Drug and Alcohol Dependence

Transcription

1 Drug and Alcohol Dependence 103S (2009) S54 S64 Contents lists available at ScienceDirect Drug and Alcohol Dependence journal homepage: Measuring collaboration and integration activities in criminal justice and substance abuse treatment agencies Bennett W. Fletcher a,, Wayne E.K. Lehman a, Harry K. Wexler b, Gerald Melnick b, Faye S. Taxman c, Douglas W. Young d a National Institute on Drug Abuse, National Institutes of Health, 6001 Executive Boulevard, Bethesda, MD 20892, United States b National Development and Research Institutes, Inc., New York, NY 10010, United States c George Mason University, Manassas, VA 20110, United States d Bureau of Governmental Research, University of Maryland, College Park, MD 20740, United States article info abstract Article history: Received 21 March 2008 Received in revised form 19 December 2008 Accepted 5 January 2009 Available online 23 Febuary 2009 Keywords: Interorganizational relationships Systems integration Criminal justice Cross-agency collaboration Substance abuse treatment Individuals with substance abuse problems who are involved in the criminal justice system frequently need community-based drug and alcohol abuse treatment and other services. To reduce the risk of relapse to illicit drugs and criminal recidivism, criminal justice agencies may need to establish collaborations with substance abuse treatment and other community-based service providers. Although there are many variations of interorganizational relationships, the nature of these interagency collaborations among justice agencies and treatment providers has received little systematic study. As a first step, we present an instrument to measure interagency collaboration and integration activities using items in the National Criminal Justice Treatment Practices Surveys conducted as part of the Criminal Justice Drug Abuse Treatment Studies (CJ-DATS). Collaboration and integration activities related to drug-involved offenders were examined between substance abuse treatment providers, correctional agencies, and the judiciary. The measurement scale reliably identified two levels of collaboration: less structured, informal networking and coordination and more structured and formalized levels of cooperation and collaboration. An illustration of the use of the systems integration tool is presented. Published by Elsevier Ireland Ltd. 1. Introduction The need for substance abuse treatment services in the criminal justice system is well established. Over 7 million adults were under correctional supervision in the United States in 2005 (Glaze and Bonczar, 2006). Many of these have substance abuse or addiction disorders. In 2004, 45% of federal and 53% of state inmates met criteria for drug abuse or dependence (exclusive of alcohol problems), and over 60% of inmates had used illicit drugs regularly (Mumola and Karberg, 2006). Slightly over 36% of admissions to publicly funded substance abuse treatment in 2005 were referred by criminal justice sources, mostly probation/parole officers (Substance Abuse and Mental Health Services Administration, 2007: Tables 3.5 and 3.12). Fewer offenders receive treatment than need it. Mumola and Karberg (2006) report that only 15 17% of inmates meeting abuse or dependence criteria had received substance abuse treatment in prison. Recent estimates indicate that treatment is available to fewer than 10% of offenders in correctional settings on a daily Corresponding author. Tel.: ; fax: address: bfletche@nida.nih.gov (B.W. Fletcher). basis (Taxman et al., 2007a). Perhaps not surprisingly, there is a high failure rate for offenders returning to their communities after incarceration. Langan and Levin (2002) reported that within three years of release, 67% of drug offenders were rearrested for a new offense, 47% were reconvicted for a new crime, and about 49% were back in prison serving a new sentence or on a technical violation of release requirements. Substance abuse is a robust predictor of recidivism (Belenko, 2006; Bonta et al., 1998; Dowden and Brown, 2002). Efforts to integrate substance abuse treatment with criminal justice have a long history, beginning with the compulsory treatment of heroin addiction in 1930s-era Federal narcotics farms. More recently, partnerships between criminal justice and substance abuse treatment contribute to such as Treatment Alternatives to Street Crime (TASC, now the national Treatment Accountability for Safer Communities organization) (Treatment Accountability for Safer Communities, 2007; Wenzel et al., 2001), rehabilitation supervision (Paparozzi and Gendreau, 2005; Bonta et al., 2000), treatment alternatives to incarceration (Broner et al., 2003; O Callaghan et al., 2004), prison-based treatment programming (Inciardi et al., 2004; Welsh and Zajac, 2004), drug treatment courts (Turner et al., 2002), seamless probation combined with drug treatment (Alemi et al., 2006), weed and seed /$ see front matter. Published by Elsevier Ireland Ltd. doi: /j.drugalcdep

2 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 S55 initiatives (Office of Justice Programs, 2005), and treatment prisons intended for offenders with drug problems (Olson et al., 2004; Welsh and McGrain, 2008). These initiatives reflect the premise described in the landmark report The Challenge of Crime in a Free Society (President s Commission on Law Enforcement and Administration of Justice, 1967), which proposed that the reintegration of offenders into the community required coordination, collaboration, and partnerships with community agencies. For offenders with substance abuse and addiction problems, this suggests that criminal justice and substance abuse treatment should work together to provide effective treatment services that give the individual the best chance to abstain from illicit drug use and end criminal behavior. Nevertheless, the existing criminal justice system is often characterized as fragmented, with poor coordination between the judiciary and correctional institutions, between jails, prisons, and community supervision, and between health services and criminal justice agencies (Freudenberg, 2001; Veysey et al., 1997). Much attention has been given to reducing the organizational and systemic service delivery barriers that may contribute to reentry failure. Steadman (1992) proposed that criminal justice agencies dealing with individuals whose needs exceed the agency s capabilities should be able to reach across their organizational boundaries to coordinate with other agencies that can provide the needed resources or expertise. Criminal justice and treatment service integration strategies based on standardized risk and assessment tools, using incentives and sanctions, and drug testing have been recommended (Farabee et al., 1999; Taxman, 1998; Taxman and Bouffard, 2000; Wenzel et al., 2001). Still, there are many missed opportunities for cross-agency coordination and collaboration in assessing need for substance abuse and mental health treatment and linkage to services, in planning transitional services, in allocating treatment resources to the drug-involved offender, and in linking to community-based medical care for HIV and other infectious disease (Duffee and Carlson, 1996; Hammett et al., 1998; Robillard et al., 2003; Taxman and Bouffard, 2000; Taxman et al., 2007b). Despite widespread recognition of the potential benefit of collaborative efforts, there have been few systematic efforts to study organizational models that might be useful for guiding the integration of criminal justice requirements with drug abuse treatment. In 2002, the National Institute on Drug Abuse (NIDA) began a major research program that is the focus of several of the studies in the current volume. A primary objective of this research program, the Criminal Justice Drug Abuse Treatment Studies (CJ-DATS), is to improve outcomes for offenders with substance use disorders by improving the integration of substance abuse with other public health and public safety systems. In this context, the present study develops a tool to measure the levels of interorganizational activities that occur, a first step in exploring and characterizing the types of interorganizational relationships that exist between criminal justice agencies and substance abuse treatment providers across the nation. The interagency activity measure was developed using data collected through the CJ-DATS National Criminal Justice Treatment Practice Surveys (NCJTPS; Taxman et al., 2007b). This interagency activity measure is based in part on a framework developed by Konrad (1996), described below, to describe a continuum of levels of systems integration activities across agencies. Analyses are presented to show the fit of the measure to the Konrad model, how the activities are organized, and how frequently organizations engage in the different activities. These analyses will help establish the potential usefulness of the measure in terms of describing collaboration activities and for further analyses of organizational factors that relate to more integrated collaboration efforts Cross-agency collaboration efforts Relatively few studies have examined interorganizational factors related to substance abuse treatment in the criminal justice system. Apart from the drug treatment criminal justice nexus, however, there is a substantial body of work on systems and services integration efforts. Early federal initiatives during the 1970s tended to focus on administrative-level systems integration efforts, such as interagency agreements, co-location of services, centralized intake and assessment, new co-funding strategies, administrative coordination or consolidation, and shared management information systems. Major obstacles to successful system-level integrations were encountered, including size and complexity of the systems; bureaucratization and specialization contributing to organizational silos; difficulties of integration itself; and a lack of knowledge of how integration might best be accomplished (Kusserow, 1991). In his review of 20 years of systems and services integration efforts, Kusserow (1991) concluded that the substantial efforts made over that time had limited or inconclusive institutional impact. Later efforts concentrated on services integration strategies such as case management, case conferences and case review panels, individualized assessments and services plans, case monitoring and outcome monitoring, and giving the service provider more control over resources (Kahn and Kamerman, 1992). Both