DAWSON, COMPTON, AND GRANT 751

Frequency of 5+/4+ Drinks as a Screener for Drug Use and

Drug-Use Disorders*
DEBORAH A. DAWSON, PH.D.,

WILSON M. COMPTON, M.D., M.P.E.,

AND BRIDGET F. GRANT, PH.D., PH.D.

Laboratory of Epidemiology and Biometry, National Institute on Alcohol Abuse and Alcoholism, 5635 Fishers Lane, MSC 9304, Bethesda,
Maryland 20892-9304
751
ABSTRACT. Objective: The objective of this study was to test the
ability of a question on frequency of drinking 5+ (for men) or 4+ (for
women) drinks to screen for drug use and drug-use disorders (DUDs)
in a general population sample. Method: Using data collected in 2001-
2002 from a representative U.S. adult population sample (N = 43,093),
including a subsample of those with past-year emergency-department use
(n = 8,525), past-year frequency of drinking 5+/4+ drinks was evaluated
as a screener for drug use and DUDs for four categories of illicit drugs.
Results: Sensitivities and specicities of the 5+/4+ drinks screener were
72.4% and 76.6% for any drug dependence, 71.9% and 77.3% for any
DUD, and 63.3% and 78.9% for any drug use in the general population.
Sensitivities and specicities were higher for marijuana and cocaine/
crack and lowest for illicit prescription drugs. Optimal screening cut-
points were once a month or more for cocaine/crack dependence, either
once or more a month or seven or more times a year for cocaine/crack
DUDs, seven or more times a year for cocaine/crack use, and once or
more a year for the other drug use and DUD measures. Sensitivity and
specicity were similar among adults who had visited an emergency
department in the past year, and the optimal screening cutpoints were
identical. Conclusions: Past-year frequency of drinking 5+/4+ drinks
was quite accurate as a screener for past-year marijuana and cocaine/
crack use and DUDs, but it was less accurate for illicit prescription drug
use and DUDs. Its drug-screening potential can be thought of as added
value from an item already likely to be asked in the interest of detecting
problem drinking. Future work may consider using the alcohol consump-
tion screener as a starting point, with follow-up questions to assess illicit
drug use among those who screen positive. (J. Stud. Alcohol Drugs, 71,
751-760, 2010)
Received: November 30, 2009. Revision: March 15, 2010.
*The study on which this article is based, the National Epidemiologic
Survey on Alcohol and Related Conditions, is sponsored by the National
Institute on Alcohol Abuse and Alcoholism, National Institutes of Health,
U.S. Department of Health and Human Services, with supplemental support
from the National Institute on Drug Abuse. This research was supported by
the Intramural Program of the National Institutes of Health, National Insti-
tute on Alcohol Abuse and Alcoholism. All authors are federal government
employees, and none of the authors has any nancial conict of interest to
report. The views and opinions expressed in this article are those of the authors
and should not be construed to represent the views of any of the sponsoring
organizations, agencies, or the U.S. government. All authors had full access
to all of the data in the study and take responsibility for the integrity of the
data and the accuracy of the data analysis.

Correspondence may be sent to Deborah A. Dawson at the above ad-

dress or via email at: [email protected]. Wilson M. Compton is with
the Division of Epidemiology, Services and Prevention Research, National
Institute on Drug Abuse, Bethesda, MD.
I
LLICIT DRUG USE IS COMMON in the United States.
Recent data indicate that approximately 20 million indi-
viduals 12 years of age and older used at least one illicit
drug in the past month, and 2.8% had a past-year illicit drug-
use disorder (DUD; Substance Abuse and Mental Health
Services Administration, 2009). Use of illicit drugs is associ-
ated with numerous adverse health consequences affecting
multiple organs and body systems (Khalsa et al., 2002, 2008)
and accounts for 0.5% of deaths and 1.8% of the burden
of disease in developed regions of the world (Rehm et al.,
2006). Drug use and DUDs are also associated with signi-
cant social, mental, and emotional impairment (Compton et
al., 2007) and staggering economic costs (Cartwright, 1999;
Ofce of National Drug Policy, 2004).
Because of their severe consequences, it is a crucial
public health concern to identify and offer interventions
for illicit drug use and DUDs. Although illicit drug use and
DUDs are not uncommon, most persons with these behaviors
do not automatically request treatment; in fact, most indi-
viduals with diagnosable DUDs do not seek care (Compton
et al., 2007; Substance Abuse and Mental Health Services
Administration, 2009). One way to address this issue is to
provide assertive outreach in settings where drug users and
persons with DUDs are likely to be identied. Primary care,
emergency department (ED), and similar health care settings,
especially those that have a large number of adolescent or
young-adult patients, are examples of such logical venues
for drug screening. Recent data from a nationally repre-
sentative sample of U.S. adults indicated that rates of using
illicit drugs at least monthly were similar among those who
did and did not report any past-year primary care use, 3.2%
and 3.6%, respectively, but were higher among those who
reported ED use than those who did not, 6.6% versus 3.2%
(Cherpitel and Ye, 2008). These survey results of drug-use
prevalence among primary care patients are broadly compa-
rable to rates from actual samples of primary care and ED
patients (e.g., rates of 3.2%-5.2% in primary care samples;
Manwell et al., 1998; Mertens et al., 2005). The prevalence
of drug use in ED samples has varied as a function of the
population being served by the ED and the type of screening,
752 JOURNAL OF STUDIES ON ALCOHOL AND DRUGS / SEPTEMBER 2010
but testing of physical specimens in a sample of Tennessee
ED patients found that the proportion of positive screens of
physical specimens ranged from 15% for marijuana to 3.1%
for barbiturates (Rockett et al., 2003). A study of 2,366
inner-city Canadian ED patients found that approximately
18% had used street drugs (Cummings et al., 2006), whereas
a Swiss ED study of 2,304 women and 2,688 men reported
rates of past-year marijuana and other illicit drug use of
5.0% and 1.4% for women and 13.2% and 3.8% for men
(Fleming et al., 2007).
Psychiatric and general health care clinicians have been
encouraged to screen patients routinely for alcohol and illicit
drug problems to identify and intervene with those who have
clinically signicant levels of symptoms and to provide early
intervention for those at risk for adverse conditions. Whereas
some primary and specialty care physicians routinely ask
new patients about alcohol and illicit drug use, this does not
always entail quantication or use of a standard screening
instrument (Friedmann et al., 2001; Gunderson et al., 2005;
Maheux et al., 1999; Schermer et al., 2003). Questioning
about illicit drug use is more common in specialty areas
such as psychiatry and obstetrics/gynecology than in fam-
ily medicine practices (Friedmann et al., 2001), but even in
specialty settings, accurate and efcient assessment is often
lacking. To receive the highest level of accreditation from the
American College of Surgeons, trauma centers already are
required to identify alcohol problems (Gentilello, 2007), and
there have been proposals to extend this requirement to in-
clude identication of drug problems (Martins et al., 2007).
However, a recent study of emergency physicians showed
that only a minority routinely ask about illicit drug use, far
fewer than ask about tobacco or alcohol use (Williams et al.,
2000). Studies of primary and emergency physician attitudes
toward substance-misuse screening indicate concerns about
the sensitivity of the questions, inadequate training, and
the time required and lack of reimbursement for this effort
(Friedmann et al., 2001; Gentilello, 2005; Schermer et al.,
2003; Yoast et al., 2008). Concerns regarding asking about
or documenting an illegal practice also may deter physicians
from asking about drug use, whereas similar concerns would
not exist for questions about alcohol use, at least among
adults age 21 and older.
Although a number of brief substance use screeners exist,
the trend under the time constraints of clinical care is toward
ever-shorter screening instruments. A number of recent stud-
ies have examined a single-item screener asking about the
frequency of drinking ve or more (5+) drinks in a single
day for men or four or more (4+) drinks in a single day for
women. This single question has performed almost on a par
with longer instruments in screening for alcohol-use disor-
ders (AUDs) and hazardous drinking in general population,
ED, and patient samples (Dawson et al., 2010; Smith et al.,
2009; Stewart et al., 2008; Williams and Vinson, 2001). This
screener also showed surprisingly high levels of sensitivity
in screening for illicit drug use in a recent study of patients
in four trauma centers in Los Angeles County. Specically,
at a cutpoint of drinking 5+/4+ drinks once or more in the
past 30 days, this question had a sensitivity and specicity
of 74% and 57%, respectively, for past-year marijuana use;
67% and 67%, respectively, for past-year cocaine/crack use;
and 63% and 53%, respectively, for past-year illicit pre-
scription drug use. The overall percentage of cases screened
correctly varied from 62% to 68%, and it generally did a
better job of identifying drug users than nonusers (i.e., its
sensitivity exceeded its specicity; Ramchand et al., 2009).
Why might a question on heavy episodic drinking per-
form well as a screener for drug use and DUDs? The answer
lies in the strong associations between alcohol and drug use
and their associated disorders. Stinson et al. (2005) reported
that 55.2% of individuals with a DUD had an AUD, com-
pared with 7.5% of those without a DUD. Dawson et al.
(2008) demonstrated that the odds of incident drug use and
DUD increased in a linear manner with the frequency of
drinking 5+/4+ drinks, even after adjusting for a host of oth-
er risk factors, and Miller et al. (2007) found a similar linear
association between frequency of heavy episodic drinking
and current marijuana, cocaine, and inhalant use among high
school students. Likewise, heavy episodic drinking has been
linked with the frequency of marijuana use among young
women in a primary care sample (Rose et al., 2007) and with
a threefold to sixfold increase in the prevalence of marijuana
and other drug use in an ED sample (Fleming et al., 2007).
Thus, whereas it might seem surprising to ask about alcohol
use to screen for drug use and DUDs, the feasibility of such
a screener is supported by the strength of the associations
between alcohol and drug use.
The purpose of this study is to determine how well the
single-item 5+/4+ drinks screener works to identify persons
with illicit DUDs in a nationally representative sample of
U.S. adults, thus extending the research of Ramchand et
al. (2009), as well as in a subsample who reported having
gone to an ED for care at least once in the preceding year
(although not an ED sample per se). The study examines the
performance of the screener in terms of its ability to identify
past-year drug use, DUD (abuse or dependence), and drug
dependence for any illicit drug and for three specic drugs:
marijuana (cannabis), cocaine (including crack), and illicit
prescription drugs. These illicit substances were chosen be-
cause marijuana and cocaine are the most prevalent specic
drugs and because illicit prescription drugs are an emerging
problem in the United States (Compton and Volkow, 2006).
Method
Sample
This analysis is based on data from the 2001-2002 Wave
1 National Epidemiologic Survey on Alcohol and Related
Conditions (NESARC), designed by the National Institute
DAWSON, COMPTON, AND GRANT 753
on Alcohol Abuse and Alcoholism (Grant et al., 2003a).
The Wave 1 NESARC sample (N = 43,093, response rate =
81%) represented U.S. adults age 18 years or older residing
in households and selected noninstitutional group quarters
in all 50 states and the District of Columbia. The NESARC
data were weighted to reect design characteristics (includ-
ing oversampling of Blacks, Hispanics, and young adults)
and nonresponse. Weighted data were further adjusted to
match the civilian, noninstitutionalized population on so-
cioeconomic variables based on the 2000 U.S. census. Data
were collected in personal interviews using a computer-
assisted survey instrument administered by highly trained
and experienced lay interviewers. All potential respondents
were informed in writing about the nature of the survey,
statistical uses of the survey data, voluntary aspect of their
participation, and federal laws that rigorously provide for
the condentiality of identiable survey information. Only
respondents consenting to participate after receiving this in-
formation were interviewed. The research protocol, including
informed consent procedures, received a full ethical review
and approval from the U.S. Census Bureau and the U.S. Of-
ce of Management and Budget. The analyses for this study
were conducted in 2009, using the total NESARC sample
(N = 43,093) and a subsample of individuals who reported
having visited an ED for care in the 12 months preceding the
interview (n = 8,525).
Measures
The 5+/4+ drinking measure used in this analysis reects
past-year frequency of drinking 5+ drinks in a single day
for men and 4+ drinks in a single day for women, based on
all types of alcohol combined: During the last 12 months,
about how often did you drink (ve or more/four or more)
drinks in a single day? Response categories were (a) every
day, (b) nearly every day, (c) 3-4 times a week, (d) 2 times
a week, (e) once a week, (f) 2-3 times a month, (g) once a
month, (h) 7-11 times in the last year, (i) 3-6 times in the last
year, (j) 1-2 times in the last year, and (k) never in the last
year. These questions were asked only of current drinkers
who in initial screening questions reported having consumed
at least one alcoholic drink in the last 12 months, and among
these it was directed only to those whose largest quantity of
drinks consumed on any day was 5+/4+ or unknown, 5,596
men and 3,879 women. Frequency or 5+/4+ drinking was
automatically set to never for 16,147 past-year abstainers,
7,471 men whose largest quantity of drinks was four or less,
and 10,000 women whose largest quantity of drinks was
three or less. The questions on 5+/4+ drinks were preceded
by questions that asked for the overall frequency of drinking,
the usual and largest number of drinks consumed in a single
day, and the frequency of consuming the largest quantity.
In addition, the whole sequence of questions on all types of
alcohol combined was preceded by similar sets of questions
specic to drinking coolers, beer, wine, and distilled spirits.
In a random subsample of NESARC respondents reinter-
viewed approximately 10 weeks after the initial interview,
the intraclass correlation coefcient for test-retest reliability
was 0.69 for frequency of drinking 5+ drinks (Grant et al.,
2003b).
Drug use and drug-use disorders
The NESARC asked respondents if they ever used 10
categories of illicit drugs: sedatives, tranquilizers, opioids,
amphetamines, cannabis, cocaine including crack, hal-
lucinogens, inhalants/solvents, heroin, and all other drugs
combined. Illicit prescription drug use entailed use without
or beyond the limits of a prescription. Individuals who ever
used each drug type were asked: Did you use (drug type)
in the last 12 months only, before the last 2 months only, or
during both time periods? Past-year use comprised use in
the last 12 months only or during both time periods. DUDs
were dened in accordance with the criteria from the Diag-
nostic and Statistical Manual of Mental Disorders, Fourth
Edition (DSM-IV; American Psychiatric Association, 1994),
using the National Institute on Alcohol Abuse and Alcohol-
isms Alcohol Use Disorders and Associated Disabilities
Interview ScheduleDSM-IV Version (Grant et al., 2001),
a structured diagnostic interview designed to be administered
by lay interviewers. A total of 33 symptom item indicators
represented the seven dependence and four abuse criteria;
past-year symptoms were asked only of past-year users.
Respondents who endorsed a given past-year symptom were
asked to identify the drug(s) associated with the endorsed
symptom. To be classied with past-year drug dependence
for any specic drug, respondents had to meet at least three
dependence criteria for that drug in the 12 months preced-
ing interview. To be classied with any DUD, respondents
had to satisfy at least one abuse criterion or at least three
dependence criteria (Compton et al., 2007; Grant et al.,
2004). Any-drug measures (i.e., dependence, DUD, and
use of any drug) required that the behavior or condition be
positive for at least one of the 10 specic drug types; any
prescription drug measures required a positive behavior/
condition for at least one prescription drug type (sedatives,
tranquilizers, opioids, amphetamines). Test-retest reliability
of past-year DUDs ranged from = .79 for any DUD to .91
for cocaine-use disorder. The reliability of past-year drug
use varied from = .86 for cocaine/crack and .77 for mari-
juana to .50-.82 for various types of illicit prescription drugs
(Grant et al., 1995).
Analysis
Sensitivity, specicity, and positive predictive values were
based on weighted data generated for various screener cut-
points using SUDAAN (Research Triangle Institute, 2002),
754 JOURNAL OF STUDIES ON ALCOHOL AND DRUGS / SEPTEMBER 2010
T
A
B
L
E

