Effects of program and patient characteristics on retention of drug treatment patients

https://doi.org/10.1016/S0149-7189(01)00027-1Get rights and content

Abstract

Objective: To examine effects of program and patient characteristics on patient retention in residential drug treatment programs, outpatient drug-free programs (ODF), and methadone maintenance (MM) programs.

Data sources/study setting: Patient data were based on admission and discharge records for individuals entering treatment programs in Los Angeles County during 1992 and 1993. Program data were collected from program directors via a mail survey. The study sample included 26,047 patients in 87 programs. The dependent variable was patient completion of a critical threshold of treatment (360 days for MM and 180 days for the other two modalities). We applied logistic regression hierarchical linear modeling analysis for each modality.

Principal findings: Threshold retention rates were generally low in all three modalities (18.1% for residential programs, 22.9% for ODF, and 13.6% for MM). An articulated programmatic focus and low caseload increased patient retention in residential programs. A lower level of group therapy focus increased patient retention in ODF programs. A low programmatic focus and a low percentage of recovering staff were associated with high retention rates among MM patients. For ODF programs, none of the slopes showed random effects, while for residential and MM programs, some program factors contributed to the explanation of the random effects in several slopes (e.g., drug use severity).

Conclusion: Program practice and service provision played important roles in determining patient retention in treatment. Service providers and planners should consider these key factors to improve retention of patients, which is likely to increase overall treatment effectiveness and efficiency.

Introduction

Like the treatment of many chronic diseases or psychological disorders (e.g., diabetes, hypertension, mental illness), treatment for drug dependence is beset with problems of patient non-compliance with intervention protocols (McLellan, Woody, & Metzger, 1996). Many drug abusers prematurely quit treatment, and subsequent relapse to drug use and related negative behaviors are common among dropouts from all types of drug treatment programs (Hubbard et al., 1989, Simpson, 1979, Simpson et al., 1997). For example, the national Drug Abuse Treatment Outcome Study (DATOS) reported that more than 50% of patients in residential and outpatient drug-free (ODF) programs remained in treatment less than 3 months (Simpson et al., 1997). Studies have consistently demonstrated that treatment retention is positively correlated with such post-treatment outcomes as reduced drug use and improved psychosocial functioning (Anglin and Hser, 1990, De Leon, 1990–91, Hubbard et al., 1989, Hubbard et al., 1997, Simpson, 1979). Thus, premature dropout represents a major obstacle to effective treatment and is a major concern in the drug treatment field. It is paramount to identify key factors that influence patient retention in treatment in order to effectively reduce dropout rates. Gaining improved knowledge of these influential factors is a critical step toward improved treatment planning and intervention strategies, which can result in better outcomes at the individual level as well as for the treatment system as a whole. The focus of the present study is to identify program characteristics and patient characteristics that influence patient retention in treatment.

Drug treatment programs vary in terms of modality/setting, service provision, therapeutic approaches, patient populations, and patient outcomes (Hubbard et al., 1989). Factors possibly influencing patients’ length of stay can be broadly discussed in terms of program characteristics and patient characteristics. Past literature on treatment effectiveness has consistently shown positive outcomes associated with drug treatment, but most studies have focused on patient characteristics to account for the variation in treatment outcomes (Anglin & Hser, 1990). Patients who have less extensive drug use and higher levels of social adjustment (e.g., employed, less criminally involved) generally are more likely to have favorable outcomes. Although the effects of program characteristics have seldom been quantitatively evaluated, most researchers and clinical observers believe that substantial differences in outcome can be attributed to program policies and protocols, the quality of therapeutic staff and program management, the breadth of services provided, and general staff and facility-wide morale (Anglin and Hser, 1990, Ball and Ross, 1991, McGlothlin and Anglin, 1981). Because program factors are relatively more amenable to policy influence, the rest of this section focuses on program factors.

The most salient program characteristic that has been addressed in the literature is the program's modality or setting, such as residential care, short-term detoxification, methadone maintenance (MM), and outpatient care. Some of the most intensive residential settings are therapeutic community programs, which typically provide a controlled residential environment for 6–18 months with intensive personality restructuring and socialization efforts. Most detoxification or inpatient programs are delivered within a hospital setting (with the exception of methadone detoxification programs) and have more emphasis on medical care during the typically short period of stay. Outpatient methadone maintenance programs displace heroin dependence by substituting methadone within the context (ideally) of medical, social, and psychological services. Especially in an era of HIV/AIDS concerns, methadone maintenance patients are expected to stay with programs for lengthy periods, perhaps indefinitely. Outpatient drug-free programs, although they vary widely on many dimensions, basically provide 12 months or more of drug education and counseling.

