Evaluating Substance Abuse in Persons with Severe Mental Illness

Full text

U.S. Department of Health and Human Services
Substance Abuse and Mental Health Services Administration
Center for Mental Health Services
www.samhsa.gov
Human Services Research Institute
2269 Massachusetts Avenue
Cambridge, MA 02140
www.tecathsri.org
C E N T E R
@
EVALUATION HSRI
t h e
Evaluating Substance Abuse
in Persons with Severe
Mental Illness
1995
Prepared by:
Kim T. Mueser, Robert E. Drake, Robin E. Clark,
Gregory J. McHugo, Carolyn Mercer-McFadden,
& Theiman H. Ackerson
Kim T. Mueser, Ph.D.
New Hampshire-Dartmouth Psychiatric Research Center
Main Building
105 Pleasant St.
Concord, NH 03301
Telephone: (603) 271-5747 FAX: (603) 271-5265

This Toolkit is one of a series of such kits commissioned by the Evaluation Center@HSRI. The
Center is a grant program of the Substance Abuse and Mental Health Services Administration,
Center for Mental Health Services. The mission of the Evaluation Center is to provide technical
assistance related to the evaluation of adult mental health system change.
The Center offers six programs all of which are designed to enhance evaluation capacity.
The programs are: the Consultation Program, consultation tailored to the needs of individual
projects; the Topical Evaluation Networks, which provide a forum for ongoing dialogue via
electronic conferencing; the Toolkit Program, which provides evaluators with tested
methodologies and instruments related to specific topics; the Materials Program, an evaluation
materials program which supplies evaluators with original papers on selected topics and
identifies relevant literature in the field; the Mini Grant Program that provides seed grants for
significant evaluations in the area of adult mental health system change; and the Training
Program designed to enhance the evaluation skills of producers and consumers of evaluations.
The Toolkits are designed to provide evaluators with complete descriptions of
methodologies and instruments for use in evaluating specific topics. Based on information from a
needs assessment study conducted by the Center and on feedback from evaluators in the fields,
we have identified a number of important topics that evaluators are frequently interested in
examining. Expert consultants have been engaged to review the background of these topics and
to compile Toolkits that provide evaluators with state-of-the-art evaluation techniques to use in
their own work.
The Evaluation Center@HSRI is also interested in supporting “user groups” for its Toolkits.
These groups will provide a forum for Toolkit users to share their expertise and experiences with
the Toolkits. If you would like to participate in a user group, please fill in the form enclosed and
return it to the Evaluation Center@HSRI.
We hope that this Toolkit on evaluating substance abuse in persons with severe mental
illness will be helpful to those evaluators who are interested in assessing the impact of system
changes on the life circumstances of persons with severe mental illness.
H. Stephen Leff, Ph.D. Virginia Mulkern, Ph.D.
Director Associate Director

Table of Contents
I. Introduction_______________________________________________ p. 1
Scope of the Problem..........................................................................................................p. 2
Definitions.....................................................................................................................p. 2
Prevalence.....................................................................................................................p. 3
Consequences of Substance Abuse................................................................................p. 6
Evaluation Difficulties........................................................................................................p. 7
Evaluation vs. Treatment Planning.....................................................................................p.10
The Recovery Process.........................................................................................................p.11
II. Clinician Rating Scales______________________________________ p. 12
Introduction.........................................................................................................................p.12
Specific Clinician Rating Scales.........................................................................................p.13
Alcohol & Drug Use Scales ..........................................................................................p.14
The Substance Abuse Treatment Scale .........................................................................p.19
Necessary Data for Valid Clinician Ratings .......................................................................p.21
Self-Report Measures....................................................................................................p.22
Clinician Ratings Based on Direct Observations .........................................................p.26
Collateral Reports.........................................................................................................p.26
Urine Drug Tests...........................................................................................................p.27
Assessments from Other Treatment Settings.................................................................p.27
Frequency of Clinical Assessments ....................................................................................p.27
Setting .................................................................................................................................p.29
III. Training __________________________________________________ p. 30
Introduction to the Concepts...............................................................................................p.30
Description of the Specific Scales ......................................................................................p.31
Practice & Discussion Using Each Scale............................................................................p.31
Reliability & Validity Checks.............................................................................................p.33
IV. Data Processing & Analysis __________________________________ p. 35
Dissemination/Public Policy...............................................................................................p.35
V. Bibliographies _____________________________________________ p. 44
General References .............................................................................................................p.44
Clinician Rating Scale References......................................................................................p.48
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness

INTRODUCTION
Over the past decade families, clinicians, and mental health
administrators have become increasingly aware of the problem of substance
abuse in persons with severe mental illness (Lehman & Dixon, 1995; Minkoff
& Drake, 1991; Ridgely et al., 1986). Previously, psychiatric patients were
rarely asked about their use of alcohol or drugs, nor were the possible effects of
substance abuse on the course of the disorder given more than cursory
consideration. Moreover, clinicians who did suspect that their patients might
have a problem with substance abuse were limited by the lack of validated
instruments for assessing substance use disorders in persons with severe mental
illnesses and by the lack of effective treatments.
Fortunately, substantial progress has been made in recent years in
understanding the scope of the problem of co-occuring substance use disorders,
in the development of reliable and valid measures for evaluating substance
abuse in people with severe psychiatric disorders, such as schizophrenia and
bipolar disorder, and in providing effective treatment for persons with both
disorders. This toolkit provides the information needed to assess the presence of
substance use disorders in persons with a psychiatric disorder, the severity of
the alcohol and drug abuse, and where on the continuum of recovery from
substance abuse patients fall. In our review, we have placed a premium on
measurement tools that are psychometrically sound, user friendly, and time
efficient to administer. At the same time, we highlight the limitations of
existing instruments and discuss possible threats to the validity of assessments.
We begin with a review of the scope of the problem of substance use
disorders in persons with severe psychiatric disorders, including prevalence
rates and impact on the course of illness and adjustment. Next, we discuss
problems inherent in the measurement of substance abuse in psychiatric clients,
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 1

and consider the difference between assessment and treatment planning. We
then review the recovery process for persons with a substance use disorder, as
such a process has implications for the measurement of these disorders, and we
describe specific rating scales that can be used to monitor the recovery process.
Methodological and training aspects of assessing substance use disorders in
severely mentally ill persons are also discussed, as well as strategies for the
processing and analysis of obtained data. Finally, we consider the public policy
and dissemination implications of conducting substance use assessments on this
population.
Scope of the Problem
Definitions
The diagnostic term “substance use disorder” refers to a habitual pattern
of alcohol or illicit drug use that results in significant impairments in areas of
adjustment, such as work, social relationships, economic well-being,
involvement in the legal system, or physical health. Traditionally, substance use
disorders have been divided into two mutually exclusive classifications—
substance abuse and substance dependence—with the latter diagnosis
representing the more severe disorder. Although there is some evidence that the
abuse/dependence distinction may be etiologically important (Noordsy et al.,
1994) and prognostically useful in the population of severely ill psychiatric
patients (Bartels et al., 1995), the same assessment issues pertain to both
classifications. For the purposes of this review we will follow nomenclature
from the Diagnostic and Statistical Manual of Mental Disorders (DSM) and
refer to a person with either substance abuse or substance dependence as having
a substance use disorder.
The time-frame for which a substance use disorder is assessed can have
important treatment implications. Generally, assessment techniques focus on
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 2

providing “lifetime” or “current” diagnoses of a substance use disorder. The
specific DSM criteria for specifying the course of a substance use disorder
change with each new edition. The DSM-IV uses at least one month without
abuse or dependence to indicate early remission and at least one year to indicate
sustained remission. Although persons with a lifetime substance use diagnosis
that is in remission may be seen as not requiring substance use-related treatment
services, their high vulnerability to relapses of their substance use indicates that
these patients often require ongoing treatment and assessment.
The term “comorbidity” refers to the presence or co-occurence of two
different medical conditions. Thus, persons with a psychiatric disorder (such as
schizophrenia) and a substance use disorder (such as alcohol abuse) can be
described as having comorbid disorders. They are also sometimes referred to
as having a dual diagnosis. In the next section, we review the research on the
prevalence of comorbid substance use disorders in persons with severe
psychiatric illnesses.
Prevalence
Estimates of the prevalence of substance use disorders in persons with
severe mental illness vary considerably, from as low as 10% to as high as over
65% (Safer, 1987; Goodwin & Jamison, 1990; Mueser et al., 1990; Mueser,
Bennett, & Kushner, 1995). The high variability in prevalence rates appears to
be due to differences across studies in factors such as the treatment setting in
which patients are sampled (e.g., community mental health center, acute
inpatient, chronic inpatient), whether the community is urban or rural, the
demographic mix of the study sample (e.g., proportion of males), and the
methods for assessing psychiatric and substance use disorders (e.g., structured
clinical interview, chart review) (Galanter, Castaneda, & Ferman, 1988). For
example, young males are significantly more prone to develop a substance use
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 3

disorder (Mueser, Yarnold, & Bellack, 1992), so that samples comprised of a
high proportion of these clients, as is the case with many studies of “young,
chronic” mental clients (Pepper, Kirshner, & Ryglewicz, 1981; Safer, 1987),
tend to yield high estimates of the prevalence of substance use disorders.
Similarly, clients assessed in an emergency room setting are more likely to have
a substance use disorder than clients living in state hospitals (Ritzler et al.,
1977; Barbee et al, 1989). Thus, the actual rate of substance use disorders in
persons with severe psychiatric disorders is determined, in large part, by the
mix of clients receiving treatment in that setting. Table 1 contains a list of some
of the most important predictors of substance abuse in persons with psychiatric
disorders.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 4

