Abstract
California’s Proposition 36 offers nonviolent drug offenders community-based treatment as an alternative to incarceration or probation without treatment. The study objective was to examine how substance abuse treatment providers perceive the impact of Proposition 36 on their clinical decision making. Program surveys were completed by 115 treatment programs in five California counties to assess the impact of the law on clinical decision making, and five focus groups were conducted with 37 treatment providers to better understand their perspectives. Compared to residential programs, outpatient programs reported that the policy impacted them to a greater extent in terms of drug testing, reporting to criminal justice personnel, and determining client discharge. Providers in the focus groups particularly highlighted their changing roles in assessing clients’ treatment needs and determining the best routes of care for them. The findings indicate that alternate strategies for determining treatment placement and continuing care should be developed.
California’s Proposition 36, enacted as the Substance Abuse and Crime Prevention Act of 2000 (SACPA), has been in operation for more than 5 years. This law allows (under certain conditions) adults convicted of nonviolent drug-related offenses to choose community-based drug treatment in lieu of incarceration or probation without treatment. Offenders on probation or parole who violate drug-related conditions of probation or parole can also receive treatment. Additionally, Proposition 36 clients are entitled to three chances to succeed in treatment, meaning they can stay out of jail or prison until the third time they violate conditions of their treatment program (e.g., no show to treatment, test positive for drugs). The initiative mandated a statewide evaluation of its fiscal impact and effectiveness which was conducted by UCLA. However, the statewide evaluation did not provide for an in-depth investigation of the effect of Proposition 36 on local drug treatment service delivery systems, and little is known about the consequences of similar drug policies and programs on the treatment delivery system. The Treatment System Impact Study was designed to address this gap by exploring the effects of the implementation of SACPA on California’s treatment service delivery system, including counties and treatment providers. The findings will elucidate the effects of the policy on the treatment service delivery system and can inform policy and practice at the national, state, and local levels.
California has one of the largest drug treatment systems in the nation, with more than 900 community-based programs serving approximately 250,000 patients annually,1 and Proposition 36 introduced an unprecedented number of clients into this system.2 Overall, treatment admissions increased by 11% to 34% during the first year in four of the five counties studied, with an 11% increase statewide. In the law’s second year, the number of Proposition 36 offenders entering drug treatment continued to increase in all counties and statewide (50% increase over year 1). In year 2, Proposition 36 clients constituted 20% of the state’s treatment system, which is a significant proportion given the brief time since implementation. This expansion occurred primarily in outpatient drug programs as opposed to residential programs.2 Treatment completion rates of clients who entered treatment in the first 3 years of implementation ranged from 32.0% to 34.4%, and treatment completion was related to lower re-arrest rates.3 Although SACPA implementation was not associated with statewide crime trends, those who were eligible for SACPA in its first 2 years were more likely to have a new arrest for drug or property offenses in an ensuing 30-month period compared to a similar group of offenders in the pre-SACPA-era.3 Nevertheless, UCLA’s evaluation showed that SACPA yielded cost savings to both state and local governments.3 Specifically, they reported an overall savings of $2.50 for every $1 invested and a savings of $4 for every $1 invested for individuals who completed treatment.
Under the law’s general guidelines, individual counties determined the specifics for the law’s implementation, and there was considerable county variation on how Proposition 36 was implemented.4 For example, the types and availability of treatment modalities and levels of care differ across counties (e.g., some counties do not provide narcotics replacement treatment). Some counties utilize centralized assessment center(s) to conduct assessments and determine and assign individuals to specific programs and levels of care, while other counties rely on individual programs to make those decisions. Agencies involvement and interactions with one another also differ tremendously across counties. For example, counties vary in the extent to which substance abuse treatment agencies work collaboratively with criminal justice and mental health agencies.
Drug-abusing offenders are not new to community-based treatment programs, as almost half of outpatient treatment admissions before Proposition 36 were either referred by the Criminal Justice System (CJS) or otherwise involved in the CJS.5 Tensions between the two systems, with treatment providers focusing on therapeutic principles and CJS personnel focusing on punishment and supervision, have been well documented.6,7 Nevertheless, compliance with the law has necessitated and formalized greater interaction among agencies, including sharing of client information and making treatment decisions.8 For example, the state has made available funds for urine testing because positive urines are clear evidence of program violations. Urine testing has become a routine component of CJS supervision and is often used by treatment providers to make therapeutic decisions. This paper addresses how treatment providers perceive shifts in their roles and responsibilities as a result of the interpenetrating CJS that does not typically have clinical care as a goal. The findings can be used to inform practice at the state and local levels, as well as by other states that are considering instituting such policy as Proposition 36.
