Chapter 1: Health Services Utilization by Individuals with Substance Use and Mental Disorders

Health Services Research

The Institute of Medicine (IOM, 1995) defined health services research as "a multidisciplinary field of inquiry, both basic and applied, that examines the use, costs, quality, accessibility, delivery, organization, financing, and outcomes of health care services to increase knowledge and understanding of the structure, processes, and effects of health services for individuals and populations" (pp. 3 and 17). In early 2000, K. N. Lohr and D. M. Steinwachs co-chaired an ad hoc committee to develop the following revised definition for the Association for Health Services Research (AHSR), now the Academy for Health Services Research and Health Policy:

Health services research is the multidisciplinary field of scientific investigation that studies how social factors, financing systems, organizational structures and processes, health technologies, and personal behaviors affect access to health care, the quality and cost of health care, and ultimately our health and well-being. Its research domains are individuals, families, organizations, institutions, communities, and populations. (Lohr & Steinwachs, 2002, p. 16)

Health services research strives to identify the most effective ways to organize, manage, finance, and deliver high quality care; reduce medical errors; and improve patient and/or client safety.¹ A variety of disciplines study health services research issues. They include medicine, biostatistics, economics, epidemiology, management, law, nursing, pharmacy, psychology, and sociology. Health services research involves integrating epidemiologic, sociologic, economic, and other analytical sciences so that the relationships between need, demand, availability, utilization, and outcome of health services may be better understood. Its ultimate research goal is maximizing the efficiency and efficacy of health services.

In general health care, health services research has been used extensively to study health care systems and to define best practices. It has provided evidence of both what works and what does not. Health care utilization has been studied widely to determine the factors that influence treatment-seeking behaviors. The use of information gained from this research has saved billions of dollars and improved health care for many Americans.

Health services research had its beginnings in the 17^th century as physicians sought to understand variations in hospital mortality rates (McCarthy & White, 2000). Until the 20^th century, studies were limited in size and scope. In 1898, the American Medical Association (AMA) Committee for Scientific Research was established to provide grants for fostering medical research (AMA, 2004). The National Health Survey conducted in 1935–1936 and other studies underscored the disparities in health status and access to medical care associated with income (IOM, 1992). Government programs developed during the Great Depression shifted responsibility for many social programs to the Federal Government. During the 1940s and 1950s, as the number of physicians increased and specialization of medical practice also increased, health services research became more important. In the 1960s and 1970s, health services research became increasingly institutionalized with the development of a National Institutes of Health (NIH) Health Services Research Study Section and the journal Health Services Research (McCarthy & White, 2000). With the passage of Medicare legislation in 1965 and concurrent regional medical programs, there was a need for oversight and better understanding of how health services utilization would be affected. In 1966, the Comprehensive Health Planning Act was passed by Congress, and Federal funding for health services research began.

It should be noted that prior to the development of biostatistical methods and consistent data collection techniques, health services research could be done on only a small scale. The development of the fields of biostatistics and epidemiology, coupled with rapid improvements in computer hardware, software technologies, and data collection methodologies during the 1970s, enabled more rigorous study of such issues as access, selection of providers, service delivery, financing, and outcomes. In 1973, legislation enabling the creation of the National Center for Health Statistics (NCHS) was enacted. During the 1970s and 1980s, health economics developed, and there was increased awareness that certain patterns of medical practice were associated with better health outcomes.

From 1971 to 1982, the RAND Health Insurance Experiment was undertaken (Newhouse, 1999; Newhouse & Insurance Experiment Group, 1994). The conclusion from this study was that while the use of co-payments reduced the use of health services, there was no apparent change in the health status of most families (Freund, 1994; Normand, 1994). The "sick poor" (i.e., persons with low incomes who were ill in particular ways at the start of the experiment) were the exception. For them, access to "free" care helped in regard to those conditions. These findings had a great impact on health care reform in the United States.

Over the past three decades, large health care studies have been sponsored by the Federal Government and also by foundations such as the Henry J. Kaiser Family Foundation and the Robert Wood Johnson Foundation. Many Federal agencies, such as the Centers for Disease Control and Prevention (CDC) and the NIH maintain large ongoing studies of the Nation's health, and longitudinal datasets are now available to enable the ongoing study of various health conditions, as well as special population groups.

Health Services Research in the Behavioral Health Sciences

The great societal costs of substance use and mental illnesses, coupled with large amounts of Federal, State, and local resources devoted to the amelioration of these problems, resulted in the use of health services research methodologies to seek effective models of short- and long-term care for people with acute and chronic substance use and mental disorders. The following describes the data collection efforts and evaluations that focused on mental health services and substance abuse treatment.

