Development of a Needs-Based Planning Model for

Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada: Final Report 2010-2014

Brian Rush Joël Tremblay Chantal Fougere Renée Behrooz Wendi Perez and Julia Fineczko

Health Systems and Health Equity Research Group Centre for Addiction and Mental Health

July 24, 2014

1.0 INTRODUCTION AND BACKGROUND .......................................................................................................................... 1 1.1 TIERED FRAMEWORKS TO SUPPORT TREATMENT SYSTEM PLANNING....................................................................................... 4 1.1.1 Continuum-of-care ........................................................................................................................................... 4 1.1.2 Beyond the Continuum-of-Care to Tiered Frameworks ................................................................................... 6 1.2 THE ROLE OF HEALTH PROMOTION AND PREVENTION IN NEEDS-BASED PLANNING ..................................................................12 1.3 RECONCILING THE TIERED FRAMEWORK AND THE CONTINUUM OF CARE APPROACH .................................................................14 1.3.1 Brief Interventions ..........................................................................................................................................17 1.3.2 Computer and Mobile-Based Services and Supports ......................................................................................20 1.4 GOING FROM TIERED FRAMEWORKS TO NEEDS-BASED PLANNING .........................................................................................22 1.4.1 Needs-based planning models ........................................................................................................................24 1.4.2 Extending the 1990 “Rush Model” to the full tiered framework ....................................................................27 1.4.3 Project Objectives ...........................................................................................................................................28 1.4.4 Assumptions Underlying Model Development ................................................................................................29 1.4.5 Evaluating Project Impact ..............................................................................................................................31 2.0 DEFINING NEED ...........................................................................................................................................................32 2.1 CONCEPTUAL AND MEASUREMENT ISSUES ........................................................................................................................32 2.2 A FEASIBLE MEASUREMENT MODEL FOR NEEDS-BASED PLANNING IN CANADA .......................................................................39 3.0 DEFINING THE TREATMENT SYSTEM FOR REQUIRED CAPACITY ESTIMATION .........................................................45 3.1 SERVICES REQUIRED TO DELIVER THE UNIVERSAL TREATMENT SYSTEM FUNCTIONS...................................................................61 3.2 SCHEMATIC DIAGRAM OF NEEDS-BASED PLANNING MODEL FOR SUBSTANCE USE SERVICES AND SUPPORTS .................................67 4.0 POPULATING THE PARAMETERS OF THE MODEL ......................................................................................................72 4.1 NATURALISTIC HELP-SEEKING (P2 ....................................................................................................................................72 4.2 GENERIC SERVICES (E.G. SBIRT; ADDICTION LIAISON, HEALTH EDUCATOR) (P1, P3 - P5) ..............................................................79 4.3 DELPHI PROCESS TO ALLOCATE THE OVERALL HELP-SEEKING POPULATION TO TREATMENT SERVICE CATEGORIES P6 – P23....................82 4.3.1. Selection of Delphi Participants ....................................................................................................................83 4.3.2. Data Collection Instrument ............................................................................................................................84 4.3.3 Data Collection ...............................................................................................................................................86 4.3.4 Data Analysis ..................................................................................................................................................89 4.3.5 Results .............................................................................................................................................................90 4.3.6 Discussion .......................................................................................................................................................95 5.0 PILOT TESTING AND GAP ANALYSIS .........................................................................................................................102 5.1 PILOT SITE RESULTS ....................................................................................................................................................103 5.2 LESSONS LEARNED FROM PILOTING: CHALLENGES IN THE APPLICATION OF THE NEEDS-BASED PLANNING MODEL .........................122 5.3 KEY LESSONS LEARNED................................................................................................................................................125 6.0 LIMITATIONS, SPECIAL CONSIDERATIONS AND NEXT STEPS ...................................................................................126 7.0 REFERENCES ..............................................................................................................................................................133 APPENDIX A: RESEARCH TEAM AND NATIONAL ADVISORY COMMITTEE MEMBERS ..................................................149 APPENDIX B: PROJECT LOGIC MODEL……………………………………………………………………………………………………………………151 APPENDIX C: CATEGORY DEFINITIONS AND CCHS 1.2 VARIABLES ................................................................................152 APPENDIX D: DELPHI MATERIALS ...................................................................................................................................153 APPENDIX E: SARNIA-LAMBTON NEEDS-BASED PLANNING PRESS RELEASES ..............................................................164 APPENDIX F: APPROXIMATE POPULATION OF HOMELESS PEOPLE IN CANADA ..........................................................167 APPENDIX G: APPROXIMATE POPULATION (AGES 15 AND OVER) LIVING ON RESERVES IN CANADA ........................168 APPENDIX H: HEALTH PLANNING REGIONS IN CANADA ...............................................................................................169

1.0 Introduction and Background1

It is now well established that a relatively small proportion of people in the community who experience substance use problems seek assistance from the specialized sector of services that has been commissioned specifically to provide treatment and support to people with these challenges. The data supporting this assertion are drawn largely from general population surveys that variously define the need for treatment and also inquire about formal and informal help seeking within a defined timeframe. Such data on the so-called “treatment gap” have supported the case for a more comprehensive view of the substance use treatment system. It is argued that a discernible impact at a population level is not likely to be achieved only through provision of services commissioned specifically to serve people with the most severe and complex needs such as specialized substance use treatment agencies (e.g., Babor, Stenius, & Romelsjo, 2008). A broader population health approach is needed; one that engages multiple sectors such as health, social welfare, criminal justice, and education in a comprehensive system of services and supports. This means building service capacity in the settings where people with substance use problems are more typically engaged (e.g., primary care, emergency departments, criminal justice). It also means implementing early intervention, health promotion and prevention policies and services for those at risk of developing these problems, and working to link these initiatives to the treatment system. Policies and programs designed to reduce stigma and discrimination of people with substance use problems are also critical since they can impact help-seeking and participation in treatment and early intervention services. One can locate the roots of this comprehensive view on treatment systems in seminal reports from the early 1990s (most notably the Institute of Medicine report (1990)) which called for “broadening the base of treatment” in order to achieve wider coverage and yield positive outcomes at a population level. Figure 1, adapted from Babor and colleagues, offers a perspective on the wide range of service delivery settings and contexts to be considered in broadening the base of treatment and integrating services into a coherent treatment system. 1

This introduction draws heavily upon a paper by Rush (2010) on the evolution of tiered frameworks for planning substance use treatment systems. Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

1

Figure 1: Service Delivery Mechanisms and Contexts for a Comprehensive Substance Use Treatment System*

*Adapted from Babor et al., 2008.

Paralleling this more comprehensive perspective of what comprises the treatment system, has been a broader understanding of the nature of substance use problems. It is now commonly recognized that the construct of “substance use problems” is multi-dimensional, comprised of substance use (frequency, quantity and variability), substance abuse (essentially negative consequences of use), and substance dependence (Hasin et al. 2006; Rehm, 2008). In addition, evidence from studies involving people from the general population and clients in treatment/health care settings also shows that heavy substance use, abuse and/or dependence frequently co-occur with mental health problems, physical illness and a range of psychosocial needs. Thus, the overall problem profile is complex and exists in varying degrees of severity. This heterogeneity is not well-captured in historical classification systems. One approach to the conceptualization of problem severity suggests that it consists of three inter-related dimensions: acuity, chronicity and complexity (Reist & Brown 2008). Acuity refers to short duration and/or urgent risks or adverse consequences (e.g., accidents or criminal charges) that are associated with the index problem (e.g., heavy substance use or dependence).

Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

2

Chronicity refers to the development or worsening of long duration or enduring conditions (e.g., major depression or other severe mental disorders, chronic pain, Hepatitis C). Complexity refers to the degree of co-occurrence of the acute or chronic index problems and/or the existence of health and social factors such as homelessness, unemployment, family dysfunction that complicate the process of addressing the index problem(s). Complexity is a concept that is being applied more frequently to individual assessment and treatment planning in the field of psychosomatic medicine (Huyse et al., 2006); the planning and implementation of various strategies for integrating mental health and substance use services with broader health care services and systems (e.g., Kathol et al., 2009); and risk-adjustment for outcome monitoring and costing purposes (Hermann et al., 2007). Here we apply it to needs-based planning at a population level. The severity of substance use problems thus represents the cumulative gestalt of acuity, chronicity, and complexity, akin to the concept of “level of burden” (Aldworth et al., 2010) or “multi-morbidity” (Angst et al., 2002). For planning purposes, it is also important to consider how these acute and chronic problems converge at different points in time in the individual’s life course and thereby influence the trajectory of help-seeking and service utilization – often referred to as the “treatment career” (Anglin, et al. 1997; Chi & Weisner, 2008).

Figure 2: Population by Problem Severity


3

Figure 2 illustrates the distribution of substance use problem severity within the general population, a distribution that can be described as a population health pyramid. The highest levels of severity are associated with the fewest number of people but who need the most costly specialized and/or intensive care. Those with lower levels of problem severity are more numerous and their needs can often be met by less intensive or less specialized care that is more widely available in a variety of health and social service contexts, as well as more informal community and/or family networks of support. Internet and mobile-based services are also becoming more widely available (Bewick et al, 2008; Cunningham & Van Mierlo, 2009; Lapham at al, 2012) including the provision of direct and quite structured treatment services for people with severe substance use disorders (Campbell et al., 2014). The bottom of the pyramid reflects people at low risk – the target population for secondary and primary prevention. The population health pyramid underlies what is often referred to as the Chronic Care Model, to be discussed in more detail below. Simply put, the broad “treatment system” must be planned in such a way as to respond effectively and efficiently to this full spectrum of acute, chronic and complex needs.

1.1 Tiered Frameworks to Support Treatment System Planning For substance use treatment systems, there is no generally accepted conceptual framework to guide planning and resource allocation; indeed the form such a framework would take depends on the prevailing views of substance use problems and their treatment; extant knowledge of evidence-based practice; the purpose to which the model will be put; and, perhaps most importantly, the social, political and cultural context for model development, implementation and evaluation.

1.1.1 Continuum-of-care The concept of the “continuum-of-care” underlies one such conceptual framework that has been brought to bear for several years in the substance use field, and which continues to hold currency in many parts of the world. Briefly, one can view the continuum-of-care as being organized along categories of service delivery that correspond to a mix of services and expected flow of clients into and through various treatment settings and functions (e.g., intake, screening,


4

assessment and treatment planning, withdrawal management/detoxification, stabilization, treatment intervention, continuing care2). A systems framework based on a continuum-of-care approach rests upon a model of problem severity that is also on a continuum. Thus, a range of treatment settings offer interventions of varying intensity and structure (e.g., social versus medical withdrawal management, community treatment3, day/evening treatment, short/long term residential) that are accessed by clients on the basis of problem severity and other matching criteria such as stability of the person’s life situation, psychiatric comorbidity, and status of his/her environment with respect to relapse prevention (Toche-Manley et al., 2011). Such an approach underlies the widely used ASAM criteria for matching people seeking help to various levels of care (Gastfriend, 2003; American Association of Community Psychiatry, 2009; Rush, in press). It also provides the foundation for assessment and matching protocols in several jurisdictions, including Ontario (Cross & Sibley-Bowers, 2002). The work in the Netherlands on system development based on client-treatment matching has also been noteworthy in this regard (Schippers et al., 2002). Past system design efforts that hinge on the continuum-of-care model include the “CoreShell Model”, whereby centralized functions of intake, assessment and case management (the core) match and link clients to the array of treatment services required for the overall client population (the shell) (Glaser, 1974; Marshman, 1978). Another more recent system design framework that is also based on continuum-of-care principles is the “stepped care” approach such that clients are assigned on the basis of assessment to the least intensive and intrusive level of care, and then “step-up” if outcomes are not positive and, when appropriate, “step-down” for the maintenance of gains and ongoing support (Breslin et al., 1998; Sobell & Sobell, 2000). From a historical perspective, the continuum-of-care model was a significant advance over a “one-sizefits-all-approach” to delivery of substance use treatment services, for example, the 28-day “Minnesota Model” of residential treatment (Winters et al., 2000). As useful as it has been, however, the continuum-of-care approach appears to have now been subsumed under the broader systems approach described earlier, since the continuum of care traditionally has included

2

The concept of continuing care has been expanded to include “Recovery Monitoring Check-ups” (Dennis, Scott and Funk, 2003; Rush, 2008) which aim to return people to treatment based on their status at an outcome monitoring follow-up contact. 3 “Community treatment” is the term used in Ontario, for what several other jurisdictions call outpatient services. Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

5

only the specialized sector of substance use services to the exclusion, for example, of primary care or mental health settings. 1.1.2 Beyond the Continuum-of-Care to Tiered Frameworks As noted earlier, one of the first conceptual frameworks for substance use treatment systems that moved beyond the continuum-of-care approach was advanced in 1990 in a seminal report from the Institute of Medicine (1990) in the U.S., and which drew heavily on Canadian research and expertise. A more recent approach is referred to as the “tiered framework” or “tiered model”, a systems modeling approach that has found its way in the last decade into planning documents for both mental health and substance use services from several countries, including the UK (National Treatment Agency for Substance Misuse, 2006), Australia (National Mental Health Strategy, 2004), Canada (Hollander & Prince, 2008; National Treatment Strategy Working Group, 2008), and Europe (Baldacchino & Corkery, 2006). The roots and evolution of tiered frameworks have been described by Rush (2010). Briefly, such frameworks have their origins in the Chronic Care Model (CCM) for the treatment and management of chronic illnesses such as diabetes or other long-term health conditions (Wagner, 1998; Bodenheimer et al., 2002) as well as integrated service delivery models that sought to operationalize the CCM. In particular, the Continuity-of-Care Model (McGonigle et al., 1992) and the so-called Kaiser Triangle (Wallace, 2005) defined “levels of chronic care” that were based on level of risk and problem severity, a fundamental aspect of tiered approaches for substance use treatment, and mental health services generally. The Continuity-of-Care Model and the Kaiser Triangle both share the idea of service delivery tiers matched to the distribution of severity at the population level. These models all build upon the well-known articulation of primary, secondary, and tertiary care based on level of risk and problem severity. The National Treatment Agency for Substance Misuse in the UK led the way internationally in the development and application of tiered frameworks in the substance use field. The essential idea was to define a set of tiers (as in the Kaiser or Continuity-of-Care models); locate various elements of a comprehensive treatment system in the various tiers; and then work with local jurisdictions to, over time, address system gaps through funding and/or more integrated policies


6

and care planning. The first attempt, in 2002, defined four tiers on the basis of a combination of setting, interventions and the agency responsible for providing the interventions (National Treatment Agency, 2002). This “mixed bag” of criteria for allocating elements of the treatment system to a particular tier resulted in considerable confusion and variability in application, in particular an overly rigid interpretation of the tiers. Of particular concern was the view that certain types of service providers were “slotted into” one particular tier even though they provided services that may span more than one tier and provide services to people across quite a range of severity4. A revised model was released in 2006 which defined the four tiers on the basis of “interventions” offered within them, and provided greater clarity around the nature of these interventions, the settings in which they may be located, and the competencies required for them to be successfully offered to clients and their families. Inherent to the definitions of tiers 1 through 4 is the population distribution of severity as reflected in the Kaiser triangle. In addition to articulating these tiers and their various elements, the UK framework identified several critical features in support of the client’s “treatment journey” recognizing that treatment is more of a process than an event, and offering important citations to longitudinal research on treatment trajectories. These critical features included treatment engagement, treatment delivery (including maintenance), community integration (which underpins both delivery and treatment maintenance, or completion), and treatment completion (for those who choose to be drug free). Several concrete options were recommended to operationalize these features, such as “keyworkers”, who are dedicated practitioners responsible for ensuring the client’s care plan is delivered and reviewed; and customized “integrated care pathways” that are dynamic and flexible to changing client needs. Such concrete options for service continuity are reminiscent of the details embedded in the Chronic Care Model and the Kaiser Triangle, and reinforce the critical importance of linkage and transitions across the tiers, as well as the systemlevel supports that are needed to sustain these linkage mechanisms (e.g., e-health capability to transfer assessment and treatment information).

4

The original document had explicitly stated this was not the case, however.


7

In 2008, a Canadian report entitled Systems Approach to Substance Use in Canada: Recommendations for a National Treatment Strategy was released (www.nts-snt.ca). A five-tiered framework in support of a broader systems approach was a key element, and drew substantively upon the UK approach. As in the UK, the framework was based on the idea that different service categories were embedded in each tier. These categories aligned with different levels of problem severity of the help-seeking population. That said, significant challenges arose with definitional issues (i.e., what elements of the treatment system really fit into what tiers? How much flexibility in interpretation was tolerable in planning services?). The challenges were reminiscent of the UK “tier trap” whereby different types of service models were seen as belonging to a tier even though they could also provide functions across other tiers and address multiple levels of severity. However, major differences and improvements upon the UK framework, included: 

The addition of a fifth tier focused on prevention and health promotion;



A broader set of criteria defining the tiers;



A clearer role for natural, informal systems of support such as family and friends, and community structures such as neighborhood associations;



A clearer role for self-help (e.g., which was conceptualized as largely Tier 1 on the basis of its open nature). In 2008 an Ontario report on system design was developed in support of the 10-year

mental health and addictions strategy. This report adapted the tiered model from the national treatment strategy to incorporate addictions, mental health and problem gambling5. The integrated tiered model is shown in Figure 3, and its roots further explicated by Rush (2010). In this framework there were five tiers comprised of ‘functions’. A function was defined as a higher-order grouping of like services or interventions aimed at achieving similar outcomes and targeted at a particular level of problem severity. Thus, the term ‘function’ was interpreted broadly, including components along the continuum of care (e.g., outpatient or residential 5

This built upon work underway at the same time in Alberta, Canada that also aimed at one integrated tiered model for mental health and substance use services and which separated three sub-populations: people with mental health problems, people with substance use problems substance abuse, and people with co-occurring disorders (Fraser, 2009).


8

treatment); a multidisciplinary team providing specialized care (e.g., Assertive Community Treatment); a class of interventions (e.g., screening, self-management, pharmacotherapy); a type of risk management/reduction (e.g., emergency medical care, psychosocial crisis intervention, needle exchange); a population-based initiative (e.g., health promotion); or any of a variety of types of general counseling and support (e.g., continuing care, case management, support groups). Since a range of functions from more than one tier may be provided within one program/service, a function is thereby distinguished from the program or service in which it is embedded (e.g., a primary care service; a substance use program; a community mental health agency). The functions were grouped within tiers that reflected an increasing degree of specialization with respect to the nature of the function provided and the competency requirements of the service provider to address mental health, substance use, and/or gambling problems. This increased degree of specialization also was considered to correspond to increased problem severity (as described in Figure 3) such that the higher the tier, the higher the severity of the target population and the fewer the number of people in need of the specialized service. Figure 3: Integrated tiered framework for mental health, substance use and problem gambling CORE SYSTEM PRINCIPLES FOR ACCESS AND INTEGRATION ACROSS FUNCTIONS

Any Door is the Right Door

TIER 5 Highly specialized care functions targeted to individuals with complex problems

CORE SERVICE SYSTEM PRINCIPLES

TIER 4 Specialized care functions targeted to people assessed/diagnosed as in need of more intensive or specialized care.

Simultaneous/ Sequential Tier Involvement

Consumer Involvement Cultural Competence

Graduated Integration Linkages across Tiers & with other Service Systems

TIER 3 Treatment planning, risk/crisis management and support functions targeted to individuals with identified problems. TIER 2 Early Intervention and Self-Management functions targeted to people at risk TIER 1 Population-based health promotion and prevention functions targeted at the general population

Policy Leadership Funding

System Supports

Determinants of Health Family Involvement Harm Reduction Psychosocial Supports SelfManagement

Performance Management & Accountability Information Management Research & Knowledge Exchange

MOVING THE SYSTEM FORWARD: PLANNED CHANGE MANAGEMENT Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

9

The functions associated with each of the tiers of the Ontario tiered model were described as follows: Tier 1: Population-based health promotion and prevention functions targeted at the general population  This tier is comprised of functions that are designed to enhance natural systems and networks of support for individuals, families and communities. This includes an emphasis on the social determinants of health as well as education and policy functions aimed at the general public with the objective of promoting healthy lifestyles and preventing the development of mental health, substance use or gambling problems. Tier 2: Early intervention & self-management functions targeted to people at risk  This tier is comprised of functions targeted to people with emerging or unidentified problems. The functions include screening/identification, information & referral, brief interventions, brief psychotherapy, psychopharmacy, self-management, motivational and peer support functions. Tier 3: Treatment planning, risk/crisis management and support functions targeted to individuals with identified problems  This tier is comprised of functions targeted to people with identified problems who are not engaged in, or have completed specialized treatment. These functions may serve as a doorway to higher tier, specialized care functions and lower tier, self-management and mutual aid functions. Examples of these Tier 3 functions include comprehensive assessment/diagnosis, outreach/engagement, and case management. They also include general support functions (e.g., continuing care, supportive counseling, support groups, walk-in services) as well as functions designed to reduce the risks and consequences associated with the identified problems (e.g., emergency/acute medical care, psychosocial crisis intervention, and needle exchange).


10

Tier 4: Specialized-care functions targeted to people assessed/diagnosed as in need of more intensive or specialized care  This tier is comprised of, but not limited to, most of the functions generally considered to be part of the specialized mental health, substance use and problem gambling treatment systems. The functions include ambulatory and structured residential interventions, including pharmacotherapy, psychotherapy, and may involve multidisciplinary teams (e.g., ACT). These are specialized treatment functions intended to be delivered by individuals with special training to people who have been assessed/diagnosed as requiring this level of specialization. The function is unrelated to setting (e.g., a primary care physician providing pharmacotherapy for alcohol dependence or depressive disorders is providing a Tier 4 function). Tier 5: Highly specialized-care functions targeted to individuals with complex problems  These are functions designed for people with particularly complex or severe mental health, substance use or gambling problems or combinations of these problems (e.g., inpatient medical withdrawal management; comprehensive inpatient/residential concurrent disorder services; inpatient forensic services; long-term inpatient psychiatric care). As in the UK model, and the Canadian adaptation for the Systems Approach (National Treatment Strategy Working Group, 2008), the fact that people can enter this comprehensive service and support system at multiple points is of critical importance (i.e., the concept of “any door is the right door”). Thus, people may access the system by way of any of the five tiers and, upon entry, should be linked to other functions within or across tiers according to their needs. The system is also to be operationalized in such a way as to facilitate transitions across the tiered functions as dictated by the individual’s needs. Thus, no part of the system “owns” the person; they are a client of the entire system. The set of core service/system principles described in the right hand and left hand side-bars of Figure 3 refer to these and other fundamental principles and values that are applicable to all of the functions across the five tiers. Another key principle embedded in the tiered model is that various programs or settings can provide multiple functions and across multiple tiers. Coordination and continuity across


11

functions are critical to ensuring the system works for the person and his or her family. Especially important in this regard are: (a) the development of linkages to facilitate service integration (e.g., case management), and (b), the application of the concept of ‘graduated integration’ (i.e., gauging the level of integration of services according to the severity of the individual case). Service integration is distinct from system integration, the latter referring to the regional and provincial/territorial structures and processes that provide the infrastructure for the organization and delivery of integrated clinical and psychosocial services (Rush & Nadeau, 2011). These are represented by the ‘System Supports’ section at the bottom of Figure 3 and provide the foundation for all service delivery within and across the five tiers; for example, leadership, policy, funding, and performance measurement systems. In summary, the tiered framework as developed in Ontario, and adapted from earlier national and international iterations (Rush, 2010), was intended as a planning tool to guide the development and implementation of an integrated system of service functions for people with mental health, substance use and gambling problems. It incorporated a distinction between service-level integration and system-level integration and described a broader vision of a comprehensive, integrated system based on an alignment of service functions with the level of severity in the population. It was also based on a population health approach that included increased emphasis on health promotion, prevention, early intervention, and reduction of stigma and discrimination.