As a result of their process evaluation of treatment programs for offenders,9 Taxman and Bouffard10 proposed a conceptual framework for the development of an integrated service delivery system (i.e., a system that integrates the criminal justice and substance abuse treatment systems). They note the potential tension in the integrated system between “external control-oriented services” such as criminal justice-based testing and monitoring and “internal control-oriented services” such as clinical care. Their framework includes the following six decision-making domains that are implicated when the two systems are combined: (1) screening and assessment, (2) treatment placement criteria, (3) treatment progress and continuum of care criteria, (4) supervision/monitoring components, (5) drug testing, and (6) criteria for discharge due to noncompliance and criteria for completion of treatment and supervision. The objective of this study is to use Taxman and Bouffard’s conceptual framework to examine how substance abuse treatment providers qualitatively characterize the impact of Proposition 36 on their clinical decision making. Quantitative data were also collected to provide descriptive data regarding the impact of Proposition 36 on clinical decision making. Utilizing the quantitative data, the perceptions of providers working in outpatient and residential treatment programs were also compared, as prior research has shown that Proposition 36 has increased treatment admissions to outpatient programs more so than residential programs.2 Hence, we hypothesized that outpatient programs would perceive the impact of Proposition 36 to be greater on their clinical decision making than would residential programs.
Methods
Study design
The Treatment System Impact (TSI) Study was designed to investigate the impact of Proposition 36 on local treatment systems and their linkages to the respective criminal justice systems. Using a purposive sampling strategy aimed at maximizing variation in a small sample of counties, five diverse counties were selected along dimensions hypothesized to have important implications for each county’s alcohol and drug treatment system and participant outcomes including geography (e.g., urban, rural; northern, central, southern), population (e.g., large, small), and Proposition 36 implementation strategies (e.g., urine testing, treatment options).4 Counties eligible for selection had to have admitted at least 100 Proposition 36 clients in the first implementation year (1 July 2001 to 30 June 2002) and participated in a prior treatment outcome study. The five counties selected were Kern, Riverside, Sacramento, San Diego, and San Francisco.
Program surveys
All programs serving Proposition 36 clients in the five study counties were invited to participate in a survey from December 2005 through August 2006 to assess the impact of the law on local treatment service delivery systems during the first 4 years of implementation. All questionnaires were mailed directly to programs. The survey took approximately 1 h to complete, and respondents were offered $100 for their participation. This paper presents findings from six items in the survey that assessed the impact of Proposition 36 on clinical decision making using a four-point Likert scale where 0 = Not at all, 1 = Limited extent, 2 = Moderate extent, and 3 = Great extent. The programs’ use of drug testing was also assessed in the program survey.
The 115 program surveys included in this study represented 77 outpatient and 38 residential treatment programs. In total, 126 of the 137 (92%) treatment sites contacted responded to the survey. However, 11 of these questionnaires did not complete the section on clinical decision making, resulting in a response rate of 84% for this study. The surveys were completed by nine executive directors (7.8%), 45 program directors/managers (39.1%), 27 program supervisors/coordinators (23.5%), nine clinic administrators or other administrative staff (7.8%), and eight clinical supervisors or counselors (6.9%). Seventeen of the respondents (14.8%) did not identify their role in the treatment program.
Focus groups
To augment the quantitative data collection effort, focus groups were conducted with treatment provider staff in March and April 2004 to better understand their perspectives on the impact of Proposition 36 on the local treatment service delivery system. One focus group was held in each of the five participating counties. In three of the counties (Sacramento, San Diego, and Riverside), study staff contacted the directors of eight programs in each county that were collecting client intake assessment data as part of the Treatment Outcomes component of the TSI study. In the other two counties (Kern and San Francisco), client intake assessment data collection was centralized at the county level; therefore, study staff contacted the directors of eight programs that served the most Proposition 36 clients stratified by treatment modality. No directors declined participation. A flyer about the focus group was posted at each program, indicating that if more than one staff member from a program volunteered, UCLA would randomly select one participant.
