ArticlePDF AvailableLiterature Review

Relapse to alcohol and drug use among individuals diagnosed with co-occurring mental health and substance use disorders: A review



This paper reviews the literature investigating relapse to alcohol and drug use among individuals dually diagnosed with a substance use and a co-occurring mood, anxiety, schizophrenia-spectrum, or personality disorder. Prevalence rates for each co-occurring set of disorders are discussed, followed by research studies that examine predictors of relapse to substance use within these groups. Relevant conceptual models well-suited to incorporating relapse as an outcome variable, and psychiatric factors both as predictor and outcome variables, are presented. Suggestions for future studies are provided. A priority area is developing and using consistent and well-articulated definitions of relapse across studies. Several diagnostic issues surfaced such as using structured clinical interviews to determine diagnosis (preferably following detoxification from alcohol and/or drugs), separating individuals with only alcohol use disorders from those with alcohol and drug use disorders in analyses, reporting the rates and types of overlap in mental health diagnoses, and conducting analyses that include and exclude multiply disordered individuals. Finally, future studies that focus on isolating predictors of relapse and abstinence could make substantive contributions to improving treatment for individuals with co-occurring substance use and mental health disorders.
Relapse to alcohol and drug use among individuals diagnosed with
co-occurring mental health and substance use disorders: A review
Clara M. Bradizza *, Paul R. Stasiewicz, Nicole D. Paas
Research Institute on Addictions, University at Buffalo, United States
This paper reviews the literature investigating relapse to alcohol and drug use among individuals dually diagnosed with a
substance use and a co-occurring mood, anxiety, schizophrenia-spectrum, or personality disorder. Prevalence rates for each co-
occurring set of disorders are discussed, followed by research studies that examine predictors of relapse to substance use within
these groups. Relevant conceptual models well-suited to incorporating relapse as an outcome variable, and psychiatric factors both
as predictor and outcome variables, are presented. Suggestions for future studies are provided. A priority area is developing and
using consistent and well-articulated definitions of relapse across studies. Several diagnostic issues surfaced such as using
structured clinical interviews to determine diagnosis (preferably following detoxification from alcohol and/or drugs), separating
individuals with only alcohol use disorders from those with alcohol and drug use disorders in analyses, reporting the rates and types
of overlap in mental health diagnoses, and conducting analyses that include and exclude multiply disordered individuals. Finally,
future studies that focus on isolating predictors of relapse and abstinence could make substantive contributions to improving
treatment for individuals with co-occurring substance use and mental health disorders.
D2005 Elsevier Ltd. All rights reserved.
Keywords: Drug use; Co-occurring mental health; Relapse; Alcohol use
1. Introduction
Rates of substance abuse among individuals dually diagnosed with a mental health and substance use disorder are
exceedingly high, with estimates as high as 50% for individuals diagnosed with a bipolar or schizophrenic disorder
(Kessler et al., 1997; Mueser et al., 1990). These rates highlight the magnitude of the problem faced by dually
diagnosed individuals, treatment providers, and the larger treatment community who are collectively engaged in the
process of reducing relapse to alcohol and drug use in this difficult-to-treat population. Given the scope of substance
abuse among mentally ill individuals and its associated problems, several important questions arise. Are individuals
able to maintain abstinence or nonproblem levels of substance use following treatment? If not, what are the rates of
relapse to substance use for individuals with different mental health disorders? Which factors predict relapse to
substance use vs. the maintenance of abstinence?
Among individuals diagnosed with a substance use disorder but with no assessed mental illness, rates of relapse
vary by time since treatment. At 3 months post-treatment, 40–60% of individuals in treatment for alcohol problems
0272-7358/$ - see front matter D2005 Elsevier Ltd. All rights reserved.
* Corresponding author.
E-mail address: (C.M. Bradizza).
Clinical Psychology Review 26 (2006) 162 – 178
relapse to a first drink, whereas by 12 months this rate increases to 70–80% (Hunt, Barnett, & Branch, 1971;
Lowman, Allen, Stout et al., 1996). For heroin users and smokers, the 3-month rate of relapse to first use is about 60%
and the 12-month rate is approximately 75% (Hunt et al., 1971).
This review paper attempts to examine relapse rates among individuals dually diagnosed with a substance use
disorder and an Axis I and/or Axis II mental disorder, including affective disorders (i.e., major depression, bipolar
disorder), anxiety disorders (i.e., post-traumatic stress disorder [PTSD], panic disorder, social anxiety disorder [SAD],
obsessive–compulsive disorder [OCD]), personality disorders (i.e., antisocial personality disorder [APD], borderline
personality disorder [BPD]), or a schizophrenia-spectrum disorder. Most studies reviewed examine relapse to
substance use among individuals diagnosed with a mental illness that were subsequently assessed for a substance
use disorder (SUD). Only rarely are studies cited in which individuals present for treatment of a SUD and are
determined to have a mental illness. One possible explanation for this bias is that most studies of individuals
diagnosed with an SUD exclude persons with a mental illness or do not assess for these disorders. As a result, the
majority of studies examining relapse among the dually diagnosed are among those presenting for treatment of a
mental illness.
2. Mood disorders
2.1. Comorbidity of mood and substance use disorders
As a category, mood disorders (e.g., major depression, dysthymia, bipolar disorder) are the most prevalent Axis I
mental disorder in the general population (Regier et al., 1988). Epidemiological studies assessing lifetime prevalence
of substance use and mood disorders have found high rates of comorbidity. The Epidemiological Catchment Area
study (ECA; Regier et al., 1990) reported that 32% of individuals with any affective disorder were comorbid for a
substance use disorder. More specifically, 27% of individuals with major depression and 56% with bipolar disorder
met criteria for an alcohol or drug use disorder. Rates of alcohol use disorders (AUD) were similarly high, with 16%
of those diagnosed with major depression and 43% with bipolar meeting criteria. A second study, the National
Comorbidity Survey (NCS; Kessler et al., 1997) found that among individuals with an alcohol dependence disorder,
53% of women and 28% of men met criteria for any affective disorder diagnosis. Among the affective disorders, 48%
of women and 24% of men met criteria for major depression, and 6% of both men and women were diagnosed with a
manic disorder. Therefore, about one-half of women and one-fourth of men with an AUD are also grappling with
major depression and about 1 in 20 are diagnosed with bipolar disorder. The vast majority of the literature examining
substance abuse among individuals with affective disorders has focused on the relationship between depression and
SUDs, reflecting the much higher prevalence rates as compared with bipolar disorder.
2.2. Depression and relapse to alcohol use
The high comorbidity rates for depression and substance use has fueled a substantial literature in this area with an
emphasis on the effects of depression on alcohol consumption and problems. However, much less research has
focused on examining the impact of depression on alcohol relapse. Overall, the results of these studies tend to support
the idea that depression negatively impacts relapse to alcohol use; however, support for this finding is equivocal. A
few studies have enrolled only men while others have enrolled both genders. Two articles have focused on examining
the relationship between depression and alcohol relapse among men over a 12-month post-treatment period. Curran,
Flynn, Kirchner, and Booth (2000) recruited 298 men from a Veteran’s Administration (VA) hospital inpatient
substance abuse treatment program. Those classified as mildly-to-moderately depressed on the Beck Depression
Inventory (BDI score = 14–19; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961) at 3 months post-treatment were
nearly three times more likely to relapse to alcohol use over the follow-up period as compared to nondepressed men.
Men who were severely depressed (BDI score = 20+) were nearly five times more likely to relapse. In addition, the
timing of the depressive symptoms and relapse was important. Depressive symptoms that occurred following alcohol
treatment or a period of abstinence were much better predictors of relapse than depression early in treatment (Curran
& Booth, 1999).
These results are supported by studies that have recruited samples of both men and women (Bobo, McIlvain,
& Leed-Kelly, 1998; Glenn & Parson, 1991; Greenfield et al., 1998). Greenfield and colleagues followed 61 men
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 163
and 40 women for 1 year following discharge from inpatient alcohol treatment. They found that a diagnosis of
major depression at treatment entry was associated with both a shortened time to first drink and relapse (defined
as 3 or more drinks per drinking day for women and 5 or more for men). However, no relationship was found
between depressive symptomatology (i.e., BDI score) and resumption of alcohol use. Similarly, Hasin et al.
(1996) used survival analysis with time-dependent covariates to assess the effects of remissions and relapses to
major depressive disorder on alcohol relapse. This analytic technique allows the investigators to assess the effects
of certain predictor variables that change over time (i.e., depression diagnosis status) on a given outcome (i.e.,
alcohol relapse). Depression and alcohol use disorder diagnoses were assessed in a sample of 127 (76 men and
51 women) individuals using the Schedule for Affective Disorders and Schizophrenia (SADS; Endicott & Spitzer,
1978). They found that changes in depression diagnosis status over a 5-year period significantly predicted relapse
to alcohol use such that improvements in depression status resulted in a reduced chance of a return to alcohol
use. Rather than using a depression diagnosis, two studies that assessed the relationship between depressive
symptomatology and alcohol relapse following treatment found that depressive symptoms were a significant
negative predictor of alcohol relapse. Bobo et al. (1998) administered the Center for Epidemiological Studies
Depression Scale (CES-D; McDowell & Newell, 1996; Radloff, 1977) to 385 men and 190 women receiving
residential substance abuse treatment. At the 12-month follow-up, the relative risk of alcohol relapse was between
1.21 and 1.28 for those who were depressed at study intake, indicating a significantly increased risk of relapse.
Similarly, a study enrolling 58 men and 45 women examined the ability of depressive symptoms, neuropsycho-
logical performance, psychosocial adjustment, treatment history, and childhood attention deficit disorder symp-
tomatology to predict alcohol relapse 14 months following discharge from inpatient treatment (Glenn & Parson,
1991). The authors found that depressive symptomatology (as assessed by BDI score at study intake) was the best
single predictor of alcohol relapse.
These studies have some commonalities, including a follow-up period of about 1 year post-treatment and an
assessment of depressive symptomatology or a depression diagnosis that is relatively close in time to the follow-up
period in which alcohol relapse is being assessed. Taken together, these studies indicate that depression is a successful
predictor of earlier relapse to alcohol use. However, not all studies have found significant relationships between
depression and alcohol relapse.
Five studies found no significant impact of depression on drinking relapse following alcohol and/or drug treatment.
These studies shared some common characteristics. Most utilized a structured diagnostic interview (Hasin, Endicott,
& Keller, 1989; Powell et al., 1992; Sellman & Joyce, 1996) at study intake to determine the presence of an alcohol
and major depression diagnosis. Lifetime depression criteria were most commonly used, with only one study relying
on a current diagnosis of depression (Hasin et al., 1989). The length of the post-treatment follow-up period ranged
from 6 to 24 months. Two of the studies recruited only men (Powell et al., 1992; Sellman & Joyce, 1996) while the
three others recruited mixed gender samples (Hasin et al., 1989; Miller, Hoffman, Ninonuevo, & Astrachan, 1997;
Miller, Klamen, Hoffman, & Flaherty, 1996).
In sum, studies that found significant relationships between depression and alcohol relapse were more likely to
have used depression symptomatology or a diagnosis that was assessed relatively close in time to alcohol relapse.
Studies finding no significant relationship were more likely to have used a lifetime diagnosis of depression that was
assessed at study intake, which occurred months or even years prior to the follow-up period during which the authors
were attempting to predict alcohol relapse.
2.3. Depression and relapse to cocaine use
The majority of the literature examining the impact of depression on drug relapse has focused on cocaine use. The
four identified studies (Alterman et al., 2000; Brown et al., 1998; Carroll, Nich, & Rounsaville, 1995; McKay et al.,
1997) examined the impact of a diagnosis of major depression or depressive symptomatology on relapse to cocaine
use either at the end of treatment (Carroll et al., 1995) or during a post-treatment follow-up period. The follow-ups
were relatively brief, ranging from 3 months (Brown et al., 1998) to 6–7 months (Alterman et al., 2000; McKay et al.,
1997). One study found that depression was significantly related to a reduced likelihood of relapse to cocaine use.
McKay et al. (1997) provided either intensive outpatient or standard group counseling aftercare to 98 VA men who
completed an intensive outpatient program. They examined a number of possible predictors of cocaine use during the
6-month follow-up period, including depression. Participants who had a lifetime diagnosis of major depression were
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178164
more likely to remain abstinent during the first 6 months of aftercare, as compared with those who were not
In contrast, three studies have found no evidence of a relationship between depression and cocaine relapse.
