Skip Navigation

Link to  the National Institutes of Health  
The Science of Drug Abuse and Addiction from the National Institute on Drug Abuse Archives of the National Institute on Drug Abuse web site
Go to the Home page

NIDA Home > About NIDA > Organization > HSR   

Health Services Resource (HSR)

Part A: Comorbidity of Mental Disorders with Drug Abuse/Dependence

Annotated Bibliography
May, 1998

Barry S. Brown, Ph.D.
University of North Carolina at Wilmington


Hyperlinks to sections within this text:

Part A:
General Population Surveys Regarding Comorbidity
Prevalence of Mental Disorders in Drug Abuse Populations
Prevalence of Drug Abuse/Dependence in Psychiatric Populations
Comorbidity and HIV Risk Behaviors
Psychotherapeutic Approaches and Comorbidity

Part B:
Other Psychosocial Interventions and Comorbidity
Pharmacotherapeutic Approaches and Comorbidity
Treatment Outcomes for Comorbid Populations
Costs Associated with Comorbidity
Literature Reviews Relating to Comorbidity



Merikangas, K. R., Angst, J., Eaton, W., Canino, G., Rubio-Stepic, M., Wacker, H., Wittchen, H. -U., Andrade, Essau, C., Whitaker, A., Kraemer, H., Robins, L. N., & Kupfer, D. J. (1996). Comorbidity and boundaries of affective disorders with anxiety disorders and substance misuse: Results of an International Task Force.  British Journal of Psychiatry, 168 (Suppl. 30), 58–67.

The authors report the results of five epidemiologic studies from the United States (mainland and Puerto Rico), Switzerland (Zurich and Basel), and Germany (Munich). All studies involved use of structured interviews with adult samples to derive psychiatric diagnoses consistent with DSM-III and DSM-III-R criteria. All studies sought to obtain estimates of lifetime and current prevalence of disorders. Drug use/dependence (but not alcoholism) was strongly linked to depression in all five studies. There was no relationship between drug use/dependence and depression by gender, although there was a relationship between alcohol use/dependence and depression for women. The onset of depression was equally likely to precede as to follow alcoholism (comparable data for drug use/misuse are not reported).

Milby, J. B., Sims, M. K., Khuder, S., Schumacher, J. E., Huggins, N., McLellan, A. T., Woody, G., & Haas, N. (1996). Psychiatric comorbidity: Prevalence in methadone maintenance treatment. American Journal of Drug and Alcohol Abuse, 22, 95–107.

This follow-up study included 102 former methadone maintenance clients drawn from three widely dispersed programs. Follow-up occurred 5-7 years from the time of first contact. At time of follow-up, contact subjects were administered a standardized interview designed to elicit the presence or absence of DSM-III-R affective and/or anxiety disorders, as well as the Beck Depression Inventory (BDI) and the Spielberger State-Trait Anxiety Scale. Of the 102 subjects, 94 were interviewed face to face, and the remainder were interviewed by phone. A high rate of interviewer reliability was established. At time of follow-up, 63% of subjects were in methadone maintenance programs. Fifty-eight percent of subjects showed evidence of affective disorders, with 31% showing major depression and 13% bipolar disorder. Nearly as many (55%) showed evidence of anxiety disorders, with 31% showing psychiatric stress disorder, 21% generalized anxiety, and 20% phobic disorder. Moreover, 37% showed evidence of both an anxiety and an affective disorder. Subjects in and out of methadone treatment did not differ in terms of prevalence of anxiety and affective disorders.

Regier, D. A., Boyd, J. H., Burke, J. D., Rae, D. S., Myers, J. K., Kramer, M., Robins, L. N., George, L. K., Karno, M., & Locke, B. (1988). One-month prevalence of mental disorders in the United States—Based on five Epidemiologic Catchment Area sites. Archives of General Psychiatry, 45, 977–986.

Prevalence rates of psychiatric diagnoses of a probability sample of 18,571 adults (18 years of age and over) were drawn from households in catchment areas in five cities. The National Institute of Mental Health (NIMH) Diagnostic Interview Schedule (DIS) was used to understand psychiatric status in accord with diagnostic criteria contained in the DSM-III. Diagnostic status was calculated for the period 1 month prior to the administration of the DIS. Past month prevalence for substance use disorders (including alcohol and drug abuse/dependence) was reported by 3.8% of the household population, with 1.3% meeting criteria for drug abuse/dependence. For the 6 months prior to interview, 6% met criteria for substance abuse, and 2% met criteria for drug abuse/dependence. Lifetime prevalence was found to be 16.4% for substance abuse and 5.9% for drug abuse/ dependence. Drug abuse/dependence was most likely to be seen in household subjects aged 18-24 (3.5%). Within that age group, drug abuse/dependence was diagnosed in 4.8% of men and 2.4% of women.

Regier, D. A., Farmer, M. E., Rae, D. S., Locke, B. Z., Keith, S. J., Judd, L. L., & Goodwin, F. G. (1990). Comorbidity of mental disorders with alcohol and other drug abuse—Results from the Epidemiologic Catchment Area (ECA) study. Journal of the American Medical Association, 264, 2511–2518.

Data were obtained from 20,291 adults (18 years of age and over) sampled in an epidemiological study conducted between 1980 and 1984 assessing the prevalence of mental, alcohol, and drug abuse disorders in the general population (i.e., in community and institutional populations). The National Institute of Mental Health (NIMH) Diagnostic Interview Schedule (DIS) was used to determine the existence of diagnosable disorders. This study addressed comorbidity in terms of rates of co-occurring mental disorder and/or drug and alcohol disorder. Over all subjects, the prevalence rates for any disorder (alcohol, drug abuse, or mental health—ADM) were 16% for the month preceding interview, 20% for the 6 months preceding interview, and 33% for lifetime. Lifetime alcohol abuse/dependence was found for 14% of the population and lifetime drug abuse/dependence for 6% of the population. Lifetime prevalence rates for dual disorder were 3.1% for mental and alcohol disorders, 1.5% for mental and other drug disorders, 1.1% for drug and alcohol disorders, and 1.7% for all three ADM disorders. However, of adults with lifetime histories of mental disorder, 22% also had lifetime histories of alcohol abuse/dependence, and 15% also had lifetime histories of drug abuse/dependence. The rate for lifetime alcohol disorder was twice as high for those with lifetime mental disorder than for those without lifetime mental disorder; the rate for lifetime drug disorder was four times as high for those with lifetime mental disorder as for those without lifetime mental disorder. Half of those adults with lifetime drug abuse/dependence showed evidence of lifetime mental disorder and were four times as likely to show mental disorder as adults without lifetime drug disorder; half of adults with lifetime drug abuse/dependence showed evidence of lifetime alcohol abuse/dependence and were seven times as likely to show alcohol disorder as those without lifetime drug disorder. Also, 28% of those with lifetime schizophrenia met criteria for lifetime drug disorder and were six times as likely to show evidence of drug disorder than were those without lifetime schizophrenia; 42% of those with lifetime antisocial personality (ASP) diagnoses showed evidence of lifetime drug abuse/dependence and were 13 times as likely to show drug disorder as those lacking lifetime ASP diagnoses; 12% of those showing any lifetime anxiety disorder showed evidence of lifetime drug abuse/dependence and were 2.5 times more likely to show drug disorders than those lacking lifetime anxiety disorder; and 19% of those showing any lifetime affective disorder showed evidence of lifetime drug abuse/ dependence and were five times as likely to show drug disorders as those not showing lifetime affective disorder. Also, 28% of adults with lifetime drug disorders showed lifetime anxiety disorder, 26% showed lifetime affective disorder, and 18% showed lifetime ASP. Persons with drug disorders and mental health disorders were significantly more likely to attend treatment than those with drug disorders alone.