1
.




P
a
s
t
-
y
e
a
r

p
r
e
v
a
l
e
n
c
e

o
f

i
l
l
i
c
i
t

d
r
u
g

d
e
p
e
n
d
e
n
c
e
,

d
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r
,

a
n
d

d
r
u
g

u
s
e
,

b
y

p
a
s
t
-
y
e
a
r

f
r
e
q
u
e
n
c
y

o
f

d
r
i
n
k
i
n
g

5
+
/
4
+

d
r
i
n
k
s

C
a
s
e
s
a

2
V
a
r
i
a
b
l
e

n

T
o
t
a
l

N
e
v
e
r

1

-
2

/
y
r
.

3

-
6

/
y
r
.

7

-
1
1

/
y
r
.

1
x
/
m
o
.

2

-
3

/
m
o
.

1

/
w
k
.

2

/
w
k
.

>
2

/
w
k
.

p
U
.
S
.

a
d
u
l
t
s

1
8

a
n
d

o
l
d
e
r
,
c
a
s
e
s
,
a
n

4
2
,
9
2
3

3
3
,
6
8
4

1
,
6
3
3

1
,
3
9
5

7
0
8

9
8
9

1
,
1
1
5

1
,
0
7
6

9
1
2

1
,
4
1
1

D
r
u
g

d
e
p
e
n
d
e
n
c
e


A
n
y

i
l
l
i
c
i
t

d
r
u
g

2
4
8

0
.
6

(
0
.
1
)

0
.
2

(
0
.
0
)

0
.
9

(
0
.
2
)

0
.
3

(
0
.
2
)

1
.
0

(
0
.
4
)

1
.
3

(
0
.
4
)

1
.
8

(
0
.
5
)

1
.
8

(
0
.
7
)

2
.
6

(
0
.
7
)

5
.
1

(
0
.
7
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

1
3
2

0
.
3

(
0
.
0
)

0
.
1

(
0
.
0
)

0
.
7

(
0
.
2
)

0
.
2

(
0
.
2
)

0
.
9

(
0
.
4
)

0
.
9

(
0
.
3
)

0
.
8

(
0
.
3
)

1
.
3

(
0
.
7
)

1
.
7

(
0
.
5
)

2
.
2

(
0
.
5
)

<
.
0
0
1


C
o
c
a
i
n
e
/
c
r
a
c
k

4
9

0
.
1

(
0
.
0
)

<
0
.
1

(
0
.
0
)

0
.
1

(
0
.
0
)

<
0
.
1

(
0
.
0
)

0
.
0

(
0
.
0
)

0
.
4

(
0
.
2
)

0
.
4

(
0
.
2
)

0
.
3

(
0
.
2
)

0
.
7

(
0
.
5
)

1
.
5

(
0
.
3
)

.
0
0
1


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

8
8

0
.
2

(
0
.
0
)

0
.
1

(
0
.
0
)

0
.
3

(
0
.
1
)

0
.
1

(
0
.
1
)

0
.
3

(
0
.
2
)

0
.
3

(
0
.
2
)

0
.
7

(
0
.
4
)

0
.
7

(
0
.
3
)

0
.
2

(
0
.
1
)

1
.
9

(
0
.
4
)

.
0
0
4

D
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r


A
n
y

i
l
l
i
c
i
t

d
r
u
g

7
7
5

2
.
0

(
0
.
1
)

0
.
7

(
0
.
1
)

2
.
6

(
0
.
4
)

2
.
9

(
0
.
5
)

5
.
1

(
0
.
9
)

3
.
9

(
0
.
6
)

6
.
9

(
1
.
0
)

5
.
9

(
0
.
1
)

9
.
9

(
1
.
3
)

1
2
.
2

(
1
.
1
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

5
5
8

1
.
5

(
0
.
1
)

0
.
4

(
0
.
0
)

2
.
2

(
0
.
4
)

2
.
3

(
0
.
5
)

4
.
0

(
0
.
8
)

3
.
3

(
0
.
6
)

5
.
2

(
0
.
8
)

4
.
6

(
0
.
9
)

7
.
6

(
1
.
1
)

9
.
1

(
0
.
9
)

<
.
0
0
1


C
o
c
a
i
n
e
/
c
r
a
c
k

9
7

0
.
3

(
0
.
0
)

0
.
1

(
0
.
0
)

0
.
1

(
0
.
1
)

0
.
1

(
0
.
0
)

0
.
3

(
0
.
2
)

0
.
7

(
0
.
3
)

0
.
9

(
0
.
3
)

0
.
5

(
0
.
3
)

1
.
4

(
0
.
5
)

3
.
0

(
0
.
5
)

<
.
0
0
1


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

2
3
4

0
.
6

(
0
.
1
)

0
.
3

(
0
.
0
)

0
.
5

(
0
.
2
)

0
.
6

(
0
.
2
)

1
.
1

(
0
.
4
)

0
.
7

(
0
.
3
)

1
.
8

(
0
.
6
)

1
.
7

(
0
.
4
)

1
.
6

(
0
.
7
)

4
.
4

(
0
.
7
)

<
.
0
0
1

D
r
u
g

u
s
e


A
n
y

i
l
l
i
c
i
t

d
r
u
g

2
,
4
5
7

6
.
2

(
0
.
2
)

3
.
0

(
0
.
1
)

9
.
4

(
0
.
9
)

1
0
.
8

(
1
.
1
)

1
2
.
5

(
1
.
4
)

1
4
.
6

(
1
.
4
)

1
8
.
3

(
1
.
2
)

1
6
.
7

(
1
.
4
)

2
5
.
5

(
2
.
0
)

2
6
.
9

(
1
.
6
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

1
,
6
1
0

4
.
1

(
0
.
2
)

1
.
5

(
0
.
1
)

7
.
4

(
0
.
8
)

6
.
9

(
0
.
8
)

9
.
1

(
1
.
2
)

1
0
.
5

(
1
.
3
)

1
3
.
6

(
1
.
0
)

1
2
.
5

(
1
.
3
)

1
2
.
6

(
1
.
3
)

2
1
.
2

(
1
.
5
)

<
.
0
0
1


C
o
c
a
i
n
e
/
c
r
a
c
k

2
1
7

0
.
6

(
0
.
0
)

0
.
1

(
0
.
0
)

0
.
3

(
0
.
2
)

0
.
3

(
0
.
1
)

1
.
0

(
0
.
4
)

1
.
4

(
0
.
4
)

1
.
7

(
0
.
4
)

1
.
8

(
0
.
5
)

3
.
4

(
0
.
8
)

5
.
6

(
0
.
8
)

<
.
0
0
1


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

1
,
2
2
8

3
.
2

(
0
.
1
)

1
.
8

(
0
.
1
)

3
.
5

(
0
.
5
)

4
.
3

(
0
.
7
)

5
.
9

(
1
.
1
)

6
.
8

(
0
.
9
)

8
.
9

(
1
.
1
)

8
.
0

(
1
.
1
)

1
0
.
6

(
1
.
5
)

1
3
.
9

(
1
.
2
)

<
.
0
0
1
U
.
S
.