Table 1
Predictors of Substance Use Disorders
in Clients with Severe Psychiatric Illnesses
Patient Predictors Who is Most Vulnerable
Gender Males
Age Young
Education Clients with lower educational levels
Race/SES No differences for alcohol abuse.
Minorities and clients with lower SES are
more likely to abuse illicit drugs (e.g.,
cocaine, marijuana); White and higher
SES clients are more likely to abuse
prescription drugs (e.g., sedatives)
Premorbid functioning
social/sexual
Clients with higher premorbid
functioning
Setting Predictors Where Abuse is Most Likely
Population density Urban
Treatment setting Clients presenting to emergency rooms are
most likely to have substance use
disorders; patients in state hospitals are
less likely
Although specific estimates vary, there is overwhelming evidence that
persons with severe mental disorders are at increased risk for substance use
disorders. The most comprehensive study on the comorbidity of psychiatric and
substance use disorders was conducted as part of the Epidemiological
Catchment Area (ECA) study (Regier et al., 1990), in which over 20,000
persons living in the community or institutional settings were assessed. The
ECA study found that all people with a psychiatric disorder were more prone to
substance abuse, but persons with severe mental illness were especially
vulnerable. For example, clients with schizophrenia were more than four times
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 5

as likely to have had a substance use disorder during their lifetime, and those
with bipolar disorder were more than five times as likely to have such a
diagnosis, than persons in the general population.
The results of the ECA study, combined with numerous other prevalence
studies, indicate that persons with severe psychiatric illness are more likely to
have problems with alcohol and drug use than less ill clients or people with no
psychiatric disorder. Overall, about half of all persons with a severe psychiatric
illness have had a substance use disorder at sometime during their lives, and
between 25% and 35% have a current substance use disorder. By comparison,
less than 20% of people in the general population have a substance use disorder
during their lives. The high rate of substance use disorders among psychiatric
clients underscores the importance of accurate assessment in these persons. As
described in the next section, there is a high cost, both clinically and
economically, of the failure to diagnose and treat substance use disorders in this
population.
Consequences of Substance Abuse
Substance abuse among persons with severe mental illness can have
negative clinical effects, such as precipitating relapses and rehospitalizations,
and increasing suicidality and violence; it can lead to adverse medical
consequences, such as vulnerability to HIV+ infection, and precipitate
psychosocial instability, such as financial problems, housing loss, and
homelessness; furthermore, substance abuse can result in higher service
utilization, treatment costs, and economic burden to the family (Bartels et al.,
1992; Bartels et al., 1993; Clark, 1994; Cournos et al., 1991; Drake, Osher, &
Wallach, 1989; Yesavage & Zarcone, 1983). The impact of substance abuse on
symptoms is often so marked that clinicians are advised to first explore
substance use when a psychiatric client presents with an otherwise unexplained
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 6

symptom exacerbation. Despite the serious consequences of substance abuse,
there are reasons to be optimistic. In most cases, the impact of substance abuse
appears to be temporary, and dually diagnosed clients who attain stable
remission improve clinically and resemble non-abusing clients (Zisook, et al.,
1992). Thus, successfully reducing substance abuse may result in positive
outcomes in areas such as symptoms, community tenure, and service utilization.
Evaluation Difficulties
A wide range of different problems complicate the assessment of
substance use disorders in persons with severe psychiatric disorders (Drake,
Alterman, & Rosenberg, 1993). The major problem in most psychiatric settings
is that clinicians simply fail to take a careful history of substance use. For
example, a study by Ananth et al. (1989) found that 84% of substance disorder
diagnoses that were detected on structured interview were missed when clients
were evaluated in emergency settings and on entrance to a state hospital.
Clients who are in our research studies on dual disorders are identified as
having a substance disorder on only a minority of discharge summaries from a
variety of hospitals. Taking a careful history of clients’ alcohol and drug use
behavior does not guarantee detection of a substance use disorder, but it is the
most important first step in the evaluation process. Many clients who do not
volunteer information about their use of substances freely admit to alcohol or
drug abuse when a history is taken, enabling the clinician to establish a
diagnosis.
Some clients are willing and able to describe their substance use
behavior, while others are not. When evaluated directly, people with severe
mental illness are prone to the usual problems that accompany self-report (e.g.,
recalling the details of past behavior, responding to the demand characteristics
of the situation). In addition to these problems of self-report, however, they
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 7

may have difficulty participating in a structured interview during a symptom
exacerbation or in crisis (Barbee et al., 1989). Another common difficulty is
that cognitive, psychotic, and mood-related distortions characteristic of
psychiatric disorders can interfere with accurate recall. Furthermore, it is often
difficult or impossible to discern the causal effects of substance use on
psychiatric clients, since they often experience multiple stressors during times
of crisis; thus, for example, it may not be possible to determine the role(s)
played by substance abuse in precipitating any of the elements of a crisis which
include medication noncompliance, a symptom exacerbation, an episode of
homelessness, and a hospitalization.
In addition to problems understanding the effects of substance use,
psychiatric clients are prone to denial when they have experienced severe
sanctions, such as having been extruded from a program or a housing setting,
because of substance abuse. Denial is more common for marijuana, cocaine,
and other illicit drugs than for alcohol (Stone et al., 1993; Galletly, Field, and
Prior, 1993), probably because of laws prohibiting possession and use of such
substances. Minimization often occurs due to genuine confusion about the
effects of substance use. People with severe mental illnesses have typically had
a number of terrible experiences in their lives, and substance abuse, although
deleterious, may not be easily identified as a causal agent. In part, this is
because clients are typically aware of the short-term positive effects of
substance use, such as decreases in anxiety and depression, improved sleep, and
temporary feelings of well-being, rather than the long-term negative effects
which may be more difficult to detect, such as increases in hallucinations,
suicidal thoughts, and interference with the ability to manage one’s life.
Another critical problem in evaluating substance abuse in psychiatric
clients is that the usual standards for assessment are different in these persons
compared to people with a primary substance use disorder, but no psychiatric
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 8

illness. In other words, the dimensions of assessment—pattern of use,
consequences, the dependence syndrome, and subjective distress—are all quite
different for persons with severe mental illness than for those without a
psychiatric disorder. Thus, people with a severe mental illness tend to incur
adverse consequences on using relatively small amounts of alcohol or other
drugs (Janowsky et al., 1973; Treffert, 1978; Knudsen & Vilmar, 1984; Drake,
Osher, and Wallach, 1989; Lieberman, Kinon, & Loebel, 1990). The
consequences that they experience, although often typical for persons with
severe mental disorder, are not the consequences that are assessed on standard
instruments for primary substance abusers. For example, psychiatric clients
often encounter difficulties managing their illness, complying with prescribed
medications, budgeting disability funds, maintaining housing, and participating
in rehabilitation. On the other hand, most clients do not encounter problems
with jobs, spouses, and revocation of driver’s licenses because they are rarely
employed, married, or own their own vehicles.
Because of their sensitivity to the effects of alcohol and other drugs,
psychiatric clients often do not develop the syndrome of physiological
dependence, including tolerance or withdrawal when they stop using the
substance (Drake et al., 1990). Finally, due to the salience of other problems in
their lives and the difficulty in making accurate causal connections between
substance abuse and adjustment, they often have little subjective distress
regarding alcohol and other drug use. For these reasons, standard instruments
developed for primary substance abusers are usually inadequate to the task of
assessing these problems in persons with severe mental illness. For example,
instruments like the Addiction Severity Index (McLellan et al., 1990), which
rely on pattern of use and subjective distress, often fail to detect the extent of
substance abuse problems in this population.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 9

At this point in time, there is a pressing need to develop new instruments
for the assessment of substance abuse in patients with severe psychiatric
disorders. Until more refined instruments are available, we recommend taking a
multimodal approach. Such an assessment recognizes that no single instrument
and no one source of information is sufficient to diagnose a substance use
disorder accurately in this population. Rather, the most accurate assessment
process makes use of several instruments, pays attention to issues of relevance
for this population (e.g., effects on symptoms, treatment compliance, housing
stability), obtains information from multiple sources (e.g., patient, relatives,
case manager, drug screens), and includes a repeated, longitudinal component.
In this toolkit we describe some simple scales and strategies for assessing and
monitoring substance abuse over time.
Evaluation vs. Treatment Planning
The approach described here addresses the task of monitoring clients
who are in treatment. It is not intended to serve the function of a comprehensive
assessment for the purpose of treatment planning. We have reviewed the more
complex approach suitable for a thorough clinical assessment elsewhere (Drake
& Mercer-McFadden, 1995). Clinical assessment links the four tasks of
detection, classification, detailed assessment, and treatment planning in a
process of reciprocal feedback. The goal is to involve the client in an effort to
identify and address all of the biological, psychological, social, and
environmental factors that sustain the abusing behaviors.
The Recovery Process
An understanding of the process of recovery from a substance use
disorder can aid clinicians in monitoring substance abuse and progress in
treatment. Longitudinal research on persons with a primary drug or alcohol use
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 10

disorder indicates that these disorders are usually chronic over the lifetime,
associated with increased mortality rates, and receive only limited amounts of
treatment (Vaillant, 1983, 1988; Hser, Anglin, & Powers, 1993). Despite the
persistence of these disorders, some persons do achieve sustained abstinence,
with about 2-5% reaching stable remission per year, and 1-2% returning to
substance abuse. Less information is available about the natural course of
clients with severe mental illnesses and substance use disorders, but one long-
term study (seven years) of dually diagnosed clients indicates a rate of recovery
similar to that in primary substance abusers (Bartels et al., 1995). However,
there is also encouraging evidence suggesting that integrated substance abuse
and mental health treatment can accelerate the rate of remission in dual
diagnosis clients (Drake, Mueser, Clark, & Wallach, in press; Mueser, Drake, &
Miles, in press).
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 11