Brooner, R. K., King, V. L., Kidorf, M., Schmidt, C. W., & Bigelow, G. E. (1997). Psychiatric and substance use comorbidity among treatment-seeking opioid users. Archives of General Psychiatry, 54, 71–80.

The psychiatric status of 716 consecutive admissions to a methadone maintenance program was studied. Clients were completing diagnostic assessments and were without disabling medical problems. Diagnoses were made using the Structured Clinical Interview for DSM-III-R. Clients also were administered the Addiction Severity Index (ASI) and the Neo-Personality Inventory (NEO-PI) to determine selected personality traits. Reliability between interviewers was satisfactorily high. Forty-seven percent of clients (48% of men and 47% of women) had a lifetime psychiatric diagnosis in addition to substance use disorder; of those clients, 38.5% had two or more additional diagnoses. Lifetime antisocial personality disorder was found in 25% of clients; lifetime major depression was found in 16% of clients. In terms of a current diagnosis, 39% of opiate clients had a current psychiatric diagnosis in addition to substance use disorder, 35% showed personality disorder, and 8% had an Axis I disorder (3.2% major depression and 3.4% dysthymia). Thirty-three percent of women and 16% of men showed lifetime Axis I diagnoses; 24% of women compared to 9% of men showed lifetime major depression. Forty-one percent of men and 28% of women showed lifetime Axis II disorders other than substance abuse/dependence; 40% of men showed evidence of lifetime ASP compared to 15% of women. Lifetime major depression could be diagnosed as occurring independent of substance use/dependence in only 3.5% of clients; lifetime dysthymia was diagnosed as occurring independent of substance use/dependence in only 2% of clients. Regardless of whether psychiatric diagnosis was independent of or consequent to substance use/dependence, comorbidity was significantly more likely in clients with earlier age of first use of any drugs, with use of opiates, and with a greater number of lifetime drug use diagnoses. Comorbid clients also were significantly more likely to show greater substance use and psychosocial problems measured by scores on the ASI and NEO-PI.

Calsyn, D. A., Fleming, C., Wells, E. A., & Saxon, A. J. (1996). Personality disorder subtypes among opiate addicts in methadone maintenance. Journal of Addictive Diseases, 10, 3–8.

The authors studied 196 male and 113 female consecutive admissions to an outpatient methadone maintenance program. Clients had an average of more than 13 years of opiate use. During the first month of treatment, all clients were administered the Millon Clinical Multiaxial Inventory (MCMI) in addition to a structured interview schedule. In accord with scoring interpretations based in earlier research and the suggestions of the test’s author, clients were grouped into personality disorder subtypes. It is important to note that Millon subtypes are not identical to DSM diagnoses. In terms of Millon subtypes—which are not mutually exclusive—32% of methadone clients could be characterized as showing affective disturbance, 36% narcissistic-antisocial disorder, and 17% psychotic symptoms. Female clients were significantly more likely than males to show evidence of affective and psychotic disorders. African American clients were more likely to show evidence of psychotic symptoms and less likely to show evidence of affective disorders than Hispanic and white clients. Over an 18-month period, retention was unrelated to affective disorder or psychotic symptoms. Methadone clients categorized as withdrawn-negativistic according to the Millon (13% of admissions) were significantly more likely than all other subtypes to be retained in treatment.

Craig, T. J., & DiBuono, M. (1996). Recognition of comorbid psychopathology by staff of a drug detoxification unit. American Journal on Addictions, 5, 76–80.

This article studied the capacity of staff of a hospital-based drug detoxification unit to determine and refer clients with a concurrent psychiatric disorder. Staff were nursing personnel and "addiction therapists" whose number and backgrounds were not specified. Clients with "significant" but unspecified psychopathology were screened out and referred to other treatment units prior to the start of the study. Drug abuse treatment staff were encouraged to refer clients for psychiatric evaluation whenever psychiatric difficulty was suspected. Clients referred for evaluation (N = 131) and a random sample of clients who were not referred (N = 91) were studied. One staff psychiatrist interviewed all clients referred and not referred and assigned DSM-III-R diagnoses as appropriate. It is unclear whether diagnoses were made blindly. Psychiatric diagnoses for both Axis I and II disorders (nonsubstance use/dependence) were made for 84% of the referred sample and 21% of the nonreferred sample. Depressive disorder was the most common diagnosis in the referred sample (40%), but constituted only 7% of the nonreferred sample. Personality disorders were the most common diagnoses in the nonreferred sample (10%) and the second most common diagnoses in the referred sample (30%). Cocaine and heroin (but not other opiate) users were the least likely to receive diagnoses for other than substance use/dependence. Whereas 21% of the nonreferred sample was not identified as showing psychiatric disorder by drug abuse treatment staff, 16% of the referred clients were inappropriately identified as showing psychiatric disorder. The authors conclude that although staff are "relatively successful" in identifying psychiatric disorder, a substantial number of clients nonetheless would be missed in screening conducted by drug abuse treatment staff alone.

Dolan, M. P., Black, J. L., Malow, R. M., & Penk, W. E. (1991). Clinical differences among cocaine, opioid, and speedball users in treatment. Psychology of Addictive Behavior, 5, 78–84.

The authors studied a sample of 144 male drug abusers entering a 30-day inpatient drug treatment program. Subjects had to have IQs of at least 80, be literate, and show no evidence of organic or functional psychosis to qualify for inclusion in the research. Subjects were administered the Raven Progressive Matrix test, the Minnesota Multiphasic Personality Inventory (MMPI), and the Drug Abuse Screening Test (DAST). Subjects were categorized into cocaine users (N = 33), opioid users (N = 90), and speedball users (N = 21) based on drug preference, frequency of use, and DSM-III-R criteria for drug dependence. Opioid users had significantly higher IQs than the other drug-using groups. Speedball users showed significantly greater deviance than the other groups on scales of the MMPI, specifically on scales measuring hypochondriasis (Hs), hysteria (Hy), depression (D), psychopathic deviance (Pd), and social interaction (Si). When age and race were covaried out, Hy, Pd, and Si still differentiated speedball users from opiate users, while Pd and Si differentiated speedball users from cocaine users. Thus, although all groups showed some level of psychopathology in terms of elevations on the several MMPI scales, speedball users displayed significant deviance from their drug-using peers.

Flynn, P. M., Luckey, J. W., Brown, B. S., Hoffman, J. A., Dunteman, G. H., Theisen, A. C., Hubbard, R. L., Needle, R., Schneider, S. J., Koman, J. J., Atef-Vahid, M., Karson, S., Palsgrove, G. L., & Yates, B. T. (1995). Relationship between drug preference and indicators of psychiatric impairment. American Journal of Drug and Alcohol Abuse, 21, 153–166.

The psychological and behavioral functioning of heroin (N = 146) and cocaine (N = 136) admissions to treatment programs in Washington, DC, were compared. At treatment entry, all clients were administered a structured interview exploring background and demographic variables as well as a battery of clinical measures. Two weeks after treatment entry, clients were administered the Millon Clinical Multiaxial Inventory, Version II (MCMI-II). Clients were grouped as heroin or cocaine users based on their reports of favorite drug. Self-reports of drug preference were highly correlated with reports of frequency of use of the drugs named. In comparison to heroin users, cocaine users were significantly younger, were more likely never to have married, and were more likely to have been using drugs for a shorter period. Whereas substantial proportions of both heroin and cocaine users reported behaviors indicating psychological problems, cocaine users were significantly more likely to report several behaviors. Thus, 34% of heroin users reported periods of depression compared to 51% of cocaine users; 19% of heroin users reported suicidal ideation compared to 29% of cocaine users; 14% of heroin users reported suspiciousness compared to 26% of cocaine users; and 23% of heroin users reported physical abuse compared to 33% of cocaine users. MCMI-II scores indicated that cocaine users also were significantly more likely to show avoidant (i.e., inhibited and self-doubting) and self-defeating behaviors than heroin users.