1.2 The Role of Health Promotion and Prevention in Needs-Based Planning The focus of this project is on the planning of substance use services and supports on the basis of a systematic analysis of population needs. This includes the needs for health promotion and prevention. However, doing full justice to the health promotion and prevention component is a major challenge as this can involve planning and delivery of health promotion and prevention functions to people already engaged with the service delivery system, as well as those more ‘upstream’ in the general population in terms of risks and harms. As a result an important limitation of this project, in terms of scope, is that it is not possible to fully address the needs and related resource requirements for Tier 1 functions that are aimed at the general population. The Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

12

importance of this limitation cannot be overemphasized for several reasons, not the least of which is that these population-level programs and policies will have a larger impact on the overall health of the population vis a vis substance use and addiction than the efforts of more ‘downstream’ services and supports. These programs/policies may include those aimed at the broad determinants of health (e.g. housing, income support, education, neighborhood safety and green space), as well those aimed at substance use and addiction specifically (e.g., drinking age, alcohol pricing and availability, server intervention, school-based programs, control of drug supply, pharmaceutical-related policy with respect to opiate and other medication). Research in the tobacco area suggests it is a combined ecological approach that will be most effective. Recent analysis suggests the importance of alcohol control measures above and beyond screening and treatment-oriented interventions (Heather, 2012).

To summarize, ‘health services’ is but one element of a comprehensive approach to health promotion and prevention and, in terms of population health, is overall less important than focusing on the environmental and social determinants of health, and the specific individual, familial, and community risk factors for substance use and addiction. This is a core message of the Ottawa Charter (World Health Organization, 1986), which places the primary emphasis on salutogenic versus pathogenic responses to substance-related risks and harms. That said, there are no doubt advantages in combining needs assessment, planning, and resource allocation processes to span both population-level health promotion and prevention and substance use treatment services and supports. In some jurisdictions (e.g. Nova Scotia) this level of integration may extend to specification of job duties of managers and staff, thereby challenging isolated estimation of resource needs for the treatment service component alone. More commonly in Canada and elsewhere there are different divisions/departments of government responsible for public health and health care which calls for significant collaborative planning efforts. Collaboration is needed to involve substance use services and supports in population-based health promotion and prevention programs and policies. Each jurisdiction/community that applies a needs-based planning model for substance use services and supports must: (a) consider the relative balance of resources to be devoted to population-level health promotion and prevention, and resources for substance use services and supports; and (b) embed, to the extent possible, the resource requirements and the Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

13

manner in which health promotion and prevention functions will be embedded inside and alongside the substance use treatment services and supports. 1.3 Reconciling the Tiered Framework and the Continuum of Care Approach An obvious assumption underlying the tiered framework, or any such planning model for substance use services, is that treatment services and supports “work”, that is to say they accrue positive benefits to the people being treated, their families and social networks, and the community as a whole. This assumption is unequivocally supported by research evidence (e.g., Timko et al., 1999; Martin & Rehm, 2012; Rush, 2012; Lev-Ran et al., 2012) including evidence on the return on investment in economic terms (e.g. California Department of Alcohol and Drug Programs, 2008). Treatment research questions now focus on who does best with what treatment options (i.e. interventions, levels of care) and thus call for studies on the treatment environment, societal control/coercion, engagement strategies, therapeutic relationship and the role of clientlevel factors such as problem severity/complexity and readiness to change. Are the tiered framework and the continuum-of-care model compatible with each other as frameworks for planning these evidence-based treatment services? Some reconciliation is important since the continuum-of-care model still underpins the planning of substance use services in Canada and elsewhere and the tiered framework appears to be increasingly popular as a planning tool. Reconciliation, however, requires further refinement of the thinking behind the tiered framework, including the Ontario version. One must start from the premise that, first and foremost, the tiers in the framework represent levels of severity, or perhaps more precisely, the levels of risks and harms related to substance use as distributed in the general population. Given that, one can then consider the tiers from a service planning point of view as a collection of functions that are required in comprehensive treatment systems in order to minimize the levels of risks and harms of people in these categories. While this was the intention in the Ontario model, the definition of functions was still mixed with service delivery settings, and many of these settings were typically aligned with the older continuum-of-care model. In short, the Ontario model still confused functions and settings. More clarity in these definitions is needed in order to better align the needs-based planning model that combines the tiered framework and the continuum-of-care model. Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

14

Table 1 shows, at a conceptual level, the relationship between the tiers as defined by risks and harms in the general population, and functions to be delivered in a comprehensive treatment system. We build upon these functions in a subsequent section and, at this point, name these tiers and functions for illustration purposes only. Two points are critical. The first is that for purposes of system planning, the populations needing each of these core functions in the treatment system are “nested”. For example, the prevention and health promotion function should be delivered to people in ALL the severity tiers; early intervention for those in Tiers 2 and up; and so on to Tier 5, such that the target population for Tier 5 is the only one needing treatment functions aimed at severe addiction, health-related and mental health co-morbidity. This nesting of functions has significant implications for staffing and other resource requirements and, therefore, the costs of services to deliver these functions. Table 1: Conceptual view of the relationship between tiers in the general population and functions of a comprehensive treatment system6 Universal Functions

Treatment Treatment of of Complex Abuse and Co-occurring Dependency Disorders

Prevention and Support

Early Intervention

Risk Reduction

Tier 5: Complex/High Severity

X

X

X

X

Tier 4: Chronic Harms

X

X

X

X

Tier 3: Active Risk/Harm

X

X

X

Tier 2: Moderate Risk

X

X

Tier 1: Low Risk

X

F

Severity

X

6

We would like to acknowledge David Brown on our national Project Advisory Committee for contributing this framework to the project. Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

15

The second key consideration in reconciling the tiered framework and the continuum-ofcare is that there are many service delivery models that can provide the various functions to the appropriate target population represented by the tiers. The traditional continuum-of-care approach was broadly based on a set of service delivery categories, variously defined as outpatient treatment, intensive outpatient or day/evening treatment and residential treatment; the ASAM model of treatment matching and client placement being a good example. Prevention and health promotion was not typically included. It is also important to recognize that these characterizations of treatment services, and their precise label and definition, are inherently culture and context dependent. Precise definitions aside, the various types of services within a substance use treatment system often deliver more than one core function that may be aimed at multiple populations. A residential treatment centre may, for example, co-locate a counseling staff or nurse liaison in an emergency room of a general hospital and thus provide both counseling and other structured but brief interventions for substance abuse and dependency, as well as outreach and early intervention. They may also provide a needle exchange service, as well as safe sex education aimed at HIV-AIDS prevention. Another example is a primary care physician who not only proactively screens patients for a wide range of substance use and related problems but also provides brief intervention to individuals at low to moderate risk; ongoing counseling to those with substance abuse or dependence; medication management and other support for home-based withdrawal management for those needing this level of care, and medication management and counseling for mild-to-moderate, co-occurring depressive disorders or opiate dependence. Clearly it is inappropriate to “place” residential treatment services, or well-trained and experienced primary care physicians into one and only one tier. It is the functions they offer that align with tiers of problem severity NOT the overall treatment service, program or organization. Thirdly, we must acknowledge that the continuum-of-care as traditionally conceptualized by treatment settings is roughly aligned with functions that are targeted at increasing levels of severity or risks and harms. Conventional practice wisdom holds that individuals experiencing higher levels of risk and harm, more complex substance use-related problems, and whose environment presents challenges for relapse prevention will have better outcomes in residential treatment services compared to non-residential services. Similarly, conventional practice wisdom


16

maintains that the same holds true for non-residential services that vary in duration and intensity of interventions and program structure; the more intense structured programs such as day/evening programs being required for those with a higher degree of severity and complexity than non-residential services where the client attends weekly or bi-weekly appointments, or a small number of very brief treatment sessions. Referral criteria for the more intensive day/evening and residential services typically include severity of dependence, social stability including homelessness, environmental risk for relapse (i.e., heavy alcohol or drug use in the home or immediate social network), and mental health co-morbidity, including suicide risk. While this conventional practice wisdom regarding the so-called “matching hypothesis” is challenged by considerable research that fails to pinpoint the critical matching characteristics, it remains a strong enough assumption to form the basis of the widely used ASAM criteria, including their use in managed care and insurance related purposes in the US (Gastfriend, 2003). It also forms the basis of admission and discharge criteria in Ontario and several other Canadian jurisdictions. 1.3.1 Brief Interventions The discussion of the research evidence underlying client placement along the continuum of care is now further complicated by the need to include brief interventions delivered in generic health care services such as primary care, emergency departments and medical units psychiatric services, as well as other settings such as schools, universities/colleges, and recently, the Internet. Brief intervention may be combined with formal or informal screening and referral to treatment (e.g. SBIRT; Saitz, 2012). Rationale for Screening, Brief Intervention, and Referral to Treatment (SBIRT) The rationale for SBIRT is very strong and it is based largely on the fact that there are a significant number of people who span the spectrum of substance use risks and harms that are currently in contact with a range of non-specialist services, but whose risks and harms remain unidentified (Mitchell et al., 2012). This spectrum is similar to the tiered framework of need (i.e. the population pyramid) although the percentage of people in each category will vary considerably by service delivery setting and population served. For example, the percentage of people with more severe substance dependence, other problems, and co-morbidity will be higher in


17

emergency departments, mental health services and correctional settings than in general medical practice, community health centers, or a counselling department in a college/university or high school. Similarly, the percentage of people in the lower tiers of risk and harms will also vary. Thus, SBIRT interventions have the potential to identify many people with a wide range of risks and harms and provide opportunistic intervention. For some, the goal should be to resolve the problems or reduce the risk/harm with a brief intervention; for others, the goal should be to motivate and support the person to engage in more formal, specialized treatment services and supports.

Is the screening component effective? The evidence is strong that systematic screening will identify many people across the spectrum of risks and harms that would not otherwise be identified. Optimal screening protocols include a validated screening instrument appropriate to the population and the service delivery setting (e.g., the AUDIT, ASSIST, GAIN-SS), and many tools are readily available that are appropriate across a wide developmental age span (e.g. the GAIN-SS (Rush et al., 2013; Rush, in press)). The advantages of validated tools notwithstanding, “screening” can also be undertaken in a less systematic way by increased vigilance to the signs and symptoms of substance use risk and harm (Health Canada, 2001) or by the use of single, validated questions (Smith et al., 2010; Corson et al., 2004; Ramchand et al., 2009). Even with less systematic screening, research shows that increased vigilance followed by engagement with a liaison specialist can be highly effective in identifying and linking new individuals with moderate-to-severe substance use problems into the specialist treatment system for the first time (Blanchette-Martin et al., 2011).

“Brief intervention” is not uniformly defined in terms of structure and duration in the relevant research literature (Saitz, 2012). The briefest approach, typically preceded by a systematic screening process and aimed at non-treatment seeking populations, is usually conducted in one 5 – 15 minute session (Bertholet et al., 2005). While staged models of an increasing number and duration of sessions have also been evaluated, the most common approach is a single session. The evidence is overwhelmingly supportive for brief interventions and brief treatment for both alcohol and other drug use in a range of generic services, particularly for short-term outcomes (Bertholet Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

18

et al., 2005; Babor et al., 2007; Moyer et al., 2002; Cherpitel et al., 2010; McQueen et al., 2011; Kaner et al., 2009; Bien et al., 1993; Madras et al., 2009). Although most studies have involved adult populations, SBIRT has also been shown to be effective for adolescents in a smaller number of studies (Tait & Hulse, 2005).

The vast majority of studies have examined alcohol and drug use and consequences as key outcomes. Other studies use brief interventions to increase motivation for, and engagement in formal treatment. These interventions are often aimed at non-treatment seekers at higher risk levels in terms of consumption or problems/dependence. Recently, brief intervention has been distinguished from “brief treatment”, the latter involving more sessions of longer duration (e.g., 2 to 4 sessions of 20 – 30 minutes) (Babor et al., 2007; Moyer et al., 2002). It is generally accepted that brief intervention, brief treatment, and extended treatment (e.g., ongoing outpatient counselling visits) should be aimed at people of increasing severity levels, although some studies show brief treatment can be effective with some clients with quite severe problems (Field & Caetano, 2010; Cobain et al., 2011). The research is clouded by the fact that some studies have aimed to evaluate “brief treatment” but the intervention may include a few sessions spread over the same length of time as extended treatment, and also include a process of pre-treatment detoxification and/or a comprehensive assessment (Moyer et al., 2002). Thus, the distinction between brief treatment and the usual-care control condition can be quite blurred.

Is the referral-to-treatment component effective? The evidence is very strong that not only will a SBIRT program identify many people who may need more structured treatment of longer duration and intensity, but many people will in fact follow through with the recommendation if the intervention includes a well-implemented motivational component. Babor and colleagues (2007) suggest that the amount/intensity of the brief intervention may be less important in this area than duration of treatment and inclusion of a component focused on treatment engagement and continuity of care. More research is needed on who benefits most from brief interventions delivered outside specialized substance use treatment services by trained health care and social services


19

professionals, especially with longer term follow-up and drug abusing populations better represented (Field et al., 2010). More data are also needed on the persistence of related outcomes. The need for more research notwithstanding the existing data suggest that: (a) brief intervention (BI or SBIRT) should be formally considered as part of the community’s continuum of care, and (b) they are not necessarily targeted only at people experiencing mild to moderate levels of risk and harm/severity. They should also aim to improve not only alcohol and drug-related outcomes but also engagement in treatment when indicated. 1.3.2 Computer and Mobile-Based Services and Supports As with the addition of brief intervention services to the traditional continuum-of-care, there is another “new kid on the block” in the provision of substance use services and supports; namely those based on Internet and mobile phone technology. There are several ways in which Internet and mobile telecommunications are being used to assist people with health problems, including substance use and mental health problems. The evidence base around these applications is advancing. As of yet, there is no widely accepted categorization of these applications, but the following broad grouping is helpful in a needs-based planning context: 

Mobile telecommunications, in particular text-messaging (SMS)



Internet-based applications, including websites

The kinds of services and supports that can be offered through either of these technologies (or in combination) include: 

Unassisted access to health information (e.g., a website or portal to other websites)



Self-completed screening or diagnostic tests, or structured interventions, such as CBT, with automated feedback



Therapist-assisted counseling (e.g., questions may be posted and a professional responds confidentially; sometimes called e-counselling)



Chat lines, open forums or social networking (e.g., Facebook, Twitter) for mutual aid support or sharing of information with or without therapist mediation


20



Text messaging or emailing to deliver health-related messages, encourage adherence to interventions being delivered by traditional means, provide follow-up support, or obtain evaluation feedback



Mixed methods, for example, using text messaging in conjunction with a manual, diaries, brief telephone supporter and/or weekly counseling appointments



Treatment extenders which provide structured interactive modules that are Internetdelivered and which supplement face-to-face outpatient counselling

It is beyond the scope of the present summary to review the rapidly expanding literature on the use of Internet and mobile communication technologies for prevention, health promotion and health care, including mental health and addictions (see for example Cole-Lewis & Kershaw, 2010; Chou et al., 2009; Budman, 2000; Boulos et al., 2011; Bjerke et al., 2008; Bewick et al., 2008; Andrew et al, 2010; Cunningham & Van Mierlo, 2009; Cunningham et al., 2010; Fjeldsoe et al., 2009; Gold et al., 2011; Koski-Jannes et al., 2007; Robinson et al., 2006; Selby et al., 2010; Shapiro et al., 2010; Wood & Wood, 2009, Lapham et al, 2012; Campbell et al., 2014). Overall, the results are sufficiently encouraging to warrant formal consideration of Internet and mobile technology in the context of needs-based planning and expanding views of the continuum-of-care. Summary: Given the above discourse, the concept of the continuum-of-care based on selected service categories that are matched roughly to problem severity remains valid for the purposes of needs-based planning in the Canadian context. However, the situation is clearly more complex with the inclusion of screening and brief intervention or brief treatment in generic services, and interventions delivered by Internet/mobile technology which can be effective across a wide range of severity. Further the tiered framework for system planning compels one to consider a wide range of nested functions and the role of ALL human service sectors that can organize the delivery of these core system functions in various treatment services (i.e. levels of care) in order to cover all members of the general population organized by severity, risks and harms. In short, the tiered framework expands the thinking and planning based on the continuum-of-care by defining, planning, delivering and evaluating this continuum well beyond the traditional specialized sector. Essentially the tiered framework makes substance use services and supports everyone’s business in order to deliver a comprehensive array of core functions to people across a wide spectrum of


21

severity and contribute to a population-level reduction in alcohol and drug-related harm (Heather, 2012). Figure 4: Reconciling the tiered framework and the continuum of care.

Tiered Services and Sectors (with specialized substance use services seen largely in Tiers 3 to 5)

Tiered Populations (based on risks , harms/severity)

Tiered Functions (universal functions aligned with risk and harms/severity)

Specialized and generic services/settings organized to deliver multiple functions to multiple populations

1.4 Going from Tiered Frameworks to Needs-based Planning Specialized substance use services and supports have traditionally been funded without a comprehensive systems-level, needs-based planning model to help allocate resources by service type and target population, and according to population needs. Annual funding for treatment programs has largely been driven by the budget allocation from previous years with the result that gaps or imbalances in services relative to actual population needs are perpetuated over time. Typically, new resources for substance use treatment become available as a result of new, periodic government strategies or targeted funding opportunities (e.g., services for youth, homeless, and people with co-occurring disorders). However, there are many factors that underlie funding decisions – local advocacy efforts for example – and funds may not be allocated equitably within a jurisdiction on the basis of population needs. In Canada, there is evidence of a substantial gap between the need for substance use services and current availability (National Treatment Strategy Working Group, 2008). However, due to a lack of comprehensive substance use treatment Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

22

information systems and population-level data, the exact size and nature of this gap is unknown across Canada as a whole, and within specific Canadian jurisdictions. Adding to this challenge is the fact that planning efforts to date for substance use services and supports in Canada have been largely focused on the specialized sector of services and have not embraced the broader systems approach articulated in the 1990 Institute of Medicine report, or the more recent tiered frameworks described above. The need for a broader systems approach to treatment related information systems and planning models was recognized in the Systems Approach report (National Treatment Strategy Working Group, 2008). Among the recommendations was a call to “…establish a process for reporting and sharing data on the capacity and use of services and supports based on the [NTS] tiered model…” (pg. 32). As a first step, the National Treatment Indicators Working Group was convened in 2009, under the leadership of the Canadian Centre on Substance Abuse (CCSA), to develop national indicators of treatment utilization. This began with a short list of data elements commonly collected by Canadian jurisdictions on client characteristics and services provided. There is also an effort to align these indicators to the extent possible with indicators from other jurisdictions, such as the US, Europe, and Australia. Beginning with the specialized sector of substance use services, the goal of the Canadian project is to develop the capacity to collect and report information across the broader system of services and supports for substance use as advocated by the tiered model. The National Treatment Indicators Working Group is serving as an advisory body to CCSA and to the National Treatment Strategy Leadership Team in the initiation, implementation, and further development of national treatment indicators focused on service utilization. The working group also aims to support Health Canada in international reporting relating to services and supports for substance use. The Working group released its last report in 2014, available at http://www.ccsa.ca/Resource%20Library/NTS-2014-National-Treatment-Indicators-Report-en.pdf. With a process in place to develop the national and regional picture regarding treatment service utilization and current capacity, a second component is required that would estimate population-based service delivery requirements. With capacity requirements in hand (i.e., need), as well as information on the current capacity (i.e. supply), one would have the basis for a population-level, needs-based gap analysis to inform resource allocation efforts (see Figure 5). Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

23

When applied at the local/regional level, this gap analysis could eventually be supplemented by one more component, namely the impact at the population level of treatment systems that vary in capacity and configuration. Impacts would include substance use-related morbidity, mortality, justice-related outcomes (e.g., driving under the influence), and prevention drivers such as lower per capita consumption of alcohol and other drugs (Babor et al, 2008). With these three components in place (utilization, need, and impact) one would have the core components of a comprehensive systems-level outcomes management system akin to what is feasible at the clinical level (Toche-Manley et al., 2011). In the present project we focus on Component 2 – estimating service requirements based on population needs.

