EDITORIAL New Deviancy Theory and the Healthcare System’s Role in Creating, Labeling, and Facilitating Unauthorized Prescription Drug “Abuse” Although the “new” deviancy theories are now more than 50 years old, they continue to inform our under- standing of contemporary patterns of illicit drug use.1 The lines between licit and illicit forms of drug use have become increasingly blurred as the “new” deviancy theories were developed: the “misuse” of prescription opiate analgesics and the expansion of marijuana prescribing have pulled much of the discussion sur- rounding these activities into the medical sphere. It is our contention that the application of classic sociolog- ical and criminological theory lends valuable insights into contemporary patterns of unauthorized prescrip- tion drug “abuse.” Deviance is a sociological concept that refers to behaviors and beliefs that deviate from the norms, standards, and expectations of a given society.1 It is a broader concept than crime and is distinct from the notion of “difference” in that it contains the implicit likelihood of authoritative intervention or sanction: that is, it refers to behaviors and beliefs that are stigmatized. “New” deviancy theories challenged established ways of thinking about such phenomena by rejecting the idea that there is a distinct, unambiguously deviant minority whose behavior can be explained as a result of individ- ual pathology or social dysfunction. In place of the traditional “correctionalist” orientation, an “apprecia- tive stance” was advocated that is committed to faithful understanding of the world as seen by the subject. Viewed from this perspective, it was argued that deviance is meaningful behavior involving choice and that there is an underlying continuity between normalcy and deviance.2 Such continuity is evident in the use of prescription medications, which is deemed legitimate when it is authorized by a physician to treat a medical ailment, but is likely to be deemed deviant if patients continues to use when there is no longer a medical need to do so—either for pleasure or because they have become dependent. Howard Becker provided the most famous statement of the “new” deviancy position when he noted that “deviance” is not a quality of the act the person commits, but rather a consequence of the application by others of rules and sanctions to an “offender”: deviant behavior, in other words, “is behavior that people so label.”3 In his seminal work, Becoming a Marihuana User, Becker describes a series of learned steps that he deemed necessary for someone to become a regular drug user: . . . No one becomes a user without 1 - learning to smoke the drug in a way which will produce real effects; 2 - learning to recognize the effects and connect them with the drug use (learning, in other words, to get high); and 3 - learning to enjoy the sensations he perceives.4 Due to the illegality of marijuana use thr ...

1. EDITORIAL
New Deviancy Theory and the Healthcare
System’s Role in Creating, Labeling, and
Facilitating Unauthorized Prescription
Drug “Abuse”
Although the “new” deviancy theories are now more
than 50 years old, they continue to inform our under-
standing of contemporary patterns of illicit drug use.1
The lines between licit and illicit forms of drug use have
become increasingly blurred as the “new” deviancy
theories were developed: the “misuse” of prescription
opiate analgesics and the expansion of marijuana
prescribing have pulled much of the discussion sur-
rounding these activities into the medical sphere. It is
our contention that the application of classic sociolog-
ical and criminological theory lends valuable insights

2. into contemporary patterns of unauthorized prescrip-
tion drug “abuse.”
Deviance is a sociological concept that refers to
behaviors and beliefs that deviate from the norms,
standards, and expectations of a given society.1 It is a
broader concept than crime and is distinct from the
notion of “difference” in that it contains the implicit
likelihood of authoritative intervention or sanction: that
is, it refers to behaviors and beliefs that are stigmatized.
“New” deviancy theories challenged established ways of
thinking about such phenomena by rejecting the idea
that there is a distinct, unambiguously deviant minority
whose behavior can be explained as a result of individ-
ual pathology or social dysfunction. In place of the
traditional “correctionalist” orientation, an “apprecia-
tive stance” was advocated that is committed to faithful
understanding of the world as seen by the subject.
Viewed from this perspective, it was argued that

3. deviance is meaningful behavior involving choice and
that there is an underlying continuity between normalcy
and deviance.2 Such continuity is evident in the use of
prescription medications, which is deemed legitimate
when it is authorized by a physician to treat a medical
ailment, but is likely to be deemed deviant if patients
continues to use when there is no longer a medical need
to do so—either for pleasure or because they have
become dependent.
Howard Becker provided the most famous statement
of the “new” deviancy position when he noted that
“deviance” is not a quality of the act the person
commits, but rather a consequence of the application
by others of rules and sanctions to an “offender”:
deviant behavior, in other words, “is behavior that
people so label.”3 In his seminal work, Becoming a
Marihuana User, Becker describes a series of learned
steps that he deemed necessary for someone to become a
regular drug user:

4. . . . No one becomes a user without 1 - learning to
smoke the drug in a way which will produce real
effects; 2 - learning to recognize the effects and
connect them with the drug use (learning, in other
words, to get high); and 3 - learning to enjoy the
sensations he perceives.4
Due to the illegality of marijuana use throughout the
United States at the time, would-be users had to contend
with powerful forces of social control. It was, Becker
noted, by being a part of a user group that participants
could gain access to supply, keep their use a secret, and
gain access to justifications and rationalizations.
Use of prescription medications has many interesting
contrasts and similarities with the processes Becker
describes in relation to marijuana use. Marijuana and
opiates have the potential to create both euphoric and
dysphoric sensation. Physicians may spend considerable
effort educating patients about the risks and benefits of

5. DOI: 10.1111/papr.12458
© 2016 World Institute of Pain, 1530-7085/16/$15.00
Pain Practice, Volume 16, Issue 7, 2016 791–793
the drug—helping them to perceive the effects and to
make sense of the experience. In this way, the informed
consent process replaces the role of the drug user group
described by Becker. As part of the process of guiding
patients and helping them to learn how to use prescrip-
tion drugs, we might infer that physicians might inad-
vertently facilitate the transition to “abuse.” From an
ethical perspective, two major principles of medical
practice seem at odds: the principals of primum non
nocere, or “do no harm,” and “patient autonomy.” In
respecting one of these principles, the physician violates
the other. How such principles are understood might
influence the way clinicians frame instructions for use,
side effects, and the risk profiles of prescription drugs.
The role that rationalizations and justifications play

6. in supporting deviant behavior was famously high-
lighted by David Matza and Gresham Sykes.5 Insisting
that “juvenile delinquents” do not subscribe to an
oppositional morality, these authors argued that delin-
quency is motivated by exaggerated adherence to widely
held subterranean values, emphasizing excitement and
hedonistic leisure over formal values and work. Matza
and Sykes also highlighted the role that neutralization
techniques play in sustaining deviant behavior by
warding off the guilt associated with such activities.5
These techniques include denial of responsibility, denial
of injury, denial of the victim, condemnation of the
condemners, and appeals to higher loyalties. It follows
that neutralization techniques only need to be applied
when behavior is deviant, and always when it is illegal.
During the initiation of prescription medications, these
techniques are unnecessary, but are likely to be activated
if use progresses beyond the point of medical need.

7. Based on this perspective, one can assert that prescrip-
tion use becomes deviant once the user needs to employ
neutralization techniques: the use of such techniques
signifies an implicit recognition that the behavior falls
outside of what is considered legitimate or acceptable
and is moving toward recreational use or dependency.
Drawing on these insights, clinicians might consider
assessing the use of neutralization techniques to diag-
nose “inappropriate” drug use. Addressing patients’
assumptions and beliefs is already a core part of
psychotherapy in the addictions. Further, understanding
the patient’s value system can help direct the informed
consent discussion to explicitly confront the sensation of
feeling high as part of the side effect profile of these
drugs, especially with regard to opioids.
Jock Young drew attention to the socially constructed
nature of deviance in his book The Drugtakers.6
Adopting a relativist position, Young argued that the

8. same activity might be labeled as simultaneously deviant
and normal depending on whose standards are being
applied. It is, in other words, the context surrounding
the action as well as the larger societal norms that
constructs the definition. This type of subjective assess-
ment of deviancy has direct parallels with the interplay
between physician and patient. It underscores some of
the largest practical difficulties when labeling/diagnos-
ing use, “misuse,” and “abuse” or, in the sociological
rhetoric, deviancy. There is a dynamic context for drug
use: at one moment, it can be to treat pain alone, and
another to enjoy the high or to meet a dependence, while
many times it achieves all three. As with deviancy, the
diagnoses of pain and/or dependency are subjective and
context specific.
Edwin Lemert’s7 distinction between primary and
secondary deviance is pertinent here. Highlighting the
importance of social reaction, Lemert notes that primary

9. deviance is commonplace and managed within a socially
acceptable identity, while secondary deviance is inter-
nalized and becomes part of the core definition of the
self. An example of secondary deviance would be when
somebody who uses drugs comes to define themselves as
an “addict.” Interaction with significant others is an
important influence and may lead to the normalization
or acceptance of the deviation as peripheral to identity
or may stimulate a symbolic reorganization of the self
around the deviant act. The distinction between primary
and secondary deviance parallels exactly the transition
from authorized use of medication to treat pain to
viewing the use of the drug or the addiction as the
pathology in and of itself. Furthermore, Lemert
describes secondary deviance as, “Adjustment to the
overt and covert problems created by the consequent
societal reaction to him,” which corresponds with the
way modern welfare systems give social support to

10. chronic patients due to their disability.8 The chicken and
egg debate about whether welfare support incentivizes/
creates long-term disability remains contentious.9
Harold Finestone showed how addiction is shaped by
the broader social context in his influential ethnographic
study of black heroin users in 1960s in Chicago. “With
little prospect of achieving or identifying with status
positions in larger society”, he argued, “the Cat [heroin
user] is the personal counterpart of an expressive social
movement.”10 According to Finestone, this form of
secondary deviance was an expressive, productive
adaptation to cope with systemic racism, segregation,
and exclusion from the formal economy. The heroin
792 � LEVIN AND SHINER
scene provided the basis of a countercultural identity
built around “cool” and “kicks” as well as the need to
“hustle” (to maintain the lifestyle). For the Cat, the

11. taboo and the desire to put himself beyond the compre-
hension of the “square” were motivating and unifying.10
The development of subcultures around prescription
medications requires an ethnographic study of its own.
“New” deviancy theories developed in opposition to
the prevailing dogma that there was a deviant minority
whose behavior could be explained as a result of intrinsic
pathology or social dysfunction. Modern medical
research tends to emphasize inherent pathology, neuro-
chemical pathways, and social determination in much the
same way as the very earliest deviancy theorist. Applying
“new” deviancy to this modern phenomenon can gener-
ate a novel understanding of the topic. The main
contribution of the “new” deviancy theories was to draw
attention to the counterproductive nature of stigmatizing
and exclusionary forms of social control: far from
eliminating “deviance,” such responses often serve to
entrench it. This does not mean that social control is

12. necessarily a bad thing, however, and we would do well
to heed the distinction Braithwaite draws between
shaming that is stigmatizing and counterproductive and
that which is reintegrative and crime reducing.11 Young
made a similar distinction when he claimed “the subcul-
ture of drugtaking” has “the only viable authority to
control the activity of its members” (p. 221). Rather than
harassing and undermining existing drug subcultures, he
advocated a policy of maintaining such cultures and
encouraging users to adapt their habits by providing
them with what he called “positive propaganda”—
accurate, credible information about the effect of drugs.
Physicians treating patients whose use of prescription
medication is blurring into recreational or dependent use
are well placed to fulfill such a role. These lessons can
guide policy makers seeking to address the larger issues
contribution to this problem.
ACKNOWLEDGEMENTS

13. This article is based on a coursework submitted to the
London School of Economics and Political Science as part
ofthe MSc HealthPolicy,Planningand Financingin2015.
ROLE OF FUNDING SOURCES
The authors received no funding to support this work.
CONTRIBUTORS
David Levin worked under the guidance of Michael
Shiner to develop the major concepts for this work.
There was continued feedback and interaction between
the two authors. David Levin wrote the first draft of the
article, and all authors contributed to and have
approved the final article.
CONFLICTS OF INTEREST
The authors state that they have no conflict of interest.
David Levin, MD, MSc, BESc (Mech)*,†
*Department of Anesthesiology, University of Toronto,
Toronto,
Ontario, Canada
†Health Policy, Planning and Financing, London School of
Hygiene

14. and Tropical Medicine, London School of Economics, London,
U.K.
Michael Shiner, PhD‡
‡Department of Social Policy, London School of Economics,
London,
U.K. E-mail: [email protected]
REFERENCES
1. Sumner C. Deviance. In: McLaughlin E, Muncie J, eds.
The Sage Dictionary of Criminology. London: Sage;
2013:135–136.
2. Matza D. Becoming Deviant. Englewood Cliffs, NJ:
Prentice-Hall; 1969.
3. Becker H. Outsiders: Studies in the Sociology of
Deviance. New York: The Free Press; 1963.
4. Becker HS. Becoming a marihuana user. Am J Sociol.
1953;59:235–242.
5. Matza D, Sykes GM. Juvenile delinquency and sub-
terranean values. Am Sociol Rev. 1961;26:712–719.
6. Young J. The Drugtakers: The Social Meaning of Drug
Use. London: MacGibbon and Kee; 1971.
7. Lemert EM. Social Pathology: A Systematic Approach