Carroll et al. (1995) provided 12 sessions of cognitive–behavioral treatment and desipramine (alone and in
combination) to 109 (80 men and 29 women) patients diagnosed with a cocaine dependence disorder. They
administered both the BDI and the Hamilton Rating Scale for Depression (HAM-D; Hedlund & Vieweg, 1979)to
assess extent of depressive symptoms and determined that 35% (37) of the participants had at least mild depressive
symptomatology. Although the investigators did not assess relapse, they did measure consecutive days of
abstinence during the treatment period. At post-treatment, depressed individuals did not differ from non-depressed
individuals regarding the number of consecutive days of cocaine abstinence. Alterman et al. (2000) followed 160
VA men who completed either a 1-month intensive day hospital or 1-month inpatient drug treatment program for 7
months. They found that a diagnosis of lifetime major depression was unrelated to cocaine relapse during the
follow-up period. Similarly, a study tracking 65 men and 24 women through the first 3 months following 2 weeks
of either residential or day hospital substance abuse treatment found that both a diagnosis of current major
depression and HAM-D scores did not predict cocaine relapse status at either the 1-month or 3-month follow-ups
(Brown et al., 1998). Overall, these studies provide little evidence of a significant relationship between depression
and relapse to cocaine use; however, one study did find a significant negative relationship and further studies are
warranted. A shortcoming common to all studies was brief follow-up periods; longer periods (especially 2+ years)
would be useful in gauging longer-term effects of depression on cocaine relapse. In addition, three of the five
studies recruited only men. More studies that emphasize recruitment of substantial samples of women are needed in
order to assess the extent to which these findings apply to women. If future studies were to support a negative
relationship, the mechanisms by which greater depression reduces cocaine relapse would be a useful and potentially
productive avenue of investigation.
2.4. Depression and relapse to opioids/other drugs
One study (Kosten, Rounsaville, & Kleber, 1986) has examined the effect of depression on relapse to opioids. In
this study, 268 opiate-addicted individuals (196 men and 72 women) were followed for 2.5 years following treatment
intake. At treatment admission, the results of the SADS interview determined that 26% (69) of the sample met criteria
for a current major depressive disorder. No significant relationship was found between depression and relapse to
opioid use during the follow-up period.
Two studies examined the role of depression in relapse to substance use among samples of individuals with alcohol
and a variety of drug use disorders. A study by Charney, Paraherakis, Negrete, and Gill (1998) supports the findings
of several other studies that a depression diagnosis results in a longer duration of abstinence post-treatment.
Interviews were conducted with 75 (46 men and 29 women) individuals entering substance abuse treatment who
were diagnosed using the Structured Clinical Interview for DSM-IV (SCID; Spitzer, Williams, Gibbons, & First,
1994) with major depression (14%), dysthymia (7%) or bipolar disorder (4%), and a primary alcohol (53%), cocaine
(23%), benzodiazepine (11%), or heroin (9%) use disorder. Participants were interviewed at treatment entry and again
at 3 months. At both assessment points, the BDI and the HAM-D scales were administered to assess depression. At
the 3-month follow-up, individuals who were diagnosed with depression or who had high scores at intake on the
HAM-D attained a longer duration of abstinence with their primary substance of abuse. In contrast, a study examining
the impact of a lifetime diagnosis of depression (as assessed by clinical interview) on relapse found no significant
impact of depression at the 3-month follow-up (Miller et al., 1999). Participants were 2029 (1501 men and 528
women) individuals attending evening outpatient substance abuse treatment at one of 33 treatment centers. Thirty-
eight percent were diagnosed with a mood disorder (i.e., depression, dysthymia, bipolar disorder) and substance
dependence diagnoses included alcohol (76%), marijuana (16%), and cocaine (13%). Analyses indicated that lifetime
depression diagnosis did not significantly predict substance abuse relapse or continuous abstinence during the follow-
up period.
Very few studies have been conducted with opiate-addicted, depressed individuals to draw any conclusions. More
research is needed assessing the effect of depression on relapse outcomes, particularly long-term abstinence and
relapse. Additionally, it is difficult to draw conclusions based on the results of studies with participants who have
mixed substance use disorders.
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 165
2.5. Bipolar disorder and relapse to alcohol and drug use
As noted earlier, bipolar disorder has the highest rate of co-morbidity with substance use disorders with
more than half of bipolar individuals also diagnosed with an alcohol or drug use disorder. Despite this
exceeding high rate of comorbidity, no studies were found that examined relapse to alcohol or drugs among
bipolar individuals. The vast majority of published studies have continued to examine the prevalence rates
of alcohol and/or drug use problems among these individuals (Chengappa, Levine, Gershon, & Kupfer,
2000; Frye et al., 2003; Reich, Davies, & Himmelhoch, 1974). Progress in this area would be furthered by
complementary qualitative and quantitative studies that focus on determining appropriate measures and
examining relapse episodes in more detail. For example, qualitative studies could serve to provide valuable
information regarding appropriate existing measures in assessing relapse and could also provide invaluable
assistance in measurement development. Subsequent quantitative studies assessing rates of relapse to alcohol
and drug use among bipolar individuals during and following treatment, predictors of substance use relapse,
and predictors of abstinence would provide valuable information in setting the stage for appropriate
3. Anxiety disorders
3.1. Comorbidity of anxiety and alcohol use disorders
The ECA survey (Regier et al., 1990) found that individuals with any anxiety disorder had a 50% increase in the
odds of being diagnosed with a lifetime alcohol use disorder (alcohol abuse or dependence; odds ratio [OR] = 1.5).
With regard to specific anxiety disorders, panic disorder presented a much greater increase in the odds of being
diagnosed with alcohol dependence (OR =3.3). For obsessive–compulsive disorder, the odds ratio associated with
alcohol dependence was somewhat lower at 2.5.
Similarly, the NCS study (Kessler et al., 1997) found that the risk of alcohol dependence was greater for women
(OR = 2.9) and men (OR = 2.3) with panic disorder and men (OR = 3.9) and women (OR = 3.0) with generalized
anxiety disorder. Interestingly, both the ECA and NCS surveys found more significant comorbid relationships with
alcohol dependence as compared to alcohol abuse alone.
In a third investigation, Kushner, Sher, and Erikson (1999) conducted a prospective study of the relationship
between anxiety and alcohol use disorders in male and female college students assessed during freshman (year
1), and then again at years 4 and 7. Prospectively, the odds of developing a new alcohol dependence disorder
at year 7 increased from 3.5 to 5.0 for those diagnosed with an anxiety disorder at years 1 and 4.
Interestingly, the cross-sectional data from this study revealed comorbidity rates comparable to those from
the ECA and NCS community-based surveys previously mentioned. For example, there was a two-to-threefold
increase in the risk of incurring either an alcohol use or an anxiety disorder given the presence of the other
comorbid condition.
In summary, the three survey studies reviewed document similar rates of comorbidity between anxiety and alcohol
use disorders. In most cases, the relationship between alcohol use and anxiety disorders appears to be stronger for
alcohol dependence as compared with alcohol abuse alone. The following sections will review studies that have
examined relapse to substance use for specific anxiety disorders.
3.2. Comorbidity of post-traumatic stress and substance use disorders
Among men with a lifetime diagnosis of an AUD, Kessler et al. (1997) found that the odds of having PTSD were
0.45 for those with an alcohol abuse disorder as compared with 3.30 for those with alcohol dependence. Among
women, the odds of having PTSD were 1.01 for those diagnosed with alcohol abuse and 3.60 for those with an
alcohol dependence diagnosis. Although the odds ratios for alcohol abuse were larger for women as compared with
men, this difference was not statistically significant. These data indicate that PTSD is more likely to occur among
individuals diagnosed with an alcohol dependence disorder as compared with an alcohol abuse disorder. Clinical
studies support these findings indicating that approximately one third of patients with an SUD have a comorbid PTSD
diagnosis (Ouimette, Moos, & Brown, 2002).
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178166
3.3. Post-traumatic stress disorder and relapse to substance use
Although there are few published studies examining relapse to substance use among SUD–PTSD individuals, the
results consistently indicate that the presence of comorbid PTSD is associated with worse substance use outcomes. In
one of the first published studies, Brown, Stout, and Mueller (1996) compared alcohol and drug use outcomes among
substance dependent women (N= 31) with and without comorbid PTSD following inpatient substance abuse
treatment. Approximately 45% of the women (N= 14) received a DSM-III-R diagnosis of alcohol dependence
only, 19% (N=6) wer e diagnosed with substance dependen ce other than alcohol, and 36% (N= 11) were polysub-
stance dependent. Forty-two percent of the sample (N= 13) met criteria for current PTSD. At the 3-month post-
treatment interview, 70% of the women were determined to have relapsed, defined as any use of alcohol and/or drugs.
Although relapse rates did not differ by PTSD status, PTSD women did relapse more rapidly than non-PTSD women.
In contrast, psychiatric distress was not found to be a significant predictor of substance abuse relapse. Based on the
results of this pilot study, Brown and colleagues speculated that failure to treat concurrent trauma predisposes PTSD
substance abusers to return quickly to using substances to manage their PTSD symptoms. Brown (2000) examined 6-
month outcomes for 29 SUD–PTSD women receiving inpatient substance abuse treatment. All participants met
criteria for some form of current substance dependence as determined by the substance dependence modules of the
SCID-IV. Approximately 35% received a diagnosis of alcohol dependence only, 35% were diagnosed with drug
dependence only, and 31% were dependent on both alcohol and drugs. In addition, participants were diagnosed with
PTSD as determined by the Clinician Administered PTSD Scale (CAPS; Blake et al., 1995). At the 6-month post-
discharge interview, approximately half the women (52%) had relapsed at some point to alcohol and/or drug use.
Relapse was defined as 7 or more days of heavy drinking or any drug use during the follow-up period. Based on the
CAPS, participants were classified at follow-up as either unremitted (i.e., they still met diagnostic criteria for PTSD)
or remitted (i.e., they no longer met diagnostic criteria for PTSD). Approximately one-quarter of the sample (24%)
had remitted from PTSD at follow-up. Both substance use (e.g., percent days abstinent, number of years of
problematic use) and PTSD variables (e.g., baseline severity of re-experiencing, avoidance/numbing) were examined
as predictors of relapse status at follow-up. The only significant predictor of substance use relapse was the baseline
severity of PTSD re-experiencing symptoms. The odds of relapsing at follow-up increased by a factor of 2 for each
additional re-experiencing symptom, up and above the first symptom. Also, this same variable predicted PTSD status
(remitted vs. unremitted) at follow-up. For each additional re-experiencing symptom above the required one
symptom, the odds of not remitting from PTSD increased by a factor of 2.8. Although this study is limited to a
small sample of female patients, the results suggest that re-experiencing symptoms may play an important role in
determining relapse to alcohol and drug use for individuals with comorbid PTSD and a SUD.
As part of a multisite evaluation of VA substance abuse treatment, Ouimette, Moos, and Finney (1997) followed
the post-treatment course of male patients diagnosed with SUD–PTSD (N= 159) compared to patients with a SUD
only (N= 2497) and those with other comorbid Axis I psychiatric disorders (SUD–PSY; N= 314). Although there
were three waves of follow-up (i.e., 1, 2, and 5 years), only the results of the 2- and 5-year post-treatment follow-ups
are reviewed below as they specifically include information about patients’ SUD remission status (i.e., relapse).
Remission was defined as abstinence from illicit drug use and either alcohol abstinence or non-problem alcohol use.
To be categorized as remitted, a patient must have (a) abstained from all 13 assessed drugs, (b) had no problems
related to drug or alcohol abuse, and (c) consumed 3 oz (88.79 ml) or less of alcohol per day on maximum drinking
days in the past month. The results of the 2-year follow-up revealed that SUD–PTSD patients had poorer long-term
substance use, psychiatric, and psychosocial outcomes than SUD-only and SUD–PSY patients (Ouimette, Finney, &
Moos, 1999). With regard to substance use outcomes, SUD–PTSD patients were less likely to be in remission than the
SUD–PSY patients, and a trend emerged for the SUD–PTSD patients to be less likely to have remitted than the SUD-
only patients. Taken together, these findings suggest that poorer outcomes may be specific to PTSD rather than to
psychiatric comorbidity in general. Moreover, the increased psychiatric distress experienced by the SUD–PTSD
group places them at increased risk for substance abuse relapse over the long term.