CLINICIAN RATING SCALES
Introduction
A large proportion of persons with severe psychiatric disorders are
affected by substance abuse, and clinical research points to the need to treat
psychiatric and substance use disorders in an integrated fashion. Accordingly,
all mental health clinicians who work with these individuals must develop
competence in the detection and treatment of substance use disorders. Simply
referring clients with a substance use disorder to other clinicians or other
treatment facilities never worked very well in the first place, and is no longer a
credible approach. Moreover, as mental health clinicians develop expertise in
the assessment and management of substance use disorders, their ability to
monitor substance abuse accurately assumes even greater importance, for
several reasons.
First, considering clients’ high vulnerability to substance use disorders,
monitoring is necessary to identify who needs substance abuse treatment
services and to pinpoint the possible causes of symptom exacerbations and
other crises. Second, substance use behavior needs to be repeatedly evaluated
over extended periods of time and in different settings in order to monitor
response to treatment. Third, regular monitoring is necessary even for clients
whose substance abuse is in remission, since they continue to be at high risk for
relapse of their substance use disorder. We will describe several clinical scales
for assessing alcohol and drug use in psychiatric clients, and for evaluating the
stage of treatment for clients’ substance use disorder.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 12

Specific Clinician Rating Scales
Our clinician rating scales were originally developed for case managers
to use in monitoring their clients; some are now incorporated as part of
standardized data collection across the New Hampshire mental health system.
We subsequently began to use these scales for research purposes and have
repeatedly demonstrated their reliability and validity.
Alcohol and Drug Use Scales The Clinician Rating Scales (CRS) for
alcohol and drug use, shown in Tables 2 and 3 (p. 15 and 16), were developed
to enable clinicians to assess and monitor substance use in persons with severe
mental illness. The scales were based on DSM-III-R criteria, but can be
modified in accordance with changes in diagnostic criteria in subsequent
revisions of the DSM. Case managers who follow their clients closely in the
community have access to multimodal assessment data about their use of
alcohol and drugs, including self-reports, observations across different
situations, collateral reports from significant others and friends, and medical
evaluations from different treatment settings. Case managers can easily be
trained to incorporate these data into their CRS ratings in order to monitor
clients’ substance use disorders over time. Because of the problems of self-
report and poor validity of standard instruments with this population, reviewed
above, clinicians’ ratings that incorporate multiple perspectives are usually
superior to assessments based on client self-reports alone.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 13

Table 2
Client Name: ___________________________
Date of Rating: __________________
Clinician Alcohol Use Scale
Please rate your client’s use of alcohol over the past six months according to the following
scale. If the person is in an institution, the reporting interval is the time period prior to
institutionalization. You should weight evidence from self-report, interviews, behavioral
observations, and collateral reports (family, day center, community, etc.) in making this rating.
____ 1 = ABSTINENT Client has not used alcohol during this time interval.
____ 2 = USE WITHOUT IMPAIRMENT Client has used alcohol during this time
interval, but there is no evidence of persistent or recurrent social, occupational,
psychological, or physical problems related to use and no evidence of recurrent
dangerous use.
____ 3 = ABUSE Client has used alcohol during this time interval and there is evidence of
persistent or recurrent social, occupational, psychological, or physical problems related
to use or evidence of recurrent dangerous use. For example, recurrent alcohol use leads
to disruptive behavior and housing problems. Problems have persisted for at least one
month.
____ 4 = DEPENDENCE Meets criteria for moderate plus at least three of the following:
greater amounts or intervals of use than intended, much of time used obtaining or using
substance, frequent intoxication or withdrawal interferes with other activities, important
activities given up because of alcohol use, continued use despite knowledge of
substance-related problems, marked tolerance, characteristic withdrawal symptoms,
alcohol taken to relieve or avoid withdrawal symptoms.
For example, drinking binges and preoccupation with drinking have caused client to
drop out of job training and non-drinking social activities.
____ 5 = DEPENDENCE WITH INSTITUTIONALIZATION Meets criteria for severe
plus related problems are so severe that they make noninstitutional living difficult. For
example, constant drinking leads to disruptive behavior and inability to pay rent so that
client is frequently reported to police and seeking hospitalization.

Table 3
Client Name: ___________________________
Date of Rating: __________________
Clinician Drug Use Scale
Please rate your client’s use of drugs over the past six months according to the following scale.
If the person is in an institution, the reporting interval is the time period prior to
institutionalization. You should weight evidence from self-report, interviews, behavioral
observations, and collateral reports (family, day center, community, etc.) in making this rating.
____ 1 = ABSTINENT Client has not used drugs during this time interval.
____ 2 = USE WITHOUT IMPAIRMENT Client has used drugs during this time interval,
but there is no evidence of persistent or recurrent social, occupational, psychological, or
physical problems related to use and no evidence of recurrent dangerous use.
____ 3 = ABUSE Client has used drugs during this time interval and there is evidence of
persistent or recurrent social, occupational, psychological, or physical problems related
to use or evidence of recurrent dangerous use. For example, recurrent drug use leads to
disruptive behavior and housing problems. Problems have persisted for at least one
month.
____ 4 = DEPENDENCE Meets criteria for moderate plus at least three of the following:
greater amounts or intervals of use than intended, much of time used obtaining or using
substance, frequent intoxication or withdrawal interferes with other activities, important
activities given up because of drug use, continued use despite knowledge of substance-
related problems, marked tolerance, characteristic withdrawal symptoms, drugs taken to
relieve or avoid withdrawal symptoms. For example, binges and preoccupation with
drugs have caused client to drop out of job training and non-drug social activities.
____ 5 = DEPENDENCE WITH INSTITUTIONALIZATION Meets criteria for severe
plus related problems are so severe that they make noninstitutional living difficult. For
example, constant drug use leads to disruptive behavior and inability to pay rent so that
client is frequently reported to police and seeking hospitalization.
Mark drugs used: __ Cannabis __ Cocaine __ Hallucinogens __ Opiates
__ PCP __ Stimulants __ Sedatives/Hypnotics/Anxiolytics
__ Over-the-counter __ Other ____________________

The CRS encompasses a simple classification system that corresponds to
DSM-III-R criteria and also to severity in terms of clinical distinctions that are
considered meaningful for this population. Thus, as described in Tables 2 and 3,
the categories of abstinent, use without impairment, abuse, dependence, and
dependence with institutionalization comprise the CRS. An unusually large
proportion of clients with severe mental illness abstain from alcohol or drug
use, particularly those patients with poor premorbid functioning and more
severe symptoms (Ritzler et al., 1977; Mueser et al., 1990; Dixon et al., 1991;
Arndt et al., 1992). This isolation may be due to their severe social isolation and
lack of awareness of social norms, including potentially destructive norms,
which renders them less likely to be exposed to substance use and less able to
maintain a pattern of regular use (Cohen & Klein, 1970). Non-problematic use
is documented because these clients tend to develop substance abuse if they
continue using. Therefore, these clients are important candidates for education
and early intervention to prevent the development of a substance use disorder
(Drake & Wallach, 1993).
Abuse, according to DSM-III-R criteria (American Psychiatric
Association, 1987), is defined as a pattern of substance use that leads to
significant impairment or distress in vocational, social, emotional, or medical
functioning, or results in recurrent use in situations which are physically
hazardous. These criteria can easily be tailored to persons with severe mental
illness because they typically experience some negative effects of their
substance abuse, such as inability to manage funds, maintain housing, or
participate in rehabilitation. Dependence involves greater severity of the
addiction process and is operationalized in terms of DSM-III-R criteria: e.g.,
greater amounts or intervals of use than intended, much of time used obtaining
or using substance, frequent intoxication or withdrawal interferes with other
activities, important activities given up because of substance use, continued use
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 16

