as with road traffic
accidents (Hall et al., 1994) cannabis may amplify other risk
factors.
12 Association between cannabis use, injuries and accidents
Cannabis is the most widely used, and earliest first used, illicit
drug in the UK (Aust et al., 2002), and it is the one most likely to be
continued into adulthood (Williams and Parker, 2001). It is frequently
used in combination with other illicit drugs and with alcohol and is
increasingly part of young adult life (Parker et al., 2002) among
otherwise conforming individuals (Williams and Parker, 2001). It is,
therefore, increasingly important to understand not only any cognitive
impact of cannabis use, but also any consequent safety implications
within the context of other real world factors both as a result of acute
intoxication and the longer-term sequelae of chronic use.
Acknowledgement
Funded by the Health and Safety Executive.
References
Alleyne B C, Stuart P, Copes R (1991) Alcohol and other drug use in
occupational fatalities. Journal of Occupational Medicine 33(4):
496500
Ashton C H (2001) Pharmacology and effects of cannabis: a brief review.
British Journal of Psychiatry 178(101): 101106
Aust R, Condon J (2003) Geographic Variation in Drug Use: Key Findings
from the 2001/02 British Crime Survey England and Wales, Develop-
ment and Statistics Directorate, Home Office
Aust R, Sharp C, Goulden C (2002) Prevalence of Drug Use: Key Findings
from the 2001/2002 British Crime Survey. Research, Development and
Statistics Directorate, Home Office
Block R I, Ghoneim M M (1993) Effects of chronic marijuana use on
human cognition. Psychopharmacology 110: 219228
Borghi C, Dormi A, DAddato S, Gaddi A, Ambrosioni E, Brisighella
Heart Study Working Party (2004) Trends in blood pressure control
and antihypertensive treatment in clinical practice: the Brisighella
Heart Study. Journal of Hypertension 22(9): 17071716
Bosma H, Peter R, Siegrist J, Marmot M (1998) Two alternative job stress
models and the risk of coronary heart disease. American Journal of
Public Health 88(1): 6874
Calafat A, Bohrn K, Juan M, Kokkevi A, Maalste N, Mendes F, Palmer A,
Sherlock K, Simon J, Stocco P, Sureda M, Tossmann P, van de Wijn-
gaart C, Zavatti P (1999) Night life in Europe and recreative drug use.
Valencia: IREFREA & European Commission
Chipman M (1995) Risk factors for injury: similarities and differences for
traffic crashes and other causes. Accident Analysis and Prevention
27(5): 699706
Coambs R B, McAndrews M P (1994) The effects of psychoactive sub-
stances on workplace performance. In: MacDonald S, Roman P (eds),
Drug Testing in the Workplace. Plenum Press, New York, pp. 77102
Dell T, Berkhout J (1998) Injuries at a metal foundry as a function of job
classification, length of employment and drug screening. Journal of
Safety Research 29: 914
Drummer O H, Gerostamoulos J, Batziris H, Chu M, Caplehorn J, Robert-
son M D, Swann P (2003) The involvement of drugs in drivers of
motor vehicles killed in Australian road traffic crashes. Accident
Analysis and Prevention 943: 110
Fergusson D M, Horwood J L J, Beautrais A L (2003a) Cannabis and edu-
cational achievement. Addiction 98: 16811692
Fergusson D, Swain-Campbell N, Horwood J (2003b) Risky driving
behaviour in young people: prevalence of personality characteristics
and traffic accidents. Aust N Z J Public Health 27(3): 337342
Goddard E, Higgins V (2000) Drug use, Smoking and Drinking among
Young Teenagers in 1999. HMSO, London
Golding J F (1992) Cannabis. In: Smith A, Jones D (eds), Handbook of
Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009
Human Performance: Health and Performance, Vol. 2. Academic
Press, New York, p. 175
Hall W, Solowij N, Lemon J (1994) The Health and Psychological Con-
sequences of Cannabis Use. National Drug Strategy Monograph, Series
No. 25, Australian Government Publishing Service, Canberra
Hall W, Solowij N (1998) Adverse effects of cannabis. Lancet 352:
16111616
Heishman S J, Arasteh K, Stitzer M L (1997) Comparative effects of
alcohol and marijuana on mood, memory and performance. Pharmacol-
ogy, Biochemistry and Behavior 58: 93101
Jarman B (1984) Underprivileged areas: validation and distribution of
scores. British Medical Journal 289: 15871592
Kelleher L M, Stough C, Sergejew A A, Rolfe T (2004) The effects of
cannabis on information-processing speed. Addictive Behaviors 29:
12131219
Kelly E, Darke S, Ross J (2004) A review of drug use and driving: epi-
demiology, impairment, risk factors and risk perceptions. Drug and
Alcohol Review 23(3): 319344
Larson G E, Alderton D L, Neideffer M, Underhill E (1997) Further evid-
ence of dimensionality and correlates of the Cognitive Failures Ques-
tionnaire. British Journal of Psychology 88: 2938
Leavitt J, Webb P, Norris G, Struve F, Straumanis J, Fitzgerald M J,
Nixon F, Patrick G, Manno J (1993) Performance of chronic daily
marijuana users on neuropsychological tests. In: Harris L (ed.), Prob-
lems of Drug Dependence 1992, National Institute on Drug Abuse
Research Monograph 132, US Government Printing Office, Washing-
ton, DC, p. 179
Leavitt J, Webb P, Norris G, Struve F, Straumanis J, Patrick G, Fitzgerald
M J, Nixon F (1992) Differences in complex reaction time between
THC users and non-user controls. In: Harris L (ed.), Problems of Drug
Dependence 1991, National Institute on Drug Abuse Research Mono-
graph 119, US Government Printing Office, Washington, DC, p. 452
Leitner M, Shapland J, Wiles P (1993) Drug usage and drugs prevention.