Figure 5: Schematic representation of population-based gap analysis

Estimates of current service utilization (demand and current capacity (supply))

[minus]

Estimates of populationbased service requirements (need)

=

Estimates of population based unmet need (gap)

1.4.1 Needs-based planning models

Rush (1990) published a needs-based planning model focused exclusively on specialized alcohol use services and supports in Ontario and it followed four steps: 1. Determine the geographic area and size of the population served. The model was developed to estimate the required capacity of treatment services for people aged 15 and over. At the time, the regional health planning districts in Ontario (n=40) were the focus of the treatment projections, although any aggregate of these districts was also seen as appropriate. 2. Estimate the number of people with substance use problems/disorders within each population unit, referred to as the “in-need” population. The focus at the time was exclusively on alcohol and a broad, non-diagnostic perspective on alcohol problems. The prevalence estimation Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

24

method was based on alcohol sales data and estimated per capita consumption in each jurisdiction, and using a cut-off of high risk drinking at 35 standard drinks or more per week (approximately 475 grams or more of absolute alcohol). This estimation method, and the exclusion of drugs other than alcohol, was due to the lack of other options, in particular methods based on population survey data. 3. Estimate the number of individuals from step two that should be planned for and treated in a given year. An earlier attempt at a needs-based planning model (Ford, 1985) estimated that the proportion of the in-need population that should be planned for on an annual basis was 20%. This can be referred to as the “demand” or “help-seeking” population. This was an estimate based on estimated relapse rates following treatment, incidence rates of alcohol dependence (i.e., new cases per year), and the goal to ensure that the treated rate would at least exceed the combined impact of recidivism and incidence (i.e., over time penetrate and reduce the in-need population). Survey data were also examined on the proportion of the inneed population that are likely to seek help from the specialized sector of substance use services and a range of values for the demand population was subsequently proposed – 5%, 10% or 15%. In the application of the initial “Rush model”, this estimate could be adjusted locally based, for example, on the nature and scope of local case-finding activity and accessibility of local substance use services. 4. Estimate the number of individuals from step three that will require service from each component of the specialized treatment system. What was required at this stage was a breakdown of the continuum-of-care (specialized alcohol services only) into component parts (assessment/referral, detoxification, case management, outpatient treatment, day treatment, short and long term residential treatment, and aftercare), and specification of the “ideal” pathways between them. This stage drew on published research on patient characteristics and treatment matching, the cost-effectiveness of various treatment settings, and rates of treatment completion. The treatment system, and the flow within it, was modeled in a conceptual diagram and then statistical parameters estimated at various junction points. The final output was an estimate of the number of individuals requiring service in each component of the continuum-of-care, in each planning jurisdiction, and for the province of Ontario as a Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

25

whole. A range of low, medium and high estimates was derived by changing the assumptions for help-seeking (10, 15 or 20%).

This forecasting model was used extensively in local jurisdictions in Ontario as a planning tool but never became enshrined in that province’s addictions services policy as a funding formula. It did, however, become a formal component of funding processes in other parts of Canada, most notably British Columbia, Alberta and Quebec7. Two international applications are also particularly noteworthy. In Australia, the model was used extensively and extended to incorporate a further step that estimated the actual resource requirements for trained FTE equivalents within each type of treatment setting. This important advancement requires critical assumptions and specification of the actual intervention models to be deployed within the various settings, for example, the desired duration of initial screening and assessment protocols; the relative use of group versus individual approaches for delivering interventions; and the ideal duration of day/evening and residential treatment programs. In Birmingham UK, David Best and colleagues extended the Rush forecasting model using a tiered framework and estimated in some detail the optimal linkage between the specialized substance use sector and the criminal justice system. The Birmingham team also incorporated feedback loops derived from documented client trajectories across tiers and proposed the use of routinely-collected outcome monitoring data to re-calibrate the parameters of the forecasting model over time based on the outcomes of clients in the treatment settings to which they were matched. In the end, therefore, the ultimate test of the validity of the model was to be based on the outcomes achieved by clients based on a system of calibration like that described at the program/clinical level vis a vis outcomes-based management (Toche-Manley et al., 2011).

Also in the UK, Drummond and colleagues (2005) were commissioned to assess the need for alcohol treatment in England and the gap between current and required capacity at a national and regional level. They estimated need for treatment based on the prevalence of alcohol dependence. Non-dependent use of alcohol was not included, nor was other drug use or 7

The model remains in use today in Quebec Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

26

dependence. Demand for treatment was based on the past-year utilization of alcohol specialist services (i.e., recent access) and “potential demand” was based on the percentage of people with alcohol dependence seeking help from some non-specialist service provider (e.g., primary care physician). The study was replicated in Scotland in subsequent years (Drummond et al., 2009).

International work on needs-based planning models continues with a project currently underway in Australia to assess the size and nature of the treatment gap based on epidemiology of alcohol and drug abuse and dependence and required care pathways (Ritter, 2012; Ritter et al., under review). A request for proposals was recently released for an update of treatment forecasting models in the UK and aimed at building upon the work of Rush (1990) in Canada. A funding proposal to the National Institute on Drug Abuse (NIDA) is underway for the development of a model appropriate for use in the United States. The model is also currently being adapted for application in the Brazilian context.

1.4.2 Extending the 1990 “Rush Model” to the full tiered framework The present project aims to update the 1990 “Rush model”, at that time developed for the specialized sector of alcohol use services. Major areas for enhancement include: incorporation of drugs other than alcohol; more direct estimation of population-level needs with population survey data; and a broader view of the treatment system based on the tiered framework and new service categories within the continuum-of-care (e.g., home/mobile withdrawal management). The project has been funded through Health Canada’s Drug Treatment Funding Program (DTFP); a program of the National Anti-Drug Strategy (http://www.nationalantidrugstrategy.gc.ca/). The funding period for the project has extended through April 1, 2010 to March 31, 2014, and the project has been implemented through the Centre for Addiction and Mental Health (http://www.camh.net/)8. In addition to the project team located at CAMH, project leadership has been bolstered by Dr. Joël Tremblay from the Université de Québec à Trois-Rivières in a co-Principal Investigator role and also supported by a national Advisory Committee comprised 8

The project team awaits a decision on a proposal to Health Canada to extend the work through to March 31, 2016.


27

of treatment system experts and other key stakeholders from across Canada. This includes, among others, representatives from British Columbia, Alberta and Québec; jurisdictions which had been significant users of the 1990 model for treatment system planning. Also included are representatives at the policy as well as clinical levels, and evaluation experts in substance use systems (see Appendix A for Advisory Committee membership). The project has also drawn upon an International Expert Advisory Panel.

1.4.3 Project Objectives The main questions of interest in this project are: “What is the level of need in the Canadian population for substance use services and supports and what are the capacity requirements to address this need?” The first goal of the project is to estimate the number of people requiring services from each component of the system organized in relationship to problem severity (acuity, chronicity and complexity) according to the tiered framework. The next phase of the project (planned for 2014-16, pending funding) will then turn these capacity requirements, expressed as the required number of individuals to be treated, into service delivery requirements (e.g., number of beds, trained counselors for community services treatment; day/evening treatment slots). It is expected that the results of this research and development process will be useful for three purposes: 

System-level planning and resource allocation so as to achieve more equitable resource distribution and maximum population impact with available resources;



Monitoring and trending key features of treatment system design (i.e., inputs) as part of broader evaluation and performance measurement frameworks;



Advocating for additional resources to address identified gaps. A program logic model for the project was developed to align with the outcomes expected

for the DTFP program nationally (see Appendix B). The anticipated outcomes for the project in the immediate and intermediate terms include:


28



Increased access to models and toolkits for needs-based planning and allocating resources for substance use services and supports;



Increased awareness of decision-makers across Canada of needs-based planning models and their relative advantage over existing approaches;



Enhanced provincial/territorial (P/T) commitments to affect system change in DTFP treatment systems’ investment areas using needs-based planning models.

Longer-term outcomes include: 

Across Canadian jurisdictions, increased use of needs-based planning models for substance use services and systems;



Increased ability to systematically allocate resources to better meet needs of individuals accessing services in all relevant health sectors;



Improved decisions for resource allocation for substance use services and systems;



Increased PT capacity to plan and evaluate substance use treatment systems’ capacity and performance; and



Strengthened evidence-informed substance use treatment systems including: o

System-level outcomes such as a more balanced continuum of services (e.g., residential/community services), improved continuity of care, increased penetration to in-need populations, and improved population health outcomes;

o

Client-level outcomes such as reduced harms associated with substance use and improved outcomes related to health and quality of life.

1.4.4 Assumptions Underlying Model Development 1. Substance use “problems” vary along a continuum of severity comprised of risks, harms and complexity. Substance use related problems exact a high cost to society. 2. A needs-based system of services will be more effective and cost-effective than a system that develops over time solely on the basis of other factors (e.g., advocacy, political pressure, short term priorities). An effective and cost-effective system of services will have an impact at the population-level on the nature and level of substance use problems in the community.


29

3. People experiencing substance use-related problems benefit from assistance in managing these problems. The benefit to individuals, and society as a whole, of providing assistance through public resources outweigh the cost. 4. Many options exist to provide assistance and there is value in providing a range of options in the community and organizing them in such a way that the most costly of these options are aimed at those people who need them and will benefit most from them. While client choice plays a key role in accessing required services, policy-based incentives at the system level, and the use of client-centered, motivational approaches at the individual level are needed to ensure the most intensive and costly interventions are available to those who need them, and not “taken up” by those with less demonstrated need. 5. Characteristics of people in the community can be described and summarized in such a way that reflects the need for different service options and the percentage of people at different levels of problem severity who will seek some form of assistance. 6. Communities will benefit from guidelines on the nature and required capacity of different options for assisting people with substance use problems, as well as support in using these guidelines (e.g., consultation, advice, tool kits, and statistical data). 7. The nature and scope of prevention, health promotion, harm reduction, early intervention and stigma and discrimination programs in the community can influence help seeking behaviour and, therefore, the percentage of people in the community who will seek help. Thus, there is a close connection between planning systems of treatment and support, and systems of prevention and health promotion. 8. A relatively small percentage of people who need assistance with substance use problems seek help from formal helping sources and professionals. Although informal support from family, friends, co-workers, etc. are also helpful, the current supply and mix of helping services and professionals is not sufficient to address the unmet needs in the community. A broader base of services and supports is needed. 9. Unmet need can be met in many ways. One can add new formal services or professionals to the system; adjust the current mix of services or professionals and increase the role of informal supports; and/or increase efficiency through innovative strategies such as more efficient screening and assessment, or the effective use of the Internet and information technology. All these strategies may be needed to some degree, therefore calling for systematic, needs-based planning processes and models and thoughtful prioritization of solutions to address the identified gaps.


30

10. Needs-based planning processes and models need to be tailored to community context, for example, urban/rural/remote communities, and communities of different cultural context. Special sub-populations also need to be considered, such as First Nations, women and youth. There is no one-size-fits-all planning model. 11. Use of these guidelines and supportive tools and processes will contribute to the design and implementation of a service delivery system better aligned with the needs of community members and more needs-based allocation of resources. 12. Statistical, needs-based planning models compliment rather than replace other needs assessment and priorization methods (e.g., key informant opinion; social indicators; geomapping of service availability/accessibility in relation to need).

1.4.5 Evaluating Project Impact In 2011, the project team conducted a baseline survey to determine the current practices in planning and monitoring the performance of substance use services and supports across Canada at the level of the local health region and the province/territory (Health Systems and Health Equity Research Group, 2011). One objective was to determine what data and planning tools or processes were being used to identify gaps in the treatment system and ensure resources were being equitably allocated on the basis of community need. This survey found a high perceived need for better quality local data and more quantitatively oriented planning tools. This baseline survey will allow for a future assessment of the uptake and impact of the needs-based planning model under development in this project on planning processes and treatment system development across Canada.


31

2.0 Defining Need

2.1 Conceptual and Measurement Issues Need: There is no universal interpretation of the term “need” for community needs assessment purposes (Aoun et al., 2004). Needs can be defined in a variety of ways, for example, they can be recognized and unrecognized, private and public, short and long term, and can be seen from the perspective of a variety of groups such as individual, family, group, organization and community, as well as client, program, provider network and regulatory bodies. “Needs” focus on the nature and scope of the problem and are distinct from “wants” which are expressed desires, for example, for help, for resources, etc. Demand refers to actual help seeking. “Supply” is related to availability, accessibility and acceptability. In an ideal situation, needs are balanced with wants, demand, and supply. Bradshaw distinguished between “felt need” (what people say they want or what they think their problems are that need addressing); “expressed need” that is demonstrated by peoples’ use of services; “normative need”, that is need determined by experts based on research or professional opinion, and “comparative need” whereby one person’s (or group’s) needs are evaluated in relation to the position of others (Bradshaw, 1994). In the substance use field, and mental health generally, all these types of needs have been investigated. Normative need has been commonly used in population survey approaches whereby formal diagnostic criteria are applied to responses in community surveys and respondents who meet criteria for one or more mental disorders (including substance use disorders) are considered to be in need of treatment. This was the approach adopted in the earliest psychiatric epidemiological surveys (Regier et al., 1984) and continues to be used in many reports intended to illustrate the “treatment gap” (e.g., Yacoubian, 2003; Cunningham & Blomqvist, 2006; Wang et al., 2005; Drummond et al., 2005). “Met need” is assumed when these same people report use of services (although there are many variations of the specific questions asked), and “unmet need”, or the treatment gap, results when diagnostic criteria are met and services have not been accessed. See Kohn et al. (2004) for a synthesis of psychiatric epidemiological studies defining the treatment gap in this manner. Across


32

37 studies and 25 countries the percentage of people determined to be in need but not accessing mental health services averaged 76.2%. It is now widely accepted that “need” is a multi-dimensional construct and that normative need defined by diagnostic criteria alone is but one approach. Going beyond diagnostic criteria, investigators draw upon other constructs such as complexity/comorbidity, daily functioning, quality of life, and mental distress, although a normative judgment is still made regarding the need for treatment – it’s just done by “experts” using a broader set of criteria and conditions than diagnosis alone. Felt need, or more precisely the “perception of the need for care”, is a complex construct that is influenced by many factors. This includes normative beliefs in the prevailing community and culture; lack of recognition of a “problem” or that the “problem” is related to substance use; lack of awareness that help is available or accessible; belief about the effectiveness of treatment; and feelings of embarrassment, fear or stigmatization concerning seeking help (Sareen et al., 2007; ten Have et al., 2010). In some instances, severe impairment may also result in lack of insight into the severity of their symptoms, for example, with individuals experiencing schizophrenia and also abusing alcohol or other drugs. Some surveys have used a validated instrument on perceived need, the most common being the Perceived Need for Care Questionnaire developed for mental health survey work in Australia (Meadows et al., 2000; Henderson et al., 2000). One specific question posed to survey participants is: “Have you ever felt that you needed help for emotions, mental health, or alcohol drug use. One reason for the high interest in perceived need is because epidemiological surveys have consistently shown that, among those who meet diagnostic criteria for a mental disorder, including substance use disorders, a significant number also report that they do not need treatment. By the same token, many people who report needing services for emotional, mental or alcohol/drug problems do not meet criteria for mental or substance use disorders (Druss et al., 2007). Further, some people may meet some diagnostic criteria but not the full set required for the diagnosis. These are referred to as sub-threshold cases. Many others may not meet any of the diagnostic criteria at all. Thus, “need for treatment” is not a construct that is really present or absent. It is a complex concept best conceptualized as a “series of overlapping


33

constructs, including symptoms, disease burden, treatment effectiveness, and consumer perceptions” (Druss et al., 2007, pg. 1197). Several studies using survey data on perceived need find that those who have other cooccurring conditions and higher severity of the condition in question, including substance abuse, are more likely to perceive the need for treatment (e.g., Codony et al., 2009; Mojtabai et al., 2002). That said, many mild or moderate cases, variously defined, also report a perceived need for treatment and support. While many factors have been found to predict perceived need for mental health or substance use treatment, the specific results depend on the population studied (for example, older adults, adolescents, men versus women, cultural groups). A major challenge in navigating this literature is that people with substance use disorders may not be separated in the analysis since it is more common that substance use disorders/problems are included as one of several mental disorders/problems (e.g., Henderson et al., 2000; Andrews et al., 2001), or omitted from the study analysis altogether (e.g., Brugha et al., 2004). Studies that separate perceived need for mental and emotional problems from alcohol and other drug problems are rare and have been done almost exclusively in the United States. There are, however, a few studies that have made the important distinction between substance abuse and other mental disorders in the data analysis, or focus on substance abuse exclusively (Shepard et al., 2005; Proudfoot & Teesson, 2002; Huang et al., 2006; Teesson et al., 2006; Wu, 2010). In sum, the need for substance use treatment can be defined normatively with diagnostic or other criteria and/or by self-report of perceived need. It follows then that the level of “unmet need” varies with the threshold of severity used to define need for care and the precise questions that are posed about perceived need. To paraphrase Mechanic (2003), the crux of the matter is not the operational definition of need per se but rather to find a definition that clinicians, decisionmakers, researchers and perhaps even the public can agree on as a credible concept. Treatment Outcome and Natural Recovery: A fundamental premise underlying needsbased planning is that treatment through formal publicly-funded treatment services is required for the recovery and improvement or substance use problems. Although the assumption that “treatment works” is well-supported in the literature (Martin & Rehm, 2012; Rush, 2012; Lev-Ran


34

et al., 2012) it does raise questions about the difference between transient need and persistent need, a distinction that requires longitudinal panel data to fully explicate (Timko et al., 1999; Stockdale et al., 2006). When a person is identified with a perceived or normative need at one point in time but not at a subsequent time, and without intervening service use, it is defined as transient need. In the study by Stockdale et al. (2006) about 40 to 60% of those “in need” of mental health or addiction services at Time 1 were not “in need” three years later (i.e., transient need). However, for more than half of respondents with Wave 1 “probable clinical need”, this need persisted over time. Interestingly, about 40% of those with no perceived or “probable clinical need” at Time 1 received some form of care (monitoring, assessment, or treatment in Wave 1, Wave 2, or both). In the substance use field transient need is at the heart of the concept of “natural recovery” (essentially remission of substance abuse or dependence without intervention). Some argue that this needs to be carefully considered in forecasting the need for treatment services, essentially downplaying the role of formal treatment and yielding much more conservative estimates of required treatment capacity. Timko et al. (1999) followed up a cohort of 466 people who had alcohol-related problems, but who had not yet had formal treatment at baseline. By the 8-year follow-up individuals had self-selected into one of four groups: no treatment; completed treatment in years 1 - 3; additional treatment (help in years 1 – 3 and more help in years 4 – 8); and delayed treatment (no help until years 4 – 8). Those who entered treatment in years 1 – 3 had the best outcome at year 8, despite being more severe at baseline. Clearly, treatment was better than no treatment. In addition, however, by the 8-year follow-up, 26% of the untreated participants had been abstinent for at least the prior 6 months, and 54% were free of any drinkingrelated problems. This would suggest that for some percentage of people, their drinking problem can resolve without formal treatment. Looking at it from another perspective, it can be argued that all forms of intervention influence the individual to some degree and help move him/her further along their (natural) recovery pathway (Edwards, 2000). Ultimately, it comes down to the person’s own internal processes and there is no clear way to demarcate the relative influence of personal reflection, informal supports and encouragement, formal coercion or social control, and formal treatment Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

35

advice and structured interventions. From the perspective of the tiered framework, these natural informal supports are included as a legitimate part of the treatment system and, for that reason, estimates of the prevalence “natural recovery” are not defined and factored into the service capacity requirements. Service Use: Other studies have focused on service use as the outcome of interest in the analysis and find that, at the individual level, it is strongly predicted by perceptions of need as well as co-occurring conditions and problem severity (Vasiliadis et al., 2009; Mojtabai, 2009; Rockett et al., 2005; Burgess et al., 2009; Jacobi et al., 2004). As with studies on perceived need, many survey respondents with mild or moderate cases, variously defined, also report using services. Further, the individual or contextual predictors of service use depend on the population studied as well as the way service use is broken down in the survey items or in the analysis (e.g., formal versus informal; specialized substance use services versus generalist services such as primary care physicians). Many studies identify important mediating and moderating factors, for example, substance abuse mediating the relationship between mental health and service use (Maulik et al., 2010), or involvement in the criminal justice system predicting higher likelihood of receiving services for severe mental illness (McAlpine & Mechanic, 2000). Service availability is also an important mediating variable (Kovess-Masfety et al., 2007), as are several demographic characteristics (e.g., Tempier et al., 2009; Schmidt et al., 2007) and psychosocial challenges such as interpartner violence (Lipsky & Caetano, 2008). Importantly, different moderators and mediators can emerge from the analysis of service utilization depending on how service use is modeled in the analysis (e.g., yes/no versus scaled intensity, or continued use) (Elhai & Ford, 2007). The question “what percentage of people with substance use problems actually seek help?” has been tackled by many researchers around the world including Canada. It is very challenging to synthesize the results of this large body of research into a single percentage, or even a small range, given varying definitions of seeking help (e.g., specialized versus non-specialist substance use services) and substance use problems (e.g., alcohol versus or including other drugs; dependent versus or including abuse/problems), and varying time frames for reporting (e.g., past year or lifetime). Results may also be jurisdiction-specific, depending on the prevailing norms about seeking help, existence of anti-stigma and discrimination programs, scope of screening and Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

36

case-finding initiatives, and insurance payment requirements, for example. The wording of the survey question also complicates matters since some surveys ask about seeking help for substance use concerns (the precise wording varies), and others ask about seeking help for substance use, mental health or emotional concerns (that is, not responding only for substance use). This has been a challenge in the present project since the latter type of question is used in the most relevant Canadian surveys. Time frame: Consideration of the time element also requires one to determine, for planning purposes, the relative value of lifetime versus 12-month data on symptoms/diagnosis, related impairments, perceived need for care and service use. Generally speaking, when assessing perceived need or use of services, 12-month data are preferred over lifetime data, if for no other reason than to minimize recall bias. In contrast, other experts in this area feel that it is better to document lifetime disorders and co-morbidities when using this information to predict service use. Kessler and colleagues (2001) have shown that there is often a significant lag between the onset of need as defined by diagnosis or sub-threshold diagnosis, and subsequent service use. In Kessler’s analysis, depending on the country providing data, 50% to 85% of people needing treatment eventually seek treatment based on lifetime data. In that study, the use of 12-month data for both need and service use would have yielded considerably lower estimates of treatment demand and, therefore, service requirements. That said, 12-month data are considerably more relevant for annualized budgeting and service planning and have been used in the present project. Community-level factors: Various conceptual models have been advanced to help understand need and consider its role in predicting service use. The Anderson model is the most commonly used and it distinguishes between need variables (e.g., diagnostic criteria, severity, perceived need), predisposing variables (e.g., gender, age, SES), and enabling variables (e.g., social and family supports). Although this model has been criticized for its inability to “explain” service utilization with a high degree of statistical accuracy, it does provide a useful framework to organize one’s thinking and to encourage reflection on the multiple factors that underlie a decision to seek treatment. Importantly, the Anderson model points one to both individual-level factors (e.g., gender, SES) as well as community-level contextual factors, such as the role of private health care, the proportion of the population with insurance coverage, existing policy that may restrict access, Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

37

accessibility of services, and prevailing cultural norms regarding professional versus traditional, alternative forms of healing. In short, service use is influenced by causal factors at both the individual and community levels, making it a challenge to measure with either survey data or community-type indicators alone. This can, however, be accomplished with fairly sophisticated multi-level statistical models (Diaz-Granados et al., 2010); an approach not yet applied in substance use services research. Age Considerations: The usual interpretation of the “population” for purposes of estimating needs for substance use services includes all people living in a particular jurisdiction irrespective of age. In many respects, this is the optimal interpretation since it allows for consideration of the life course trajectories of problematic substance use and co-occurring conditions – trajectories that often begin in early childhood – as well as trajectories of service utilization, again often beginning in early adolescence, if not before. The practical reality, however, is that health and social services, including mental health and substance use services and supports, are often funded through departments of government with specific age mandates and each with their own needs assessment and decision-making processes. Although this separation is often cited as a major challenge to continuity-of-care in the transition from adolescence to young adulthood, separate funding silos remain entrenched in Canadian jurisdictions, and undoubtedly the majority of other countries. Although there is flexibility in some jurisdictions depending on the individual person seeking help, the usual cut-off for access to children and youth services is 16 years of age. Another reality that motivates the choice of age ranges for needs-based planning is the availability of survey data upon which to estimate normative and perceived need, as well as service use. The best source of Canadian survey data on needs for mental health and substance use services (the 2002 and 2012 CCHS 1.2 survey) included respondents age 15 and over, for example. Accordingly, the present project focuses on ages 15 and over. Geographic Boundaries: For needs assessment purposes another practical consideration concerns the geographic region for which service capacity requirements are to be developed. Inter-provincial differences in need and use of services for mental and/or substance use problems have been documented across Canada (Afifi, 2005). However, health service planning now occurs at a sub-provincial level in many of the provinces and territories (i.e., the level of health Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

38

authorities) and a recent study showed important differences in mental health service use that were associated with both individual and regional characteristics at that geo-planning level (DiazGranados et al., 2010). For the present project, a baseline survey of current planning processes and tools at the local planning region level across Canada was undertaken and established the geographic and jurisdictional boundaries of these planning regions for the purposes of needsbased planning for substance use services and supports (Health Systems and Health Equity Research, 2011). The present project has been focused at the local health planning region. Although higher-level rollup of the results is possible (e.g., to the provincial level).