Federal involvement with mental health research began in 1855 with the creation in Washington, DC, of Saint Elizabeth's Hospital, a facility for persons with mental illnesses (National Library of Medicine, 2001). A Division of Mental Hygiene was established in the Public Health Service (PHS) in 1930 and became in 1949 the National Institute of Mental Health (NIMH) (National Library of Medicine, 1998). After passage of legislation in the 1970s that provided for community-based treatment, there was increased interest in demonstrating the benefits of these large expenditures, and a call for research to show the most efficacious treatments. Thus began a large-scale research agenda that evolved over the next three decades and focused on understanding substance abuse treatment and mental health services, including the need for treatment, service utilization, outcomes, and financing. Several national evaluations were undertaken, and important national databases were established. They are briefly described in the following sections.

DARP, TOPS, and DATOS

One of the first behavioral health services research studies was the Drug Abuse Reporting Program (DARP). Designed to evaluate the emerging community-based substance abuse treatment programs, as well as the use of methadone to treat addiction, DARP began in 1968 and was planned as a client reporting system to establish a research database for treatment programs funded by the National Institute on Drug Abuse (NIDA) (Sells, 1974). DARP measured treatment outcomes on 44,000 clients admitted to 52 treatment programs from 1969 to 1973. Methadone maintenance, therapeutic community, outpatient drug-free, and outpatient detoxification programs were studied. Findings from DARP demonstrated the effectiveness of community-based treatment in reducing substance use and criminal behaviors (Hubbard et al., 1989). The study also provided useful data on the natural history of heroin use in a treated population and evidence that addicted clients treated with methadone had better outcomes if they remained in treatment for longer than 90 days (Joe, Simpson, & Hubbard, 1991; Simpson & Savage, 1980).

The Treatment Outcome Prospective Study (TOPS) was designed to expand on DARP and to provide longitudinal data on clients to allow the assessment of short- and long-term treatment outcomes and to obtain more data on client attributes, program environments, and services delivered in treatment (Allison, Hubbard, Craddock, & Rachal, 1982; Ginzburg, 1978). Outpatient methadone, short- and long-term residential, and outpatient drug-free programs in operation from 1979 to 1981 were studied.

As in earlier studies, TOPS findings suggested that treatment was cost-effective and cost-beneficial in reducing the use of heroin and other illicit drugs (Harwood, Hubbard, Collins, & Rachal, 1988). Levels of predatory crime declined during treatment and remained lower than at baseline (Harwood et al., 1988; Hubbard et al., 1989). TOPS data showed that clients with substance use disorders have a great need for mental health services (Allison et al., 1982). TOPS data also indicated that clients with more ancillary services, particularly mental health, employment, and general services, had improved outcomes (Joe et al., 1991).

The Drug Abuse Treatment Outcome Study (DATOS) was a prospective cohort study designed to evaluate treatment effectiveness in typical, stable, community-based substance abuse treatment organizations operating from 1991 to 1993. DATOS obtained data on approximately 20,000 clients in four types of treatment programs: outpatient methadone, long-term residential, outpatient drug-free, and short-term residential (Tims, Fletcher, & Hubbard, 1991). Adult clients were interviewed at admission, during treatment, and at 12 months after the termination of treatment (except for clients receiving methadone long term who were interviewed approximately 24 months after admission if they were still in treatment during the follow-up phase of the study). DATOS used a standard set of instruments to provide diagnostic profiles of clients (Fletcher, Tims, & Brown, 1997; Flynn, Craddock, Hubbard, Anderson, & Etheridge, 1997).

Findings from DATOS indicated that in each type of treatment, clients lowered their drug use from pretreatment baselines at 12 months after treatment. DATOS also was able to document the marked decrease in services provided to clients in substance abuse treatment accompanied by an increase in unmet service need in the decade since clients entered the TOPS programs. There was a shift from more expensive targeted services to the provision of core services. However, DATOS found that substance abuse treatment programs appeared to have improved how they delivered services of counseling, treatment planning, and use of aftercare (Hubbard, Craddock, Flynn, Anderson, & Etheridge, 1997; Simpson, Joe, & Brown, 1997). DATOS was particularly important in that it was able to document the change in substance use patterns and examine outcomes for community-based cocaine abuse treatment. Moreover, it was the first national study conducted after the start of the AIDS epidemic and also after the shift to Substance Abuse Prevention and Treatment (SAPT) block grant funding.