Halikas, J. A., Crosby, R. D., Pearson, V. L., Nugent, S. M., & Carlson, G. A. (1994). Psychiatric comorbidity in treatment-seeking cocaine abusers. American Journal on Addictions, 3, 25–35.

The authors of this paper studied the psychiatric diagnoses of 207 adults who met DSM-III-R criteria for cocaine dependence and were admitted to a cocaine treatment program in response to newspaper advertisements for study participants. To be eligible for the study, clients currently had to be using cocaine, but could not be considered suicidal; have histories of head injury or seizures, schizophrenia, schizoaffective disorder, or bipolar disorder; or currently be taking any psychotropic medication. At intake, clients were given the Diagnostic Interview Schedule (DIS), the Addiction Severity Index (ASI), the Drug Impairment Rating Scale for Cocaine, and the Minnesota Cocaine Craving Scale. Of the 207 clients, 62% met DSM-III-R criteria for a current disorder other than substance use/dependence. Affective disorders were reported for 17% of the sample, anxiety disorders for 30%. The most frequent disorders identified were phobic syndromes (27%), posttraumatic stress (18%), and major depression (16%). (Interrater reliabilities for the DIS are not reported.) Lifetime disorders (73% of the sample) were most commonly antisocial personality (40% lifetime, not reported for current functioning), phobia (34%), posttraumatic stress (27%), and major depression (23%). Anxiety disorders preceded first regular drug use for 76% of clients with histories of anxiety disorder, but first regular drug use preceded affective disorder for 65% of clients with histories of affective disorder

Joe, G. W., Knezek, L., Watson, D., & Simpson, D. D. (1991). Depression and decision-making among intravenous drug users. Psychological Reports, 68, 339–347.

This study included 145 out-of-treatment injection drug users located through outreach efforts in Dallas, Texas. All subjects were administered the AIDS Initial Inventory (AIA), an interview schedule exploring respondents’ HIV risk-taking behaviors, as well as the Beck Depression Inventory (BDI) and depression and decision-making scales developed by the investigators. Based on the BDI, 83% of subjects showed depression, with 42% showing severe depression, 30% showing moderate depression, and 12% showing mild depression where cut-offs were employed to minimize the risk of false negatives. Where action was taken to minimize the risk of false positives, 23% were categorized as showing severe depression, with 39% showing moderate and 21% showing mild depression. Female, younger, and less educated injection drug users were significantly more likely to show depression than their counterparts. In addition, depression was correlated significantly to decision making such that the greater the depression, the lower the confidence in one’s ability to make decisions.

Kadden, R. M., Kranzler, H. R., & Rounsaville, B. J. (1994). Validity of the distinction between "substance-induced" and "independent" depression and anxiety disorders. American Journal of Addictions, 4, 107–117.

This study explored the distinction between substance-induced psychopathology and psychopathology independent of substance abuse (i.e., intrinsic to the individual's functioning). One hundred subjects were recruited from an inpatient (N = 83) and a day treatment (N = 17) substance abuse program during their first week of treatment. Treatment was provided for alcohol dependence, drug dependence, or a combination of alcohol and drug dependence. The authors report that analysis of the relation between comorbid disorder categories and differing substance use categories (i.e., alcohol dependence, drug dependence, or alcohol and drug dependence) found no differences in comorbid disorder categories by substance use categories. An average of 11 days after treatment entry, subjects received a diagnostic interview to determine diagnosis with specific regard to substance use disorder, depressive disorder, phobic disorder, and anxiety disorder. Subjects also were administered the Addiction Severity Index (ASI) and the Beck Depression Inventory (BDI). Two sets of diagnoses were obtained. One was dependent simply on the presence of the minimal number of diagnostic criteria required to satisfy a particular category; the second required the diagnostic criteria either to precede the substance use disorder by 6 months or to have occurred during a 6-month period of abstinence. The number of diagnoses for major depression and/or anxiety disorder under conditions of independence from substance use disorder decreased to less than 25% of the number made for the same subjects without regard to presence of substance use disorder. That is, study findings suggest the importance of ruling out drug effects in making a diagnosis for other than substance use disorder. However, as discussed by the authors, agreement on diagnoses of major depression and/or diagnosis between independent raters for the same subjects dropped from an acceptable level for substance-induced comorbid disorders to a marked reduction in agreement for independently occurring comorbid disorders. On objective measures, there was an overall decline on BDI and ASI scores from baseline. However, the group with comorbid disorder in association with substance use disorder showed an increase in BDI scores at follow-up, whereas the group with comorbid disorder independent of substance use disorder showed a substantial decrease. The group with independent comorbid disorder did not constitute a more severe risk group. All groups (no comorbid disorder, comorbid disorder in association with substance use disorder, and comorbid disorder independent of substance use disorder) showed decreases on all seven scales of the ASI, including psychological functioning. The authors believe that study problems preclude them from clearly establishing the need for making distinctions between substance-induced major depression and/or anxiety and the independent expression of major depression and/or anxiety, but recommmend continuing study in this area.

Kranzler, H. R., Satel, S., & Apter, A. (1994). Personality disorders and associated features in cocaine-dependent inpatients. Comprehensive Psychiatry, 35, 335–340.

The authors studied 50 newly admitted inpatients who met DSM-III-R criteria for cocaine dependence and were not seen as being acutely psychotic. The clients were administered the Structured Clinical Interview for DSM-III-R for Personality Disorders (SCID-II), a Cocaine Experience Questionnaire, the computerized version of the Diagnostic Interview Schedule for DSM-III-R (DIS), the Wisconsin Scales of Psychosis Proneness, the Timeline Followback Questionnaire (a measure of frequency of use of alcohol and other drugs tied to significant life events), and a Family History Questionnaire. Seventy percent of cocaine-dependent clients received at least one personality disorder (Axis II) diagnosis other than substance use, most typically borderline (34%), antisocial (28%), and/or narcissistic (28%). Those receiving an Axis II diagnosis did not differ from those who did not in demographic or background characteristics, or for current or past drug use history. However, those clients showing Axis II disorder were significantly more likely to show evidence of depressive and anxiety disorders as well as scores on the Wisconsin Scales indicating psychosis proneness. In addition, family history of alcoholism was correlated with clients’ substance use and anxiety disorders.

Marlowe, D. B., Husband, S. D., Lamb, R. J., Kirby, K. C., Iguchi, M. Y., & Platt, J. J. (1995). Psychiatric comorbidity in cocaine dependence. Diverging trends, Axis II spectrum, and gender differentials. American Journal on Addictions, 4, 70–81.

This study involved 100 consecutive admissions to an outpatient cocaine treatment program to determine the nature and extent of psychiatric comorbidity. Subjects were administered the Structured Clinical Interview for DSM-III-R-Patient Edition (SCID-P) and the SCID-II to determine personality disorder. Interrater reliability was not assessed but was thought to be high based on training provided and the findings of previous comparable studies. Ninety-one of the 100 clients received at least one DSM-III-R diagnosis additional to cocaine use/dependence and an average of nearly three diagnoses. Lifetime diagnoses most frequently involved personality disorder (73%), with antisocial (23%), borderline (22%), and paranoid (21%) the most frequent individual categories. In addition to cocaine use disorder, 37% met criteria for lifetime alcohol dependence, 23% marijuana dependence, and 11% opiate dependence. Only 13% were diagnosed as showing lifetime affective disorder; 11% met DSM-III-R criteria for current affective disorder; and 8% met criteria for lifetime and current anxiety disorder. Males were significantly more likely to be diagnosed with antisocial personality disorder and lifetime alcohol dependence than females.