Programs vary in their orientation toward drug dependence and its treatment. Some program philosophies argue that past experience with drug use is necessary for understanding and engaging patients; therefore, there is an emphasis on recruiting individuals ‘in recovery’ as counselors. Programs also vary in their emphasis on the importance of medication and of counseling. Some programs believe in the importance of skills building, and some believe that environmental factors are primary influences on dependence and emphasize the necessity of making changes in the patient's environment. Virtually no research has empirically examined how programs’ philosophies and approaches influence patient outcomes, and studies on the effects of staff qualification are also limited.

Research on treatment effectiveness has increasingly emphasized the importance of specifying the types and amounts of services provided during treatment, and recent studies (Hoffman et al., 1994, McLellan et al., 1993) have demonstrated that the amount of services received by patients exerts considerable impact on patient outcomes. For example, Hoffman et al. (1994) found that neither demographic differences nor dynamic variables, such as levels of patient motivation or recent patterns of drug or alcohol use, were significantly related to patient retention in treatment or levels of treatment exposure. A number of comparisons between the provision of ‘standard treatment services,’ relative to adding ‘enhancements’ to the standard treatment condition, showed that the inclusion of enhancements had a significant impact on treatment retention for cocaine users. Similarly, McLellan et al. (1996) demonstrated that patients who received a broader array and frequency of services stayed in treatment longer and showed 15% better outcomes than patients who did not. Hoffman et al. (1994) concluded that one successful approach to enhance patient retention and treatment exposure is to increase the frequency, intensity, and types of treatment services offered by programs.

Even though drug treatment counselors may be the most important members of the drug treatment staff (Hser, 1995, Winick, 1990–91), scant information is available about their roles and functions in the delivery of treatment services (Onken & Blaine, 1990). Researchers describe counselors as traditionally being in positions of low-skill expectation, low pay, and low prestige (Gustafson, 1991, Onken and Blaine, 1990). Additionally, they have relatively heavy caseloads. These stressors have resulted in a rather high level of burnout and rapid turnover among drug treatment counselors (Millman, 1989). We anticipated that programs with high counselor turnover rates and high levels of caseload per counselor would negatively impact service quality, and thus would have less success in retaining patients in treatment.

Several other program factors may also influence patient retention. For example, use of contingency management strategies has been shown to increase retention and compliance (Higgins et al., 1991, Higgins et al., 1993). Some studies (Maddux, Prihoda, & Desmond, 1994) have shown that elimination of fees significantly increased retention among patients in methadone maintenance programs.

The literature indicates that a more systematic investigation is needed to fully assess the various aspects of programs, patients, and their interaction as influencing factors on patient outcomes. In this article, we address four major research questions: What is the extent of variation in patient retention among programs of the same modality? What program characteristics are associated with better patient retention? What patient characteristics are associated with better patient retention? Finally, we explore potential interaction effects of program and patient characteristics, or cross-level interactions, to address whether the strength of association between a patient characteristic (e.g., drug use severity) and patient retention is similar across programs. For example, is drug use severity a more important predictor of patient retention in some programs than in others?

Section snippets

Data description

Subjects were patients included in the California Alcohol and Drug Data System (CADDS). The CADDS was developed by the California State Department of Alcohol and Drug Programs (ADP) in response to data collection requirements contained in the federal Anti-Drug Abuse Act of 1988. All treatment programs receiving public funding (state or federal) and those which are state-licensed are required to report the number and characteristics of patients to CADDS. Each patient treatment record contains

Patient characteristics

Patient characteristics, by modality, are presented in Table 1. More than two-thirds of patients in both residential and ODF programs were younger than 35-year-old, while about two-thirds of MM patients were older than 35. Gender was about equal in residential programs, whereas about two-thirds of the ODF patients were women and two-thirds of the MM patients were men. African-Americans constituted the largest group in residential programs (46.4%) and in ODF (43.2%), but represented the smallest

Discussion

Keeping patients in treatment for periods of sufficient duration is an important intermediate goal of drug treatment. To achieve this goal, it is important to identify both the main and interaction effects of program and patient characteristics that prolong treatment retention. In addition, it is important to understand how these effects differ within various treatment modalities. The present study applied the hierarchical linear model approach to examine patient retention as influenced by

Acknowledgements

Support for this project was provided by Contract no. U461-37 with Caliber Associates under funding from the Substance Abuse and Mental Health Services Administration (SAMHSA) and by grants P50DA07699 and DA09652 from the National Institute on Drug Abuse (NIDA). Drs Hser, Chou, and Anglin are also supported by Research Scientist Development Awards (DA00139, DA00215, DA00146, respectively) from NIDA. Interpretation of the results and conclusions in this paper, however, do not necessarily

Yih-Ing Hser, Ph.D., is Adjunct Professor in the Department of Psychiatry and Biobehavioral Sciences at UCLA. She received her PhD in Cognitive Psychology from UCLA. Her major research interests include human learning and memory, drug use epidemiology, treatment process and outcome evaluation, health services research, and development and application of statistical methodologies.

References (26)

  • R.M. Etheridge et al.

    The relationship of counseling and self-help participation to patient outcomes in DATOS

    Drug & Alcohol Dependence

    (1999)
  • Y.-I. Hser et al.