CODAP and TEDS

The Client Oriented Data Acquisition Process (CODAP) began in 1973 and initially was developed to satisfy requirements outlined in the Drug Abuse Office and Treatment Act of 1972. At that time, 13 separate Federal agencies were involved in the provision of substance abuse treatment, and the resultant duplication of effort and conflicts arising from the situation were best resolved by the creation of a single, ongoing management reporting system, agreed upon by all Federal agencies (Blanken, 1989). This system was CODAP.

Designed to monitor drug treatment need and use, CODAP provided current information describing clients and the treatment given to them in order to aid in planning, management, and evaluation activities (NIDA, 1982). Programs for treatment of alcohol abuse were not included. All program recipients of Federal funds for drug abuse treatment and rehabilitation services were required to participate in CODAP. Data collection began in 1973. Admission reports for approximately a quarter of a million clients were submitted annually by 1,800 to 2,000 participating treatment programs (Office of Applied Studies [OAS], 2004). CODAP included no client, staff, or program surveys, and it depended on treatment units to provide data. Discharge data provided parallel information, as well as the reason for discharge. Although CODAP was not able to produce client outcome data, data available pertaining to client characteristics, geographic distribution, and drug use patterns and trends over time also greatly increased national knowledge about the epidemiology of drug use and the provision of treatment services.

The Alcohol, Drug Abuse, and Mental Health Services (ADMS) block grant program, implemented in 1981, transferred Federal funding from individual programs to the States for distribution, and required no data reporting (OAS, 2004). Thus, CODAP was no longer a required reporting process.

In 1988, the Comprehensive Alcohol Abuse, Drug Abuse, and Mental Health Amendments established a revised SAPT block grant and mandated Federal data collection on clients receiving treatment for either alcohol or drug use disorders. Work on the Treatment Episode Data Set (TEDS) began in 1989 with the issuance of 3–year development grants to States. TEDS includes admissions to facilities that are licensed or certified by State substance abuse agencies to provide substance abuse treatment (or are administratively tracked for other reasons) (OAS, 2004).

DSRS, SROS, and ADSS

In 1990, the Office of National Drug Control Policy (ONDCP) requested that NIDA conduct a nationally representative study of the treatment system. The Drug Services Research Survey (DSRS) was designed to obtain information on drug abuse treatment providers and client characteristics to supplement information from the 1989 National Drug and Alcoholism Treatment Unit Survey (NDATUS).

The DSRS had two phases. In Phase I, a representative probability sample of 1,183 drug treatment facilities was drawn from a comprehensive list of organized substance abuse treatment facilities (the 1989 NDATUS list). During Phase II, a representative subsample, stratified by facility type, of 120 drug treatment facilities was selected. The DSRS provided a picture of treatment participation and client characteristics, including demographics, prior treatment history, drug use history, and discharge status of clients at each sampled treatment facility (Batten et al., 1992, 1993).

The Services Research Outcomes Study (SROS) was sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA) and was a 5–year postdischarge follow-up of a broadly representative sample of treatment facilities and clients based on the DSRS facility sample. Specifically, 1,799 clients from a sample of 99 drug treatment facilities were interviewed. The study compared client behavior in the 5 years before treatment with behavior 5 years after treatment with respect to drug and alcohol use, criminal involvement, employment, physical and mental health, and other behaviors. SROS analyzed treatment results in light of client characteristics and the type and cost of treatment services the clients received. It provided an examination of multiple treatment episodes before and after treatment in a 1990 population. SROS confirmed that both substance use and criminal behavior were reduced following inpatient, outpatient, and residential treatment for drug use disorders (OAS, 1998).

The Alcohol and Drug Services Study (ADSS) built on the work of the 1990 DSRS study with a more complex sampling frame, an enhanced sampling design, and improved measures of financing and organization. Sponsored by OAS, ADSS was designed to collect information on the characteristics of substance abuse treatment facilities and their clients and to study the relationships among facility characteristics, treatment services, and clients in treatment. OAS was interested in developing better estimates of client length of stay and the costs of treatment and to describe the posttreatment status of clients (OAS, 2003b).

The ADSS sample was selected using a multistaged, stratified design, with selection of 2,393 facilities in Phase I, a selection of a subset of Phase I responding facilities, and a selection of client discharge records in Phase II and client follow-up in Phase III. Facilities in the sampling frame were stratified by treatment type: hospital inpatient, nonhospital residential, outpatient predominantly alcohol, outpatient predominantly methadone, other outpatient, and combined treatment types (OAS, 2003b).