In an extension of the 2-year follow-up study, Ouimette, Moos, and Finney (2003) examined predictors of
longer term (5 years) remission in 100 male SUD–PTSD inpatients. This report focused on only those patients with
comorbid PTSD who completed all three waves of follow-up as described above. Overall, the receipt of PTSD-
focused treatment shortly after discharge from substance abuse treatment (i.e., during the first 3 months) predicted
patients’ SUD remission status 5 years later. Also, attendance at the 12-Step self-help groups in year 1 was
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 167
positively associated with 5-year remission. This finding extends their earlier work, where 12-Step involvement in
year 1 was positively associated with remission at either or both the 1-year and 2-year follow-ups (Ouimette et al.,
2001). Finally, the receipt of SUD outpatient care in the first year did not predict later remission. Based on these
findings, the authors suggest that receipt of PTSD-focused treatment immediately after SUD treatment may
enhance long-term remission.
In summary, SUD–PTSD patients appear to relapse more quickly following substance abuse treatment, severity of
PTSD re-experiencing symptoms is associated with relapse to substance use, and receipt of PTSD treatment shortly
after SUD treatment is related to better substance abuse outcomes (i.e., remission) 5 years later. There also is evidence
to suggest that improvements in PTSD symptoms may be an important factor in the remission of substance use
disorders. Brown, Read, and Kahler (2003) classified patients into three groups: (1) PTSD unremitted (i.e., met
diagnostic criteria at baseline and follow-up); (2) PTSD remitted (i.e., met diagnostic criteria at baseline but not at
follow-up); and (3) No PTSD control group. After controlling for baseline levels of psychiatric distress and percent
days abstinent, patients with unremitted PTSD were found to have poorer substance use outcomes than those who
were never diagnosed with PTSD. There were no significant differences between the PTSD remitted group and the No
PTSD group. These data support the use of integrated mental health and substance abuse treatments for SUD–PTSD
patients. A study that employs random assignment of SUD–PTSD patients to interventions would be an important
next step in this area. Ideally, future studies should include both genders, employ multiple measures of relapse (e.g.,
time to first drink, time to first heavy drinking day) and include treatment process measures that have the potential to
elucidate the potentially complex relationships between changes in PTSD symptom status and relapse to alcohol and
drug use.
3.4. Comorbidity of panic and substance use disorders
Among those individuals with anxiety disorders in the ECA study, those with panic disorders with and without
agoraphobia (PD) presented the greatest odds of being diagnosed with any substance use disorder (OR = 2.9; Regier et
al., 1990). A second line of evidence supporting the strong relationship between SUD and PD comes from research on
the prevalence of anxiety disorders among substance abuse patients. In a review of studies examining the prevalence
of panic-related anxiety among alcoholics, Cox, Norton, Swinson, and Endler (1990) reported that between 9% and
60% of the alcoholic samples reviewed met criteria for agoraphobia or panic disorder. Despite studies demonstrating
the high co-occurrence of panic and substance use disorders, little systematic study of the impact of panic disorder on
substance abuse treatment outcomes has been conducted.
3.5. Panic disorder and relapse to alcohol use
Published studies to date have examined the relationship between a diagnosis of panic disorder or symptomatology
and relapse to alcohol use–no studies were found that examined relapse to drug use. In a study of patients admitted to
a residential substance abuse treatment facility, LaBounty, Hatsukami, Morgan, and Nelson (1992) assessed for
phobic and panic symptoms during a 6-month post-treatment interview. They identified 35 alcoholics who reported
symptoms of phobia, panic, or both. This group was compared to 35 matched controls who indicated no anxiety
problems. Although more individuals in the anxiety problem group reported relapsing to cope with negative emotions
(e.g., tension, depression), rates of relapse did not differ between the two groups. At the 6-month follow-up, 34%
(N= 12) of the anxiety group and 29% (N= 10) of matched controls had relapsed. Unfortunately, several methodo-
logical problems, including the absence of a valid and reliable diagnostic measure and the failure to define relapse,
limit the conclusions that can be drawn from this study.
In a prospective study of a representative national sample of patients seeking detoxification from alcohol in
Iceland, those with agoraphobia or panic disorder reporting for admission for the first or second time had a greater
than fivefold increased risk for readmission (for detoxification) within the 28-month follow-up period (Tomasson &
Vaughn, 1998). A strength of this study was the use of a structured clinical interview to determine diagnoses.
However, the diagnostic interviews were conducted immediately following the acute withdrawal period (4 or more
days after admission) and this may have increased the rate of anxiety disorder diagnoses reported in this study. For
example, it is possible that many alcoholics demonstrate anxiety disorders caused by intoxication or withdrawal from
alcohol, and these anxiety states are likely to improve markedly during the first several weeks to 1 month of
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178168
abstinence (Schuckit, 1996). Therefore, some uncertainty exists as to whether the increased risk for readmission was
related to the co-occurring anxiety disorder or to a more severe form of alcohol dependence or withdrawal.
In a controlled trial of CBT conducted with 231 alcoholics with comorbid panic disorder, Bowen, D’Arcy, Keegan,
and Senthilselvan (2000) randomly assigned patients entering a 4-week alcohol inpatient treatment program to either
standard alcohol treatment alone (control) or to a group receiving CBT for panic disorder plus standard alcohol
treatment. The CBT consisted of 6 sessions (12 h) of group-based treatment for panic management training. Problem
drinking and anxiety symptoms were measured at baseline and at 3, 6, and 12 months post-treatment. Relapse was
defined as any use of alcohol during the prior 3-month period. At the 12-month follow-up, the relapse rates for the
two groups did not differ. In addition, there was comparable improvement of symptoms of panic and agoraphobia.
The authors’ comment that active anxiety-treatment ingredients in the standard alcohol treatment program (e.g.,
relaxation, discussion of ways to cope with stress) may have limited the ability to detect differences between the
Despite the prevalence of substance use disorders among individuals with panic disorder, there is a dearth
of research investigating relapse to alcohol use and no studies examining relapse to drug use among
individuals diagnosed with SUD–PD following treatment. Future studies should be focused on using structured
clinical interviews to assess diagnoses that occur several weeks following detoxification. In addition, studies
conducting randomized clinical trials should carefully delineate general alcohol treatment from treatment for
panic disorder.
3.6. Comorbidity of social anxiety and substance use disorders
Data from the NCS study indicates that among men with a lifetime diagnosis of an AUD, the odds of having social
phobia were .085 for those with alcohol abuse and 3.11 for those with alcohol dependence disorder. Among women,
the odds of having social phobia were 2.24 for those with alcohol abuse and 2.63 for those with alcohol dependence.
Again, gender differences in the odds ratios for alcohol abuse and alcohol dependence were not statistically
3.7. Social anxiety disorder and relapse to alcohol use
Similar to panic disorder, all published studies located for this review examined the relationship of social anxiety
disorder (also referred to as social phobia) and relapse to alcohol use–no published studies were found that included
drugs. Two studies have examined remission or relapse among individuals with comorbid AUD and SAD disorders.
In a retrospective study, Thevos, Roberts, Thomas, and Randall (2000) used the Project MATCH data set (Project
MATCH Research Group, 1997) to compare the drinking outcomes of 397 AUD–SAD individuals seeking outpatient
alcohol treatment with a matched sample of 397 alcoholics without SAD. Patients were assigned (albeit not randomly)
to one of three treatment conditions: (a) Cognitive–Behavioral Therapy (CBT), (b) Twelve-Step Facilitation Therapy
(TSF), or (c) Motivational Enhancement Therapy (MET). The study was designed to test the hypothesis that SUD–
SAD patients treated with CBT would have better drinking outcomes than those treated with TSF; MET patients were
not included in the analyses. Female outpatients with SAD delayed relapse to drinking (defined either as time to first
drink or time to first heavy drinking day) when treated with CBT as compared with TSF; the reverse was true for
female outpatients with an AUD only. There were no between-group differences for male outpatients. Thus, the
hypothesis that individuals with SUD–SAD would have better drinking outcomes when treated with CBT rather than
TSF was supported for women but not for men. It is important to note that patients were not randomly assigned to
treatment condition in this study. Moreover, CBT did not specifically address social phobia and patients were only
offered 8 out of a possible 12 sessions; therefore, they received only the core CBT sessions and did not receive the
sessions targeting mood disturbance. As a result, it was unclear whether CBT that targets both SUD and SAD would
result in better treatment outcomes than CBT for alcoholism only.
In a subsequent article, Randall, Thomas, and Thevos (2001) investigated this question. They proposed that
simultaneous treatment of SAD and alcoholism, as compared with treatment of alcoholism alone, would improve
alcohol use and social anxiety outcomes in comorbid patients. Ninety-three individuals with SAD seeking outpatient
treatment for an AUD were randomly assigned to receive either 12 weeks of individual CBT for alcoholism only or
concurrent treatment for both alcohol and social anxiety disorder. Outcome data were collected at the end of treatment
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 169
and at 3 months post-treatment. The alcohol-only treatment group received twelve 60-min CBT sessions modeled
after the CBT therapy manual used in Project MATCH (Project MATCH Research Group, 1997). The dual treatment
group received twelve 90-min sessions that addressed both social anxiety and alcohol use disorders. No differences
were found between the two treatment groups in treatment attendance or completion. With regard to drinking
outcomes, both treatment groups improved on drinking measures at the end of treatment assessment. At the 3-month
follow-up, patients who received the dual treatment had worse outcomes than patients in the alcohol-only treatment
group with respect to percentage days abstinent, percentage days heavy drinking, and total number of drinks
consumed. Time-to-event measures (e.g., time to first drink, time to first heavy drinking day) were not included in
this study. Overall, the results were counter to the original hypothesis, which states that the dual treatment group
would demonstrate superior outcomes, presumably because treatment of the social phobia would decrease social
anxiety thereby reducing the need to self-medicate with alcohol. One possible explanation is that the dual treatment
protocol included homework assignments that required exposure to anxiety-eliciting social situations; therefore,
patients in this group may have engaged in drinking to cope with these situations. However, if this were true, then the
problem may have been a failure to achieve optimal exposure (viz., anxiety reduction) to the anxiety eliciting cues. To
the extent that drinking provides relief from anxiety, the process of extinction is interrupted and the behavior of
drinking is negatively reinforced by the reduction in anxiety. Therefore, future studies will need to conduct a more
sensitive assessment of the relationship between in vivo exposure and alcohol consumption. Another possibility is
that the dual treatment group fared worse because of the longer treatment sessions (90 min vs. 60 min for the alcohol-
only group). For example, the authors speculated that patients in the dual treatment group may have lost interest or
had poorer attention during treatment sessions, although the treatment completion rates do not support this contention.
These studies provide some initial data regarding alcohol relapse among individuals with social anxiety disorder.
However, studies are needed that specifically assess time-to-relapse variables such as time to first drink or time to first
heavy drinking day. In addition, studies that include drugs, particularly more commonly used substances such as
cocaine and marijuana, would make a significant contribution to this area.
3.8. Other anxiety disorders and relapse to substance use
In a randomized, placebo-controlled double-blind study, Tollefson, Montague-Clouse, and Tollefson (1992)
investigated the use of buspirone (a selective serotonergic agonist) in the treatment of 51 recently detoxified
alcoholics with comorbid generalized anxiety disorder (GAD). Diagnoses of GAD were completed after a minimum
of 4 weeks of abstinence from alcohol had been achieved. Eligible participants were randomized to receive either 5
mg of buspirone or a placebo daily. Ten follow-up visits occurred over a 24-week period at which time measures of
alcohol use, alcohol craving, anxiety, serum buspirone levels, and toxicology screens were obtained. Relapse was
defined by positive toxicology screens for both alcohol and other drugs. Of the 14 subjects (27%) completing all 24
weeks of treatment, three patients had a positive serum alcohol screen, two were positive for marijuana, and one was
positive for cocaine. There were no significant differences between groups in positive toxicology screens.
Fals-Stewart and Schafer (1992) examined relapse to substance use in 60 individuals dually diagnosed with SUD
and obsessive–compulsive disorder (OCD). Individuals were voluntarily admitted to a drug-free therapeutic com-
munity and were randomly assigned to one of three treatment conditions. One group received a combined intervention
that addressed their OCD symptoms and substance abuse; a second group received only substance abuse treatment;
and an attention control group received substance abuse treatment and training in progressive muscle relaxation.
Substance abusers with OCD who participated in the combined intervention group had improved outcomes compared
to the other two groups – they stayed in treatment longer, demonstrated a greater reduction in OCD symptom severity,
and were more likely to remain abstinent during a 1-year follow-up. In addition, among patients who did relapse
during the follow-up period, those in the combined treatment group had longer periods of abstinence following
discharge. Completing the diagnostic assessment 1 month following admission and detoxification likely improved the
accuracy of the diagnoses and is a notable strength of this study.