systems integration and services integration can be effective in improving outcomes for individuals with multiple needs. Friedmann and his colleagues examined a type of systems integration, how drug treatment providers linked their patients to other service providers. Formal referral linkages were more important than informal linkages in getting drug treatment patients to other service providers (Friedmann et al., 2001b), but providing transportation was even more effective (Friedmann et al., 2001a). One major five-year demonstration effort which used a quasi-experimental design to implement and evaluate systems integration for agencies in nine cities serving homeless individuals with substance use and mental disorders measured systems level outcomes and adherence to study aims as well as individual outcomes. These investigators found that although their attempts were at least partially successful in achieving the system-level aims of improved access to a wider range of services (Morrissey et al., 2002), there was limited impact on the outcomes of most interest, namely, improvement in the quality of life of the clients served by these agencies. This was attributed in part to an inadequate base of resources that could be linked together (Dennis et al., 2000). It was also found that many agencies which had successfully implemented integrated systems and services (such as integrated housing and support systems) abandoned these efforts following the end of the five-year project. Integration efforts which were sustained generally had agency staff who believed in systems integration and who had the time and ability to network (Steadman et al., 2002). Taxman and Bouffard (2000, 2002) evaluated an organizational boundary-spanning services integration strategy to build a seamless system of care between jail-based substance abuse treatment and community-based treatment for offenders funded through the U.S. Department of Justice Residential Substance Abuse Treatment (RSAT) block grant program. They found that despite formal agreements to coordinate between jails, parole and probation agencies, and local public health agencies, the operational practices needed to transcend interorganizational boundaries were not well implemented in the six jurisdictions they examined. With one exception, the sites transitioned fewer than 15% of their clients to communitybased treatment. Most efforts were placed into providing clinical services rather than in creating processes that bridged organizational boundaries. The drug court model posits that judicial supervision coordinated with comprehensive substance abuse treatment and other

3 S56 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 services can arrest criminal behavior linked to abuse of alcohol and drugs. Collaborative linkages between the drug court and treatment providers are a key component of the drug court model (Drug Courts Standards Committee, 1997). Wenzel et al. (2004) evaluated collaborative linkages and their barriers in 14 drug courts across the United States. The 11 dimensions they measured were (1) accommodation of the other organization s practice standards; (2) use of case management services; (3) staff cross training; (4) formalization of interagency agreements; (5) resource sharing; (6) joint assessment of clients; (7) joint planning of client service goals; (8) client referrals; (9) mutual sensitivity to concerns of the other organization; (10) sharing of information about clients; and (11) staff meetings. Wenzel et al. (2004) reported that linkages between drug courts and treatment providers were relatively strong across most dimensions, and that drug court administrators and treatment providers agreed on the nature of linkages and barriers to them. Resource sharing was least likely to be integrated across systems; however, providers indicated that enhanced communication would help most to strengthen drug court-treatment linkages. Case management is another services integration strategy that can be effective in helping clients obtain access to needed services as well as in improving retention and reducing criminal behavior (Rapp and Goscha, 2004; Rapp et al., 1998; Siegal et al., 2002; Vanderplasschen et al., 2004). Case managers can also be effective boundary spanners (Dvoskin and Steadman, 1994). Boundary spanners individuals who link with and coordinate services both within and outside their own agencies are recognized as an important element in the successful linkage between criminal justice and mental health treatment (Grudzinskas et al., 2005; Steadman, 1992) and in criminal justice and reentry programming (Byrne et al., 2002; Pettus and Severson, 2006; Taxman and Bouffard, 2000) Models of cross-agency collaboration In general, agencies form collaborations in order to enhance each other s capacity for mutual benefit and to achieve a common goal (Himmelman, 2001). Much work has been carried out conceptualizing the dynamics of interorganizational relationships. With regard to why such relationships are formed and how they are sustained, Huxham (1996) proposed that interorganizational relationships are created when they provide collaborative advantages when the product of agencies working together is greater than could be achieved independently. This advantage may occur when organizational partnerships provide resources, expertise, or leadership to achieve cross-cutting goals (Huxham and Vangen, 2000; Lasker et al., 2001; Mitchell and Shortell, 2000; Taxman, 1998; Weiss et al., 2002). Agencies may be motivated to collaborate in order to provide needed services, to reduce costs by sharing resources, or to improve efficiency or cost-effectiveness (Rivard and Morrissey, 2003), or because such collaboration provides opportunities to improve agency or professional standing (Wells et al., 2005). Wells et al. (2005) also note that interorganizational relationships develop or evolve through stages of formation, implementation, and maintenance over time. Bolland and Wilson (1994) and Heflinger (1996) stress the functional roles of service delivery, administrative functions, and planning activities in interagency coordination networks. The development of interorganizational relationships can involve multiple organizational dimensions and can differ in both level and intensity of effort. Konrad (1996) proposed a hierarchical services integration framework describing levels of integration activities ranging from informal, less structured activities to more formal and more highly structured ones. In her model, cross-agency activities may be focused in various service delivery, administrative, and planning areas, and the level of integration may vary in each of these. Informal activities are less likely to be guided by established agreements, protocols, or procedures; they are usually less frequent and may occur on an as-needed basis. At higher levels of formalization, some activities may become officially sanctioned and governed by written agreements that specify rules, boundaries, goals, and activities. Higher levels of formalization tend to encompass less structured activities as well. Konrad conceptualized systems integration activities on a continuum ranging from informal, less structured activities to more formalized, more structured activities; she described five anchor points along this integration continuum: information sharing/communication (less structured), cooperation and coordination, collaboration, consolidation, and fully integrated (more structured). 1. Information sharing and communication: Relationships between agencies are not formally structured. Agency representatives may share general information about, services, and clients. Communications may be less frequent or ad hoc. Activities may include sharing informational brochures, educational presentations, newsletters, or joint staff meetings. 2. Cooperation and coordination: Cross-agency activities are somewhat more structured. Agencies may work together to change procedures or structures to help make more successful. Activities may include reciprocal client referrals and follow-up processes, verbal agreements to hold joint staff meetings, mutual agreements to provide priority responses, or joint lobbying for legislative change or funding requests. 3. Collaboration: Although temporary or brief collaboration can operate informally, ongoing collaborations are usually more structured. Autonomous agencies and work together with a common goal, product, or outcome. Examples include partnerships with written agreements, goals, formalized operational procedures, and possibly joint funding, staff cross training, or shared information systems. 4. Consolidation: Consolidated systems may be those under an umbrella organization or those with some centralized functions (e.g., program or financial administration). Line authority for or services is contained within different divisions or agencies. Cross-program collaboration, coordination, cooperation, and information sharing are more frequent and often more structured activities. An example might be a government agency with responsibility for different human service. 5. Integration: A fully integrated system has a single authority that is comprehensive in scope, operates collectively, addresses client needs in an individualized fashion, and is multi-purpose and cross-cutting. Categorical lines are transparent with fully blended activities and pooled funding. The client perceives service delivery as seamless, with little or no organizational barriers to access. An example might be a one-stop agency with unified intake and assessment, case management and many services provided in one location. Management and operational decisions are the responsibility of a single entity. A similar framework was proposed by Himmelman (1996, 2001). Horwath and Morrison (2007) adopted Konrad s (1996) model to examine critical elements of effective collaborative efforts for improving services for vulnerable children. Messeri et al. (2003) expanded the Konrad framework to include Kagan and Neville s (1993) locus of integration activity (client-, program-, policy-, or funding-centered) and to incorporate service and systems integration strategies employed by Dennis et al. (2000). Messeri and his colleagues used this expanded model to characterize the success of service integration efforts in HIV service networks among 33 recipients of HIV/AIDS service delivery demonstration grants awarded in 20 states in They found that agencies attempted to implement both systems and services integration activities, but system-level activities such as implementing written agreements