Five focus groups (one in each of the five counties) were conducted, involving 37 individuals total. Approximately half (54%) of the participants were women; 73% identified as white, 19% Hispanic, 3% black, and 5% other race/ethnicity. The average age of the focus group participants was 45±10 years, with a range of 24 to 63 years. The majority (97%) had at least some college-level education, with 19% having an advanced degree. Participants had worked in the alcohol/drug treatment field an average of 9±7 years, and they had worked at their current agencies an average of 5±5 years. Participants reported their primary role(s) as individual/group counselor (70%), intake/assessment counselor (35%), supervisor of counseling staff (32%), or other (e.g., associate executive director; 19%). More than half (59%) of the participants had multiple responsibilities (e.g., intake/assessment and individual/group counseling).
Focus groups were held at county agency facilities. Informed consent was completed by all participants, and background information was collected through a brief, anonymous, pencil-and-paper questionnaire. Each focus group, which lasted approximately 2 hours, was digitally recorded, and an assistant took notes throughout. During the session, participants were asked to discuss their programs’ treatment approaches/orientations, impact of Proposition 36 on the delivery of treatment services (e.g., type, quantity, and quality of services for Proposition 36 and non-Proposition 36 clients, provision of services for Proposition 36 clients with special needs), and administration of Proposition 36 at their programs (e.g., procedural and staff changes, what has worked well, what has not worked well, lessons learned). Participants from programs that allowed compensation were paid $100 for their participation.
Analytic approach
With regard to the program survey, the impact of Proposition 36 on clinical decision making was analyzed by treatment modality. Group differences were examined using chi-square analysis. For the focus groups, the digital recordings were transcribed verbatim by a transcription service. Researchers checked each transcript against the recordings for accuracy. Transcripts were then analyzed in ATLAS.ti, a qualitative data analysis software program. Initially, transcripts were coded using broad topic-driven categories, and responses within these categories were reviewed by the research team using the constant comparative method where results are compared across data sources. Based on this review, specific content-driven codes were developed that were relevant to most or all of the interviews and that corresponded roughly to the domains proposed by Taxman and Bouffard. The definition of codes was discussed among three researchers and documented. After coding was completed, the analysis was reviewed and confirmed by another researcher. Relationships between the secondary codes were examined using the program’s network diagramming feature that allows the user to graphically depict overlapping themes (i.e., codes). Themes and patterns emerging from the coded data are illustrated in the results with verbatim quotes. Because there was considerable in-practice overlap among the Taxman and Bouffard domains (e.g., supervision/monitoring and drug testing were always intertwined in respondent narratives), three larger categories for discussion were pursued: (1) screening/assessment and treatment placement, (2) supervision/monitoring, testing, and discharge/completion, and (3) treatment progress and continuum of care.
Results
According to participants in the focus groups, Proposition 36 affected clinical decision making with regard to assessment and placement, client progress and the continuum of care, supervision and monitoring (including drug testing), and treatment completion and discharge. All of the focus groups discussed the impact of the policy on assessment and placement, while the other topics were raised in at least two of the five focus groups. Each domain will be explicated below, with illustrative quotes when relevant. The impact of Proposition 36 on clinical decision-making was also compared across outpatient and residential treatment programs using chi-square analysis (see Table 1).
Table 1.
Impact of Proposition 36 on clinical decision making and consequences of positive drug tests