3.9. Comorbid anxiety and substance use disorders: summary and future directions
Although anxiety disorders are common among individuals with a SUD, there are few randomized controlled trials
that address relapse among this comorbid population. The most common methodological problem in this area is the
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178170
timing of the anxiety disorder diagnosis session, as symptoms of alcohol dependence and withdrawal overlap with
anxiety disorder symptoms. Some have suggested postponing the diagnosis of anxiety disorders for 1 month
following treatment entry (Schuckit, 1996). This would potentially decrease the number of false positive diagnoses
and reduce the heterogeneity of comorbid treatment samples. Future studies should also include multiple measures of
relapse and should begin to establish the base rates of remission from anxiety disorder symptoms that occur with only
substance abuse treatment. For example, it is possible that existing empirically supported substance abuse treatments
(e.g., CBT, 12-Step) include active ingredients (e.g., stress reduction coping skills, discussion of negative affect
situations) that contribute to the reduction of anxiety symptoms. Also, fairly high drop-out rates among the
experimental treatment group are not uncommon (e.g., Bowen et al., 2000), suggesting that future research investigate
individuals’ motivations for addressing the anxiety disorder and the staging or timing of the substance abuse treatment
vs. anxiety disorder treatment.
4. Personality disorders
4.1. Comorbidity of personality and substance use disorders
Antisocial personality disorder (APD) and borderline personality disorder (BPD) are the most common comorbid
Axis II diagnoses among men and women with a substance use disorder. In a multisite sample of 366 substance
abusers in treatment, Morgenstern, Langenbucher, Labouvie, and Miller (1997) found that approximately 58% of the
sample met criteria for at least one personality disorder. Antisocial personality disorder (APD) was the most prevalent
single personality disorder with 23% of the sample meeting criteria. BPD was the most prevalent personality disorder
among women (36%) and APD was the most prevalent personality disorder among men (26%). In addition, most
participants who met criteria for a personality disorder had at least two personality disorder diagnoses. Other studies
investigating the prevalence of APD and BPD among substance abusers in treatment (Cacciola, Alterman, Rutherford,
& Snider, 1995; Carrol, Ball, & Rounsaville, 1993; Longabaugh, Rubin, Malloy, Beattie, Clifford, & Noel, 1994;
Rounsaville et al., 1998), and among substance abusers in the general population (Regier et al., 1990), have reported
similar findings. Because of their high prevalence among substance abusers, studies have examined the prognostic
significance of APD and BPD on relapse to substance use.
4.2. Personality disorders and relapse to substance abuse
Tomasson and Vaglum (2000) examined the association between comorbid Axis I disorders and relapse (defined as
abstinence vs. non-abstinence) at 16 and 28 months post-treatment in alcoholics with APD. The vast majority (87%)
of individuals relapsed during the follow-up period. Relapse was best predicted by the number of prior treatment
admissions (odds ratio [OR] = 1.3) and the presence of an Axis I affective disorder was associated with a reduced risk
of relapse (OR = 0.2). The latter finding is consistent with the results of another study (Woody, McLellan, Luborsky, &
O’Brien, 1985), and suggests that APD alcoholics who can experience and report affective symptoms may be more
capable of establishing a working alliance, and therefore achieve better alcohol outcomes.
A study comparing residential addiction treatment for clients meeting diagnostic criteria for APD (with childhood
onset) with those reporting syndromal levels of antisocial behavior only in adulthood (adult antisocial behavioral
syndrome, AABS; Goldstein et al., 2001) found that those with APD returned to drug use significantly earlier than
those with AABS. At the 6-month follow-up interview, 40% of the APD individuals had lapsed compared to 32% of
the AABS group. The number of days of consecutive substance use following the first lapse did not differ between the
In a 2-year follow-up study, Cooney, Kadden, Litt, and Getter (1991) examined time to first heavy drinking day in
97 alcoholics recruited from an inpatient treatment program and randomly assigned to aftercare group treatment with
either coping skills training or interactional therapy. Relapse occurred significantly more slowly when high sociopathy
patients, as assessed by the California Personality Inventory (CPI; Gough & Bradley, 1996), received coping skills
treatment than when low sociopathy patients received interactional therapy. The authors speculate that patients with
more sociopathy may have benefited more from the coping skills treatment because it did not require the development
of strong interpersonal relationships among group members and because it provided specific anger management
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 171
Using a randomized clinical trial design, McKay, Alterman, Cacciola, Mulvaney, and O’Brien (2000) assigned 127
cocaine-dependent patients (N= 46 with comorbid APD) to either a 20-week standard group or individua lized relapse
prevention continuing care intervention after completing an initial treatment intervention. Follow-up interviews were
conducted at 3, 6, and 12 months. The Cocaine Relapse Interview (CRI; McKay, Rutherford, Alterman, & Cacciola,
1996) was used to gather information on experiences during the week prior to specific episodes of cocaine use.
Relapse was defined as 1 day or more of cocaine use that was preceded by an abstinent period of at least 2 weeks.
Overall, the analyses indicated that APD patients did not have worse substance use outcomes. With regard to relapse
episodes, the APD and non-APD patients differed on only one subscale; APD patients reported fewer interpersonal
problems in the week before the relapse as compared with non-APD patients. Given the large number of statistical
tests conducted, the authors caution that this finding may have been due to chance.
Previous studies have also examined personality traits as predictors of substance use relapse. Fisher, Elias, and Ritz
(1998) used the NEO-Personality Inventory (Costa & McCrae, 1985) to assess personality characteristics of 108
inpatient substance abusers. One year following discharge from treatment, patients who were both low in conscien-
tiousness and high in neuroticism had the greatest risk of relapsing to substance use. In this study, relapse was defined
as a return to heavy drinking or drug use. However, an objective criterion for relapse based on the quantity and
frequency of use was not established. Rather, the definition of relapse was based on the client’s self-report of re-
establishing a pattern of ongoing use of alcohol or drugs. Patients who acknowledged having had a slip (e.g., having
two drinks on one occasion, but then returning to abstinence) were not classified as having relapsed. The results from
this study indicate that high emotional instability and behavioral disinhibition are associated with a return to heavy
use of alcohol and other drugs.
Janowsky, Boone, Morter, and Howe (1999) administered the Tridimensional Personality Questionnaire (TPQ;
Cloninger, 1987) to 62 substance-dependent patients in a residential detoxification program. Most patients (70%)
abused or were dependent on both alcohol and cocaine, with the remainder being equally divided in using only
alcohol or cocaine. Relapse was defined as consistent drinking and/or other drug use for 2 or more days. Forty-two
(68%) patients were interviewed at 2 weeks and at 1 month post-discharge via telephone by a substance abuse
counselor who also served as the patient’s therapist during detoxification. The results show that low TPQ persistence
scale scores predicted short-term relapse. The authors conclude that future research should attempt to replicate these
findings with a larger sample and with a longer follow-up period.
Due to the small number of studies examining relapse in substance abusers with comorbid APD or BPD, few
definitive statements about relapse can be made at this time. However, the studies reviewed in this section suggest
several potential avenues for future research. First, it may be beneficial to assess for the presence of other Axis I
disorders, particularly affective disorders, as relapse outcomes may vary for APD substance abusers with and without
a co-occurring affective disorder. Second, personality traits in addition to diagnostic categories should be assessed as
they are associated with relapse. Third, additional randomized clinical trials that employ longer follow-up periods
(i.e., 12–24 months) and greater specificity with regard to defining relapse are needed.
5. Schizophrenia-spectrum disorders
5.1. Comorbidity of schizophrenia and substance use disorders
Epidemiological studies indicate that individuals with schizophrenia are over three times (OR =3.3) more likely to
have an alcohol use disorder and over six times (OR =6.2) more likely to have a drug use diagnosis as compared with
individuals in the general population (Regier et al., 1990). These high rates of substance use are supported by clinical
studies indicating that between 20% and 65% of individuals with a severe mental illness (SMI) also have an alcohol
and/or drug use disorder (Ananth et al., 1989; Drake et al., 1990; Mueser et al., 1990; Safer, 1987). These high rates
indicate that a great deal of research examining the etiology, maintenance of substance use, and factors affecting
relapse to alcohol and drug use among schizophrenic individuals is warranted.
5.2. Schizophrenia-spectrum disorders and relapse to alcohol and drug use
No studies were found that examined relapse to alcohol or drug use among individuals diagnosed with a
schizophrenia-spectrum (i.e., schizophrenia, schizophreniform or schizoaffective) and substance use disorder.
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178172
Three studies have been conducted that provide some background for future investigations. Hasin, Endicott, and
Keller (1991) conducted a 5-year follow-up study with 127 (76 men and 51 women) psychiatric inpatients diagnosed
with an affective disorder and alcohol dependence. A diagnosis of schizoaffective disorder was one of the few
demographic, diagnostic or alcohol-specific variables found to predict a longer time to alcohol remission (defined as
at least 6 months with no evidence of alcohol problems). Although the authors did not assess relapse per se, these
findings indicate that individuals experiencing both schizophrenic and affective symptoms have a more difficult time
achieving abstinence.
Cuffel and Chase (1994) examined data from the ECA study in which participants were interviewed twice
approximately 1 year apart (i.e., Wave 1 and Wave 2) in order to assess changes in the status of mental health and
substance use disorders using the DIS for DSM-III (Regier et al., 1993). This study examined the data of 168
individuals who met criteria for a schizophrenia or schizophreniform disorder at the Wave 1 assessment. In addition,
all met criteria for a substance use disorder either at Wave 1 or 2. No participant demographic information was
provided in the publication. They found that a high percentage of participants were abstinent both in the year prior to
the Wave 1 (85%) and Wave 2 (79%) interviews. Although the prevalence rates remained steady, individuals both
began and terminated substance use during this 1-year period. Among those who abused alcohol or cannabis at Wave
1, one-third reported no substance abuse at Wave 2; among those who met a drug use diagnosis at Wave 1, one-half
were no longer abusing drugs at Wave 2. Consistent with the literature on non-mentally ill substance abusers,
schizophrenic individuals who persisted in meeting a substance use disorder diagnosis at Wave 2 were significantly
more likely to be younger and male. They also reported a significantly greater number of depressive symptoms, as
compared to those who terminated or did not begin abusing substances. The number of schizophrenic symptoms
endorsed at either Wave 1 or 2 was not associated with beginning, continuing or terminating substance abuse. These
results are supported by a study in which psychiatric inpatients dually diagnosed with a severe mental illness and
alcohol and/or drug use disorder reported that depressive symptoms including feelings of hopelessness, worthlessness
and helplessness were significantly more likely to be endorsed as triggers for substance use relapse as compared with
psychotic symptoms such as hearing auditory and visual hallucinations (Bradizza, Stasiewicz, & Carey, 1998).
In summary, these results suggest that amongst individuals with schizophrenia-spectrum and substance use
disorders, affect may play a larger role than schizophrenic symptoms with respect to alcohol abuse. However,
none of the studies reviewed were specifically aimed at assessing alcohol or drug relapse. Studies focused on
assessing the base rates of relapse following substance abuse and/or dual diagnosis treatment would provide a
valuable starting point for this area. These studies should include explicit definitions of relapse assessed in a variety of
ways (e.g., time to first alcohol or drug use, time to first heavy drinking day). Additionally, given the results of the
studies reviewed, measures of affect, stress and depression should be included in studies examining predictors of
substance abuse relapse in this population.
6. Psychiatric disorders and substance abuse relapse: conceptual models
During the past 25 years, a number of relapse models have been proposed that incorporate cognitive elements (e.g.,
Annis, 1986; Litman, 1986; Ludwig & Wikler, 1974; Marlatt & Gordon, 1985; Tiffany, 1990) as predictors of alcohol
and drug relapse. Although each model has been used to explain alcohol and drug relapse among primary substance
abusers, there are several reasons why these models are not ideally suited for dually diagnosed individuals. Of
primary importance, none of these models are specifically designed to account for the effects of psychiatric
functioning on post-treatment alcohol and drug involvement. In addition, these models do not easily account for
the interrelationships between psychiatric variables and other factors such as pretreatment alcohol and drug
involvement, treatment variables, stressors, and coping skills. Moos and his colleagues (Moos & Finney, 1983;
Moos, Finney, & Cronkite, 1990) have proposed a systems model that attempts to understand relapse and post-
treatment functioning in alcoholic patients by assessing the interrelationships between social background factors (e.g.,
age, gender, race, ethnicity, SES, education, and employment history), intake symptomatology (e.g., type and severity
of problem drinking and psychological, social and occupational functioning), treatment factors (e.g., type, number
and duration of treatment sessions, therapist–client alliance), stressors (e.g., job status, death of family member,
interpersonal conflict), coping responses (e.g., behavioral vs. cognitive, approach vs. avoidance), and family
environment (e.g., marital conflict, child–parent problems). Numerous studies conducted over the past 20 years
have provided significant support for this model (e.g., Connors, Maisto, & Zywiak, 1996, 1998; Moos et al., 1990).