    Predicting perceived quality of drug treatment programs using latent variable models

    Evaluation and Programming Planning

    (1998)
  • M.D. Anglin et al.

    Treatment of drug abuse

  • J.C. Ball et al.

    The effectiveness of methadone maintenance treatment: Clients, programs, services, and outcome

    (1991)
  • R.C. Blair et al.

    An investigation of the robustness of the t test to unit of analysis violations

    Educational and Psychological Measurement

    (1983)
  • A.S. Bryk et al.

    Hierarchical linear models: Applications and data analysis methods

    (1992)
  • A.S. Bryk et al.

    HLM: Hierarchical linear and nonlinear modeling with the HLM/2L and HLM/3L programs

    (1996)
  • G. De Leon

    Retention in drug-free therapeutic communities

    NIDA Research Monograph

    (1991 1991)
  • J.S. Gustafson

    Do more… and do it better: Staff-related issues in the drug treatment field that affect the quality and effectiveness of services

  • S.T. Higgins et al.

    A behavioral approach to achieving initial cocaine abstinence

    American Journal of Psychiatry

    (1991)
  • S.T. Higgins et al.

    Achieving cocaine abstinence with a behavioral approach

    American Journal of Psychiatry

    (1993)
  • J.A. Hoffman et al.

    Comparative cocaine abuse treatment strategies: Enhancing patient retention and treatment exposure

    Journal of Addictive Diseases

    (1994)
  • Y.-I. Hser

    Drug treatment counselor practices and effectiveness: An examination of literature and relevant issues in a multilevel framework

    Evaluation Review

    (1995)
  • Cited by (53)

    • Sober living house characteristics: A multilevel analyses of factors associated with improved outcomes

      2019, Journal of Substance Abuse Treatment
      Citation Excerpt :

      Further, because most recovery housing models operate by residents paying out-of-pocket for expenses associated with operating the residence (as in the SLH model), there is often a direct relationship between the number of residents in the house and fees charged per resident. Lack of staff education and training as well as high client-to-staff ratios have been found to be associated with worse outcomes among those in residential substance use treatment settings (Grella & Stein, 2006; Hser, Joshi, Maglione, Chou, & Anglin, 2001), but it is unclear whether the potential benefits of experiential knowledge and affordability may result in different effects in recovery housing settings. Despite growing recognition of its importance to a robust recovery-oriented system of care, increasing evidence of its effectiveness, and renewed efforts to establish best practices, recovery housing is still all too often viewed with skepticism by the lay media, addiction professionals, and substance use treatment researchers (Polcin, Mericle, Callahan, Harvey, & Jason, 2016).

    • Predictive factors for treatment retention in methadone programs in Indonesia

      2012, Journal of Substance Abuse Treatment
      Citation Excerpt :

      Social characteristics include family support (Dobkin et al., 2002; Siddal & Conway, 1988), community support (Brown et al., 2004), and peer support (Booth et al., 2004). Some studies found that interactions between these factors were also significant in predicting treatment success (Ball and Ross, 1991; Chou, Hser, & Anglin, 1998; Hser et al., 2001). Indonesian studies also identify potential predictors of treatment retention.

    View all citing articles on Scopus

    Yih-Ing Hser, Ph.D., is Adjunct Professor in the Department of Psychiatry and Biobehavioral Sciences at UCLA. She received her PhD in Cognitive Psychology from UCLA. Her major research interests include human learning and memory, drug use epidemiology, treatment process and outcome evaluation, health services research, and development and application of statistical methodologies.

    Vandana Joshi, Ph.D. is a researcher at the UCLA Drug Abuse Research Center. Her research interests include drug use epidemiology, substance use among adolescents, cohort analysis, and multilevel modeling.

    Margaret A. Maglione is a project director at RAND with extensive experience in health services research and health policy analysis. She is currently managing a 5-year Healthy Aging Project funded by the Health Care Financing Administration (HCFA). The project involves several meta-analyses on prevention issues as well as designing intervention demonstrations. Ms. Maglione previously spent 5 years at UCLA's Drug Abuse Research Center (DARC), working in HIV prevention, treatment utilization, and program evaluation. She also spent several years in AIDS policy and planning during the early years of the epidemic. Ms. Maglione has a BA in economics from Columbia University and a Masters in Public Policy from the University of Southern California.

    Chih-Ping Chou, Ph.D., is Associate Professor of Research in Department of Preventive Medicine, University of Southern California. His research interests include multivariate statistical analyses and methodologies for prevention research.

    M. Douglas Anglin, Ph.D., is Professor-in-Residence in the Department of Psychiatry and Biobehavioral Sciences at UCLA. He has been conducting research on substance abuse and treatment evaluation since 1972 and is author or co-author of over 150 published articles. He has been project director or principal investigator on over 30 federally funded studies. He is currently Associate Director of the UCLA Integrated Substance Abuse Programs.

    View full text