15. to the Theory of Sociopathic Behavior. 1st ed. New York:
McGraw-Hill; 1951.
8. Fitzcharles MA, Ste-Marie PA, Gamsa A, Ware MA,
Shir Y. Opioid use, misuse, and abuse in patients labeled as
fibromyalgia. Am J Med. 2011;124:955–960.
9. Clayton S, Bambra C, Gosling R, Povall S, Misso K,
Whitehead M. Assembling the evidence jigsaw: insights from a
systematic review of UK studies of individual-focused return to
work initiatives for disabled and long-term ill people. BMC
Public Health. 2011;11:170.
10. Finestone H. Cats, Kicks, and Color. Indianapolis, IN:
Bobbs-Merrill, College Division; 1957.
11. Braithwaite J. Crime, Shame and Reintegration. Cam-
bridge, UK: Cambridge University Press; 1989.
Levin and Shiner � 793
Copyright of Pain Practice is the property of Wiley-Blackwell
and its content may not be
copied or emailed to multiple sites or posted to a listserv
without the copyright holder's

16. express written permission. However, users may print,
download, or email articles for
individual use.
http://informahealthcare.com/dep
ISSN: 0968-7637 (print), 1465-3370 (electronic)
Drugs Educ Prev Pol, 2016; 23(6): 462–470
! 2016 Informa UK Limited, trading as Taylor & Francis Group.
DOI: 10.3109/09687637.2016.1167168
Feeding on fear: Edible marijuana and disproportionality in US
media
Michael H. Eversman
Department of Social Work, Rutgers University, Newark, NJ,
USA
Abstract
Aims: US marijuana policy is in flux, as some states have
legalized it for medical and recreational
use despite long-standing federal prohibition. Edible marijuana
(edibles) allows consumption
without smoke inhalation, yet concern surrounding dosing and
its effect is concerning. Cohen’s
moral panic framework informs how and why some societal
problems are disproportionately
constructed and stigmatized, while other important problems are
tolerated or ignored. This
qualitative study explores the construction of disproportionality
within textual US news

17. coverage of legal edibles. Methods: A three-year sample of 349
articles was retrieved and
analysed. Analysis was informed a priori by the moral panic
criteria of disproportionality, and
data reflect headlines and story text purporting to gauge
problems surrounding edibles.
Findings: Using reports of emergencies, poison centre calls and
anecdotes, disproportionality
surrounding edibles was constructed using indefinite numeric
adjectives while avoiding or
downplaying totals and presenting statistics with little context
to gauge danger relative to
other substances. Edibles were also depicted as causing
fatalities, occluding other contributing
causes. Conclusions: Deviance was ascribed to marijuana by
exaggerating its toxicity when
eaten and downplaying important contextual factors.
Disproportionality surrounding marijuana
supports bases vested in maintaining criminalization policies
while drawing attention from
other relevant social problems.
Keywords
Cannabis, drugs, policy, qualitative research,
substance abuse
History
Received 13 January 2016
Revised 6 March 2016
Accepted 14 March 2016
Published online 27 April 2016
Introduction

18. Evidence of worldwide marijuana (cannabis sativa/indica)
smoking and eating for medicinal purposes, religious rituals
and consciousness transcending has existed for millennia, yet
its benefits remain challenged and its dangers are often
exaggerated (Bostwick, 2012; Cohen, 2009). From ‘‘reefer
madness’’ during the 1930s to its inclusion in the War on
Drugs, marijuana in contemporary US society has long been
immersed in sociopolitical conflict (Berkey, 2011; Musto,
1973; Sloman, 1979/1998). Despite long-standing federal
prohibition, individual states began legalizing medical mari -
juana in 1996 and recreational use in 2012. Presently, 23
states and the District of Columbia have legalized medical
marijuana, while four states and D.C. have legalized recre-
ational use.
Edible marijuana products – long a mainstay for medical
users who prefer not to smoke – have proven popular in the
recreational market. In edible form, marijuana can be
consumed without smoke inhalation, considered perhaps its

19. biggest health risk (Lamarine, 2012). Yet concern surround-
ing proper dosage and the unique high of edible marijuana has
garnered attention, as it has been implicated in a handful of
serious incidents, emboldening those opposed to expanded
legalization.
Cohen’s (1972/2002) social constructivist framework of
moral panic helps inform how and why some social problems
are constructed such that they serve particular interests
(political, business, etc.). During moral panic, the true threat
posed by certain social problems is exaggerated, typically via
sensationalized media coverage. From this construction of the
problem, it becomes increasingly stigmatized and those
associated with it politically weakened. In the wake of the
disproportionate danger ascribed to the problem, the occur-
rence of other, often more serious social problems is accepted
or ignored. Given the contentious sociopolitical history
surrounding marijuana in the United States and the presence
of vested interests opposed to liberalized marijuana policies,

20. examining its depiction is relevant. Informed by Cohen’s
moral panic framework, this study examines textual US news
coverage of edible marijuana, specifically the qualitative
construction of disproportionality within media depiction
of it.
Edible marijuana: Background
The therapeutic value of marijuana was first formally
acknowledged in the United States by its inclusion in the
1850 United States Pharmacopeia drug reference manual
(through 1942 when it was removed), and it was long
recommended for various disorders (Sloman, 1979/1998).
Edible marijuana products (‘‘edibles’’) originated within the
medical marijuana community and have long been a staple
choice for patients who prefer to consume it without smoking
Correspondence: Michael H. Eversman, Department of Social
Work, 418
Hill Hall, Rutgers University, Newark, NJ 07102, USA. E-mail:
[email protected]
(Chapkis & Webb, 2005). However, since the 1970 federal

21. Controlled Substances Act deemed marijuana a Schedule I
substance with high abuse potential and no medicinal value,
large-scale study of therapeutic efficacy was effectively
stifled (Bostwick, 2012). Yet throughout the 1970s, some
states began decriminalizing marijuana and supporting
research, and with the AIDS crisis in the 1980s, San
Francisco enacted medical marijuana, setting the stage for
California’s passage of the Compassionate Use Act in 1996
(Berkey, 2011).
In addition to voluminous anecdotal patient reports
supporting the therapeutic benefits of marijuana (Chapkis
& Webb, 2005), clinical research supports its indication for
controlling nausea and vomiting and promoting weight gain,
and treating muscular neuropathy and spasticity from
conditions such as multiple sclerosis (Grant, Atkinson,
Gouaux, & Wilsey, 2012), though others suggest the benefits
may be overstated (Whiting et al., 2015). While early-age
marijuana use is concerning, lifetime dependence rates

22. (9–10%) compare favourably to those of other recreational
substances such as nicotine (32%) and alcohol (15%)
(Bostwick, 2012; Cohen, 2009). Marijuana dependence is
associated with psychosis and the onset of schizophrenia,
and while causality is unclear, heavy use is associated with
cognitive impairment, depression and severe psychiatric
illness. Claims of a ‘‘gateway theory’’ – that marijuana use
leads to the use of ‘‘harder’’ drugs – are largely unsubstan-
tiated (Bostwick, 2012; Cohen, 2009).
Legalized recreational marijuana laws have expanded the
number of marijuana users who can consume edible mari-
juana, and while the exact number of users is unknown, sales
information suggests that it is popular. In 2014, Colorado sold
almost five million units of medical and recreational edible
marijuana products, and approximately, 45% of marijuana
sales in that state involve edible marijuana products (Centers
for Disease Control, 2015; Colorado Department of Revenue,
2015). However, due largely to unclear dosing guidelines and

23. differences between eating marijuana and smoking it, wider
consumption of edible marijuana has fostered concerns about
increased emergency visits for intoxication and paediatric
exposures and fears of increased youth initiation and use
(MacCoun & Mello, 2015).
Moral panic: Background
Originating with British sociologist Stanley Cohen (1972/
2002), moral panic is a social constructionist framework for
understanding social problems and deviance, notably how,
why and by whom particular phenomena are defined as
deviant. According to moral panic, the dangers of some social
problems are depicted disproportionately to their ‘‘true’’
extent and ultimately draw attention from other equally or
more dangerous problems (Cohen, 1972/2002; Goode & Ben-
Yehuda, 1994/2009). The existence of moral panic requires
three elements: an enemy, a victim and societal consensus to
act. The occurrence of the problem becomes linked to an
enemy or ‘‘folk devil’’ and is seen to threaten victims –

24. typically children or ‘‘middle class’’ society. From the panic
generated by the problem now linked to the enemy, consensus
emerges that the threat poses wide scale societal danger
unless action is taken – usually legislation (Cohen, 1972/
2002).
A key application of moral panic lies in scrutinizing the
information – typically media coverage – provided about a
particular problem. Functioning as information processors,
media outlets frame and define (construct) social problems as
deviant and worthy of attention (Cohen, 1972/2002). The
elements of moral panic can be found within media coverage
and include concern about the threat of the problem, hostility
targeted at those (enemies) deemed responsible for the
problem, consensus that something be done to address the
problem, disproportionality about the scope of the threat
posed by the problem and volatility, such that panic about the
problem often vanishes suddenly (Cohen, 1972/2002). As
detailed below, this study emphasizes the element of dispro-

25. portionality, considered the core aspect of moral panic
(Goode & Ben-Yehuda, 1994/2009).
Despite its utility, criticisms of moral panic should also
be noted. For example, some point to its underlying
constructionist orientation and remind us that subjective
formulations of a particular social problem do not mitigate
its objective harms (Goode, 1990). Others argue that
changes in media consumption and production render
moral panics ‘‘everyday’’ occurrences and that the frame-
work has long been removed from its original formulation
and its application complicated by the multiplicity of
societal interests and mechanisms of social control
(McRobbie & Thornton, 1995). More recent critics note
that the meaning of moral panic is and has always been open
to interpretation. For example, a core feature of moral panic
is the irrationality of disproportionate reaction to particular
social deviance, yet Stanley Cohen himself has suggested the
idea of ‘‘good’’ moral panics, whereby societal reaction to

26. phenomena can be legitimate and proportionate (David,
Rohloff, Petley. & Hughes, 2011).
Drug moral panics in textual US news media
Several analyses of textual US news media have examined
drug moral panics. Most examine methamphetamine
(Armstrong, 2007; Jenkins, 1994; Linnemann, 2009; Omori,
2013; Weidner, 2009), crack cocaine (Chiricos, 1996;
Hartman & Golub, 1999; Reinarman & Levine, 1997), or
both drugs (Cobbina, 2008), while others have considered
heroin (Agar & Reisinger, 2000; Denham, 2008), LSD
(Goode, 2008), MDMA (Baldwin, Miller, Stogner, & Hach,
2012), and marijuana (Goode & Ben-Yehuda, 1994/2009).
Methamphetamine moral panics
Moral panics surrounding methamphetamine (meth) are
considered to have occurred episodically in the United
States since the late 1980s. Jenkins (1994) discussed a brief
panic in 1989 surrounding a form of meth known as ‘‘ice’’.
Though largely confined to Hawaii (called the ‘‘Hawaii

27. epidemic’’), media coverage and Congressional hearings
fanned concerns and drew national attention but ebbed after
just a few months. Armstrong (2007) declared a meth panic
occurred in the Midwestern United States in the early 2000s,
documenting disproportionality by finding inconsistency
between the volume of meth-related articles and government
DOI: 10.3109/09687637.2016.1167168 Feeding on fear: Edible
Marijuana 463
epidemiologic data. Noting also that concerns surrounding
the environmental hazards of meth labs are exaggerated
relative to those surrounding livestock industry waste, for
example, the author concluded that panic surrounding meth
deflects attention from the widespread poverty in many rural
US communities. Similarly, Weidner (2009) studied a meth
panic in the Midwest that occurred between 1997 and 2005
by comparing print media portrayal of meth to objective
indicators including usage rates, treatment admissions and

28. laboratory seizures, finding that the frequency of related
stories was excessive, while also describing claims about
meth’s addictiveness as exaggerated. Studying a similar
period (1995–2007), Linnemann (2009) noted how identify-
ing moral panic also reveals inequalities spanning race, class
and gender and found differences in media portrayals of
gender such that women who use meth were criticized as
maternal failures, while such failures among men who use
meth were generally ignored, and they were instead
portrayed as rational actors trying to earn a living. Omori
(2013) considered the years 2000–2007 to examine the
influence of print news on state antimeth legislation,
suggesting that media coverage acts in concert with law
enforcement to exaggerate the meth problem and drive
policy formation.
Crack cocaine moral panics
It is widely accepted that moral panic surrounding crack
cocaine occurred in the United States. during the 1980s and

29. 1990s. Chiricos (1996) noted that news coverage of crack
increased during the 1980s despite data indicating its use was
decreasing. This increased attention fuelled panic and largely
obscured the economic decline and rising crime in many US
urban areas, while simultaneously justifying harsh criminal
penalties and expansion of the prison industrial complex.
Similarly, Reinarman and Levine (1997) surmised that fear
and panic generated by sensationalized media accounts in the
1980s and early 1990s misrepresented evidence about crack,
functioning to legitimize the War on Drugs and support a
conservative political agenda. Considering whether inaccurate
news claims about crack would ‘‘self-correct’’, Hartman and
Golub (1999) analysed newspaper and magazine content
spanning 1985–1995, finding that numerous exaggerations
and falsehoods – widely repeated – became accepted as truth.
Cobbina (2008) considered depictions of race and class
surrounding crack and meth, finding that stories about crack
(portrayed as a black drug) were twice as likely to emphasize

30. violence and call for harsh penalties than were stories on
methamphetamine (portrayed as a white drug), which more
often emphasized public health problems.
Other drug moral panics
Agar and Reisinger (2000) examined newspaper coverage of
heroin between 1992 and 1998 and found that linking white
suburban youth with heroin use induced moral panic. Denham
(2008) explored whether coverage of heroin during the 1990s
mentioned a cultural icon (model, actor or musician)
associated with heroin use and/or referenced a motion picture
addressing heroin and found that contrary to media coverage,
survey data did not indicate a significant increase in heroin
use. Citing the link of psychedelic drugs to political
subversion during the 1960s, Goode (2008) claims that
stigmatization of LSD users made outrageous claims of its
danger believable. In a case study of MDMA and club drug
use, Baldwin et al., (2012) concluded that passage of an
antirave ordinance in a Florida city in the late 1990s consisted

31. of an ‘‘interest group’’ moral panic (Goode & Ben-Yehuda,
1994/2009) driven by influential community stakeholders.
Moral panic surrounding marijuana occurred in the United
States during the 1930s, when sensationalized media accounts
and the popular film ‘‘Reefer Madness’’ heightened fears and
deviance (Goode & Ben-Yehuda, 1994/2009). By enflaming
fears that marijuana use induced racial minorities to commit
crimes, state and local governments enacted marijuana
prohibition, culminating in the federal Marihuana Tax Act
of 1937 (Musto, 1973). Marijuana’s current sociopolitical
dynamic (and stigma) is in flux, as states increasingly legalize
recreational use alongside medical use while federal prohib-
ition remains. Informed by Cohen’s moral panic framework,
this qualitative study explores US news coverage of edibles.
Specifically, it considers the presence and construction of
disproportionality in a sample of textual news story headlines
and articles since passage of recreational legalization (2012–
2015).

