Highlights

Topics include:
Introduction
Summary of Sampling Design
Overview of 1-Year Follow-up
Treatment Retention and Follow-up Outcomes
Treatment Engagement and Process
Treatment History and Outcomes
Treatment of Cocaine Dependence
Cost and Benefits of Cocaine Treatment

Introduction

Results from a series of core studies published in the  Psychology of Addictive Behaviors (Simpson & Curry, eds., 1997), Archives of General Psychiatry (Simpson et al, 1999), and Drug and Alcohol Dependence (Simpson & Brown, eds., 1999) describe some of the highlights and major findings from this project.

Return to top

Summary of Sampling Design 

A total of 96 treatment programs in 11 large U.S. cities were chosen to reflect typical community-based treatment services available to the public (see Overview of Sampling Plan). Geographic location and type of program as well as the type of clients they served were considered in designing the sampling plan. Participating programs were located in Chicago, Houston, Miami, Minneapolis, Newark, New Orleans, New York, Phoenix, Pittsburgh, Portland, and San Jose. The overall methodology used to compile the database used in DATOS studies, including client samples and program characteristics, used to meet research objectives is described by Flynn, Craddock, Hubbard, Anderson, and Etheridge (1997).

• Intake Sample: A total of 10,010 clients entered the 96 treatment programs that participated in DATOS during 1991-1993. Overall, this treatment sample was 66% male; 47% African American; and 13% Hispanic, with a mean age of 33 years. However, these and other client characteristics varied across modalities, reflecting their different therapeutic and operational characteristics.
   
• Follow-up Sample: 4,229 of the eligible clients who completed the two-stage intake interviews were selected for follow-up (using a stratified random design). Altogether, 74% (n=3,147) were located, including 70% (n=2,966) who were successfully interviewed, 1.5% (n=64) who were deceased, and 2.7% (n=117) who refused to participate. Gender, ethnicity, and average age were not significantly different between the intake and follow-up samples.

Data Collection

• During-Treatment Interviews were completed at 1 month after admission, and again at 3 and 6 months. Questions were included about service delivery and client satisfaction.
   
• Follow-up Interviews were administered at 12-months after treatment termination (although for long-term methadone clients treated for more than 12 months in DATOS, the follow-up interview had to be scheduled at 24 months after admission). It replicated much of the Intake and focused on key behaviors during the year before the follow-up interview. (An additional follow-up is planned for approximately 48 months posttreatment.)
   
• Treatment Process Questionnaires were sent to counselors and program directors at 75 selected programs approximately 8 months after during-treatment data were collected. Questions were asked about program and treatment structure, philosophy, available services, policy, staffing, treatment planning, and aftercare. Program directors were asked about program organization and financing. Overall, 71 questionnaires (95%) were completed.

Procedures

• Specially trained and supervised interviewers were hired to carry out Intake and During-Treatment data collection in order to assure that programs were not inconvenienced by the research. Trained professional survey interviewers conducted follow-up surveys in the community. Quality control procedures were in place for each step of data management, data editing and entry, and document control.
   
• Clients were compensated $10 for each Intake and During-Treatment interview, and $15 for a Follow-up interview.
  

Treatment Modalities Represented

• Outpatient Methadone Treatment (OMT) programs administered the medication methadone to reduce cravings for heroin, in addition to providing counseling and case management services. Some provided long-term methadone maintenance for clients and others used methadone to taper to abstinence, but all programs had expected stays of 2 years or longer. Private for-profit methadone clinics, nonprofit community-based programs, hospital-based outpatient clinics, and county-managed programs were represented. There were 29 OMT programs with 1,540 clients in the DATOS sample.
   
• Long-Term Residential (LTR) programs offered drug-free treatment in a residential setting, with planned stays ranging from 4 months to 2 years. LTR programs in DATOS included traditional therapeutic communities, modified therapeutic communities, and other programs requiring in-residence treatment; most expected clients to stay at least 9 months. There were 21 LTR programs with 2,774 clients in the DATOS sample.
   