2.2 A Feasible Measurement Model for Needs-Based Planning in Canada The 2002 Canadian Community Health Survey (CCHS) 1.2 questioned over 36,000 Canadians about mental health and substance use issues (Statistics Canada, 2002). This comprehensive survey used a sampling frame that represented the general Canadian population aged 15 and over, but with several important exclusions to be discussed in more detail later (e.g., people who are homeless, First Nations reserve populations, institutionalized people). As a starting point, the data from this survey were used in the present project to develop estimates of the in-need population. We have used the CCHS 1.2 survey data to estimate the number of people in the population of each of the five tiers of the tiered framework needing substance use treatment and pending funding approval plan to update our data using the results from the more recent 2012 survey. Our goal was to develop these estimates for each of the identified planning regions in the country, drawing upon our baseline survey of these planning regions and the literature reviewed above, while recognizing and documenting the limitations in these survey data. We summarize these limitations in a subsequent section, some of which can be addressed in future attempts to improve the needs estimation model, as well as the survey items themselves. Notably, some of the relevant questions in the survey were revised for our purposes for the 2012 iteration of this Canadian mental health survey.


39

Descriptive statistics on the survey data that are available for public use were run on a large number of variables to familiarize the project team with the data. There were many variables considered as possible indicators of need. Our list of variables for the initial exploratory analysis included:  distress scale (K10)  physical health rating  mental health rating  life satisfaction rating  measures of two week disability  alcohol/drug interference flags9  alcohol/drug problems and dependence  total number of mental disorders10  interference from mental disorders  12 month utilization of any services  chronic health conditions causing restrictions Correlations were generally weak to moderate (.25 to .45). Since the sample size was very large, even small correlations were found to be statistically significant. As one would expect, many of the indicators of problem severity were correlated with the measures of service utilization. An important consideration is that we are producing estimates for local planning areas using national survey data. The CCHS was designed to produce national estimates with only some provinces adding to the sample size for their jurisdiction. We accounted for the age and gender distribution of each local area by calculating percentages of people in each need category (tier) by age and gender. However, national estimates of need may not accurately reflect local needs with absolute precision. Indeed, one expects there to be important regional variations. The percentage of local residents within various age groups in each of the five need categories may also vary between regions. At the same time, using local regional health data to make more precise 9

Flags were based on pre-established cut-off points on the interference scales A subset of mood and anxiety disorders were included; psychotic disorders and personality disorders were notable exceptions from the perspective of the goals of the present project 10


40

projections was not feasible due to a lack of identifying information in the CCHS public data file, as well as the small sample sizes in many regions. Need Categories As described earlier we conceptualized the tiered framework as being stratified by problem severity rather than treatment sectors, services or settings per se. Our aim was to create need categories that were logical, had face agreement with present theoretical and clinical knowledge, and would be useful in separating groups of people according to their need for services as per the tiered framework. The approach was to develop definitions of need within each tier that most treatment system planners and clinicians could identify with for planning purposes – an approach that is based partly on practicality and face validity. This categorization gave us mutually exclusive categories – each person was classified into only one category. The categories conform to a population pyramid, such that there were fewer people in each category than in the one below it, and the need for more intensive treatment and support functions was considered to increase for people in the higher compared to lower tiers. The categories are described below11: Category 1 These respondents were abstainers and light to moderate drinkers or drug users. These are people who need no treatment interventions per se, but rather, primary prevention and harm reduction through health promotion, and exposure to reduced stigma and discrimination programs. Category 2 These respondents were heavy/binge drinkers or heavy drug users who reported few problems related to their substance use and did not meet the DSM criteria for alcohol or drug dependence. Category 3 These respondents experienced four or more substance use related problems OR met the criteria for substance abuse or dependence.

11

Category definitions and CCHS 1.2 variables can be found in Appendix C. Precise code based on the SPSS analysis can be found in Appendix J, bound separately. Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

41

Category 4 These respondents experienced several substance use related problems or met the criteria for substance abuse or dependence AND: 1. Had a positive response to the question "During the past 12 months, was there ever a time when you felt that you needed help for your emotions, mental health or use of alcohol or drugs but didn't receive it?" OR 2. Utilized formal health services because of mental health or substance use issues within the past 12 months OR 3. Showed significant interference in some aspect of their lives from their drug or alcohol use as indicated by the flag variables for alcohol or drug interference (at least 4 out of 10 on any of the 5 interference questions for each of drugs and alcohol).

Category 5 Respondents in this top category were judged to be in need of specialized and intensive medical/psychiatric service functions. People placed in this category met all the criteria of Category 4 AND the criteria listed below: 1. Met the DSM criteria for 2 or more (of five) mental diagnoses (major depression, manic episode, panic, social phobia, agoraphobia without panic) and; 2. Had 1 or more mental disorders with significant interference (using the mental health interference flag variable) for at least one of these disorders and; 3. Had a physical or mental condition that reduced ability sometimes/often in 1 of 4 areas (home, work, school, leisure). With respect to mental health, it is important to recognize that people in Categories 1-4 may experience various mental health related challenges (e.g., psychological distress and/or a specific mental disorder) but not at the level required for inclusion in Category 5. Indeed some mental health-related issues are likely based on the extant literature on co-occurring disorders and these needs must be taken into account in service provision (e.g., ensuring a basic level of CDcapability). The definitions of the five severity categories were applied to the national dataset using SPSS (version 15.0). Individual standardized weights supplied in the CCHS dataset were used


42

to weight the data. Figure 6 shows the distribution of people within the five categories for Canada as a whole. Figure 6: Distribution of the Canadian population across the five severity categories (tiers).

The percentage of people that would fall into each severity category by age and gender was calculated for the entire national sample and projected for local planning tables. This resulted in a total of three tables for each local health planning region (male, female and total), as well as for each province/territory as a whole. We also report the data in four age categories: 15-19, 2024, 25-59, and 60+ (see Appendix J, bound separately). Statistics Canada 2010 census data estimates were used to obtain the population by age and gender in each of the local health planning regions in Canada. The age- and sex-specific weighted percentages in the five tiers were applied to each of the planning areas to obtain the local estimates of the number of people in each category. Wherever possible, resulting tables corresponded to the latest local health planning regions. However, as these health planning regions are continually being revised, there may be some discrepancies between more recent geographic breakdowns and the ones we utilized. Ultimately, health area groupings were chosen to correspond to the planning areas used in our baseline survey of needs-based planning for


43

substance use services and supports in Canada (Health Systems and Health Equity Research Group, 2011). Example – Applying the Model to a Local Health Region For illustrative purposes, we have chosen to use the Champlain Local Health Integration Network (LHIN) geographic health planning area in Ontario. Geographic Health Planning Area: Champlain LHIN, Ontario Total population: 1 044 481 people (aged 15 and over) (2010 Census estimates) The In-Need Population Table: Table 2 shows the estimated total (e.g., male and female combined) for the ‘in need population’ ages 15 and over for the Champlain LHIN, broken down by four age categories, 15-19, 20-24, 25-59 and over 60. The numbers were derived from 2010 Canadian population estimates and the CCHS 1.2 survey as described above, and yield estimates of how the population within the region can be distributed across each of the five need categories. Reviewing the total population collapsed across five need categories and across all age categories (shaded bottom row) yielded estimates of 80.6%, 10.4%, 6.2%, 2.6%, and 0.2% of the total population aged 15 and over, across Tiers 1 – 5, respectively12. The total number of people in need in each tier is shown at the bottom of Table 2.

12

Rounding error causes minor deviation from the national data in Figure 6


44

Table 2: Population Pyramid Table - Ontario LIHN and Males and Females Combined

Total Population Pyramid Projections Using 2010 Population Estimates Example Champlain LHIN Ontario Need Categories 1

2

3

4

5

Total

58230 71.6 50652 56.7 503454 79.7 229473 94.6

10811 13.3 15454 17.3 73207 11.6 9616 4.0

7230 8.9 16776 18.8 37609 6.0 2760 1.1

4921 6.1 5943 6.7 15939 2.5 623 0.3

140 0.2 467 0.5 1177 0.2 0 0.0

81332 100 89291 100 631385 100 242473 100

841808

109087

64376

27426

1784

1044481

80.6

10.4

6.2

2.6

0.2

100

Age 15-19 %s 20-24 %s 25-59 %s 60+ %s Count Total Percent

3.0 Defining the Treatment System for Required Capacity Estimation

The literature on tiered frameworks, studies focused on defining and measuring the need for substance use services and supports, and our operationalization of the many conceptual and measurement issues with Canadian population survey data, set the stage for the next step in our model development – defining the capacity requirements. However, this next step first begs the important question “what are the services for which we are estimating capacity”? Although the project is grounded in the conceptual tiered framework to ensure a broad systems approach (Rush, 2010), this framework also needs to be operationalized in concrete terms for purposes of planning and resource allocation. Our approach was to first revisit the idea of key functions of a system of substance use services and supports that one would consider to be universal across cultures and jurisdictions, although the precise interpretation of these functions may vary across these cultures and


45

jurisdictions. We consider these functions to comprise the essential prescriptive aspect of a comprehensive substance use treatment system. However, as articulated earlier in Section 1.2, these functions can be organized and delivered in a wide variety of non-prescriptive service delivery models, including but definitely not limited to, the currently utilized levels of care/service delivery categories embedded in the ASAM criteria, and similar models for client placement in North America and elsewhere. While innovation is strongly encouraged in how the core universal functions are delivered to people in need of substance use services and supports, our starting place for projecting capacity requirements is based on currently employed service categories that represent a continuum of care conceptualization. This is based on the evidence and clinical practice wisdom that, in general, services within this continuum-of-care should be matched to problem severity to maximize treatment outcome in the context of a stepped care model. However, consistent with the tiered framework these service categories no longer apply to only those service providers specifically mandated for the provision of substance use treatment and support. In short, the service categories for which we will project required capacity align along a re-conceptualized continuum of care, but the services can be located or co-located in virtually any part of the human service delivery system, including through the Internet. Ultimately, the focus is on workforce distribution within a wide range of settings and competencies for delivering the core, universal functions in support of substance use treatment across the entire treatment system, interpreted very broadly. This must also include consideration of Internet and mobile based services and supports, as well as generic services (e.g. SBIRT; addiction liaison and consultation) implemented in a wide variety of settings and sectors. These universal functions are: I. Prevention, Health Promotion and Addressing Stigma and Discrimination Prevention and health promotion are closely related terms and often used together (World Health Organization, 1998). Health promotion is the broader term and in the context of a comprehensive treatment system refers to activities undertaken by service planners, funders, providers and even clients and their families to enable people who have sought help as well as the community as a whole to increase control over and to improve their health. Prevention, or more


46

precisely “disease” prevention, covers measures that not only prevent the occurrence of disease (i.e., risk reduction) but also to arrest its progress and reduce its consequences once established. In this core function we are thinking of primary prevention as distinct from secondary prevention (encompassed in early intervention below) and tertiary prevention (encompassed under interventions). In practice these may overlap, for example, smoking-related policies and activities that may either prevent a client from smoking in the first place or which contribute to smoking cessation or reduction in use and improvement in health outcomes, including other substance and consequences. Activities and policies supporting the function of prevention and health promotion may be at both the community and service delivery level. As noted earlier, this report and the planning model under development are focused primarily on prevention and health promotion services that can be delivered within the context of substance use services and supports while recognizing the importance of population-level initiatives such as advocacy and support for healthy public policy (e.g., use of alcohol in local recreational facilities); support for the creation of supportive environments (e.g., improved community housing and recreational opportunities for youth); strengthening community action for health (e.g., participating on a local public health coalition); behavioural risk reduction; development of personal skills (e.g., support for local health education campaigns activities aimed at increasing health literacy/learning and improving access to services, housing, education and learning opportunities. Many of these same types of activities, and others such as a behavioural risk reduction (e.g. smoking cessation, physical fitness), also apply within the context of delivering substance use services, and are relevant for people at all levels of severity. The function of prevention and health promotion also embodies the concept of self-help, which refers to lay people (i.e., non-health professionals) mobilizing the necessary resources to promote, maintain or restore their health or that of community members. At the individual level this includes “self-care” or “mutual aid” such as self-medication or involvement in supportive groups and various spiritual practices. Again, these are relevant at the service delivery level. Addressing stigma and discrimination has been a major topic of concern and priority for the Mental Health Commission of Canada (Martin & Johnston, 2007) and other national and


47

international bodies. There is an ongoing debate about the use of the terms “stigma” and “discrimination”, with many preferring the single term “discrimination”. It is beyond our purpose here to engage in the debate over terminology but rather, to emphasize the importance of efforts to eliminate/reduce the impact of behaviours, attitudes and lack of knowledge and social conditions on policies that marginalize people or groups through adverse social judgment. There are different types of stigma – health-related (i.e., exclusion at a time when most in need of inclusion, acceptance or compassion); self stigma (i.e., internalized negative attitudes or beliefs about their own condition); and courtesy stigma (i.e., stigma-by-association experienced by those closely associated with stigmatized people) (Martin & Johnston, 2007). Given the scope of these definitions, it is not surprising that a wide variety of interventions have been planned and implemented, ranging from those targeted at the whole population (e.g., working through mass media), to those directly involving people experiencing mental health and substance use problems and/or their families. As above, our primary focus in the planning model is on substance use services. While evaluation evidence needs to improve across the constellation of options, we note here: (a) the close connection between reduced stigma and discrimination at a population level and access to required services and supports; and (b) the role that the overall system of substance use services and supports has to play in reducing the impact of stigma and discrimination among those who access services. This may be through training and education of staff; involving people with lived experience in program planning, governance and evaluation; ensuring organizational and program policies that are non-discriminatory; and provision of person-centered services with a focus on self-confidence, self-esteem and personal recovery. Managers and staff of substance use services may also become involved in a wide range of community-level activities that provide education, challenge stereotypes, and dispel myths of substance use and related challenges such as mental illness.

II. Harm reduction The primary focus of harm reduction is on people who are already experiencing some risk or harm due to their substance use. The most appropriate interventions, whether macro or micro, are those geared to supporting movement from more to less harm. The following is a commonly Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

48

accepted definition of harm reduction: "harm reduction is any policy or program designed to reduce drug-related harm without requiring the cessation of drug use” (CAMH, 2003). Given this definition, at a conceptual level as well as a practical level in the planning and implementation of programs and policies, there is considerable overlap between harm reduction and health promotion, as well as early identification and intervention. Even the treatment interventions delivered to those with severe substance use problems can be grounded in a harm reduction philosophy, for example, flexibility in substance use goals and education related to safe sex practices. The focus of harm reduction policy and programs is the reduction of harmful consequences without necessarily requiring any reduction in substance use, since a change in mode of administration or pattern of substance use may also reduce harm. This focus on reducing harm separates it from zero-tolerance approaches to substance use. Further, while a sense of urgency may have targeted many treatment programs at those substance users who are currently experiencing harm, harm reduction is applied at all levels and all stages of use. Interventions may also be aimed at the individual or community level. At the community level this includes, for example, server intervention programs which decrease public drunkenness; needle and syringe exchange programs which prevent the transmission of HIV among injection drug users; and, environmental controls on tobacco smoking which limit the exposure to second hand smoke. Treatment services may advocate for or otherwise support such initiatives in their community. Our focus in this report is at the service delivery level whereby treatment services may institute policies or undertake many types of activities that support the harm reduction function of the overall system. This may include the following: 

Prioritization of each individual's goals with an emphasis on an immediate and realizable reduction in substance-related harm rather than hoped for long-term outcomes such as abstinence. Harm reduction thus recognizes the central role of the consumer in determining the extent and nature of health care services to be provided.



Provision of a wide range of options, for example, for treatment of opiate dependence, offering options such as drug substitution, drug maintenance and interventions that adopt safer methods of use.


49



A strong emphasis on therapeutic alliance exemplified through a client-centred agreement between a client and their clinician about the treatment approach to be taken based on the expressed needs and desires of the client. Grounded in the knowledge that their very relationship has the power to facilitate positive change, the treatment professional accepts that the client may make less than optimal choices for their health in the short term.



Goal choice can range from abstinence, to the reduction in use of their primary substance, to abstinence from their primary substance but continued use of other substances. Relapse policy whereby a single lapse, several occasions of use, or a return to more regular use is not regarded as a reason to exclude or discharge a client from treatment. The review of treatment goals is ongoing between client and therapist. Harm reduction goals for substance use per se are particularly important for clients with both substance use and severe mental disorders.



Provision of safe sex education and free condoms to help prevent the transmission of infectious disease such as AIDS/HIV and Hepatitis C.



Provision of a drug treatment court as an initiative to keep seriously dependent substance users out of prison, provide treatment, and help to integrate them back into the community. A harm reduction approach to drug use in prison includes methadone treatment and provision of clean needles. The above is by no means an exhaustive list of approaches to operationalize the harm

reduction function of a comprehensive treatment system. The mix of policies and activities in support of harm reduction is culture and context-bound. Further, as part of a commitment to client-centred care, harm reduction is one approach in a broader spectrum that may also embrace an abstinence-based philosophy. III. Early identification and Intervention This function of a treatment system is well-supported by evidence that treatment interventions will be more successful if undertaken “early”. In this context “early” tends to be interpreted as the person using substances at a moderate to high level of risk or who are already experiencing biopsychosocial consequences that, while challenging, do not meet the criteria for substance abuse or dependence. However, “early” may also be interpreted as the person being of young age, or at an early developmental stage. In this context, early identification and intervention Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

50

has both risk reduction (i.e. secondary prevention goals) as well as primary prevention goals, to the extent that screening and intervening for mental health problems (e.g., attention challenges and behavioural problems like oppositional defiant disorder or conduct disorders), may prevent the future onset of substance use problems (Adair, 2009). It is important to distinguish screening that may occur as part of the intake and assessment process of a treatment service (e.g., screening all prospective substance use clients for cooccurring mental disorders) (Rush & Castel, 2011) and a screening process that is embedded in more generic services and undertaken opportunistically among non-treatment seekers (i.e., targeted screening in a primary care or emergency room setting; i.e., SBIRT) (Babor et al, 2007). The literature on SBIRT is summarized in an earlier section (see pages 17-20). Screening in the context of treatment intake and assessment is considered below as part of the function of Problem Identification, Assessment of Needs and Strengths, and Individualized Treatment Planning. The early identification and intervention function can also be operationalized in a treatment system via the Internet and mobile technologies. This has the potential for wide coverage and impact. The substance use treatment system field is in the early stages of the research and development process to identify and disseminate Internet-based approaches (see section 1.3.2). This is also true for mental health generally.

IV. Provision of Information, Engagement, and Linkage Supports, Outreach and Case Management The aspect of this treatment system function this is concerned with the “provision of information” has much in common with the health literacy elements of health promotion. Clients, or prospective clients of treatment services, need to be informed of options available in the program with which they are engaged, as well as other programs in the community. This can be operationalized at a community level through initiatives to increase community awareness, as well as organizations funded to respond to requests for substance use treatment information with subsequent referral/linkage to treatment options (see for example, http://www.connexontario.ca). At the individual level, substance use services may have a standard approach with respect to informing clients, or prospective clients, of their rights as well as Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

51

mutually agreed upon responsibilities upon acceptance into the program, including all associated costs. In addition, clients may have access to agency-level information on treatment outcomes, client satisfaction or other performance-related measurement and evaluation information. Ideally, this information is benchmarked against similar providers and available in a client-friendly format. With respect to the functions related to treatment engagement this is an area of critical importance to retaining clients in the treatment service for sufficient duration and with sufficient exposure to treatment interventions to have an impact. Adopting a motivational interviewing approach to all client contacts at the treatment entry stage is critical. In addition, it is important to recognize the role of coercion in the treatment entry process (e.g., the high percentage of clients that are mandated to present for treatment by the justice or child welfare system) (Rush & Wild, 2003), and that one uses engagement strategies appropriate for these clients (e.g., providing brief treatment options that meet immediate needs but which may stimulate interest and motivation in more intensive interventions at another time). Many services are organized around a stages-ofchange approach that seeks to measure the person’s level of motivation (e.g., precontemplation, contemplation) and triage them accordingly into different treatment options. Other motivational frameworks can also be used that focus on the type/source of the motivation for treatment (e.g., identified, introjected or external) (Ryan & Deci, 2000; Urbanoski & Wild, 2012) rather than the level or stage of motivation. The engagement function can also be addressed by ensuring a welcoming attitude among all staff as well as the creation of a welcoming physical environment (e.g., non-institutional lookand-feel; physical layout; posters with content reflecting a diversity of people across age, gender, cultural and ethnic heritage, for example. Engagement is also impacted by the overall length and efficiency of the treatment entry process, including the intake, screening and assessment tools and processes (Hilton, 2011; Quanbeck et al., 2013). Trained “engagement specialists” may also be employed and incorporated into the intake process to assist in removing barriers to treatment entry, such as transportation, child care, work commitments, and basic necessities such as toiletries and appropriate clothing for appointments or overnight stays in residential programs (Scott& Dennis, 2009).