32. Methods
Theoretical application of moral panic
While determining the occurrence of full moral panic (i.e. the
presence of all elements) with edibles is beyond the scope of
this analysis, three important aspects are utilized. First, the
presence and qualitative construction of disproportionality in
textual news media headlines and articles are examined.
While moral panic consists of five elements (concern,
hostility, consensus, disproportionality and volatility) dispro-
portionality is central such that ‘‘the concept of the moral
panic rests on disproportionality (Goode & Ben-Yehuda,
1994/2009, p. 38, italics in original). A key indicator of its
presence lies with distorted figures and statistical claims cited
to describe the scope of, and risk posed by, the problem
(Cohen, 1972/2002; Goode & Ben-Yehuda, 1994/2009).
Second, the paper explores how disproportionality can be
constructed to establish (or re-establish) social deviance
surrounding edibles and marijuana per se. Moral panic theory

33. regards the construction of deviance as transactional,
facilitated by information processors – media outlets with
particular political and commercial interests (Cohen, 1972/
2002). Lastly, moral panic holds that disproportionate atten-
tion given to some social problems draws attention away from
other, often more serious social problems (Goode & Ben-
Yehuda (1994/2009). Thus, the paper considers how dispro-
portionately constructed around edible marijuana usage serves
to occlude other social problems present (but downplayed) in
media coverage of edibles.
Sampling
The sample consisted of 349 unique (i.e. occurring once)
textual US news media articles published between
1 November 2012 (approximating when the first states
464 M. H. Eversman Drugs Educ Prev Pol, 2016; 23(6): 462–
470
legalized recreational use) and 31 October 2015. Articles
were retrieved from two comprehensive full-text newspaper

34. databases – Access World News and Factiva which yielded
articles from newspapers across the United States – and
Google News, a computer-generated aggregator of worldwide
news stories which yielded articles from internet media, such
as magazines, broadcast outlet web pages, web-based news-
paper editions and blogs. Duplicate articles were excluded –
notably syndicated pieces published in multiple sources – and
only the article from the originating media source is included.
All articles were published in US media or the US version of
an international source.
Articles were retrieved based on the occurrence of
keywords appearing anywhere in its text. Keywords included
‘‘edible marijuana’’, ‘‘marijuana edibles’’, ‘‘marijuana
infused’’, ‘‘pot infused’’, ‘‘pot laced’’, ‘‘pot candy’’ and
‘‘marijuana candy’’. Included articles focused on edible
products legally available via medical and/or retail outlets
in the United States. While all retrieved articles referenced
legalized edibles, the sample includes only articles in which it

35. was a main focus such that an iteration of it appears in the
headline and/or is discussed in the lead paragraph or more
than half of the article text. News stories, columns, opinions
(editorials, op-eds) and letters to the editor (in which the
writer indicated an institutional or professional credential)
were included. Excluded articles were those addressing
‘‘homemade’’ (i.e. illicit) edibles and/or edibles not otherwise
specified as legal, business news and articles addressing
cooking and/or dining with marijuana. Blog content was
included only from blogs affiliated with a mainstream media
outlet (typically online newspaper editions). The author
conducted all database retrieval.
Sample description
Figure 1 shows the bimonthly distribution of all 349 articles
spanning 1 November 2012 through 31 October 2015.
The sampled articles were selected from 191 unique media
outlets throughout the United States, most of which (177)
yielded less than five articles. Table 1 lists the fourteen media

36. outlets from which five or more articles were sampled.
Almost two-thirds (62%) of the 349 sampled articles
consisted of news stories (n¼216), while columns comprised
nearly one-third (31%, n¼109). Editorials (n¼21) and
letters to the editor (n¼3) occurred less frequently.
Regional distribution was led by articles from national
media outlets (28%, n¼100) and outlets in Colorado (26%,
n¼92). Articles from outlets in other legalized states
[Washington (n¼23), Oregon (n¼20), California (n¼14)]
as well the District of Columbia (n¼11) were also well
represented.
Data analysis
Retrieved articles were saved as word processing documents
and entered as qualitative data analysis software files (Atlas
TI, v. 7.5, Corvallis, OR). First-level data coding was then
initiated, during which all articles were examined and
individual recording units identified and coded a priori.
Recording units are textual passages that are discretely
determined by ‘‘a definable boundary, or symbolic meaning’’
(Riffe, Lacy, & Fico, 1998). As such, a recording unit is best
understood as the most basic expression of meaning contained

37. within the data, coded to reflect this meaning. All retrieved
textual content (headlines and articles) was examined using
the analytic software which allowed the author to highlight
and tag (code) identified passages (i.e. recording units)
reflecting an element of moral panic (concern, consensus,
hostility, disproportionality and/or volatility). Familiarity with
these elements allowed the author to identify content a priori.
The study reports only recording units coded to reflect
disproportionality, which was considered to be present in any
textual references ‘‘cited to measure the scope of the
problem’’ (Goode & Ben-Yehuda, 1994/2009), including
numeric claims (i.e. statistics) and non-numeric claims (i.e.
claims of ‘‘an increase’’ in edible related incidents).
Second-level analysis involved identifying, arranging and
categorizing first-level recording units to reflect thematic
content (Coleman & Unrau, 2005). Specifically, recording
units were grouped to reflect conceptually larger notions of
disproportionality. For example, one grouping of recording

38. units reflected that disproportionality was constructed by
citing statistical claims out of context, and thus, they were
categorized as ‘‘disproportionality, statistics out of context’’.
Recording units (in the form of story headlines and text)
deemed by the author to best convey larger thematic ideas are
presented below. Further, the data are displayed chronologic-
ally (spanning the sample period). Article headlines are
presented in capital letters, while article text is quoted and/or
block formatted throughout.
Results
Disproportionality constructed around the danger of edibles
used indefinite numeric adjectives to describe the occurrence
of incidents involving their use while avoiding or downplay-
ing low total numbers. Statistics were reported with little
context to gauge the severity of similar incidents (i.e.
emergency exposures) with other substances, while stories
depicted edibles as the central or sole cause of involved
incidents. By linking the benign to the serious, the danger of

39. edibles was highlighted, while other causes of reported
incidents were typically ignored or downplayed. Incidents
reported included ER visits, poison control centre calls and
fatal and non-fatal anecdotes.
Indicators of disproportionality first emerged in April 2013
with news of a Children’s Hospital Colorado study which
determined that ER admissions for marijuana exposure in
children under age 12 went from zero before 2009 to 14 from
2009–2011 when medical marijuana was enacted in Colorado.
Headlines and articles raised concerns about child access to
edibles but provided no context to properly gauge the extent
relative to accessing other drugs:
COLO. KIDS GET INTO POT CANDY, PROMPT
CALLS FOR CHILDPROOF PACKAGING. UPI.com
(United Press) – 2 April 2013
Children’s Hospital Colorado saw 14 children come into
the emergency room after ingesting cannabis candy in the
two years following medical marijuana legalization in the

40. state, reports the Denver Post. Studies of ER charts by
DOI: 10.3109/09687637.2016.1167168 Feeding on fear: Edible
Marijuana 465
Colorado doctors show looser pot laws leading to child-
hood ingestion, often from mistakenly eating tempting
‘‘edibles’’ like marijuana-infused gummy worms or
brownies. . .
Related articles quoted a doctor affiliated with the study
as having seen ‘‘a dramatic increase’’ in paediatric marijuana
exposures due ‘‘mostly’’ to edibles and treating ‘‘one to two
kids a month’’ without reporting rates of paediatric exposures
to other substances. The doctor is further quoted to say that
marijuana related calls to the state poison control centre
‘‘have doubled since 2009’’ but no specific numbers are
provided. The above study garnered more attention after its
formal publication in May, 2013, and headlines again used
indefinite numeric claims, while multiple articles similarly
cited ‘‘a significant increase’’ in youth marijuana exposure:

41. CASES OF CHILDREN ACCIDENTALLY INGESTING
POT ON RISE, STUDY SAYS. Los Angeles Times – 28
May 2013
SURGE IN CHILDREN ACCIDENTALLY EATING
MARIJUANA-LACED FOODS. Targeted News Service
– 29 May 2013
STUDY SUGGESTS LINK BETWEEN EDIBLE POT
AND OVERDOSE AMONG KIDS. KPLU.org – 28 May
2013
A new study conducted in Colorado shows an increase in
kids seen at Emergency Departments ever since medical
marijuana laws were liberalized in 2009. In more than
three years prior to that, zero kids went to the ER for
marijuana overdose. But in the two years after the changes,
14 kids were confirmed to have overdosed on marijuana.
Half the poisonings in Colorado were linked to edibles.
Despite articles referencing the ‘‘admittedly small’’ total
number of paediatric exposure emergencies, they were

42. nonetheless recast as ‘‘small but growing’’ or otherwise
‘‘significant’’:
WHAT HAPPENS WHEN CHILDREN IN COLORADO
FIND THEIR GRANDPARENTS’ POT BROWNIES.
National Journal.com – 28 May 2013
Studying the patient records of one children’s hospital in
the state, the incidence of marijuana consumption in
children younger than 12 hopped up from zero cases
(between 2005 and 2009) to 14 cases (between 2009
Figure 1. Article frequency and key events.
0
10
20
30
40
50
60
Ar�cle Frequency & Key Events

43. Frequency
4/2/14: 1st fatality
4/17/14: 2nd fatality
6/2/14: NY Times column
4/21/15: 3rd fatality
Table 1. Media outlets (n¼14) yielding five or more articles.
Outlet name
N ¼ Source
articles/total
sample
Percentage
total article
sample
Denver Post 36 10.4
Denver CBS Local.com 11 3.2
Associated Press (National) 10 2.9
Associated Press (State Wire, Colorado) 8 2.3
Washington Post 8 2.3
USA Today 7 2
Oregon Live.com 7 2
New York Times 7 2
Forbes 6 1.7
Los Angeles Times 6 1.7
Reuters 6 1.7
CBS News.com 5 1.4
Huffington Post 5 1.4
Seattle Times 5 1.4

44. 127/349 33.5
Four or fewer articles were sampled from each of 177 other
media
outlets.
466 M. H. Eversman Drugs Educ Prev Pol, 2016; 23(6): 462–
470
and 2011). The number is small, but the authors find it to
be statistically significant.
By April 2014 media coverage intensified after news
broke of the March 11 death of a nineteen year old college
student visiting Colorado who leapt from his hotel window.
Use of edibles was involved, and the story received national
coverage. Headlines and articles portrayed edibles as the
direct cause of death:
STUDENT DIES IN COLORADO FALL AFTER
EATING POT COOKIE.
NBC News.com – 2 April 2014
TEEN JUMPS TO HIS DEATH AFTER EATING

45. MARIJUANA-LACED COOKIE FROM COLORADO
SHOP. Hartford Examiner (CT) – 2 April 2014
COLLEGE STUDENT FALLS TO DEATH AFTER
EATING POT COOKIE IN DENVER. Atlanta Examiner
– 3 April 2014
IN THE FIRST FATALITY LINKED TO LEGAL
MARIJUANA, THE MAN WHO JUMPED ATE POT
COOKIES: EDIBLES CITED IN DEATH REPORT.
The Denver Post – 3 April 2014
(Coroner’s office spokeswoman) said (the decedent) had
no known physical or mental-health issues, and toxicology
tests for other drugs or alcohol came back negative. ‘‘We
have no history of any other issues until he eats a marijuana
cookie and becomes erratic and this happens’’, she said.
‘‘It’s the one thing we have that’s significant’’.
This incident drew sustained attention and was com-
monly cited in subsequent coverage of edibles and marijuana
in general, highlighting the dangers and risks of legalization:

46. ERs SEE MORE PATIENTS AS LAWMAKERS
REVIEW EDIBLES’ POTENCY:
DEATH PUTS FOCUS ON RISK. The Denver Post – 4
April 2014
DEATH RAISES NEW ALARMS OVER POT: EDIBLE
FORM OF DRUG A CONCERN IN COLORADO. The
Baltimore Sun – 9 April 2014
The death, involving a victim with no history of mental
problems or suicidal tendencies, was linked to ‘‘marijuana
intoxication ‘‘. The case has become a grim exhibit in a
growing case file as Colorado health officials wonder
whether, in the rapid rollout of legalized marijuana,
adequate attention was paid to potential health risks of
its use, especially in the little-scrutinized area of edible
marijuana.
Soon after (14 April 2014) news broke of a second
Colorado death involving edibles whereby a husband shot
his wife in their home. Articles noted that the exact role of