• Outpatient Drug-Free (ODF) programs are characterized by a wide range of therapeutic approaches such as cognitive-behavioral, insight-oriented, supportive, and 12-step. All planned for clients to stay in treatment for at least 3 months, and most expected 6 months or longer. Therapeutic community managed outpatient programs, nonprofit community programs, mental health and short-term managed programs, and private for-profit programs were included. There were 32 ODF programs with 2,574 clients in the DATOS sample.
   
• Short-Term Inpatient (STI) programs generally kept clients in-residence for up to 30 days, with a focus on medical stabilization, abstinence, and lifestyle changes. They included free-standing nonprofit and for-profit short-term programs, public and nonprofit hospital programs, and county-managed programs. Due to changes in insurance coverage and a national trend toward "managed care" during the time this project was being conducted, however, planned duration of treatment became shorter over time. Most of the STI programs studied have now closed or been converted to other types of facilities. There were STI 14 programs with 3,122 clients in the DATOS sample.

Return to top

Overview of 1-Year Follow-up

Clients treated in all four modalities studied in DATOS – that is, OMT, LTR, ODF, and STI – showed large and significant improvements during the 1-year follow-up (N= 2,966; Hubbard, Craddock, Flynn, Anderson, & Etheridge, 1997). Overall, major outcome indicators for drug use, illegal activities, and psychological distress were each reduced on average by about 50%. However, there were notable distinctions between clients admitted to different types of treatment (and there were further variations even between programs of the same general type) as well as in the length of time they remained in treatment. For these reasons, the general findings summarized below are useful for addressing overall questions about "treatment effectiveness," but they do not indicate who benefits most from which treatment, and why. These are among the special topics of other studies now in progress.

Outpatient Methadone Treatment (OMT)

• Admissions to OMT were 60% male, 52% African American or Hispanic, and 82% were over 30 years of age. In addition, 67% had graduated from high school (or had a GED), 40% were married or living as married, 3% were referred to treatment by the criminal justice system, and 10% had private health insurance. Previous drug treatments were reported by 77%; of these, 92% had accumulated more than 3 months in treatment.
   
• Principal indicators of problems in pretreatment functioning (affecting more than 1 of 5 OMT admissions) were weekly heroin use (89%), weekly cocaine use (42%), no full-time work (85%), and illegal activity (29%).
   
• Follow-up outcomes (see OMT Outcomes Chart) showed a 69% drop in the number of weekly heroin users and a 48% decline in weekly cocaine users.
   
• Unemployment did not change significantly, but illegal activity declined 52%. Further tabulations showed a drop from 63% to 21% in those jailed in the year before versus after treatment.
   
• Further treatment during follow-up was reported by 74% of the OMT sample, suggesting the need for a longer follow-up period in order to describe posttreatment outcomes for this sample of OMT clients.

• Admissions to STI were 67% male, 61% African American or Hispanic, and 64% were over 30 years of age. In addition, 72% had graduated from high school (or had a GED), 38% were married or living as married, 5% were referred to treatment by the criminal justice system, and 38% had private health insurance. Previous drug treatments were reported by 47%; of these, only 9% had accumulated more than 3 months in treatment.
   
• Compared to LTR and ODF treatments, this specialized treatment modality admitted clients who were 3 to 10 times more likely to have private health insurance, more highly educated and employed, and 7 to 8 times less likely to be a criminal justice referral. STI admissions also had only a small fraction of the treatment exposures reported by LTR and ODF clients. Finally, STI programs as represented in DATOS have largely disappeared as a form of treatment available to the public, and cautions are recommended against making simple (i.e., unadjusted) comparisons of outcomes involving this modality.
   
• Principal indicators of problems in pretreatment functioning (affecting more than 1 of 5 STI admissions) were weekly cocaine use (67%), weekly marijuana use (30%), heavy alcohol use (48%), no full-time work (67%), suicidal ideation (31%), and illegal activity (26%).
   
• Follow-up outcomes (see STI Outcomes Chart) showed a 69% drop in the number of weekly cocaine users, a 63% reduction in the number of weekly marijuana users, and a 58% decline in heavy drinkers.
   