despite knowledge of substance-related problems, marked tolerance,
characteristic withdrawal symptoms, substance taken to relieve or avoid
withdrawal symptoms. Other criteria, which are more typical of clients with
severe mental disorder, should probably also be included in this definition.
Evidence from at least two studies indicates that the abuse-dependence
distinction may be particularly important for these clients (Bartels, Drake, and
Wallach, 1995; Noordsy et al., 1994). Finally, when clients have difficulty
maintaining themselves outside of institutional or homeless settings because of
their involvement with substances, they are rated as severely dependent.
The CRS is reliable, sensitive, and specific when used by case managers
who follow their mentally ill clients over time in the community (Drake, Osher,
& Wallach, 1989; Drake et al., 1990). Test-retest reliabilities over one to two
weeks on small samples have been close to 100%. Inter-rater reliabilities,
established by comparing ratings of clinical case managers and team
psychiatrists, have yielded Kappa coefficients between .85 and .95 for current
use disorder (Drake, Osher, and Wallach, 1989). An independent study used the
CRS to rate recent and past alcohol and drug use disorders, each separately, and
found intraclass correlation coefficients ranging between .58 - .82, (Mueser et
al., 1995). When CRS ratings were compared to consensus diagnoses generated
by a team of experienced psychiatrists using all clinical, research, and treatment
data available for each client to establish a current diagnosis of substance abuse
or dependence, the CRS achieved a high sensitivity (94.7%) and specificity
(100%) (Drake et al., 1990).
The ratings refer to an individual’s particular pattern of substance use.
As Table 3 indicates, categories of abuse should include not just the usual
groups of abused drugs, but also over-the-counter medications (e.g.,
antihistamines, “diet” pills) and prescribed medications (e.g., benzodiazepines),
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 17

two types of substances that are often abused by persons with severe mental
illness.
The Substance Abuse Treatment Scale The Substance Abuse
Treatment Scale (SATS) was developed to assess and monitor the progress that
persons with severe mental illness make toward recovery from substance use
disorder. Empirical observations by clinicians and clients’ self-reports indicated
that persons with severe mental illness typically recover from substance use
disorders in a sequential fashion: First they become engaged in some type of
treatment relationship. Second, they develop motivation to moderate or
eliminate their use of alcohol or drugs. Third, they adopt active change
strategies to attain controlled substance use or, more typically, abstinence.
Fourth, they endeavor to maintain specific changes and build supports to
prevent relapses. These observations led Osher and Kofoed (1989) to postulate
four stages in the recovery process, which they called engagement,
persuasion, active treatment, and relapse prevention. Clinicians who have
used this four-stage model in New Hampshire since 1989 observed that they
were actually able to differentiate early and late aspects of each stage, thus
expanding the model to a total of eight stages—pre-engagement, engagement,
early persuasion, late persuasion, early active treatment, late active treatment,
relapse prevention, and recovery—that corresponded to progress and treatment
needs. These eight stages were defined with operational criteria, as shown in
Table 4 (next page).
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 18

Substance Abuse Treatment Scale
Instructions: This scale is for assessing a person’s stage of substance abuse treatment, not for
determining diagnosis. The reporting interval is the last six months. If the person is in an
institution, the reporting interval is the time period prior to institutionalization.
1. Pre-engagement The person (not client) does not have contact with a case manager,
mental health counselor or substance abuse counselor.
2. Engagement The client has had contact with an assigned case manager or counselor but
does not have regular contacts. The lack of regular contact implies lack of a working
alliance.
3. Early Persuasion The client has regular contacts with a case manager or counselor but
has not reduced substance use more than a month. Regular contacts imply a working
alliance and a relationship in which substance abuse can be discussed.
4. Late Persuasion The client is engaged in a relationship with case manager or counselor,
is discussing substance use or attending a group, and shows evidence of reduction in use
for at least one month (fewer drugs, smaller quantities, or both). External controls (e.g.,
Antabuse) may be involved in reduction.
5. Early Active Treatment The client is engaged in treatment, is discussing substance use
or attending a group, has reduced use for at least one month, and is working toward
abstinence (or controlled use without associated problems) as a goal, even though he or she
may still be abusing.
6. Late Active Treatment The person is engaged in treatment, has acknowledged that
substance abuse is a problem, and has achieved abstinence (or controlled use without
associated problems), but for less than six months.
7. Relapse Prevention The client is engaged in treatment, has acknowledged that substance
abuse is a problem, and has achieved abstinence (or controlled use without associated
problems) for at least six months. Occasional lapses, not days or problematic use, are
allowed.
8. In Remission or Recovery The client has had no problems related to substance use for
over one year and is no longer in any type of substance abuse treatment.

Recovery from a substance use disorder is a longitudinal process that
takes place over months or years. When clinicians do not understand the
longitudinal process, they often bring unrealistic expectations to the interaction,
offer interventions for which the client is not ready, and become frustrated. Use
of the SATS reminds the clinician of the longitudinal process and permits the
identification of treatment options that are appropriate for the client’s current
stage of recovery. Other advantages of using the SATS to assess and monitor
clients are that it allows the clinician to evaluate progress before abstinence is
obtained and permits monitoring over time of specific patients and programs
(McHugo et al., in press).
Use of the SATS does not imply that recovery is a linear process.
Substance abuse is a chronic, relapsing disorder. Clients typically backslide and
cycle between stages, particularly early in treatment, as a natural part of the
recovery process. Nevertheless, at any one point in time, treatment needs to be
provided which is matched to the client’s current stage of recovery (Drake &
Noordsy, 1994). Thus, for example, a client who is homeless and living in a
shelter must typically be engaged in a collaborative treatment relationship, or
working alliance, before he or she will be interested in pursuing substance
abuse treatment. As another example, once the client is engaged in a treatment
relationship, he or she must have some motivation to pursue abstinence before
successfully participating in one or more active, abstinence-oriented
interventions. Before motivation is present, motivational interventions are more
appropriate than strategies designed to reduce alcohol and drug use.
Initial studies of the SATS (McHugo et al., in press) indicate high inter-
rater and test-retest reliability, with intraclass correlations typically around 0.9.
Clinician ratings of the SATS also correspond strongly to ratings made by
researchers, as well as to clinician ratings of substance use, and to client self-
reports about alcohol and drug use. Correlations are in the 0.3 to 0.6 range on
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 20

these measures of similar constructs, used to assess convergent validity. As a
measure of discriminant validity, SATS ratings are correlated with assessments
of progress in other functional domains in the 0 to 0.3 range.
The SATS can be used as either a process or an outcome measure.
As a process measure, the SATS yields useful information to clinicians as to
their most proximate goals in therapy and the techniques that may aid in
helping a client progress to the next stage of treatment. Thus, the most
immediate goal when working with a client in the pre-engagement phase is to
work towards the next stage, engagement, by establishing an interpersonal,
helping relationship. Efforts to convince the client to address his or her
substance abuse problem before such a relationship is established usually fail
and may drive the person away from treatment. As an outcome measure, the
SATS enables clinicians and program evaluators to assess the success (or lack
thereof) of treatment for substance use disorders. A total lack of change or
multiple backslidings over many years, as evident from repeated assessments
with the SATS, might be used to question the interventions or programs being
used to treat those specific clients. In sum, the SATS can be used to guide the
clinician’s therapeutic work and to inform clinicians and program evaluators as
to whether progress is evident in particular clients or groups of clients.
Necessary Data
for Valid Clinician Ratings
We have briefly described in the preceding section the need for
information from multiple sources for clinicians to make reliable and valid
ratings on the CRS and SATS. This procedure relies on the clinician’s actively
pursuing, obtaining, and synthesizing information from a wide array of different
sources. It assumes that case managers or other clinicians using these scales
know their patients well, understand the various clinical presentations of
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 21

substance use disorders and the recovery process, and are unbiased in their
assessments. These assumptions are supported by previous research we have
conducted on the use of these scales by clinicians in a variety of mental health
settings. In this section we elaborate on the necessary types of data, including
self-report measures, direct observations, collateral reports, urine drug tests, and
assessments from other treatment settings.
Self-report Measures No single self-report instrument has great
validity in this population, but such assessments can provide invaluable
information about some clients’ use of alcohol and drugs. To obtain specific
information about clients’ recent substance use, we recommend assessing the
pattern of use over the past six months using the Time-Line Follow-Back
(TLFB) method (Sobell et al., 1980). An example of a TLFB assessment form
is provided in Table 5 (p. 26). The TLFB involves having the client estimate the
specific amount of alcohol and different types of drugs consumed each month
over the past six months. Although these estimates may be biased towards
underreporting, they are nevertheless useful in characterizing the pattern of
abuse in clients who admit to at least some alcohol or drug use.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 22

Table 5
Drug/Alcohol 6-Month Follow-Back Calendar
I.D. Number: _____________________________ Date: ____/____/____
Instructions to Interviewer: Probe for patterns of alcohol and drug use, starting with information from the past 30 days obtained on the
ASI. Work backwards, month by month, and emphasize days of abstinence within each month.
ALCOHOL
Month
1 2 3 4 5 6
Kind
How Much
How Often
DRUGS
Month
1 2 3 4 5 6
Kind
How Much
How Often

Once a pattern of substance use has been established, specific
consequences of use can be evaluated by employing a checklist derived from
the DSM. We also recommend supplementing the items on this checklist with
additional items that are frequent problems in persons with severe mental
illnesses, such as those listed in Table 6 (next page). Self-report information,
when combined with knowledge of common consequences of substance abuse
in the psychiatric population, is often sufficient to evaluate the severity of a
substance use disorder.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 24