52

There are many forms of “outreach” models, all of which share the feature of extending the point of service contact into the client’s (or prospective clients) natural environment. This can include street services for marginalized youth or homeless populations; engagement with parents in the home to support participation by youth in treatment; or co-located mental health/addiction workers in schools. With respect to the linkage function, this can be operationalized formally via case management or “wrap-around” services that support clients by linking them with other services in the community. Some treatment systems have created specific positions referred to as “linkage managers” or “system navigators” recognizing the difficulties some clients, or prospective clients, have accessing services and experiencing continuity of their care across multiple service providers. This is particularly needed for the most severe and marginalized client populations, including those with severe co-occurring mental disorders that may experience challenges accessing integrated mental health and substance use treatment (Rush & Nadeau, 2011). Another important development with respect to the linkage function comes from the area of outcome monitoring with the inclusion of a “return-to-treatment” protocol as part of the routine follow-up (Scott & Dennis, 2009; Rush et al., 2012).

V. Problem Identification, Assessment of Strengths and Needs, and Individualized Treatment and Support Planning This function encompasses both screening and assessment, further articulated below, as well as individualized treatment and support planning. Screening is the use of evidence-based procedures and tools to identify individuals with problems, or those who are at risk for developing problems. The goal of screening is to detect these problems and to set the stage for subsequent assessment and treatment – not to provide a detailed description of problem areas or make a diagnosis. There are two stages in screening (Rush & Castel, 2011; Rush, in press): i. Stage 1 – Risk Assessment/Case Finding: The use of single, highly predictive items (Smith et al., 2010; Corson et al., 2004; Ramchand et al., 2009), or very short tools to determine Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

53

either the level of risk associated with substance use (e.g., Saunders et al., 1993) and/or the possibility that a client has any disorder/problem or broad groups of disorders/problem areas. The GAIN- Short Screener (Dennis, Chan, & Funk, 2006; Rush et al., 2013) is an example of a Stage 1 Screener.

ii. Stage 2 - Case Definition: The use of more detailed tools to tentatively identify one or more specific disorders or problem areas, for example, the Psychiatric Diagnostic Screening Questionnaire (Zimmerman & Mattia, 2001), the Modified Mini (Alexander et al., 2008); or for adolescents the Problem Oriented Screening Instrument for Teenagers (Knight et al., 2001; Knight et al., 2003).

Assessment and Treatment Planning: Information captured by assessment tools provides a more individualized identification of the nature and extent of a client’s problem areas and strengths. This information is used to help develop a treatment plan with the client, and can determine which services the client needs to be referred to. As with the screening, there are also two stages in assessment: i. Stage 1 - Information Gathering and Placement: Clients are administered a tool that captures information required to initiate a treatment plan, including the need for more targeted referral/placement for subsequent detailed assessment and treatment planning, for example the GAIN Q3 in the GAIN suite of tools (www.chestnut.org).

ii. Stage 2 - Case Conceptualization/Comprehensive Treatment Planning: Clients are administered tools that identify and describe problem areas and strengths, and how they are interrelated. The resulting information needs to be adequate for case conceptualization, formulation and/or diagnosis and individualized treatment planning. Additional referral may also be indicated. Examples of tools for this purpose include the GAIN-I in the GAIN family of tools (www.chestnut.org) or the Addiction Severity Index (McLellan et al., 1992).


54

Stage 2 Assessment is typically grounded in the present context of the person’s life situation and is problem-focused. This does not mean, however, that one does not include critical underlying factors such as trauma, including inter-generational trauma and related neuropsychological factors. A thorough health assessment, including psychiatric assessment may also be required and is especially indicated for individuals presenting with more complex cooccurring conditions. VI. Delivery of Substance Use Specific and Biophysical Interventions and Supports There is a wide variety of interventions specifically aimed at reducing substance use and ameliorating related problems. The recent research reviews confirm the effectiveness of some, but not all of them (Martin & Rehm, 2012; Lev-Ran et al., 2012). In these recent reviews, the following groupings of interventions were identified: 

Motivational enhancement therapy



Social skill training



Relapse prevention therapy



Behavioural self-control training



Brief interventions



Community reinforcement approach



Cue exposure treatment



Marital/family therapy



Twelve step facilitation



Psychotherapy



Case management



General counseling



Pharmacotherapy. Of the non-pharmacological interventions, those with the strongest empirical support are

motivational enhancement therapy, a variety of cognitive-behavioural interventions and brief interventions. Martin and Rehm (2012) conclude that there is little basis on which to recommend one of the available modalities over another, but good reason to select among them.


55

Martin and Rehm (2012), and Rush (2012) in his editorial commentary on these recent review papers, emphasize the need to consider “therapist effects” in the interpretation and application of this literature on treatment effectiveness. Therapeutic alliance may account for as much as 30% of the variance in treatment outcome, a finding that emphasizes, for example, the importance of empathy, communication, and trust-building in the delivery of interventions. In addition to “therapist effects”, there are many aspects of the treatment service itself that may impact on the effectiveness of the interventions being delivered. These domains have been identified as important aspects of the clients “perception of care” and include issues related to access and entry into the service (e.g., convenience, welcoming); their participation in the treatment process; their rights (e.g., right to privacy and a complaint process); the program environment (e.g., safety, accommodation for disability); discharge planning and continuity-ofcare (e.g., being informed of where to get subsequent support). In addition to interventions specifically for people experiencing high risk substance use and related problems, substance use services and supports typically provide supportive interventions for family members, or more broadly speaking, significant others, who are impacted by their relationship with the person. These may be delivered irrespective of the involvement of the person with the substance use problem. Several authors also note the gap between the interventions with strong evidence of treatment effectiveness (i.e., what we know) and what is routinely delivered in practice settings (i.e., what we do) (Miller, 2007; National Center on Addiction and Substance Abuse at Columbia University, 2012; McGlynn et al., 2003; Lamb et al., 1998). This suggests that needs-based planning include a review of interventions currently being delivered in a given jurisdiction and contrasts this with those identified as most effective in terms of research evidence for the population being served.

VII. Delivery of Substance Use Specific and Highly Integrated Psychosocial, Medical and Psychiatric Interventions and Supports There is a large and still-growing literature on people with co-occurring substance use and mental health problems (Health Canada, 2001) and challenges meeting their needs. In that report, Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

56

and other subsequent literature on the topic (Rush & Nadeau, 2011), the conversation has expanded to include co-morbidity related to physical health, and this level of complexity is firmly embedded in the tiered framework underlying the needs-based planning model.

The idea has been proposed that all substance use services and supports should have a basic capability for treating and supporting people with complex conditions (Skinner, 2005; McGovern et al., 2007). This includes displaying positive attitudes and values independent of problem complexity; competencies for screening and referral; basic, non-diagnostic assessment of substance use and mental health problems and their relationship; interventions such as motivational interviewing, harm reduction and relapse prevention; and case management, including facilitating access to food and shelter, and other basic needs. However, in addition to such CD-capable programs, other programs are needed which fully integrate substance use and mental health treatment and support for individuals with high severity on both issues. This includes provision of, or coordinated access to, care for physical health or other challenges. In these “CD-enhanced” programs, specialized assessment methods may be employed (e.g., psychiatric assessment), as well as specialized therapies such as medication, dialectical behaviour therapy, and other elements of evidence-based treatment for people with severe mental illness and co-occurring substance use disorders such as stage-based assessment, fully integrated treatment planning, CD-family groups and CD-sensitive referral to self-help groups (Mueser et al., 2003). These CD-enhanced services also typically provide consultation on concurrent disorders to other programs and services, as well as clinical training and education. Substance use services can be assessed with structured rating scales to assess whether they are CD-Capable, or CD-Enhanced (McGovern et al., 2007). Similarly, mental health services can be assessed with other tools for their capacity with respect to substance use. Primary care services can be assessed with respect to both mental health and substance use from a CD perspective (http://ahsr.dartmouth.edu/docs/DDCAT_Toolkit.pdf).


57

VIII. Continuing Care/Recovery Monitoring As noted earlier in this report, the literature on substance use services and supports advocates a conceptual shift to a chronic disease or chronic care paradigm that acknowledges the likelihood of variable stages of recovery (e.g., “relapse”) and multiple service episodes over time. This model is especially appropriate for individuals at higher levels of severity. As with other chronic, relapsing conditions there is a need for some level of service to continue after an official discharge. There are many terms applied to these continuing services, for example, continuing care, aftercare, and more recently, recovery monitoring checkups (RMC) (Dennis et al., 2003; Rush et al., 2008). The term “extended interventions” is a catch-all term to apply to post-treatment interventions longer than six months in duration (McKay, 2005). The RMC intervention involves a routine post-treatment follow-up, with a protocol to facilitate treatment re-entry based on defined criteria and, of course, the persons interest in additional services and supports. Results are very strong in terms of health outcomes and overall treatment engagement (Dennis et al., 2003; Rush et al., 2008). The literature on the effectiveness of these continuing services is reasonably strong, but also points toward adaptive protocols that can be adjusted up or down in response to changes in symptoms and functioning over time (as in a stepped care model). Examples of these continuing services include connection to self-help groups such as Alcoholics Anonymous, telephone or periodic face-to-face contacts, regular “alumni” meetings, and more recently, e-mail, text messaging or other Internet/mobile-based interventions such as a web forum with or without therapist support.

Mapping the functions by tier: Table 3 maps these various universal treatment functions onto the five severity tiers, again emphasizing the nested nature of these function, for example, prevention and harm reduction being critical across all tiers. In the next section we turn our attention to the categories of substance use services in which these functions may be delivered. While the needs based planning Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada

58

model will then estimate the required capacity of each of these service categories, system planners must also pay critical attention to operationalizing these universal functions with the best evidence available. These various elements are in fact critical for resource estimation. An analogy would be such that the estimation of the need for different kinds of food outlets in a given community must take into account that the various types of food outlets and food-related services must adhere to essential practices to facilitate healthy meals and sustain a healthy population.


59

Table 3: Mapping the core universal treatment functions onto the five tiers of severity in the tiered framework Function

Prevention, health promotion and addressing stigma and discrimination

Harm reduction

Early identification and intervention

Provision of information, engagement and linkage supports, outreach and case management

Problem identification, assessment of strengths and needs, and individualized treatment and support planning

Delivery of substance use specific and biopsychosocial interventions and supports

Delivery of substance use specific and highly integrated psychosocial, medical and psychiatric interventions and supports

Continuing care/recovery monitoring

Tier 5

X

X

X

X

X

X

X

X

Tier 4

X

X

X

X

X

X

X

X

Tier 3

X

X

X

X

X

Tier 2

X

X

X

Tier 1

X

X

X

X

X


60

3.1 Services Required to Deliver the Universal Treatment System Functions As noted above, the services required to deliver the universal treatment functions described above can span a wide range of sectors, including but not limited to, the specialized substance use service sector. We engaged our project Advisory Committee in identifying the service categories used in their respective jurisdictions, and developed a common set of categories for use in our pan-Canadian model for estimating capacity requirements. We also sought to synchronize our service categories with the work of the Canadian Centre on Substance Abuse’s National Treatment Indicators project to ensure roll-up and eventual comparison to current treatment service utilization, so as to eventually yield a population-based gap analysis. We also further explored service categories used at the local level across Canada via our baseline survey of planning regions (see Appendix I, bound separately). Lastly, we aimed for synergy with the more widely-used ASAM levels of care and placement criteria while also including a service category consistent with the burgeoning literature on Generic Services (e.g. SBIRT). SBIRT is a type of service currently outside the scope of the ASAM categorization system, but widely supported by research evidence if appropriately placed in many generic service delivery sectors (Babor et al., 2007). We also aimed to incorporate Internet and Mobile-based services and supports in some fashion recognizing the strengths but also the current gaps in the evidence base. The service categories required to deliver the universal treatment functions are as follows: 1. Generic Services (e.g. SBIRT) These services take a comprehensive, integrated approach to identify people who have developed or are at risk of developing substance use disorders (Babor et al., 2007). In addition, prevention and health promotion, and as described earlier, brief screening tools are employed opportunistically or universally with non-treatment seeking clients, or identified sub-populations, presenting for assistance or otherwise engaged in generic service delivery settings (e.g., primary care, emergency rooms, social welfare, employee assistance programs, correctional services, or schools, universities and colleges). They are used to identify those at high risk for developing substance use problems that can be effectively ameliorated with brief, but structured intervention, and if needed, referral to treatment. Those with more severe problems, defined on

61

the basis of risk - levels and/or patterns of consumption as well as acuity, or chronicity and complexity of concomitant problems, are referred to treatment for more comprehensive, targeted screening and assessment, and intervention.

2. Withdrawal Management Services These services assist with voluntary withdrawal from substances and are provided at three levels of care – home-based/mobile; community/medical residential; and hospital/complexity enhanced. In addition to various intervention activities, withdrawal management services provide activities aimed at the universal functions of prevention, health promotion and stigma and discrimination reduction; harm reduction; early identification and intervention; provision of information, engagement and linkage supports; problem identification, screening and assessments of needs and strengths and individualized treatment planning; and continuing care. They may provide and/or be closely associated with services offering interventions. Provision of these universal functions has important implications for resource planning – staffing and bed requirements – as well as treatment outcomes.

i. Home-based/mobile: This involves voluntary withdrawal management with support provided in a client’s home or other safe accommodation. It may also involve visits to a central location (e.g., addictions program) during the day, while returning home at night. This service may involve a medical assessment by a physician and regular monitoring by a nurse and health care worker during the withdrawal process in order to provide medical management and support. Before the client is “discharged”, case workers seek to ensure that the client and/or those supporting the client are connected to other substance use treatment services. This category of withdrawal management includes Daytox and sobering centres.

ii. Community/medical residential: This involves voluntary withdrawal management in a nonhospital residential setting. While the environment and supportive services are largely nonmedical, this type of service may involve a medical assessment by a physician and regular monitoring by a nurse and health care worker during the withdrawal process to provide medical management and support. Treatment can be provided with or without drug therapy. 62

Before the client is discharged, case workers ensure that the client and/or those supporting the client are connected to other substance use treatment services.

iii. Hospital/complexity enhanced: This involves assistance with voluntary withdrawal management where care is provided within the structure of a health care setting with a high level of medical and psychiatric capability. Treatment can be provided with or without drug therapy, but typically involves medication management, for example, physical stabilization and withdrawal, and for co-occurring mental disorders. Before the client is discharged, case workers ensure that the client and/or those supporting the client are connected to other substance use treatment services.

3. Community Services and Supports These substance use services are provided on a regular basis (e.g., weekly, bi-weekly) for clients that live elsewhere and are offered at three levels of care: Community Minimal, Community Moderate, and Community Intensive. In addition to counseling and other intervention activities all community services and supports should include activities aimed at the universal functions of prevention and health promotion; harm reduction; provision of information, engagement and linkage supports; problem identification, assessments of needs and strengths and individualized treatment planning; and continuing care/recovery monitoring. There are all important factors for estimating staffing and other resource requirements as well as outcome determination.

i. Community Low Barrier Services: This involves a very limited number of sessions of substance use specific counseling or support activities in individual or group formats. Minimal intervention techniques are also used for clients who are unable or unwilling to access more structured substance use treatment services. Some examples include motivational interviewing and encouraging moderate use of substances or abstinence, or engaging in brief and structured efforts to help pregnant clients stop using substances (Barry, 1993). Community outreach and case management are also included for planning purposes and gap analysis, as well as brief treatment for people who are mandated to participate in substance use treatment by the courts or other social control agents. 63

ii. Community Scheduled Moderate: A scheduled course of one – two hour sessions of addiction-specific counseling in group sessions or individual formats13. iii. Community Scheduled Intensive: A structured schedule of addiction-specific counseling activities taking place over most or all of the days of the week, or evenings.

4. Residential Services and Supports For these services, clients temporarily reside in an environment where substance use treatment services and/or supports are provided. For purposes of model development and application, the project Advisory Committee agreed on three different levels of residential services: supportive recovery/stabilization, residential treatment, and complexity enhanced residential treatment. In addition to counseling and other intervention activities, all residential services include activities aimed at the universal functions of prevention and health promotion; harm reduction; problem identification, assessments of needs and strengths and individualized treatment planning; provision of information, engagement and linkage supports; and continuing care/recovery monitoring. As with withdrawal management and community treatment provision of these universal functions has important implications for resource planning – staffing and bed requirements – as well as treatment outcomes.

i. Supportive recovery/stabilization: This typically includes accommodation and related recovery support provided in a stable substance free environment. Activities may include coaching for daily living, community reintegration, participating in mutual aid supports, and some counseling and case management. Highly structured interventions or programs are not offered in house - counseling and case management may be offered. Stabilization services may be offered to clients prior to assessment, withdrawal management, or pre- or post-treatment in community or residential treatment services and there may be designated stabilization beds that are used flexibly based on individual needs and program 13

Initial model development and pilot testing included, in this category, opioid substitution services with clear provision for counselling component. In the proposal for an additional two years of funding the aim is to introduce a separate service category for opioid substitution.

64

occupancy. These beds are meant for temporary use for either short-term stabilization or as a transition bed. A stabilization or flex bed is for short-term use (generally 24-48 hours), and can be offered either with or without medical supervision. These beds may serve as alterative to a shelter bed, or may provide an alternative location for police to bring an individual who may be publicly intoxicated. A transition bed can be used by clients that may be transitioning into or out of a residential or withdrawal management bed, or who are awaiting housing. Generally, these beds can be occupied for up to 30 days, and can be located in a variety of settings (e.g., at their own stabilization facility, in a hospital, or in a withdrawal management centre).

ii.

Residential treatment: In this type of residential service clients are engaged in a structured, scheduled program of interventions and activities specifically designed to treat substance use problems and/or co-occurring mental disorders. Clients reside on-site and have access to 24-hour support and an alcohol and drug-free residential treatment milieu, including individual and group counselling and many other types of treatment-related activities and interventions.

iii.

Complexity enhanced (medical/psychiatric): A structured, scheduled program of substance use treatment activities and intervention provided for clients with significant medical needs, co-occurring disorders, or other complex needs (e.g., infections, traumatic brain injury, developmental disability). Clients have access to individualized medical or psychiatric care and have 24-hour access to residential support. Drug therapy is a normative element of treatment interventions. In addition to these residential services and supports, the project team and Advisory

Committee struggled with how best to incorporate the housing needs of people with substance use problems. Housing needs are themselves on a continuum ranging from short-term, low threshold shelter14, to supervised supported housing where structured support and counselling may be accessed in-house by residents. Rather than viewing this housing continuum as a sub14

Low-threshold signifies minimal house rules, for example, use of substances may be allowed as long as it does not contribute to disruptive behaviour in the setting.

65

category of residential substance use services and supports per se, it was seen as more appropriately falling within the wide range of psychosocial supports needed by many people with substance use problems. For the present, estimating the capacity requirements along the housing continuum is seen as outside the scope of the needs-based planning model. 5. Internet and Mobile Services and Supports As noted earlier in the report, Internet and mobile-based technologies are emerging as critically important in the delivery of substance use services and supports and mental health services broadly. These technologies are being harnessed to distribute educational/health literacy material, and a range of self-administered and therapist-assisted interventions. At present Internet and mobile services and supports are outside the scope of the needs-based planning model in terms of estimating capacity requirements. They should however be part of the broader needs assessment and planning process.

6. Mutual Aid Resources Mutual aid resources such as Alcoholics Anonymous, Narcotics Anonymous etc. are widely recognized as a key component of substance use treatment and support systems, albeit not formally included in planning and funding processes. However, their important role is acknowledged given the evidence that they are widely used, often serving as the initial access point and continuing support and are an important adjunct to formal services. Many services offer mutual aid groups on site, and/or incorporate these groups in client’s continuing care plans. The research on the effectiveness of Alcoholics Anonymous, for example, shows positive benefits of participation (Humphreys et al., 2004).

6. Substitution Treatment Services As noted above feedback from stakeholders at several pilot sites highlighted the need to incorporate opioid substitution services as a separate service category. This was s significant challenge in many of the pilot site in that there were challenges estimating both need for treatment (e.g., survey data underestimated) as well as current capacity (e.g., prescribing physicians not reporting cases to the same database as other treatment providers.

66

We will work to expand on the demand and capacity requirements for these substitution services in the next stage of NBP model development.

3.2 Schematic Diagram of Needs-Based Planning Model for Substance Use Services and Supports The diagram in Figure 7 is a cross-sectional representation of how people (aged 15 and over) may be distributed throughout the above service categories from the perspective of their needs as reflected in the 5-tiered framework. The movement of clients through the substance use treatment system is seen as complex and dynamic and the diagram aims to represent the potential movement of clients into and between the services. That said, the diagram remains a very simplistic representation of what are, in reality, very complex service utilization trajectories influenced by many factors, including the individual’s profile of needs and strengths, community context, service availability and treatment outcomes. To minimize the complexity of the diagram and the statistical modeling process, each tier is best seen as a cross-sectional representation of these trajectories. In other words, at the present time we will consider neither the health outcomes of services received nor the trajectory of people into or out of the categories of severity (Tier). The challenge remains, from a modeling perspective, to adequately represent highly complex trajectories in order to derive numerical capacity estimates that include flow of people across and within the many service categories. The approach taken in the estimation procedures outlined below was to estimate flow across the large categories of withdrawal management services, community services and supports and residential services and supports within a given year, but not within each of these categories (e.g., community intensive to community moderate, or residential services to supportive recovery). This is less than ideal but acknowledges current limitations of both empirical data and expert opinion for more detailed estimation.