Milby, J. B., Sims, M. K., Khuder, S., Schumacher, J. E., Huggins, N., McLellan, A. T., Woody, G., & Haas, N. (1996). Psychiatric comorbidity: Prevalence in methadone maintenance treatment. American Journal of Drug and Alcohol Abuse, 22, 95–107.

This follow-up study included 102 subjects who had been admitted to three methadone maintenance treatment programs 5-7 years earlier. At time of interview, 51% were abstinent for illicit drugs, 54% were employed full time, and 63% were in methadone treatment. Subjects were administered a structured interview incorporating diagnostic questions consistent with decision making for DSM-III-R diagnoses. At least one affective disorder was present in 58% of subjects, with major depressive disorder the most common (31%). At least one anxiety disorder was present in 55% of subjects, with posttraumatic stress disorder the most common (31%) and general anxiety disorder the next most common (21%). Subjects were predominantly veterans of the Vietnam War. Finally, 36% of subjects showed evidence of both anxiety and affective disorders, while 26% showed no evidence of any disorder other than substance use. Neither retention in treatment nor abstinence from illicit drugs was found to be associated with anxiety or affective disorders.

Miller, N. S., Klamen, D., Hoffmann, N. G., & Flaherty, J. A. (1996). Prevalence of depression and alcohol and other drug dependence in addictions treatment populations. Journal of Psychoactive Drugs, 28, 111–124.

The authors studied 6,355 voluntary admissions to 38 inpatient and 19 outpatient treatment programs affiliated with the Comprehensive Assessment and Treatment Outcome Research (CATOR) registry. To be included in the study, the client had to have a current DSM-III-R diagnosis of substance use disorder and had to have undergone evaluation for a diagnosis of major depression. The treatment programs were described as 12-step-based abstinence-oriented programs. Diagnoses were based on interviews conducted at client intake. The client population differs from others in its comparatively high rates of full-time employment (64%), at least high school education (85%), marital status (43% married), and income level (37% had personal incomes of $20,000 or more and 54% had family incomes of $20,000 or more). In addition, 31% reported having received previous psychiatric treatment, and 24% had received treatment specific to depression. On the basis of the intake interview, 44% of clients had a lifetime diagnosis of major depression. Major depression was most likely to be associated with opioid, stimulant, and prescription drug use as specific drugs of abuse; with number of drugs used; and with early onset of alcohol and marijuana use. Relapse at 1 year following treatment was equivalent for clients with and without diagnoses of major depression. Relapse was significantly associated with lower rates of lifetime depression for male cocaine users.

Rounsaville, B. J., Weissman, M. M., Kleber, H., & Wilber, C. (1982). Heterogeneity of psychiatric diagnosis in treated opiate addicts. Archives of General Psychiatry, 39, 161–166.

The authors studied 533 opiate-using admissions to a Connecticut multimodality facility. The subjects constituted a largely convenience sample of 157 new admissions, 204 consecutive new admissions, 120 methadone maintenance clients (at least 3 months in program), and 60 Latino clients composed of both treatment applicants (N = 30) and therapeutic community residents (N = 30). (Note that the figures add to 541; the difference between the total of subsamples and the reported total sample is not explained.) The Schedule for Affective Disorders and Schizophrenia-Lifetime (SADS-L) was used to make diagnoses on the basis of the Research Diagnostic Criteria (RDC), in terms of both current and lifetime functioning. Data for all subsamples were combined after they were determined to be comparable with regard to rates of psychiatric disorder. With regard to current functioning, 24% of the total sample showed evidence of major depression, 3% other affective disorders, 9% phobia, 1% panic disorder, 1% generalized anxiety, and 14% alcoholism. Overall, 70% met criteria for some current disorder including the criteria for personality disorder. With regard to lifetime functioning, 54% of the total sample met the criteria for major depression, 35% for alcoholism, 27% for antisocial personality, 19% for intermittent depression, 17% for labile personality, 10% for phobic disorder, 8% each for minor depression and schizotypal personality, 7% for hypomanic disorder, and 5% for anxiety disorder. Overall, 87% of opiate addicts met the criteria for some lifetime psychiatric disorder other than drug abuse, with 74% meeting criteria for some affective disorder. Among demographic differences in psychiatric diagnosis, women were more likely than men to show major depression and selected other affective disorders as well as both panic and phobic disorders, whereas men were more likely to show evidence of antisocial personality disorder. Education was associated with a greater likelihood to show psychiatric disorder generally; marital separation and divorce were associated with major depression, cyclothymia, and phobic disorder. Whites were generally most likely (relative to Latinos and African Americans) to give evidence of psychiatric disorder and African Americans least likely. Also, 52% of opiate addict subjects gave evidence of two or more psychiatric diagnoses other than drug abuse, while 35% gave evidence of one diagnosis other than drug abuse. In particular, depression and alcoholism were linked. The authors suggest the need for careful diagnosis as well as the importance of providing psychological and pharmacological treatment as needed and responding to the potential for alcohol abuse in clients.

Rutherford, M. J., Cacciola, J. S., Alterman, A. A., & McKay, J. R. (1996). Assessment of object relations and reality testing in methadone patients. American Journal of Psychiatry, 153, 1189–1194.

This study included 240 admissions to two urban methadone programs. Three weeks after entry, subjects were administered the Structured Clinical Interview for DSM-III-R (SCID), the Structured Interview for DSM-III Personality Disorders, and the Bell inventory measuring object relations (in terms of scales assessing alienation, insecure attachment, egocentricity, and social incompetence) and reality testing (in terms of scales measuring reality distortion, uncertainty of perception, and hallucinations and delusions). No relationships were found between Bell inventory scores regarding reality testing or object relations and methadone clients' diagnoses for affective disorders, panic disorder, or anxiety disorders. No relationships were found between Bell inventory scores and number of Axis I disorders shown by methadone clients. However, when Bell inventory scores and number of diagnosed personality disorders (e.g., schizoid, avoidant, antisocial) were compared, a significant relationship was found between number of disorders and both reality testing and object relations scale scores. Negative scale scores increased with number of disorders present. Reality testing scale scores were particularly disturbed for the group with three or more personality disorders. Scores on the object relations scales and on the uncertainty of perception scale differentiated between groups showing no combination of Axis I and Axis II disorders and those showing such a combination. The authors conclude that it is important to explore for comorbid Axis II disorders among methadone clients in accord with that group's display of problems in terms of object relations and reality testing.

Weissman, M. M., Slobetz, F., Prusoff, B., Mezritz, M., & Howard, P. (1976). Clinical depression among narcotic addicts maintained on methadone in the community. American Journal of Psychiatry, 133, 1434–1438.

This article describes a study of 106 methadone maintenance clients in one Connecticut program, of whom 62 were new admissions and 44 had been in treatment for an average of 1 year. The two client groups were combined for study, since there were no demographic or clinical differences between them. Client functioning was rated by a clinician using the Raskin Depression Scale and the Hamilton Rating Scale. In addition, information on client functioning was obtained using the Symptom Check List (SCL-90) and the Social Adjustment Scale; information on client history was obtained using a semistructured interview to explore life events. The majority of clients (64%) reported parents or siblings as showing histories of mental illness, and 54% reported they themselves had received mental health treatment, with 24% reporting mental health treatment prior to their becoming addicted. On the basis of the Raskin Scale, 32% of clients showed evidence of clinical depression. A comparison of depressed and nondepressed clients (in accord with Raskin Scale scores) showed that depressed clients were significantly more likely to have histories of alcohol abuse, but did not differ on other drug use or length of time using heroin; were more symptomatic for all items on the SCL-90 and the vast majority of Hamilton Scale scores; showed greater maladjustment in social roles (e.g., family unit, social, and leisure activities); and reported a greater number of stressful events occurring in the preceding 6 months (e.g., illness in family, conflict with spouse). The authors recommend an expanded use of antidepressant medication as consistent with their findings.