• Unemployment showed no significant change, but suicidal ideation fell by 48% and illegal activity declined 58%. Further tabulations showed a drop from 49% to 20% in being jailed in the year before versus after treatment, and those with any arrests decreased from 26% to 20%.
   
• Further treatment during follow-up was reported by 25% of the STI sample.

Return to top

Treatment Engagement and Process

A "treatment process model" was developed in previous work by Simpson, Joe, Rowan-Szal, and Greener (1997) to represent essential elements of treatment readiness and engagement indicators as predictors of retention and outcomes. Joe, Simpson, and Broome (1999) tested this model using different therapeutic settings represented in DATOS, including long-term residential (LTR; n = 1,362), outpatient drug-free (ODF; n = 866), and outpatient methadone (OMT; n = 981) treatments. Findings supported the model.

• Motivated clients developed better relationships with their counselors and stayed in treatment longer.
   
• Clients who attended more counseling sessions and discussed a broader range of topics in sessions stayed longer in ODF and LTR.
   
• Clients with more severe background problems (like hostility or cocaine use) had difficulty developing a working relationship with their counselors, attended fewer sessions, and discussed fewer topics.

Reference:

Simpson, D. D. Joe, G. W., Rowan-Szal, G. A. & Greener, J. M. (1997). Drug abuse treatment process components that improve treatment. Journal of Substance Abuse Treatment, 14, 565-572.  [Abstract]

• Clients with higher motivation at admission developed more confidence and commitment to treatment, as did clients who had better relationships with counseling staff and who attended more counseling sessions.
   
• Programs where clients had greater overall levels of confidence and commitment used more social and health services, maintained consistent attendance at counseling sessions, and served clients with more similar kinds of needs.

Return to top

Treatment History and Outcomes

There was considerable diversity in patterns of drug use and in treatment histories for the DATOS admission sample (Anglin, Hser, & Grella, 1997).

• For about half of the clients, DATOS was their first treatment experience; the other half averaged about 3.5 prior episodes of treatment.
   
• Individuals in STI and ODF were least likely to report prior treatment (about 50%), while those in LTR and OMT were 10-25% more likely to have previous treatment experience.
   
• Across all modalities, the average age at first treatment admission was 30 years, and the average interval between initiation of regular drug use and first treatment was 7 years.
   
• Admissions to OMT reported longer addiction and treatment careers, while clients in STI and ODF reported shorter and less severe histories.
   
• ODF admitted the most eclectic mix of clients, including a higher proportion of individuals who were not dependent on cocaine, heroin, or alcohol.
   
• Cocaine- and alcohol-dependent individuals were more likely than heroin users to have been treated in STI, perhaps reflecting the growth in short-term chemical dependency treatment in response to the cocaine epidemic of the 1980s and the historical use of this modality for alcohol treatment.
   
• Cocaine- and heroin-dependent individuals who had no prior treatment experience were more likely than those with prior treatment history to enter STI.
   
• Higher levels of prior treatment were associated with more severe addiction career characteristics, injection drug use, and criminal activities at treatment admission.
   
• Treatment approaches should focus on strategic interventions that recognize and address the diversity of client treatment histories in order to maximize effectiveness.

Hser, Grella, Hsieh, Anglin, & Brown (1999) contrasted cocaine-abusing clients who were in treatment for the first time (n=406) and who had extensive histories of prior treatment (n=383) in an effort to identify factors associated with better outcomes in each group. Treatment history was defined as the number of treatment episodes, total length of time in previous treatment, and the number of years between the client’s first and the current DATOS treatment episode.

• Treatment-experienced clients had more severe drug problems, greater recognition of their drug problems, greater perceived needs for services upon admission to treatment in DATOS, and generally poorer treatment outcomes as compared with first-time clients.
   
• Early engagement in DATOS treatment was associated with higher levels of posttreatment abstinence, regardless of prior treatment history.
   
• Across modalities, treatment-experienced clients received fewer individual counseling sessions while in treatment, were less likely to comply with program rules, and had more unmet service needs. However, they were more likely to be abstinent following treatment if they received more individual counseling sessions and complied with program rules (in outpatient drug-free treatment) and if they had higher levels of rapport with their counselors (in outpatient methadone treatment).
   