a
d
u
l
t
s

1
8

a
n
d

o
l
d
e
r

w
i
t
h
p
a
s
t
-
y
e
a
r

E
D

u
t
i
l
i
z
a
t
i
o
n
,
c
a
s
e
s
,
n

8
,
5
1
1

6
,
5
9
9

3
5
1

2
4
6

1
4
7

1
9
4

2
0
7

2
3
4

1
9
8

3
3
5


D
r
u
g

d
e
p
e
n
d
e
n
c
e


A
n
y

i
l
l
i
c
i
t

d
r
u
g

1
0
7

1
.
6

(
0
.
2
)

0
.
6

(
0
.
0
)

2
.
0

(
0
.
9
)

0
.
8

(
0
.
7
)

1
.
4

(
1
.
1
)

2
.
4

(
1
.
1
)

5
.
6

(
2
.
4
)

4
.
6

(
2
.
4
)

4
.
0

(
1
.
8
)

1
1
.
0

(
2
.
1
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

4
8

0
.
7

(
0
.
2
)

0
.
2

(
0
.
1
)

1
.
1

(
0
.
7
)

0
.
7

(
0
.
6
)

0
.
9

(
0
.
9
)

1
.
3

(
0
.
9
)

2
.
7

(
1
.
4
)

4
.
0

(
2
.
3
)

2
.
1

(
0
.
9
)

4
.
1

(
1
.
2
)

.
0
0
6


C
o
c
a
i
n
e
/
c
r
a
c
k

1
9

0
.
4

(
0
.
1
)

0
.
1

(
0
.
1
)

0
.
5

(
0
.
5
)

0
.
0

(
0
.
0
)

0
.
0

(
0
.
0
)

0
.
6

(
0
.
6
)

0
.
4

(
0
.
4
)

0
.
9

(
0
.
9
)

1
.
6

(
1
.
6
)

3
.
6

(
1
.
2
)

.
0
8
8


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

5
1

0
.
8

(
0
.
1
)

0
.
4

(
0
.
1
)

0
.
9

(
0
.
6
)

0
.
1

(
0
.
1
)

1
.
4

(
1
.
0
)

1
.
4

(
0
.
9
)

2
.
6

(
2
.
0
)

1
.
5

(
1
.
0
)

0
.
3

(
0
.
3
)

4
.
4

(
1
.
3
)

.
0
3
0

D
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r


A
n
y

i
l
l
i
c
i
t

d
r
u
g

2
5
4

3
.
4

(
0
.
3
)

1
.
3

(
0
.
2
)

3
.
0

(
1
.
0
)

1
.
9

(
0
.
9
)

9
.
8

(
3
.
0
)

4
.
8

(
1
.
6
)

1
2
.
3

(
3
.
0
)

8
.
2

(
2
.
6
)

1
6
.
7

(
3
.
6
)

1
9
.
3

(
2
.
6
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

1
6
8

2
.
3

(
0
.
2
)

0
.
7

(
0
.
1
)

2
.
1

(
0
.
9
)

1
.
5

(
0
.
8
)

8
.
2

(
3
.
0
)

3
.
1

(
1
.
2
)

8
.
5

(
2
.
3
)

7
.
3

(
2
.
5
)

1
0
.
4

(
2
.
6
)

1
3
.
1

(
2
.
1
)

<
.
0
0
1


C
o
c
a
i
n
e
/
c
r
a
c
k

3
5

0
.
6

(
0
.
1
)

0
.
2

(
0
.
1
)

0
.
6

(
0
.
5
)

0
.
0

(
0
.
0
)

0
.
6

(
0
.
5
)

0
.
6

(
0
.
6
)

0
.
4

(
0
.
4
)

1
.
0

(
0
.
9
)

2
.
8

(
1
.
8
)

5
.
9

(
1
.
5
)

.
0
1
6


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

1
0
6

1
.
5

(
0
.
2
)

0
.
7

(
0
.
2
)

1
.
4

(
0
.
7
)

0
.
4

(
0
.
3
)

2
.
4

(
1
.
2
)

2
.
4

(
1
.
2
)

4
.
1

(
2
.
2
)

2
.
7

(
1
.
2
)

4
.
4

(
2
.
5
)

9
.
0

(
2
.
0
)

.
0
0
1

D
r
u
g

u
s
e


A
n
y

i
l
l
i
c
i
t

d
r
u
g

6
8
0

8
.
9

(
0
.
4
)

4
.
2

(
0
.
3
)

1
2
.
9

(
2
.
1
)

9
.
8

(
2
.
4
)

1
9
.
6

(
3
.
9
)

2
3
.
6

(
3
.
7
)

2
3
.
6

(
3
.
6
)

2
1
.
3

(
3
.
4
)

3
2
.
0

(
4
.
4
)

3
7
.
6

(
3
.
6
)

<
.
0
0
1


M
a
r
i
j
u
a
n
a

4
3
4

5
.
9

(
0
.
4
)

2
.
1

(
0
.
2
)

9
.
8

(
1
.
8
)

5
.
7

(
1
.
7
)

1
7
.
3

(
3
.
7
)

1
4
.
5

(
3
.
3
)

1
6
.
4

(
2
.
9
)

1
6
.
0

(
3
.
2
)

2
8
.
1

(
4
.
7
)

2
7
.
8

(
3
.
2
)

<
.
0
0
1


C
o
c
a
i
n
e
/
c
r
a
c
k

6
8

1
.
1

(
0
.
2
)

0
.
4

(
0
.
1
)

0
.
7

(
0
.
5
)

0
.
0

(
0
.
0
)

1
.
9

(
0
.
9
)

2
.
2

(
1
.
2
)

1
.
6

(
0
.
9
)

2
.
5

(
1
.
3
)

4
.
0

(
2
.
0
)

9
.
8

(
2
.
0
)

.
0
0
1


A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

3
7
7

5
.
1

(
0
.
3
)

2
.
7

(
0
.
3
)

6
.
8

(
1
.
6
)

3
.
6

(
1
.
7
)

8
.
8

(
2
.
8
)

1
3
.
0

(
3
.
3
)

1
2
.
7

(
3
.
1
)

9
.
1

(
2
.
2
)

1
5
.
1

(
3
.
6
)

2
2
.
5

(
3
.
0
)

<
.
0
0
1
N
o
t
e
s
:

F
i
g
u
r
e
s

i
n

p
a
r
e
n
t
h
e
s
e
s

a
r
e

s
t
a
n
d
a
r
d

e
r
r
o
r
s

o
f

e
s
t
i
m
a
t
e
s
;

s
t
a
n
d
a
r
d

e
r
r
o
r
s

o
f

0
.
0

d
e
n
o
t
e

<
.
0
5
.

Y
r
.

=

y
e
a
r
;

m
o
.

=

m
o
n
t
h
;

w
k
.

=

w
e
e
k
;

r
x

=

p
r
e
s
c
r
i
p
t
i
o
n
;

E
D

=

e
m
e
r
g
e
n
c
y

d
e
p
a
r
t
m
e
n
t
.

a
N
u
m
b
e
r
s

o
f

c
a
s
e
s

r
e

e
c
t

m
i
s
s
i
n
g

d
a
t
a

o
n

f
r
e
q
u
e
n
c
y

o
f

d
r
i
n
k
i
n
g

5
+
/
4
+

d
r
i
n
k
s

f
o
r

1
7
0

c
a
s
e
s

(
0
.
4
%
)

i
n

t
h
e

t
o
t
a
l

s
a
m
p
l
e

a
n
d

1
4

c
a
s
e
s

(
0
.
2
%
)

i
n

t
h
e

E
D

s
u
b
s
a
m
p
l
e
.
F
r
e
q
u
e
n
c
y

o
f

d
r
i
n
k
i
n
g

5
+
/
4
+

d
r
i
n
k
s
DAWSON, COMPTON, AND GRANT 755
a software package that uses Taylor series linearization to
adjust variance estimates for complex, multistage sample
designs. At any given cutpoint, sensitivity reects the pro-
portion of individuals actually positive for the condition of
interest whose screener score was greater than or equal to the
cutpoint, and specicity reects the proportion of individuals
actually negative for the condition whose screener score was
lower than the cutpoint. The positive predictive value is the
proportion of those with a positive screen who are positive
for the condition of interest. As a measure of screening ef-
ciency, the positive likelihood ratio, which is the ratio of true
to false positives (sensitivity divided by 1 minus specicity),
was calculated for each cutpoint. Areas under receiver-opera-
tor characteristic curves (AUCs) that plot sensitivity versus 1
minus specicity at each screener cutpoint were calculated as
measures of overall performance (Swets and Pickett, 1982)
using ROCKIT 0.9B (Metz, 2003) for maximum-likelihood
estimates of semiparametric binormal curve AUC.
Results
As shown in Table 1, less than 1% (0.6%) of U.S. adults
met the criteria for past-year drug dependence, 2.0% had a
past-year DUD (dependence and/or abuse), and 6.2% were
past-year illicit drug users. Prevalence rates for specic
drugs were highest for marijuana, next highest for any il-
licit prescription drug, and lowest for cocaine/crack. In the
subpopulation of adults who had used the services of an ED
in the year preceding interview, rates of drug use and DUDs
were slightly higher than in the general population. In both
the general population and among those who had used an
ED, the rates of drug use and DUDs increased as a function
of past-year frequency of consuming 5+/4+ drinks, and these
associations were highly signicant as indicated by the p for
chi-square tests of association, p < .001 in most cases and p
>.05 only for cocaine/crack dependence in the ED subpopu-
lation (p = .088).
The performance of the single item 5+/4+ drinks screener
varied as a function of drug type (Table 2). Performance was
best for cocaine (AUC = .887-.897) and marijuana (AUC
= .839-.854), worst for the pooled category of any illicit
prescription drug (AUC = .748-.766), and intermediate for
the pooled category of any drug (AUC = .799-.833). Within
each drug category, differences in performance according to
the specic gold standard being considered were small and
within sampling error (i.e., their standard errors indicated
overlapping 95% condence intervals) but suggested that
the screener might be slightly more accurate in screening for
DUDs rather than drug use. Positive predictive values were
low, especially for cocaine/crack, reecting the low preva-
lence of illicit drug use and DUDs in the general U.S. adult
population, even among those engaging in heavy episodic
drinking.
For the drug categories of any drug, marijuana, and any
illicit prescription drug, the optimal screening cutpoint was
drinking 5+/4+ drinks once or more a year. A cutpoint of
drinking 5+/4+ drinks three or more times a year was a
reasonable alternative for situations that favor specicity
over sensitivity (e.g., when the costs of an increase of ap-
proximately 5% in false positives outweigh the value of
identifying an additional 5%-10% of true positives). For the
any-drug and marijuana measures, achieving a specicity of
80% or more generally required accepting a sensitivity of
less than 70% (i.e., of identifying less than 70% of the indi-
viduals truly positive for the drug use or DUD in question),
and positive likelihood ratios were generally in the range
of 3 to 4 for the optimal cutpoints. That is, these cutpoints
would identify three to four times as many true positives as
false positives. For the any illicit prescription drug category,
the positive likelihood ratios were less than 3 at the optimal
cutpoints, indicating a low level of screening efciency, and
all specicities were associated with sensitivities of 60% or
less. In other words, the screener was incapable of identify-
ing more than 60% of the individuals with illicit prescription
drug use or DUDs at any cutpoint.
For cocaine, the optimal screening cutpoint for depen-
dence was drinking 5+/4+ drinks once or more a month. The
cutpoints of once or more a month and seven or more times
a year performed equally well in screening for any cocaine
DUD, and a cutpoint of seven or more times a year was op-
timal for any cocaine use. At the cutpoint of once or more
a month, sensitivity and specicity were 76.0% and 86.0%,
respectively, for cocaine dependence and 73.9% and 86.1%,
respectively, for any cocaine-use disorder. At a cutpoint of
seven or more times a year, sensitivity and specicity for
cocaine use were 77.6% and 84.5%, respectively.
Among individuals who went to an ED for care in the
past year (Table 3), the single-item screener performed
almost as accurately as in the general population and more
so for some drugs and/or target conditions. All differences
between the general population and ED subsample were
small and lay within sampling error; moreover, the optimal
screening cutpoints were the same in the two samples.
Discussion
Data from a nationally representative sample of U.S.
adults revealed that a single question about past-year fre-
quency of drinking 5+/4+ drinks performed well as a screen-
er for past-year marijuana- and cocaine-use disorders, as well
as for monthly or more frequent use of these drugs. It was
considerably less accurate in screening for illicit prescription
drug use or disorders. This may reect the fact that alcohol
use is contraindicated when using some types of prescription
drugs, or perhaps it indicates that the prescription-drug-use
phenotype has a different set of predictors than alcohol and
the other DUDs (Colliver et al., 2006). The particularly
strong performance of the single-item screener in predict-
756 JOURNAL OF STUDIES ON ALCOHOL AND DRUGS / SEPTEMBER 2010
D
r
u
g

d
e
p
e
n
d
e
n
c
e
D
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r

(
a
b
u
s
e

a
n
d
/
o
r

d
e
p
e
n
d
e
n
c
e
)
D
r
u
g

u
s
e
T
A
B
L
E

2
.