Alexander, M. J., Craig, T. J., MacDonald, J., & Haugland, G. (1994). Dual diagnosis in a state psychiatric facility. American Journal on Addictions, 3, 314–324.

The authors studied 50 consecutive admissions to a state psychiatric facility to determine the extent of substance use problems in a psychiatric population. Based on diagnostic interview, case manager reports, or client self-report, clients were grouped as showing either dual diagnosis (N = 24) or single psychiatric diagnosis only (N = 25). (One client’s data were not useable.) For clients to be categorized as dually diagnosed, alcohol or drug problems had to be evident within 6 months of treatment entry. Problems were rated on a four-point continuum of severity based on the extent to which substance use created problems for the individual and the community. Nearly half (49%) of clients showed a current alcohol or drug use problem. Dually diagnosed clients were significantly more likely than single diagnosis clients to be male, to be younger, and to have been admitted to first psychiatric treatment earlier. Dually diagnosed clients were significantly more likely to have childhood histories including trauma or the potential for trauma in terms of exposure to physical abuse, parental mental illness, and extended drug or alcohol use in the home. Additionally, dual diagnosis clients reported greater conflict with family members, greater residential instability, and greater involvement with both the criminal justice and health care systems. Over all clients, 35% were currently using marijuana, 27% crack, 20% other forms of cocaine, and 6% heroin.

Brady, K., Casto, S., Lydiard, R. B., Malcolm, R., & Arana, G. (1991). Substance abuse in an inpatient psychiatric sample. American Journal of Drug and Alcohol Abuse, 17, 389–397.

One hundred consecutive admissions to a VA inpatient psychiatric facility were studied. To understand substance use disorders, a modified version of the Structured Clinical Interview for DSM-III-R was administered by any of three interviewers (interrater reliability not reported), and a nonblind chart review was conducted to clarify history of psychiatric treatment and psychosocial functioning. Based on diagnostic interviews, 29% of psychiatric admissions met DSM-III-R criteria for a substance use disorder (i.e., could be viewed as dually disordered). Dually disordered clients were significantly younger than those singly diagnosed. Over all admissions, alcohol was the drug of choice for 68%, stimulants (not further clarified) for 17%, marijuana for 7%, and opiates for 1%. Only 40% of admissions with current or historic substance use problems had ever received treatment for their substance abuse problem.

D’Mello, D. A., Boltz, M. K., & Msibi, B. (1995). Relationship between concurrent substance abuse in psychiatric patients and neuroleptic dosage. American Journal of Drug and Alcohol Abuse, 21, 257–265.

This article examined the drug use behavior of 58 consecutive admissions to a psychiatric inpatient facility. Clients were predominantly schizophrenic (29%) and showed bipolar disorder (25%), major depression (15%), and schizoaffective disorder (10%). The remainder showed adjustment and organic disorders. Drug abuse was determined using the substance abuse section of the Schedule for Affective Disorders and Schizophrenia (SADS). Overall, 58% of psychiatric patients showed evidence of diagnosable drug use (70% of men and 50% of women). In terms of psychiatric diagnosis, drug use was shown by 58% of schizophrenic clients, 55% of bipolar manic clients, and 56% of major depression clients.

Grilo, C. M., Becker, D. F., Fehon, D. C., Edell, W. S., & McGlashan, T. H. (1996). Conduct disorder, substance use disorders, and coexisting conduct and substance use disorders in adolescent inpatients. American Journal of Psychiatry, 153, 914–920.

The authors studied 165 consecutive adolescent (ages 12-18) admissions to an inpatient and private psychiatric hospital to explore the characteristics of adolescents showing evidence of dual disorder as compared to those not showing dual disorder. Three groups of clients were identified based on DSM-III-R diagnoses: (a) adolescents showing conduct disorder only (N = 25), (b) adolescents showing substance use disorder only (N = 24), and (c) adolescents showing both substance use and conduct disorders (N = 54). Shortly after admission, clients were given the Schedule for Affective Disorders and Schizophrenia for School-Age Children–Epidemiologic Version (K-SADS-E) and the Personality Disorder Examination. To be diagnosed with a personality disorder, the adolescent had to give evidence of a disorder having been significantly present for a period of 3 or more years. Antisocial personality disorder was not considered because of client age. Interrater reliability was high for both interview schedules (kappas of .77 and .84). Subjects also were rated with the DSM-III-R Global Assessment of Functioning Scale. The group showing dual disorder had a significantly greater proportion of males than the group showing substance use disorder only. Clients in the conduct disorder only group were significantly more likely to show evidence of attention deficit disorder and had an earlier age of first psychiatric contact. Clients showing evidence of conduct disorder with substance use disorder and conduct disorder only were significantly more likely to show anxiety symptoms than clients showing evidence of substance use disorder only. However, differences associated with anxiety symptoms were associated with gender, such that women were responsible for the higher rates of anxiety in both groups. Groups did not differ on demographic variables other than gender or on scores on the Global Assessment of Functioning Scale.

Lehman, A. F., Myers, C. P., Corty, E., & Thompson, J. W. (1994). Prevalence and patterns of "dual diagnosis" among psychiatric inpatients. Comprehensive Psychiatry, 35, 106–112.

This study involved 435 consecutive admissions who consented to participate in a study at two publicly funded psychiatric hospitals serving an inner-city population. All subjects were given the Structured Clinical Interview for DSM-III-R Inpatient Version (SCID-P) within 1 week of admission. Interviewers all had a minimum of masters degrees and 1 year experience and training in the SCID-P; the interviewers showed an interrater reliability of .81 in diagnoses provided. Over half the subjects (56%) had substance abuse diagnoses with alcohol (39%), the most frequently abused drug, followed by cocaine (18%) and marijuana (11%). Thirty-five percent of subjects showed psychiatric syndromes associated with their drug use and had no histories of psychiatric syndromes prior to their drug use. The most common psychiatric diagnoses consequent to drug use were major depression (15%) and brief psychotic disorders (8%). The 56% of clients with current substance abuse disorder includes some subjects with substance use as their only disorder; 54% met the criteria for dual diagnosis (i.e., a current substance use disorder and current Axis I disorder). However, when the authors deleted those subjects whose Axis I disorder followed their substance use disorder, only 25% of subjects showed independent disorders. The remaining subjects, who typically would be characterized as part of the dual disorder group, showed syndromes that were "definitely due to" substance use (13%) or "possibly related to" substance abuse (16%). Subjects whose psychiatric diagnosis appeared to be consequent to substance use differed significantly from those whose psychiatric diagnosis appeared to be independent of their substance use in terms of significantly higher rates of major depression and lower rates of schizophrenia and anxiety disorders as well as significantly fewer and shorter psychiatric hospitalizations. Opiate and cocaine use, as opposed to marijuana use, were significantly more likely to be associated with psychiatric diagnoses appearing consequent to substance use. The authors emphasize the importance of precise diagnosis to assure both accurate reporting and appropriate services.

Miller, N. S., Belkin, B. M., & Gibbons, R. (1994). Clinical diagnosis of substance use disorders in private psychiatric populations. Journal of Substance Abuse Treatment, 11, 387–392.

This study analyzed the extent of substance abuse/dependence in a randomly selected sample of 200 adult psychiatric patients admitted to inpatient units of a private university-affiliated psychiatric hospital. No patients were admitted to this acute care facility solely for alcohol or other drug use. Diagnoses were made at discharge by psychiatrists blind to the nature of the study. Findings were reported separately for each of the two units to which patients were assigned. Both substance use and other psychiatric diagnoses (i.e., dual diagnoses) were seen as appropriate for 25% of patients on one unit and for 33% of patients on the second unit; psychiatric but not substance use diagnosis was reported for 67% of patients on one unit and for 75% on the second unit. Males on both units were significantly more likely to show evidence of comorbid disorders involving varying combinations. Substance use was found with equal frequency in psychotic and nonpsychotic patients. Alcohol and drug use were not associated with one Axis I or II diagnosis more than any other. The authors findings support a view of high prevalence of drug use in private psychiatric populations. That relationship has been reported for public hospital psychiatric populations as well.