Table 6
Checklist of Common Consequences of Substance Abuse
in Persons with Severe Mental Illness
Consequence Examples
Housing instability getting evicted from apartment, group
home, family
Symptom relapses apparently
unrelated to life stressors
increases in psychotic symptoms,
worsening of depression, mania
Treatment noncompliance failing to attend medication or other clinic
appointments
Violent behavior or threats of violence getting into fights, throwing objects,
cursing at others
Sudden, unexplained mood shifts depression and hopelessness, anger,
euphoria, anxiety, expansiveness
Suicidal ideation or attempts thoughts or talk about hurting or killing
oneself, contemplating death, thinking of
plans to hurt oneself
Cognitive impairments increased confusion, memory problems,
difficulty planning ahead not related to a
stress-induced symptom relapse
Difficulty budgeting funds frequent attempts to borrow money,
stealing money, pawning one’s own or
others’ possessions
Prostitution trading sex for money, food, clothing, or
drugs/alcohol
Social isolation increased avoidance of others
Social difficulties frequent arguments with family, friends
Employment difficulties frequently tardy or absent, arguments with
employer or other employees, having pay
docked, job loss
Hygiene and health problems deterioration in personal hygiene and
grooming, medical problems, weight loss
Legal problems arrests for disorderly conduct, drunken
driving, possession of illicit drugs,
shoplifting
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 25

Clinician Ratings Based on Direct Observations One of the most
critical sources of information about substance abuse is the clinician’s own
observations of clients’ behavior at the mental health center or other treatment
settings. For example, if clients appear for appointments or attend groups when
they are intoxicated, there is strong evidence that they have a substance use
disorder. Other behavior changes may also provide clues about a possible
substance abuse problem, such as missed appointments, unexplained symptom
relapses, sudden interpersonal conflicts, or budgeting problems in a client who
is ordinarily able to manage his or her money (see Table 6 for other common
consequences). Although observations of clients in treatment settings are
useful, information gleaned across different situations and at different times of
the day in non-treatment settings is also very helpful. Such information is
available to clinicians whose work is not solely clinic-based and who have the
flexibility to meet with clients in more naturalistic settings (e.g., at their homes,
restaurants, parks).
Collateral Reports Clinicians are frequently privy to a limited and
biased sample of behavior based on their own contacts and observations of
clients. This over-reliance on a select sample of behavior can sometimes be
overcome by obtaining collateral reports from others who have regular contact
with the client. Other treatment providers, as well as shelter workers, housing
staff, and family members are the most commonly available people, but reports
may be available from others as well (e.g., friends, members of the clergy, law
enforcement officials). When obtaining collateral reports about clients’
substance use behavior, it is useful for the clinician to review with the
informant some of the common consequences of substance abuse in persons
with severe mental illness (Table 6), and the specific criteria included in the
CRS (Tables 2 and 3). This discussion may highlight for the informant critical
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 26

behaviors characteristic of a substance use disorder, improving their ability to
aide in the monitoring of these problem behaviors. An important goal when
soliciting collateral reports is to develop a working relationship with others who
are familiar with the client’s behavior outside of the usual treatment setting, so
that ongoing information can be obtained from these same sources.
Urine Drug Tests Urine drug tests cannot inform clinicians about the
consequences of substance use, but they can identify which clients have been
recently using substances. Our experience has been that urine drug screens are
more likely to be resisted by the clinicians who must administer them than by
the clients who provide samples. Therefore, once obstacles within a given
treatment setting have been overcome, such screens can be readily obtained,
and they provide a unique insight into clients’ substance use. We recommend
regular testing whenever the clinical situation suggests possible substance abuse
and regular testing (e.g., at least every month) for those who are in the process
of recovery (Drake, Alterman, & Rosenberg, 1993).
Assessments from Other Treatment Settings Finally, clinicians need
to be aware of all information available about clients’ substance use history in
records from other treatment settings. Clients are often inconsistent about what
they tell different treatment providers, and an accurate assessment can only be
made when all possible sources of information have been compiled. For
example, general medical records may provide information on alcohol-related
problems.
Frequency of Clinical Assessments
Substance use disorders in both the general population and among
persons with severe mental illness tend to be chronic, often life-long conditions.
Because of the severity and persistence of these disorders, they tend to improve
with treatment at exceedingly slow rates. Stable changes often appear after
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 27

years, rather than weeks or months, of attempts to change. The short-term
picture, i.e., what happens over 30 days following an intervention, is not
strongly predictive of stable changes. Therefore, for the purposes of both
clinical and program evaluation, assessments need to be conducted on a regular
basis over long periods of time. We recommend conducting formal clinician
assessments (CRS, SATS) on all clients in a mental health program every six
months, although on-line clinicians should conduct informal assessments on a
more frequent basis (e.g., monthly) in order to best meet clients’ needs.
Furthermore, we recommend that routine assessments be conducted for at least
a two-year period on any client who has a history of substance use disorder,
even if that disorder is currently in remission. Long-term follow-up assessments
are especially important in order to evaluate the success of programs aimed at
improving the course of dually diagnosed clients. Most of the available
evidence suggests that brief programs lasting one year or less tend to produce
only transient improvements in substance use disorder in this population.
For example, our studies in New Hampshire show a slow but steady
progression toward attaining stable abstinence, so that few clients appear to
improve markedly over any six month interval, but significant progress can be
observed over two or three years (Drake, McHugo, & Noordsy, 1993; Drake,
Mueser, Clark, & Wallach, in press; McHugo et al., in press). These studies
document that recovery occurs slowly, in stages, over years. By three years, one
third to one half have typically achieved substantial abstinence, and many
others have moved into active, abstinence-oriented treatment with reduction in
their use.
Setting
Substance abuse is an extremely environmentally sensitive disorder
(Galanter, Castaneda, & Ferman, 1988; Moos et al., 1990). This means that a
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 28

client’s substance use behavior in one environment may not generalize to
another setting. Thus, abstinence in an institutional setting, whether prison or
hospital, or in a residential treatment setting, is not predictive of abstinence in
less restrictive settings in the community, as such clients often relapse when
they return to their usual community living situations. The implications of this
limitation are two-fold. First, assessments of substance use behavior need to be
routinely conducted when a client’s environment has changed, because there is
little generalization of assessments across different settings. Second,
intervention for clients with substance use disorders in highly restrictive
environments must also extend the treatment into clients’ natural settings if
treatment gains are to be maintained. The failure to provide a continuity of care
from inpatient or residential-based treatments for substance use disorders may
be one reason why such approaches have not been found to have long-term
impact (Drake, Mueser, Clark, & Wallach, in press). Thus, from the perspective
of program evaluation, substance use disorders require ongoing assessment,
especially following a change from a more restrictive to a less restrictive living
arrangement.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 29

TRAINING
The training of clinicians in the use of the substance abuse assessment
scales described here can be divided into five steps: introduction to the
concepts, description of the specific scales, practice and discussion using each
scale, and reliability and validity checks. The essential components of each of
these steps are reviewed below.
Introduction to the Concepts
Clinicians need to be introduced to three broad concepts before they can
learn to provide valid ratings on the CRS and SATS: 1) the prevalence of
substance use disorders in the general and psychiatric populations; 2) the
defining characteristics of substance use disorders (i.e., the definitions of
“substance abuse” and “substance dependence”); and 3) the concept of “stages
of treatment.” The introduction to these concepts is best achieved by a
combination of assigned readings, didactic presentations, and group
discussions. For assigned readings on the prevalence of substance use disorders
in the severely mentally ill, we recommend the chapter by Mueser, Bennett, and
Kushner (1995). For the definition of substance use disorders, we recommend
reading the relevant sections from the DSM on substance use disorders and the
article by Drake, Alterman, and Rosenberg (1993) on assessment. Concerning
the stage of treatment concept, we recommend Osher and Kofoed’s (1989)
article. These articles are reprinted with permission in Appendix A.
We have had the most success training clinicians when we have assigned
the relevant articles in advance and started each session with a brief, didactic
review of the content of the article. This introduction is then followed by an
open discussion of the concept in which the main purpose is to elicit clinicians’
understanding of the material, correct misconceptions, and enable them to see
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 30

the relevance of the information to their clients. Although this educational
component of training precedes instruction in the actual use of the scales,
trainers need to be alert to opportunities to provide additional education to
clinicians about these concepts throughout all aspects of the training process.
Descript ion of the Specific Scales
After clinicians have become familiar with the core concepts behind the
scales, they are introduced to the scales themselves. Copies of the scales are
handed out or projected on an overhead screen, a brief overview of the entire
scale is provided, and then the specific items on each scale are reviewed in the
order they appear. As with all aspects of the training process, we suggest that
trainers periodically stop and ask clinicians to paraphrase their understanding of
specific points, in order to evaluate their comprehension of the presented
material. Furthermore, eliciting specific case examples from clinicians during
the course of describing the scales can help them understand how different
points on each scale translate into actual clinical cases. When introducing the
specific scales, discussion of the two CRS scales (alcohol and drug) can be
combined into a single session. However, we caution against including a
discussion of the SATS in the same session, as the function of this scale is
somewhat different, and discussion of all three scales in a single session runs a
significant risk of information overload for the participants.
Practice & Discussion Using Each Scale
When clinicians understand the purpose and components of a scale, they
are ready to begin learning how to use it. Before clinicians rate their own
clients, it is useful to give them practice rating clinical vignettes which the
trainers have prepared in advance. These vignettes serve to describe
prototypical cases that illustrate a particular rating on the scale. Clinicians must
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 31