4 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 S57 Table 1 Factor loadings for common activities of corrections with substance abuse treatment and of substance abuse treatment with corrections. Corrections re-drug treatment (N = 430) Drug treatment re-corrections (N =217) Factor Factor Less structured cooperation and coordination a. We share information on offender treatment services We share information on offender needs for treatment services b. Our organizations employ similar requirements for program eligibility Our organizations have agreed to similar requirements for program eligibility for some c. We have written agreements providing space for substance abuse services We have written agreements providing space for substance abuse services for some d. We hold joint staffings/case reporting consultations e. We have modified some program/service protocols to meet the needs of each agency We coordinate policies and procedures to accommodate each other s requirements f. We hold joint staff meetings g. We have written protocols for sharing offender information We have written protocols for sharing client information More structured collaboration and consolidation h. We have developed joint policy and procedure manuals i. Our organizations have pooled funding for some offender substance abuse services Our organizations have pooled funding for offender substance abuse services j. We share budgetary oversight of some treatment We share budgetary oversight of treatment k. We share operational oversight of some treatment We share operational oversight of treatment l. Our organizations cross-train staff on substance abuse issues Notes: Language differs slightly between the correctional facility director s survey and treatment program director s survey for some items. Where there are two versions, the first version is the corrections survey and the second version is the treatment director s survey. Item f appears only in the program directors survey. or creating formal linkage arrangements were more likely to be abandoned or not fully implemented. Initiatives to integrate services and systems have often been undertaken with the assumption that more integration is better (see, for example, Hill and Lynn, 2003; Randolph et al., 2002), but this may not always be the case (Provan and Milward, 2001). Himmelman (2001) points out that collaborative strategies exist along a continuum, and those employed reflect both the needs of the agencies and the current status of the agencies interorganizational relationships. Guidance has been published on how to establish interorganizational relationships, but surprisingly little research has been conducted on measuring existing services or systems integration activities. A brief, reliable instrument to measure the level of interagency collaborative activities is needed to provide a better understanding of the nature of existing, naturally evolving interrelationships and how they are in turn related to other organizational factors. The goal of this study is to develop and evaluate a measurement tool for assessing the level of services and systems integration activities in criminal justice and substance abuse treatment agencies. 2. Method The data used in this study were collected under the National Criminal Justice Treatment Practices Survey (NCJTPS), a nationally representative survey of organizational characteristics and substance abuse treatment practices for offenders with drug problems. Respondents from four organizational levels were surveyed; these levels were a census of state public safety agency directors (designated S1), a census of directors of state substance abuse treatment agencies and administrators of adult and juvenile correctional agencies with offender substance abuse treatment oversight (S2), a sample of directors and administrators of correctional facilities, parole/probation offices, and outpatient treatment serving criminal/juvenile justice clients (S3A and S3B); and a sample of line staff from these same facilities and offices (S4). For the present study, we employed items from S3A, the survey of correctional facility administrators, and S3B, the treatment program directors survey A detailed description of the NCJTPS survey methodology, sampling plan, and samples is provided by Taxman et al. (2007b) Development of the interagency activity measure We developed measures of collaboration and integration for inclusion in the surveys of correctional facility administrators (S3A) and treatment program directors (S3B). We characterize collaboration and integration activities as those activities that describe how an agency interacts with another agency when they have overlapping target populations. These activities can range on a continuum from few or none to

5 S58 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 an extensive and formalized set of activities in multiple domains. Activities that are incorporated into an agency s mission, staffing, or operational planning are expected to reflect a higher level of systems integration than the informal or less structured activities that agency staff initiate. Models that informed the items for the integration scale were the triad of networking, coordination, and collaboration developed by Carter et al. (2002); the continuum of informal to formal integration across 12 dimensions (partners, target population, goals, policy and legislation, governance and authority, service delivery, stakeholders, planning and budgeting, financing, outcomes and accountability, licensing and contracting, and information systems and data management) developed by Konrad (1996); the drug court integration model tested by Wenzel et al. (2001, 2004); and boundaryless system models focusing on policies, procedures, and operations across the main programming areas of corrections (assessment, treatment placement, planning, services, compliance management, and outcomes) described by Taxman (1998) and Taxman and Bouffard (2000). Based on these models, we developed a series of questions to measure the level of interagency integration. Organizational centralization, defined as having a single agency responsible for planning, administering, and delivering both correctional services and substance abuse treatment services, represents the most unified end of the integration continuum described by Konrad (1996). There are few examples of this type of arrangement in the criminal justice system. In most jurisdictions corrections and drug treatment agencies are separate entities. Hence, our construction of a measure of integration for criminal justice and drug treatment delivery systems does not include fully integrated systems. Since organizational integration is usually a by-product of multiple agencies working together, it is possible that there are as many forms of collaboration or coordination as there are stakeholder agencies with shared, cross-cutting goals. Our measure of integration elicited information about activities involving the four major stakeholder agencies working with drug-involved offenders: drug treatment agencies, the judiciary, correctional institutions, and probation or parole agencies. Survey respondents were asked if they engaged in each of 11 integration activities with the other types of organizations (Table 1) Questionnaire item identification A literature review was conducted to define features of specialized correctional drug treatment and to identify important factors in providing services to offenders with substance use disorders. This review was followed by a number of informal discussions with executives and staff from one state agency and with members of several professional organizations representing criminal justice interests, particularly those involving alternatives for drug-involved offenders. These discussions confirmed that coordination of services and collaboration with other criminal justice and treatment agencies were considered to be important programming elements for drug-involved offenders. Important areas for coordination included sharing basic information through formal or informal networking or information systems; procedures to identify eligible clients; sharing resources (including space, funds, training, and staff); and modifying services for better coordination across multiple agencies. We developed our survey items on target population, administration, service delivery, stakeholders, budgeting, information sharing, planning, and service delivery based on the literature and our discussions with corrections professionals Scaling principles Survey respondents in a given agency (drug treatment, judicial, correctional institution, or probation/parole) indicated whether they engaged in activities with each of the three other types of agencies potentially involved in offender drug treatment. The interagency activity measure was designed as an additive scale with 11 items. The items were developed with the intent of distinguishing activities related to the Konrad (1996) integration continuum (ranging from informal, less structured information sharing, communication, and cooperative activities to increasingly formalized and structured activities of coordination, collaboration, and consolidation) based on analyses of survey data Pretesting of the questionnaire items The interagency activity measure was developed collaboratively by members of the CJ-DATS Research Centers and NIDA scientific officers. We asked correctional and treatment program managers in a state county to review this measure as part of our ongoing discussions with them. Based on these reviews, a dichotomous item format was adopted for the final survey versions. The correctional facility administrators (S3A) survey included a list of 11 items representing common activities that can occur between agencies. Respondents were asked to check all activities that apply to their working relationship with substance abuse treatment, the judiciary, and correctional agencies on issues specific to offender substance abuse treatment. Community corrections respondents were asked to indicate their activities with institutional corrections, and jail/prison respondents were asked to indicate their activities with community corrections. For the analyses reported in this paper, we combined community corrections reports of activities shared with institutional corrections with institutional corrections reports of activities shared with community corrections. The items