Brooner, R. K., Bigelow, G. E., Strain, E., & Schmidt, C. W. (1990). Intravenous drug abusers with antisocial personality disorder: Increased HIV risk behavior. Drug and Alcohol Dependence, 26, 39–44.

The author studied 100 injection drug users recruited from clients retained in a methadone maintenance program at least 1 month (N = 66) and persons recruited from the community (N = 34). A diagnostic interview and an interview assessing HIV risk-taking behavior were administered. The most common substance use diagnoses involved opiates (93%) or cocaine (70%) with multiple substance use diagnoses the most likely. In addition, 50% of subjects met criteria for lifetime disorder other than substance abuse, 36% met criteria for antisocial personality disorder, and 16% met criteria for major depression (20% of the treatment sample). Antisocial personality diagnosis was associated with significantly greater needle sharing. In further analysis making use of multiple regression analyses, antisocial personality disorder was the only significant predictor of needle sharing. Cocaine use was the only significant predictor of frequency of injection. The authors conclude that, within the group of drug injectors, antisocial personality disorder places the individual at increased risk for contracting and spreading HIV infection.

Brooner, R. K., Greenfield, L., Schmidt, C. W., & Bigelow, G. E. (1993). Antisocial personality disorder and HIV infection among intravenous drug abusers. American Journal of Psychiatry, 150, 53–58.

A total of 272 injection drug users were recruited for study of the relationship of drug use, psychopathology, and HIV infection. Of the 272 subjects, 140 were obtained from methadone programs, and 132 were not in drug treatment and were recruited from the community. All subjects had been using injection drugs for at least 3 months prior to recruitment, had used injection drugs continuously for at least 3 months at some time in their lives, and had less than 12 months of treatment in the preceding 4-year period. Structured and semistructured interviews were administered to determine DSM-III-R diagnoses specific to antisocial personality and substance abuse disorders, and to HIV risk behaviors. In addition, bloods were drawn for HIV testing. Overall, 44% of injection drug users met the DSM-III-R criteria for antisocial personality disorder. Subjects showing antisocial personality disorder were significantly more likely to be male and to be out of treatment. Subjects diagnosed as showing antisocial personality disorder also were significantly more likely to be involved in needle risk behaviors (e.g., with regard to frequency of drug injections and of needle sharing) and to be HIV infected. In addition, subjects diagnosed as showing antisocial personality disorder had significantly higher rates of use of cocaine, alcohol, sedatives, and hallucinogens. Thus, antisocial drug users pose a particular risk to themselves and to their communities.

Camacho, L. M., Brown, B. S., & Simpson, D. D. (1996). Psychological dysfunction and HIV/AIDS risk behavior. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology, 11, 198–202.

This study included 834 injection drug users admitted to three methadone maintenance programs in three different cities. Baseline measures involved the administration of an 88-item questionnaire including, in part, measures of depression, anxiety, and hostility, and an overall measure of psychological dysfunction. In addition, clients were given an interview assessing demographic and background characteristics as well as HIV risk-taking behaviors. Psychological dysfunction was significantly related to needle risk behaviors (use of "dirty works," number of people with whom needles are shared, and number of times needles are shared with strangers) and to sexual risk behaviors (number of partners, unprotected sex with injection drug users, and trading sex for drugs). Cocaine use (in addition to opiate use) also was significantly related to needle and sex risk behaviors. In additional analysis, psychological dysfunction was significantly related to needle and sex risk behaviors independent of cocaine use or other client characteristics.

Ferrando, S. J., Wall, T. L., Batki, S. L., & Sorensen, J. L. (1996). Psychiatric morbidity, illicit drug use and adherence to zidovudine (AZT) among injection drug users with HIV disease. American Journal of Alcohol and Drug Use, 22, 475–487.

A retrospective study was conducted of the clinical charts of 57 HIV-infected methadone maintenance clients receiving AZT in association with their drug abuse treatment program. Of the 57 clients, 46 (81%) were given Axis I diagnoses consequent to psychiatric review and psychiatric follow-up consultation. The clients requiring psychiatric follow-up (N = 46) were compared to those not requiring follow-up on measures of adherence to AZT treatment, illicit drug use measured through urine tests given a minimum of once monthly, and change in immune status as measured by clients' T-cell counts. The study covered a 4-month period of drug treatment. Clients requiring psychiatric attention were significantly less likely to adhere to AZT regimens than were those without Axis I diagnoses. The groups did not differ in terms of drug-positive urines or in levels of change in T-cell counts. The authors express concern about the generally poor rates of compliance with medication regimens and the need for increased attention to mental health issues in an effort to increase compliance and to counter the possible negative effects of awareness of HIV infection on psychological functioning.

Lipsitz, J. D., Williams, J. B. W., Rabkin, J. G., Remien, R. H., Bradbury, M., el Sadr, W., Goetz, R., Sorrell, S., & Gorman, J. M. (1994). Psychopathology in male and female intravenous drug users with and without HIV infection. American Journal of Psychiatry, 151, 1662–1668.

This article reports the results of a study of injection drug users recruited from a New York methadone maintenance program (N = 53) and from New York medical clinics (i.e., a sample composed of persons both in and out of treatment [N = 170]). All subjects had to have injected drugs at least 10 times in their lives and had to have injected at some time in the preceding year; in addition, methadone clients had to have been in treatment at least 3 months. All subjects knew their HIV status at time of recruitment, but were retested as a part of the study protocol. Individuals with active AIDS were excluded from study; HIV-negative and HIV-positive subjects with no or mild to moderate symptoms were retained for study. Subjects were administered a diagnostic interview designed to clarify DSM-III-R diagnosis, depression and anxiety scales of the Symptom Check List (SCL-90R), the Global Assessment of Functioning Scale, a measure of level of stress felt by the subject, the Marlow-Crowne Social Desirability Scale, and an assessment of stage of HIV disease. It was found that 124 subjects (56%) were seropositive and 99 (44%) seronegative. The only significant diagnostic category for the full sample was current depression (26%), with no differences between seropositive and seronegative subjects in rates of depression. However, when comparisons were made within gender, seropositive males showed a significantly greater number of diagnoses of depression than seronegative males; no comparable differences were found between seropositive and seronegative females. Scores on depression and anxiety scales of the SCL-90R did not differ significantly for seropositive and seronegative samples by gender or overall; however, scores did differ between males and females regardless of serostatus, with women obtaining significantly higher scores on both measures. Similarly, there were no differences between seropositive and seronegative groups on the measure of stress, but there were significant differences between males and females regardless of serostatus, with females showing greater levels of stress. As with SCL-90 scales, scores on the Global Assessment of Functioning Scale approached scores for psychiatric patients. Whereas HIV-positive males had significantly lower scores on this measure than HIV-negative males, HIV-positive females had significantly higher scores on this measure than HIV-negative females. Stage of HIV disease was associated with psychological functioning such that advanced stage of disease was significantly related to a diagnosis of depression, as well as elevated scores on measures of depression, anxiety, and stress. In addition, methadone clients were twice as likely to be diagnosed as showing depression as injection drug users who were not in treatment.

Malow, R. M., Corrigan, S. A., Pena, J. M., Calkins, A. M., & Bannister, T. M. (1992). Mood and risk behavior among drug-dependent veterans. Psychology of Addictive Diseases, 6, 131–134.