| All programs (n=115) | Outpatient (n=77) | Residential (n=38) | χ2 value | ||
|---|---|---|---|---|---|
| Prop. 36 has influenced: | |||||
| Whether clients are discharged from the program | Not at all | 32 (28.6) | 12 (16.0) | 20 (54.1) | 18.8** |
| Limited extent | 30 (26.8) | 25 (33.3) | 5 (13.5) | ||
| Moderate extent | 23 (20.5) | 16 (21.3) | 7 (18.9) | ||
| Great extent | 27 (24.1) | 22 (29.3) | 5 (13.5) | ||
| Reporting of treatment noncompliance to CJ personnel based on the potential legal consequences to clients | Not at all | 30 (26.3) | 18 (23.7) | 12 (31.6) | 8.9* |
| Limited extent | 15 (13.2) | 9 (11.8) | 6 (15.8) | ||
| Moderate extent | 21 (18.4) | 10 (13.2) | 11 (28.9) | ||
| Great extent | 48 (42.1) | 39 (51.3) | 9 (23.7) | ||
| Duration of treatment | Not at all | 20 (17.4) | 10 (13.0) | 10 (26.3) | 4.9 |
| Limited extent | 11 (9.6) | 9 (11.7) | 2 (5.3) | ||
| Moderate extent | 32 (27.8) | 20 (26.0) | 12 (31.6) | ||
| Great extent | 52 (45.2) | 38 (49.4) | 14 (36.8) | ||
| Provision of services (e.g., type and intensity) | Not at all | 31 (27.0) | 15 (19.5) | 16 (42.1) | 6.7 |
| Limited extent | 10 (8.7) | 7 (9.1) | 3 (7.9) | ||
| Moderate extent | 32 (27.8) | 24 (31.2) | 8 (21.1) | ||
| Great extent | 42 (36.5) | 31 (40.3) | 11 (28.9) | ||
| Frequency of drug testing | Not at all | 27 (23.7) | 12 (15.8) | 15 (39.5) | 10.4* |
| Limited extent | 27 (23.7) | 22 (28.9) | 5 (13.2) | ||
| Moderate extent | 23 (20.2) | 14 (18.4) | 9 (23.7) | ||
| Great extent | 37 (32.5) | 28 (36.8) | 9 (23.7) | ||
| Referrals to specific providers (i.e., for specialized services) | Not at all | 33 (28.9) | 20 (26.3) | 13 (34.2) | 1.8 |
| Limited extent | 29 (25.4) | 18 (23.7) | 11 (28.9) | ||
| Moderate extent | 33 (28.9) | 24 (31.6) | 9 (23.7) | ||
| Great extent | 19 (16.7) | 14 (18.4) | 5 (13.2) | ||
| Consequences of positive drug tests for Prop. 36 clients: | |||||
| Adjustment is made to the client’s treatment plan | Not at all | 18 (16.2) | 0 (0.0) | 18 (51.4) | 48.6** |
| Limited extent | 8 (7.2) | 5 (6.6) | 3 (8.6) | ||
| Moderate extent | 23 (20.7) | 18 (23.7) | 5 (14.3) | ||
| Great extent | 62 (55.9) | 53 (69.7) | 9 (25.7) | ||
| Change is made to client’s level of care (e.g., transfer from outpatient to day treatment) | Not at all | 22 (20.2) | 2 (2.7) | 20 (58.8) | 47.7** |
| Limited extent | 22 (20.2) | 18 (24.0) | 4 (11.8) | ||
| Moderate extent | 43 (39.4) | 34 (45.3) | 9 (26.5) | ||
| Great extent | 22 (20.2) | 21 (28.0) | 1 (2.9) | ||
| Frequency of drug testing is increased | Not at all | 19 (17.1) | 1 (1.3) | 18 (51.4) | 43.8** |
| Limited extent | 27 (24.3) | 22 (28.9) | 5 (14.3) | ||
| Moderate extent | 37 (33.3) | 28 (36.8) | 9 (25.7) | ||
| Great extent | 28 (25.2) | 25 (32.9) | 3 (8.6) |
p<0.05;
p<0.0001
Assessment and placement
With centralized assessment (i.e., “gate teams”) for Proposition 36 clients, some county providers questioned the accuracy of the diagnostic process. This process seemed particularly challenging with dual diagnosis clients. As a result, the conundrum that providers sometimes faced was that clients were improperly assessed and then assigned to inappropriate treatment levels.* For example, a provider in one county stated, “I’m wondering how in the world did the gate team… assess them to bring them in? We have people coming in at level two who…we know that level three would be just the right thing for them, based on an assessment you do, which is an ASI [addiction severity index] Lite. Yet they’re coming in at level two. The assessment process with the gate team, sometimes it’s questionable how and for what reason they send them in that level.” Other providers from this county suggested that the criminal justice system decided where clients should be placed, with minimal regard for a clinical diagnosis. Whether clients were sometimes inappropriately assessed by the gate team or placed without adequate diagnosis, consideration of what the treatment providers felt the clients needed was missing from the process. Providers in this focus group felt that their assessments were often not solicited or respected: “But they won’t …call [us] and say, ‘You know, I respect your assessment and he’s here with me and we have done our assessment, and I truly believe he should be in this level.’ It’s not going to happen.”