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 173
A multivariate approach to understanding the interrelationships among factors and outcomes holds promise for
integrating data, and provides a theoretical rationale for conducting research studies. A majority of the studies
examining relapse among individuals with co-occurring disorders have examined only the direct effects of small
numbers of variables (e.g., diagnosis, coping responses, or treatment) on outcomes. Moos et al.’s systems model
allows variables assessing related constructs to be grouped conceptually into factors allowing for the exploration of
complex and potentially informative interrelationships among factors and their impact on outcomes. To date, few
studies have used such models to guide study design and analysis of findings. This has contributed to a lack of focus
in the literatures examining relapse to substance use among individuals with mental health disorders. In addition, the
absence of a strong conceptual model has made it difficult to integrate study findings into the existing literature. A
systems approach provides a conceptual basis for prioritizing research questions, resolving methodological issues and
guiding data analysis. Four studies that have used Moos et al.’s model to examine the interrelationship of factors and
their relationship to outcomes among samples of dually diagnosed individuals are described below.
Two studies utilizing the same sample examined the interrelationships between depressive symptoms, drinking to
cope with distress, negative life events, family support, and drinking behavior and problems. Holohan, Holahan,
Moos, Cronkie, and Randall (2004) followed 424 depressed individuals who entered treatment for major depression
and 424 matched community controls over a 10-year period. The depressed sample reported more drinking problems,
engaged in more drinking to cope, experienced more negative life events, and less family support, as compared with
the community sample. Over time, more negative life events and less family support were associated with more
drinking to cope. Within the depressed sample, individuals who were more likely to drink to cope at baseline
demonstrated a stronger relationship between depressive symptoms and both alcohol consumption and drinking
problems, than individuals who scored lower on drinking to cope measures (Holohan, Holahan, Moos, Cronkite, &
Randall, 2003).
Similarly, Moggi, Ouimette, Moos, and Finney (1999a) examined the relationship of general and substance-
specific coping and extent of dual diagnosis treatment climate to 1-year post-treatment outcomes in 981 male,
substance-abusing patients diagnosed with a comorbid psychiatric disorder. The sample included 142 psychotic
(schizophrenia, paranoid psychosis and affective psychosis) individuals and the remaining 839 with affective, anxiety
and personality disorders. They found that both types of coping were associated with abstinence at follow-up and that
extent of dual diagnosis treatment environment was related to post-treatment psychiatric functioning. A second study
by Moggi, Ouimette, Moos, and Finney (1999b) used this same sample to examine the influence of psychotic
symptoms on outcomes such as abstinence, psychiatric symptoms, employment, and community residence. They
found that patients with more severe psychiatric disorders showed less improvement as compared with patients with
less severe psychiatric diagnoses, indicating that extent of psychiatric symptoms are associated with important
outcomes among these individuals including alcohol-related outcomes.
By providing information regarding the interrelationships among factors and outcomes, these studies provide
initial evidence that a systems model such as that proposed by Moos et al. (1990) can provide a useful conceptu-
alization for examining the relationship of psychiatric, pretreatment alcohol and drug use, treatment, and post-
treatment factors on alcohol involvement and relapse. However, there are ways in which this model could be modified
to more clearly incorporate other pre- and post-treatment factors relevant to dually diagnosed populations.
For example, Moos et al.’s model could be modified to isolate psychiatric and substance use variables as separate
factors both in the pretreatment and post-treatment portions of the model. Background characteristics (e.g., demo-
graphics), psychiatric symptomatology (e.g., psychiatric diagnosis, psychological symptoms), and alcohol/drug
involvement and problems (e.g., daily substance use, extent of alcohol and drug problems) could be viewed as the
most distal factors that influence psychiatric functioning and substance use relapse both directly and indirectly
through a treatment factor (e.g., amount of therapeutic contact, quality of therapeutic alliance). Treatment would have
an influence on outcomes through its effect on coping skills and stressors. In turn, coping skills and stressors would
serve to impact the type and frequency of relapse environments that are encountered by the individual, which
subsequently could impact both substance use relapse and psychiatric outcomes. This proposed model is not intended
as an exhaustive description of all possible relationships between the relevant factors but as a heuristic for guiding
investigations. Although Moos et al. (1990) included psychiatric and substance use factors as part of their model
(intake symptomatology), the separation of these factors both as predictors and outcomes has several advantages.
Conceptualizing psychiatric and substance use variables as separate factors can improve study design by helping
guide the choice of instruments used to assess variables underlying the factors. It also promotes a more clear
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178174
evaluation of the differential impact of psychiatric symptomatology and substance use variables on separate
psychiatric and substance use outcomes. In addition to these changes, incorporating family environment into the
more general category of relapse environment allows for the inclusion of other relevant, proximal variables such as
setting, context, cognitions and affect that can influence outcomes, particularly those related to substance abuse
7. Future directions
This review has covered alcohol and drug relapse among a wide range of Axes I and II disorders. Despite the
diversity of these disorders, there are common issues and conclusions that can be highlighted with respect to
substance use relapse. An important caveat, however, is that drawing conclusions across studies in this literature
carries the problem of conclusions based on different definitions of relapse. For example, among the studies reviewed
in this paper, definitions of relapse varied and included time to first use of alcohol or drugs, time to first heavy
drinking day (with varying definitions of heavy drinking), 7 or more days of heavy drinking during the follow-up
period, and positive toxicology screens for alcohol and/or drugs, to name several. In other studies, no definition of
relapse was provided. Assuming equivalent follow-up periods, different definitions of relapse may lead to different
findings (Miller, 1996). In the dual-diagnosis literature, there are currently too few studies of relapse within specific
diagnostic groups (e.g., bipolar, social phobia) to warrant confidence in comparisons of relapse rates between studies
using different definitions. In addition, very few studies used multiple definitions of relapse, a point that is addressed
in more detail below. The question of whether or not different definitions of relapse will lead to different substantive
findings will need to be addressed by future research. One of the most widespread problems across many studies is the
absence of well-articulated definitions of relapse. This does not suggest that one uniform definition of relapse is
needed. In fact, several definitions that are used consistently across studies are preferable, as they allow for a more
multidimensional analysis of relapse. A few of the more common definitions that appear in the addictions treatment
research literature include time to first drink/drug use, time to first heavy drinking day, number of relapse events
occurring in specific time interval, and number of days to first 4 consecutive drinking days. Miller (1996) points out
several other parameters that should be considered when deciding upon a definition of relapse. Relapse cannot occur
without a preceding period of abstinence. How long should an individual be abstinent before a return to alcohol/drug
use is labeled a relapse? In the case of multiple substance use, the use of which drug constitutes a relapse? All drugs or
only the drug (e.g., alcohol) that is the focus of treatment? Although there is no consensus regarding the bbestQ
definition of relapse, researchers who undertake a studies with dually diagnosis individuals will be faced with
decisions about the optimal way to define relapse for their sample.
Second, studies should determine the presence of a diagnosis using a structured clinical interview (e.g., SCID,
DIS), and also assess the number and type of symptoms. Both have shown promise in predicting substance abuse
relapse; however, there is enough inconsistency across studies to indicate they are conceptually distinct.
Third, studies should consider a delay in determining the presence of a diagnosis until approximately 3–4 weeks
following the onset of abstinence. This would allow for improved accuracy in obtaining mental health diagnoses by
reducing the occurrence of substance-induced disorders that dissipate following detoxification.
Fourth, the issue of multiple psychiatric diagnoses should be allocated greater attention. Studies involving dually
diagnosed individuals often find that individuals who meet criteria for one mental health disorder will often meet
criteria for one or more other Axis I or II disorders. This can complicate drawing conclusions from analyses involving
individuals with multiple diagnoses. Authors should ensure that the rate and type of diagnostic overlap is clearly
described and that analyses are conducted to reflect both the inclusion and exclusion of these multiply diagnosed
individuals in all major analyses. Analytic and methodological strategies for examining the issue of overlapping
psychiatric diagnoses such as hierarchical models for handling shared and unique variance among internalizing and
externalizing disorders have proven useful for understanding common and distinct etiological factors among these
disorders (e.g., Krueger, 1999; Krueger, & Piasecki, 2002).
Fifth, studies that assess individuals with both alcohol and drug problems should separate individuals with only an
alcohol use disorder from those who have alcohol and drug use disorders. In addition, relapse to alcohol and drug use
should be examined separately.
Finally, this area could benefit greatly from studies that focus on isolating predictors of relapse to alcohol and drug
use, and also predictors of abstinence in these populations. The information gleaned from these studies could make
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 175
substantive contributions to improving treatment for individuals with co-occurring substance use and mental health
Alterman, A. I., McKay, J. R., Mulvaney, F. D., Cnaan, A., Cacciola, J. S., Tourian, K. A., et al. (2000). Baseline prediction of 7-month cocaine
abstinence for cocaine dependence patients. Drug and Alcohol Dependence,59, 215 – 221.
Ananth, J., Vandewater, S., Kamal, M., Bordsky, A., Gamal, R., & Miller, M. (1989). Missed diagnosis of substance abuse in psychiatric patients.
Hospital and Community Psychiatry,40, 297 – 299.
Annis, H. M. (1986). A relapse prevention model for treatment of alcoholics. In W. R. Miller, & N. Heather (Eds.), Treating addictive behaviors
(pp. 407 – 433). New York7Plenum Press.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry,
4, 53 – 63.
Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman, F. D., Charney, D. S., et al. (1995). The development of a Clinician
Administered PTSD Scale. Journal of Traumatic Stress,8, 75 – 90.
Bobo, J. K., McIlvain, H. E., & Leed-Kelly, A. (1998). Depression screening scores during residential drug treatment and risk of drug use after
discharge. Psychiatric Services,49(5), 693 – 695.
Bowen, R. C., D’Arcy, C., Keegan, D., & Senthilselvan, A. (2000). A controlled trial of cognitive behavioral treatment of panic in alcoholic
inpatients with comorbid panic disorder. Addictive Behavior,25, 593 – 597.
Bradizza, C. M., Stasiewicz, P. R., & Carey, K. B. (1998). High-risk alcohol and drug use situations among seriously mentally ill inpatients: A
preliminary investigation. Addictive Behaviors,23(4), 555 – 560.
Brown, P. J. (2000). Outcome in female patients with both substance use and post-traumatic stress disorders. Alcoholism Treatment Quarterly,18(3),
127 – 135.
Brown, P. J., Read, J. P., & Kahler, C. W. (1993). Comorbid posttraumatic stress disorder and substance use disorders: Treatment outcomes and the
role of coping. In P. Ouimette, & P. J. Brown (Eds.), Trauma and substance abuse: Causes consequences, and treatment of comorbid disorders
(pp. 171–188).
Brown, P. J., Stout, R. L., & Mueller, T. (1996). Posttraumatic stress disorder and substance abuse relapse among women: A pilot study.
Psychology of Addictive Behavior,10(2), 124 – 218.
Brown, R. A., Monti, P. M., Myers, M. G., Martin, R. A., Rivinus, T., Dubreuil, M. E., et al. (1998). Depression among cocaine abusers in
treatment: Relation to cocaine and alcohol use and treatment outcome. American Journal of Psychiatry,155, 220 – 225.
Cacciola, J. S., Alterman, A. I., Rutherford, M. J., & Snider, E. C. (1995). Treatment response of antisocial substance abusers. Journal of Nervous
and Mental Disease,183(3), 166 – 171.
Carroll, K. M., Ball, S. A., & Rounsaville, B. J. (1993). A comparison of alternate systems for diagnosing antisocial personality disorder in cocaine
abusers. Journal of Nervous and Mental Disease,181, 436 – 443.
Carroll, K. M., Nich, S., & Rounsaville, B. J. (1995). Differential symptom reduction in depressed cocaine abusers treated with psychotherapy and
pharmacotherapy. The Journal of Nervous and Mental Disease,183(4), 251 – 259.
Charney, D. A., Paraherakis, A. M., Negrete, J. C., & Gill, K. J. (1998). The impact of depression on the outcome of addictions treatment. Journal
of Substance Abuse Treatment,15(2), 123 – 130.
Chengappa, K. N. R., Levine, J., Gershon, S., & Kupfer, D. J. (2000). Lifetime prevalence of substance or alcohol abuse and dependence among
subjects with bipolar I and bipolar II disorders in a voluntary registry. Bipolar Disorder,1, 191 – 195.
Cloninger, C. R. (1987). A systematic method for clinical description and classification of personality variants: A proposal. Archives of General
Psychiatry,44, 573 – 588.