47. edibles in the murder was unclear. The murder was
committed with a firearm, and early reports indicated the
husband may have taken other medications and that the
couple was enduring increasing marital stressors, yet
headlines again depicted edibles as the direct, singular
cause of the shooting:
DENVER POLICE WANT TO KNOW WHETHER
MURDER SUSPECT CONSUMED POT
Los Angeles Times – 15 April 2014
POLICE: MAN ATE POT CANDY BEFORE SHOOTING
WIFE. Honolulu Star-Advertiser – 17 April 2014
A POLICE AFFIDAVIT SAYS THE HUSBAND
SUSPECTED OF KILLING HIS WIFE HAD,
ACCORDING TO HER 911 CALL, JUST EATEN POT
CANDY: MAN ‘‘TOTALLY HALLUCINATING’’. The
Denver Post – 18 April 2014
MAN ATE POT CANDY BEFORE SHOOTING HIS
WIFE DEAD: COPS. Huffington Post.com – 18 April

48. 2014
DENVER MAN ACCUSED OF KILLING WIFE MAY
HAVE EATEN POT CANDY. NBC News.com – 18 April
2014
POLICE: MAN EATS POT CANDY, KILLS. The Key
West Citizen (FL) - April 18, 2014
DAD TURNS VIOLENT AFTER EATING MARIJUANA
CANDY FROM (STORE NAME). Atlanta Examiner – 18
April 2014
Media accounts soon after emphasized the ‘‘ease of
access’’ to edibles purportedly underlying paediatric expos -
ures, and headlines again described an increase using
indefinite terms:
STONED TODDLERS RAISE RED FLAGS. Hawaii
Tribune-Herald (Hilo, HI) – 25 April 2014
CASES OF KIDS EATING POT PRODUCTS RISING.
The Denver Post – 6 May 2014
THIS YEAR’S SURGE IN ACCIDENTAL CASES

49. INVOLVING KIDS IS ON PACE TO MORE THAN
DOUBLE THE TOTAL FOR 2013.: CHILDREN’S
HOSPITAL SEEING RISE IN MARIJUANA
INGESTION. The Denver Post – 22 May 2014
By using anecdotal accounts, articles also generated
disproportionality surrounding child access through interviews
in which, for example, paediatric exposures are described by
medical personnel as ‘‘rising at an alarming rate.’’ First-hand
accounts from doctors – who were variously described as
‘‘sounding an alarm’’ – provided salient details of danger:
DOCTORS WORRY ABOUT AN INCREASE IN KIDS
CONSUMING CANNABIS. Fox21news.com – 27 May
2014
‘‘There are long-term effects, but the biggest thing is
respiratory depression. Their (children) vitals need to be
constantly monitored, and that’s the biggest one because if
you can’t breathe you can’t live. It’s a scary thing
sometimes,’’ Dr. (name) said. In most cases reported,

50. children ate edible products. ‘‘The lollipops, the brownies,
DOI: 10.3109/09687637.2016.1167168 Feeding on fear: Edible
Marijuana 467
the candies, everything is just way too enticing to kids.
And who’s gonna stop at one brownie?’’
In June, 2014 a high profile first person account of an
unpleasant experience with edibles was published in the New
York Times. Although the amount consumed was unspecified
(‘‘I nibbled off the end. . .and then nibbled some more’’
having had ‘‘a bite or two’’), the column was frequently
recounted in subsequent media coverage, notably the authors
claim of experiencing ‘‘a hallucinatory state for eight hours’’
and having become ‘‘convinced I had died’’. Headlines about
the incident were sensationalized:
COLUMNIST SUFFERS POT PARANOIA AS DOPERS
ADVANCE. Accuracy in Media – 4 June 2014
STAR NEWSPAPER COLUMNIST VISITS
COLORADO, TRIES MARIJUANA

51. EDIBLE, FREAKS OUT. The Gazette: Blogs (CO) – 4
June 2014
AND THEY CALL THIS A RECREATIONAL DRUG?
The Deseret News (UT) – 9 June 2014
Called the latest in ‘‘a string of incidents’’, the danger
of edibles were further exaggerated by linking this non-fatal
anecdote to the two April deaths:
POT EDIBLES COULD CAUSE MORE THAN A HIGH.
The News Tribune (WA), 23 June 2014
Edibles can even be a problem for adults – as illustrated in a
recent article by New York Times columnist (name). . .who
took ill after ingesting a small amount of pot-infused candy.
In March, a 19-year-old college student who had eaten a pot
cookie died after jumping off a Denver hotel balcony. And
the shooting death of a Denver woman has been linked to
her husband’s ingestion of pot-infused candy.
In March, 2015 a third death was linked to edible
consumption, in which a young man shot himself while visiting
Colorado on family vacation. Dubbed a ‘‘MARIJUANA

52. SUICIDE’’ by at least one headline, others again suggested
edibles were the singular, direct cause of death:
OKLAHOMA MAN SHOOTS SELF AFTER EATING
POT CANDIES. The Denver Post – 27 March 2015
MAN COMMITS SUICIDE AFTER CONSUMING POT-
INFUSED CANDIES. Las Vegas Review-Journal – 27
March 2015
DID POT GUMMY BEARS KILL THIS MAN?
Associated Press State Wire: (CO) – 27 March 2015
Referred to as ‘‘The latest of a handful of deaths’’ and
reflecting ‘‘The growing number of deaths’’ linked to edibles,
headlines and articles emphasized an unchallenged attribution
by grieving family members that the death was caused by
marijuana ingestion, ignoring other factors such as firearm
availability:
MARIJUANA EDIBLES BLAMED FOR KEYSTONE
DEATH. Denver.CBSLocal.com – 25 March 2015
The family of a. . .man who shot himself Saturday night-

53. . . .is blaming his suicide on his ingestion of edible
marijuana candies. ‘‘It was completely a reaction to the
drugs,’’ (decedent’s mother) said about her son’s
suicide. . .(The coroner spokeswoman) says the preliminary
cause of death is a self-inflicted gunshot wound. As for the
impact of the marijuana edibles, she said, ‘‘That’s what
we’ve heard consistently.’’
In June 2015, findings of an academic study (Onders,
Casavant, Spiller, Chounthirath, & Smith, 2015) reported on
rates of paediatric (aged six and younger) marijuana expos-
ures in the United States. In their findings, the study authors
noted that the low rate (5.9 per every 1,000,000) indicates
these exposures are ‘‘rare’’ and that between 2006 and 2013
the total numbers increased from roughly 100 to 250, yet
headlines and articles reported these numbers in percentages:
MARIJUANA LEGALIZATION MEANS MORE
CHILDREN ACCIDENTLY CONSUMING POT
PRODUCTS, EXPOSURE ROSE 148% SINCE 2006.
IBTimes.com – 8 June 2015

54. AS POT BECOMES LEGAL, MORE KIDS EAT IT.
Cincinnati.com – 14 June 2015
A study released this week found toddlers and infants are
being exposed to marijuana nearly 147 percent more often
than in 2006, prompting researchers to recommend states
take swift action to address child safety when marijuana is
legalized. Cincinnati.com – 14 June 2015
Similarly, headlines in July 2015 and beyond proclaim
that Poison Control calls involving edibles have increased by
noting an ‘‘ALARMING INCREASE’’ or that
‘‘REPORTS. . .SPIKE’’, while articles downplayed the low
overall numbers and instead emphasized that they represent
‘‘An upward trend’’, ‘‘An increasing number’’, and ‘‘An
emerging health risk’’. The headline and text from a story
reporting on ‘‘14 potential marijuana poisonings’’ among
youth 19 and younger:
MARIJUANA ‘POISONING’ INCIDENTS ON THE
RISE – CHOCOLATES, CANDIES APPEAL TO
CHILDREN. The Spokesman-Review (Spokane, WA) –

55. 28 July 2015
Reports of kids eating marijuana-infused cookies and
candies are on the rise statewide and in Spokane County,
where a 4-year-old spent the night in intensive care after
eating a product belonging to a parent. The number of ‘‘pot
poisonings’’ is relatively small, but the increase is
troubling to public health officials.
Similarly, a headline to a story which reported the total
number of marijuana calls (of which an unspecified ‘‘many’’
are attributed to edibles) increased from 54 in 2009 to 136 in
2013:
POISON CONTROL CALLS UP AFTER
LEGALIZATION OF POT.
Reuters.com – 16 October 2015
468 M. H. Eversman Drugs Educ Prev Pol, 2016; 23(6): 462–
470
Discussion

56. While considering the presence of all moral panic elements is
beyond the scope of this analysis, there is evidence of media
generated disproportionality and other key features of moral
panic surrounding edible marijuana. Although these media
accounts can be said to be objectively true, presented as they
are – without context to gauge their frequency and severity
relative to similar incidents with other legal substances –
distorts the danger posed by edible marijuana and increases
social deviance surrounding its very nature. Indeed, unchal -
lenged suggestions that edible marijuana consumption leads
directly to death and/or violent behaviour are reminiscent of
‘‘Reefer Madness’’ era hype. Attributing the cause of such
incidents solely to the intrinsic properties of edible marijuana
without focusing on dosage, the possibility of irresponsible
use, and the presence of other causative factors serves to
increase stigma and reinforce stereotypes of the danger of
marijuana. Such a climate becomes more favourable to
marijuana control policies that maintain criminalization or

57. otherwise hinder liberalization.
Depicting the two April 2014 deaths as due solely to edible
consumption established what Agar and Reisinger (2000)
called a topic framework for subsequent coverage, whereby
the danger of edibles becomes taken for granted by the
audience. From this comes what Cohen (1972/2002) called
‘‘sensitization’’, a cueing process that occurs during moral
panic that ‘‘. . .has the effect of increasing the awareness of
items of a similar nature which he (the reader) might
otherwise have ignored’’ (p.59), or as Goode and Ben-
Yehuda (1994/2009) describe it ‘‘. . .the process whereby
harm, wrongness, or deviance is attributed to the. . .phenom-
enon that is routinely ignored when the same consequences
are caused by or attributed to more conventional conditions
(p. 156, emphasis added). For example in the U.S. there are
over 30,000 annual ER visits (and many hospitalizations) for
children aged six and younger due to pharmaceutical medi -
cation exposure and over half a million calls to Poison Control
hotlines for children aged 19 and younger involving
prescribed and over the counter pharmaceuticals (Ferguson,

58. Osterthaler, Kaminski, & Green, 2015; Lovegrove et al.,
2014). Further, there are over 2000 annual alcohol poisoning
deaths – more than 40 of which involve youth aged 15–20
(Kanny et al., 2015). By comparison, the number of fatalities
attributed to marijuana in this sample pales. If the same media
intensity were given to alcohol fatalities, such stories would
exceed five a day.
Incidents of suicide, domestic violence and gun violence
have multiple causes and antecedents such that no single
variable can fully explain their occurrence. Yet by suggesting
that edibles are the lone, immediate cause in the incidents
covered, the tragic regularity with which they occur goes
unnoticed. In 2013 adolescents and young adults committed
suicide at a rate of 10.9 per 100,000, and more than 41,000
suicides occurred in the United States overall. A firearm was
used in more than half (American Foundation for Suicide
Prevention, 2015). Similarly, between 2001 and 2012, over
6000 women were murdered by an intimate partner using a

59. gun (Center for American Progress, 2014). Such numbers
make these incidents tragically ‘‘routine’’, and sadly it is the
involvement of legal marijuana that accounts for the intense
and sustained media attention received.
Goode and Ben-Yehuda (1994/2009) describe elite-engin-
eered moral panic as driven and maintained by particular elite
interests (law enforcement, political, business) to divert
attention from societal problems such elites may have to
account for, or which if given attention, will otherwise hurt
their interests. Certainly those vested in continued marijuana
criminalization policies and limiting its availability as a
therapeutic and recreational substance benefit from dispropor -
tionate danger ascribed to it. Other interests are served when
attention is drawn from the public health problems associated
with the misuse of widely available legal pharmaceutical and
recreational substances, and firearms violence. Constructing
disproportionality around the danger of marijuana enables
these concerns to ‘‘hide in plain sight’’. Additionally, dispro-

60. portionality in media coverage diverts attention from policy-
related harms. For example, drug criminalization policies
divert attention from the fact that prohibition creates an
environment in which people, fearing the associated social and
legal repercussions, may avoid seeking medical attention for
accidental ingestion or overdose.
As other states consider legalization policies they look to
the experience in Colorado for guidance (Miller, 2016), and
news reports from that state greatly influence the general
discussion of marijuana policy. Support for marijuana legal-
ization in the United States has consistently increased in
recent years, with poll data indicating a majority (58%) of
Americans support it, the highest margin in 46 years (Jones,
2015). Support for legalization is strong among youth aged
18–34 (68%) and has risen sharply amongst all groups except
those aged 70 and older (People Press.org, 2015. With
legalization likely to continue expanding among states, it
appears marijuana is now less deviant than ever in the post