are shown in Table 1. The treatment program directors survey (S3B) included a list of 12 items representing common activities that can occur between agencies. As shown in Table 1, 11 of the items were essentially the same as those in the S3A survey with some minor wording differences. In addition, the treatment program directors survey had one item not included in the correctional facility administrators survey: We hold joint staff meetings. Treatment program directors were asked to check each of the 12 activities that apply to their working relationships with the judiciary and with community corrections Samples Two distinct samples were examined. The first was corrections facility directors survey respondents (S3A), comprised of 430 corrections facility administrators representing adult prisons (N = 98), adult jails (N = 57), adult community corrections agencies (N = 134), juvenile secure residential facilities (N = 49), local juvenile jails (N = 32), and community corrections agencies (N = 60). The second sample was treatment program directors survey (S3B) respondents, and included 217 substance abuse treatment directors (122 from county-level facilities and 95 from prison-based ). The response rates for S3A and S3B were 64.8% and 54.5% respectively. While low, these response rates are typical for mailed organizational surveys. Nonresponse patterns were examined to determine whether they were associated with organizational factors such as organizational climate, resource need, and leadership style (Taxman et al., 2007b). These analyses found no systematic differences between survey respondents and non-respondents. Taxman et al. (2007b) describes survey methodology and procedures to reduce response bias Analyses A primary question of interest was whether the factor structure of the survey items in the interagency activity measures would support the Konrad (1996) model. The 11 items from S3A, the correctional facility administrators survey, were factor analyzed separately for substance abuse treatment, judiciary, and other corrections. Likewise, the 12 items from S3B, the treatment program directors survey, were factor analyzed separately for the judiciary and community corrections. The percentage of respondents participating in the different activities was then examined for the activities that were represented in the factors suggested by the factor analysis. Our model is posited to be hierarchical in the sense that lower levels of integration are characterized by having less structured interagency activities, while higher levels of integration between the correctional and treatment agencies should have more structured or formalized arrangements between agencies. Participation rates were examined across items to examine the hierarchical nature of the model. Although it might seem intuitive that more activities would reflect a higher level of integration, this is not necessarily so. Integration may occur among multiple dimensions of organizational function, such as organizational goals, service implementation, planning and budgeting, client outcomes, information systems, and data management. A large number of informal interagency activities across several functional areas might reflect a lower level of integration than formal, highly structured activities in a single domain. However, the NCJTPS surveys do not allow us to discriminate between levels of activities by domain. Under this constraint, we expected the number of activities (and not just their degree of structure) to be related to the level of interagency integration, and we counted the total number of checked activities. The distributions of interagency activity levels and the relationships between level of integration and participation in activities representing lower and higher levels of integration were examined. 3. Results 3.1. Factor analysis We conducted exploratory factor analyses (SAS Proc Factor) using tetrachoric correlations to account for the discrete dichotomous activity items (Knol and Berger, 1991). Table 1 shows orthogonal rotated factor loadings for the corrections administrators and treatment administrators, respectively. Factor analyses were also conducted for the activities with judicial and other corrections agencies. Factors for these analyses (not shown) were very similar to those for substance abuse treatment. For corrections administrators, two factors emerged. Table 1 shows loadings for the factors for activities with Substance Abuse Treatment. The first factor included 6 of the 11 items represented the less structured items at the levels of Cooperation and Coordination. The first item, We share information on offender needs for treatment services, reflects informal networking activities represented in Konrad s information sharing and communication level. Other items included in the first factor were agreeing to similar requirements for program eligibility, written agreements for pro-

6 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 S59 Table 2 Percent of shared activities of corrections with substance abuse treatment (SA), judiciary, and jail/prison or community corrections and of substance abuse treatment with judiciary and community corrections. SA Corrections re-judiciary corrections (N = 430) Drug treatment re-judiciary corrections (N =217) Low structure cooperation and coordination a. We share information on offender treatment services 57% 46% 66% 58% 59% We share information on offender needs for treatment services b. Our organizations employ similar requirements for program 44% 23% 43% 24% 34% eligibility Our organizations have agreed to similar requirements for program eligibility for some c. We have written agreements providing space for substance 40% 11% 31% 21% 34% abuse services We have written agreements providing space for substance abuse services for some d. We hold joint staffings/case reporting consultations 38% 15% 33% 23% 34% e. We have modified some program/service protocols to meet 30% 13% 30% 22% 31% the needs of each agency We coordinate policies and procedures to accommodate each other s requirements f. We hold joint staff meetings 12% 19% g. We have written protocols for sharing offender information 44% 24% 50% 41% 45% We have written protocols for sharing client information High structure collaboration and consolidation h. We have developed joint policy and procedure manuals 17% 6% 22% 10% 16% i. Our organizations have pooled funding for some offender 22% 7% 21% 11% 12% substance abuse services Our organizations have pooled funding for offender substance abuse services j. We share budgetary oversight of some treatment 19% 4% 17% 7% 11% We share budgetary oversight of treatment k. We share operational oversight of some treatment 23% 7% 21% 7% 16% We share operational oversight of treatment l. Our organizations cross-train staff on substance abuse issues 30% 7% 26% 14% 25% Notes: Language differs slightly between the correctional facility director s survey and treatment program director s survey for some items. Where there are two versions, the first version is the corrections survey and the second version is the treatment director s survey. Item f appears only in the program directors survey. viding space for substance abuse services, holding joint staffings, modifying some program protocols to meet needs of each agency and having written protocols for sharing offender information. Although the correspondence is not unequivocal, these activities roughly represent support for common objectives, coordinated services, and communication components of the Coordination level of the Konrad model and typically represent a lower level of interagency integration activities. We labeled this factor Low Structure. The second factor, labeled High Structure, tended to represent the more structured collaboration and consolidation levels of activities. The collaboration level includes sharing of resources and activities, formalized agreements, and joint goal setting. The consolidation level includes administrative functions overseen by separate administrative units. Activities loading on the second factor included developing joint policy and procedure manuals, pooled funding for some services, sharing budgetary and operational oversight of some, and cross-training staff on substance abuse issues. Factor loadings for the 12 items for treatment program directors are also shown in Table 1. The two factors that emerged were very similar to those for correctional facility administrators. The results shown in Table 1 are for activities applying to working relationships with community corrections administrators. Results for the judiciary (not shown) were similar. One item for treatment program administrators which was not asked of corrections administrators ( We hold joint staff meetings ) loaded on the first factor and is representative of collaboration activities. The major difference between corrections administrators and treatment administrators is for the item on cross-training staff on substance abuse issues, which loaded on the High Structure factor for corrections administrators and on the Low Structure factor for treatment staff administrators. However, for subsequent analyses, it was included with the High Structure factor since it more closely falls under collaborative activities as described by Konrad Activity frequencies Table 2 show the percent of respondents who reported each activity with substance abuse treatment agencies, judiciary, and other corrections agencies for corrections administrators and with judiciary and corrections agencies for treatment administrators. For corrections administrators, Table 2 shows that the activities that loaded on the Low Structure factor were reported more frequently than any of the activities that loaded on the High Structure factor. For each type of organization, sharing information was the most frequent cross-agency activity. Agencies reported less structured activities more frequently than the more structured activities. The results for treatment agency administrators shown in Table 2 were similar with only one exception. The most frequent activity was networking, with 58% of treatment administrators reporting sharing information with judiciary agencies and 59% with corrections agencies. Having less structured or informal activities was reported with greater frequency than conducting more structured activities, with one exception. Holding joint staff meetings, a less structured activity, was reported less frequently than any of the more structured activities.