Connors, G. J., Maisto, S. A., & Zywiak, W. H. (1996). Understanding relapse in the broader context of post-treatment functioning. Addiction,91,
S173 – S189.
Connors, G. J., Maisto, S. A., & Zywiak, W. H. (1998). Male and female alcoholics’ attributions regarding the onset and termination of relapses and
the maintenance of abstinence. Journal of Substance Abuse,10, 27 – 42.
Cooney, N. L., Kadden, R. M., Litt, M. D., & Getter, H. (1991). Matching alcoholics to coping skills or interactional therapies: Two-year follow-up
results. Journal of Consulting and Clinical Psychology,59, 598 – 601.
Costa, P. T., & McCrae, R. R. (1985). The NEO personality inventory manual. Odessa, FL7Psychological Assessment Resources.
Cox, B. J., Norton, G. R., Swinson, R. P., & Endler, N. S. (1990). Substance abuse and panic-related anxiety: A critical review. Behavior Research
and Therapy,28(5), 385 – 393.
Cuffel, B. J., & Chase, P. (1994). Remission and relapse of substance use disorders in schizophrenia: Results for a one-year prospective study. The
Journal of Nervous and Mental Disease,182(6), 342 – 348.
Curran, G. C., & Booth, B. M. (1999). Longitudinal changes in predictor profiles of abstinence from alcohol use among male veterans. Alcoholism:
Clinical and Experimental Research,23, 141 – 143.
Curran, G. M., Flynn, H. A., Kirchner, J., & Booth, B. M. (2000). Depression after alcohol treatment as a risk factor for relapse among male
veterans. Journal of Substance Abuse Treatment,19(3), 259 – 265.
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.
Endicott, J., & Spitzer, R. L. (1978). A diagnostic interview: The schedule for affective disorders and schizophrenia. Archives of General
Psychiatry,35, 837 – 844.
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178176
Fals-Stewart, W., & Schafer, J. (1992). The treatment of substance abusers diagnosed with obsessive–compulsive disorder: An outcome study.
Journal of Substance Abuse Treatment,9, 365 – 370.
Fisher, L. A., Elias, J. W., & Ritz, K. (1998). Predicting relapse to substance abuse as a function of personality dimensions. Alcoholism: Clinical and
Experimental Research,22, 1041 – 1047.
Frye, M. A., Altshuler, L. L., McElroy, S. L., Suppes, T., Keck, P. E., Denicoff, K., et al. (2003). Gender differences in prevalence, risk, and clinical
correlates of alcoholism comorbidity in bipolar disorder. American Journal of Psychiatry,160, 883 – 889.
Glenn, S. W., & Parson, O. A. (1991). Prediction of resumption of drinking in posttreatment alcoholics. International Journal of the Addictions,
26(2), 237 – 254.
Goldstein, R. B., Bigelow, C., McCusker, J., Lewis, B. F., Mundt, K. A., & Powers, S. I. (2001). Antisocial behavioral syndromes and return to drug
uses following residential relapse prevention/health education treatment. American Journal of Drug Alcohol Abuse,27(3), 453 – 482.
Gough, H. G., & Bradley, P. (1996). CPI manual. (3rd ed.). Palo Alto, CA7Consulting Psychologists Press.
Greenfield, S. F., Weiss, R. D., Muenz, L. R., Vagge, L. M., Kelly, J. F., Bello, L. R., et al. (1998). The effect of depression on return to drinking: A
prospective study. Archives of General Psychiatry,55(3), 259 – 265.
Hasin, D., Endicott, J., & Keller, M. B. (1989). RDC alcoholism in patients with major affective syndromes: Two-year course. American Journal of
Psychiatry,146(3), 318 – 323.
Hasin, D., Endicott, J., & Keller, M. B. (1991). Alcohol problems in psychiatric patients; 5-year course. Comprehensive Psychiatry,32(4),
303 – 316.
Hasin, D., Tsai, W. Y., Endicott, J., Mueller, T. I., Coryell, W., & Keller, M. (1996). The effects of major depression on alcoholism. American
Journal on Addictions,5, 144 – 155.
Hedlund, J. L., & Vieweg, B. W. (1979). The Hamilton rating scale for depression. Journal of Operational Psychiatry,10(2), 149 – 165.
Holohan, C. J., Holahan, C. K., Moos, R. H., Cronkite, R. C., & Randall, P. K. (2003). Drinking to cope and alcohol use and abuse in unipolar
depression: A 10-year model. Journal of Abnormal Psychology,112(1), 156 – 159.
Holohan, C. J., Holahan, C. K., Moos, R. H., Cronkie, R. C., & Randall, P. K. (2004). Unipolar depression, life context vulnerabilities, and drinking
to cope. Journal of Consulting and Clinical Psychology,72(2), 269 – 275.
Hunt, W. A., Barnett, L. W., & Branch, L. G. (1971). Relapse rates in addiction progams. Journal of Clinical Psychology,27, 455 – 456.
Janowsky, D. S., Boone, A., Morter, S., & Howe, L. (1999). Personality and alcohol/substance-use disorder patient relapse and attendance at self-
help group meetings. Alcohol and Alcoholism,34(3), 359 – 369.
Kessler, R. C., Crum, R. M., Warner, L. A., Nelson, C. B., Schulenberg, J., & Anthony, J. C. (1997). Lifetime co-occurrence of DSM-III-R
alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Archives of General Psychiatry,54,
313 – 321.
Kosten, T., Rounsaville, B. J., & Kleber, H. D. (1986). A 2.5 follow-up of depression, life crises, and treatment effects on abstinence among opioid
addicts. Archives of General Psychiatry,43, 733 – 738.
Krueger, R. F. (1999). The structure of common mental disorders. Archives of General Psychiatry,56, 921 – 926.
Krueger, R. F., & Piasecki, T. M. (2002). Toward a dimensional and psychometrically-informed approach to conceptualizing psychopathology.
Behaviour, Research and Therapy,40(5), 485 – 500.
Kushner, M. G., Sher, K. J., & Erickson, D. J. (1999). Prospective analysis of the relation between DSM-III anxiety disorders and alcohol use
disorders. American Journal of Psychiatry,156(5), 723 – 732.
LaBounty, L. P., Hatsukami, D., Morgan, S. F., & Nelson, L. (1992). Relapse among alcoholics with phobic and panic symptoms. Addictive
Behaviors,17, 9 – 15.
Litman, G. K. (1986). Alcoholism survival: The prevention of relapse. In W. R. Miller, & N. Heather (Eds.), Treating addictive behaviors
(pp. 391 – 405). New York7Prenum Press.
Longabaugh, R., Rubin, A., Malloy, P., Beattie, M., et al. (1994). Drinking outcomes of alcohol abusers diagnosed as antisocial personality disorder.
Alcoholism: Clinical and Experimental Research,18(4), 778 – 785.
Ludwig, A. M., & Wikler, A. (1974). Craving and relapse to drink. Quarterly Journal of Studies on Alcohol,35, 108 – 130.
Lowman, C., Allen, J., Stout, R. L., et al. (1996). Section II. Marlatt’s taxonomy of high-risk situations for relapse: Replication and extension.
Addiction,91(Supplement), S51 – S71.
Marlatt, G. A., & Gordon, J. R. (1985). Relapse prevention: Maintenance strategies in the treatment of addictive behaviors . New York7Guilford.
McDowell, I., & Newell, C. (1996). Measuring health: A guide to rating scales and questionnaires. New York, NY7Oxford University Press.
McKay, J. R., Alterman, A. I., Cacciola, J. S., Mulvaney, F. D., & O’Brien, C. P. (2000). Prognostic significance of antisocial personality disorders
in cocaine-dependent patients entering continuing care. Journal of Nervous and Mental Disease,188(5), 287 – 296.
McKay, J. R., Alterman, A. I., Cacciola, J. S., Rutherford, M. J., O’Brien, C. P., & Koppenhaver, J. (1997). Group counseling versus individualized
relapse prevention aftercare following intensive outpatient treatment for cocaine dependence: Initial results. Journal of Consulting and Clinical
Psychology,65(5), 778 – 788.
McKay, J. R., Rutherford, M., Alterman, A. I., & Cacciola, J. S. (1996). Development of the cocaine relapse interview: An initial report. Addiction,
91, 535 – 548.
Miller, N. S., Hoffmann, N. G., Ninonuevo, F., & Astrachan, B. M. (1997). Lifetime diagnosis of major depression as a multivariate predictor of
treatment outcome for inpatients with substance use disorders from abstinence-based programs. Annals of Clinical Psychiatry,9(3), 127 – 137.
Miller, N. S., Klamen, D., Hoffman, N. G., & Flaherty, J. A. (1996). Prevalence of depression and alcohol and other drug dependence in addictions
treatment populations. Journal of Psychoactive Drugs,28(2), 111– 124.
Miller, N. S., Ninonuevo, F., Hoffmann, N. G., & Astrachan, B. M. (1999). Prediction of treatment outcomes: Lifetime depression versus the
continuum of care. American Journal on Addictions,8(3), 243 – 253.
Miller, W. R. (1996). What is a relapse? Fifty ways to leave the wagon. Addiction,91, S15 – S27.
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178 177
Moggi, F., Ouimette, P. C., Moos, R. H., & Finney, J. W. (1999a). Dual diagnosis patients in substance abuse treatment: Relationship of general
coping and substance-specific coping to one year outcomes. Addiction,94(12), 1805 – 1816.
Moggi, F., Ouimette, P. C., Moos, R. H., & Finney, J. W. (1999b). Effectiveness of treatment for substance abuse outcomes. Journal of Studies on
Alcohol,60, 856 – 866.
Moos, R. H., & Finney, J. W. (1983). The expanding scope of alcoholism treatment evaluation. American Psychologist,38, 1036 – 1044.
Moos, R. H., Finney, J. W., & Cronkite, R. C. (1990). Alcoholism treatment: Context, process and outcome. New York7Oxford University Press.
Morgenstern, J., Langenbucher, J., Labouvie, E., & Miller, K. J. (1997). The comorbidity of alcoholism and personality disorders in a clinical
population: Prevalence and relation to alcohol typology variables. Journal of Abnormal Psychology,106(1), 74 – 84.
Mueser, K. T., Yarnold, P. R., Levinson, D. F., Singh, H., Bellack, A. S., Kee, K., et al. (1990). Prevalence of substance use in schizophrenia:
Demographic and clinical correlates. Schizophrenia Bulletin,16(1), 31 – 56.
Ouimette, P. C., Moos, R. H., & Finney, J. W. (1997). Twelve-step and cognitive-behavioral treatment for substance abuse: A comparison of
treatment effectiveness. Journal of Consulting and Clinical Psychology,65, 230 – 240.
Ouimette, P. C., Finney, J. W., & Moos, R. H. (1999). Two-year posttreatment course and coping of patients with substance abuse and posttraumatic
stress disorder. Psychology of Addictive Behaviors,13, 105 – 114.
Ouimette, P. C., Humphreys, K., Moos, R. H., Finney, J. W., Cronkite, R., & Federman, B. (2001). Self-help participation and functioning among
substance use disorder patients with PTSD. Journal of Substance Abuse Treatment,20, 25 – 32.
Ouimette, P. C., Moos, R. H., & Brown, P. J. (2002). Posttraumatic stress disorder-substance use disorder comorbidity: A survey of treatments and
proposed practice guidelines. In P. C. Ouimette, & P. J. Brown (Eds.), Trauma and substance abuse: Causes, consequences, and treatment of
comorbid disorders (pp. 91 – 110). Washington, DC7American Psychological Association.
Ouimette, P. C., Moos, R. H., & Finney, J. W. (2003). PTSD treatment and 5-year remission among patients with substance use and posttraumatic
stress disorders. Journal of Consulting and Clinical Psychology,71(2), 410 – 414.
Powell, B. J., Penick, E. C., Nickel, E. J., Liskow, B. I., et al. (1992). Outcomes of co-morbid alcoholic men: A 1-year follow-up. Alcoholism:
Clinical and Experimental Research,16(1), 131 – 138.
Project MATCH Research Group. (1997). Matching alcohol treatments to client heterogeneity: Project MATCH posttreatment drinking outcomes.
Journals of Studies on Alcohol,58, 7 – 29.
Radloff, L. S. (1977). The CES-D scale: A self report depression scale for research in the general population. Applied Psychological Measurements,
1, 385 – 401.
Randall, C. L., Thomas, S., & Thevos, A. K. (2001). Concurrent alcoholism and social anxiety disorder: A first step toward developing effective
treatments. Alcoholism: Clinical and Experimental Research,25, 210 – 220.