P
e
r
f
o
r
m
a
n
c
e

o
f

s
i
n
g
l
e
-
q
u
e
s
t
i
o
n

s
c
r
e
e
n
e
r

f
o
r

p
a
s
t
-
y
e
a
r

d
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r

a
n
d

d
r
u
g

u
s
e
:

U
.
S
.

a
d
u
l
t
s

1
8

y
e
a
r
s

o
f

a
g
e

a
n
d

o
l
d
e
r
F
r
e
q
.

o
f

d
r
i
n
k
i
n
g
5
+
/
4
+

d
r
i
n
k
s

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R
A
n
y

i
l
l
i
c
i
t

d
r
u
g

O
n
c
e

a

y
e
a
r

7
2
.
4

(
3
.
3
)

7
6
.
6

(
0
.
4
)

1
.
9

(
0
.
1
)

3
.
1

(
0
.
2
)

7
1
.
9

(
2
.
2
)

7
7
.
3

(
0
.
4
)

6
.
1

(
0
.
3
)

3
.
2

(
0
.
1
)

6
3
.
3

(
1
.
4
)

7
8
.
9

(
0
.
1
)

1
6
.
5

(
0
.
6
)

3
.
0

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

6
5
.
9

(
3
.
5
)

8
0
.
9

(
0
.
4
)

2
.
1

(
0
.
2
)

3
.
5

(
0
.
2
)

6
6
.
3

(
2
.
1
)

8
1
.
5

(
0
.
4
)

6
.
8

(
0
.
4
)

3
.
6

(
0
.
1
)

5
6
.
8

(
1
.
3
)

8
3
.
1

(
0
.
4
)

1
8
.
2

(
0
.
6
)

3
.
4

(
0
.
1
)

7

t
i
m
e
s

a

y
e
a
r

6
4
.
2

(
3
.
6
)

8
4
.
4

(
0
.
3
)

2
.
5

(
0
.
2
)

4
.
1

(
0
.
2
)

6
1
.
2

(
2
.
3
)

8
5
.
0

(
0
.
3
)

7
.
7

(
0
.
5
)

4
.
1

(
0
.
2
)

5
0
.
7

(
1
.
3
)

8
6
.
4

(
0
.
3
)

1
9
.
8

(
0
.
7
)

3
.
7

(
0
.
1
)

O
n
c
e

a

m
o
n
t
h

6
1
.
3

(
3
.
7
)

8
6
.
2

(
0
.
3
)

2
.
7

(
0
.
3
)

4
.
4

(
0
.
3
)

5
6
.
5

(
2
.
4
)

8
6
.
8

(
0
.
3
)

8
.
0

(
0
.
5
)

4
.
3

(
0
.
2
)

4
7
.
0

(
1
.
3
)

8
8
.
1

(
0
.
3
)

2
0
.
8

(
0
.
8
)

3
.
9

(
0
.
1
)

T
w
i
c
e

a

m
o
n
t
h

5
6
.
1

(
3
.
7
)

8
8
.
8

(
0
.
3
)

3
.
0

(
0
.
3
)

5
.
0

(
0
.
4
)

5
1
.
5

(
2
.
3
)

8
9
.
3

(
0
.
2
)

8
.
9

(
0
.
6
)

4
.
8

(
0
.
2
)

4
1
.
1

(
1
.
3
)

9
0
.
4

(
0
.
2
)

2
2
.
1

(
0
.
9
)

4
.
3

(
0
.
2
)

O
n
c
e

a

w
e
e
k

4
7
.
8

(
3
.
9
)

9
1
.
5

(
0
.
2
)

3
.
4

(
0
.
4
)

5
.
6

(
0
.
5
)

4
2
.
0

(
2
.
3
)

9
1
.
9

(
0
.
2
)

9
.
6

(
0
.
7
)

5
.
2

(
0
.
3
)

3
3
.
0

(
1
.
2
)

9
2
.
8

(
0
.
2
)

2
3
.
4

(
1
.
0
)

4
.
6

(
0
.
2
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
2
6
,

9
5
%

C
I

[
.
7
9
5
,

.
8
5
4
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
3
3
,

9
5
%

C
I

[
.
8
1
8
,

.
8
4
7
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
9
9
,

9
5
%

C
I

[
.
7
8
9
,

8
1
0
]
M
a
r
i
j
u
a
n
a

O
n
c
e

a

y
e
a
r

7
8
.
3

(
3
.
9
)

7
6
.
5

(
0
.
4
)

1
.
1

(
0
.
1
)

3
.
3

(
0
.
2
)

7
7
.
0

(
2
.
4
)

7
7
.
1

(
0
.
4
)

4
.
7

(
0
.
3
)

3
.
4

(
0
.
1
)

7
2
.
7

(
1
.
5
)

7
8
.
4

(
0
.
4
)

1
2
.
6

(
0
.
5
)

3
.
4

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

6
9
.
5

(
4
.
3
)

8
0
.
7

(
0
.
4
)

1
.
2

(
0
.
2
)

3
.
6

(
0
.
2
)

7
0
.
4

(
2
.
4
)

8
1
.
3

(
0
.
4
)

5
.
3

(
0
.
3
)

3
.
8

(
0
.
2
)

6
5
.
0

(
1
.
4
)

8
2
.
5

(
0
.
4
)

1
3
.
8

(
0
.
5
)

3
.
7

(
0
.
1
)

7

t
i
m
e
s

a

y
e
a
r

6
7
.
1

(
4
.
6
)

8
4
.
3

(
0
.
3
)

1
.
4

(
0
.
2
)

4
.
3

(
0
.
3
)

6
4
.
7

(
2
.
5
)

8
4
.
8

(
0
.
3
)

5
.
9

(
0
.
4
)

4
.
2

(
0
.
2
)

5
9
.
0

(
1
.
6
)

8
6
.
0

(
0
.
3
)

1
5
.
3

(
0
.
6
)

4
.
2

(
0
.
1
)

O
n
c
e

a

m
o
n
t
h

6
2
.
1

(
5
.
0
)

8
6
.
1

(
0
.
3
)

1
.
4

(
0
.
2
)

4
.
5

(
0
.
4
)

5
9
.
7

(
2
.
7
)

8
6
.
6

(
0
.
3
)

6
.
2

(
0
.
4
)

4
.
3

(
0
.
2
)

5
5
.
0

(
1
.
7
)

8
7
.
7

(
0
.
3
)

1
6
.
1

(
0
.
6
)

4
.
5

(
0
.
2
)

T
w
i
c
e

a

m
o
n
t
h

5
5
.
2

(
5
.
0
)

8
8
.
6

(
0
.
3
)

1
.
5

(
0
.
2
)

4
.
8

(
0
.
5
)

5
3
.
9

(
2
.
7
)

8
9
.
1

(
0
.
2
)

6
.
8

(
0
.
5
)

4
.
9

(
0
.
3
)

4
8
.
5

(
1
.
6
)

9
0
.
1

(
0
.
2
)

1
7
.
3

(
0
.
7
)

4
.
9

(
0
.
2
)

O
n
c
e

a

w
e
e
k

4
8
.
7

(
5
.
3
)

9
1
.
4

(
0
.
2
)

1
.
8

(
0
.
3
)

5
.
7

(
0
.
6
)

4
4
.
0

(
2
.
7
)

9
1
.
8

(
0
.
2
)

7
.
3

(
0
.
6
)

5
.
4

(
0
.
4
)

3
9
.
4

(
1
.
5
)

9
2
.
5

(
0
.
2
)

1
8
.
5

(
0
.
9
)

5
.
3

(
0
.
2
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
5
1
,

9
5
%

C
I

[
.
8
1
9
,

.
8
8
0
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
5
4
,

9
5
%

C
I

[
.
8
3
9
,

.
8
6
7
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
3
9
,

9
5
%

C
I

[
.
8
2
9
,

.
8
4
8
]
C
o
c
a
i
n
e
/
c
r
a
c
k

O
n
c
e

a

y
e
a
r

7
9
.
7

(
7
.
8
)

7
6
.
4

(
0
.
5
)

0
.
5

(
0
.
1
)

3
.
4

(
0
.
3
)

7
9
.
2

(
5
.
7
)

7
6
.
4

(
0
.
4
)

0
.
9

(
0
.
1
)

3
.
4

(
0
.
2
)

8
2
.
3

(
3
.
5
)

7
6
.
6

(
0
.
4
)

1
.
9

(
0
.
2
)

3
.
5

(
0
.
2
)

3

t
i
m
e
s

a

y
e
a
r

7
6
.
6

(
8
.
1
)

8
0
.
6

(
0
.
4
)

0
.
5

(
0
.
1
)

4
.
0

(
0
.
4
)

7
6
.
8

(
5
.
8
)

8
0
.
7

(
0
.
4
)

1
.
1

(
0
.
1
)

4
.
0

(
0
.
3
)

7
9
.
7

(
3
.
6
)

8
0
.
9

(
0
.
4
)

2
.
3

(
0
.
2
)

4
.
2

(
0
.
2
)

7

t
i
m
e
s

a

y
e
a
r

7
6
.
0

(
8
.
1
)

8
4
.
2

(
0
.
3
)

0
.
6

(
0
.
1
)

4
.
8

(
0
.
5
)

7
6
.
0

(
5
.
8
)

8
4
.
3

(
0
.
3
)

1
.
3

(
0
.
2
)

4
.
8

(
0
.
4
)

7
7
.
6

(
3
.
7
)

8
4
.
5

(
0
.
3
)

2
.
7

(
0
.
3
)

5
.
0

(
0
.
3
)

O
n
c
e

a

m
o
n
t
h

7
6
.
0

(
8
.
1
)

8
6
.
0

(
0
.
3
)

0
.
7

(
0
.
1
)

5
.
4

(
0
.
6
)

7
3
.
9

(
5
.
9
)

8
6
.
1

(
0
.
3
)

1
.
4

(
0
.
2
)

5
.
3

(
0
.
4
)

7
4
.
5

(
3
.
8
)