be familiar with how such clear-cut cases should be rated before they are
prepared to evaluate other, less obvious cases. Table 7 provides a series of
clinical vignettes that illustrate each point on the SATS. These specific
vignettes could be used for training, or other vignettes could be generated that
more closely approximate the clients receiving services in a particular setting.
Vignettes can be used in two different ways when training clinicians.
First, a vignette can be provided to illustrate a particular point on the rating
scale, thus translating an abstract idea into a clinical example. Second,
clinicians can be presented with a vignette and requested to identify which
rating would best correspond to that example. This allows clinicians to “test”
themselves and provides information to the trainer about their understanding of
the use of the scale. When training is conducted in a group format, the
responses of different clinicians can be discussed and clarifications can be made
concerning the correct answer. We recommend using clinical vignettes both to
illustrate the rating scales and to provide clinicians initial practice in using the
scales.
After clinicians have demonstrated sufficient understanding of the use of
the scale with clinical vignettes, they can begin practicing the scale on their
own clients. At first it is preferable to ask each clinician to rate a small number
of each of their clients (2-4 clients) and to bring back their ratings to the group.
It is optimal if each clinician who is providing ratings can rate clients who can
also be rated by another clinician. This will permit comparison of the different
ratings. Clinicians should be requested not to discuss with each other how a
client would be rated until after they have made their ratings. Ratings should
then be discussed in the group, with an effort made at reconciling differences
through reaching a consensus. The process of rating clients and then discussing
the ratings usually needs to continue for several training sessions. Trainers can
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 32

keep track of agreement to determine when clinicians are able to use the scales
accurately.
Reliability & Validity Checks
Reliability and validity checks need to be conducted on a regular basis
for all clinicians who provide ratings on these scales. We recommend that
routine reliability and validity checks be conducted every six months. The
reliability of ratings refers to the level of agreement achieved by independent
raters for a particular scale (i.e., how consistent a rating is across different
clinicians). To assess reliability, different clinicians should provide independent
ratings on the same clients. It is best if the clinicians are not aware that a
reliability check is being conducted, since such information could influence
their ratings. Ratings of the same clients can then be compared across different
clinicians. Usually, ratings of at least ten different clients need to be obtained to
determine whether they are reliable, and ratings of more clients are preferred.
Although there are complicated statistical formulas for evaluating level of
agreement between raters, a simple approach is to determine the percentage of
agreement between different pairs of clinicians. A level of 80% agreement is
considered acceptable. If some clinicians fall below this level, additional
training sessions should be considered. If many clinicians fall below this level a
special meeting may be required to evaluate whether the clients rated posed
special problems in using the scales, or whether there are significant differences
in how clinicians understand the scales are to be used.
The validity of ratings refers to how accurate the ratings are in terms of
what they are intended to measure. In other words, does a rating really reflect
the severity of a client’s substance use disorder or his/her stage of treatment?
Unlike the question of reliability, there is no simple answer to this question.
However, by examining other measures which are believed to be related to
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 33

substance abuse, one can gather information pertinent to the validity of a
clinician rating. For example, if a urine drug screen reveals the presence of
cocaine in the urine of a client, one would expect that client to have a rating on
the CRS-Drug of at least “2” (recent use) and probably higher. If a client was
arrested for “drunk and disorderly conduct,” then he or she would be expected
to have a rating of at least “3” (alcohol abuse) on the CRS-Alcohol scale. If a
client has regular contact with the case manager and has been attending groups
for dually diagnosed persons, he or she would be rated at least a “3” (Early
Persuasion) on the SATS. The specific information available to assess the
validity of clinician ratings will vary across clients and programs. Despite the
difficulty in evaluating the validity of clinician ratings, such checks are
essential to be confident that the rating scales are being used as intended, and in
order to troubleshoot problems related to their use.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 34

DATA PROCESSING & ANALYSIS
The processing and analysis of ratings based on the clinician scales is
straightforward, as the scales themselves are quite simple. Improvements in
client’s substance use disorders (CRS) and stage of treatment (SATS) over time
can be examined by computing within group t-tests (two time points) or
repeated measures analyses of variance (more than two time points), with the
ratings of each scale treated as a continuous dependent variable. Evaluations
can also be conducted using these scales to determine whether clients have
changed their group membership over time. For example, on the CRS a rating
of “3” or higher denotes that the client has a substance use disorder. Thus,
ratings of “1” or “2” can be collapsed to form a “no disorder group,” and ratings
of “3,” “4,” or “5” can be collapsed to form a “substance use disorder group,”
and subsequent analyses could be performed to evaluate whether the number of
clients with a disorder changed as a function of treatment. Similarly, ratings on
the SATS could be combined to represent the four stages of recovery from
substance abuse (“1” and “2,” “3” and “4,” and so on).
Dissemination/Public Policy
Growing emphasis on the cost-effectiveness of treatment highlights the
need for valid, easy-to-use outcome measures that can be applied consistently
across a wide variety of settings. Better measurement of substance abuse and
other outcomes has tremendous potential to improve the effectiveness of
treatment. If used consistently, valid and reliable client outcome measures are
likely to have a broad impact on financing of care, on demand for treatment and
on provider training.
Private corporations and government agencies are relying increasingly
on outcome data to make financial decisions. Maine, New Hampshire, and other
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 35

states have already begun experimenting with reimbursement mechanisms that
tie payments to client outcomes. Among private insurers and providers,
treatment outcomes are rapidly becoming a factor in competition for contracts
and for clients. Large managed care firms are rushing to set up departments for
measuring outcomes. With these high stakes, the validity and reliability—even
the use—of outcome measures are likely to be scrutinized carefully and
challenged repeatedly. In such an environment it is critical that providers use
the best measurement techniques possible and that they apply them
appropriately and consistently across clients and settings.
From the purchaser’s perspective, it is essential to account for the
limitations of outcome measures that were discussed in the preceding sections.
Caution is especially important when outcomes are measured over short periods
of time. Until we understand better the longitudinal course of substance abuse
and mental illness, payers should be careful about tying strong financial
incentives to outcomes. Incentives that are too strong could encourage
providers to emphasize short-term results at the expense of more lasting
improvements or to exclude from treatment those clients who are more severely
impaired.
Assessment of substance abuse in people with mental illness, like other
outcome measures, is not free. Although the approaches we have discussed are
relatively inexpensive, the cost of careful evaluation can range from a small
addition to the workload of clinicians and administrative staff for collection and
analysis of data to expenditures for special interviewers, urine tests, computer
equipment, software, and expert data analysis. Because the quality of
measurement often increases with expenditures for more sophisticated
techniques, there is a tradeoff between cost and quality. Ultimately, providers
and payers must decide what level of accuracy and cost they are willing to
accept.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 36

Additional interest in measuring and monitoring substance abuse by
people with mental illness is already increasing the demand for specialized
training in treatment of dual disorders. To be effective, such training should be
available on a continuous basis. Ongoing formal instruction from an outside
trainer would strain the meager training budgets of most providers, but some
organizations have addressed the problem by identifying staff members with the
requisite skills and interest to be in-house trainers and by supplementing their
efforts with outside trainers from time to time.
Despite increasing awareness of the need to monitor outcomes carefully,
students in most clinical training programs get very little training in the theory
and techniques of outcome measurement. Even when such instruction is a part
of their training, it rarely focuses on the specific problems of dual disorders.
Increasing demands to show results from expensive clinical interventions make
it doubly important that present and future providers have an understanding of
the basic principles of outcome measurement. Incorporating into clinical
curricula training in how to evaluate substance abuse and in how to use such
information for improving treatment interventions would help prepare future
clinicians to function effectively in a world in which outcomes achieved, rather
than services provided, are the measure of success.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 37