Regier, D. A., Boyd, J. H., Burke, J. D., et al. (1988). One-month prevalence of mental disorders in the United States: Based on five epidemiological
catchment area sites. Archives of General Psychiatry,45, 977 – 986.
Regier, D. A., Farmer, M. E., Rae, D. S., Locke, B. Z., Keith, S. J., Judd, L. L., et al. (1990). Comorbidity of mental disorders with
alcohol and other drug abuse: Results from the Epidemiologic Catchment Area (ECA) Study. Journal of American Medical Association,
264, 2511– 2518.
Regier, D. A., Narrow, W. E., Rae, D. S., Manderscheid, R. W., Locke, B. Z., & Goodwin, F. K. (1993). The de facto US mental and addictive
disorders service system: Epidemiologic Catchment Area prospective 1-year prevalence rates of disorders and services. Archives of General
Psychiatry,50, 85 – 94.
Reich, L. H., Davies, R. K., & Himmelhoch, J. M. (1974). Excessive alcohol use in manic-depressive illness. American Journal of Psychiatry,131,
83 – 86.
Rounsaville, B. J., Kranzler, H. R., Ball, S., Tennen, H., Poling, J., & Triffleman, E. (1998). Personality disorders in substance abusers: Relation to
substance use. Journal of Nervous and Mental Disease,186, 87 – 95.
Safer, D. (1987). Substance abuse by young adult chronic patients. Hospital and Community Psychiatry,38, 511 – 514.
Schuckit, M. A. (1996). Alcohol, anxiety and depressive disorders. Alcohol Health and Research World,20, 81 – 85.
Sellman, J. D., & Joyce, P. R. (1996). Does depression predict relapse in the 6 month following treatment for men with alcohol dependence?
Australian and New Zealand Journal of Psychiatry,30(5), 573 – 578.
Spitzer, R. L., Williams, J. B. W., Gibbon, M., & First, M. B. (1994). Structured clinical interview for DSM-IV—Patient edition. Washington, DC7
American Psychiatric Press.
Thevos, A. K., Roberts, J. S., Thomas, S. E., & Randall, C. L. (2000). Cognitive behavioral therapy delays relapse in female socially phobic
alcoholics. Addictive Behaviors,25(3), 333 – 345.
Tiffany, S. T. (1990). A cognitive model of drug urges and drug-use behavior: Role of automatic and nonautomatic processes. Psychological
Review,97, 147 – 168.
Tollefson, G. D., Montague-Clouse, J., & Tollefson, S. L. (1992). Treatment of comorbid generalized anxiety in a recently detoxified alcoholic
population with a selective serotenergic drug. Journal of Clinical Psychopharmacology,12, 19 – 26.
Tomasson, K., & Vaughn, P. (1998). The role of psychiatric comorbidity in the prediction of readmission for detoxification. Comprehensive
Psychiatry,39(3), 129 – 136.
Tomasson, K., & Vaglum, P. (2000). Antisocial addicts: The importance of additional axis I disorders for the 28-month outcome. European
Psychiatry,15(8), 443 – 449.
Woody, G. W., McLellan, A. T., Luborsky, L., & O=Brien, C. P. (1985). Sociopathy and psychotherapy outcome. Archives of General Psychiatry,
42, 1081 – 1086.
C.M. Bradizza et al. / Clinical Psychology Review 26 (2006) 162–178178
... 6 In addition to being a risk factor for the development of chronic pain, 2 individuals with AUD undergoing acute alcohol withdrawal exhibit greater pain sensitivity, [7][8][9] which in turn can lead to relapse in recovering from AUD. 2 Unfortunately, there are few effective therapies to treat AUD-mediated pain and individuals undergoing AUD treatment programs have a very high, 70-85%, relapse rate. 10,11 Therefore, new therapeutic approaches are needed to treat AUD, in particular AUD-associated pain. ...
... 8,67,68 In addition, reduction in pain during treatment for AUD is associated with lower risk for relapse. 69 With AUD causing 2.3 million deaths per year, 70 and with a relapse rate of ∼80% one year postwithdrawal, 10,11,71 there is an urgent need for new therapies to treat AUD. Our results suggest that consumption of probiotics may reduce pain in individuals with AUD, which could suppress their tendency to relapse. ...
Full-text available
Alcohol use disorder (AUD) is a major health problem that causes millions of deaths annually world-wide. AUD is considered to be a chronic pain disorder, that is exacerbated by alcohol withdrawal, contributing to a high (∼80%) relapse rate. Chronic alcohol consumption has a marked impact on the gut microbiome, recognized to have a significant effect on chronic pain. We tested the hypothesis that modulating gut microbiota through feeding rats with probiotics can attenuate alcohol-induced muscle mechanical hyperalgesia. To test this hypothesis, rats were fed alcohol (6.5%, 4 days on 3 days off) for 3 weeks, which induced skeletal muscle mechanical hyperalgesia. Following alcohol feeding, at which time nociceptive thresholds were ∼37% below pre-alcohol levels, rats received probiotics in their drinking water, either Lactobacillus Rhamnosus GG (Culturelle) or De Simone Formulation (a mixture of 8 bacterial species) for 8 days; control rats received plain water to drink. When muscle mechanical nociceptive threshold was evaluated 1 day after beginning probiotic feeding, nociceptive thresholds were significantly higher than rats not receiving probiotics. Mechanical nociceptive thresholds continued to increase during probiotic feeding, with thresholds approaching pre-alcohol levels 5 days after starting probiotics; nociceptive threshold in rats not receiving probiotics remained low. After probiotics were removed from the drinking water, nociceptive thresholds gradually decreased in these two groups, although they remained higher than the group not treated with probiotic (21 days after ending alcohol feeding). These observations suggest that modification of gut microbiota through probiotic feeding has a marked effect on chronic alcohol-induced muscle mechanical hyperalgesia. Our results suggest that administration of probiotics to individuals with AUD may reduce pain associated with alcohol consumption and withdrawal, and may be a novel therapeutic intervention to reduce the high rate of relapse seen in individuals with AUD attempting to abstain from alcohol.
... One-third of AUD individuals also exhibit depressive symptoms (27). Selective serotonin reuptake inhibitors (SSRIs) are widely used to treat depression, and while clinical trials have examined their efficacy in treating comorbid AUD, these trials have failed to find a clinically significant effect. ...
Full-text available
Alcohol Use Disorder (AUD) is a chronic psychiatric disorder marked by impaired control over drinking behavior that poses a significant challenge to the individual, their community, the healthcare system and economy. While the negative consequences of chronic excessive alcohol consumption are well-documented, effective treatment for AUD and alcohol-associated diseases remains challenging. Cognitive and behavioral treatment, with or without pharmaceutical interventions, remain the most commonly used methods; however, their efficacy is limited. The development of new treatment protocols for AUD is challenged by difficulty in accurately measuring patterns of alcohol consumption in AUD patients, a lack of a clear understanding of the neuropsychological basis of the disorder, the high likelihood of AUD patients relapsing after receiving treatment, and the numerous end-organ comorbidities associated with excessive alcohol use. Identification and prediction of patients who may respond well to a certain treatment mechanism as well as clinical measurement of a patient's alcohol exposure are bottlenecks in AUD research which should be further addressed. In addition, greater focus must be placed on the development of novel strategies of drug design aimed at targeting the integrated neural pathways implicated in AUD pathogenesis, so that next-generation AUD treatment protocols can address the broad and systemic effects of AUD and its comorbid conditions.
... Current treatments for SUDs usually include a combination of in-and/or outpatient psychotherapeutic, psychosocial, or pharmacological interventions. However, high rates of attrition (Brorson et al., 2013), relapse (Brandon et al., 2007;Fleury et al., 2016), suicide (Poorolajal et al., 2016), and mental health co-morbidity (Bradizza et al., 2006;Wu & Blazer, 2014) limit the efficacy of these services. Identifying active components of interventions used in SUD treatment that positively affect outcomes is vital for the advancement of further treatment options. ...
Full-text available
Objective: Increasing regular physical activity (PA) behaviour may be an effective adjunct intervention for substance use disorder (SUD) treatment. This systematic review aims to identify promising behaviour change techniques (BCTs), namely, BCTs present in the design of interventions evidencing significant short-term and/or long-term (d ≥ 0.15 for objective measures and d ≥ 0.36 for self-report measures) increase in PA and/or reduction of substance use, secondary psychological measures, and retention in the PA intervention. Method: PRISMA guidelines were followed, and the search was performed on March 11, 2021 across databases including MEDLINE, PsycINFO, SPORTDiscus, Cochrane Library, CINAHL, ProQuest, Web of Science Core Collection, Google Scholar; Open Grey, and ProQuest Dissertations & Theses. Studies were included if they measured PA, included participants aged ≥18 years, were randomised control trials, and if participants were diagnosed with SUDs. The Cochrane RoB 2.0 Tool was used to assess risk of bias. BCTs from eligible studies were extracted, coded, and ranked according to their proportional presence across studies. Results: The final synthesis included k = 61 studies with N = 12,887 participants. High heterogeneity across outcome measures, interventions and control conditions was found. In total, 477 applications of BCTs were identified. Instruction on how to perform the behaviour, social support (unspecified), behavioural practice/rehearsal, problem solving, pharmacological support, goal setting (behaviour), self-monitoring (behaviour), and biofeedback were the eight most frequently used promising BCTs across studies. Conclusions: Incorporating the eight most promising BCTs identified in this review in future PA interventions in SUD populations may improve SUD outcomes.
... Over the course of this work, I was privileged to work alongside the people under my care, and their families, to come to understand some of their motivations and hopes for the future. Many of the goals set when support planning with individuals involved making manageable steps towards regaining confidence and autonomy, but progress was rarely quick or simple, and relapse was common (Ascher-Svanum et al, 2010;Bradizza et al, 2006). This process was especially difficult when it came to managing psychotropic medications and finding a balance between assisting people to have improved well-being and quality of life on effective medication, as well as mitigating dependence. ...
Conference Paper
Aims: Recent studies have shown that mood can bias perceived reward value, with this effect being strongest in individuals with more mood instability. Spontaneous use of mental imagery has been highlighted as an important feature in generating and maintaining mood symptoms in bipolar disorder. We examined whether mental imagery influencing motivation biases perceived reward value during learning, and to what extent effects are modulated by mood symptoms. Method: 50 healthy participants completed a brief, online-based manipulation in which they generated mental images related to goal-attainment and goal-failure with a view to increasing and decreasing motivation, respectively. We quantified the efficacy of this manipulation on mood and motivation, as well as on the perception of reward stimuli encountered in two learning blocks. Participants performed each block under one of the two types of imagery, thus using a within-participants design. To test for bias in perceived reward value, participants were subsequently asked to indicate their preference in pairwise choices between all stimuli encountered. Trait mood instability (HPS), propensity towards imagery (SUIS), and depression symptoms (PHQ-9) were included in analyses to test for modulatory effects on biased preference. Results: Goal-oriented mental imagery effectively impacted subjective motivation, with higher ratings in the goal-attainment imagery block, compared to goal-failure. Depression symptoms, but not mood instability, were observed to have a modulating effect on change in motivational state. The degree to which momentary motivation was impacted by imagery was positively associated with bias in perceived reward value, and further modulated by depression symptoms. Conclusions: Our findings indicate that goal-oriented mental imagery is effective in impacting motivational state in healthy individuals reporting more depression symptoms, and that motivational state in turn modulates reward perception. Insights are offered to aid development of interventions using mental imagery as an emotional and motivational “amplifier” to improve depressed mood.
... Despite the availability of behavioral interventions for substance use in the majority of correctional institutions, the rate of relapse upon release is as high as 80% within the first year (Bradizza et al., 2006;Walitzer & Dearing, 2006). This indicates that the current interventions offered in corrections may not be sufficiently effective in reducing substance use and its associated morbidities. ...
Background The prevalence of alcohol use disorder (AUD) is estimated to be ten times higher amongst individuals in the criminal justice system compared to the general population. Alcohol use is also one of the strongest modifiable risk factors for recidivism. One intervention that has been shown to be effective in reducing alcohol consumption in the general population is medication assisted treatment (MAT) (e.g., naltrexone, disulfiram), and this critical review synthesized the existing evidence on MAT for AUD in correctional settings. Methods Empirical, peer-reviewed studies on approved medications for AUD in correctional populations were searched in major databases. One hundred and sixty-two articles were initially screened, and 14 eligible articles were included in the final review. Four articles examined disulfiram, and ten articles examined naltrexone. Results The studies on disulfiram were considerably older, predating contemporary scientific standards. In terms of outcomes, disulfiram in combination with substantial contingencies in a supervised setting significantly reduced alcohol-related measures of consumption and recidivism, and had acceptable safety and tolerability. All naltrexone studies showed significant reductions on alcohol-related measures, but the effects on recidivism were mixed. The naltrexone studies indicated that it was highly acceptable and well-tolerated. In addition, offenders receiving naltrexone had significantly greater medication adherence, treatment attendance and treatment duration compared to placebo. Conclusions A small number of studies on pharmacological interventions for AUD in the correctional population suggest that MAT is useful for addressing alcohol consumption, although report mixed results regarding recidivism. On balance, the evidence was more convincing for naltrexone in reducing alcohol-related outcomes than disulfiram and it may also be a more feasible intervention in correctional settings. Further research on MAT to address AUD in correctional populations with larger sample sizes, longer duration, and in combination with behavioural interventions is warranted.