The ADSS cost study examined treatment costs with validated cost data from a nationally representative sample of substance abuse treatment facilities and applied its costing methodology to a representative sample of facilities. Its findings supported other studies' findings that nonmethadone outpatient treatment is substantially less expensive than residential treatment (OAS, 2003a). It also found that the cost per enrolled client day in outpatient methadone treatment was only marginally higher than the cost per enrolled client day in nonmethadone outpatient treatment. One of the most important conclusions from the ADSS cost study was the variability in unit costs within a type of treatment.

NTIES

The National Treatment Improvement Evaluation Study (NTIES) was a congressionally mandated 5–year longitudinal study of the impact of drug and alcohol treatment on clients in treatment units that received public support in 1990–1991 from SAMHSA through its Center for Substance Abuse Treatment (CSAT) demonstration grant-funding program. All recipient facilities were recruited to provide facility-level information on services, staffing, and costs. Ten percent of the facilities were purposively selected for inclusion in a client follow-up study that covered the 1993–1994 admission cohort and included a 1–year follow-up study. Client-level data were obtained for over 5,000 clients at treatment intake, at treatment exit, and 12 months after treatment exit. The study was designed to address two issues: (a) the amount of treatment required to achieve successful outcomes; and (b) the extent to which favorable outcomes persist following termination from treatment.

Findings indicated better treatment outcomes for clients in outpatient methadone treatment with longer stays in treatment who were still in the program at follow-up (Koenig, Denmead, Nguyen, Harrison, & Harwood, 1999). NTIES provided support for expansion of methadone treatment, expansion of aftercare services for clients treated with methadone, and expansion of ancillary services, such as transportation and day care to support retention in treatment. Practice implications suggested the need to develop aftercare plans prior to discharge and to provide information and referral services to help discharged clients locate alternative treatment services. NTIES underscored the need to study factors that result in increased client retention and better treatment.

NCS

The National Comorbidity Survey (NCS) was a collaborative epidemiological investigation designed to study the prevalence and correlates of disorders defined and described in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R) (American Psychiatric Association [APA], 1987), as well as the patterns and correlates of service utilization for these disorders. The NCS was the first survey to administer a structured psychiatric interview to a nationally representative sample. The survey was carried out in the early 1990s with a household sample of more than 8,000 respondents aged 15 to 54 years. Baseline NCS respondents were reinterviewed in 2001–2002 (NCS-2) to study the patterns and predictors of the course of mental and substance use disorders and to evaluate the effects of primary mental disorders in predicting the onset and course of secondary substance use disorders (Harvard Medical School, Office of Public Affairs, 2003; Kessler et al., 2003).

The study showed that as many as half of the U.S. population met criteria for a mental disorder at some time in their lives and that, in any given year, such mental disorders were highly concentrated in a relatively small portion of the population (5 to 8 percent) (Kessler et al., 1997). In addition, findings from the NCS indicated that 6 percent of women and 9 percent of men met criteria for alcohol dependence and that 23 percent of women and 26 percent of men met criteria for tobacco dependence (Kandel, Warner, & Kessler, 1998).

In summary, changes in the patterns of substance use, the nature of the treatment population, and the treatment system have added new challenges to behavioral health services research. The evaluations and databases developed over the past three decades provide a good foundation to study the nature and impact of these changes. Most were developed to answer specific research and practice questions relevant at the time, and each study was costly and required long periods of time for planning, pilot testing, data collection, and data analysis. Although study methodologies improved greatly over the decades, most of the studies described above had similar methodological problems, which are inherent in studying those with substance use and mental disorders and treatment services for those populations. For example, treatment facilities and modalities change. To ensure that programs will be in operation at the end of data collection, it is prudent to select large well-established programs for study. Difficulties in following clients after treatment completion or discharge exist because such clients are highly mobile and often want anonymity after treatment completion. Some changes in the treatment milieu, occasioned by changes in financing structures and the impact of managed care, were largely unanticipated. Nonetheless, the evaluations and databases provide a rich source of information for behavioral health services researchers. To make optimal use of this information, researchers should be aware of the limitations of these datasets, as well as the societal conditions and treatment structures in place at the time of the evaluations. These factors should be considered in interpreting data on the organization and impact of treatment.

Overview of Chapters and Key Issues

The large expenditures of State and Federal dollars on substance abuse treatment and mental health services have spawned increased demand for accountability and determination of best practices. In response, the Federal Government has supported the creation of large databases that have been used to gain a better understanding of clients served and services provided. The chapters in this compendium use data from a variety of sources, including many of those described above, and provide important new knowledge.