7 S60 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 Table 3 Number of shared activities of corrections with substance abuse treatment (SA), judiciary, and jail/prison or community corrections and of substance abuse treatment with judiciary and community corrections. SA Judiciary Corrections Number of shared activities of corrections a with substance abuse treatment (SA), judiciary, and jail/prison or community corrections None (no activities) 37% 50% 24% 1 6 activities (Low Structure) 35% 43% 54% 7 11 activities (High Structure) 28% 7% 22% Number of shared activities of substance abuse treatment b with judiciary and community corrections None (no activities) 36% 30% 1 7 activities (Low Structure)) 57% 56% 8 12 activities (High Structure) 8% 14% a Based on 430 corrections survey respondents. b Based on 217 substance abuse survey respondents Total number of interagency integration activities Table 3 shows the total number of interagency activities reported by substance abuse treatment agencies, judiciary agencies, and other corrections agencies for corrections administrators and for judiciary and corrections agencies for treatment administrators. Our interagency integration model is hierarchical in nature, with participation more likely in informal networking, cooperation, and coordination activities than participation in more structured or more formalized interagency activities. Since the interagency activity items were indeed generally ordered along a continuum, respondents who checked more structured activities were also very likely to check most or all of the less structured activities. We were able to use the number of interagency activities checked to classify respondent agencies into Low Structure and High Structure levels. Based on the factor analyses, we defined Low Structure for corrections administrators as checking 1 to 6 activities. High Structure was defined as checking 7 or more activities. With an additional item in the scale for treatment administrators, 1 7 items defined the Low Structure level of interagency integration and checking 8 Table 4 Percent of shared activities by level of activity structure (Low vs. High) of Corrections Programs with Substance Abuse Treatment agencies (from facility directors survey), and by Substance Abuse Treatment with Community Corrections (from program directors survey). Corrections re-drug treatment (N = 430) Drug treatment re-corrections (N =217) Informal (1 6) Formal (7 11) Informal (1 7) Formal (8 12) Low structure cooperation and coordination a. We share information on offender treatment services 86% 95% 80% 100% We share information on offender needs for treatment services b. Our organizations employ similar requirements for 53% 89% 38% 90% program eligibility Our organizations have agreed to similar requirements for program eligibility for some c. We have written agreements providing space for 44% 86% 37% 90% substance abuse services We have written agreements providing space for substance abuse services for some d. We hold joint staffings/case reporting consultations 37% 89% 37% 90% e. We have modified some program/service protocols to 23% 80% 32% 90% meet the needs of each agency We coordinate policies and procedures to accommodate each other s requirements f. We hold joint staff meetings 14% 81% g. We have written protocols for sharing offender information We have written protocols for sharing client information 51% 92% 55% 100% High structure collaboration and consolidation h. We have developed joint policy and procedure manuals 6% 53% 11% 68% i. Our organizations have pooled funding for some offender substance abuse services Our organizations have pooled funding for offender substance abuse services j. We share budgetary oversight of some treatment We share budgetary oversight of treatment k. We share operational oversight of some treatment We share operational oversight of treatment l. Our organizations cross-train staff on substance abuse issues 14% 60% 3% 68% 7% 57% 3% 61% 12% 69% 8% 81% 23% 79% 25% 77% Notes: Language differs slightly between the correctional facility director s survey and treatment program director s survey for some items. Where there are two versions, the first version is the corrections survey and the second version is the treatment director s survey. Item f appears only in the program directors survey.