HMSO, London
MacDonald S, Anglin-Bodrug K, Mann R E, Erickson P, Hathaway A,
Chipman M, Rylett M (2003) Injury risk associated with cannabis and
cocaine use. Drug and Alcohol Dependence 72: 99115
MacDonald S, Wells S, Giesbrecht N, Cherpitel C (1999) Demographic
and substance use factors related to self-reported violent and accidental
injuries: results from an emergency room study. Drug and Alcohol
Dependence 55: 5361
Macleod J, Oakes R, Copello A, Crome I, Egger M, Hickman M,
Oppenkowski T, Stokes-Lampard H, Davey Smith G (2004) Psycho-
logical and social sequelae of cannabis and other illicit drug use by
young people: a systematic review of longitudinal, general population
studies. Lancet 363: 15791588
National Assembly for Wales (2000) Indices of Deprivation of Electoral
Divisions in Wales, 2000 (J3). National Assembly for Wales, Cardiff
Normand J, Salyards S D, Mahoney J J (1990) An evaluation of pre-
employment drug testing. Journal of Applied Psychology (Washington,
DC) 75(6): 629639
OHare D, Wiggins M, Batt R, Morrison D (1994) Cognitive failure analy-
sis for aircraft accident investigation. Ergonomics 37: 18551869
Parker H, Williams J, Aldridge J (2002) The normalization of sensible
recreational drug use: further evidence from the North West England
Longitudinal Study. Sociology 36: 941964
Parrott A C (1987) Assessment of psychological performance in applied
situations. In: Hindmarch I, Stonier PD (eds), Human Psychopharma-
cology: Measures and Methods, Vol. 1. John Wiley & Sons, Chich-
ester, UK

Parrott A, Morinan A, Moss M, Scholey A (2004) Understanding Drugs
and Behaviour. John Wiley & Sons Ltd, Chichester, UK
Peat M A (1995) Financial viability of screening for drugs of abuse. Clini-
cal Chemistry 41(5): 805808
Pope Jr H G, Yurgelun-Todd D A (1996) The residual cognitive effects of
heavy marijuana use in college students. JAMA 275: 521527
Ramsay M, Partridge S (1998) Drug Misuse Declared in 1998: Results
from the British Crime Survey. Home Office, London
Ramaekers J G, Berghaus G, van Laar M, Drummer O H (2004) Dose
related risk of motor vehicle crashes after cannabis use. Drug and
Alcohol Dependence 73: 109119
Seymour A, Oliver J S, Chen C (1999) The role of drugs and alcohol in
impaired drivers and fatally injured drivers in the Strathclyde police
region of Scotland, 19951998. Forensic Science International 103:
89100
Siegrist J (1996) Adverse health effects of high-effort/low-reward con-
ditions. Journal of Occupational and Health Psychology 1(1): 2741
Simpson S A, Wadsworth E J K, Moss S C, Smith A P (2005 Minor
injuries, cognitive failures and accidents at work: prevalence and asso-
ciated features. Occupational Medicine 55: 99108
Smith A, Johal S, Wadsworth E, Peters T, Davey Smith G (2000) The
scale of occupational stress: the Bristol Stress and Health at Work
Study. HSE, London
Smith A, Wadsworth E, Moss S, Simpson S (2004) The scale and impact
of drug use by workers. Health and Safety Executive Research Report
No. 192, HSE Books, London
Solowij N (1995) Do cognitive impairments recover following cessation of
cannabis use?. Life Sciences 56: 21192126
Downloaded from http://jop.sagepub.com by Juan Martell Muoz on February 17, 2009
Association between cannabis use, injuries and accidents 13
Solowij N (1998) Cannabis and Cognitive Functioning. Cambridge: Cam-
bridge University Press
Stenbacka M (2003) Problematic alcohol and cannabis use in adolescence
risk of serious adult substance abuse? Drug and Alcohol Review
22(3): 277286
Swedish Council for Information in Alcohol and Other Drugs (SCIAOD)
(2000) European School Survey Project on Alcohol and Other Drugs:
Summary of the 1999 Findings. SCIAOD, Stockholm
Townsend P, Phillimore P, Beattie A (1988) Health and Deprivation:
Inequality and the North. Croom Helm, London
Wadsworth E J K, Simpson S A, Moss S C, Smith A P (2004a) Recre-
ational drug use: patterns from a South Wales self-report survey.
Journal of Psychopharmacology 18(2): 228237
Wadsworth E J K, Moss S C, Simpson S A, Smith A P (2004b) Factors
associated with recreational drug use. Journal of Psychopharmacology
18(2): 238248
Wadsworth E J K, Moss S C, Simpson S A, Smith A P (2006) Cannabis
use, cognitive performance and mood in a sample of workers. Journal
of Psychopharmacology 20(1): 1423
Wadsworth E J K, Simpson S A, Moss S C, Smith A P (2003) The Bristol
Stress and Health Study: accidents, minor injuries and cognitive fail-
ures at work. Occupational Medicine 53: 392397
Williams L, Parker H (2001) Alcohol, cannabis, ecstasy and cocaine:
drugs of reasoned choice amongst young adult recreational drug users
in England. International Journal of Drug Policy 12: 397413