The authors studied 170 consecutive male admissions to a Veterans Administration drug treatment program meeting DSM-III-R criteria for drug dependence and showing no evidence of organic brain syndrome. Subjects were not being maintained on methadone or prescribed psychotropic medications and were predominantly involved with cocaine (69%). Indeed, most of those who were not dependent on cocaine by DSM-III-R criteria were nonetheless using cocaine at some level. At least 10 days after admission, subjects were administered semistructured individual interviews assessing demographic and background information, a structured interview to clarify DSM-III-R substance dependence diagnosis, the State-Trait Anxiety Scale measuring anxiety and depression, the Beck Depression Inventory (BDI), and a structured questionnaire measuring HIV risk-taking behaviors. Needle risk behaviors were shown by 14% of the sample (needle risk was a composite of patterns of injection drug use, sharing, and needle cleaning) and were found to be related to scores on the BDI. Sex risk behaviors were shown by 78% of the sample (sexual risk was a composite of number and type of partners, type of sexual activity, and condom use) and were not related to BDI scores. A high depression group created from BDI scores was significantly more likely to be engaged in needle risk behaviors than either a moderate or low depression group. Similarly, the group classified as showing evidence of low anxiety on the basis of test scores was significantly less likely to be involved in needle risk behaviors than either of the other two groups, which did not differ from each other. Finally, the group characterized by DSM-III-R criteria as dependent on more than one substance (typically cocaine and one or more other substances) was significantly more likely to show needle but not sexual risk taking than were other DSM-III-R substance dependence groups. Overall, investigators report a relationship between psychological symptoms and needle risk taking, but no relationship to the more pervasive sexual risk taking.

Simpson, D. D., Knight, K., & Ray, S. (1993). Psychosocial correlates of AIDS-risk drug use and sexual behaviors. AIDS Education and Prevention, 5, 121–130.

This study explored the relation between HIV risk-taking behavior and the psychological functioning of a sequential sample of 194 out-of-treatment injection drug users identified through outreach. Subjects were administered the AIDS Initial Assessment (AIA) and a brief trailer form. Both forms explored needle and sexual risk taking as well as standard demographic and background information. Subjects also were given a battery of measures developed by the authors to assess self-esteem, depression, anxiety, motivation (for change and for drug treatment), decision making, and risk taking. Composite needle and sex risk indices were developed and related to psychosocial measures. Needle risk was associated with high depression and anxiety, low self-esteem, and low decision-making confidence as well as high desire for help (a component of motivation) and high drug problems. Psychosocial measures were unrelated to sexual risk index. Thus, concerns regarding needle risk, as reflected in the psychosocial measures employed, may help explain the comparative malleability of needle risk behaviors as compared to sexual risk taking.


Kleinman, P. H., Woody, G. E., Todd, T. C., Millman, R. B., Kang, S. -Y., Kemp, J., & Lipton, D. S. (1990).  Crack and cocaine abusers in outpatient psychotherapy.   In L. S. Onken & J. D. Blaine (Eds.), Psychotherapy and counseling in the treatment of drug abuse (pp. 24–35).  Washington, DC: U.S. Government Printing Office.

The effectiveness of three psychotherapeutic forms with 148 cocaine abusers who had either spouses (80%) or parents (20%) willing to participate in the client's treatment was studied. Subjects were randomly assigned to supportive-expressive therapy, family therapy, or paraprofessionally led group therapy. At intake, subjects were administered the Structured Clinical Interview for DSM-III-R (SCID), a drug history interview, the Symptom Check List (SCL-90), and the Addiction Severity Index (ASI). All subjects met diagnostic criteria for both lifetime and current substance (cocaine) use disorder; 72% used cocaine in the form of crack. In addition, 47% of subjects also met the diagnostic criteria for clinical depression with phobic disorders (number and percent unspecified), the only other Axis I disorder reported. Axis II disorders reported were passive-aggressiveness (20%), conduct disorder (19%), antisocial personality (18%), and paranoia (18%). Earlier onset of drug use (i.e., tobacco, marijuana, and cocaine) was associated with a diagnosis of major depression. Only therapist assignment appeared to be related to retention in treatment for one or more sessions.

Rounsaville, B. J., Glazer, W., Wilber, S. H., Weissman, M. M., & Kleber, H. D. (1983). Short-term interpersonal psychotherapy in methadone-maintained opiate addicts. Archives of General Psychiatry, 40, 629–636.

The authors studied 72 methadone-maintained opiate addicts in treatment for a minimum of 6 weeks and randomly assigned to either short-term interpersonal psychotherapy (IPT) consisting of at least 1 hour per week of counseling provided by a psychiatrist or psychologist (N = 37) or low-contact treatment consisting of 20 minutes per month of review by a psychiatrist of the client's social situation (N = 35). To qualify for study, subjects had to meet Research Diagnostic Criteria (RDC) for a diagnosis other than drug abuse using information collected from the Schedule for Affective Disorders and Schizophrenia–Lifetime (SADS–L). Outcomes for the two treatment groups were measured at 12 and 24 weeks after entry into the treatment conditions using (a) behavioral assessments including urinalyses, alcohol use, arrests, and employment; (b) locus of control; (c) the Maudsley Personality Inventory (MPI); and (d) change in the four problem areas selected by the client at outset as of greatest significance to him or her. In addition, assessment was made at 4-week intervals of (e) social functioning (using a social adjustment scale); (f) depression (using the Raskin Depression Scale); and (g) psychiatric symptoms (using the Symptom Check List [SCL-90]). The two groups did not differ at baseline in terms of psychiatric diagnosis, with 50% showing mood disorders (typically labile personality) and 17% showing antisocial personality as the dominant disorders. Dropout over the 24 weeks of study was greater in the IPT group (62%) than in the low-contact group (46%), with the IPT clients significantly more likely to withdraw voluntarily. The greater demands placed on IPT clients are seen as responsible for their greater rate of voluntary withdrawal. At 12 and 24 weeks, considering only those clients receiving a minimum of three sessions, both IPT and low-contact groups had significant reductions in Raskin Depression Scale scores, SCL-90 scores, and ratings of problem areas. There were no differences between groups in degree of improvement obtained. Similarly, there were no differences in behavioral outcomes between groups at 12 and 24 weeks. In further analysis comparing only treatment completers in the two groups, there were no significant differences between IPT and low-contact clients on any of the measures employed. Separating out depressed from nondepressed clients (using the Raskin Depression Scale) also did not indicate any advantage for IPT over low contact. The authors suggest that several features of their research design may have limited their capacity to demonstrate a psychotherapeutic effect.

Woody, G. E., Luborsky, L., McLellan, A. T., O'Brien, C. P., Beck, A. T., Blaine, J., Herman, I., & Hole, A. (1983). Psychotherapy for opiate addicts—Does it help? Archives of General Psychiatry, 40, 639–645.

In a study of the effectiveness of psychotherapy with opiate-dependent clients, 110 male methadone maintenance clients between 18 and 55 years of age were randomly assigned to (a) drug abuse counseling plus supportive-expressive therapy (SE [N = 32]); (b) drug abuse counseling plus cognitive-behavioral therapy (CB [N = 39]); and (c) drug abuse counseling only (DC [N = 39]). To be eligible for study, clients had to be neither psychotic nor organically impaired, be in methadone treatment a minimum of 2 months and a maximum of 6 months, have given written agreement to participate in the study (60% of clients), and have completed baseline measures and at least three counseling/therapy appointments within the first 6 weeks of treatment (60% of those agreeing to participate). Measures employed included the Addiction Severity Index (ASI), the Beck Depression Inventory (BDI), the Symptom Check List (SCL-90), the Maudsley Personality Inventory (MPI), the Shipley Institute of Living Scale, and the Schedule for Affective Disorders and Schizophrenia–Lifetime (SADS–L) and Change (SADS–C). The measures were administered at baseline and at 1 and 7 months after start of treatment. Retention and urinalysis data were also available from all clients located and interviewed at follow-up (92% of the sample). Over the 24 weeks of study, DC clients attended an average of 17 counseling sessions; SE clients attended an average of 12 sessions with counselors and 12 with therapists; CB clients attended an average of 12 sessions with counselors and 9.5 with therapists. All groups showed marked and typically significant positive self-reported change in behaviors (drug use, crime days, and licitly derived income) at 7-month follow-up. All groups showed significant diminutions in drug-positive urines generally and opiate-positive urines specifically. However, the DC group obtained a significantly greater number of opiate-positive urines than the therapy groups. At 7 months, SE clients, compared to the other groups, achieved significantly better scores on measures of employment and earned income and on the BDI and MPI, but significantly poorer scores on a composite measure of legal difficulties. CB clients, compared to the other groups, showed significantly poorer scores on a composite measure of drug use, had more days using heroin, and had higher levels of illicit income (at the .07 level of significance). DC clients, compared to the other groups, showed a significantly greater number of days using sedatives and poorer SCL-90 scores. The authors conclude that clients in all three groups derive benefit from methadone maintenance once they become engaged in treatment, and that some proportion of maintenance clients are interested in psychotherapy and can derive significant benefit from that addition to standard treatment.