8
6
.
3

(
0
.
3
)

3
.
0

(
0
.
3
)

5
.
4

(
0
.
3
)

T
w
i
c
e

a

m
o
n
t
h

6
8
.
3

(
8
.
1
)

8
8
.
5

(
0
.
3
)

0
.
8

(
0
.
2
)

5
.
9

(
0
.
7
)

6
7
.
6

(
5
.
9
)

8
8
.
6

(
0
.
3
)

1
.
6

(
0
.
2
)

5
.
9

(
0
.
5
)

6
8
.
2

(
3
.
8
)

8
8
.
8

(
0
.
3
)

3
.
3

(
0
.
3
)

6
.
1

(
0
.
4
)

O
n
c
e

a

w
e
e
k

6
0
.
0

(
7
.
9
)

9
1
.
3

(
0
.
2
)

0
.
9

(
0
.
2
)

6
.
9

(
0
.
9
)

5
8
.
2

(
6
.
0
)

9
1
.
4

(
0
.
2
)

1
.
8

(
0
.
3
)

6
.
8

(
0
.
7
)

5
9
.
7

(
4
.
0
)

9
1
.
5

(
0
.
2
)

3
.
8

(
0
.
4
)

7
.
0

(
0
.
5
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
8
7
,

9
5
%

C
I

[
.
8
3
0
,

.
9
2
9
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
9
7
,

9
5
%

C
I

[
.
8
6
4
,

.
9
2
3
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
9
3
,

9
5
%

C
I

[
.
8
7
4
,

.
9
1
0
]
A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

O
n
c
e

a

y
e
a
r

6
0
.
3

(
6
.
6
)

7
6
.
4

(
0
.
4
)

0
.
6

(
0
.
1
)

2
.
6

(
0
.
3
)

6
0
.
1

(
3
.
9
)

7
6
.
5

(
0
.
4
)

1
.
6

(
0
.
2
)

2
.
6

(
0
.
2
)

5
6
.
5

(
2
.
0
)

7
7
.
4

(
0
.
4
)


7
.
6

(
0
.
4
)

2
.
5

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

5
5
.
1

(
6
.
7
)

8
0
.
7

(
0
.
4
)

0
.
7

(
0
.
1
)

2
.
9

(
0
.
4
)

5
6
.
7

(
3
.
9
)

8
0
.
8

(
0
.
4
)

1
.
8

(
0
.
2
)

3
.
0

(
0
.
2
)

5
1
.
8

(
1
.
9
)

8
1
.
6

(
0
.
4
)


8
.
5

(
0
.
5
)

2
.
8

(
0
.
1
)

7

t
i
m
e
s

a

y
e
a
r

5
4
.
0

(
6
.
7
)

8
4
.
2

(
0
.
3
)

0
.
8

(
0
.
1
)

3
.
4

(
0
.
4
)

5
3
.
5

(
4
.
0
)

8
4
.
4

(
0
.
3
)

2
.
1

(
0
.
2
)

3
.
4

(
0
.
3
)

4
7
.
0

(
1
.
9
)

8
5
.
1

(
0
.
3
)


9
.
5

(
0
.
5
)

3
.
2

(
0
.
1
)

O
n
c
e

a

m
o
n
t
h

5
1
.
7

(
6
.
5
)

8
6
.
0

(
0
.
3
)

0
.
8

(
0
.
2
)

3
.
7

(
0
.
5
)

5
0
.
3

(
4
.
0
)

8
6
.
2

(
0
.
3
)

2
.
2

(
0
.
2
)

3
.
6

(
0
.
3
)

4
3
.
6

(
2
.
0
)

8
6
.
9

(
0
.
3
)


9
.
9

(
0
.
6
)

3
.
3

(
0
.
2
)

T
w
i
c
e

a

m
o
n
t
h

4
8
.
6

(
6
.
5
)

8
8
.
6

(
0
.
3
)

1
.
0

(
0
.
2
)

4
.
3

(
0
.
6
)

4
7
.
6

(
4
.
1
)

8
8
.
7

(
0
.
3
)

2
.
5

(
0
.
3
)

4
.
2

(
0
.
4
)

3
8
.
3

(
1
.
9
)

8
9
.
4

(
0
.
3
)

1
0
.
6

(
0
.
7
)

3
.
6

(
0
.
2
)

O
n
c
e

a

w
e
e
k

4
0
.
0

(
6
.
5
)

9
1
.
3

(
0
.
2
)

1
.
0

(
0
.
2
)

4
.
6

(
0
.
8
)

3
9
.
6

(
4
.
6
)

9
1
.
4

(
0
.
2
)

2
.
8

(
0
.
3
)

4
.
6

(
0
.
5
)

3
0
.
5

(
1
.
9
)

9
2
.
0

(
0
.
2
)

1
1
.
2

(
0
.
8
)

3
.
8

(
0
.
3
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
6
4
,

9
5
%

C
I

[
.
6
9
0
,

.
8
2
7
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
6
6
,

9
5
%

C
I

[
.
7
2
4
,

.
8
0
4
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
4
8
,

9
5
%

C
I

[
.
7
2
9
,

.
7
6
5
]
N
o
t
e
s
:

F
i
g
u
r
e
s

i
n

p
a
r
e
n
t
h
e
s
e
s

r
e
p
r
e
s
e
n
t

s
t
a
n
d
a
r
d

e
r
r
o
r
s

o
f

e
s
t
i
m
a
t
e
s
.

S
h
a
d
e
d

c
e
l
l
s

d
e
n
o
t
e

c
u
t
p
o
i
n
t
s

t
h
a
t

r
e
s
u
l
t

i
n

h
i
g
h
e
s
t

c
o
m
b
i
n
e
d

v
a
l
u
e
s

o
f

s
e
n
s
i
t
i
v
i
t
y

(
s
e
n
s
.
)

a
n
d

s
p
e
c
i

c
i
t
y

(
s
p
e
c
.
)
,

w
i
t
h

s
e
n
s
.

c
o
n
-
s
t
r
a
i
n
e
d

t
o

b
e

a
b
o
v
e

6
0
%

i
f

p
o
s
s
i
b
l
e
.

F
r
e
q
.

=

f
r
e
q
u
e
n
c
y
;

P
P
V

=

p
o
s
i
t
i
v
e

p
r
e
d
i
c
t
i
v
e

v
a
l
u
e
;

P
L
R

=

p
o
s
i
t
i
v
e

l
i
k
e
l
i
h
o
o
d

r
a
t
i
o
;

R
O
C

=

r
e
c
e
i
v
e
r

o
p
e
r
a
t
o
r

c
h
a
r
a
c
t
e
r
i
s
t
i
c
;

A
U
C

=

a
r
e
a

u
n
d
e
r

r
e
c
e
i
v
e
r
-
o
p
e
r
a
t
o
r

c
h
a
r
a
c
t
e
r
i
s
t
i
c

c
u
r
v
e
;

M
L
E

=

m
a
x
i
m
u
m

l
i
k
e
l
i
h
o
o
d

e
s
t
i
m
a
t
i
o
n
;

C
I

=

c
o
n

d
e
n
c
e

i
n
t
e
r
v
a
l
;

r
x

=

p
r
e
s
c
r
i
p
t
i
o
n
.
DAWSON, COMPTON, AND GRANT 757
D
r
u
g

d
e
p
e
n
d
e
n
c
e
D
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r

(
a
b
u
s
e

a
n
d
/
o
r

d
e
p
e
n
d
e
n
c
e
)
D
r
u
g

u
s
e
T
A
B
L
E

3
.




P
e
r
f
o
r
m
a
n
c
e

o
f

s
i
n
g
l
e
-
q
u
e
s
t
i
o
n

s
c
r
e
e
n
e
r

f
o
r

p
a
s
t
-
y
e
a
r

d
r
u
g
-
u
s
e

d
i
s
o
r
d
e
r

a
n
d

d
r
u
g

u
s
e
:

U
.
S
.

a
d
u
l
t
s

1
8

y
e
a
r
s

a
n
d

o
l
d
e
r

w
i
t
h

p
a
s
t
-
y
e
a
r

E
D

u
t
i
l
i
z
a
t
i
o
n
F
r
e
q
.

o
f

d
r
i
n
k
i
n
g
5
+
/
4
+

d
r
i
n
k
s

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R

%

S
e
n
s
.

%

S
p
e
c
.

%

P
P
V

P
L
R
A
n
y

i
l
l
i
c
i
t

d
r
u
g

O
n
c
e

a

y
e
a
r

7
0
.
4

(
5
.
5
)

7
4
.
9

(
0
.
7
)

4
.
3

(
0
.
6
)

2
.
8

(
0
.
2
)

7
1
.
5

(
3
.
5
)

7
5
.
8

(
0
.
7
)


9
.
5

(
0
.
8
)

3
.
0

(
0
.
2
)

6
5
.
4

(
2
.
3
)

7
8
.
1

(
0
.
7
)

2
2
.
6

(
1
.
3
)

3
.
0

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

6
4
.
4

(
5
.
6
)

7
9
.
6

(
0
.
7
)

4
.
9

(
0
.
7
)

3
.
2

(
0
.
3
)

6
7
.
4

(
3
.
5
)

8
0
.
6

(
0
.
7
)

1
0
.
9

(
1
.
0
)

3
.
5

(
0
.
2
)

5
8
.
5

(
2
.
3
)

8
2
.
6

(
0
.
6
)

2
4
.
8

(
1
.
5
)

3
.
4

(
0
.
2
)

7

t
i
m
e
s

a

y
e
a
r

6
2
.
7

(
5
.
9
)

8
3
.
0

(
0
.
6
)

5
.
6

(
0
.
8
)

3
.
7

(
0
.
4
)

6
5
.
6

(
3
.
6
)

8
3
.
9

(
0
.
6
)

1
2
.
6

(
1
.
1
)

4
.
1

(
0
.
3
)

5
4
.
9

(
2
.
4
)

8
5
.
9

(
0
.
6
)

2
7
.
6

(
1
.
7
)

3
.
9

(
0
.
2
)

O
n
c
e

a

m
o
n
t
h

6
1
.
1

(
5
.
8
)

8
4
.
9

(
0
.
5
)

6
.
1

(
0
.
9
)

4
.
0

(
0
.
4
)

6
0
.
1

(
3
.
7
)

8
5
.
7

(
0
.
5
)

1
2
.
9

(
1
.
2
)

4
.
2

(
0
.
3
)

5
0
.
7

(
2
.
4
)

8
7
.
6

(
0
.
5
)

2
8
.
6

(
1
.
8
)

4
.
1

(
0
.
3
)

T
w
i
c
e

a

m
o
n
t
h

5
7
.
1

(
5
.
8
)

8
7
.
5

(
0
.
5
)

6
.
9

(
1
.
1
)

4
.
6

(
0
.
5
)

5
6
.
4

(
3
.
8
)

8
8
.
3

(
0
.
5
)

1
4
.
6

(
1
.
4
)

4
.
8

(
0
.
4
)

4
3
.
7

(
2
.
3
)

8
9
.
8

(
0
.
5
)

2
9
.
6

(
2
.
0
)

4
.
3

(
0
.
3
)

O
n
c
e

a

w
e
e
k

4
7
.
8

(
6
.
0
)

9
0
.
1

(
0
.
5
)

7
.
2

(
1
.
3
)

4
.
8

(
0
.
7
)

4
6
.
8

(
4
.
1
)

9
0
.
7

(
0
.
4
)

1
5
.
1

(
1
.
7
)

5
.
0

(
0
.
5
)