Participants in some of the focus groups noted that placement of women and the homeless was particularly challenging. Other counties faced problems with inappropriate placement of their dually diagnosed clients due to minimal screening for co-occurring mental health disorders and/or the lack of regard for the provider’s assessment. For example, one provider noted, “One of the things that constantly comes up for us…is to have an outpatient counselor do a basic assessment of their client that’s sinking and send them back to [the gatekeeper] and then have [the gatekeeper] totally discount what the counselor has said….It’s been an ongoing battle when a competent qualified alcohol and drug counselor makes an assessment with a client that is seeking treatment, and refers them for, say, level three or detox—a homeless, indigent individual—and [the gatekeeper] says, ‘Well, we got to let them fail.’” Interestingly, and in contrast, one county noted that their dually diagnosed clients were experiencing increased access to psychiatrists, and they found that their probation department was making efforts to be flexible with these clients. Providers from this county noted that assessment and placement improved once the process was moved from probation back to the treatment providers.
According to the program survey, 64.3% of programs reported that Proposition 36 impacted the provision of services provided to a moderate or great extent and 45.6% reported the law impacted referrals to specialized services to a moderate or great extent. Proposition 36 did not differentially affect outpatient and residential programs in terms of the provision of services provided or referral for specialized services (p>0.05).
Supervision and monitoring, drug testing, and discharge
Three focus groups discussed the impact of Proposition 36 drug testing on their clinical care. In one county, residents had to leave the program and go several places to get tested and receive their results. Providers pointed out that this process detracted from treatment because clients had to do this four to five times within 45 days. In another county where testing was done within the treatment facilities, providers noted that they spent a great deal of time testing their Proposition 36 clients who were resentful that they were tested more than voluntary clients. Providers in this focus group also noted that they had lost the freedom to determine when clients should be maintained in or discharged from treatment, as this decision was being based more on drug test results than on the providers’ assessments of what would be clinically best for the clients. These providers viewed this as “moving backwards” and wanted to be left to make their own decisions: “Let us make the decision because we work with them every day and we know the ones that are out there that are not participating in the program when they are here, and [we are] discharging them. But [the criminal justice people] will ask, ‘Well, why did they get discharged after three but this one’s not?’” As a result of less flexible treatment decision making, providers noted that their clinical work had become “canned treatment” and they felt that this was potentially a “liability” for the clients.
Program surveys revealed that Proposition 36 impacted programs to a moderate or great extent in a number of ways, including client discharge from the program (44.6%), reporting of treatment noncompliance to criminal justice personnel (60.5%), treatment duration (73%), and frequency of drug testing (52.7%). Compared to residential programs, outpatient programs reported that Proposition 36 impacted their programs to a greater extent with regard to whether clients were discharged from their programs (χ2=18.8, p<0.0001), whether treatment noncompliance was reported to criminal justice personnel based on the potential legal consequences to clients (χ2=8.9, p<0.05) and the frequency of drug testing (χ2=10.4, p<0.05). It did not differentially affect outpatient and residential programs in terms of treatment duration (p>0.05).
Most programs (n=79, 68.7%) typically reported Proposition 36 clients to CJS on their first positive drug test. Before reporting, other programs typically allowed one positive drug test (n=7, 6.1%), two positive drug tests (n=11, 9.6%), three positive drug tests (n=4, 3.5%), five positive drug tests (n=1, 0.9%), or six positive drug test (n=5, 4.3%). Eight programs (7%) did not respond to this question. Programs also varied in the number of positive drug tests typically allowed before a Proposition 36 client was discharged from the treatment unit. Twenty-one (18.3%) programs typically discharged clients after the first positive drug test, whereas other programs typically allowed one positive drug test (n=15, 13.0%), two positive drug tests (n=17, 14.8%), three positive drug tests (n=29, 25.2%), or four positive drug test (n=22, 19.1%) before discharge. Eleven programs (9.6%) did not respond to this question. The treatment consequences of positive drug tests for Proposition 36 clients were compared across outpatient and residential treatment programs using chi-square analysis (see Table 1). Compared to residential programs, outpatient programs reported that positive drug tests were more likely to lead to an adjustment to the client’s treatment plan and level of care and to an increase in the frequency of drug testing (all p<0.0001).