Chapter 2, "Access to Substance Abuse Treatment and Mental Health Services: A Literature Review," examines literature on access to treatment for substance use disorders and to services for mental illnesses.
Chapter 3, "Substance Use and Mental Disorder Discharges from U.S. Community Hospitals in the Early 1990s, Revisited," examines trends in discharges from community hospitals nationwide of those with substance use and mental disorders between 1990 and 1995.
Chapter 4, "Do Client Characteristics Affect Admission to Inpatient Versus Outpatient Alcohol Treatment in Publicly Monitored Programs?" uses the 1996 Treatment Episode Data Set (TEDS) to examine the association between disorder severity and admission to publicly monitored inpatient versus outpatient alcohol treatment among adult males.
Chapter 5, "Client Choice among Standard Outpatient, Intensive Outpatient, Residential, and Inpatient Alcohol Treatment in State-Monitored Programs," examines client characteristics that are associated with client choice of treatment setting.
Chapter 6, "Length of Stay among Female Clients in Substance Abuse Treatment," examines factors associated with retention of female clients in treatment, using ADSS Phase II data.
Chapter 7, "A Hybrid Cost Function for Outpatient Nonmethadone Substance Abuse Treatment Facilities," uses economic modeling of administrative data from ADSS Phase II to estimate a cost function for outpatient nonmethadone substance abuse treatment facilities.
Chapter 8, "Effects of Reporting Requirements on Estimates from the Treatment Episode Data Set (TEDS)," examines potential biases that may exist in conducting research using TEDS data.
Chapter 9, "Conclusions and Implications," summarizes the chief findings of this compendium and discusses the implications of these findings with suggestions for future research.

The remaining discussion focuses on key issues from these chapters. These chapters provide new knowledge on such issues as access to treatment, treatment financing and costs, treatment retention, treatment choice, and some limitations of commonly used data sources.

Access, Financing, and Costs

Many of the Federal-sponsored evaluations mentioned earlier focused on issues related to populations served, services provided (including specialty services), characteristics of facilities providing treatment, and, to the extent possible, financing issues. This compendium further explores the financing of and access to substance abuse treatment and mental health services. Chapter 2 examines the literature on access to treatment for substance use and mental disorders, and it reviews a variety of attributes (with a special emphasis on financial factors) that may affect a client's ability to access treatment. In the chapter, the shift to outpatient treatment is discussed, as well as its possible ramifications. The importance of managed care and financing on access to treatment is underscored, especially among those clients with more serious behavioral health disorders. The chapter also includes a discussion of the ability of government-sponsored behavioral health care systems to improve access for those clients with more serious disorders.

Chapter 7 uses economic modeling of administrative data from Phase II of ADSS, described earlier, to estimate a cost function for outpatient nonmethadone substance abuse treatment facilities with a sample representing 9,166 facilities. In the chapter, the practice of using costs estimated from a small number of nonrandomly selected facilities to conduct cost-benefit analyses is called into question. Importantly, it is suggested not only that larger facilities may be able to provide care at a lower price than smaller facilities, but also that higher costs may be appropriate when dealing with high-risk populations, such as clients receiving supplemental security income (SSI).

The importance of considering client characteristics that affect access, as well as treatment-seeking behaviors and treatment retention, has been extensively reported upon in the health services research literature. Chapter 2 reviews the literature on client demographics and treatment-seeking behavior. TEDS data from 1996 are used in Chapter 4 to examine the association between disorder severity and admission to publicly monitored inpatient versus outpatient alcohol treatment. Subjects studied are adult males with alcohol as a primary substance of abuse who were admitted for rehabilitation treatment in a State-monitored program in one of nine States.

In Chapter 4, clients who either paid for treatment out-of-pocket or who received publicly funded treatment are examined, as is how the source of payment varied across States. Multivariate logistic regression models were used for each State, with results indicating that greater disorder severity increased the odds of inpatient admission (an exception was co-occurring mental disorders). Those clients who were employed were less likely to have an inpatient admission, while those who were homeless, who had one prior treatment episode, and/or who were referred by an alcohol or drug treatment provider were more likely to have an inpatient admission.

Treatment Retention

As described earlier, several large research studies have indicated that a longer length of stay (LOS) is associated with better client outcomes. In Chapter 6, the ADSS Phase II dataset is used to derive a nationally representative sample of female clients discharged from substance abuse treatment facilities to examine factors associated with treatment retention in one of four types of treatment facilities: nonhospital residential treatment only, outpatient methadone treatment only, outpatient nonmethadone treatment only, or a combination of types of care. Lengths of stay in treatment are compared based on demographic, socioeconomic, and organizational characteristics. As reported in earlier evaluations, the total number of services provided to clients in treatment was found to be declining. The study described in Chapter 6 finds that with regard to women's treatment, facilities offering child care services and treatment at women-only facilities were associated with longer lengths of stay.