Woody, G. E., McLellan, A. T., Luborsky, L., & O’Brien, C. P. (1985). Sociopathy and psychotherapy outcome. Archives of General Psychiatry, 42, 1081–1086.

Further analysis of a study of the effectiveness of psychotherapy (see Woody et al., 1983, this volume) explored the significance of a diagnosis of sociopathy for psychotherapy outcome. Employing DSM-III diagnoses derived from Schedule for Affective Disorders and Schizophrenia–Lifetime (SADS–L) interviews, 110 male methadone clients who had been provided supportive-expressive or cognitive-behavioral psychotherapy were subdivided into four groups: (a) opiate dependence only (N = 16); (b) opiate dependence and major depressive disorder (N = 16); (c) opiate dependence and both major depressive disorder and antisocial personality disorder (ASP) (N = 17); and (d) opiate dependence and ASP (N = 13). Differences between groups in response to psychotherapy were assessed in terms of change for each group (paired t-tests) between baseline and 7 months later (including 6 months of planned exposure to psychotherapy) in terms of 22 outcome criteria. Significant change was obtained for 11 of 22 criteria for the opiate dependence only group, for 9 of 22 criteria for the opiate dependence plus major depressive disorder group, for 11 of 22 criteria for the opiate dependence plus both major depressive disorder and ASP, but for only 3 of 22 criteria for the opiate dependence plus ASP group. Moreover, whereas behavior changes for other client groups embraced several domains (i.e., drug use, psychiatric symptoms, employment, legal status), changes for methadone clients showing ASP only were confined to drug use and legal status. The authors conclude that psychotherapy is an appropriate and effective intervention for a wide range of methadone clients, including those showing ASP as long as those clients also show evidence of depressive disorder; however, psychotherapy is not indicated for clients showing ASP in the absence of Axis I disorder.

Woody, G. E., McLellan, A. T., Luborsky, L., & O’Brien, C. P. (1995). Psychotherapy in community methadone programs: A validation study. American Journal of Psychiatry, 152 (9), 1302–1308.

To replicate the findings from an earlier study of the effectiveness of psychotherapy with methadone clients (see Woody et al., 1983, this volume), 123 methadone clients from three community methadone programs were randomly assigned (using a 2:1 ratio) to receive supportive-expressive psychotherapy plus standard counseling (N = 82) or standard counseling only (N = 41). Unlike the earlier study, clients assigned both to standard counseling and to supportive-expressive psychotherapy saw two treatment providers: in the case of standard counseling, a primary and secondary counselor; in the case of supportive-expressive psychotherapy, a primary counselor and a psychotherapist. Also in contrast to the earlier study, a substantial minority of clients were female (40%), and all clients sampled showed moderate to high levels of psychiatric symptoms. The two treatment groups differed significantly on 2 of 38 baseline comparisons. Clients assigned to the supportive-expressive group (SE) had shorter periods of opiate use and longer periods of voluntary sustained abstinence than those assigned to standard (i.e., drug) counseling (DC). To qualify as study subjects, clients had to attend a minimum of three sessions with therapists and counselors during the first 6 weeks of study. The course of SE and/or DC covered 6 months. Follow-ups were conducted at 1 and 6 months posttreatment (i.e., after the 6 months of therapy and/or counseling). Measures administered at baseline and at the two follow-ups included the Addiction Severity Index (ASI), several psychological tests (Beck Depression Inventory, Maudsley Personality Inventory, Shipley Institute of Living Scale, and Symptom Check List) as well as the Schedule for Affective Disorders and Schizophrenia–Lifetime and monthly urinalyses. There were no differences between SE and DC groups in terms of sessions attended, with both groups attending about one per week. Similarly, at 1 month after receipt of the SE or DC regimen, both groups showed significant and comparable gains in terms of drug and alcohol use, legal status, and employment as well as psychological functioning. At 6-month follow-up, the SE group showed a larger number of significant differences (baseline to follow-up) than the DC group, specifically in terms of selected drug, alcohol, employment, and psychological measures. In addition, the SE group had significantly fewer cocaine positive urines than the DC group over the 6-month period. This study indicates the potential for psychotherapy in methadone programs, but also shows the problems inherent in the "facts of life" of methadone programming, including subtherapeutic methadone dosing, staff turnover, and difficulty in recruiting psychotherapists.

Woody, G. E., McLellan, A. T., Luborsky, L., O'Brien, C. P., Blaine, J., Fox, S., Herman, I., & Beck, A. T. (1984). Severity of psychiatric symptoms as a predictor of benefits from psychotherapy: The Veterans Administration-Penn study. American Journal of Psychiatry, 141, 1172–1177.

In further analysis of a study of the effectiveness of psychotherapy as compared to counseling only with methadone maintenance clients (see Woody et al., 1983, this volume), the maintenance clients were divided into three groups based on psychiatric severity using a composite baseline score derived from the several measures of psychiatric functioning employed (Addiction Severity Index, Symptom Check List-90, Beck Depression Inventory, Maudsley Neuroticism Scale, and Schedule for Affective Disorders and Schizophrenia–Change and Lifetime, the general assessment scale only). Baseline to 7-month outcome data were explored for 110 clients who had been randomly assigned to drug abuse counseling plus supportive expressive therapy (SE [N = 32]), drug abuse counseling plus cognitive behavioral therapy (CB [N = 39]), and drug abuse counseling only (DC [N = 39]). Clients in each group were subdivided to high (psychiatric) severity, mid-severity, and low severity. The numbers in these subgroups varied between 10 and 18. Low-severity clients showed substantial positive behavior change (drug use, employment, criminal activity) regardless of group assignment; mid-severity clients showed positive behavior change across groups, but with DC had scores reflecting greater psychiatric problems than the other groups; high-severity clients assigned to DC had poorer outcomes than therapy groups at 7 months in terms of employment (composite measure), days of opiate use, criminal justice functioning (composite measure), and psychiatric functioning (composite measure and several individual measures). Moreover, DC clients evidencing high severity at baseline showed a lower number of positive changes in behavior and psychiatric functioning from baseline to 7 months than the SE and CB groups, although high-severity clients in all groups were less likely to evidence change than their counterparts showing mid and low severity. The authors conclude that the selective use of psychotherapy can add importantly to the effectiveness of drug abuse counseling and treatment.

About NIDA Contents

Archive Home | Accessibility | Privacy | FOIA (NIH) | Current NIDA Home Page
National Institutes of Health logo_Department of Health and Human Services Logo The National Institute on Drug Abuse (NIDA) is part of the National Institutes of Health (NIH) , a component of the U.S. Department of Health and Human Services. Questions? See our Contact Information. . The U.S. government's official web portal