61. ‘‘reefer madness’’ era, and whatever fear and opposition
remains seems to surround edibles.
Study limitations include the sampling procedures. First,
the databases utilized may not have contained all relevant
textual news coverage of edibles in the United States, and
thus, relevant articles may be excluded. The keywords used to
retrieve articles may not have included all iterations of
legalized edibles, and implicit references to them or articles
discussing them by some other form were not included. In
addition, the exclusion of articles about illicit edibles might
overlook policy-related news and consideration of policy
reforms in reaction to depictions of edibles. Furthermore, the
sample excluded content from non-journalistic sources such
as social media and so-called ‘‘niche’’ blogs. Though textual
news media is a valid proxy for overall news media (Smith
et al., 2008), non-textual media (television, internet and radio
broadcasts) were not included. Finally, the study did not
discern whether and how media outlets may construct

62. marijuana coverage differently, for example whether patterns
exist based on geographic region or ownership structure.
Future studies should continue to examine media depic-
tions of marijuana, particularly as state legalization expands
along with calls to reschedule marijuana at the Federal level.
Concerns for youth marijuana use should receive attention, as
should attention to media coverage of other social problems
(child welfare, suicide, domestic violence and gun violence)
DOI: 10.3109/09687637.2016.1167168 Feeding on fear: Edible
Marijuana 469
in which marijuana use may be involved. Such studies should
particularly examine the depiction of marijuana relative to
these other concerns. Attention should also be paid to the
depiction of marijuana in influential non-textual media and
social media sources.
Declaration of interest
The author declares that no funding was received for this

63. research and there are no conflicts of interest.
References
Agar, M.H., & Reisinger, H.S. (2000). Read all about it: Media
construction of a heroin epidemic. Substance Use & Misuse, 35,
1551–1571. doi: 10.3109/10826080009148230.
American Foundation for Suicide Prevention, (2015). Facts and
Figures.
October 7, 2015. Retrieved from
https://www.afsp.org/understanding-
suicide/facts-and-figures.
Armstrong, E.G. (2007). Moral panic over meth. Contemporary
Justice
Review, 10, 427–442. doi: 10.1080/10282580701677519.
Baldwin, J.M., Miller, B.L., Stogner, J., & Hach, S. (2012). The
night the
raving died: the social construction of a local drug panic.
Deviant
Behavior, 33, 675–698. doi: 10.1080/01639625.2011.636723.
Berkey, M. (2011). Mary Jane’s new dance: the medical
marijuana legal
tango. Cardozo Public Law, Policy, and Ethics Journal, 9, 417–
450.
doi: http://trove.nla.gov.au/version/171489159.
Bostwick, J.M. (2012). Blurred boundaries: the therapeutics and
politics
of medical marijuana. Mayo Clinic Proceedings, 87, 172–186.
doi:
http://dx.doi.org/10.1016/j.mayocp.2011.10.003.

64. Center for American Progress, (2014). Fact sheets: Protecting
women
from gun violence. Retrieved from
https://www.americanprogress.org/
issues/guns-crime/report/2014/10/14/92119/protecting-women-
from-
gun-violence/.
Centers for Disease Control and Prevention (2015). Notes from
the field:
death following ingestion of an edible marijuana product –
Colorado,
March 2014. Retrieved from
http://www.cdc.gov/mmwr/preview/
mmwrhtml/mm6428a6.htm?-s_cid¼mm6428a6_x.
Chapkis, W., & Webb, R.J. (2005). Mother’s milk and the
muffin man:
Grassroots innovations in medical marijuana delivery systems.
Journal of Ethnicity in Substance Abuse, 4, 183–204. doi:
10.1300/
J233v04n03_08.
Chiricos, T. (1996). Moral panic as ideology: drugs, violence,
race, and
punishment. In Critcher, C. (Ed.), Critical readings: Moral
panics and
the media (pp. 103–123). Maidenhead: Open University Press.
Cobbina, J.E. (2008). Race and class differences in print media
portrayals of crack cocaine and methamphetamine. Journal of
Criminal Justice and Popular Culture, 15, 145–167.
Cohen, S. (1972/2002). Folk devils and moral panics: The
creation of the
mods and rockers (3rd ed.). New York: Routledge.

65. Cohen, P.J. (2009). Medical marijuana: the conflict between
scientific
evidence and political ideology. Part one of two. Journal of Pain
and
Palliative Care Pharmacotherapy, 23, 4–25. doi: 10.1080/
15360280902727973.
Coleman, H., & Unrau, Y.A. (2005). Analyzing qualitative data.
In
Grinnell, R.M., & Unrau, Y.A. (Eds.) Social work research and
evaluation: Quantitative and qualitative approaches (7th ed.)
(pp. 404–420). New York: Oxford University Press.
Colorado Department of Revenue (2015). Annual update.
Retrieved from
https://www.colorado.gov/pacific/sites/default/files/2014 MED
Annual Report_1.pdf.
David, M., Rohloff, A., Petley. J., & Hughes, J. (2011). The
idea of moral
panic – ten dimensions of dispute. Crime, Media, Culture, 7,
215–228.
doi: 10.1177/1741659011417601.
Denham, B.E. (2008). Folk devils, news icons, and the
construction of
moral panics: Heroin chic and the amplification of drug threats
in
contemporary society. Journalism Studies, 9, 945–961. doi:
10.1080/
14616700802227811.
Ferguson, R.W., Osterthaler, K., Kaminski, S., & Green, A.
(2015).
Medicine safety for children: An in-depth look at poison center

66. calls.
Washington D.C.: Safe Kids Worldwide Watch.
Goode, E. (1990). The American drug panic of the 1980’s:
Social
construction or objective threat? The International Journal of
the
Addictions, 25, 1083–1098. doi:10.3109/10826089009058874.
Goode, E., & Yehuda, N. (1994/2009). Moral panics: The social
construction of deviance (2nd ed.). Oxford, UK: Blackwell.
Goode, E. (2008). Moral panics and disproportionality: The case
of LSD
use in the sixties. Deviant Behavior, 29, 533–534. doi: 10.1080/
01639620701839377.
Grant, I., Atkinson, J.H., Gouaux, B., & Wilsey, B. (2012).
Medical
marijuana: Clearing away the smoke. Open Neurology Journal,
6,
18–25. doi: 10.2174/1874205X01206010018.
Hartman, D.M., & Golub, A. (1999). The social construction of
the crack
epidemic in the print media. Journal of Psychoactive Drugs, 31,
423–433. doi: 10.1080/02791072.1999.10471772.
Jenkins, P. (1994). ‘‘The ice age’’: The social construction of a
drug
panic. Justice Quarterly, 11, 7–31.
doi:10.1080/07418829400092111.
Jones, J. (2015). In U.S., 58% back legal marijuana use.
Retrieved from
http://www.gallup.com/poll/186260/back-legal-marijuana.aspx.

67. Kanny, D., Brewer, R., Mesnick, J., Paulozzi, L., Naimi, T., &
Lu, H.
(2015). Vital signs: Alcohol poisoning deaths — United States,
2010–
2012. Retrieved from
http://www.cdc.gov/mmwr/preview/mmwrhtml/
mm6353a2.htm.
Lamarine, R.J. (2012). Marijuana: Modern medical chimaera.
Journal of
Drug Education, 42. 1–11. doi: 10.2190/DE.42.1.a.
Linnemann, T. (2009). Mad men, meth moms, moral panic:
Gendering
meth crimes in the Midwest. Critical Criminology, 18, 95–110.
doi:
10.1007/s10612-009-9094-8.
Lovegrove, M.C., Mathew, J., Hampp, C., Governale, L.,
Wysowski,
D.K., & Budnitz, D.S. (2014). Emergency hospitalizations for
unsupervised prescription medication ingestions by young
children.
Pediatrics, 134, 1009–1016. doi: 10.1542/peds.2014-0840.
MacCoun, R.J., & Mello, M.M. (2015). Half-baked: The retail
promo-
tion of marijuana edibles. The New England Journal of
Medicine, 372,
989–991. doi: 10.1056/NEJMp1416014.
McRobbie, A., & Thonrton, S.L. (1995). Rethinking ‘moral
panic’ for
multi-mediated social worlds. British Journal of Sociology, 46,
559–574. doi:10.2307/591571.

68. Miller, J. (2016). In Colo., a look at life after marijuana
legalization. Retrieved from
https://www.bostonglobe.com/metro/
2016/02/21/from-colorado-glimpse-life-after-marijuana-
legalization/
rcccuzhMDWV74UC4IxXIYJ/story.html.
Musto, D. (1973). The American disease: Origins of narcotic
control.
New Haven: Yale University Press.
Omori, M.K. (2013). Moral panics and morality policy: The
impact of
media, political ideology, drug use, and manufacturing on
metham-
phetamine legislation in the United States. Journal of Drug
Issues, 43,
517–534. doi: 10.1177/0022042613491101.
Onders, B., Casavant, M.J., Spiller, H.A., Chounthirath, T., &
Smith,
G.A., (2015). Marijuana exposure among children younger than
six
years in the United States. Clinical Pediatrics, 55, 428–436.
DOI:
10.1177/0009922815589912.
People Press.org (2015). In debate over legalizing marijuana,
disagree-
ment over drug’s dangers. Retrieved from http://www.people-
press.-
org/2015/04/14/in-debate-over-legalizing-marijuana-
disagreement-over-
drugs-dangers/.

69. Reinarman, C., & Levine, H.G. (1997). The crack attack:
Politics and
media in the crack scare. In Reinerman C., Levine H.G., (Eds.),
Crack
in America: Demon drugs and social justice (pp. 18–51).
Berkeley:
University of California Press.
Riffe, D., Lacy S., & Fico, F. (1998). Analyzi ng media
messages: Using
quantitative content analysis in research. Mahwah: Lawrence
Erlbaum Associates.
Sloman, L. (1979/1998). Reefer madness: A history of
marijuana in
America. New York: St. Martin’s Griffin.
Smith, K.C., Wakefield, M.A., Terry-McElrath, Y., Chaloupka,
F.J.,
Flay, B., Johnston, L., . . . Siebel, C. (2008). Relation between
newspaper coverage of tobacco issues and smoking attitudes and
behaviour among American teens. Tobacco Control, 17, 17–24.
Weidner, R.R. (2009). Methamphetamine in three small
Midwestern
cities: Evidence of a moral panic. Journal of Psychoactive
Drugs, 41,
227–239. doi:10.1080/02791072.2009.10400533.
Whiting, P.F., Wolff, R.F., Deshpande, S., Di Nisio, M., Duffy,
S.,
Hernandez, A.V., . . . Kleijnen, J. (2015). Cannabinoids for
medical
use: A systematic review and meta-analysis. Journal of the
American
Medical Association, 313, 2456–2473. doi:

70. 10.1001/jama.2015.6358.
470 M. H. Eversman Drugs Educ Prev Pol, 2016; 23(6): 462–
470
Copyright of Drugs: Education, Prevention & Policy is the
property of Taylor & Francis Ltd
and its content may not be copied or emailed to multiple sites or
posted to a listserv without
the copyright holder's express written permission. However,
users may print, download, or
email articles for individual use.
Feeding on fear: Edible marijuana and disproportionality in US
mediaEdible marijuana: BackgroundDrug moral panics in
textual US news mediaMethodsResultsDiscussionReferences
Substance Use & Misuse, 45:865–887
Copyright © 2010 Informa Healthcare USA, Inc.
ISSN: 1082-6084 (print); 1532-2491 (online)
DOI: 10.3109/10826080903443610
Original Article
Nonmedical Prescription Drug Use Among US
College Students at a Midwest University: A Partial
Test of Social Learning Theory
ROBERT L. PERALTA AND JENNIFER L. STEELE
Department of Sociology, The University of Akron, Akron,
Ohio, USA

71. We estimate the prevalence of nonmedical prescription drug
(NMPD) use and test social
learning theory as an explanation for NMPD use based upon
data from a large pilot
study. Data were collected from 465 college students at a
Midwestern university in
the USA using a self-administered questionnaire. The sample
was predominantly white
(88%), 43% were female and the mean age was 22. Most
participants (80%) were
not members of social fraternities or sororities. A majority of
students did not report
NMPD use: 39.4% of respondents reported lifetime NMPD use,
31% reported past-
year use, and 14.4% reported past-month use. Multivariate
regression results partially
supported social learning theory as an explanation for lifetime
NMPD use. Limitations
and suggestions for future research are suggested.
Keywords social learning theory; nonmedical prescription drug
use; NMPD use; pre-
scription drug use and misuse; college students; deviance
Introduction
Relatively recently, calls for the use of social network theories
(the equivalent of social
learning theory) have been called for to better understand how
drug use occurs from a socio-
logical vantage point (see Valente, Gallaher, and Mouttapa,
2004 for a thorough overview).
In response to this call for research, we test social learning
theory as an explanation for
nonmedical prescription drug (NMPD) use using data from a

72. large pilot study collected at
a Midwest university in the United States. In the past decade,
NMPD use among college
students has been steadily rising (McCabe, West, and Wechsler,
2007). According to a 2004
report by the National Survey on Drug Use and Health, the
number of Americans initiating
The authors thank Dr.’s Keith Durkin (Ohio Northern
University) and Michelle Brown (Ohio
University) for early substantive contributions. We also thank
Dr. Stan Einstein, editor in Chief, Sub-
stance Use and Misuse, Dr. Cheryl Elman (The University of
Akron), and the anonymous reviewers
at Substance Use and Misuse for their suggestions and
comments on initial drafts. Finally, we thank
Dr. Christopher Eustis, Dr. Robert Jeantet, and Mr. Herman
Jara-Droguett of the Modern Languages
Department of The University of Akron for abstract translation.
Address correspondence to Dr. Robert L. Peralta, Olin Hall,
Akron, OH 44325-1905, Department
of Sociology, The University of Akron; E-mail:
[email protected]
865
866 Peralta and Steele
(i.e., used at least once) the nonmedical use of prescription
opioids over the age of 12 years
(2.4 million) exceeded individuals initially beginning the use of
marijuana (2.1 million)
or cocaine (1.0 million) with young adults (18–25) having the
highest rates of NMPD use