• These findings suggest the importance of considering treatment processes and aftercare in developing and implementing strategies to improve treatment outcomes for clients at differing stages of their treatment careers.

Age differences were examined by Grella, Hser, Joshi, and Anglin (1999) as a moderator of the relationships between client characteristics, treatment retention, and treatment outcomes in DATOS. Separate structural equation models were tested for 551 clients from 19 long-term residential (LTR) programs and 571 clients from 27 outpatient drug-free (ODF) programs. Younger adults (less than 30 years of age) comprised 51% of LTR subjects, and 39% of ODF subjects.

• Longer retention in treatment and higher self-efficacy to resist drug use had a positive effect on abstinence for both groups, however, the relationship between treatment retention and abstinence at follow-up was stronger for younger adults in both modalities.
   
• Both age groups reduced their contact with other drug users following treatment, but the influence of drug-using peers was more strongly related to lowered feelings of self-efficacy to resist drug use among younger adults in LTR and among older adults in ODF.
   
• Older adults in LTR and ODF and younger adults in LTR who spent more time in DATOS treatment had stronger feelings that they would be able to resist drug use following treatment.
   
• Older adults in LTR who had longer prior treatment histories had lower levels of self-efficacy to resist drug abuse following DATOS treatment.
   
• The findings suggest that age-specific treatment protocols need to be implemented to address lowered self-efficacy among older adults with longer treatment histories, the influence of negative reference groups, and to increase treatment retention, particularly for younger adults.

Return to top

Treatment of Cocaine Dependence

Cocaine use is the most common drug problem of patients entering treatment for illicit drug use.  In a national sample from 55 treatment programs, problem severity of patients at admission was found to be directly related to cocaine relapse in the year following discharge, and treatment retention also was a significant predictor among moderate-to-high problem groups.  Among the highest severity patients, 90 days or longer in residential programs was needed to improve outcomes.  Findings suggest patient assessments should play a central role in the selection of appropriate settings and duration of treatment to maximize outcomes. 

One-year follow-up interviews with a national sample of 1605 patients treated for cocaine dependence in 55 programs showed (again) that longer treatment stays are related to better outcomes. Overall, 1 of 4 (24%) reported relapses to weekly cocaine use and another 18% obtained further treatment in the year after discharge in DATOS due to continuing problems. The sample included 542 from 19 long-term residential (LTR) programs, 548 from 24 outpatient drug-free (ODF) programs, and 605 from 12 short-term inpatient (STI) treatment programs.

• Although all patients in the study met clinical criteria for cocaine dependency, not all were daily or heavy users in the year before treatment. In LTR, 49% used cocaine daily before treatment, dropping to 12% in the year afterwards. In addition, heavy drinking (3 or more days per week) dropped from 42% to 16%, and incarcerations in jail dropped from 79% to 35% in the year before versus after treatment.
   
• In ODF, 28% used cocaine daily before treatment, dropping to 9% in the year afterwards. Heavy drinking (3 or more days per week) dropped from 34% to 16%, and incarcerations in jail dropped from 73% to 28%.
   
• In STI, 45% used cocaine daily before treatment, dropping to 8% in the year afterwards. Heavy drinking (3 or more days per week) dropped from 45% to 15%, and incarcerations in jail dropped from 53% to 20%. (Only 33% of the STI group were unemployed on a full-time job in the year prior to admission, compared to 59% in LTR and 51% in ODF.)

However, comparisons of relapse to weekly cocaine use rates between different types of client and treatment programs were the main focus of the study. Problem severity at intake (PSI, defined by 7 scores on drug history and psychosocial indicators) was assessed and found to differ across types of treatment programs; on average, LTR treated the most severe cases and STI the least problematic cases. PSI scores were predictive of relapse to weekly cocaine use after treatment, but different programs were not all equally effective. While low-severity patients did about equally well regardless of the type of treatment they received or how long they stayed, outcomes for medium-to-high problem patients improved significantly if they were treated for at least 3 months.

• Problem indicators included multiple drug use (27% of the total sample), alcohol dependence (51), criminal activities (60%), unemployment at a full-time job (47%), low social support from family and friends (56%), depression or anxiety (66%), and no insurance (76%).