8 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 S61 or more items was categorized as High Structure. Since a substantial number of respondents reported no activities in common with other agencies, we also created a separate No Activities category. Table 3 shows the distribution for corrections administrators for integration activities with substance abuse agencies, judiciary agencies, and other corrections agencies. There was a relatively even distribution into No, Low, and High categories. It is notable that nearly a quarter of the corrections agencies reported no shared activities with their community corrections or institutional corrections counterparts, and even more (37%) reported no shared activities with substance abuse treatment. Half of corrections administrators reported no integration activities with judiciary agencies and 43% were classified as having a less structured level of integration with judiciary agencies. Only 7% of corrections administrators reported having a high level of integration activities with judiciary agencies. Table 3 also shows the classification into No, Low, and High levels of integration for substance abuse treatment administrators with judiciary agencies and corrections agencies. As with the correctional facility administrators, it is noteworthy that despite working with criminal justice-involved clients, 30% of substance abuse treatment reported having no activity not even sharing information on offender treatment needs with corrections agencies. More structured levels of integration were relatively rare between the judiciary and either corrections administrators (7%) or substance abuse treatment providers (8%). Fewer (14%) of substance abuse treatment program directors reported having more structured activities with corrections agencies than correctional facility administrators reported for drug treatment agencies (22%). Table 4 shows the percent participation in each of the interagency activities for corrections administrators and substance abuse treatment agencies classified by Low and High levels of structure in interagency activities. As shown in Table 4, corrections administrators classified as being at the Low level of interagency activity structure were very likely to report sharing information (86%) and from 23% to 53% reported carrying out other (different) less structured activities. As expected, participation by Low Structure administrators in High Structure activities was very low. Administrators classified as having High Structure had very high rates of participation in each of the less structured activities and moderately high rates of participation in each of the more structured activities. Results were similar for treatment administrators in terms of integration with corrections agencies (Table 4). Eighty percent of administrators classified as being in the Low Structure category participated in sharing information. Participation in other low structure activities ranged from 14% for holding joint staff meetings to 55% for having written protocols for sharing client information. Very few administrators in the Low Structure level participated in more structured activities. As expected, there was a high percentage of administrators in the High Structure category who engaged in the less structured activities. 4. Discussion This study was undertaken in order to develop a better understanding of the types and levels of interorganizational relationships that exist between drug abuse treatment providers and criminal justice agencies, including prisons, community corrections, and the judiciary. As a long-term objective, we hope to understand not just what these interagency activities are, but which ones are more likely to be effective in achieving their goals and how they are sustained over time. The first step, and the one reported on in this study, is to develop a measure that can be used to describe levels of interagency integration activities that currently exist between these agencies. The measure that we report on was based in part on the systems integration continuum described by Konrad (1996). Our analyses supported the integration measure and the hierarchical nature of our integration model. Administrators were much more likely to engage in less structured, more informal interagency integration activities such as information sharing, networking, and services coordination than in more structured types of interagency activities such as sharing funding, co-developing procedural manuals, or sharing budgetary or operational oversight. Factor analyses separated the NCJTPS items into two factors that matched closely to the levels of Low Structure Cooperation and Coordination, and High Structure Collaboration and Consolidation. Agencies that participated in 6 or fewer activities (classified as Low Structure) had very low participation in any of the high structure collaboration and consolidation activities. Programs that participated in 7 or more activities at the more structured levels were more likely to have participated in all of the less structured cooperation and coordination activities. The results reported here also show that for corrections administrators, most cross-agency activities at the High Structure level are more likely to occur with substance abuse treatment administrators and with other corrections agencies than with the judiciary. Results also suggested that treatment administrators were more likely to have a High Structure level of integration activities with corrections agencies than with judicial agencies. For both correctional facility administrators and treatment program directors, having a High Structure level of integration with the judiciary was relatively rare Limitations This paper describes the measurement of cross-system collaboration and integration activities based on reports by correctional program directors and substance abuse treatment providers. Response rates to the surveys (65% for correctional administrators and 55% for treatment directors), while typical for this type of survey, are low and the results may not be generalizable to other respondents. Similarly, there could be systematic bias in reporting activities across agencies. Administrators and directors may be more likely to know about formal activities than informal ones. Although underreporting of informal activities would not affect the hierarchical nature of the model, systematic underreporting of formal activities could affect the hierarchical structure of the measurement model and the relationships between informal levels of integration and other aspects of organizational functioning. Our factor analyses did not reproduce Konrad s four integration levels (omitting the fully integrated level), but we note that the NCJTPS items were not specifically written to represent the four levels. Item responses were dichotomous and only measured whether a given activity was part of the respondent s working relationships. We recognize that we have not measured the qualitative dimensions of collaborative activities. For example, sharing information, an activity we considered low structure, may be informal and cursory or systematic and comprehensive. Items that quantified the strength of a relationship or frequency of activities might have allowed finer differentiation into more integration levels; or it may be that more items are needed in the scale to be able to differentiate the four constructs. Our model implies that collaborative activities exist along a continuum from none to many, but we do not know whether participating in more activities produces better client outcomes. It may be that fewer activities produce better outcomes if the activities are carefully selected and well implemented. More research is needed to determine which dimensions of collaboration and integration are

9 S62 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 most important to improving shared goals across agencies, particularly those related to improved outcomes for the drug-involved offender. The National Criminal Justice Treatment Practices Survey is cross-sectional. It allows us to identify existing levels of activities occurring between agencies, but not whether or how such crossagency activities change over time, either toward or away from more integration of systems and services. Also, our survey items provide little about the nature of the specific collaborative activities themselves, such as how effective or useful they are, how frequently they occur, whether they are usual or exceptional, whether they are initiated by the correctional facility or by the substance abuse treatment provider, or what external factors (such as media attention, legislative or regulatory oversight, or budgetary considerations) may facilitate or hinder such activities. More research is needed to determine what factors might drive the reported activities. Nevertheless, while there is more work to be done in the area of cross-agency collaboration and integration, we believe that this study represents a useful first step in understanding how criminal justice and drug abuse treatment organizations interact and the nature of their shared activities Future research directions The framework for the present study was intended to represent