Table 7
Vignettes for Substance Abuse Treatment Scale
1 = Pre Engagement
The person does not have contact with a case manager, mental health counselor,
or substance abuse counselor.
John was seen by Emergency Services after being picked up by the
police for disturbing the peace. He had been drinking heavily and was yelling
loudly at passerby’s to “stop looking at him”. He had no particular residence
and no visible means of support. From old hospital records it was found that he
had been in a state psychiatric hospital for 20 years and had been discharged 5
years ago. After a brief period of hospitalization for stabilization on
medications and detoxification he was referred to the community support
program at the local mental health center (MHC). He did not keep any
appointments at the center but is often seen in the company of other clients of
case management.
Jeanne, a woman of indeterminate age, lives in a SRO building and has
high visibility in the local community because of her “weird” behaviors which
become worse when she is using substances. Police and local merchants have
called the MHC about her and several attempts have been made by MHC
outreach staff to get her into the center. She continues to refuse these offers.
2 = Engagement
The client has had contact with an assigned case manager or counselor but does
not have regular contacts. The lack of regular contact implies lack of a working
alliance.
Lionel, a young single man who has been diagnosed in the past as
suffering from schizophrenia, occasionally shows up at the mental health center
and demands to see someone. He knows he has a case manager but cannot
remember his name. He last saw his case manager 3 months ago when he
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 38

wanted to get fuel assistance. His contacts are infrequent, and usually involve
wanting money, food or cigarettes. Lionel smokes marijuana on a daily basis
but does not speak with his case manager about it.
After a brief hospitalization at the local psychiatric unit following a
psychotic episode, Pamela, a young college student, was assigned a case
manager. She saw her case manager on 2 occasions following discharge but has
not been seen for several months at the MHC and has not responded to phone
calls or letters. The client‘s mother has called the case manager and says that
she is worried about Pamela‘s increasing paranoia and indiscriminate use of
substances.
3 = Early Persuasion
The client has regular contacts with a case manager or counselor but has not
reduced substance use more than a month. Regular contacts imply a working
alliance and a relationship in which substance abuse can be discussed.
Julie sometimes initiates contact with her case manager and usually
keeps her appointments. Most of her contacts are in regards to basic needs. She
is able to listen when her case manager brings up her binges and other
substance use but does not contribute to the conversation or acknowledge a
problem. The case manager’s approach is to increase Julie‘s awareness of
substance use without any demands for abstinence.
Fred has been a client of the MHC for many years. He was a long time
resident of the state hospital prior to his involvement at the MHC.
Fred continues to drink at least a quart of wine daily and is not compliant with
taking his Haldol. He does meet weekly with his case manager and sometimes
calls when in crisis. The meetings usually deal with concrete needs and
activities of daily living.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 39

4 = Late Persuasion
The client is engaged in a relationship with a case manager or counselor, is
discussing substance use or attending group, and shows evidence of reduction in
use for at least one month (fewer drugs, smaller quantities, or both). External
controls (e.g., Antabuse) may be involved in reduction.
Ezekial, a young man with a history of schizoaffective disorder and
heavy marijuana use, was placed in a group home. His mother became
representative payee to control his funds. Since his placement, his relationship
with the case manager has improved. He attends weekly sessions and is about to
start a substance abuse group. It appears that his substance use has decreased so
as not to be a daily occurrence. Ezekial is able to discuss in his sessions what
the effects of substances are and on rare occasions verbalizes a goal of
abstinence.
Star lives in a supported apartment with two other clients of the mental
health center. She attends a day treatment program at the MHC 3 days a week
and sees her case manager twice a month. Star attends a “Double Trouble” AA
group once or twice a month in the community. Her case manager reports the
number of “parties” at the apartment have decreased considerably and Star has
not been binging as much.
5 = Early Active Treatment
The client is engaged in treatment, is discussing substance use or attending a
group, has reduced use, for at least one month, and is working toward abstinence
(or controlled use without associated problems) as a goal, even though he or she
may still be abusing.
Joe is a 44 year old twice divorced father of two who has a 20 year
history of bipolar disorder and polydrug abuse. In the past year, he has taken
more responsibility for his substance abuse. He is beginning to discuss it with
his case manager and in weekly group meetings at the MHC. He has started to
chart his weekly use and though not abstinent he says that eventually he wants
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 40

to be clean and sober. He complies with his psychiatric medications and is
attempting to make social contacts with non abusers.
Crystal is a grandmother with years of polysubstance abuse. Her
psychiatric symptoms are controlled with medication which she receives every
other week from the MHC nurse. She sees her case manager at least monthly.
Six months ago she went on a binge of drinking and also smoking crack. She
was out of control, was brought to the ER, and scared her daughter and her 2
grandchildren. Since that incident she has contracted with her case manager and
her daughter not to use crack and is trying to cut down on her drinking. She
wants to be able to still drink in a controlled manner, but if this does not work
then she states that she would have abstinence as a goal. She has begun to
attend AA again and is calling her case manager weekly to report her progress
and discuss her concerns.
6 = Late Active Treatment
The person is engaged in treatment, has acknowledged that substance abuse is a
problem, and has achieved abstinence (or controlled use without associated
problems), but for less than six months.
Gina is a young single woman with bipolar disorder who is active in NA
and AA for her cocaine addiction. She has been abstinent for 2 months and
prior to that has had a 5-month and a 4-month period of abstinence. After her
last lapse she asked to be in a more structured living situation associated with a
treatment program. She knows that cocaine is her drug problem and uses this as
a focus of her weekly meetings with her case manager. Her goals include
abstinence and getting to work.
Jonathan has been actively engaged in the case management program at
the MHC for over one year. During this time he has made much progress on his
daily abuse of alcohol and has now been abstinent for 3 months. With the help
of his case manager and the weekly substance abuse groups, he realizes that his
delusions and his behavior are affected by his substance abuse. He now takes
his psychiatric medications regularly.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 41

7 = Relapse Prevention
The client is engaged in treatment, has acknowledged that substance abuse is a
problem, and has achieved abstinence (or controlled use without associated
problems) for at least six months. Occasional lapses, not days of problematic use,
are allowed.
Vanessa a middle-aged woman with a bipolar disorder sees her case
manager weekly. She has been sober for 2 years with one lapse of 2 days
several months ago. She became depressed over a love relationship, loss of a
job and financial problems, and slipped. Following this she went into an 8-week
day treatment program and has continued to work with her case manager in
treatment to deal with these issues.
Sky is active in AA, where he has a sponsor, and also attends the weekly
substance abuse group at the MHC. He actively engages other clients and
confronts them about their abuse. He has been clean and sober for 2 and 1/2
years. He still has cravings but has utilized his case manager and community
support system to get through these periods. Sky has completed a year of
college and is active in the mental health consumer group.
8 = Recovered
The client has had no problems related to substance use for over one year and is
no longer in any ty
pe of substance abuse treatment.
Jefferson is a long-term client of the mental health system. He has an
excellent relationship with his case manager where the focus is on social skills
and maintaining himself in the community. For many years he had a heavy
alcohol dependency but has not used any substances in over 22 months and has
no craving to do so. He is maintained on his injection of Prolixin and his social
needs are met by the consumer run drop-in center.
Arianne began abusing cocaine following her first psychotic break in
college. Her polydrug abuse spanned 10 years but with the help of the
appointment of a guardian, enforced medication compliance and payeeship, she
gradually became engaged with her case manager. Since she was not
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 42

comfortable attending groups she and her case manager confronted the
substance abuse problem along with stabilizing her psychiatric symptoms. She
has not had any substances in over 3 years, works 10 hours a week at the
newspaper, and sees her case manager monthly.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 43

Bibliography
General References
American Psychiatric Association (1987). Diagnostic and Statistical Manual of Mental Disorders
(Third Edition - Revised) (DSM-III-R). Washington, DC: Author.
Arndt, S., Tyrrell, G., Flaum, M., & Andreasen, N.C. (1992). Comorbidity of substance abuse
and schizophrenia: The role of pre-morbid adjustment. Psychological Medicine
, 22, 379-
388.
Ananth, J., Vandewater, S., Kamal, M., Broksky, A., Gamal, R., & Miller, M.: Missed diagnosis
of substance abuse in psychiatric patients.
Hospital and Community Psychiatry
, 4, 297-
299, 1989.
Barbee, J.G., Clark, P.D., Crapanzano, M.S., Heintz, G.C., & Kehoe, C.E. (1989). Alcohol and
substance abuse among schizophrenic patients presenting to an emergency service.
Journal of Nervous and Mental Disease
, 177, 400-407.
Bartels, S.J., Drake, R.E., & McHugo, G.J. (1992). Alcohol abuse, depression, and suicidal
behavior in schizophrenia. American Journal of Psychiatry 149: 394-395.
Bartels, S.J., Drake, R.E., & Wallach, M.A. (1995). Long-term course of substance use disorders
among patients with severe mental illness.
Psychiatric Services
46: 248-251.
Bartels, S.J., Teague, G.B., Drake, R.E., Clark, R.E., Bush, P., & Noordsy, D.L. (1993).
Substance abuse in schizophrenia: Service utilization and costs.
Journal of Nervous and
Mental Disease, 181, 227-232.
Clark, R.E. (1994) Family costs associated with severe mental illness and substance use: A
comparison of families with and without dual disorders.
Hospital and Community
Psychiatry, 45, 808-813.
Cohen, M., & Klein, D.F. (1970). Drug abuse in a young psychiatric population. American
Journal of Orthopsychiatry, 40, 448-455.
Cournos, F., Empfield, M., Horwath, E., McKinnon, K., Meyer, I., Schrage, H., Currie, C., &
Agosin, B. (1991). HIV seroprevalence among patients admitted to two psychiatric
hospitals.
American Journal of Psychiatry
, 149, 1225-1229.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 44