• Low-level severity (0 to 3 problems) described 26% of admissions to LTR, 43% of ODF, and 64% of STI; on the other hand, 24% of LTR, 11% of ODF, and 6% of STI were high-severity patients (with 6-7 problems).
   
• Overall, 20% of low-severity patients reported relapse to weekly cocaine use, regardless of whether they had short-term or long-term stays in treatment. ("Long-term" was defined as at least 90 days for LTR or ODF, and at least 21 days for STI.)
  
  
• Among patients with medium-to-high problem severity at intake (i.e., 4 or more problems), 40% of those with short-term treatment stays relapsed to weekly cocaine use, compared to 28% of those with long-term stays.

  
For the most severe cases (with 6-7 problems), however, treatment setting became highly significant. In particular, we found that these patients needed at least a "minimum dose" of more intensive services in LTR (usually therapeutic community) programs. After staying in one of these programs for at least 3 months, 15% relapsed in the 1-year follow-up. By comparison, relapse rates were 29% for similar patients treated in ODF settings for 3 months or longer and 38% of those treated for prescribed stays of at least 21 days in STI programs.

The results of this study, of course, lead to many other questions about the particular treatment services involved, the therapeutic engagement process, influences of treatment history and social context, and cost benefits. These are some of the topics addressed in the next wave of DATOS articles contained in special issue of Drug and Alcohol Dependence (Simpson & Brown, eds., 1999). A report on the first national study of treatment outcomes in England is also included, which replicates many of the DATOS findings (see the NTORS Web site at www.ntors.org.uk for more details.)

References:

Simpson, D., Joe, G., Fletcher, B., Hubbard, R., & Anglin, D. (1999). A national evaluation of treatment outcomes for cocaine dependence. Archives of General Psychiatry, 56, 507-514.   [Abstract]

Simpson, D. & Brown, B. (Eds). (1999). Special issue on treatment process and outcome studies from DATOS. Drug and Alcohol Dependence.  [Summary]

Return to top

Cost and Benefits of Cocaine Treatment

The average net economic benefit from an episode of LTR treatment was $10,344. Net benefits from an episode of ODF treatment ($795) were of a lesser magnitude. However, the magnitude of return on each dollar of investment in ODF was similar to LTR. Because of the considerably lower crime costs of ODF clients before treatment, and the lower cost associated with ODF treatment due to its lesser intensity than LTR which is appropriate for its client population, ODF returns on treatment investments also provide evidence of the value of public investments in ODF treatment. The sample was selected from programs for which financial data were available and is a subset of the client sample described by Simpson et al. (1999). There were 300 clients from 10 LTR programs and 202 from 9 ODF programs.

• All subjects met clinical criteria for cocaine dependence or were daily cocaine users in the year before treatment. The majority of subjects were African American (62%), male (66%), and about 32 years of age.

Tangible costs of 9 crimes published in the literature (Rajkumar & French, 1997) and reported by clients before, during, and after treatment were used to estimate the cost of crime to society, net benefits, and cost-benefit ratios. These tangible costs included crime victim costs, criminal justice system costs, and crime career costs. Detailed explanations of how these costs were calculated are described by Rajkumar and French.

• The highest crime group was LTR with per client crime costs of $20,743 before and $4,605 after treatment. These cost reductions from before to after treatment were 78%.
   
• ODF client crime costs were $3,494 before and $2,503 after treatment representing a reduction of 28%.
   
• Daily treatment costs were $72 in LTR and $9 in ODF.

Simpson, D. & Brown, B. (Eds). (1999). Special issue on treatment process and outcome studies from DATOS. Drug and Alcohol Dependence.  [Summary]

Simpson, D., Joe, G., Fletcher, B., Hubbard, R., & Anglin, D. (1999). A national evaluation of treatment outcomes for cocaine dependence. Archives of General Psychiatry, 56, 507-514.  [Abstract]

Return to top

Site Contents

Research Centers
NIDA | NDRI | TCU | UCLA

DATOS Home | About DATOS | Background |
Highlights | Special Topics |
Publications | Web Posters | What's New