the available range of integration activities from none to consolidation, but at any given point on this continuum there are a considerable number of activities that can better define the nature of the relationships between agencies. For example, within the range from coordination to collaboration, multiple partnerships and networking arrangements can exist. Collaborative activities may be more readily initiated by staff members to address client needs, but these arrangements may be difficult to sustain over time without an ongoing organizational commitment to them. Systems integration can require fundamental organizational change among multiple agencies and perhaps at multiple organizational levels. More research is needed to identify and describe these working relationships and how they contribute to outcomes at the individual, program, and system levels. Research is needed to determine how the level of systems integration affects access to or effectiveness of services for druginvolved offenders. Henderson et al. (2007) found that more formalized systems integration in state agencies was associated with greater use of evidence based practices in local corrections facilities. Service integration strategies such as case management, individualized treatment planning, or linking clients to other service providers usually do not require the fundamental changes in interagency relationships that characterize systems integration (Dennis et al., 2000). However, Dodds et al. (2004) note that achieving integrated services often requires system-level changes. The relationship between the level of integration and organizational factors in criminal justice and drug abuse treatment agencies is further examined by Lehman et al. (2009), who found that more structured or formalized integration was related to facility size, resource network, access to treatment services, and number of services provided to offenders. Our measure of integration encompasses multiple organizational domains, but we do not measure the level of integration within given domains (such as goal setting, information management, or budgeting). It may be that integration is more likely for some domains than others; there may also be a progression of integration activities across domains. For example, successful information sharing across agencies may require collaboration in the areas of goal setting, data management functions, and budgeting. An area for future research is to determine whether some integrative activities precede or predispose others, and whether or how these processes are related to the successful maintenance of collaborative activities Summary Integration of systems and services is often stated as an important organizational goal to improve outcomes for drug-involved offenders. The few integration efforts that have been systematically studied have produced mixed results. Given the record of previous systems and services integration efforts, it was thought necessary to establish a better understanding of how agencies work together before contemplating future efforts to increase the levels of systems or services integration. The first step in this effort is to develop a way to measure existing interorganizational relationships between criminal justice and substance abuse treatment agencies. Our results show that the systems integration activities rated in the correctional facility directors and treatment program directors surveys largely support the Konrad model. The factor analyses were consistent with Konrad s levels of integration constructs and the rate of participation in the activities support a hierarchical systems integration concept. These results support the use of an overall systems integration index that is a simple count of the number of integration activities an agency participates in, given that a low number of activities strongly suggest participation in informal integration activities and unlikely participation in activities suggesting formal levels of systems integration. The overall systems integration index is a measure that can be used for further study to examine organizational antecedents to systems integration as well as outcomes of systems integration. Role of funding source This study was funded under a cooperative agreement from the U.S. Department of Health and Human Services, National Institutes of Health, National Institute on Drug Abuse (NIH/NIDA). The corresponding author is the NIDA Program Scientist with CJ-DATS. Taxman and Young were funded by NIDA grant U01 DA16213, and Wexler and Melnick were funded by NIDA grant U01 DA The contents are solely the responsibility of the authors and do not necessarily represent the official views of the Department of Health and Human Services, NIH/NIDA, or other participants in CJ-DATS. Conflict of interest The authors have no conflicts of interest. Acknowledgements The authors gratefully acknowledge the collaborative contributions by federal staff from NIDA, members of the Coordinating Center (University of Maryland at College Park, Bureau of Governmental Research and George Mason University), and the nine Research Center grantees of the NIH/NIDA CJ-DATS Cooperative (Brown University, Lifespan Hospital; Connecticut Department of Mental Health and Addiction Services; National Development and Research Institutes, Inc., Center for Therapeutic Community Research; National Development and Research Institutes, Inc., Center for the Integration of Research and Practice; Texas Christian University, Institute of Behavioral Research; University of Delaware, Center for Drug and Alcohol Studies; University of Kentucky, Center on Drug and Alcohol Research; University of California at Los Angeles, Integrated Substance Abuse Programs; and University of Miami, Center for Treatment Research on Adolescent Drug Abuse). Contributors: Bennett Fletcher and Wayne Lehman designed the study. Harry Wexler and Gerald Melnick consulted on the study and

10 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 S63 provided critical input into the study design. Faye Taxman and Doug Young designed the survey items, fielded the national survey, and drafted the description of the items in this report. Bennett Fletcher reviewed the literature and wrote the first draft of the manuscript. Wayne Lehman conducted the data analyses and drafted the analyses section. All authors contributed to and have approved the final manuscript. References Alemi, F., Taxman, F., Baghi, H., Vang, J., Thanner, M., Doyon, V., Costs and benefits of combining probation and substance abuse treatment. J. Ment. Health Policy Econ. 9, Belenko, S., Assessing released inmates for substance-abuse-related service needs. Crime Delinq. 52, Bolland, J.M., Wilson, J.V., Three faces of integrative coordination a model of interorganizational relations in community-based health-and-human-services. Health Serv. Res. 29, Bonta, J., Law, M., Hanson, K., The prediction of criminal and violent recidivism among mentally disordered offenders: a meta-analysis. Psychol. Bull. 123, Bonta, J., Wallace-Capretta, S., Rooney, J., A quasi-experimental evaluation of an intensive rehabilitation supervision program. Crim. Justice Behav. 27, Broner, N., Nguyen, H., Swern, A., Goldfinger, S., Adapting a substance abuse court diversion model for felony offenders with co-occurring disorders: initial implementation. Psychiatr. Q 74, Byrne, J.M., Taxman, F.S., Young, D., Emerging roles and responsibilities in the Reentry Partnership Initiative: New ways of doing business. Department of Justice Report U.S. Department of Justice, Washington, DC. Carter, M., Ley, A., Steketee, M., Gavin, F., Stroker, R., Woodward, W., Collaboration: A Training Curriculum to Enhance the Effectiveness of Criminal Justice Teams. State Justice Institute, Washington, DC. Dennis, D.L., Steadman, H.J., Cocozza, J.J., The impact of federal systems integration initiatives on services for mentally ill homeless persons. Ment. Health Serv. Res. 2, Dodds, S., Nuehring, E.M., Blaney, N.T., Blakley, T., Lizzotte, J.M., Lopez, M., Potter, J.E., O Sullivan, M.J., Integrating mental health services into primary HIV care for women: The Whole Life project. Public Health Rep. 119, Dowden, C., Brown, S.L., The role of substance abuse factors in predicting recidivism: a meta-analysis. Psychol. Crime Law 8, Drug Court Standards Committee, Defining Drug Courts: The Key Components. Drug Courts Program Office, U.S. Department of Justice, Washington, DC. Duffee, D.E., Carlson, B.E., Competing value premises for the provision of drug treatment to probationers. Crime Delinq. 42, Dvoskin, J.A., Steadman, H.J., Using intensive case-management to reduce violence by mentally-ill persons in the community. Hosp. Community Psychiatry 45, Farabee, D., Prendergast, M., Cartier, J., Wexler, H., Knight, K., Anglin, M.D., Barriers to implementing effective correctional drug treatment. Prison J. 79, Friedmann, P.D., Lemon, S.C., Stein, M.D., 2001a. Transportation and retention in outpatient drug abuse treatment. J. Subst. Abuse Treat. 