Dixon, L., Haas, G., Weiden, P.J., Sweeney, J., & Francis, A.J. (1991). Drug abuse in
schizophrenic patients: Clinical correlates and reasons for use. American Journal of
Psychiatry, 148, 224-230.
Drake RE, Alterman AI, Rosenberg SR: Detection of substance use disorders in severely
mentally ill patients.
Community Mental Health Journal
, 29, 175-192, 1993.
Drake, R.E.; McHugo, G.M.; and Noordsy, D.L. Treatment of alcoholism among schizophrenic
outpatients: Four-year outcomes. American Journal of Psychiatry 150: 328-329, 1993.
Drake, R.E., & Mercer-McFadden, C. (1995). Assessment of substance abuse among persons
with severe mental disorder. In Lehman, A.F., Dixon, L. (eds.),
Double Jeopardy:
Chronic Mental Illness and Substance Abuse. Chur, Switzerland: Harwood Academic
Publishers (pp. 47-62).
Drake, R.E., Mueser, K.T., Clark, R.E., & Wallach, M.A. (in press). The natural history of
substance disorder in persons with severe mental illness.
American Journal of Ortho
Psychiatry.
Galanter, M., Castaneda, R. & Ferman, J. (1988). Substance abuse among general psychiatric
patients: Place of presentation, diagnosis and treatment. American Journal of Drug and
Alcohol Abuse, 14, 211-235.
Galletly, C.A., Field, C.D., & Prior, M. (1993). Urine drug screening of patients admitted to a
state psychiatric hospital.
Hospital and Community Psychiatry
, 44, 587-589.
Goodwin, F.K., & Jamison, K.R. (1990). Manic Depressive Illness. New York: Oxford,
University Press.
Hser, Y.-I., Anglin, D., & Powers, K. (1993). A 24-year follow-up of California narcotics
addicts.
Archives of General Psychiatry, 50, 577-584.
Janowsky, D.S., El-Yousef, M.K., Davis, J.M, & Sekerke, H.J. (1973). Provocation of
schizophrenic symptoms by intravenous administration of methylphenidate.
Archives of
General Psychiatry, 28, 185-191.
Knudsen, P., & Vilmar, T. (1984). Cannabis and neuroleptic agents in schizophrenia. Acta
Psychiatrica Scandinavica, 69, 162-174.
Lehman, A.F., & Dixon, L. (eds.) (1995). Double Jeopardy: Chronic Mental Illness and
Substance Abuse. Chur, Switzerland: Harwood Academic Publishers.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 45

Lieberman, J.A., Kinon, B.J., & Loebel, A.D. (1990). Dopaminergic mechanisms in idiopathic
and drug-induced psychoses. Schizophrenia Bulletin, 16, 97-110.
McLellan, A.T., Luborsky, L., O’Brien, C.P., & Woody, G.E. (1980). An improved diagnostic
instrument for substance abuse patients: The Addiction Severity Index.
Journal of
Nervous and Mental Disease, 168, 26-33.
Minkoff, K.; & Drake, R.E. (eds.) (1991). Dual Diagnosis of Major Mental Illness and Substance
Use Disorder. New Directions For Mental Health Services, 50. San Francisco: Jossey-
Bass.
Moos, R.H., Finney, J.W., & Cronkite, R.C. (1990). Alcoholism Treatment: Context, Process,
and Outcome. New York: Oxford University Press.
Mueser, K.T., Bennett, M., Kushner, M.G. (1995) Epidemiology of substance abuse among
persons with chronic mental disorders. In Lehman, A.F., & Dixon, L. (eds.), Double
Jeopardy: Chronic Mental Illness and Substance Abuse, Chur, Switzerland: Harwood
Academic Publishers (pp. 9-25).
Mueser, K.T., Drake, R.E., & Miles, K.M. (in press). The course and treatment of substance use
disorder in patients with severe mental illness. National Institute of Drug Abuse (NIDA)
Research Monographs: Comorbid Mental and Addictive Disorders: Treatment and HIV-
Related Issues.
Mueser, K.T., Nishith, P., Tracy, J.I., DeGirolamo, J., & Molinaro, M. (1995). Expectations and
motives for substance use in schizophrenia. Schizophrenia Bulletin, 21, 367-378.
Mueser, K.T., Yarnold, P.R., & Bellack, A.S. (1992). Diagnostic and demographic correlates of
substance abuse in schizophrenia and major affective disorder.
Acta Psychiatrica
Scandinavica, 85, 48-55.
Mueser, K.T., Yarnold, P.R., Levinson, D.F., Singh, H., Bellack, A.S., Kee, K., Morrison, R.L.,
Yadalam, K.Y. (1990). Prevalence of substance abuse in schizophrenia: Demographic
and clinical correlates.
Schizophrenia Bulletin
, 16, 31-56.
Noordsy, D.L., Drake, R.E., McHugo, G.J., & Biesanz, J.C. (1994). Family history of alcoholism
in schizophrenia.
Journal of Nervous and Mental Disease
, 182, 651-655.
Osher, F.C., & Kofoed, L.L. (1989). Treatment of patients with psychiatric and psychoactive
substance abuse disorders. Hospital and Community Psychiatry
, 40, 1025-1030.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 46

Pepper, B., Kirshner, M.C., & Ryglewicz, H. (1981). The young adult chronic patient: Overview
of a population. Hospital and Community Psychiatry, 32, 463-469.
Regier, D.A., Farmer, M.E., Rae, D.S., Locke, B.Z., Keith, S.J., Judd, L.L., & Goodwin, F.K.
(1990). Comorbidity of mental disorders with alcohol and other drug abuse.
Journal of
the American Medical Association, 264, 2511-2518.
Ridgely, M.S., Osher, F.C., Goldman, H.H., & Talbott, J.A. Executive Summary: Chronic
Mentally Ill Young Adults with Substance Abuse Problems: A Review of Research,
Treatment, and Training Issues. Baltimore: Mental Health Services Research Center,
University of Maryland School of Medicine, 1987.
Ritzler, B.A., Strauss, J.S., Vanord, A., & Kokes, R.F. (1977). Prognostic implications of various
drinking patterns in psychiatric patients. American Journal of Psychiatry
, 134, 546-549.
Safer, D.J. (1987). Substance abuse by young adult chronic patients. Hospital and Community
Psychiatry, 38, 511-514.
Sobell, M.B., Maisto, S.A., Sobell, L.C., Cooper, A.M., Cooper, T., & Sanders, B. (1980).
Developing a prototype for evaluating alcohol treatment outcome studies. In Sobell, L.C.,
Sobell, M.B., & Ward, E. (eds.),
Evaluating Alcohol and Drug Abuse Treatment
Effectiveness. New York: Pergamon Press (pp. 129-150).
Stone, A., Greenstein, R., Gamble, G., & McLellan, A.T. (1993). Cocaine use in chronic
schizophrenic outpatients receiving depot neuroleptic medications.
Hospital and
Community Psychiatry, 44, 176-177.
Treffert, D.A. (1978). Marijuana use in schizophrenia: A clear hazard. American Journal of
Psychiatry, 135, 1213-1215.
Vaillant G.E. (1983). The Natural History of Alcoholism. Cambridge, MA: Harvard University
Press.
Vaillant G.E. (1988). What can long-term follow-up teach us about relapse and prevention of
relapse in addiction? British Journal of Addiction
, 83, 1147-1157.
Yesavage, J.A., & Zarcone, V. (1983). History of drug abuse and dangerous behavior in inpatient
schizophrenics.
Journal of Clinical Psychiatry, 44, 259-261.
Zisook, S., Heaton, R., Moranville, J., Kuck, J., Jernigan, T. & Braff, D. (1992). Past substance
abuse and clinical course of schizophrenia.
American Journal of Psychiatry
, 149, 552-
553.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 47

Bibliography
Clinician Rating Scale References
Bartels, S.J., Drake, R.E., & Wallach, M.A. (1995). Long-term course of substance disorders
among persons with severe mental disorder.
Psychiatric Services
, 46, 248-251.
Drake, R.E., McHugo, G.J., & Noordsy, D.L. (1993). Treatment of alcoholism among
schizophrenic outpatients: Four-year outcomes. American Journal of Psychiatry, 150,
328-329.
Drake, R. E., Mueser, K.T., McHugo, G.J. Using Clinician Rating Scales to Assess Substance
Abuse among Persons with Severe Mental Illness. In Sederer, L.I., Dickey, B.
Outcomes
Assessments in Cinical Practice. Baltimore: Williams & Wilkins (in press).
Drake, R.E., & Noordsy, D.L. (1994). Case management for people with coexisting severe
mental disorder and substance use disorder.
Psychiatric Annals
, 24, 427-431.
Drake, R.E., Osher, F.C., Noordsy, D.L., Hurlbut, S.C., Teague, G.B., & Beaudett, M.S. (1990).
Diagnosis of alcohol use disorders in schizophrenia.
Schizophrenia Bulletin
, 16, 57-67.
Drake, R.E., Osher, F.C., & Wallach, M.A. (1989). Alcohol use and abuse in schizophrenia: A
prospective community study. Journal of Nervous and Mental Disease, 177, 408-414.
Drake, R.E., & Wallach, M.A. (1993). Moderate drinking among people with severe mental
illness.
Hospital and Community Psychiatry
, 44, 780-782.
McHugo, G.J., Drake, R.E., Burton, H.L., & Ackerson, T.H. (in press). A scale for assessing the
stage of substance abuse treatment in persons with severe mental illness.
Journal of
Nervous and Mental Disease.
Mueser, K.T., Nishith, P., Tracy, J.I., DeGirolamo, J., & Molinaro, M. (1995). Expectations and
motives for substance use in schizophrenia.
Schizophrenia Bulletin
, 21, 367-378.
Toolkit for Evaluating Substance Abuse in Persons with Severe Mental Illness p. 48