Client progress and continuum of care
Two focus groups expressed concerns with the effect of Proposition 36 on client progress and the continuum of care. In one county, participants noted that the fundamental treatment approach shifted with the influx of Proposition 36 money and clients. By shifting from predominantly one-on-one treatment to group treatment, more clients were able to be accommodated by the treatment programs. Providers in this group discussed how services were “manipulated” to get the “biggest bang for the buck.” They were not convinced that this shift improved care for the clients. Participants noted that communication between the treatment and criminal justice systems regarding client progress and changes in treatment level was often challenging. As one provider noted, “The only thing I really find problematic is the fact that sometimes…the other [criminal justice] entities cannot get back to the treatment facilities with what’s going on with the client. They may send a client somewhere….So we would assume the [probation/parole officer] would call us and say, ‘This client is no longer there, so you can close them out.’” Providers in another focus group noted that it was difficult to communicate regarding the need to change clients’ treatment levels because the criminal justice entities did not see what the providers saw: “I’m seeing one picture and they’re seeing another…” Their concern was that in keeping clients in inappropriate levels of care, they would have less chance of successful completion. Several providers discussed being “client-oriented” and felt that refusal to change a client’s level of care was “punitive.” One provider noted that not allowing a client to change levels would “cause this client to fail.” Another provider responded, “Let’s at least give him the opportunity.” According to the program survey, 51% of programs reported that changing a Proposition 36 client’s level of care was not at all difficult, 42% reported it was somewhat difficult, and 7% reported it was very difficult.
Discussion
The findings from this study point to issues that may need to be considered as California enters into a new phase of policy implementation and as other states consider policy shifts similar to Proposition 36.11 Little attention has been paid to how this policy operates “on the ground” in clinics where providers have to accommodate not only a new population of clients but also a new system of care that involves collaboration with a system that was previously separate. Providers in the focus groups particularly highlighted their changing roles in assessing clients’ treatment needs and determining the best routes of care for them, emphasizing the tension between county assessment centers and treatment providers in determining initial level of care and between the criminal justice system and treatment providers in increasing intensity of ongoing care as opposed to incarceration. For providers in some counties, assessment of client populations with multiple needs (such as dually diagnosed, women, and homeless clients) was particularly challenging when not in the hands of the clinicians themselves. Further, the greater frequency of drug testing and the expectation to report positive drug tests has resulted in a reduced flexibility in responding to client needs. The quantitative results lend further support to these findings by showing that clinical decision making was impacted to a moderate or great extent by 45–73% of treatment programs depending on what aspect of decision making was being examined. Compared to residential programs, outpatient programs reported that the policy impacted them to a greater extent in terms of drug testing, reporting to criminal justice personnel, and determining client discharge. This is not surprising considering that Proposition 36 has increased treatment admissions to outpatient programs more so than residential programs.2 These unintended consequences of the policy, which need to be resolved in California, may also become problems for other states that are adopting similar policies.
With regard to treatment placement, there has been tension since the early stages of policy planning about placing clients based on their treatment needs versus their risk to society (i.e., their criminal record). According to Klein et al.,11 placement strategies potentially varied by county according to which stakeholders were advocating the strongest for which approach to placement decisions. These authors advocated tracking the data on appropriateness of placements, particularly in relation to outcomes. However, an additional obstacle to appropriate placement is the lack of availability of more intensive treatment environments, such as residential treatment. The vast majority of Proposition 36 funds were allocated to outpatient treatment programs as a cost-efficient means to accommodate the large flow of Proposition 36 offenders.2 However, outpatient treatment may not be optimal for certain subpopulations of Proposition 36 clients, such as those who are homeless or who have co-occurring substance abuse and mental health disorders. Residential programs are much more costly than outpatient programs12,13 and are more difficult to expand, especially in the short time since the implementation of Proposition 36. According to the UCLA final report of the 2001–2006 SACPA evaluation, “although the absolute number of available residential placements increased slightly after SACPA implementation, the treatment system was unable to keep pace with the increase in demand.”3 The report recommended that expanded use of residential treatment for heavy users should be prioritized.
Furthermore, in some counties, the nature of treatment itself had changed as a result of the policy, and providers were concerned that their clients’ needs were not being adequately met. This is similar to what Sherman et al.14 described as the “unplanned” changes that occur when systems change upon linkage with one another (i.e., what the same authors call “boundary spanning”). In their study, the presence of a mental health supervisor in the jails was pivotal to the success of their program. It is possible that trading places (done by some counties)—i.e., having criminal justice personnel spend time in substance abuse treatment settings and vice versa—could contribute to a greater understanding of the issues and parameters that each group faces. As Taxman and Bouffard15 note, there is a history in drug treatment court of ignoring treatment providers in decision making especially with regard to assessment and placement. Appropriate linkages must be made with regard to joint delivery of services.16-18
This study has limitations in regard to sample selection. The roles of program survey respondents varied, with some respondents in administrative roles and others in clinical roles, and it is unknown if these individuals were the best suited to provide data regarding the clinical aspects of Proposition 36. With regard to the focus groups, a potential selection bias may have occurred in that the number of treatment staff with direct contact with Proposition 36 clients identified by the Program Directors varied across the programs, and potential staff participants may have been selected for particular reasons of which researchers are not aware (e.g., providers who were more positive about Proposition 36 may have been overidentified). Because participants from only eight programs in each of the five study counties (of the 58 total counties in California) were included in the focus groups, the findings may not be generalizable to other programs within the five counties or to other counties across the state. Furthermore, due to the observational nature of this study, the changes observed in treatment can potentially be attributable to other developments in healthcare, such as shortened treatment duration, an increase in evidence-based practices, and cost savings in providing outpatient services and group interventions as opposed to residential care and individual therapy. However, the findings are consistent with the larger environmental context of Proposition 36.