Data Limitations

Most databases have important limitations; care must be taken in interpreting findings obtained from them and generalizing to the entire population of those with substance use and mental disorders. Although biases are evident in earlier evaluations, as well as in Federal-maintained databases, few previous studies have examined the impact of biases on research findings. Two chapters in this compendium focus on the potential impact of such biases. Chapter 3 is a review of the findings from a 1990–1995 study that examined discharge trends from community hospitals nationwide of patients with substance use and mental disorders. The chapter includes a critique of previous studies that used the National Hospital Discharge Survey (NHDS) and focused on community hospitalizations for substance use and mental disorders. New estimates are provided in the chapter using the Agency for Healthcare Research and Quality's Nationwide Inpatient Sample (NIS). A persuasive argument is made that the NIS is the appropriate dataset to use for making such estimates. Contrary to findings from previous research, the evidence in Chapter 3 indicates that community hospitalization of patients with substance use and mental disorder diagnoses increased between 1990 and 1995. Moreover, even though those patients were diagnosed with more complex disorders, they had short hospital stays. Because health care utilization studies (e.g., the RAND Health Insurance Experiment) can have an important impact on public policy and funding decisions, it is important to select the most appropriate dataset for studying and interpreting treatment-seeking behavior.

Chapter 8 examines biases that may be introduced into datasets, such as TEDS, as a result of State-level reporting requirements. Each State sets requirements on the types of providers or facilities that must report into the TEDS system, with some States only requiring that facilities receiving earmarked funds for substance abuse treatment report on their clients, while other States require that all facilities report regardless of their funding sources. This chapter examines whether or not selection bias occurs in the collection and reporting of data to TEDS for adult males with alcohol as a primary substance of abuse who were admitted for rehabilitation treatment. TEDS data are used from two States that collect information from both types of facilities. Findings indicate that biases do exist. Clients admitted to facilities receiving Federal earmarked funds had a different client demographic profile. Moreover, large variations in these profiles existed between States. Thus, it may be important for States to require reporting by all facilities, regardless of their funding status, if they want an accurate understanding of their substance use disorder.

Treatment Choice

Earlier research has underscored the shift of treatment services provision to outpatient settings, as well as the impact of managed care on provider treatment options. Chapter 5 examines client characteristics associated with choice of treatment setting in State-monitored facilities, including standard outpatient, intensive outpatient, short-term residential, long-term residential, and inpatient hospital rehabilitation. The chapter discusses the appropriateness of combining types of inpatient and outpatient treatment into two broad categories for analysis purposes. As in Chapter 8, this study uses 1996 TEDS data on adult males with alcohol as the primary substance of abuse in two States that collect information on all clients admitted to treatment, regardless of the use of earmarked funds. Multivariate multinomial logistic models were estimated for each State, and covariates included disorder severity and socioeconomic measures. Those admitted to the standard outpatient setting appeared to have less severe alcohol use disorders and were far more likely to be employed at admission than those admitted to other settings. These findings suggest that client treatment-setting choice should not be studied as a dichotomous choice between two types of treatment, but rather as a choice among multiple settings.

The chapters in this compendium add to the body of knowledge concerning the provision of treatment services in behavioral health care. However, it is important to note that although much of the research obtained from the general health services continuum has relevance for treating substance use and mental disorders, important differences should be considered when making policy inferences. In Chapter 9, several of the issues raised in the compendium are discussed, their implications for policymakers are presented, and areas for further research are suggested.

References

Allison, M., Hubbard, R. L., Craddock, S. G., & Rachal, J. V. (1982). Drug abuse treatment process: Preliminary examination of the TOPS data (RTI/1901/-1–068). Research Triangle Park, NC: Research Triangle Institute.

American Medical Association. (2004, January 7). Chronology of AMA history: 1847 to 1899. Retrieved February 10, 2004, from http://www.ama-assn.org/ama/pub/category/1922.html

American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (DSM-III-R) (3^rd rev. ed.). Washington, DC: Author.