These diagrams and embedded statistical algorithms were developed to assist system planners in conceptualizing client flow and estimating the need for various types of substance use services and supports. We have developed Excel charts that show the capacity requirements for

67

each tier in terms of the number of individuals in need, which facilitate their rapid calculation based on changes/experimentation with the values of parameters in the model. In the next section, we describe our process of estimating the “parameters” in the flow diagram; that is the proportions used to differentiate and allocate individuals into the various service categories in the model, within each tier.  Generic Services (e.g. SBIRT; addiction liaison and consultation)  Withdrawal management (3 levels) o Home-based/mobile o Community/medical residential o Hospital/complexity enhanced  Community Services and Supports (3 levels) o Community low barrier services o Community scheduled moderate o Community scheduled intensive  Residential Services and Supports (3 levels) o Supportive recovery/stabilization o Residential treatment o Complexity enhanced (medical/psychiatric) Capacity estimates for Internet and mobile-based services are not developed at present; nor are estimates for self-help/mutual aid and housing options. We do, however, strongly encourage formal consideration of their role during local needs assessment and planning processes. For the purposes of this report, we present only one diagram for Tiers 2 – 5, which are the same structurally, but which are populated with quite different parameters reflecting the graduated severity of the population in these tiers. Our primary emphasis is on Tiers 2 – 5 treatment services, since Tier 1 is exclusively about prevention, health promotion, harm reduction, and stigma and discrimination reduction services, and the in-need population in Tier 1 (defined previously) is the overall target population for these services and activities. As noted earlier (p.1213), this would run the gamut from alcohol and drug policy (e.g., restrictions on availability of

68

alcohol), community-level interventions (e.g., those focused on determinants of health, server intervention training, or alcohol in recreational facilities), and targeted health education (e.g., school educational curricula, prevention messages in the work place). For purposes of illustration, we will assume that Figure 7 is representing the Tier 4 population. Based on the population defined in our modeling of available population survey data, Tier 4 includes respondents experiencing several substance use related problems or who met the criteria for substance abuse or substance dependence AND: 1. Had a positive response to the question "During the past 12 months, was there ever a time when you felt that you needed help for your emotions, mental health or use of alcohol or drugs but didn't receive it?" OR 2. Utilized formal health services because of mental health or substance use issues within the past 12 months OR 3. Showed significant interference in some aspect of their lives from their drug or alcohol use as indicated by the flag variables for alcohol or drug interference (at least 4 out of 10 on any of the five interference questions for each of drugs and alcohol)

The top of Figure 7 highlights the health region under consideration as well as the specific Tier (for illustration purposes Tier 4). As previously noted community members in this category may experience a wide range of mental health challenges but not at a level sufficient to meet the criteria for Tier 5. The population aged 15 and over is derived from the most recent census data. Two “inflows” to the Total Demand Population are represented – one being “naturalistic” helpseekers who, for whatever reason, have decided to seek assistance for their substance use problem. This, in itself, is a complex process involving a wide range of factors including coercion/social control, or more voluntary acknowledgement and motivation to access services. Some may have been identified and referred from various human services, but not formally screened or previously assessed for problem identification and description. The second pathway into the “Total Demand Population” is via more formalized Generic Services (e.g. SBIRT delivered proactively in primary care or emergency room settings). Some participants will be referred on to other parts of the treatment system (e.g., withdrawal 69

management services, community services, and residential services). This may occur immediately based on exceeding a pre-determined cut-off point in the screening protocol, or may result from a poor outcome with the brief intervention within the SBIRT service. A proportion of those screened will be determined NSR, that is “No Services Required”. A proportion of the Total Demand Population flows to each of the three service categories represented in the middle of the diagram (p5, p6, and p7). The shaded portions of the flow diagram represent the core functions embedded throughout the treatment and support system. This includes, at the top of the diagram, health promotion, prevention, harm reduction, stigma and discrimination, and early identification and intervention programs and policies at a population level. These functions are also represented at an individual/program level in the middle of the diagram, as well as provision of information, engagement and linkage supports, problem identification, screening and assessment of needs and strengths, and individualized treatment support planning. These facilitate treatment entry and engagement. Delivery of substance usespecific and biophysical interventions and supports, continuing care/recovery monitoring, and highly integrated professional, medical and psychiatric interventions and supports are shown at the bottom of the diagram and represent additional functions embedded within all treatment settings and supports, highly integrated mental health and substance use interventions being for more intensive service categories and for Tier 4 and 5 populations specifically. Together, all these functions need to be considered in estimating workforce requirements and competencies needed within all service delivery settings as well as resource requirements and outcome determination.

Under each of the main service categories, the three subcategories are represented; for example, community services – low threshold, scheduled moderate, and scheduled intensive. Each are “allocated” a proportion of cases for that service category (e.g., p9, p10, and p11), for withdrawal management sub-categories). Movement of clients across the three main services is represented by the dotted lines with arrows with corresponding parameters (p18 to p23).

70

Figure 7: SCHEMATIC DIAGRAM OF NEEDS-BASED PLANNING MODEL FOR SUBSTANCE USE SERVICES AND SUPPORTS – TIERS 2 to 5 Region: ______________ Total Population of 15 and Over: ____________ I N - N E E D P O P U L A T I O N I N T H I S T I E R (N = XXXX) Population-level Health Promotion, Prevention, Harm Reduction, Stigma and Discrimination, and Early Identification and Intervention

(p1=) (p4=)

GENERIC SERVICES (e.g.SBIRT; addiction liaison, health educator)

(p2=)

NO SERVICES REQUIRED (NSR)

(p3=)

(p5=)

NATURALISTIC HELP-SEEKERS – DIRECT TO TREATMENT / SUPPORT SERVICES

TOTAL SENT TO SUBSTANCE USE SERVICES AND SUPPORTS

BRIEF INTERVENTION

T O T A L H E L P - S E E K I N G P O P U L A T I O N I N T H I S T I E R (D = XXXX) Health Promotion, Prevention, Harm Reduction, Stigma and Discrimination, Early Identification and Intervention, Provision of Information, Engagement and Linkage Supports, Problem Identification, Screening and Assessment of Needs and Strengths, and Individualized Treatment Support Planning

Internet and MobileBased Services and Supports

(p6=)

(p8=)

(p7=) (p18=)

(p 23=)

COMMUNITY SERVICES AND SUPPORTS

WITHDRAWAL MANAGEMENT SERVICES (p19=)

HOME-BASED / MOBILE

(p21=)

Mutual Aid Resources and Natural Supports

RESIDENTIAL SERVICES AND SUPPORTS (p20=)

COMMUNITY MINIMAL

(p22=)

SUPPORTED RECOVERY

(p9=)

(p12=)

(p15=)

COMMUNITY / MEDICAL RESIDENTIAL

COMMUNITY MODERATE

RESIDENTIAL SERVICES

(p13=)

(p16=)

COMMUNITY INTENSIVE

COMPLEXITY ENHANCED (MEDICAL / PSYCHIATRIC)

(p10=)

HOSPITAL / COMPLEXITY ENHANCED

(p14=)

(p17=)

(p11=)

Delivery of substance use-specific and biophysical interventions and supports, continuing care/recovery monitoring, and highly-integrated professional, medical and psychiatric interventions and support

71

4.0 Populating the Parameters of the Model

To estimate the parameters in the model which would subsequently derive the number of people needing service and supports in each service category within the model, we used two approaches. To estimate the total help-seeking population, we reviewed the literature on naturalistic help-seeking for substance use services (to estimate p2), and the literature on SBIRT (to estimate p1 and p3 - p5) and also incorporated our experience applying the model in the pilot work (described below). To then derive estimates of the total number of people requiring service in each of the nine service categories (p6 – p17) we undertook a national Delphi process to estimate the required parameters for the model in the Canadian context.

4.1 Naturalistic help-seeking (p2) In the 1990 “Rush model” the help-seeking population was estimated at 10, 15 and 20% of the in-need population. This range allowed jurisdictions to choose an estimate deemed to be appropriate for their catchment area, taking into account, for example, local problem severity (the more severe the population the higher percentage of help-seekers) and the availability and geographical distribution of services (the lower the availability and lower population density, the lower the percentage of help-seekers). In choosing this range, the difference between “probable help-seeking” versus “target help-seeking15 was also considered. With respect to “target helpseeking”, Rush (1990) relied on previous work by Ford (1985) who advanced an estimate of 20% as being an appropriate target to aim for in terms of help-seeking. This 20% figure was said to balance incidence (new cases entering the in-need population); treatment success (past cases exiting the in-need population with problems resolved); and recidivism (past cases re-entering the treatment system with continuing or new problems). The 20% estimate was deemed sufficient to achieve a population-level impact (i.e., where treatment penetration and outcome is greater than incidence and recidivism). It has been almost 30 years since Ford (1985) developed his ideal target for help-seeking and treatment system planning, and this target needed to be updated in the context of the present model (e.g., incorporating drugs other than alcohol and different levels of 15

Recent work in the UK draws the same distinction and uses the terms “amenable” and “acceptable”.

72

substance use severity). Updating these estimates was recognized to be a highly complex task. For example, an ideal target should be based on up-to-date evidence and definition of treatment success (i.e., beyond abstinence-only goals and using a chronic care model that accepts relapse and repeat treatment episodes as the typical trajectory within a chronic care model); reflect the increased reach of new treatment models such as mobile withdrawal management and street outreach services; the advent of the Internet and mobile services; and consider the potential impact of large-scale interventions aimed at prevention and health promotion, reducing stigma and discrimination, and informing and motivating people in the general population to seek help before problems become quite severe. In contrast to target help-seeking, probable help-seeking refers to the proportion of the population who need treatment and support for substance use problems who will probably seek help on their own initiative (with or without referral from health or social services professionals), or who are mandated to seek treatment under some condition of social control, such as court diversion, child custody, condition of probation or parole, or school continuance. In the needsbased planning model, we refer to this as naturalistic help-seeking. It does not include those who may be identified through a deliberate intervention such as SBIRT or less structured, opportunistic screening liaison and referral programs.

Table 4 contains a summary of the studies that were reviewed to help estimate probable help-seeking – this is parameter p2. Studies are separated in the table according to lifetime vs. 12month reporting and also according to their reason for seeking help (either specifically for alcohol or drugs, or a broader category used in many surveys for “emotional, mental health and including alcohol and drugs”). Our primary interests in the table are the data on 12-month help-seeking and when the person reports seeking help specifically for substance use problems. In other words, the top right quadrant of the table is of most relevance to the needs-based planning model. Here, there were only three population studies identified – two from the United States and one from Australia. The study by Wu et al. (2003) reports on people with no health insurance and, although interesting in that the data show lower rates of help-seeking, it is of limited relevance to the Canadian context. This leaves two studies (Wu & Ringwalt, 2004; Henderson et al., 2000), both of

73

which report very similar results with 12 - 14% of the in-need population reporting seeking help in the past 12 months. What limited data exist for Canada comes in part from the study by Cunningham and Breslin (2004), based on Ontario survey data and reporting on lifetime substance use problems and lifetime service use for such problems (35.6%). The second relevant Canadian study is reported by Urbanoski et al. (2007) based on CCHS 1.2 data and using 12-month reported help-seeking for mental health, emotional and alcohol/drug problems (13.6%)16. Following the review of this literature we opted to use our best available data on 12-month substance use problems and 12-month help-seeking (top right quadrant of the table) as an interim measure for the needs-based planning model.

16

Consultation and advice have been provided to the department of Statistics Canada responsible for the CCHS 1.2 mental health survey, and the 2012 survey recently released in 2014 improved the wording of the survey items on help-seeking, and yield the Canadian data required for an update of the planning model.

74

Table 4: Summary of literature review on help-seeking for substance use problems Timeframe Lifetime Problems and/or Lifetime Service Use Messias, Eaton, Nestadt, Bienvenu, & Samuels (2007) (40) Alcohol only – 15.6% Cunningham & Breslin (2004) (25); Ontario Alcohol only – 35.6% Service Use Specifically for Substance Use Problems

Cunningham & Blomquist (2006) (41); United States Alcohol only – lifetime past year use – 12%

12-month Problems 12-month Service Use Wu & Ringwalt (2004) (45); United States Alcohol only - 12% men; 13% women Wu, Kouzis, & Schlenger (2003) (46); United States Alcohol and drugs – “uninsured” - 9% Henderson, Andrews, & Hall (2000) (47); Australia Alcohol and drugs – 14%

Schmidt, Ye, Greenfield, & Bond . (2007) (42); United States Alcohol only – lifetime – 16% Kessler et al. (2001) (43); several countries Alcohol/Drugs – 35.5% Kovess-Masfety, et al. (2007) (44); several European countries Alcohol only – 16 - 45% (across all six countries)

Reason for Use

Jacobi, et al. (2004) (23); Germany 12-month substance use disorder, lifetime service use 23 % up to 67% for high comorbidity

Mojtabai, et al. (2002) (24); United States Alcohol and drugs – 7% but higher for comorbidity - 34% Wang, et al. (2005) (48); United States, Alcohol and dugs (any service) – 38.1% Urbanoski, Rush, Wild, Bassani, & Castel (2007) (49); Canada Alcohol and drugs (any service) – 13.6% Codony, et al. (2009) (50); several European countries Alcohol only – 21%

Service Use for “Mental Health”; Emotional, Mental, Alcohol/Drug Problems

Kohn, et al. (2004) (51); International Major review of over 35 studies on ‘treatment gap’ 21.9% Proudfoot & Teeson (2002) (52); Australia Alcohol only – 29.5% Slade, et al. (2009) (53); Australia Alcohol and drugs – 24% Burgess, et al. (2009) (54); Australia Alcohol and Drugs - 24% Teeson, Baillie, Lynskey, Manor, & Degenhardt (2006) (55); Australia and United States comparison Australia United States Male Female Male Female Drugs 17.4% 33.4% 15.2% 21.4% Alcohol 9.2% 16.2% 7.1% 12.9%

75

To summarize, we used what we considered to be the best data available from the US, Canada and Australia, and have estimated probable help-seeking at 13% of the in-need population. The next task was to estimate this percentage for each Tier or need category. To do this, we considered the 13% estimate to be based on our Tier 4 category, since the study population was defined on the basis of alcohol or drug dependence, and excluding those meeting criteria for substance abuse, but not dependence. We then re-examined the literature summarized in Table 4 and separated the reported findings by problem severity – one category higher based on mental health co-morbidity (Tier 5) and one category lower based on abuse, but not dependence (Tier 3)17. Results consistently showed a 2-3 fold increase in the probability of help-seeking for the category of survey respondents meeting criteria for substance dependence plus one or more mental disorders (and the more mental disorders, the higher the increase in help-seeking). We thus increased the estimate for Tier 5 in the needs-based planning model by a factor of 2.5 to yield an estimate of 32.5%. The few studies that separated cases for substance abuse, but not dependence, generally showed a 2-fold decrease in the rate of help-seeking for the less severe sub-population. We considered this population most like our description of Tier 3 severity, and derived an estimate of 6% help-seeking. No study reported data on help-seeking for substance use problems for cases below threshold of substance abuse or dependence criteria such as harmful or hazardous drinking or drug use. Based on the proportionate increases/decreases in help-seeking across Tiers 3 – 5, we projected an estimate of 2% for Tier 2.

The final results of this exercise are shown in Table 5. These were the values for parameter p2 in the needs-based planning model in each of Tiers 2 – 5 that we used going into the pilot testing phase.

17

The majority of cases in Tier 3 meet criteria for substance abuse but not dependence

76

Table 5: Probable help-seeking estimates by tier based on projections from literature review Tier (need category)

% of probable help-seeking by Tier

Tier 1

Non-help-seeking

Tier 2

2%

Tier 3

6%

Tier 4

13%

Tier 5

32.5%

As described in more detail below, the project team has worked with nine pilot sites from 2011 to 2014 to help validate the NBP model and explore its value to local planning processes. The piloting process has been a valuable experience and has served to inform us of adjustments or revisions required in the application of the NBP model. Based on our experience pilot testing the model, we came to view our estimates derived from the survey literature as a minimum expectation for coverage of the treatment system. Specifically we realized that we needed flexibility in model application in some communities that may already be achieving this minimum capacity but which may nonetheless report other indicators of need for more services. Other factors also seemed to be coming into play. In short, much like in the earlier “Rush model” we saw the need to develop a range of possible values for the help-seeking parameter P2. The following considerations were brought to bear in the local sites wanting to use a helpseeking estimate higher than our minimum value from the literature. These considerations included:  There were waiting lists for existing services  Local residents were being referred out-of-region due to a lack of local options/capacity  Local data were available showing a high proportion of cases in non-specialist services with substance use related diagnoses/problems (e.g., hospital admissions, ER visits), and efforts were being planned to increase case finding and referral  High levels of social indicators related to the determinants of health such as employment, housing, low income 77

 Strong public health and prevention programs including programs aimed at stigma and discrimination which may encourage people to seek help  A high percentage of First Nations people in the local population as they are underrepresented in the underlying need estimates  A significant local population working for the armed forces since they too are underrepresented in the underlying survey data In response to these observations we developed a range of help-seeking estimates using the probably help seeking estimates in Table 5 as the minimal requirement for treatment capacity forecasting. Taking into account the various scenarios listed above we began working with the remaining pilot sites in 2013-14 (Chaudière-Appalaches Health Region, Quebec, North East LHIN, Ontario, North Simcoe Muskoka LHIN, Ontario) to choose a help-seeking target that they felt comfortable working with. As our experience grew we saw the benefit of also allowing flexibility in the help-seeking target for different service categories, for example planning for withdrawal management services perhaps requiring one estimate due to a high level of known untreated cases in the local hospital(s) whereas planning for community treatment capacity perhaps requiring another due to local barriers to accessing treatment such as poor weather and driving conditions. The new ranges for the target help-seeking parameters are provided in Table 6. Table 6: Target Help-seeking range options by Tier based on literature and pilot work Tier

Low 5 4 3 2 1

Medium 35% 14% 11% 3% 0%

55% 22% 17% 5% 0%

High 88% 35% 27% 9% 0%

78

4.2 Generic Services (e.g. SBIRT; addiction liaison, health educator) (p1, p3 - p5) Parameters p1 and p3-p5 speak to the potential of introducing systematic SBIRT or addiction liaison and consultation programs to increase penetration of the treatment system into the inneed population. We reviewed the literature to try to derive statistical estimates to help populate the needs-based planning model.

Madras et al. (2009) implemented SBIRT services in a range of medical settings across six states. A diverse population was screened and offered score-based progressive levels of intervention (brief interventions; brief treatment; referral to specialty treatment). Screening covered both alcohol and other drugs, although the tools differed across sites. Of 459,599 people screened, 22% screened positive across a spectrum of risky/problematic use to abuse and dependence. Of these, 15.9% were recommended a brief intervention; 3.2% were recommended a brief treatment; and 3.7% were recommended referral to specialty treatment. For those recommended to brief treatment or referral to treatment, self-reported improvements in several domains were significant (general health, mental health, employment, housing, and criminal behaviour).

Krupski et al. (2010) utilized addiction professionals to proactively screen patients attending an emergency department in a large urban hospital, and provided brief intervention, brief treatment or referral to specialty treatment. A matched control group was developed who did not get screened. Results showed that those screened and getting brief intervention, regardless of getting brief treatment, were more likely to enter specialized alcohol/drug treatment than those not screened (22.5% of the control group, compared to 33.8% of the brief intervention group). The impact was particularly pronounced for those not previously in addictions treatment in the past two years (15% compared to 26.6%).

Bernstein et al. (1997) developed and tested a program model called ASSERT, an intervention based in an emergency department to increase access to primary care, preventative services and the substance abuse treatment system. Workers were trained to screen patients with a health needs assessment, and administer a brief interview based on motivational, readiness-to79

change principles, and an active referral process to the local substance abuse system. Over approximately one year, 7,118 adult patients were screened; and a substance use disorder (SUD) was identified for 41%. Of these, 37% (n=1,096) were enrolled in the ASSERT program. For the SUD-positive cases, 8,848 referrals were made to a wide range of services: primary care (36%); substance abuse treatment (23%); smoking cessation (25%); other supports services (8%). Those keeping a follow-up appointment demonstrated reduced alcohol use and other drinking-related outcomes.

D’Onofrio & Degutis (2010) adapted the ASSERT model developed originally by Bernstein and colleagues, and in an emergency department setting using a systematic screening with health questionnaire. Those with unhealthy alcohol/drug use (defined in the study) received a “brief negotiated interview” with a goal of reducing alcohol/drug use or accepting referral to specialized treatment, depending on severity. Patients referred to treatment were followed up one month later by phone or contact with the program to determine completion of the referral and enrollment in the program. Over five years, 22,534 adult emergency department patients were screened. Of these, 5,533 (24.5%) report drinking over the NIAAA low risk guidelines, and 15.7% used at least one illicit drug. 27% received brief intervention and of these, 63% were referred to treatment. 83% were followed up at one month, and 65% had enrolled in the program. Direct referrals were significantly more likely to be enrolled. The ASSERT program was deemed a major success, and is now fully funded by the hospital.

Dunn & Ries (1997) implemented an on-site integrated alcohol and drug brief intervention across several medical, surgical, and psychiatric services in a county hospital in the US. The intervention integrated assessment, feedback, and referral/recommendations. 95% of 363 patients were referred by hospital staff and 95% were confirmed with a SUD diagnosis. 79% of those accessing services had no treatment history even though results also showed that 20% were fully motivated, and another 54% were partly motivated to attend treatment. When contacted at 1-2 weeks follow-up 35% reported involvement in some substance abuse treatment or 12-step meetings.

80

Tait et al. (2004) focused on adolescents between the ages of 12 – 19 years. The project evaluated the effectiveness of a brief intervention in an emergency department and supplemented the brief intervention with a consistent support person to facilitate attendance for substance use treatment following a hospital alcohol/drug presentation. A control group of nonalcohol/drug presenting adolescents was constructed. At four months, 47% of youth in the intervention group (daily and occasional drug users) had attended treatment. A greater proportion of those in the intervention group had reduced drug use and/or reduced intravenous drug use. The outcome did not depend on participation in treatment. In other words, the brief intervention plus the support person was effective on its own.

Grothues et al. (2008) analyzed data from a previous controlled trial of a brief intervention for alcohol use disorders in a general practice setting to examine impact on help-seeking for specialized alcohol treatment services, and how this impact differed for people with and without co-occurring anxiety and/or depressive disorders. All participants met criteria for alcohol abuse or dependence. Results showed an increase in help seeking for the non-co-morbid group, but not for the co-morbid group. The authors did not report the percentage of the original sample that received the recommendation for referral to treatment, nor the actual number/percentage following through. Results were reported as odds ratios (i.e., did co-morbidity increase or decrease probability of following through).

Published evidence is quite clear that there have been a host of implementation challenges for SBIRT despite the evidence base underlying the intervention (Roche & Freeman, 2004; Nilsen et al., 2006; Johnson et al., 2010; Williams et al., 2011; Anderson et al., 2004). In Canada, we would be challenged to find even one, fairly large-scale effort to systematically implement SBIRT. This is not to say new programs won’t be developed, perhaps spurred on at least in part by this particular project.

On the other hand, there are many examples of less systematic SBIRT interventions underway in Canada. We can point to:

81



The Quebec program for addiction nurse liaison workers in emergency departments, which has now spread across the province, and some new identification and linkage programs in Canadian hospitals (e.g., St. Michael’s Hospital in Toronto);



The DART program in Richmond General in Greater Vancouver could also be considered in a similar vein;



Increased screening for substance use disorders/problems in mental health settings in many provinces, as well as correctional settings;



A new initiative in Ontario that placed mental health nurses in high schools across the province to increase early identification and linkage to community services While these are not structured SBIRT programs, they clearly have the potential to increase

the percentage of people in need who are accessing substance use services (see for example, Blanchette-Martin et al., 2011).