... Along with the intrinsic complexity of addictive behaviour and various associated factors (such as genetic, epigenetic, social and cultural) that may explain the difficulty in treating and recovering from AUD, a crucial point certainly refers to the impact of psychiatric comorbidity on the detection and management of AUD. Psychiatric comorbidity (or dual diagnosis) is a crucial problem in AUD patients because it is well known to increase the risk of relapse (Bradizza et al., 2006), making therapeutic intervention more difficult (Daley & Moss, 2002;Sterling et al., 2011;Vitali et al., 2018), for example, by decreasing the treatment compliance (Dixon, 1999) or by increasing the discontinuation of treatment (Bischof et al., 2005 ...
Addiction is a chronic relapsing disorder. Despite pharmacological and psychological interventions during rehabilitation, a majority of patients still relapse. In this seventh chapter, we present neuromodulation techniques as a complementary intervention for addiction. Firstly, while deep brain stimulation (DBS) has shown promising results, its cost–benefit–risk ratio is nonetheless too high to be implemented in routine clinical care. Secondly, repeated transcranial magnetic stimulation (rTMS) and transcranial direct courant stimulation (tDCS) over the dorsolateral prefrontal cortex (DLPFC) have shown reduced craving and relapses, but the results are mixed. To improve efficacy, new perspectives envisioned that the insula could be a promising target for rTMS and DBS in combination with cognitive remediation and while participants are exposed to key conditioned stimuli. Additionally, neurofeedback could be a useful tool in teaching patients to actively regulate their neural activity, although better controlled experimental designs and rigorous measures of brain changes are needed. Despite the heterogeneity of studies, neuromodulation techniques as complementary tools to conventional care seem to constitute a turning point in the management of addictions.
Introduction Both homelessness and substance use have increased in recent years. People experiencing homelessness (PEH) are at increased risk for health problems and early mortality, both of which can be exacerbated by substance use disorders (SUD). Specialty SUD treatment is likely needed to address substance use among PEH, and more than 232,000 PEH received treatment from U.S. publicly funded SUD programs in 2015. The objective of this paper is to develop a better understanding of the SUD services that PEH receive in publicly funded treatment programs by (1) describing the characteristics and needs of the PEH population served in publicly funded SUD treatment programs, compared to non-PEH populations; (2) determining if differences exist in treatment placement (level of care) for PEH and non-PEH; and (3) gauging how successful programs are in treating PEH compared to non-PEH. Methods Observational study using a two-way fixed effect model to determine associations among homelessness, retention, and outcomes among Medicaid beneficiaries receiving SUD treatment in California from 2016 to 2019 (n = 638,953). The study team used ordinary least squares (OLS) regression to measure the degree to which homelessness was associated with baseline characteristics, SUD services received, and treatment outcomes. Results PEH were significantly more likely than non-PEH to be having methamphetamine or heroin as their primary substance. PEH had greater frequency of primary substance use prior to entering treatment, greater ER and hospital utilization, more criminal justice involvement, and greater prevalence of mental health diagnoses and unemployment. PEH were 9.82% more likely than non-PEH to receive residential treatment and 7.11% less likely than non-PEH to receive treatment intensive outpatient modalities. Homelessness was associated with an 11.90% decrease in retention, and a 19.40% decrease in successful discharge status. These trends were consistent across outpatient, intensive outpatient, and residential modalities. Conclusions Developing SUD treatment capacity and housing supports can improve treatment outcomes for PEH. Potential strategies to improve SUD services for PEH include providing more contingency management, opioid pharmacotherapies, programming designed to treat individuals with co-occurring mental health disorders, and resources for housing options that can support PEH in their recovery.
Full-text available
Background The United States is experiencing a drug addiction and overdose crisis, made worse by the COVID-19 pandemic. Relative to other types of health services, addiction treatment and overdose prevention services are particularly vulnerable to disaster-related disruptions for multiple reasons including fragmentation from the general medical system and stigma, which may lead decisionmakers and providers to de-prioritize these services during disasters. In response to the COVID-19 pandemic, U.S. states implemented multiple policies designed to mitigate disruptions to addiction treatment and overdose prevention services, for example policies expanding access to addiction treatment delivered via telehealth and policies designed to support continuity of naloxone distribution programs. There is limited evidence on the effects of these policies on addiction treatment and overdose. This evidence is needed to inform state policy design in future disasters, as well as to inform decisions regarding whether to sustain these policies post-pandemic. Methods The overall study uses a concurrent-embedded design. Aims 1–2 use difference-in-differences analyses of large-scale observational databases to examine how state policies designed to mitigate the effects of the COVID-19 pandemic on health services delivery influenced addiction treatment delivery and overdose during the pandemic. Aim 3 uses a qualitative embedded multiple case study approach, in which we characterize local implementation of the state policies of interest; most public health disaster policies are enacted at the state level but implemented at the local level by healthcare systems and local public health authorities. Discussion Triangulation of results across methods will yield robust understanding of whether and how state disaster-response policies influenced drug addiction treatment and overdose during the COVID-19 pandemic. Results will inform policy enactment and implementation in future public health disasters. Results will also inform decisions about whether to sustain COVID-19 pandemic-related changes to policies governing delivery addiction and overdose prevention services long-term.
Background and aims: Alcohol use disorder (AUD) is a substantial problem, causing early death and great economic burden. Research has highlighted the potential positive impact of technological interventions, such as smartphone applications (app) in treatment of AUD. The aim of this study was to explore the effectiveness of a smartphone app, incorporating computerized cognitive behavioural therapy and text messaging support, on alcohol outcomes over 6 months in a post-rehabilitation setting. Methods: A total of 111 participants with AUD were recruited into this randomized controlled trial, following completion of a 30-day rehabilitation programme. The intervention group (n = 54) used the smartphone app "UControlDrink" (UCD) over 6 months with treatment as usual (TAU), and the control group (n = 57) received TAU. All subjects suffered from AUD as the primary disorder, with other major psychiatric disorders excluded. All intervention subjects used the UCD smartphone app in the treatment trial, and all subjects underwent TAU consisting of outpatient weekly support groups. Drinking history in the previous 90 days was measured at baseline and at 3- and 6-month follow-ups. Additional measurements were made to assess mood, anxiety, craving, and motivation. Results were analysed using intention-to-treat analyses. Results: Retention in the study was 72% at 3 months and 52% at 6 months. There was a significant reduction in heavy drinking days in the intervention group relative to TAU over the 6 months, p < 0.02. Conclusions: The UCD smartphone app demonstrates a significant benefit to reducing heavy drinking days over a 6-month post-rehabilitation period in AUD.
A crucial point in the recovery of substance use disorders (SUDs) is related to the difficulty for clinicians to detect and quantify the impact of psychiatric comorbidity on individual patients. The frequent co-occurrence of SUD with other psychiatric syndromes (e.g. mood, anxiety, thought disorder) raises important questions about the potential mechanisms, which are likely to vary depending on the associated disease, and the way to treat it. Based on the transdiagnostic framework, the present chapter examines the comorbidity between alcohol use disorder (AUD) and internalising disorders (e.g. depression), externalising disorders (e.g. other SUDs) and thought disorders (e.g. psychosis). Potential biopsychosocial mechanisms involved in the co-occurrence of symptoms, syndromes or dimensional traits were also considered. We discussed the fact that this dimensional approach in psychopathology offers a new horizon for research that can considerably improve the treatment of AUD in association with non-AUD symptoms and syndromes.
One of the few areas of consensus in the alcoholism treatment field involves the recognition that alcoholism is a chronic condition with a high risk of relapse. Treatment outcome studies have reported rates of 80% or more by 6 months posttreatment discharge (Armor, Polich, & Stambul, 1978; Gottheil, Thornton, Skolada, & Alterman, 1979), and drinking outcomes of individual clients have been found to be highly unstable over time (Annis & Ogborne, 1983; Finney, Moos, & Newborn, 1980, Litman, Eiser, & Taylor, 1979). It is not surprising, therefore, that, increasingly, relapse is being recognized as an important phenomenon for study.
Methodological issues involved in assessing the prevalence of substance abuse in schizophrenia are discussed, and previous research in this area is comprehensively reviewed. Many studies suffer from methodological shortcomings, including the lack of diagnostic rigor, adequate sample sizes, and simultaneous assessment of different types of substance abuse (e.g., stimulants, sedatives). In general, the evidence suggests that the prevalence of substance abuse in schizophrenia is comparable to that in the general population, with the possible exceptions of stimulant and hallucinogen abuse, which may be greater in patients with schizophrenia. Data are presented on the association of substance abuse with demographics, diagnosis, history of illness, and symptoms in 149 recently hospitalized DSM-III-R schizophrenic, schizophreniform, and schizoaffective disorder patients. Demographic characteristics were strong predictors of substance abuse, with gender, age, race, and socioeconomic status being most important. Stimulant abusers tended to have their first hospitalization at an earlier age and were more often diagnosed as having schizophrenia, but did not differ in their symptoms from nonabusers. A history of cannabis abuse was related to fewer symptoms and previous hospitalizations, suggesting that more socially competent patients were prone to cannabis use. The findings show that environmental factors may be important determinants of substance abuse among schizophrenic-spectrum patients and that clinical differences related to abuse vary with different types of drugs. © 1972, American Association for Cancer Research. All rights reserved.
Objective: This study examines a model of treatment for substance abuse and dependence for patients with substance use disorders and concomitant psychiatric disorders. The model focuses on five interrelated sets of variables (social background, intake functioning, dual diagnosis treatment orientation, patients' change on proximal outcomes, and aftercare participation) that are hypothesized to affect dual diagnosis patients' 1-year posttreatment outcomes. Method: A total of 981 male dual diagnosis patients completed assessment at intake, discharge and 1-year follow-up. The relative importance of each set of variables as predictors of outcome was estimated by constructing block variables and conducting path analyses. Results: Dual diagnosis patients had a higher abstinence rate at follow-up (39%) than at intake (2%); they also improved on freedom from psychiatric symptoms (from 60% to 68%) and employment (from 20% to 29%). At follow-up, patients in programs with a stronger dual diagnosis treatment orientation showed a higher rate of freedom from psychiatric symptoms (71%) than did patients in weaker dual diagnosis treatment oriented programs (65%); they also were more likely to be employed (34% vs 25%). More change on proximal outcomes and more aftercare participation were also associated with better I-year outcomes. Patients with less severe psychiatric disorders improved more and responded better to dual diagnosis oriented treatments than did patients: with more severe psychiatric disorders. Conclusion: Treatment programs for substance use disorders that adhere to principles of dual diagnosis treatment obtain better outcomes for dual diagnosis patients, especially for patients with less severe psychiatric disorders.
Background: The effect of depression on return to drinking among individuals with alcohol dependence is controversial. From February 1, 1993, to April 15, 1996, we consecutively recruited 40 women and 61 men hospitalized for alcohol dependence and followed them up monthly for I year to assess the effect of depression on drinking outcomes. Methods: We conducted structured interviews during hospitalization and monthly following discharge for l year to determine whether depression at treatment entry affected the likelihood of return to drinking and whether this effect differed between sexes. Using survival analysis, we examined the effect of depressive symptoms and a diagnosis of current major depression at treatment entry on times to first drink and relapse during follow-up. Results: A diagnosis of current major depression at the time of hospitalization was associated with shorter times to first drink (hazard ratio, 2.03; 95% confidence interval [CI], 1.28-3.21; P=.003) and relapse (hazard ratio, 2.12; 95% CI, 1.32-3.39; P=.002). There was no significant difference between women and men in this effect. Depressive symptoms as measured by the Beck Depression Inventory did not predict time to first drink or relapse in women or men. Conclusions: A diagnosis of current major depression at entry into inpatient treatment for alcohol dependence predicted shorter times to first drink and relapse in women and men. Our results differ from earlier reports that men and women differ in the effect of depression on return to drinking.