3
6
.
7

(
2
.
3
)

9
2
.
0

(
0
.
4
)

3
1
.
1

(
2
.
2
)

4
.
6

(
0
.
4
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
0
6
,

9
5
%

C
I

[
.
7
5
2
,

.
8
5
2
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
2
3
,

9
5
%

C
I

[
.
7
9
3
,

.
8
5
0
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
0
5
,

9
5
%

C
I

[
.
7
8
4
,

.
8
2
4
]
M
a
r
i
j
u
a
n
a

O
n
c
e

a

y
e
a
r

7
7
.
9

(
6
.
8
)

7
4
.
6

(
0
.
7
)

2
.
2

(
0
.
5
)

3
.
1

(
0
.
3
)

7
6
.
2

(
4
.
0
)

7
5
.
4

(
0
.
7
)


6
.
7

(
0
.
7
)

3
.
1

(
0
.
2
)

7
3
.
8

(
2
.
5
)

7
7
.
2

(
0
.
7
)

1
6
.
9

(
1
.
2
)

3
.
2

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

7
0
.
7

(
7
.
0
)

7
9
.
3

(
0
.
7
)

2
.
5

(
0
.
5
)

3
.
4

(
0
.
4
)

7
1
.
8

(
3
.
9
)

8
0
.
1

(
0
.
7
)


7
.
7

(
0
.
8
)

3
.
6

(
0
.
2
)

6
6
.
0

(
2
.
6
0

8
1
.
7

(
0
.
6
)

1
8
.
5

(
1
.
4
)

3
.
6

(
0
.
2
)

7

t
i
m
e
s

a

y
e
a
r

6
7
.
7

(
8
.
1
)

8
2
.
6

(
0
.
6
)

2
.
8

(
0
.
6
)

3
.
9

(
0
.
5
)

6
9
.
6

(
4
.
1
)

8
3
.
4

(
0
.
6
)


8
.
9

(
1
.
0
)

4
.
2

(
0
.
3
)

6
2
.
8

(
2
.
7
)

8
5
.
1

(
0
.
6
)

2
0
.
9

(
1
.
5
)

4
.
2

(
0
.
2
)

O
n
c
e

a

m
o
n
t
h

6
5
.
4

(
8
.
2
)

8
4
.
5

(
0
.
5
)

3
.
0

(
0
.
7
)

4
.
2

(
0
.
5
)

6
2
.
7

(
4
.
4
)

8
5
.
2

(
0
.
5
)


9
.
0

(
1
.
0
)

4
.
2

(
0
.
3
)

5
7
.
2

(
2
.
8
)

8
6
.
7

(
0
.
5
)

2
1
.
3

(
1
.
7
)

4
.
3

(
0
.
3
)

T
w
i
c
e

a

m
o
n
t
h

6
0
.
7

(
8
.
2
)

8
7
.
2

(
0
.
5
)

3
.
4

(
0
.
8
)

4
.
7

(
0
.
7
)

5
9
.
1

(
4
.
6
)

8
7
.
9

(
0
.
5
)

1
0
.
2

(
1
.
2
)

4
.
9

(
0
.
4
)

5
0
.
6

(
2
.
8
)

8
9
.
2

(
0
.
5
)

2
2
.
6

(
1
.
8
)

4
.
7

(
0
.
3
)

O
n
c
e

a

w
e
e
k

5
1
.
1

(
8
.
4
)

8
9
.
8

(
0
.
5
)

3
.
5

(
1
.
0
)

5
.
0

(
0
.
9
)

4
9
.
2

(
4
.
9
)

9
0
.
4

(
0
.
5
)

1
0
.
6

(
1
.
4
)

5
.
1

(
0
.
6
)

4
3
.
2

(
2
.
8
)

9
1
.
5

(
0
.
4
)

2
4
.
2

(
2
.
1
)

5
.
1

(
0
.
4
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
5
1
,

9
5
%

C
I

[
.
8
0
0
,

.
8
9
3
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
5
3
,

9
5
%

C
I

[
.
8
2
5
,

.
8
7
8
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
4
6
,

9
5
%

C
I

[
.
8
2
7
,

.
8
6
3
]
C
o
c
a
i
n
e
/
c
r
a
c
k

O
n
c
e

a

y
e
a
r

8
0
.
8

(
1
3
.
2
)

7
4
.
4

(
0
.
7
)

1
.
1

(
0
.
3
)

3
.
2

(
0
.
5
)

7
3
.
7

(
1
1
.
2
)

7
4
.
5

(
0
.
7
)


1
.
7

(
0
.
4
)

2
.
9

(
0
.
4
)

7
4
.
8

(
7
.
7
)

7
4
.
7

(
0
.
7
)

3
.
1

(
0
.
6
)

3
.
0

(
0
.
3
)

3

t
i
m
e
s

a

y
e
a
r

7
4
.
7

(
1
3
.
6
)

7
9
.
1

(
0
.
7
)

1
.
2

(
0
.
4
)

3
.
6

(
0
.
7
)

6
8
.
7

(
1
1
.
3
)

7
9
.
2

(
0
.
7
)


2
.
0

(
0
.
5
)

3
.
3

(
0
.
6
)

7
1
.
6

(
7
.
8
)

7
9
.
5

(
0
.
7
)

3
.
7

(
0
.
7
)

3
.
5

(
0
.
4
)

7

t
i
m
e
s

a

y
e
a
r

7
4
.
7

(
1
3
.
6
)

8
2
.
4

(
0
.
6
)

1
.
5

(
0
.
4
)

4
.
2

(
0
.
8
)

6
8
.
7

(
1
1
.
3
)

8
2
.
5

(
0
.
6
)


2
.
3

(
0
.
6
)

3
.
9

(
0
.
7
)

7
1
.
6

(
7
.
8
)

8
2
.
8

(
0
.
6
)

4
.
3

(
0
.
8
)

4
.
2

(
0
.
5
)

O
n
c
e

a

m
o
n
t
h

7
4
.
7

(
1
3
.
6
)

8
4
.
4

(
0
.
5
)

1
.
7

(
0
.
5
)

4
.
8

(
0
.
9
)

6
6
.
8

(
1
1
.
2
)

8
4
.
5

(
0
.
5
)


2
.
5

(
0
.
6
)

4
.
3

(
0
.
7
)

6
8
.
3

(
7
.
9
)

8
4
.
7

(
0
.
5
)

4
.
6

(
0
.
9
)

4
.
5

(
0
.
5
)

T
w
i
c
e

a

m
o
n
t
h

7
0
.
1

(
1
3
.
7
)

8
7
.
0

(
0
.
5
)

1
.
9

(
0
.
6
)

5
.
4

(
1
.
1
)

6
4
.
1

(
1
1
.
2
)

8
7
.
1

(
0
.
5
)


2
.
9

(
0
.
7
)

5
.
0

(
0
.
9
)

6
2
.
8

(
7
.
6
)

8
7
.
4

(
0
.
5
)

5
.
1

(
1
.
0
)

5
.
0

(
0
.
6
)

O
n
c
e

a

w
e
e
k

6
7
.
4

(
1
3
.
9
)

8
9
.
7

(
0
.
5
)

2
.
2

(
0
.
7
)

6
.
5

(
1
.
4
)

6
2
.
5

(
1
1
.
4
)

8
9
.
8

(
0
.
5
)


3
.
6

(
0
.
9
)

6
.
1

(
1
.
2
)

5
8
.
9

(
7
.
6
)

9
0
.
0

(
0
.
5
)

6
.
0

(
1
.
2
)

5
.
9

(
0
.
8
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
7
5
,

9
5
%

C
I

[
.
7
6
2
,

.
9
4
4
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
8
0
,

9
5
%

C
I

[
.
8
1
2
,

.
9
2
9
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
8
8
0
,

9
5
%

C
I

[
.
8
4
0
,

.
9
1
3
]
A
n
y

i
l
l
i
c
i
t

r
x

d
r
u
g

O
n
c
e

a

y
e
a
r

5
7
.
7

(
9
.
1
)

7
4
.
5

(
0
.
7
)

1
.
7

(
0
.
4
)

2
.
3

(
0
.
4
)

6
2
.
5

(
6
.
0
)

7
4
.
8

(
0
.
7
)


3
.
6

(
0
.
5
)

2
.
5

(
0
.
2
)

5
9
.
8

(
3
.
1
)

7
6
.
0

(
0
.
7
)

1
1
.
7

(
0
.
9
)

2
.
5

(
0
.
1
)

3

t
i
m
e
s

a

y
e
a
r

5
2
.
2

(
9
.
0
)

7
9
.
2

(
0
.
7
)

1
.
9

(
0
.
4
)

2
.
5

(
0
.
4
)

5
7
.
9

(
6
.
0
)

7
9
.
5

(
0
.
7
)


4
.
0

(
0
.
6
)

2
.
8

(
0
.
3
)

5
3
.
5

(
3
.
1
)

8
0
.
7

(
0
.
6
)

1
2
.
8

(
1
.
1
)

2
.
8

(
0
.
2
)

7

t
i
m
e
s

a

y
e
a
r

5
1
.
6

(
9
.
0
)

8
2
.
5

(
0
.
6
)

2
.
2

(
0
.
5
)

2
.
9

(
0
.
5
)

5
7
.
0

(
6
.
0
)

8
2
.
8

(
0
.
6
)


4
.
7

(
0
.
7
)

3
.
3

(
0
.
4
)

5
1
.
1

(
3
.
2
)

8
4
.
0

(
0
.
6
)

1
4
.
5

(
1
.
2
)

3
.
2

(
0
.
2
)

O
n
c
e

a

m
o
n
t
h

4
8
.
1

(
8
.
7
)

8
4
.
4

(
0
.
5
)

2
.
2

(
0
.
6
)

3
.
1

(
0
.
6
)

5
3
.
9

(
6
.
0
)

8
4
.
7

(
0
.
5
)


5
.
0

(
0
.
8
)

3
.
5

(
0
.
4
)

4
7
.
8

(
3
.
2
)

8
5
.
8

(
0
.
5
)

1
5
.
2

(
1
.
3
)

3
.
4

(
0
.
3
)

T
w
i
c
e

a

m
o
n
t
h

4
3
.
1

(
8
.
6
)

8
7
.
0

(
0
.
5
)

2
.
5

(
0
.
6
)

3
.
3

(
0
.
7
)

4
9
.
5

(
6
.
2
)

8
7
.
4

(
0
.
5
)


5
.
5

(
1
.
0
)

3
.
9

(
0
.
5
)

4
0
.
9

(
3
.
1
)

8
8
.
3

(
0
.
5
)

1
5
.
7

(
1
.
5
)

3
.
5

(
0
.
3
)

O
n
c
e

a

w
e
e
k

3
4
.
1

(
8
.
1
)

8
9
.
6

(
0
.
5
)

2
.
4

(
0
.
6
)

3
.
3

(
0
.
8
)

4
2
.
1

(
6
.
4
)

8
9
.
9

(
0
.
5
)


5
.
9

(
1
.
1
)

4
.
2

(
0
.
7
)

3
4
.
3

(
3
.
2
)

9
0
.
7

(
0
.
4
)

1
6
.
4

(
1
.
7
)

3
.
7

(
0
.
4
)

A
r
e
a

u
n
d
e
r

R
O
C


c
u
r
v
e

(
A
U
C
)

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
4
6
,

9
5
%

C
I

[
.
6
5
4
,

.
8
2
9
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
6
5
,

9
5
%

C
I

[
.
7
0
1
,

.
8
2
1
]

M
L
E

b
i
n
o
r
m
a
l

A
U
C

=

.
7
5
3
,

9
5
%

C
I

[
.
7
1
9
,

.
7
8
5
]
N
o
t
e
s
:

F
i
g
u
r
e
s

i
n

p
a
r
e
n
t
h
e
s
e
s

r
e
p
r
e
s
e
n
t

s
t
a
n
d
a
r
d

e
r
r
o
r
s

o
f

e
s
t
i
m
a
t
e
s
.