21, Friedmann, P.D., Lemon, S.C., Stein, M.D., Etheridge, R.M., D Aunno, T.A., 2001b. Linkage to medical services in the Drug Abuse Treatment Outcome Study. Med. Care 39, Freudenberg, N., Jails, prisons, and the health of urban populations: a review of the impact of the correctional system on community health. J. Urban Health 78, Glaze, L.E., Bonczar, T.P., Probation and Parole in the United States, Bureau of Justice Statistics, U.S. Department of Justice, Washington, DC, NCJ Publication No Grudzinskas, A.J., Clayfield, J.C., Roy-Bujnowski, K., Fisher, W.H., Richardson, M.H., Integrating the criminal justice system into mental health service delivery: the Worcester diversion experience. Behav. Sci. Law 23, Hammett, T.M., Gaiter, J.L., Crawford, C., Reaching seriously at-risk populations: Health interventions in criminal justice settings. Health Educ. Behav. 25, Heflinger, C.A., Measuring service system coordination in managed mental health care for children and youth. Eval. Program Plann. 19, Henderson, C.E., Young, D.W., Jainchill, N., Hawke, J., Farkas, S., Davis, R.M., Program use of effective drug abuse treatment practices for juvenile offenders. J. Subst. Abuse Treat. 32, Hill, C.J., Lynn, L.E., Producing human services: why do agencies collaborate? Public Manage. Rev. 5, Himmelman, A.T., On the theory and practice of transformational collaboration: from social service to social justice. In: Huxham, C. (Ed.), Creating Collaborative Advantage. Sage Publications, London, pp Himmelman, A.T., On coalitions and the transformation of power relations: collaborative betterment and collaborative empowerment. Am. J. Community Psychol. 29, Horwath, J., Morrison, T., Collaboration, integration and change in children s services: critical issues and key ingredients. Child Abuse Negl. 31, Huxham, C. (Ed.), Creating Collaborative Advantage. Sage, Thousand Oaks, California. Huxham, C., Vangen, S., Leadership in the shaping and implementation of collaboration agendas: How things happen in a (not quite) joined-up world. Acad. Manage. J. 43, Inciardi, J.A., Martin, S.S., Butzin, C.A., Five-year outcomes of therapeutic community treatment of drug-involved offenders after release from prison. Crime Delinq. 50, Kagan, S., Neville, P., Integrating Services for Children and Families: Understand the Past to Shape the Future. Yale University Press, New Haven, Connecticut. Kahn, A.J., Kamerman, S.B., Integrating Service Integration: An Overview of Initiatives, Issues, and Possibilities. National Center for Children in Poverty, Columbia University, New York. Knol, D.L., Berger, M.P., Empirical comparison between factor analysis and multidimensional item response models. Multivariate Behav. Res. 26, Konrad, E.L., A multidimensional framework for conceptualizing human services integration initiatives. New Dir. Eval. 69, Kusserow, R.P., Services Integration: A Twenty-year Retrospective. Office of the Inspector General, Department of Health and Human Services, Washington, DC. Langan, P.A., Levin, D.J., Recidivism of Prisoners Released in Bureau of Justice Statistics, U.S. Department of Justice, Washington, DC, NCJ Publication No Lasker, R.D., Weiss, E.S., Miller, R., Partnership synergy: a practical framework for studying and strengthening the collaborative advantage. Milbank Q 79, Lehman, W.E.K., Fletcher, B.W., Wexler, H.K., Melnick, G., Organizational factors and systems integration activities in criminal justice and drug abuse treatment agencies. Drug Alcohol Depend. 103 (Supl. 1), S65 S72. Messeri, P., Kim, S., Whetten, K., Measuring HIV services integration activities. J. HIV/AIDS Soc. Serv. 2, Mitchell, S.M., Shortell, S.M., The governance and management of effective community health partnerships: a typology for research, policy, and practice. Milbank Q 78, Morrissey, J.P., Calloway, M.O., Thakur, N., Cocozza, J., Steadman, H.J., Dennis, D., Integration of service systems for homeless persons with serious mental illness through the ACCESS program. Psychiatr. Serv. 53, Mumola, C.J., Karberg, J.C., Drug Use and Dependence, State and Federal Prisoners, Bureau of Justice Statistics, U.S. Department of Justice, Washington, DC, NCJ Publication No O Callaghan, F., Sonderegger, N., Klag, S., Drug and crime cycle: evaluating traditional methods versus diversion strategies for drug-related offences. Aust. Psychol. 39, Office of Justice Programs, The Weed and Seed Implementation Manual. U.S. Department of Justice, Washington, DC, NCJ Publication No Olson, D.E., Juergens, R., Karr, S.P., Impetus and implementation of the Sheridan Correctional Center therapeutic community. Illinois Crim. Justice Inform. Authority Program Eval. Summ. 3 (1), 1 12 (accessed August 28, 2008 at Paparozzi, M.A., Gendreau, P., An intensive supervision program that worked: service delivery, professional orientation, and organizational supportiveness. Prison J. 85, Pettus, C.A., Severson, M., Paving the way for effective reentry practice the critical role and function of the boundary spanner. Prison J. 86, President s Commission on Law Enforcement and the Administration of Justice, The Challenge of Crime in a Free Society. U.S. Government Printing Office, Washington, DC. Provan, K.G., Milward, H.B., Do networks really work? A framework for evaluating public-sector organizational networks. Public Adm. Rev. 61, Randolph, F., Blasinsky, M., Morrissey, J.P., Rosenheck, R.A., Cocozza, J., Goldman, H.H., ACCESS Natl Evaluation Team, Overview of the ACCESS program. Psychiatr. Serv. 53, Rapp, C.A., Goscha, R.J., The principles of effective case management of mental health services. Psychiatr. Rehabil. J. 27, Rapp, R.C., Siegal, H.A., Li, L., Saha, P., Predicting postprimary treatment services and drug use outcome: a multivariate analysis. Am. J. Drug Alcohol Abuse 24, Rivard, J.C., Morrissey, J.P., Factors associated with interagency coordination in a child mental health service system demonstration. Adm. Policy Ment. Health 30, Robillard, A.G., Gallito-Zaparaniuk, P., Arriola, K.J., Kennedy, S., Hammett, T., Braithwaite, R.L., Partners and processes in HIV services for inmates and ex-offenders facilitating collaboration and service delivery. Eval. Rev. 27, Siegal, H.A., Li, L., Rapp, R.C., Case management as a therapeutic enhancement: impact on post-treatment criminality. J. Addict. Dis. 21, Steadman, H.J., Boundary spanners: a key component for the effective interactions of the justice and mental health systems. Law Hum. Behav. 16, Steadman, H.J., Cocozza, J.J., Dennis, D.L., Lassiter, M.G., Randolph, F.L., Goldman, H., Blasinsky, M., Successful program maintenance when federal demonstration dollars stop: the ACCESS program for homeless mentally ill persons. Adm. Policy Ment. Health 29, Substance Abuse and Mental Health Services Administration, Treatment Episode Data Set (TEDS) : National Admissions to Substance Abuse Treatment Services. DASIS Series S-37, DHHS Publication No. (SMA) Taxman, F.S., Reducing Recidivism Through a Seamless System of Care: Components of Effective Treatment, Supervision, and Transition Services in the Community. Office of National Drug Control Policy, Washington, DC (accessed online February 20, 2007 at

11 S64 B.W. Fletcher et al. / Drug and Alcohol Dependence 103S (2009) S54 S64 Taxman, F.S., Bouffard, J.A., The importance of systems issues in improving offender outcomes: critical elements of treatment integrity. Justice Res. Policy 2, Taxman, F.S., Bouffard, J.A., Assessing therapeutic integrity in modified therapeutic communities for drug-involved offenders. Prison J. 82, Taxman, F.S., Perdoni, M.L., Harrison, L.D., 2007a. Drug treatment services for adult offenders: the state of the state. J. Subst. Abuse Treat. 32, Taxman, F.S., Young, D.W., Wiersema, B., Rhodes, A., Mitchell, S., 2007b. The National Criminal Justice Treatment Practices survey: multilevel survey methods and procedures. J. Subst. Abuse Treat. 32, Treatment Accountability for Safer Communities, About National TASC: Methodology (accessed online January 16, 2007 at org/about/about.html). Turner, S., Longshore, D., Wenzel, S., Deschenes, E., Greenwood, P., Fain, T., Harrell, A., Morral, A., Taxman, F., Iguchi, M., Greene, J., McBride, D., A decade of drug treatment court research. Subst. Use Misuse 37, Vanderplasschen, W., Rapp, R.C., Wolf, J.R., Broekaert, E., The development and implementation of case management for substance use disorders in North America and Europe. Psychiatr. Serv. 55, Veysey, B.M., Steadman, H.J., Morrissey, J.P., Johnsen, M., In search of the missing linkages: continuity of care in US jails. Behav. Sci. Law 15, Weiss, E.S., Anderson, R.M., Lasker, R.D., Making the most of collaboration: exploring the relationship between partnership synergy and partnership functioning. Health Educ. Behav. 29, Wells, R., Lemak, C.H., D Aunno, T.A., Factors associated with interorganizational relationships among outpatient drug treatment organizations Health Serv. Res. 40 (Part I), Welsh, W.N., McGrain, P.N., Predictors of therapeutic engagement in prisonbased drug treatment. Drug Alcohol Depend. 96, Welsh, W.N., Zajac, G., A census of prison-based drug treatment : implications for programming, policy, and evaluation. Crime Delinq. 50, Wenzel, S.L., Longshore, D., Turner, S., Ridgely, M.S., Drug courts a bridge between criminal justice and health services. J. Crim. Justice 29, Wenzel, S.L., Turner, S.F., Ridgely, M.S., Collaborations between drug courts and service providers: characteristics and challenges. J. Crim. Justice 32,