Several aspects of California’s unique treatment system potentially reduce the generalizability of the findings. Specifically, in California, counties are dominant in implementing policy and are very different from one another. States in which counties have less influence may experience different problems than the ones reported here. Further, this study focused on program-level factors and did not address county-level factors (i.e., rural versus urban area). Future research should take county-level factors into consideration.
Implications for Behavioral Health
Proposition 36 has been controversial in California and has been watched closely nationally. This study enhances our understanding of the policy’s unintended effects on the treatment service delivery system and, more specifically, how treatment providers perceive shifts in their roles and responsibilities as a result of the interpenetrating criminal justice system. The findings can be useful in informing policy at the national, state, and local levels. The results suggest that alternate strategies for determining treatment placement need to be developed and more intensive treatment programs need be available. Increasing the capacity for more intensive programs would allow for a step-down strategy where individuals with severe problems begin at higher levels of care and move to lower levels of care as treatment indicates rather than the current “fail first” strategy (i.e., assigning all clients to the lowest levels of care and only moving them to more intensive care after they fail or even punishing them for relapsing). Enhanced training of assessors and improved communication between the criminal justice system and the substance abuse treatment system may also improve initial placement as well as ongoing care.
Acknowledgments
The study was supported in part by the National Institute on Drug Abuse (R01DA15431). Dr. Niv was supported by NIDA Institutional NRSA (5T32DA007272-15), and Dr. Hser was supported by NIDA Senior Scientist award (K05DA017648-01A2). The content of this publication does not necessarily reflect the views or policies of NIDA. The authors wish to thank the administrators from the five participating counties (Kern: Lily Alvarez, Allen Belluomini, Etta Robin; Riverside: Frank Lewis, Al Bell, Maria Lozano; Sacramento: Toni Moore, Sharon DiPirro-Beard, Jessica Vierra; San Diego: Al Medina, Linda Bridgeman-Smith, Susan Bower; San Francisco: Tom Hagan, Michael Ford, Craig Murdock) for supplying information and supporting the study.
Footnotes
Generally speaking, level 1 refers to education, level 2 to outpatient treatment, level 3 to intensive outpatient or day treatment, and level 4 to residential treatment. Considerable county-level variation exists with regard to the content of services within each level.
Contributor Information
Noosha Niv, Department of Psychiatry and Biobehavioral Sciences, UCLA Integrated Substance Abuse Programs, 1640 South Sepulveda Blvd., Suite 200, Los Angeles, CA 90025, USA. Phone: +1-310-5287554. Fax: +1-310-4737885. noosha23@yahoo.com.
Alison Hamilton, Department of Psychiatry and Biobehavioral Sciences, UCLA Integrated Substance Abuse Programs, Los Angeles, CA, USA. Phone: +1-310-2675421. Fax: +1-310-4737885. alisonh@ucla.edu.
Yih-Ing Hser, Department of Psychiatry and Biobehavioral Sciences, UCLA Integrated Substance Abuse Programs, Los Angeles, CA, USA. Phone: +1-310-2675388. Fax: +1-310-4737885. yhser@ucla.edu.
References
- 1.Hser YI, Maglione M, Joshi V, et al. Effects of treatment program and client characteristics on client treatment retention Final report for the Center for Substance Abuse Treatment Office of Scientific Analysis and Evaluation Scientific Branch. Los Angeles, CA: University of California, Drug Abuse Research Center; 1998. [Google Scholar]
- 2.Hser YI, Teruya C, Brown AH, et al. Impact of California’s proposition 36 on the drug treatment system: Treatment capacity and displacement. The American Journal of Public Health. 2007;97:104–109. doi: 10.2105/AJPH.2005.069336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Urada D, Hawken A, Anglin MD, et al. Evaluation of the Substance Abuse and Crime Prevention Act: Final Report. [May 10, 2008]; Available at: http://www.uclaisap.org/Prop36/documents/SACPAEvaluationReport.pdf.