S
h
a
d
e
d

c
e
l
l
s

d
e
n
o
t
e

c
u
t
p
o
i
n
t
s

t
h
a
t

r
e
s
u
l
t

i
n

h
i
g
h
e
s
t

c
o
m
b
i
n
e
d

v
a
l
u
e
s

o
f

s
e
n
s
i
t
i
v
i
t
y

(
s
e
n
s
.
)

a
n
d

s
p
e
c
i

c
i
t
y

(
s
p
e
c
.
)
,

w
i
t
h

s
e
n
s
.

c
o
n
s
t
r
a
i
n
e
d

t
o

b
e

a
b
o
v
e

6
0
%

i
f

p
o
s
s
i
b
l
e
.


F
r
e
q
.

=

f
r
e
q
u
e
n
c
y
;

P
P
V

=

p
o
s
i
t
i
v
e

p
r
e
d
i
c
t
i
v
e

v
a
l
u
e
;

P
L
R

=

p
o
s
i
t
i
v
e

l
i
k
e
l
i
h
o
o
d

r
a
t
i
o
;

R
O
C

=

r
e
c
e
i
v
e
r

o
p
e
r
a
t
o
r

c
h
a
r
a
c
t
e
r
i
s
t
i
c
;

A
U
C

=

a
r
e
a

u
n
d
e
r

r
e
c
e
i
v
e
r
-
o
p
e
r
a
t
o
r

c
h
a
r
a
c
t
e
r
i
s
t
i
c

c
u
r
v
e
;

M
L
E

=

m
a
x
i
m
u
m

l
i
k
e
l
i
h
o
o
d

e
s
t
i
m
a
t
i
o
n
;

C
I

=

c
o
n

d
e
n
c
e

i
n
t
e
r
v
a
l
;