Batten, H. L., Horgan, C. M., Prottas, J. M., Simon, L. J., Larson, M. J., Elliott, E. A., Bowden, M. L., & Lee, M. T. (1993). Drug Services Research Survey, Phase I final report: Non-correctional facilities (revised; submitted to the National Institute on Drug Abuse). Waltham, MA: Institute for Health Policy, Brandeis University. [Available at http://www.oas.samhsa.gov/dsrs.htm]

Batten, H. L., Prottas, J. M., Horgan, C. M., Simon, L. J., Larson, M. J., Elliott, E. A., & Marsden, M. E. (1992). Drug Services Research Survey, final report: Phase II (submitted to the National Institute on Drug Abuse). Waltham, MA: Institute for Health Policy, Brandeis University. [Available at http://www.oas.samhsa.gov/dsrs.htm]

Blanken, A. (1989, June). Evolution of a national database for drug abuse treatment clients. In Proceedings of the Community Epidemiology Work Group: Epidemiologic trends in drug abuse. Rockville, MD: National Institute on Drug Abuse.

Comprehensive Alcohol Abuse, Drug Abuse, and Mental Health Amendments, Pub. L. No. 100–690, Title II, ch. 2, § 2052 (1988).

Comprehensive Health Planning Act, Pub. L. No. 89–749 (1966). [Also known as the Partnership for Health Act of 1966 and its 1967 Amendments]

Drug Abuse Office and Treatment Act, Pub. L. No. 92–255 (1972).

Fletcher, B. W., Tims, F. M., & Brown, B. S. (1997). The Drug Abuse Treatment Outcome Study (DATOS): Treatment evaluation research in the United States. Psychology of Addictive Behaviors, 11, 216–229.

Flynn, P. M., Craddock, S. G., Hubbard, R. L., Anderson, J., & Etheridge, R. M. (1997). Methodological overview and research design for the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11, 230–243.

Freund, D. A. (1994). "Free for All?" Lessons from the RAND Health Insurance Experiment [Book review]. Health Affairs, 13, 293–295.

Ginzburg, H. M. (1978). Recent developments in chemotherapy of narcotic addiction. Part VII. Evaluation of treatment effectiveness: defensive research-The Treatment Outcome Prospective Study (TOPS). Annals of the New York Academy of Sciences, 311, 265–269.

Harvard Medical School, Office of Public Affairs. (2003, November 11). Study challenges proposed changes to clinical definition of mental illness: Untreated, mild forms of mental illness can become more severe over time. Retrieved June 2, 2004, from http://www.hms.harvard.edu/news/releases/11_10kessler.html

Harwood, H. J., Hubbard, R. L., Collins, J. J., & Rachal, J. V. (1988). The costs of crime and the benefits of drug abuse treatment: A cost-benefit analysis using TOPS data. In C. G. Leukefeld & F. M. Tims (Eds.), Compulsory treatment of drug abuse: Research and clinical practice (pp. 209–235, NIH Publication No. 94 3713, NIDA Research Monograph 86). Rockville, MD: National Institute on Drug Abuse.

Hubbard, R. L., Craddock, S. G., Flynn, P. M., Anderson, J., & Etheridge, R. M. (1997). Overview of 1–year follow-up outcomes in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11, 261–278.

Hubbard, R. L., Marsden, M. E., Rachal, J. V., Harwood, H. J., Cavanaugh, E. R., & Ginzburg, H. M. (1989). Drug abuse treatment: A national study of effectiveness. Chapel Hill, NC: University of North Carolina Press.

Institute of Medicine (G. S. Wunderlich, Ed.). (1992). Toward a national health care survey: A data system for the 21st century. Washington, DC: National Academy Press. [Available at http://books.nap.edu/catalog/1941.html]

Institute of Medicine (M. J. Field, R. E. Tranquada, & J. C. Feasley, Eds.). (1995). Health services research: Work force and educational issues. Washington, DC: National Academy Press. [Available at http://books.nap.edu/catalog/5020.html]

Joe, G. W., Simpson, D. D., & Hubbard, R. L. (1991). Treatment predictors of tenure in methadone maintenance. Journal of Substance Abuse, 3, 73–84.

Kandel, D. B., Warner, L. A., & Kessler, R. C. (1998). The epidemiology of substance use and dependence among women. In C. L. Wetherington & A. B. Roman (Eds.), Drug addiction research and the health of women (pp. 105–130, NIH Publication No. 98–4290). Rockville, MD: National Institute on Drug Abuse.

Kessler, R. C., Crum, R. M., Warner, L. A., Nelson, C. B., Schulenberg, J., & Anthony, J. C. (1997). Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Archives of General Psychiatry, 54, 313–321.