Given the high potential of SBIRT and other structured screening, liaison and linkage initiatives and limited implementation in the Canadian context, it has been decided for the national-level needs-based planning model to set the parameters p1 and p3 to p5 at zero and then encourage and support regional adjustments to these parameters based on actual SBIRT or similar initiatives underway. This is now a key focus in pilot site applications, for example, estimating the nature and scope of practice around what might be termed “case-finding activity” and using this information to select the appropriate target range for help-seeking as shown in Table 6.

4.3 Delphi process to allocate the overall help-seeking population to treatment service categories p6 – p23 The Delphi Method is a structured group communication process that aims to examine complex problems that cannot be measured or evaluated by traditional means. The method draws on the expertise in the field of interest and through multiple iterations of controlled feedback by a facilitator, participants are encouraged to reassess their previous judgments in relation to the group responses in order to achieve consensus on a specific topic. In the present project, the 82

Delphi Method was used to arrive at a national consensus on the parameters of the needs-based planning model. Specifically, participants were asked to estimate what proportion of people presenting for treatment in each category of problem severity (Tier) would require services from the various components of an ideal system of substance use services and supports. 4.3.1. Selection of Delphi Participants An email was sent to the Canadian Executive Council on Addictions (CECA) and the Federal Provincial Territorial (FPT) Liaison Committee members to nominate one to two individuals from their jurisdiction to participate in the National Delphi process. Similarly, at each of the five project pilot sites, pilot leads were also asked to identify individuals that could take part in local Delphi exercises. Inclusion criteria required participants to have several years of experience in system treatment planning or direct substance use treatment/assessment experience.

Individual Delphi Participants Initially, 14 participants were recruited to participate from the CECA and the FPT Liaison Committee. Of these, one participant was removed, as they did not meet the inclusion criteria; one opted not to participate as they were taking part in the Delphi process through a pilot site, and one declined to participate but nominated two other participants from the same province. Furthermore, two participants were actively recruited, one by the Delphi coordinator on the needs-based project team from the initial list of 14 participants, and one by the project lead to represent Central Canada; the latter of which was not recruited from either of the above agencies. The final list of individual participants consisted of nine individuals, representing seven provinces. Two participants from the province of Manitoba completed the exercise together; all other participants completed the exercise individually. There were two individual participants from Saskatchewan, yielding an overall larger regional representation in that province.

Pilot Site Participants The number of participating individuals at the pilot sites varied across the sites and rounds of the Delphi, and are outlined in Table 7. For the provincial pilot sites of Saskatchewan, Prince Edward Island, Nova Scotia and Fraser Health Regional Health Authority participants remained the

83

same throughout each of the rounds. In Saskatchewan, both participants completed all three iterations; for Prince Edward Island, two participants completed all three iterations, with pilot members in the first round participating in either the second or final round. In the province of Nova Scotia and the Fraser Health Regional Health Authority, participating members remained the same for each of the rounds. For the Northern Health Regional Health Authority in British Columbia two members were included in the second round that did not participate in the first round.

Table 7: Delphi Participants by Pilot Site Number of Participating Members Participating Site First Round Second Round Fraser Health Regional Health, 3 3 British Columbia Northern Health Regional Health, 7 4 British Columbia Province of Saskatchewan 2 2 Province of Nova Scotia Province of Prince Edward Island

Final Round 3 2 2

6

6

6

10

5

6

Overall, nine individual participants and the 30 pilot site participants representing various regions and service delivery and planning experiences formed the Delphi panel. The individual and pilot site participants represented eight of the 13 provinces in Canada, providing rural and urban perspectives from across the nation. This diverse representation contributed to a national picture of need and required service trajectories in an ideal system of substance use services and supports.

4.3.2. Data Collection Instrument The schematic diagram of the needs-based planning model and an accompanied chart was completed by all individual and pilot site participants in the First Round of the Delphi. This diagram was a simplified version of Figure 7 of this report (see Appendix D for Delphi materials). As 84

described earlier in the report, the diagram is a cross-sectional representation of how people, aged 15 years and older may be distributed throughout the treatment system based on their needs. This diagram illustrates the flow across the main treatment categories of Withdrawal Management, Community Services, and Residential Supports, and the flow within the three levels of care for each of these categories (e.g. Home-based/mobile, Community/medical residential, Hospital/complexity enhanced for Withdrawal Management Services) for Tiers 2 – 5, which represent the varying levels of need according to the 5-tiered framework.

The accompanying chart for completion included each of the main treatment categories with their respective three levels of care listed in sequence, and client level of need identified by Tiers 2, 3, 4, and 5 listed across. Participants were also provided with space to include their rationale and comments alongside estimates for each of the treatment categories. The template was developed with input from the project Advisory Committee, and feedback from the pilot sites. Its utility for estimating the parameters of the model was tested with the members of the Fraser Health and Prince Edward Island pilot sites prior to the initiation of the National Delphi with the individual participants. This proved valuable as the instructions for the exercise were adapted to ensure all participants knew to include estimates of need for youth clients (down to the age of 15) as well as the need for outreach services, which would be included in the community treatment category- at the minimal level of care. Also the instructions were adapted to acknowledge the stepped-care model of treatment system design and the need for some clients to access more than one of the main treatment service categories in a given year (e.g., from withdrawal management to community or residential treatment).

The template for the Second and Final Round of the Delphi was structured to provide each participant with their own estimates and rationale from the previous completed round, the group estimates (including the mean, median, and range for each of the parameters), and the page number for the accompanying Group Feedback document. This Group Feedback document listed the estimates from all the panel members (pilot and individual participant) and rationale/comments for each of the parameters for Tiers 2 - 5. The template included a section alongside each of the treatment categories for participants to provide revised estimates and 85

comments/rationale for the round in progress. The template and the group feedback document were both developed by the Delphi facilitator on the needs-based planning project with input from the project leader. 4.3.3 Data Collection Round One Individual participants were emailed the project’s Interim Report, a copy of the Schematic Diagram of the Needs-Based Planning Model for Substance Use Services and Supports, and the chart and instructions for completion of the first round. The instructions included a brief summary of the Interim Report, a description of the Delphi Process and its objectives, descriptions of the levels of severity (tiers) and treatment service categories. Examples for completion of the exercise were also provided. Participants were asked to estimate the proportion of people from each of Tiers 2 - 5 that would require services from each of the three main treatment categories and for the levels of care within each of these categories. Participants were instructed that the combined estimates for the main treatment categories for each of the Tiers 2 - 5 could equal 100% or greater but that the percentage within each of these main treatment categories had to equal 100%. Participants were also instructed to draw upon their judgment and experience from their own treatment service system but to provide estimates for a treatment system, where all the treatment service categories are available for their clients. To facilitate this, participants were invited to provide responses that reflected a range instead of an exact percentage for a given parameter, again based on their local experience to the extent possible. Participants were also encouraged to provide the rationale and comments to accompany their estimates. Email reminders were sent to each of the participants during the first round to prompt exercise completion. Nine of the 14 recruited participants submitted responses for the first round, yielding a response rate of 64%. Initial responses to round one varied by participant; primarily the challenges identified included: estimates provided reflected current utilization patterns thus representing projections of demand instead of need; and estimating service need by all clients for all Tiers at the same time. As a result, participants were contacted by the Delphi coordinator through email to receive further clarification regarding completion of the exercise. Six of the eight participants had telephone contact with the coordinator and completed the first round by

86

telephone. For these participants, estimates and rationale were recorded by the coordinator and confirmed by email for accuracy. For the remaining two participants, email instruction was sufficient for the participants to make the necessary changes to their estimates in the first round.

With respect to the pilot sites, the local leads were also provided with a copy of the schematic diagram of the Needs-Based Planning Model for Substance Use Services and Supports, descriptions of the levels of severity (tiers) and treatment service categories, and follow up instructions for completion of the exercise following site visits with members of the project team. The first round of Delphi’s for each of the pilot sites was completed by teleconference, with facilitation from the project leader; the exception to this was for the provincial pilot site of Saskatchewan, which was facilitated by the Delphi coordinator. All pilot site estimates and comments for the first round were recorded by the members of the project team and completed templates were emailed to each of the pilot leads.

Round Two Individual participants and pilot site leads were each emailed their own individual template for completion and the Group Feedback document for review, together with instructions for the second round of the Delphi. Individual participants were asked to review their estimates from the first round in relation to the group’s feedback and, if they wished to, revise estimates based on input from the other participants. Leads from the pilot sites were instructed to reconvene as a group with the members that took part in the first round, in order to complete the second round. Participants were reminded that totals for the estimates for the main treatment categories could be equal to or greater than 100%, and that estimates within each of these treatment categories had to total 100%. All participants were invited to contact the Delphi coordinator and project leader to receive support or further instruction for completion of round two. Email reminders were also sent by the coordinator to all the participants to prompt completion of the exercise.

Responses for the second round were received from all nine individual participants, yielding a response rate of 100%. Of the nine participants, follow-up by email was conducted with two of the participants. In one instance the estimate totals did not equal 100% within one of the 87

main treatment categories. In the second instance a participant requested telephone support for completion of the exercise and this was provided by the coordinator.

Responses for the second round were received from all five pilot sites. Three of the pilot sites requested telephone support. This was provided by the Delphi coordinator for the province of Prince Edward Island, and by the project lead and coordinator for the province of Saskatchewan and the regional pilot site of Northern Health in British Columbia. The second round was completed by each of the leads for pilot sites Fraser Health and Nova Scotia with revised estimates sent to the respective team members for confirmation.

Final Round For the final round, individual participants and pilot site leads were again emailed their own individual template for completion and the Group Feedback document for review, alongside instruction for the completion of the final round of the Delphi. For the final round, participants were invited to further review their estimates based on the group feedback from the second round or to confirm their existing estimates. Descriptions of the levels of severity (Tiers) and treatment service categories were again provided for participants, together with a reminder that the exercise requires projections for an ideal system of substance use services and supports where all treatment service categories are available to treat and support all the levels of severity. Similar to the previous rounds, participants were invited to contact the Delphi coordinator and project leader to receive support or further instruction for completion of the final round. Email reminders were also sent to prompt completion of the exercise. For the final round, pilot site members were also invited to elect one individual to review their pilot site response and confirm with other members that the estimates and comments reflect their regional experience.

Responses for the final round were received from eight of the nine individual participants. For the eight participants, email follow-up was conducted with one of the participants, as their estimate total did not equal 100% within one of the main treatment categories. Telephone support was also provided to one participant.

88

Responses for the final round were received from all five pilot sites, with telephone support provided to one of the sites by the coordinator. Two sites elected to have one member review the pilot site response, with the leads confirming final estimates with the other pilot site members by email; one site completed the final round in a two-member group, and one site completed the round with its original members without support from the coordinator.

4.3.4 Data Analysis For each consecutive round of the Delphi the median, mean, and range was calculated for each of the estimated parameters in the flow diagram. When participants provide ranges for their estimates the mid-point was used to calculate the median and mean for each returned round.

The median response from the final round of the Delphi was used to populate the needsbased planning model. The median for each parameter was rescaled so that the sum of the proportions was equal to 100% within each of the main treatment categories, for example, for Community Services at the Minimal, Moderate and Intensive level. This was a necessary step to ensure that the national in-need projections in the model accounted for the entire help-seeking population. The median was selected over the mean to populate the model, as it allowed for the inclusion of outliers (defined as estimates falling two standard deviations outside the mean for each of the parameters) in the final parameter estimates in the model. This allowed for regional variability and unique perspectives on substance abuse and treatment to be included in the overall national-level model.

For each of the individual and pilot site participants, estimates were also analyzed with each consecutive Delphi round and major changes across the rounds summarized for individual and pilot site responses and the group as a whole. The greatest amount of change in estimates occurred in the second round, averaging 18 changes as compared to an average of just over 12 changes in the final round. The least amount of change over all three rounds for all participants occurred in Tier 2 and the greatest amount of change for all participants was observed for Tier 4. One participant was excluded from the above two analyses, as they did not respond to the final

89

round. Their estimates from the second round, however, were carried over and included in the final group estimates to populate the model.

4.3.5 Results Tables 8 to 11 show the results of the final round, with results from individual and group participants combined and displayed by Tier (one table for each Tier). The rescaled median result for each parameter has been highlighted for each Tier. These tables show the final results used to populate the planning model and generate the estimates of capacity requirements for each planning region in Canada. These estimates were used in each of the five pilot sites engaged in the project and formed the basis of a gap analysis when contrasted with current service utilization statistics.

90

Table 8: Results from the Final Round of National Delphi for Tier 2 National Delphi Estimates for TIER 2 (FINAL ROUND) Participants P6 Respondent #1 Respondent #2 Respondent #3 Respondent #4 Respondent #5 Respondent #6 Respondent #7 Respondent #8 Prince Edward Island Nova Scotia Saskatchewan Fraser Health HA Northern Health HA MEAN MEDIAN MINIMUM MAXIMUM STANDARD DEVIATION

P7 10% 8% 5% 5% 5% 20% 0% 20% 3% 5% 5% 0% 10% 7.4% 5.0% 0% 20% 0.064

80% 90% 90% 100% 100% 100% 100% 100% 98% 100% 100% 95% 100% 96.3% 100.0% 80% 100% 0.062

P8 10% 6% 5% 5% 5% 20% 0% 0% 0% 0% 5% 5% 8% 5.2% 5.0% 0% 20% 0.054

Parameters (%) Withdrawal Management Community Services and Services Supports P9 P10 P11 P12 P13 P14 90% 10% 0% 50% 50% 0% 90% 10% 0% 80% 15% 5% 90% 8% 2% 55% 30% 15% 85% 15% 0% 85% 15% 0% 90% 10% 0% 80% 20% 0% 80% 15% 5% 50% 35% 15% N/A N/A N/A 90% 10% 0% 95% 5% 0% 95% 5% 0% 100% 0% 0% 90% 10% 0% 90% 10% 0% 80% 20% 0% 90% 10% 0% 80% 15% 5% N/A N/A N/A 85% 15% 0% 80% 15% 5% 80% 15% 5% 89.1% 9.8% 1.1% 76.9% 19.6% 3.5% 90.0% 10.0% 0.0% 84.2% 15.8% 0.0% 80% 0% 0% 50% 5% 0% 100% 15% 5% 95% 50% 15% 0.058 0.045 0.020 0.152 0.122 0.055

Residential Services and Supports P15 P16 P17 80% 20% 0% 85% 15% 0% 90% 5% 5% 100% 0% 0% 80% 20% 0% 80% 15% 5% N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A 80% 15% 5% 85% 15% 0% 95% 5% 0% 86.1% 12.2% 1.7% 85.0% 15.0% 0.0% 80% 0% 0% 100% 20% 5% 0.074 0.071 0.025

91


30% 25% 10% 25% 30% 30% 40% 50% 23% 40% 30% 25% 33% 30.0% 30.0% 10% 50% 0.097

P7 100% 95% 80% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 98.1% 100.0% 80% 100% 0.056

P8 20% 23% 10% 25% 20% 30% 15% 40% 10% 15% 20% 18% 25% 20.8% 20.0% 10% 40% 0.082

Parameters (%) Withdrawal Management Community Services and Services Supports P9 P10 P11 P12 P13 P14 70% 25% 5% 30% 60% 10% 65% 30% 5% 30% 67% 3% 80% 15% 5% 35% 55% 10% 50% 45% 5% 15% 70% 15% 50% 45% 5% 30% 60% 10% 50% 30% 20% 30% 40% 30% 57% 25% 18% 30% 60% 10% 25% 70% 5% 20% 80% 0% 72% 26% 3% 27% 66% 7% 70% 30% 0% 40% 55% 5% 60% 30% 10% 30% 60% 10% 65% 25% 10% 35% 45% 20% 40% 50% 10% 30% 50% 20% 58.0% 34.3% 7.7% 29.4% 59.1% 11.6% 63.2% 31.6% 5.3% 30.0% 60.0% 10.0% 25% 15% 0% 15% 40% 0% 80% 70% 20% 40% 80% 30% 0.149 0.145 0.058 0.063 0.106 0.080

Residential Services and Supports P15 P16 P17 60% 35% 5% 45% 55% 0% 40% 50% 10% 50% 40% 10% 60% 35% 5% 50% 30% 20% 70% 25% 5% 60% 40% 0% 76% 22% 2% 75% 20% 5% 60% 30% 10% 75% 20% 5% 75% 20% 5% 61.2% 32.5% 6.3% 63.2% 31.6% 5.3% 40% 20% 0% 76% 55% 20% 0.123 0.115 0.053

92


30% 85% 20% 65% 70% 55% 60% 80% 65% 60% 49% 45% 40% 55.7% 60.0% 20% 85% 0.187

P7 80% 98% 60% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 95.2% 100.0% 60% 100% 0.119

P8 30% 60% 20% 50% 40% 30% 50% 60% 30% 30% 40% 40% 43% 40.2% 40.0% 20% 60% 0.123

Parameters (%) Withdrawal Management Community Services and Services Supports P9 P10 P11 P12 P13 P14 50% 40% 10% 20% 60% 20% 36% 49% 14% 10% 40% 50% 35% 55% 10% 10% 45% 45% 30% 55% 15% 5% 55% 40% 35% 50% 15% 15% 60% 25% 45% 35% 20% 20% 40% 40% 30% 60% 10% 5% 70% 25% 20% 70% 10% 10% 50% 40% 52% 38% 10% 10% 55% 35% 50% 40% 10% 20% 65% 15% 30% 60% 10% 20% 50% 30% 49% 41% 10% 28% 56% 15% 30% 50% 20% 20% 50% 30% 37.9% 49.5% 12.6% 14.9% 53.6% 31.6% 36.8% 52.6% 10.5% 15.0% 55.0% 30.0% 20% 35% 10% 5% 40% 15% 52% 70% 20% 28% 70% 50% 0.102 0.104 0.039 0.071 0.090 0.112

Residential Services and Supports P15 P16 P17 30% 45% 25% 36% 49% 15% 25% 50% 25% 25% 55% 20% 30% 55% 15% 30% 35% 35% 30% 60% 10% 25% 70% 5% 27% 66% 7% 65% 30% 5% 30% 50% 20% 41% 41% 17% 50% 30% 20% 34.2% 49.0% 16.9% 30.8% 51.3% 17.9% 25% 30% 5% 65% 70% 35% 0.117 0.126 0.087

93

Table 11: Results from the Final Round of National Delphi for Tier 5 National Delphi Estimates for TIER 5 (FINAL ROUND) Participants P6

P7

P8

Parameters (%) Withdrawal Management Community Services and Services Supports P9 P10 P11 P12 P13 P14 20% 60% 20% 10% 30% 60% 7% 39% 55% 20% 40% 40% 0% 40% 60% 5% 20% 75% 5% 45% 50% 5% 35% 60% 0% 40% 60% 5% 25% 70% 15% 35% 50% 10% 20% 70% 5% 40% 55% 0% 30% 70% 0% 0% 100% 0% 20% 80% 0% 35% 65% 2% 59% 39% 0% 25% 75% 15% 30% 55% 5% 40% 55% 5% 30% 65% 0% 60% 40% 20% 60% 20% 5% 40% 55% 10% 40% 50%

Residential Services and Supports P15 P16 P17 0% 40% 60% 10% 25% 65% 15% 30% 55% 5% 30% 65% 5% 25% 70% 20% 35% 45% 10% 40% 50% 0% 55% 45% 8% 30% 63% 5% 25% 70% 10% 50% 40% 10% 45% 45% 10% 30% 60%

Respondent #1 Respondent #2 Respondent #3 Respondent #4 Respondent #5 Respondent #6 Respondent #7 Respondent #8 Prince Edward Island Nova Scotia Saskatchewan Fraser Health HA Northern Health HA

50% 85% 90% 80% 90% 90% 90% 100% 95% 100% 90% 63% 83%

100% 100% 60% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

70% 50% 70% 80% 75% 90% 70% 100% 70% 70% 70% 63% 90%

MEAN

85.0%

96.9%

74.4%

4.7%

38.4%

56.9%

8.3%

33.7%

58.0%

8.3%

35.4%

56.3%

MEDIAN

90.0%

100.0%

70.0%

5.0%

40.0%

55.0%

5.3%

31.6%

63.2%

10.0%

30.0%

60.0%

MINIMUM

50%

60%

50%

0%

0%

20%

0%

20%

20%

0%

25%

40%

MAXIMUM

100%

100%

100%

20%

60%

100%

20%

60%

80%

20%

55%

70%

STANDARD DEVIATION

0.143

0.111

0.130

0.063

0.149

0.183

0.067

0.132

0.170

0.055

0.099

0.103

94

4.3.6 Discussion The Delphi proved to be an effective method for populating the parameters of the needsbased planning model and providing a national picture of service capacity requirements in Canada, according to problem severity. In particular, this methodology brought together a group of experts that are geographically dispersed and may not be able to meet and contribute to this important component of the project in a traditional group-based process. Furthermore, each panel member self-identified as an expert in the field with several years of experience in system treatment planning or direct substance use treatment/assessment experience, and this expertise was weighed uniformly among participants. Therefore, the knowledge and experience was considered equally important and relevant between all participating members irrespective of their organizational roles within the mental health and addictions services sector (e.g. director, supervisor, manager team leader). The group feedback for the second and final round was also organized to include each individual panel member’s estimates with accompanied rationale/comments. This information was presented back in its verbatim form, without introducing any potential bias on the part of the project team through synthesis of comments and the removal of outlier views. As described in the Delphi literature, we believe this format allowed for generation of additional insights from panel members, while maintaining anonymity of each participant. Furthermore, members were invited to report ranges for any of the parameters when appropriate. While this was important to identify regional variation, it also reduced the pressure to conform to the remaining panel members when perceptions of local need varied from the larger group’s response. The Delphi exercise was completed after three iterations; which was sufficient to identify points of consensus and acknowledge regional variability.

Major Themes Underlying the Delphi-Based Estimates

There were a number of themes observed during the course of the Delphi exercise that affected the final estimates with which the needs-based planning model was populated; primarily, these themes related to the matching hypothesis, the role of client preference, local variation and the reservation of services for “rare cases”.