- 4.Hser YI, Teruya C, Evans EA, et al. Treating drug-abusing offenders. Initial findings from a five-county study on the impact of California’s proposition 36 on the treatment system and patient outcomes. Evaluation Review. 2003;27:479–505. doi: 10.1177/0193841X03255774. [DOI] [PubMed] [Google Scholar]
- 5.Craddock SG, Rounds-Bryant JL, Flynn PM, et al. Characteristics and pretreatment behaviors of clients entering drug abuse treatment: 1969 to 1993. American Journal of Drug and Alcohol Abuse. 1997;23(1):55–73. doi: 10.3109/00952999709001686. [DOI] [PubMed] [Google Scholar]
- 6.Farabee D, Prendergast M, Anglin MD. The effectiveness of coerced treatment for dug-abusing offenders. Federal Probation. 1998;62:3–10. [Google Scholar]
- 7.Marlowe DB. Integrating substance abuse treatment and criminal justice supervision. Science & Practice Perspectives. 2003;2:4–14. doi: 10.1151/spp03214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Speiglman R, Klein D, Miller R, et al. Early implementation of proposition 36: Criminal justice and treatment system issues in eight counties. Journal of Psychoactive Drugs. 2003;35:133–141. doi: 10.1080/02791072.2003.10400508. [DOI] [PubMed] [Google Scholar]
- 9.Taxman FS, Bouffard JA. The importance of systems issues in improving offender outcomes: Critical elements of treatment integrity. Justice Research & Policy. 2000;2:9–30. [Google Scholar]
- 10.Taxman FS, Bouffard JA. Treatment inside the drug treatment court: The who, what, where, and how of treatment services. Substance Use & Misuse. 2002;37:1665–1688. doi: 10.1081/ja-120014426. [DOI] [PubMed] [Google Scholar]
- 11.Klein D, Miller RE, Noble A, et al. Incorporating a public health approach in drug law: Lessons from local expansion of treatment capacity and access under California’s proposition 36. The Milbank Quarterly. 2004;82:723–757. doi: 10.1111/j.0887-378X.2004.00329.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gerstein DRE, Harwood HJE, editors. Treating Drug Problems, Volume 1: A Study of the Evolution, Effectiveness, and Financing of Public and Private Drug Treatment Systems. Washington, DC: National Academy Press; 1990. [PubMed] [Google Scholar]
- 13.Ettner S, Huang D, Evans E, et al. Benefit-cost in the California treatment outcome project: Does substance abuse treatment “pay for itself”? Health Services Research. 2006;41:192–213. doi: 10.1111/j.1475-6773.2005.00466.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sherman RK, Irvin LK, Yovanoff P, et al. federal demonstration funds at the interface of mental health and criminal justice. Administration and Policy in Mental Health. 2004;32:17–29. doi: 10.1023/b:apih.0000039660.45153.3d. [DOI] [PubMed] [Google Scholar]
- 15.Taxman FS, Bouffard JA. Substance abuse counselors’ treatment philosophy and the content of treatment services provided to offenders in drug court programs. Journal of Substance Abuse Treatment. 2003;25:75–84. doi: 10.1016/s0740-5472(03)00115-6. [DOI] [PubMed] [Google Scholar]
- 16.Barreira P, Espey B, Fishbein R, et al. Linking substance abuse and serious mental illness service delivery systems: Initiating a statewide collaborative. Journal of Behavioral Health Services Research. 2000;27:107–113. doi: 10.1007/BF02287808. [DOI] [PubMed] [Google Scholar]
- 17.Grella CE, Gil-Rivas V, Cooper L. Perceptions of mental health and substance abuse program administrators and staff on service delivery to persons with co-occurring substance abuse and mental disorders. Journal of Behavioral Health Services Research. 2004;31:38–49. doi: 10.1007/BF02287337. [DOI] [PubMed] [Google Scholar]
- 18.Grudzinskas AJ, Jr, Clayfield JC, Roy-Bujnowski K, Fisher WH, et al. Integrating the criminal justice system into mental health service delivery: The Worcester diversion experience. Behavioral Sciences & the Law. 2005;23:277–293. doi: 10.1002/bsl.648. [DOI] [PubMed] [Google Scholar]