Kessler, R. C., Merikangas, K. R., Berglund, P., Eaton, W. W., Koretz, D. S., & Walters, E. E. (2003). Mild disorders should not be eliminated from the DSM-V. Archives of General Psychiatry, 60, 1117–1122.

Koenig, L., Denmead, G., Nguyen, R., Harrison, M., & Harwood, H. (1999, August). The cost and benefits of substance abuse treatment: Findings from the National Treatment Improvement Evaluation Study (NTIES). Rockville, MD: Center for Substance Abuse Treatment. [Available as a PDF at http://www.icpsr.umich.edu/SAMHDA/NTIES/NTIES-PDF/REPORTS/cost-ben.pdf].

Lohr, K. N., & Steinwachs, D. M. (2002). Health services research: An evolving definition of the field. Health Services Research, 37, 15–17.

McCarthy, T., & White, K. L. (2000). Origins of health services research. Health Services Research, 35, 375–387.

National Institute on Drug Abuse. (1982). Data from the Client Oriented Data Acquisition Process (CODAP): Annual data, 1981 (DHHS Publication No. ADM 82–1223, Statistical Series E, No. 25). Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, National Institute on Drug Abuse.

National Library of Medicine. (1998, April 27). Images from the history of the Public Health Service: Mental health and drug abuse. Retrieved February 10, 2004, from http://www.nlm.nih.gov/exhibition/phs_history/121.html

National Library of Medicine. (2001, May 5). Saint Elizabeths Hospital. Retrieved February 12, 2004, from http://www.nlm.nih.gov/hmd/medtour/elizabeths.html

Newhouse, J. P. (1999). RAND Health Insurance Experiment [in metropolitan and non-metropolitan areas of the United States], 1974–1982 [Computer file] (ICPSR version, Santa Monica, CA: The RAND Corporation [producer], 1986). Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor].

Newhouse, J. P., & Insurance Experiment Group. (1994). Free for all? Lessons from the RAND Health Insurance Experiment. Cambridge, MA: Harvard University Press.

Normand, C. (1994). Views and reviews: Free for all: Lessons from the RAND Health Insurance Experiment. British Medical Journal, 308, 1724–1725.

Office of Applied Studies. (1998). Services Research Outcomes Study (SROS) (DHHS Publication No. SMA 98–3177, Analytic Series A-5). Rockville, MD: Substance Abuse and Mental Health Services Administration. [Available at http://www.oas.samhsa.gov/analytic.htm]

Office of Applied Studies. (2003a). Alcohol and Drug Services Study (ADSS) cost study: Costs of substance abuse treatment in the specialty sector (DHHS Publication No. SMA 03–3762, Analytic Series A-20). Rockville, MD: Substance Abuse and Mental Health Services Administration. [Available as a PDF at http://www.oas.samhsa.gov/adss.htm]

Office of Applied Studies. (2003b). Alcohol and Drug Services Study (ADSS): The national substance abuse treatment system: Facilities, clients, services, and staffing. Rockville, MD: Substance Abuse and Mental Health Services Administration. [Available as a PDF at http://www.oas.samhsa.gov/adss.htm]

Office of Applied Studies. (2004). Treatment Episode Data Set (TEDS): 1992–2001: National admissions to substance abuse treatment services (DHHS Publication No. SMA 03–3778, Drug and Alcohol Services Information System Series S-20). Rockville, MD: Substance Abuse and Mental Health Services Administration. [Available at http://www.oas.samhsa.gov/dasis.htm#teds2]

Sells, S. B. (Ed.). (1974). Studies of the effectiveness of treatments for drug abuse. Volume 1: Evaluation of treatments. Cambridge, MA: Ballinger Publishing Company.

Simpson, D. D., Joe, G. W., & Brown, B. S. (1997). Treatment retention and follow-up outcomes in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11, 294–307.

Simpson, D. D., & Savage, L. J. (1980). Drug abuse treatment readmissions and outcomes: Three-year follow-up of DARP patients. Archives of General Psychiatry, 37, 896–901.

Tims, F. M., Fletcher, B. W., & Hubbard, R. L. (1991). Treatment outcomes for drug abuse clients. In R. W. Pickens, C. G. Leukefeld, & C. R. Schuster (Eds.), Improving drug abuse treatment (pp. 93–113, DHHS Publication No. ADM 91–1754, NIDA Research Monograph 106). Rockville, MD: National Institute on Drug Abuse.

Chapter 1. Health Services Utilization by Individuals with Substance Use and Mental Disorders

Health Services Research

Health Services Research in the Behavioral Health Sciences

Overview of Chapters and Key Issues

References

End Notes