r
x

=

p
r
e
s
c
r
i
p
t
i
o
n
.
758 JOURNAL OF STUDIES ON ALCOHOL AND DRUGS / SEPTEMBER 2010
ing cocaine-use disorder reects the fact that cocaine is the
DUD most strongly comorbid with AUD, with an odds ratio
of 19.2 (Stinson et al., 2005). Given that almost 80% of
individuals with a past-year cocaine-use disorder also had a
past-year AUD (Stinson et al., 2005), it is not surprising that
an AUD screener would also do a good job of screening for
cocaine-use disorder. The ndings are also supported by a
recent ED study that found a threefold to sixfold increase in
the risk of illicit drug use among men and women drinking
at risk levels consistent with the 5+/4+ denition used in this
study (Fleming et al., 2007).
The sensitivities and specicities found for the subsample
of individuals with past-year ED utilization in this study ex-
ceeded those reported in a recent study of the 5+/4+ drinks
screener in a sample of trauma center patients (Ramchand et
al., 2009). The weaker performance in that study may reect
the use of a past-30-day reference period for frequency of
drinking 5+/4+ drinks, which implies a screening cutpoint
of once a month or moremore frequent than the optimal
cutpoints for drugs other than cocaine in the current study.
One would also expect a stronger degree of association when
the time reference periods for the two are identical, as was
the case in the current study.
Not surprisingly, the frequency of drinking 5+/4+ drinks
performed less accurately in screening for DUDs than
AUDs. A previous investigation of the NESARC survey
data showed that drinking 5+/4+ drinks three or more times
a year had a sensitivity and specicity of 89.5% and 83.3%,
respectively, in relation to alcohol dependence (Dawson
et al., 2010). The present study showed that at similar lev-
els of specicity, the sensitivity of the screener for drug
dependence varied from 54.0% for illicit prescription drugs
to 76.0% for cocaine. The differential performance across
substances suggests that shared genetic and environmental
factors do not explain all of the variance in the alcohol
and drug problem behaviors and disorders. Rather, some
risk factors appear to be substance-specic and therefore
might best be addressed by means of substance-specic
screeners.
There are, however, many challenges in trying to com-
pare this study with evaluations of other screeners designed
specically to detect drug use and DUDs. These challenges
include different populations, different and often less rigor-
ous gold standards, different approaches to estimating AUC
statistics (e.g., nonparametric versus semiparametric) that
might yield slightly different values, and the fact that some
studies have used measures that combined AUDs and DUDs.
Although the World Health Organizations Alcohol, Smok-
ing and Substance Involvement Screening Test (ASSIST;
Humeniuk et al., 2008) has shown promise in populations
of primary care and drug treatment patients (Henrique et al.,
2004; Hides et al., 2009; Humeniuk et al., 2008; Newcombe
et al., 2005), most of the published studies have examined
the distinctions between use and abuse and between abuse
and dependence, measures not directly addressed in the pres-
ent study. The few studies that are more directly comparable
generally suggest that longer, drug-specic screeners yield
only slightly higher values of sensitivity and specicity than
those for the 5+/4+ drinks single-item screener. In a review
of studies examining the ve-item Severity of Dependence
Scale (Gossop et al., 1995) and the Problematic Use of
Marijuana Scale (Okulicz-Kozaryn, 2007) for screening
cannabis dependence in the general population, Piontek et
al. (2008) reported AUC values of .85 to .92, compared with
.851 in the current study. However, screening for marijuana
use using the 10-item Cannabis Use Disorders Identication
Test (Adamson and Sellman, 2003) in a clinical sample of
alcoholics resulted in AUC values of .63 to .76 (Annaheim et
al., 2008), lower than the AUC of .835 for use in the current
study. At a cutpoint of 2 or more, a cocaine-specic version
of the Severity of Dependence Scale resulted in a sensitivity
and specicity of 73% and 82%, respectively, in screening
for cocaine dependence in a cross-sectional survey of past-6-
month cocaine users (Kaye and Darke, 2002), compared with
76.0% and 86.0%, respectively, for drinking 5+/4+ drinks
once or more a month. However, at specicities greater
than 90%, the Severity of Dependence Scale screener had
higher levels of sensitivity than the single-item 5+/4+ drinks
screener. Although most published studies have not provided
standard errors for their screening measures, the fairly broad
standard errors for the ED subsample in this studya sam-
ple larger than those used in most prior studiessuggest that
few, if any, of the differences across screening instruments
would be statistically signicant.
The primary limitation of this study is the fact that both
the screening and gold standard measures were based on
respondents self-reports. Any broad tendency to withhold or
fully provide requested information might tend to upwardly
bias estimates of screening performance. Moreover, the
questions on which the single-item screener was based were
embedded in a long sequence of questions on past-year al-
cohol consumption, which may have increased the accuracy
of reporting relative to what would be obtained by actually
asking a single question on 5+/4+ drinking. Moreover, the
5+/4+ questions were not asked of all respondents but were
lled on the basis of responses to prior questions for the
majority of respondents. In addition, reporting of both 5+/4+
drinking and illicit drug use might be more honest in a con-
dential survey interview setting than in a medical setting
where the responses could be linked with individuals medi-
cal records, thus creating a bias toward better reporting than
what might be expected in a noncondential medical set-
ting. Another limitation is that the gold standard conditions
against which the screener was tested did not include drug
use with consequences that failed to meet the criteria for a
DUD. In addition, although the drug-use and DUD measures
showed generally good to excellent test-retest reliability, they
were not externally validated. Finally, whereas the NESARC
DAWSON, COMPTON, AND GRANT 759
identied past-year ED users, it did not identify past-year
primary care patients; thus, we were unable to test the single-
item screener in one of the subpopulations where it would
most likely be used. These limitations indicate the need for
caution in interpreting the results of this study and for repli-
cation in relevant subpopulations.
In summary, a single-item screener comprising the fre-
quency of drinking 5+/4+ drinks shows strong promise for
detecting marijuana and cocaine use and problems in gen-
eral population samples. Although not optimal as a screener
for illicit prescription drug use, it does a fairly good job of
screening for any drug use or DUD, suggesting that it may
also perform well for specic drugs not examined in this
study, including hallucinogens and inhalants. The virtues of
the screener include its brevity and its applicability across
drug types without the need for drug-specic wording. Ar-
guably, it is also less embarrassing to ask about a legal than
an illegal practice, and individuals may be more likely to
accurately report (and physicians to query) an activity for
which they are not at risk of legal penalties. Most impor-
tantly, its demonstrated ability to accurately screen for AUDs
and hazardous drinking (Dawson et al., 2010) means that
drug screening can be thought of as added value from an
item already likely to be asked in the interest of detecting
problem drinking. These ndings remind us of the inter-
related nature of all substance-use disorders. The highest
comorbidities for AUDs are typically with respect to other
substance-use disorders (Hasin et al., 2007), and AUDs are
very common among individuals with illicit DUDs (Comp-
ton et al., 2007).
Future investigation of the 5+/4+ drinks screener in pri-
mary care and ED samples should help to clarify its utility
in those settings. Future work might consider a simple two-
stage screening process for both alcohol and illicit drugs to
help busy clinicians rule out the large number of negative
cases while simultaneously identifying persons with prob-
lematic use. Using the alcohol consumption screener as a
starting point, follow-up questions for patients who screen
positive should include assessment of illicit and prescrip-
tion drug use as well as AUDs. The additional screening
questions may be very brief but are required to ascertain
related diagnoses and the degree of severity of involvement
with all substances. By ruling out the majority of patients
on the basis of the initial 5+/4+ drinks screen, incorporation
of additional second-stage screening questions would still
represent a reduction of the aggregate patient burden relative
to asking brief drug screeners such as the ASSIST or Drug
Abuse Screening Test10 (DAST-10) of the total patient
population, and the performance of the two approaches in
detecting drug use and DUDs could be compared. If the
promising performance of the 5+/4+ drinks screener is thus
borne out in practical application, this single-item screener
should be incorporated as a standard intake item for patients
seeking routine or emergency medical care.
References
Adamson, S. J., & Sellman, J. D. (2003). A prototype screening instrument
for cannabis use disorder: The Cannabis Use Disorders Identication
Test (CUDIT) in an alcohol-dependent clinical sample. Drug and Alco-
hol Review, 22, 309-315.
American Psychiatric Association. (1994). Diagnostic and statistical manual
of mental disorders (4th ed.). Washington, DC: Author.
Annaheim, B., Rehm, J., & Gmel, G. (2008). How to screen for problematic
cannabis use in population surveys: An evaluation of the Cannabis Use
Disorders Identication Test (CUDIT) in a Swiss sample of adolescents
and young adults. European Addiction Research, 14, 190-197.
Cartwright, W. S. (1999). Costs of drug abuse to society. Journal of Mental
Health Policy and Economics, 2, 133-134.
Cherpitel, C. J., & Ye, Y. (2008). Drug use and problem drinking associated
with primary care and emergency room utilization in the US general
population: Data from the 2005 national alcohol survey. Drug and
Alcohol Dependence, 97, 226-230.
Colliver, J. D., Compton, W. M., Gfroerer, J. C., & Condon, T. (2006).
Projecting drug use among aging baby boomers in 2020. Annals of
Epidemiology, 16, 257-265.
Compton, W. M., Thomas, Y. F., Stinson, F. S., & Grant, B. F. (2007). Preva-
lence, correlates, disability, and comorbidity of DSM-IV drug abuse
and dependence in the United States. Archives of General Psychiatry,
64, 566-576.
Compton, W. M., & Volkow, N. D. (2006). Abuse of prescription drugs
and the risk of addiction. Drug and Alcohol Dependence, 83 (Suppl.
1), S4-S7.
Cummings, G. E., Francescutti, L. H., Predy, G., & Cummings, G. (2006).
Health promotion and disease prevention in the emergency depart-
ment: A feasibility study. Canadian Journal of Emergency Medicine,
8, 100-105.
Dawson, D. A., Pulay, A. J., & Grant, B. F. (2010). A comparison of two
single-item screeners for alcohol use disorder. Alcoholism: Clinical and
Experimental Research, 34, 364-374.
Dawson, D. A., Li, T.-K., & Grant, B. F. (2008). A prospective study of risk
drinking: At risk for what? Drug and Alcohol Dependence, 95, 62-72.
Fleming, E. A., Gmel, G., Bady, P., Yersin, B., Givel, J. C., Brown, D., &
Daeppen, J.-B. (2007). At-risk drinking and drug use among patients
seeking care in an emergency department. Journal of Studies on Alcohol
and Drugs, 68, 28-35.
Friedmann, P. D., McCullough, D., & Saitz, R. (2001). Screening and inter-
vention for illicit drug use: A national survey of primary care physicians
and psychiatrists. Archives of Internal Medicine, 161, 248-251.
Gentilello, L. M. (2005). Confronting the obstacles to screening and
interventions for alcohol problems in trauma centers. The Journal of
Trauma-Injury, Infection, and Critical Care, 59 (Suppl. 3), S137-S143.
Gentilello, L. M. (2007). Alcohol and injury: American College of Surgeons
Committee on trauma requirements for trauma center intervention. The
Journal of Trauma-Injury, Infection, and Critical Care, 62 (Suppl. 6),
S44-S45.
Gossop, M., Darke, S., Grifths, P., Hando, J., Powis, B., Hall, W., & Strang,
J. (1995). The Severity of Dependence Scale (SDS): Psychometric prop-
erties of the SDS in English and Australian samples of heroin, cocaine
and amphetamine users. Addiction, 90, 607-614.
Grant, B. F., Dawson, D. A., & Hasin, D. S. (2001). The Alcohol Use Disor-
der and Associated Disabilities Interview ScheduleDSM-IV Version.
Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism.
Grant, B. F., Dawson, D. A., Stinson, F. S., Chou, S. P., Dufour, M. C., &
Pickering, R. P. (2004). The 12-month prevalence and trends in DSM-IV
alcohol abuse and dependence: United States, 1991-1992 and 2001-
2002. Drug and Alcohol Dependence, 74, 223-234.
Grant, B. F., Dawson, D. A., Stinson, F. S., Chou, S. P., Kay, W., & Pick-
ering, R. P. (2003b) The Alcohol Use Disorder and Associated Dis-
760 JOURNAL OF STUDIES ON ALCOHOL AND DRUGS / SEPTEMBER 2010
abilities Interview Schedule-IV (AUDADIS-IV): Reliability of alcohol
consumption, tobacco use, family history of depression and psychiatric
diagnostic modules in a general population sample. Drug and Alcohol
Dependence, 71, 7-16.
Grant, B. F., Harford, T. C., Dawson, D. A., Chou, P. S., & Pickering, R. P.
(1995). The Alcohol Use Disorder and Associated Disabilities Interview
Schedule (AUDADIS): Reliability of alcohol and drug modules in a
general population sample. Drug and Alcohol Dependence, 39, 37-44.
Grant, B. F., Kaplan, K., Shepard, J., & Moore, T. (2003a). Source and ac-
curacy statement for Wave 1 of the 2001-2002 National Epidemiologic
Survey on Alcohol and Related Conditions. Bethesda, MD: National
Institute on Alcohol Abuse and Alcoholism.
Gunderson, E. W., Levin, F. R., & Smith, L. (2005). Screening and interven-
tion for alcohol and illicit drug abuse: A survey of internal medicine
housestaff. Journal of Addictive Disease, 24 (2), 1-18.
Hasin, D. S., Stinson, F. S., Ogburn, E., & Grant, B. F. (2007). Prevalence,
correlates, disability, and comorbidity of DSM-IV alcohol abuse and
dependence in the United States: Results from the National Epidemio-
logic Survey on Alcohol and Related Conditions. Archives of General
Psychiatry, 64, 830-842.
Henrique, I. F., De Micheli, D., Lacerda, R. B., Lacerda, L. A., & For-
migoni, M. L. (2004). Validation of the Brazilian version of Alcohol,
Smoking and Substance Involvement Screening Test (ASSIST). Revista
da Associacao Medica Brasileira, 50, 199-206.
Hides, L., Cotton, S. M., Berger, G., Gleeson, J., ODonnell, C., Proftt, T.,
. . . Lubman, D. I. (2009). The reliability and validity of the Alcohol,
Smoking and Substance Involvement Screening Test (ASSIST) in rst-
episode psychosis. Addictive Behaviors, 34, 821-825.
Humeniuk, R., Ali, R., Babor, T. F., Farrell, M., Formigoni, M. L., Jittiwu-
tikarn, J., . . . & Simon, S. (2008). Validation of the Alcohol, Smoking
and Substance Involvement Screening Test (ASSIST). Addiction, 103,
1039-1047.
Kaye, S., & Darke, S. (2002). Determining a diagnostic cut-off on the Se-
verity of Dependence Scale (SDS) for cocaine dependence. Addiction,
97, 727-731.
Khalsa, J. H., Treisman, G., McCance-Katz, E., & Tedaldi, E. (2008). Medi-
cal consequences of drug abuse and co-occurring infections. Substance
Abuse, 29, 5-16.
Khalsa, J. H., Genser, S., Francis, H., & Martin, B. (2002). Clinical con-
sequences of marijuana. Journal of Clinical Pharmacology, 42 (11
Suppl.) 7S-10S.
Maheux, B., Haley, N., Rivard, M., & Gervais, A. (1999). Do physicians
assess lifestyle health risks during general medical examinations? A
survey of general practitioners and obstetrician-gynecologists in Quebec.
Canadian Medical Association Journal, 160, 1830-1834.
Manwell, L. B., Fleming, M. F., Johnson, K., & Berry, K. L. (1998). To-
bacco, alcohol, and drug use in a primary care sample: 90-day preva-
lence and associated factors. Journal of Addictive Diseases, 17, 67-81.
Martins, S. S., Copersino, M. L., Soderstrom, C. A., Smith, G. S., Dis-
chinger, P. C., McDuff, D. R., . . . Gorelick, D. A. (2007). Sociodemo-
graphic characteristics associated with substance use status in a trauma
center inpatient population. Journal of Addictive Diseases, 26, 53-62.
Mertens, J. R., Weisner, C., Ray, G. T., Fireman, B., & Walsh, K. (2005).
Hazardous drinkers and drug users in HMO primary care: Prevalence,
medical conditions and costs. Alcoholism: Clinical and Experimental
Research, 29, 989-998.
Metz, C. E. (2003). ROCKIT 0.9B (Beta Version). Chicago, IL: University
of Chicago Press.
Miller, J. W., Naimi, T. S., Brewer, R. D., & Jones, S. E. (2007). Binge
drinking and associated health risk behaviors among high school stu-
dents. Pediatrics, 119, 76-85.
Newcombe, D. A., Humeniuk, R. E., & Ali, R. (2005). Validation of the
World Health Organization Alcohol, Smoking and Substance Involve-
ment Screening Test (ASSIST): Report of results from the Australian
site. Drug and Alcohol Review, 24, 217-226.
Ofce of National Drug Policy. (2004). The economic costs of drug abuse
in the United States: 1992-2002 (Publication No. 207303). Washington,
DC: Executive Ofce of the President.
Okulicz-Kozaryn, K. (2007). Ocena psychometryczynch wlasciwoski test
Problemowe uzywanie marihuany (PUMP dia dorastajacych). Postepy
Psychiatrii Neurologii, 16, 105-111.
Piontek, D., Kraus, L., & Klempova, D. (2008). Short scales to assess can-
nabis-related problems: A review of psychometric properties. Substance
Abuse Treatment, Prevention, and Policy, 3, 25-34.
Ramchand, R., Marshall, G. N., Schell, T. L., Jaycox, L. H., Hambarsoom-
ians, K., Shetty, V., . . . Belzberg, H. (2009). Alcohol abuse and illegal
drug use among Los Angeles County trauma patients: Prevalence and
evaluation of single item screener. The Journal of Trauma-Injury, Infec-
tion, and Critical Care, 66, 1461-1467.
Rehm, J., Taylor, B., & Room, R. (2006). Global burden of disease from al-
cohol, illicit drugs and tobacco. Drug and Alcohol Review, 25, 503-513.
Research Triangle Institute. (2002). Software for survey data analysis (SU-
DAAN) (Version 8.0). Research Triangle Park, NC: Research Triangle
Institute.
Rockett, I. R. H., Putnam, S. L., Jia, H., & Smith, G. S. (2003). Assessing
substance abuse treatment need: A statewide hospital emergency depart-
ment study. Annals of Emergency Medicine, 41, 802-813.
Rose, J. S., Herman, D. S., Hagerty, C., Phipps, M. G., Peipert, J. F., &
Stein, M. D. (2007). Marijuana use among young women in a primary
care setting. Journal of General Internal Medicine, 22, 826-829.
Schermer, C. R., Gentilello, L. M., Hoyt, D. B., Moore, E. E., Rozycki, G.
S., & Feliciano, D. V. (2003). National survey of trauma surgeons use of
alcohol screening and brief intervention. The Journal of Trauma-Injury,
Infection, and Critical Care, 55, 849-856.
Smith, P. C., Schmidt, S. M., Allensworth-Davies, D., & Saitz, R. (2009).
Primary care validation of a single-question alcohol screening test.
Journal of General Internal Medicine, 24, 783-788.
Stewart, S. H., Borg, K. T., & Miller, P. M. (2008). Prevalence of problem
drinking and characteristics of a single-question screen. Journal of
Emergency Medicine, E-pub June 6 ahead of print.
Stinson, F. S., Grant, B. F., Dawson, D. A., Ruan, W. J., Huang, B., & Saha,
T. (2005). Comorbidity between DSM-IV alcohol and specic drug use
disorders in the United States: Results from the National Epidemiologic
Survey on Alcohol and Related Conditions. Drug and Alcohol Depen-
dence, 80, 105-116.
Substance Abuse and Mental Health Services Administration. (2009). Re-
sults from the 2008 National Survey on Drug Use and Health: National
Findings. (Ofce of Applied Studies, NSDUH Series H-36, HHS Pub-
lication No. SMA 09-4434). Rockville, MD.
Swets, J. A., & Pickett, R. M. (1982). Evaluation of diagnostic systems:
Methods from Signal Detection Theory. San Diego, CA: Academic Press.
Williams, J. M., Chinnis, A. C., & Gutman, D. (2000). Health promotion
practices of emergency physicians. American Journal of Emergency
Medicine, 18, 17-21.
Williams, R., & Vinson, D. C. (2001). Validation of a single screening
question for problem drinking. Journal of Family Practice, 50, 307-312.
Yoast, R. A., Wilford, B. B., & Hayashi, S. W. (2008). Encouraging physi-
cians to screen for and intervene in substance use disorders: Obstacles
and strategies for change. Journal of Addictive Diseases, 27, 77-97.