73. (14.5%) than any other age group (SAMPSA, 2006a, 2006b).
Recent national survey data have found that college students
may be more likely
than their noncollege peers to report misuse of prescription
drugs, specifically prescription
stimulants (Herman-Stahl, Krebs, Kroutil, and Heller, 2007;
Johnston, O’Malley, Bachman,
and Schulenburg, 2004). Research on NMPD use has tended to
focus on specific classes of
drugs. The three most commonly recognized categories of
prescription drugs examined in
the current literature are opiates, stimulants, and depressants.
There are only a handful of
studies that have examined all classes concurrently. Table 1
below displays recent findings
on NMPD use so that readers can make assessments about
generalizability.
A possible theoretical explanation for NMPD use at the college
level is social learning
theory (Akers, 1994, 1998; Sutherland and Cressey, 1960).
Social learning theory states
that deviant and/or criminal behavior and the justifications for
behavior are learned from
intimate groups. Critical conditions for social learning theory to
operate include exoge-
nous (e.g., social concerns such as the impact of friends who
use) as well as endogenous
conditions (e.g., physiological or psychological precursors to
substance use). The advan-
tage of this perspective is that socio-structural conditions (e.g.,
exposure to social norms
and socialization) are emphasized while limitations of social
learning theory center on
the problem of causal ordering and/or the ignoring of bio-

74. psychological predispositions to
substance misuse. Nevertheless, studies examining social
learning theory have proved to
be useful in explaining substance use and misuse. For example,
Triplett and Payne (2004)
and Ford (2008a) found that social learning theory was
supported in studies on adolescent
NMPD use and is a valuable criminological theory for
explaining this particular type of
drug use. Triplett and Payne (2004) also reported that social
learning theory was useful in
explaining NMPD use for a sample of adolescents. Other studies
suggest most users obtain
prescription drugs from friends: a key aspect of social learning
(Hurwitz, 2005; McCabe,
Knight, Teter, and Weschler, 2005; McCabe, Teter, and Boyd,
2005). Findings such as these
infer that peer association with drug-using acquaintances or
friends is predictive of NMPD
use (Ford, 2008a). Given these established findings, we suggest
that NMPD use among
college students can at least partially be explained by elements
of social learning theory.
Social learning theory is composed of four major components:
differential association,
definitions, imitation, and differential reinforcement (see Table
2). Differential association
states that deviant and/or criminal behavior is learned;
specifically, it is learned through
intimate social interaction (e.g., friendship networks, peer
socialization, and romantic re-
lationships). Learning criminal and/or deviant behavior includes
learning the techniques,
motives, rationalizations, and attitudes needed for committing
the violation (which is simi-

75. lar to learning pro-social behavior). An individual becomes
delinquent, for example, if he or
she poses more favorable definitions of crime over definitions
that are not favorable toward
crime. Learning criminal behavior or deviance thus follows the
same processes associated
with learning other behaviors (Akers, 1994, 1998; Ford, 2008a;
Sutherland and Cressey,
1960). Akers conducted a study examining drug use among
teenagers and concluded that
actual and anticipated rewards and punishments were significant
predictors of frequency of
drug use (Akers, 1998). Given the theory’s support in other
areas of study, it is surprising
that few studies have applied social learning theory to substance
use and misuse treatment
(Andes, 1994).
Although researchers are beginning to document the
epidemiological distribution
of NMPD use, there has been little theoretical investigation into
this particular form of
T
ab
le
1
R
ec
en
t

76. re
se
ar
ch
on
no
n-
m
ed
ic
al
pr
es
cr
ip
ti
on
dr
ug
(N
M
P
D
)

77. us
e:
a
re
so
ur
ce
ta
bl
e∗
A
ut
ho
r,
ye
ar
C
ou
nt
ry
H
yp
ot
he

78. si
s
/P
ur
po
se
C
ri
ti
ca
l
co
nc
ep
ts
U
nd
er
-p
in
ni
ng
P
op

79. ul
at
io
n
N
D
at
a
so
ur
ce
/
in
st
ru
m
en
ts
us
ed
A
na
ly
si

80. s
F
in
di
ng
s
L
im
it
at
io
ns
C
ri
ti
ca
l
un
re
so
lv
ed
is
su

81. es
B
oy
d,
M
cC
ab
e,
an
d
T
et
er
,2
00
6
U
S
A
D
es
cr
ip

82. ti
ve
st
ud
y
P
ai
n
m
ed
ic
at
io
n
E
vi
de
nc
e-
ba
se
d
S
tu

83. de
nt
s
ag
es
10
–1
8
in
a
M
id
w
es
t
pu
bl
ic
sc
ho
ol
di
st
ri

84. ct
10
17
U
se
of
qu
es
ti
on
s
ad
ap
te
d
fr
om
M
on
it
or
in
g
th

85. e
F
ut
ur
e;
sp
ec
ifi
c
qu
es
ti
on
on
no
np
re
sc
ri
be
d
us
e
of

86. pa
in
m
ed
ic
at
io
n
B
iv
ar
ia
te
an
d
lo
gi
st
ic
re
gr
es
si
on

87. ;
co
nt
en
t
an
al
ys
is
of
op
en
en
de
d
su
rv
ey
qu
es
ti
on
s

88. 16
%
li
fe
ti
m
e
us
e
(i
.e
.,
pa
in
m
ed
ic
at
io
n)
;
1%
us
e

89. in
pa
st
ye
ar
;
gi
rl
s
re
po
rt
ed
hi
gh
er
ra
te
s
of
us
e;
no
ra

90. ce
di
ff
er
en
ce
s
fo
un
d;
le
ad
in
g
so
ur
ce
s
of
pa
in
m
ed
ic

91. at
io
n
w
er
e
fr
ie
nd
s
an
d
fa
m
il
y
L
im
it
ed
op
er
a-
ti

92. on
al
iz
at
io
n
of
N
M
P
D
us
e
(s
pe
ci
fi
c
to
pa
in
m
ed
ic

93. at
io
n)
an
d
no
sp
ec
ifi
ca
ti
on
of
dr
ug
ty
pe
;c
ro
ss
se
ct
io
na

94. l
st
ud
y,
se
lf
-r
ep
or
t;
la
ck
of
qu
an
ti
ty
,
fr
eq
ue
nc
y
an

95. d
ro
ut
e
of
ad
m
in
is
tr
at
io
n
m
ea
su
re
s
P
ar
en
ta
l
aw

96. ar
en
es
s
of
N
M
P
D
us
e;
le
ve
l
of
ed
uc
at
io
n
an
d
aw
ar

97. en
es
s
ab
ou
t
N
M
P
D
us
e
in
sc
ho
ol
s
F
or
d,
20
08
a
U

98. S
A
S
oc
ia
l
co
nt
ro
l/
so
ci
al
le
ar
ni
ng
pr
ed
ic
to
r
of
N

99. M
P
D
us
e
S
ub
st
an
ce
ab
us
e/
de
li
nq
ue
nc
y
ne
xu
s
E
vi

100. de
nc
e-
ba
se
d
N
at
io
na
l
sa
m
pl
e
12
an
d
ol
de
r
55
,9
05

101. N
at
io
na
l
S
ur
ve
y
on
D
ru
g
U
se
an
d
H
ea
lt
h
20
05

102. O
L
S
an
d
lo
gi
st
ic
re
gr
es
si
on
8.
8%
pa
st
ye
ar
us
e;
pa
in

103. re
li
ev
er
us
e
m
os
t
co
m
m
on
;
N
M
P
D
us
e
si
gn
ifi
ca

104. nt
ly
as
so
ci
at
ed
w
it
h
se
lf
-r
ep
or
te
d
de
li
nq
ue
nc
y
S

105. el
f-
re
po
rt
da
ta
,
no
ni
ns
ti
tu
ti
on
-
al
iz
ed
sa
m
pl
e,
cr

106. os
s
se
ct
io
na
l
da
ta
;
li
m
it
ed
co
m
pr
eh
en
-
si
ve
ne
ss

107. of
pr
es
cr
ip
ti
on
dr
ug
s
in
su
rv
ey
Id
en
ti
fy
m
ot
iv
es
fo
r

108. us
e,
th
eo
re
ti
ca
l
pr
ed
ic
to
rs
of
us
e,
an
d
ex
pl
ai
n
in
cr

109. ea
se
s
in
us
e
ov
er
ti
m
e
F
or
d,
an
d
A
rr
as
ti
a,
20
08

110. U
S
A
S
oc
ia
l
co
nt
ro
l;
so
ci
al
le
ar
ni
ng
pr
ed
ic
to
r
of

111. N
M
P
D
us
e
C
or
re
la
te
s
of
us
e
E
vi
de
nc
e-
ba
se
d
N

112. at
io
na
l
co
ll
eg
e
st
ud
en
t
sa
m
pl
e
10
,4
01
H
ar
va
rd
S

113. ch
oo
l
of
P
ub
li
c
H
ea
lt
h
C
ol
le
ge
A
lc
oh
ol
S
ur
ve
y

114. 20
01
M
ul
ti
-
N
om
ia
l
lo
gi
st
ic
re
gr
es
si
on
11
%
li
fe
ti

115. m
e
us
e
(N
M
P
D
on
ly
);
w
hi
te
s
m
or
e
li
ke
ly
to
re
po

116. rt
N
M
P
D
us
e;
il
li
ci
t
st
re
et
dr
ug
us
e
as
so
ci
at
ed
w

117. it
h
gr
ea
te
r
ri
sk
-b
eh
av
io
r
th
an
N
M
P
D
;
su
pp
or
t

118. fo
r
so
ci
al
co
nt
ro
l
th
eo
ry
D
id
no
t
ac
co
un
t
fo
r
po
ly

119. -s
ub
st
an
ce
ab
us
e;
se
lf
-r
ep
or
t
da
ta
;
cr
os
s-
se
ct
io
na

120. l;
li
m
it
ed
co
m
-
pr
eh
en
si
ve
ne
ss
of
pr
es
cr
ip
ti
on
dr
ug

121. s
in
su
rv
ey
S
oc
ia
l
an
d
be
ha
vi
or
al
co
ns
eq
ue
nc
es
of

122. N
M
P
D
;
pe
rc
ei
ve
d
sa
fe
ty
of
N
M
P
D
us
e
co
m
pa
re

123. d
to
“s
tr
ee
t
dr
ug
s”
;
th
eo
re
ti
ca
l
pr
ed
ic
to
rs
of
us
e

124. (C
o
n
ti
n
u
ed
o
n
n
ex
t
p
ag
e)
867
T
ab
le
1
R
ec
en

125. t
re
se
ar
ch
on
no
n-
m
ed
ic
al
pr
es
cr
ip
ti
on
dr
ug
(N
M
P
D

126. )
us
e:
a
re
so
ur
ce
ta
bl
e∗
(C
on
ti
nu
ed
)
A
ut
ho
r,
ye
ar
C

127. ou
nt
ry
H
yp
ot
he
si
s
/P
ur
po
se
C
ri
ti
ca
l
co
nc
ep
ts
U
nd

128. er
-p
in
ni
ng
P
op
ul
at
io
n
N
D
at
a
so
ur
ce
/
in
st
ru
m
en

129. ts
us
ed
A
na
ly
si
s
F
in
di
ng
s
L
im
it
at
io
ns
C
ri
ti
ca
l

130. un
re
so
lv
ed
is
su
es
F
or
d
an
d
R
iv
er
a,
20
08
U
S
A
H
is

131. pa
ni
cs
le
ss
li
ke
ly
to
us
e
N
M
P
D
co
m
pa
re
d
to
w
hi
te

132. s
E
th
ni
c
an
d
ra
ce
pr
ed
ic
to
rs
of
N
M
P
D
us
e
E
vi

133. de
nc
e-
ba
se
d
N
at
io
na
l
sa
m
pl
e
12
an
d
ol
de
r
55
,9
05

134. N
at
io
na
l
S
ur
ve
y
on
D
ru
g
U
se
an
d
H
ea
lt
h
20
05
B

135. iv
ar
ia
te
an
d
lo
gi
st
ic
re
gr
es
si
on
12
.1
%
N
M
P
D
us
e

136. in
pa
st
ye
ar
,t
he
hi
gh
es
t
pr
ev
al
en
ce
w
as
op
ia
te
us
e
(1

137. 0%
);
H
is
pa
ni
cs
le
ss
li
ke
ly
to
us
e
co
m
pa
re
d
to
w
hi
te

138. s
bu
t
m
or
e
li
ke
ly
to
us
e
th
an
B
la
ck
s
an
d
A
si
an
s;

