Modeling the System of Beliefs That Affect Driving Under the Influence of Cannabis and Alcohol in Washington State

• Author: Brandon G. Scott,Jay Otto,Nicholas Ward
• Date: 01 October 2021
• Published date: 01 October 2021
• DOI: 10.1177/00220426211028567
• Subject Matter: Articles

2021, Vol. 51(4) 661 –678

https://doi.org/10.1177/00220426211028567

Journal of Drug Issues

© The Author(s) 2021

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00220426211028567

journals.sagepub.com/home/jod

Article

Modeling the System of Beliefs

That Affect Driving Under the

Influence of Cannabis and Alcohol

in Washington State

Brandon G. Scott1, Nicholas Ward1,

and Jay Otto1

Abstract

Washington state has observed increases in polydrug use in fatal crashes, primarily involving the

combination of cannabis and alcohol. The purpose of this article is to explore the belief system

associated with driving under the influence of cannabis and alcohol (DUICA) in Washington

state using structural equation modeling (SEM). A convenience sample (n = 737) of surveys

collected from adults in Washington state was analyzed using SEM to reveal the latent structure

of the belief system associated with DUICA. The results of this analysis indicated that the

reported DUICA behavior (frequency) was predicted by intention and willingness. Willingness

also predicted intention. Intention and willingness were predicted by positive attitudes

toward DUICA, as well as normative perceptions that it was acceptable to important people

and common behavior for most people. These components were themselves predicted by

corresponding beliefs (behavioral beliefs, normative beliefs, and control beliefs). Finally, these

beliefs were also influenced by the values that were most important to the respondents. Based

on these results, it is reasonable to speculate that strategies to change these beliefs may also

reduce DUICA behavior and associated fatal crashes.

Keywords

cannabis, alcohol, fatal crashes, traffic safety culture

Historically, alcohol has been the drug most often involved in fatal crashes related to impaired

driving. In 2016, 10,497 people died in alcohol-related traffic crashes—28% of all traffic-related

deaths in the United States (Centers for Disease Control and Prevention, 2018). Recently, the

presence of alcohol in drivers surveyed on weekend nights decreased significantly from 16.9%

in 2007 to a low of 8.3% in 2014 (Berning et al., 2015). In contrast, the presence of cannabis in

surveyed drivers increased from 8.6% to 12.6% over this same period (Berning et al., 2015). This

is consistent with the increasing prevalence of cannabinoids detected in fatally injured drivers

(Rudisill et al., 2014).

1Montana State University, Bozeman, USA

Corresponding Author:

Nicholas Ward, Center for Health and Safety Culture, Montana State University, P.O. Box 1745, Bozeman, MT

59717-3820, USA.

Email: nward@montana.edu

1028567JODXXX10.1177/00220426211028567Journal of Drug IssuesScott et al.

research-article2021

662 Journal of Drug Issues 51(4)

2 Journal of Drug Issues 00(0)

There is also evidence that fatal crashes related to the polydrug use of cannabis and alcohol

are increasing (Grondel et al., 2018). A recent representative survey of adults in Washington state

(which has decriminalized both recreational and medicinal cannabis) conducted by Otto et al.

(2021) indicated that approximately 8.6% of respondents reported ever driving under the influ-

ence of cannabis and alcohol (DUICA) in the past 12 months.

It is difficult for drivers to compensate for the impairment resulting from combining cannabis

and alcohol because of their synergistic effects (Sewell et al., 2009). Consequently, both the risk

and responsibility for a fatal crash are higher with the combination of cannabis and alcohol than

with either drug alone (Dubois et al., 2015). Given the significance of this form of polydrug use

to traffic safety, it is necessary to understand the beliefs of drivers that predict DUICA.

Traffic safety culture has been defined as “the shared belief system of a group of people,

which influences road user behaviors and stakeholder actions that impact traffic safety” (Ward

et al., 2019). Understanding road user beliefs associated with DUICA can inform efforts to pre-

vent this behavior. For example, Grondel et al. (2018) have suggested the decision to DUICA

may be related to beliefs that cannabis can improve driving ability and, therefore, compensate for

the impairment effects of alcohol. Indeed, this belief that cannabis use may compensate for alco-

hol impairment may explain recent evidence that DUICA is more common among cannabis users

than noncannabis users (Park & Wu, 2019). Similarly, this explanation is consistent with evi-

dence that drivers testing positive for alcohol are more likely to test positive for other drugs

(Lacey et al., 2005).

To explore the reasons that drivers may decide to DUICA, this study examined the decision to

DUICA based on a model of traffic safety culture represented by a system of cascading beliefs

described in Figure 1. Beliefs positioned early in the model influence the position of beliefs later

in the model, which then influence willingness and intention to enact the behavior. Table 2 sum-

marizes each of the beliefs. Structural equation modeling (SEM) was used to test hypotheses

examining the simultaneous effects along all the network pathways and representing each belief

type as a latent variable.1

Method

Details of the survey design to measure DUICA beliefs and the method for data collection are

contained in Otto et al. (2021). The belief model shown in Figure 1 depicts the different compo-

nents of traffic safety culture and its predicted relationship to DUICA behavior. This model inte-

grates several theories of belief-based decision making including the Value-Belief-Norm Theory

(Oreg & Katz-Gerro, 2006; Stern, 2000), the Reasoned Action Approach (Fishbein & Ajzen,

2010), and the Prototype-Willingness Model (Gerrard et al., 2008) and was found to be effective

in modeling driving under the influence of cannabis (Scott et al., 2021). In addition to providing

a framework for designing the survey, the model pathways also represented hypotheses to be

tested in the analysis of the data.

Survey Design

The survey was designed to measure the different beliefs represented in the model of traffic

safety culture (Figure 1). The definitions of these beliefs are summarized in Table 1. The specific

survey items used to measure each type of belief are listed in Table 2.

Sample Design

The survey was reviewed and approved by the Institutional Review Board of Montana State

University. The survey was implemented by NORC (formerly the National Opinion Research