139. ac
cu
lt
ur
at
io
n
im
po
rt
an
t
fo
r
H
is
pa
ni
c
us
e
S
el
f

140. re
po
rt
da
ta
;
no
ni
ns
ti
tu
ti
on
-
al
iz
ed
sa
m
pl
e,
de
fi
ni

141. ti
on
of
N
M
P
D
us
e;
li
m
it
ed
co
m
-
pr
eh
en
si
ve
ne
ss
of

142. pr
es
cr
ip
ti
on
dr
ug
s
in
su
rv
ey
C
on
co
-
m
it
an
t
so
ci
al

143. pr
ob
le
m
s
st
em
m
in
g
fr
om
N
M
P
D
us
e
H
al
l,
Ir
w
in

144. ,
B
ow
m
an
,
F
ra
nk
en
be
rg
er
,
an
d
Je
w
et
t,
20
05
U
S

145. A
D
es
cr
ip
ti
ve
st
ud
y
A
D
H
D
;
pr
es
cr
ib
ed
st
im
ul
an

146. ts
,
co
ll
eg
e
st
ud
en
ts
E
vi
de
nc
e-
ba
se
d
M
id
w
es
t
co

147. ll
eg
e
st
ud
en
ts
38
1
L
ik
er
t-
ty
pe
su
rv
ey
on
op
in
io
ns
an

148. d
at
ti
tu
de
s
to
w
ar
d
st
im
ul
an
t
us
e
A
N
O
V
A
13

149. .7
%
us
ed
st
im
ul
an
ts
th
at
w
er
e
no
t
pr
es
cr
ib
ed
;
an
ti

150. -
de
pr
es
sa
nt
s
m
os
t
co
m
m
on
dr
ug
ta
ke
n
(7
%
);
m
en

151. m
or
e
li
ke
ly
to
ha
ve
re
po
rt
ed
ta
ki
ng
st
im
ul
an
ts
fo
r
no

152. nm
ed
ic
al
pu
rp
os
es
In
it
ia
l
lo
w
re
sp
on
se
ra
te
fr
om
m
en

153. ;
es
ti
m
at
es
m
ay
be co
ns
er
va
ti
ve
;
si
ng
le
co
ll
eg
e
st
ud

154. ie
d;
lo
w
ge
ne
ra
li
za
ti
on
W
ha
t
is
th
e
im
pa
ct
of
st
im
ul

155. an
t
pr
es
cr
ip
ti
on
fo
r
ch
il
dr
en
on
th
ei
r
ad
ol
se
ce
nt
N

156. M
P
D
us
e?
R
eg
io
na
l
di
ff
er
en
ce
s,
if
an
y,
ne
ed
to
be
ex

157. pl
or
ed
868
H
er
m
an
-S
ta
hl
,
K
re
bs
,K
ro
ut
il
,
an
d

158. H
el
le
r,
20
07
U
S
A
D
es
cr
ip
ti
ve
st
ud
y
M
et
ha
m
-
ph

159. et
am
in
e
E
vi
de
nc
e-
ba
se
d
Y
ou
ng
ad
ul
ts
(1
8–
25
)
23
,6

160. 45
N
at
io
na
l
S
ur
ve
y
on
D
ru
g
U
se
an
d
H
ea
lt
h
20
02

161. M
ul
ti
va
ri
at
e
an
al
ys
is
2.
1%
us
ed
N
M
P
D
(s
ti
m
ul
an

162. ts
)
in
pa
st
ye
ar
;
w
hi
te
s
ha
d
th
e
hi
gh
es
t
ra
te
;
se

163. ll
in
g
an
d
us
in
g
il
li
ci
t
dr
ug
s
as
so
ci
at
ed
w
it
h
in

164. cr
ea
se
d
od
ds
fo
r
N
M
P
D
st
im
ul
an
t
us
e;
ps
yc
ho
lo
gi

165. ca
l
di
st
re
ss
,
se
ns
at
io
n-
se
ek
in
g,
bi
ng
e
dr
in
ki
ng
,a

166. nd
co
ll
eg
e
en
ro
ll
m
en
t
as
so
ci
at
ed
w
it
h
in
cr
ea
se
d

167. ri
sk
of
N
M
P
D
st
im
ul
an
t
us
e
N
M
P
D
su
rv
ey
m
ea
su

168. re
s
no
t
op
ti
m
al
(l
im
it
ed
de
sc
ri
pt
io
n
of
;
ch
oi
ce
s

169. fo
r
pr
es
cr
ip
ti
on
dr
ug
us
e)
;
cr
os
s-
se
ct
io
na
l;
un
kn
ow

170. n
if
cu
rr
en
t
N
M
P
D
w
as
th
e
re
su
lt
of
pa
st
pr
es
cr
ip

171. ti
on
;
li
ft
im
e
an
d
pa
st
m
on
th
m
ea
su
re
s
no
t
us
ed
T

172. he
di
m
en
si
on
s
of
ps
yc
ho
lo
gi
ca
l
st
re
ss
as
th
ey
pe
rt
ai

173. n
to
ge
nd
er
,r
ac
e,
et
hn
ic
it
y,
an
d
ag
e
ar
e
im
po
rt
an
t

174. fo
r
fu
tu
re
co
ns
id
er
at
io
n
K
ok
ke
vi
,F
ot
io
u,
an
d
R
ic

175. ha
rd
so
n,
20
07
G
re
ec
e
D
es
cr
ip
ti
ve
st
ud
y
P
re
va
le
nc

176. e
an
d
in
ci
de
nc
e
tr
en
ds
E
vi
de
nc
e-
ba
se
d
N
at
io
na

177. l
sa
m
pl
e
ag
es
12
–6
4,
ur
ba
n,
se
m
i
ur
br
an
,a
nd
ru
ra
l

178. 4,
00
0
st
ru
ct
ur
ed
qu
es
ti
on
na
ir
e,
fa
ce
to
fa
ce
L
og
is
ti

179. c
re
gr
es
si
on
9.
1%
en
ga
ge
d
in
an
y
N
M
P
D
us
e
in
th
e

180. pa
st
ye
ar
in
20
04
;
an
al
ge
si
c
us
e
m
os
t
co
m
m
on
;
fe

181. m
al
es
ou
tn
um
be
r
m
al
es
in
N
M
P
D
us
e
bu
t
di
ff
er
en

182. ce
de
cr
ea
si
ng
ov
er
ti
m
e
R
es
po
ns
e
va
li
di
ty
an
d
re
li

183. ab
il
it
y
(l
ow
pr
iv
ac
y)
;
po
ss
ib
le
un
de
r-
es
ti
m
at
io
n;

184. in
co
ns
is
te
nc
ie
s
ac
ro
ss
su
rv
ey
s
P
ea
ks
an
d
do
w
nt
ur

185. ns
in
us
e
ne
ed
ex
pl
an
at
io
n;
ef
fe
ct
s
of
tr
ea
tm
en
t
de
m

186. an
d
an
d
po
li
ce
in
te
rv
en
ti
on
s
ne
ed
ed
(C
o
n
ti
n
u
ed

187. o
n
n
ex
t
p
ag
e)
869
T
ab
le
1
R
ec
en
t
re
se
ar
ch
on
no

188. n-
m
ed
ic
al
pr
es
cr
ip
ti
on
dr
ug
(N
M
P
D
)
us
e:
a
re
so
ur

189. ce
ta
bl
e∗
(C
on
ti
nu
ed
)
A
ut
ho
r,
ye
ar
C
ou
nt
ry
H
yp
ot
he

190. si
s
/P
ur
po
se
C
ri
ti
ca
l
co
nc
ep
ts
U
nd
er
-p
in
ni
ng
P
op

191. ul
at
io
n
N
D
at
a
so
ur
ce
/
in
st
ru
m
en
ts
us
ed
A
na
ly
si

192. s
F
in
di
ng
s
L
im
it
at
io
ns
C
ri
ti
ca
l
un
re
so
lv
ed
is
su

193. es
K
ok
ke
vi
,F
ot
io
u,
A
ra
pa
ki
,a
nd
R
ic
ha
rd
so
n,
20
08

194. E
ur
op
e
D
es
cr
ip
ti
ve
st
ud
y
P
re
va
le
nc
e,
co
rr
el
at
es

195. E
vi
de
nc
e-
ba
se
d
A
do
le
sc
en
ts
fr
om
31
E
ur
op
ea
n
co
un

196. tr
ie
s
85
,0
00
S
ta
nd
a-
rd
iz
ed
su
rv
ey
s
L
og
is
ti
c
re
gr

197. es
si
on
5.
6%
li
fe
ti
m
e
no
nm
ed
ic
al
tr
an
qu
il
iz
er
or
se
da

198. ti
ve
us
e;
N
M
P
D
us
e
as
so
ci
at
ed
w
it
h
to
ba
cc
o,
al
co

199. ho
l,
an
d
il
li
ci
t
dr
ug
s;
fi
nd
in
gs
co
m
pa
ra
bl
e
to
U
S

200. ;
L
it
hu
an
ia
ha
d
th
e
hi
gh
es
t
ra
te
of
N
M
P
D
an
d
U

201. kr
ai
ne
ha
d
th
e
lo
w
es
t
T
he
us
e
of
tr
an
qu
il
iz
er
s
an

202. d
se
da
ti
ve
s
co
m
bi
ne
d
in
a
si
ng
le
qu
es
ti
on
;
li
m
it

203. ed
op
er
a-
ti
on
al
iz
at
io
n
of
N
M
P
D
;
cr
os
s-
se
ct
io
na

204. l;
se
lf
-r
ep
or
t
C
ro
ss
va
-
li
da
ti
on
th
ro
ug
h
co
nt
ex
tu

205. al
in
fo
rm
at
io
n
on
pr
es
cr
ip
ti
on
pr
ac
ti
ce
s
an
d
co
nt
ro

206. ls
in
di
ff
er
en
t
co
un
tr
ie
s
or
re
ta
il
sa
le
s
ne
ed
ed
L
eu

207. ke
fe
ld
,
M
cD
on
al
d,
M
at
ey
ok
e-
S
cr
iv
en
er
,
R
ob
er
to

208. ,
W
al
ke
r,
W
eb
st
er
,a
nd
G
ar
ri
ty
,2
00
5
U
S
A
O
xy
C

209. on
ti
n
us
er
s
w
il
l
be
m
or
e
dr
ug
in
vo
lv
ed
,
us
e
m
or

210. e
he
al
th
se
rv
ic
es
,a
nd
co
m
m
it
m
or
e
cr
im
es
th
an
no
nu

211. se
rs
O
xy
C
on
ti
n
(n
ar
co
ti
c
an
al
ge
si
c)
us
er
s
ve
rs
us

212. no
nu
se
rs
;
dr
ug
us
e
cr
im
in
al
it
y
ne
xu
s
ev
id
en
ce
-b
as

213. ed
P
ur
po
si
ve
sa
m
pl
e
of
ru
ra
l
M
id
w
es
t
pr
ob
at
io
ne

214. rs
29
5
A
dd
ic
ti
on
se
ve
ri
ty
in
de
x;
M
ia
m
i
H
ea
lt
h

215. S
er
vi
ce
s
U
ti
li
za
ti
on
Q
ue
st
io
nn
ai
re
U
ni
va
ri
at
e,

216. ch
i
sq
ua
re
s;
t-
te
st
s
O
xy
C
on
ti
n
us
er
s
si
gn
ifi
ca
nt

217. ly
yo
un
ge
r,
ex
hi
bi
te
d
hi
gh
er
le
ve
ls
of
ot
he
r
il
li
ci
t

218. dr
ug
us
e,
us
ed
m
or
e
de
to
xi
fi
ca
ti
on
se
rv
ic
es
,
re
po
rt

219. ed
hi
gh
er
le
ve
ls
of
m
en
ta
l
il
ln
es
s
an
d
cr
im
in
al
be
ha

220. vi
or
th
an
no
nu
se
rs
P
re
do
m
i-
na
nt
ly
w
hi
te
m
al
e
sa
m

221. pl
e;
di
d
no
t
co
nt
ro
l
fo
r
ag
e;
di
vi
si
on
be
tw
ee
n
O
xy

222. C
on
ti
n
us
e
an
d
no
nu
se
;
no
nr
ep
re
se
nt
a-
ti
ve
sa
m
pl

223. e;
se
lf
-r
ep
or
t
T
ar
ge
ti
ng
N
M
P
D
ab
us
er
s
fo
r
tr
ea

224. tm
en
t
fo
r
ru
ra
l
ar
ea
s;
N
M
P
D
tr
ea
tm
en
t
an
d
ru
ra

225. l
N
M
P
D
tr
ea
tm
en
t
ef
fe
ct
iv
en
es
s;
pa
tt
er
ns
of
in
it

226. ia
l
us
e
an
d
co
nt
in
ue
d
us
e
L
eu
ke
fe
ld
,
W
al
ke
r,
H

227. av
en
s,
an
d
T
al
be
rt
,2
00
7
U
S
A
pe
rc
ep
ti
on
s
an
d
op

228. ti
on
s
w
il
l
va
ry
ac
ro
ss
ty
pe
s
of
in
to
rm
an
t
gr
ou
p

229. pe
rc
ep
ti
on
s
of
ru
ra
l
N
M
P
D
us
e
th
eo
ry
ba
se
d
ke
y

230. in
fo
rm
an
ts
fr
om
A
pp
al
ac
hi
an
co
un
ti
es
in
K
en
tu
ck
y
(e

231. .g
.,
el
ec
te
d
of
fi
ci
al
s,
bu
si
ne
ss
le
ad
er
s)
70
fa
ce
-t
o-

232. fa
ce
st
ru
ct
ur
ed
in
te
rv
ie
w
s
qu
al
it
at
iv
e
an
al
ys
is
de

233. ri
ve
d
fr
om
sn
ow
ba
ll
sa
m
pl
in
g
N
M
P
D
us
e
is
no
t
ne

234. w
ph
ys
ic
al
pa
in
,
re
cr
ea
ti
on
al
us
e
an
d
cu
lt
ur
al
ac
ce

235. pt
an
ce
ar
e
pa
th
w
ay
s
to
pr
ob
le
m
us
ed
ar
ti
cl
es
an
d
op

236. ti
on
s
di
d
no
t
ve
ry
by
gr
ou
p
ex
tr
em
ly
hi
gh
ra
te
of
N
M

237. P
D
us
e
N
on
ra
nd
om
sa
m
pl
in
g
of
re
sp
on
de
nt
s
se
lf
re

238. po
rt
ed
da
ta
cr
os
s-
se
ct
io
na
l
de
si
gn
V
ie
w
s
fr
om
la
w

239. en
fo
rc
em
en
t
an
d
ju
di
ci
al
sy
st
em
ne
ed
ed
re
se
ar
ch
on
tr

240. ea
tm
en
t
an
d
pr
ev
en
ti
on
pe
rt
in
an
t
to
ru
ra
l
co
m
m
un

241. it
ie
s
in
ge
ne
ra
l
al
so
in
ne
ed
870
L
an
ke
na
u,
S
an
de

242. rs
,
B
lo
om
,
H
at
ha
zi
,
B
la
rc
on
,T
or
tu
,
an
d
C
la
tt

243. s
20
07
U
S
A
E
xp
lo
ra
to
ry
ex
am
in
at
io
n
of
co
nt
ex
t

244. an
d
pa
tt
er
ns
as
so
ci
at
ed
w
it
h
N
M
P
D
us
e
am
on
g
in

245. je
ct
io
n
dr
ug
us
er
s
(I
D
U
s)
P
ol
yd
ru
g
us
e;
ID
U
s
T

246. he
or
y-
ba
se
d
Y
ou
th
an
d
yo
un
g
ID
U
s
fr
om
3
U
S
ci
ti

247. es
21
3
F
ac
e-
to
-f
ac
e
st
ru
ct
ur
ed
in
te
rv
ie
w
s
M
ix
es

248. d
m
et
ho
ds
(d
es
cr
ip
ti
ve
un
iv
ar
ia
te
an
d
qu
al
it
at
iv
e

249. an
al
ys
is
)
95
.8
%
N
M
P
D
us
e;
op
io
id
s
m
os
t
co
m
m

250. on
;
N
M
P
D
an
in
te
gr
al
pa
rt
of
dr
ug
us
e
am
on
g
ID
U
s

251. P
ri
m
ar
il
y
m
al
e,
w
hi
te
,
he
te
ro
se
xu
al
sa
m
pl
e

252. R
es
ea
rc
h
on
in
it
ia
ti
on
,u
se
,
su
pp
ly
,a
nd
co
nt
ex
ts
am

253. on
g
ID
U
s
ne
ed
ed
M
cC
ab
e,
W
es
t,
M
or
al
es
,
C
ra
nf
or

254. d,
an
d
B
oy
d,
20
07
U
S
A
D
es
cr
ip
ti
ve
st
ud
y
E
ar
ly
on

255. se
t,
de
pe
nd
en
ce
E
vi
de
nc
e-
ba
se
d
N
at
io
na
l
sa
m
pl
e

256. of
ad
ul
ts
18
an
d
ov
er
43
,0
93
N
at
io
na
l
E
pi
de
m
io
lo
gi

257. c
S
ur
ve
y
on
al
co
ho
l-
re
la
te
d
co
nd
it
io
ns
20
01
–2
00
2

258. R
es
ul
ts
fr
om
st
ru
ct
ur
ed
di
ag
no
st
ic
in
te
rv
ie
w
s
an
al

259. yz
ed
vi
a
bi
va
ri
at
e,
ch
i
sq
ua
re
,a
nd
lo
gi
st
ic
re
gr
es
si

260. on
9.
1%
of
ad
ul
ts
li
fe
ti
m
e
N
M
P
D
us
e;
ea
rl
y
ye
ar

261. of
N
M
P
D
on
se
t
pr
ed
ic
ti
ve
of
ad
ul
t
N
M
P
D
ab
us
e

262. an
d
de
pe
nd
en
ce
;
po
ly
dr
ug
us
e
co
m
m
on
N
M
P
D
su
rv

263. ey
m
ea
su
re
s
no
t
op
ti
m
al
(i
.e
.,
co
m
pr
eh
en
si
ve
li
st

264. no
t
of
fe
re
d
to
pa
rt
ic
ip
an
ts
)
th
us
es
ti
m
at
es
li
ke
ly

265. co
ns
er
va
ti
ve
;
ba
se
d
on
se
co
nd
ar
y
an
al
ys
is
;
w
or
d

266. ch
oi
ce
di
ff
er
en
ce
s
co
m
pa
re
d
to
ot
he
r
su
rv
ey
s;
fa
il

267. ur
e
to
id
en
ti
fy
th
e
m
is
us
e
of
on
e’
s
ow
n
m
ed
ic
at
io

268. n
ve
rs
us
th
e
m
is
us
e
of
ot
he
r’
s
m
ed
ic
at
io
ns
;
re
su

269. lt
s
m
ay
no
t
ge
ne
ra
li
ze
ou
ts
id
e
th
e
U
S
A
C
on
te
xt

270. of
ea
rl
y
on
se
t
im
po
rt
an
t
to
st
ud
y;
so
ci
al
,
ps
yc
ho
lo

271. gi
ca
l,
an
d
ph
ys
io
lo
gi
ca
l
vu
ln
er
ab
il
it
ie
s
to
ea
rl
y

272. us
e
im
po
rt
an
t
to
co
ns
id
er
;
ro
le
of
fa
m
il
y
hi
st
or
ie

273. s
of
al
co
ho
li
sm
im
po
rt
an
t
to
co
ns
id
er
(C
o
n
ti
n
u
ed

274. o
n
n
ex
t
p
ag
e)
871
T
ab
le
1
R
ec
en
t
re
se
ar
ch
on
no

275. n-
m
ed
ic
al
pr
es
cr
ip
ti
on
dr
ug
(N
M
P
D
)
us
e:
a
re
so
ur

276. ce
ta
bl
e∗
(C
on
ti
nu
ed
)
A
ut
ho
r,
ye
ar
C
ou
nt
ry
H
yp
ot
he

277. si
s
/P
ur
po
se
C
ri
ti
ca
l
co
nc
ep
ts
U
nd
er
-p
in
ni
ng
P
op

278. ul
at
io
n
N
D
at
a
so
ur
ce
/
in
st
ru
m
en
ts
us
ed
A
na
ly
si

279. s
F
in
di
ng
s
L
im
it
at
io
ns
C
ri
ti
ca
l
un
re
so
lv
ed
is
su

280. es
M
cC
ab
e
an
d
T
et
er
,2
00
7
U
S
A
D
es
cr
ip
ti
ve
st

281. ud
y
P
re
sc
ri
pt
io
n
st
im
ul
an
ts
E
vi
de
nc
e-
ba
se
d
W
eb

282. -b
as
ed
M
id
w
es
t
sa
m
pl
e
3,
63
9
M
od
ifi
ed
us
e
of
th
e

283. D
ru
g
A
bu
se
S
cr
ee
ni
ng
T
es
t
(D
A
S
T
)
B
iv
ar
ia
te

284. ,A
N
O
V
A
8.
5%
li
fe
ti
m
e
N
M
P
D
us
e,
w
hi
le
pa
st
ye

285. ar
us
e
w
as
6%
;
N
M
P
D
us
er
s
m
or
e
li
ke
ly
to
re
po
rt

286. po
ly
dr
ug
us
e
(9
0%
);
m
od
e
of
N
M
P
D
in
ge
st
io
n
si
gn

287. ifi
ca
nt
N
M
P
D
su
rv
ey
m
ea
su
re
s
no
t
op
ti
m
al
(l
im
it

288. ed
de
sc
ri
pt
io
n
of
/c
ho
ic
es
fo
r
pr
es
cr
ip
ti
on
dr
ug
us
e)

289. ;
D
A
S
T
no
t
te
st
ed
in
W
eb
-b
as
ed
ve
rs
io
ns
;
no
nr
es

290. po
ns
e
ra
te
co
ul
d
be
pr
ob
le
m
at
ic
F
ut
ur
e
st
ud
ie
s
sh

291. ou
ld
ex
am
in
e
m
ul
ti
pl
e
co
ll
eg
es
fr
om
a
ra
ng
e
of
ge
og

292. ra
ph
ic
lo
ca
ti
on
s
fo
r
co
m
pa
ri
so
n
pu
rp
os
es
an
d
to
as

293. se
ss
ge
ne
ra
li
za
ti
on
;
la
rg
er
,m
or
e
di
ve
rs
e
sa
m
pl
es

294. ar
e
ne
ed
ed
;
se
xu
al
it
y,
ra
ce
,
an
d
et
hn
ic
it
y
di
ff
er

295. en
ce
s
ar
e
in
ne
ed
of
st
ud
y
M
cC
ab
e
W
es
t,
an
d
W
ec
hs

296. le
s,
20
07
U
S
A
D
es
cr
ip
ti
ve
st
ud
y
T
re
nd
s
an
d
co
ll

297. eg
e
le
ve
l
ch
ar
ac
te
ri
st
ic
s
E
vi
de
nc
e-
ba
se
d
N
at

298. io
na
l
sa
m
pl
e
of
co
ll
eg
e
st
ud
en
ts
19
93
–2
00
1
15
,2
82

299. 14
,4
28
13
,9
53
10
,9
04
C
ol
le
ge
A
lc
oh
ol
S
tu
dy
H
L
M
,m

300. ul
ti
pl
e
pa
ir
w
is
e
co
m
pa
ri
so
ns
,
ch
i
sq
ua
re
te
st
s,

301. 4.
4%
pa
st
ye
ar
N
M
P
D
us
e
in
19
93
an
d
ro
se
to
9.
97
%
in

302. 20
01
;
li
fe
ti
m
e
N
M
P
D
us
e
w
as
16
.3
4%
st
ud
en
ts
fr

303. om
H
is
to
ri
ca
ll
y
B
la
ck
sc
ho
ol
s
le
ss
li
ke
ly
to
us
e;
H

304. L
M
su
gg
es
ts
a
st
ea
dy
,
si
gn
ifi
ca
nt
ly
in
cr
ea
si
ng
tr
en

305. d
in
us
e
F
in
di
ng
s
ba
se
d
on
se
co
nd
ar
y
an
al
ys
is
;
N

306. M
P
D
us
e
su
rv
ey
m
ea
su
re
s
no
t
op
ti
m
al
;
da
ta
ar
e

307. cr
os
s
se
ct
io
na
l
an
d
no
t
lo
ng
it
ud
in
al
;
se
lf
-r
ep
or

308. t;
va
ri
at
io
ns
in
re
sp
on
se
ra
te
s
fr
om
di
ff
er
en
t
in
st
it

309. ut
io
ns
R
es
ea
rc
h
on
w
ha
t
ha
s
fa
ci
li
ta
te
d
in
cr
ea

310. se
s
in
N
M
P
D
ac
ro
ss
ti
m
e
w
ou
ld
be
us
ef
ul
as
w
ou
ld

311. re
se
ar
ch
on
th
e
so
ci
o-
st
ru
ct
ur
al
pr
oc
es
se
s
as
so
ci
at

312. ed
w
it
h
re
gi
on
al
an
d
hi
st
or
ic
al
ly
bl
ac
k
st
at
us
di
ff

313. er
en
ce
s
872
M
cC
ab
e,
C
ra
nf
or
d,
an
d
B
oy
d,
20
06
U
S

314. A
D
es
cr
ip
ti
ve
st
ud
y
A
lc
oh
ol
us
e
co
-o
cc
ur
re
nc
e
E

315. vi
de
nc
e-
ba
se
d
N
at
io
na
l
sa
m
pl
e
43
,0
93
N
at
io
na
l

316. E
pi
de
m
io
lo
gi
c
S
ur
ve
y
on
A
lc
oh
ol
R
el
at
ed
co
nd
it

317. io
ns
20
01
–2
00
2
R
es
ul
ts
fr
om
st
ru
ct
ur
ed
di
ag
no
st
ic

318. in
te
rv
ie
w
s
A
na
ly
ze
d
w
it
h
bi
va
ri
at
e,
ch
i
sq
ua
re

319. ,a
nd
lo
gi
st
ic
re
gr
es
si
on
3.
17
%
pa
st
ye
ar
N
M
P
D
us
e

320. ov
er
al
l
sa
m
pl
e;
18
–2
4
ye
ar
s
ha
d
hi
gh
es
t
pa
st
ye
ar

321. R
at
e
at
7.
51
%
;
dr
in
ki
ng
se
ve
ri
ty
si
gn
ifi
ca
nt
ly
as
so

322. ci
at
ed
w
it
h
N
M
P
D
(a
lc
oh
ol
-
de
pe
nd
en
t
18
X
m
or

323. e
li
ke
ly
to
re
po
rt
an
y
N
M
P
D
us
e)
N
M
P
D
su
rv
ey
m

324. ea
su
re
s
no
t
op
ti
m
al
(i
.e
.
co
m
pr
eh
en
si
ve
li
st
no
t

325. of
fe
re
d
to
pa
rt
ic
ip
an
ts
th
us
es
ti
m
at
es
li
ke
ly
co
ns
er

326. va
ti
ve
;
ba
se
d
on
se
co
nd
ar
y
an
al
ys
is
;
w
or
d
ch
oi
ce

327. di
ff
er
en
ce
s
co
m
pa
re
d
to
ot
he
r
su
rv
ey
s;
fa
il
ur
e
to