95

Matching hypothesis Participants reflected on the severity of substance abuse/dependence and the associated complexity of medical, psychiatric, social, and housing needs when estimating the intensity at which each of the main treatment categories should be provided, for clients represented by each of the Tiers. Comments reflecting participants rationale provided alongside other estimates illustrate that as the level of severity increases, so does their perception of need for more intensive services within all the main service categories (Withdrawal Management, Community Services, and Residential Supports). Listed below are a few of the comments made by the participating panel members: “At Tier 2, most clients are able to manage withdrawal with support from family, friends and/or mutual aid” [referring to Home-Based/Mobile level of Withdrawal Management Service category] (Respondent 1; Tier 2; Round Two). “ At this level of severity, some formal or informal intervention needed/appropriate; this could even include a discussion with a family doctor in a community setting [referring to Community Services and Supports category] (Nova Scotia pilot site; Tier 2; Round One). “The need for withdrawal management will increase from T2 to T3, as will the need for more active and supported treatment” (Respondent 6; Tier 3; Final Round). “There is a difference between the services that should be provided to those diagnosed with substance abuse and substance dependency… those with dependency should be treated aggressively, this means with the least intrusive framework but also with what current best evidence practice is” [referring to Withdrawal Management Service category] (Saskatchewan pilot site; Tier 3; Round Two). “We may want to put them in home support, but the question is, do they have the support they need or alternatively are there individuals with their own addiction issues in these home settings. Also keeping in mind that if we in fact support people in the community at the moderate intensive level, then the home setting for detox is appropriate” (Prince Edward Island pilot site; Tier 4; Final Round). “Individuals at Tier 5 have issues that would be addressed in complex setting” [referring to Complexity Enhanced/Hospital-based level of Withdrawal Management Services category] (Prince Edward Island pilot site; Tier 5; Final Round). 96

“When estimating for this category, was initially looking at level of complexity; however, even supported recovery can be helpful to some. Doesn’t always have to be the most enhanced service to be helpful to clients” [referring to the levels of care in Residential Services and Supports category] (Respondent 4; Tier 5; Final Round).

Role of client/community preference Many of the participants felt that, while the need for more intensive services was clearly correlated with increased level of severity, projections for service need also had to be balanced in concert with client preference and their decisions to engage in treatment at a level of care that may not coincide with a clinician’s assessment (e.g., a clinician’s assessment may suggest community services at an intensive level of care, however the client may prefer to engage in community services at the minimal level of care). Furthermore, the discussion that “going away for treatment” may be more highly valued by community members than receiving care while in the home setting, or returning to the home setting daily, also influenced the estimates of need provided by some participants for service categories. Listed below are some of the comments made by panel members to reflect individual and community preferences: “Tier 4 and 5 may only want minimal [Community Services and Support] treatment; “Tiers 4 and 5 may be reluctant/unable to tolerate [referring specifically to Community Support – at the Intensive level]… for Tier 5 there are concurrent disorders groups that have more latitude, they may be intensive but not as stressful as groups for the general public” (Respondent 2; Tier 4 & 5; Round One). “ [Taking] cultural context into account, so many feel that they need, or should go away – many people think they need to be put into a residential type program, so this is where they are placed [referring to Residential Services and Support category] (Northern Health pilot site, Round One).

Local variation Estimates for service need were also impacted by the local regional experiences of participants. Specifically, participants commented on the existing policies of their current specialized substance use services and the need for a client to receive services from one of the main treatment categories before being accepted into another service category (e.g. withdrawal

97

management services mandated by formal residential services). Other concerns shared by participants related to the unavailability of all service categories to meet clients’ needs and the resulting demand that this placed on certain treatment categories in order to compensate for what was missing in the existing system structure. Lastly, geographical distances and the regional challenges with safe, affordable housing options also affected participants’ estimates. Below are a number of participants’ comments that relate to the issues described: “One caution we would like to note in this section is that in Mb our 2nd stage housing resource requires clients to complete a treatment program first in order to participant in 2 nd stage housing. As a result we thought most tier 2 clients would get “supported Recovery” through the “Self-Help” groups. Our experience also would imply that not many Tier 2 clients would require 2nd stage housing based on the stability in their life areas” (Respondent 3; Tier 2; Round Two). “We sometimes do see youth at this level of service engagement (w/mgt) and often if they look like they will require or request residential services, admission to our detox facility is part of the stabilization/detox package prior to direct entry into residential services” (Fraser Health pilot site; Tier 3; Round Two). “Difficult to keep within 100% as individuals go to both supported recovery and residential services” (Respondent 2; Tier 3; Round Two). “We don’t have day programming at the intensive level, although we used to. If they needed this level they would be referred to a residential setting [referring to the Intensive level of care in the Community Services and Supports category] (Respondent 4; Tier 4; Final Round). “This change in percentage reflects client needs in tier five situations. We lack consistent/integrated P11 resources” [referring to Complexity Enhanced/ Hospital –based level of care in Withdrawal Management Services category] (Respondent 3; Tier 5; Final Round). “In an ideal world, intensive day patient programming would be available in all regions to meet the needs of this client group” [referring to Intensive level of care in Community Services and Supports category] (Respondent 7; Tier 5; Round Two). “Due to the extensive geographical distances served, if treatment service are centralized then there is more need for residential support/services” (Northern Health pilot site; Tier 3; Final Round). “Additional housing support is needed in the North. Housing with supports is needed in all communities, not just centralized locations. Size of the population plays a major role, so planning is

98

used to prevent unequal distribution of residential services and supports” (Northern Health pilot site; Tier 4; Final Round). “All of them= 100%” [referring to Community Services and Supports treatment category] (Respondent 8; Tier 3; Round One). “May give 5% to minimal because this may be all that they are able to manage” [referring to Community Services and Supports category] (Respondent 4; Tier 5; Final Round). “I like the comment that we need minimal as an option because that’s where some folks start. It’s also where some, with strong social support and personal commitment get their solution. But most need more intensive and extensive treatment” [referring to the levels of care within Community Services and Supports category] (Respondent 6; Tier 5; Final Round). “The growing ability to do home based and daytox interventions will draw more people to WMS in an ideal world. Most in Tier 5 will need more involved and supported options (Respondent 6; Tier 5; Final Round).

Reserve for rare cases When providing estimates of need for the various service categories, some participants considered clients with unique circumstance that may require their use of more or less intensive services. For example, this may have included a client with a Tier 2 level of severity that requires some level of residential services and supports because they have children at home and need a safe place for some ‘sober time’. Alternatively, a client experiencing a Tier 5 level of severity may be able to safely withdraw in the home setting, despite medical and psychiatric complexity, if an adequate level of social support is available to them. The comments below reflect some of the discussion from the panel members regarding the rare cases: “Tier 3 individuals may be on the cusp of Tier 4, or by stopping their drug, they may start needing more complex services” [referring to the levels of care for Withdrawal Management Services category] (Prince Edward Island pilot site; Tier 3; Final Round). “Still low intensity clients, but you may have some complexity therefore increased estimates to 5% in P17” [referring to the Complexity-enhanced (Medical/Psychiatric) level of care in the Residential Services and Supports category] (Prince Edward Island pilot site; Tier 3; Final Round).

99

“Five percent [for Residential Services and Supports main treatment category] because there are always some outliers”; “10% would need this level of WMS for the ‘safe setting’ element versus requiring this level of withdrawal management (e.g. medication) [referring to Community/medical Residential level of care in Withdrawal Management Services category] (Respondent 5; Tier 2; Round One). “After looking at the others comments I would agree that for a small percentage of clients some form of home based detox may be the best option and have amended my numbers to reflect that” [referring to levels of care in the Withdrawal Management Services category] (Respondent 7; Tier 5; Final Round).

Challenges There were a few challenges observed during the Delphi process. Some of these challenges related to: the structure of the form for reporting in the first round and the method for communicating instructions; maintaining consistency in the pilot site participants and completion of the exercise throughout the process; challenges with estimating for service categories not directly available in certain jurisdictions and for levels of severity (Need) that may not be commonly seeking treatment directly from the specialized system.

Instructions in Round One The Schematic Diagram of the Needs-Based Planning Model for Substance Use Services and Supports was evolving in the initial stages of the Delphi; youth 15 years and over were added to the in-need population, and outreach services was also added to the existing definition of the service category of Community Minimal services. The visual presentation of the chart accompanying the schematic diagram was reported as somewhat misleading by a number of the participants. The instruction for the first round referred to estimates across the main treatment categories and within each level of care; however, the chart was developed so that all treatment categories were listed in sequence and the Tiers 2 -5 were listed across the top of the document. One individual participant reported that the schematic diagram was easier to follow and conceptualize the flow of clients and therefore, provide estimates that were in accordance with the initial instruction.

100

Two individual participants recommended a teleconference or a brief ‘mock’ exercise to help facilitate understanding of the instruction, particularly since communication was primarily via email correspondence. The individual participants that completed the first round of the exercise with the Delphi coordinator by telephone reported that this form of guidance was helpful and that the task became clearer after the completion of the first severity category (Tier). Similar findings regarding comprehension of the exercise were reported by the members of the pilot sites during teleconferences with the project leader and/or the Delphi coordinator.

Pilot Site Participation We had initially requested consistent representation of participants in the pilot sites from the first round onto the final round. However, we found this was not feasible given the summer holidays. There were some changes between the pilot sites in completion of the second round; leads for two sites (province of Nova Scotia and Fraser Health Regional Health Authority) had completed the exercise independently and then confirmed changes with their fellow pilot site members through email. The provincial pilot sites of Saskatchewan and Prince Edward Island, and British Columbia’s Northern Health Authority had requested telephone support in completion of the second round. Similarly, pilot sites opted to complete the final round also using various methods, two sites had elected leads to review and finalize estimates, Prince Edward Island received telephone support, and Northern Health completed the final round as a two-member group, while Fraser Health completed the final round with its original members.

Conceptual Challenges Participants were asked to provide estimates for services that may not currently be available in their region, for example, complexity-enhanced withdrawal management services for those with severe co-morbidity, or home-based/mobile withdrawal management for those with mild-moderate withdrawal symptoms. While this challenge was reflected in the comments made by the panel members primarily in the first round, it continued in each of the rounds. Similarly, not all participants were familiar with providing services for each level of severity (need) and, therefore, often started with a level that they were comfortable providing estimates for then and adjusted estimates for the remaining Tiers accordingly. Specifically, a number of participants 101

expressed concern regarding estimating service need for Tier 2 level of severity, as often times these clients did not receive care in the participants’ current formal substance use service system.

The model was also developed to acknowledge movement of clients throughout the system; specifically, between the main treatment categories of Withdrawal Management, Community Services, and Residential Supports. Although helpful to participants, this was still seen as limiting as it did not include movement within the levels of care. A number of participants felt it was challenging to provide estimates within each of the main treatment categories (e.g. Supported Recovery, Residential Services, and Complexity-enhanced) that were equal to 100%, as again, the current practice in their region allowed for client movement within; with a requirement often to receive residential services at the lowest level of care before being granted access to a higher level of care in this category. Participants agreed, however, that this was too complex for the current Delphi process and needs-based planning model.

The initial timeline of the Delphi was set for 11 weeks; however, with the challenges described above, particularly those experienced in the first round, the timeline was revised early on in the process and the Delphi was completed at 18 weeks.

5.0 Pilot Testing and Gap Analysis

The Needs-Based Planning model has been piloted at nine sites across Canada from 2011 to 2014, as follows: Fraser Health, BC; Northern Health, BC; province of Saskatchewan; North Simcoe Muskoka LHIN, ON; North East LHIN, ON; Sarnia-Lambton, ON; Région de la ChaudièreAppalaches, QC; province of Prince Edward Island; and province of Nova Scotia. Descriptions of the pilot site areas and the results are provided below. The nine sites were located in six provinces and covered a wide range of geographical regions across Canada, including both urban and rural areas. The characteristics of the pilot sites varied greatly with respect to population (populations ranged from 140,000 to over 1.3 million people); diversity within the region (several sites have significant First Nations populations); and geographic diversity (remote and isolated, as well as densely

102

populated communities). We piloted the model at both the local/regional health authority and the provincial levels to determine issues in the application of the model for different health care delivery systems and planning levels.

Aside from demographic aspects, there were other factors that influenced pilot site choices. Fraser Health and Quebec had prior experience using the 1990 Rush model, and we anticipated that their familiarity with the earlier work would facilitate application of new model. Northern Health had previously used the tiered framework in their planning processes, but not specific forecasting procedures as outlined in the 1990 Rush model. Piloting in Prince Edward Island and Nova Scotia helped us assess the applicability of the model to Atlantic Canada but also (in PEI) in a jurisdiction under one centralized health authority (Health PEI).

5.1 Pilot Site Results

1. Fraser Health, British Columbia provides a wide range of integrated health care services to more than 1.5 million people living in communities stretching from Burnaby to White Rock to Hope, and offers a mixed urban and rural orientation on substance use services and planning. The communities served by Fraser Health include approximately 38,000 First Nations people, and large Asian, Indo-Canadian and Filipino populations. Fraser Health provides substance use services including prevention (e.g., needle exchange); outpatient programs (e.g., group support, case management); residential programs (e.g., support recovery homes); and withdrawal management (e.g., medical withdrawal management, stabilization beds).

103

Fraser Health Gap Analysis Withdrawal Management Service

Required Capacity

Current Capacity in Specialized Services Program name

# cases per year

Total number of cases per year

Gap

Home-based/Mobile

2,140

Riverstone

356

356

1,784

Community/Medical Residential

2,228

Creekside

1,516

1,516

712

Hospital/Complexity Enhanced

746

20

20

726

-

1,892

3,222

# cases per year


Gap

TOTAL

Community Services and Supports

Burnaby

-

5,114

Required Capacity


Community Minimal

4,618

Brief Treatment

4,945

4,945

-327

Community Moderate

6,255

Outpatient

3,409

3,409

2,846

Community Intensive

2,372

Day Treatment Dewey (youth)

185 38

223

2,149

-

8,577

4,668


Gap

776

603

1,157

293

TOTAL

Residential Services and Supports

13,245

Required Capacity

Supported Recovery

1,379

Residential Services

1,450

Complexity-enhanced/ (medical/psychiatric)

703 TOTAL

3,532

-

Current Capacity in Specialized Services Program name Inner Visions Last Door Path to Freedom Phoenix Valley House Charlford Hanna Liz/Ellendale Mollie's Westminster Maple Ridge Kinghaven Treatment Centre Peardonville House Portage (youth) Heartwood Centre

# cases per year 197 56 41 100 29 63 168 47 38 37 634 374 121 12 16

Burnaby Centre

64

64

639

-

-

1,997

1,535

 Fraser Health served as an excellent example for home-based/mobile withdrawal management service.  Conducting the gap analysis exercise in Fraser Health brought the idea of stabilization/flex beds to the forefront, and suggested that this service was not adequately communicated in the NBP

104

model. Subsequently, we discussed this service category with other pilot sites across Canada, and several, including the North East LHIN, confirmed that stabilization/flex beds should be better reflected.  Fraser Health has a large number of unregulated support recovery houses. These facilities are often for-profit, and are run independently of the Health Authority. There are no standards that these independent agencies are required to adhere to, and no mechanisms to monitor service or quality. Although these independent agencies likely provide sufficient beds to provide the ‘demand’ population with a place to stay, the services and supports they are receiving may not be adequate to meet their treatment needs. Stakeholders have indicated that this is causing a ‘quality gap’ rather than a ‘capacity gap’. This posed some challenges for the gap analysis and highlighted the need for careful consideration and exclusion of private/for-profit services.  There are a large number of correctional facilities in the Fraser Health region, and as a result, many of Fraser’s residential beds are filled with clients from outside of Fraser. This makes it challenging to properly assess the gaps in residential services, as the need estimates are based on local population data, and do not take into account the people coming from outside of the health planning region.  Pilot site stakeholders noted major gaps in: housing; services for women and women with children; and getting clients into the right kinds of beds/into the appropriate treatment service.

2. Northern Health, British Columbia Northern Health covers almost two-thirds of British Columbia and is home to approximately 300,000 people, where approximately 18% of the population is Aboriginal – the highest proportion in the province. Substance use services are available for youth, adults and the elderly, and Northern Health provides a wide range of substance use services in a variety of community and residential settings. Services vary from shortterm assessment and treatment to long-term programs for those with a serious and persistent

105

mental illness and/or substance abuse problems. Substance use services are well-integrated with mental health services. Northern Health Gap Analysis Withdrawal Management Service

Required Capacity


# cases per year


Gap

0

386

Home-based/Mobile

386


401

WM supportive recovery

336

336

65


134

Adult medical detoxification Youth medical detoxification

603 9

612

-478

-

948

-27


Gap

TOTAL

Community Services and Supports Community Minimal

-

921

Required Capacity


-

-

0

832

Methadone Youth Community Services Adult Community Services

264 284 544

1,092

36

Day Treatment

463

463

-36

-

1,555

832

# cases per year


Gap

38

38

211

-

0

261

68

68

58

-

106

530

832

Community Moderate

1,128

Community Intensive

427 TOTAL

Residential Services and Supports Supported Recovery

249


261


126 TOTAL

-

2,387

Required Capacity

636

# cases per year

Current Capacity in Specialized Services Program name Legion Wing

Youth Residential Services

-

 A comprehensive continuum of services is offered in Prince George (the largest municipality in Northern Health); however, few other services (e.g., home-based/mobile withdrawal management; residential services) are offered across the rest of the region, due to its rural/remote geography.

106

 We were unable to complete the gap analysis process in Northern Health, as the data in the region were either not available or were incomplete. Several of the issues relating to data availability in Northern Health included: o Not all service providers reported data to SYNAPSE, the region’s data collection system. This includes the eight First Nations Friendship Centres and ten supported housing facilities. o Mental Health and Addiction services are very integrated in Northern Health; intake and services are offered for clients with both mental health and substance use problems, making it challenging to determine service utilization specifically for substance use problems. The data in the electronic data system do not allow one to tease out specific diagnoses, making it particularly challenging to determine current utilization of Community Minimal services for example. o Only approximately 25-35% of clients have a diagnosis on file, as staff doing client intake may not be considered qualified to provide a diagnosis. This also leads to challenges in determining how many clients are accessing services specifically for substance use problems.  Housing is one of the biggest issues in the region, and stakeholders felt that the supports from BC Housing were not adequate to meet the needs of the local population. Housing is a particular problem in Prince George, Northern BC’s largest city. Many people coming from more rural/remote regions, or from First Nations reserves usually come through Prince George, and as a result, the homeless and transient population is fairly large in the city, and housing supports are needed.  Pilot site stakeholders noted major gaps in: housing; short-term residential services; home/mobile detoxification; and brief treatment.  Due to the rural/remote geography of the region, ‘visualizing’ how home/mobile withdrawal management might operate was considered to be challenging, and this service would likely not be applicable in many of the sparsely populated areas of the Health Authority. 107

 Due to challenges related to data collection and mental health and addiction integration, we determined that the NBP model may not be a useful planning tool in Northern Health.

3. Province of Saskatchewan is home to approximately 1 million people and offers a mixed urban and rural orientation on substance use services and planning. Approximately 15% of Saskatchewan’s population identifies as Aboriginal. Substance use services in Saskatchewan are provided through the province’s 12 Regional Health Authorities, and include detoxification services (e.g., stabilization); inpatient services (e.g., intensive supports services); long-term residential services (e.g., life skill training); and outpatient services (e.g. counseling, screening and assessment). Saskatchewan Gap Analysis Withdrawal Management Service

Required Capacity


Home-based/Mobile

1,358

Scaled up from AMIS


1,411

AMIS and ADG combined


469

TOTAL


3,238

Required Capacity

Scaled up from AMIS

-

# cases per year


Gap

15

15

1,343

2,920

2,920

-1,509

185

185

284

-

3,120

118

# cases per year


Gap


Community Minimal

2,917

Combined AMIS and ADG

4,390

4,390

-1,473

Community Moderate

3,965


4,327

4,327

-362

Community Intensive

1,500


731

731

769

-

9,448

-1,066


Gap

78

78

796

TOTAL


8,382

Required Capacity

-

Current Capacity in Specialized Services Program name # cases per year

Supported Recovery

874

Scaled up from AMIS


918

Combined AMIS and ADG data

1,371

1,371

-453


442

Calder Centre Saskatoon only

147

147

295

-

1,596

639

TOTAL

2,235

-

108

 The data collection systems currently being used in Saskatchewan are comprehensive and provided adequate detail to conduct the gap analysis. As we were able to modify the Saskatchewan service utilization data, we eliminated data for First Nations living on reserve and all brief detox clients. o “Brief detox” data were eliminated, as this falls into the category of stabilization/flex beds, suggesting the need to better reflect this service in the planning model (as we found in Fraser Health, BC) o Data for First Nations living on reserve were also eliminated, as First Nations reserve populations were not included in the 2002 CCHS 1.2 survey; thus this population is not represented in the NBP model’s estimates. As the proportion of First Nations people in Saskatchewan is fairly high, we determined that inclusion of this population in the service utilization data may skew gap analysis results.  Current service utilization was not too far from model estimates; these results suggest that working with the new three ranges of help seeking estimates would provide improved gap analysis results (i.e., help set new targets).  Saskatchewan is currently working toward comprehensive screening for Mental Health/Substance Use early identification and intervention in primary care – results from this screening project could potentially help inform the SBIRT portion of the model, by providing details on how actual Canadian screening and brief intervention can increase the proportion of clients that are captured ‘early’ by health care providers.

4. Province of Prince Edward Island is home to approximately 142,000 people, and offers both an urban, but mainly rural, orientation on substance use services and planning at the provincial level. Community-based substance use services are provided on an outpatient basis, and inpatient and outpatient treatment are both offered at the Provincial Addictions Treatment Facility. A continuum of services is provided in PEI, including assessment and referrals, family programs, inpatient and outpatient detoxification and early intervention programs. 109

PEI Gap Analysis Withdrawal Management Service

Required Capacity


# cases per year

Gap

Home-based/Mobile

176

-

546

-370


183

-

593

-410


61

-

135

-74

420

-

1,274

-854

TOTAL


Required Capacity


# cases per year

Gap

Community Minimal

385

-

0

385

Community Moderate

515

-

1,112

-597

Community Intensive

195

-

165

30

1,095

-

1277

-182

TOTAL


Required Capacity


# cases per year

Gap

Supported Recovery

114

-

108

6


119

-

0

119


57

-

122

-65

290

-

230

60

TOTAL

 Gap analysis results initially showed an oversupply in almost all services categories, and particularly in withdrawal management services. Pilot site staff indicated that these results were unexpected, as feedback from clients and family members, and also relayed through members of the Legislative Assembly suggested that more services were needed. After discussion with key stakeholders we determined that the high service utilization observed in PEI, and particularly in withdrawal management, was due to the fact that all clients were required to go through withdrawal management prior to accessing services, including clients waiting to access Methadone Maintenance Treatment (MMT). As a result, clients who may not have required

110

withdrawal management services were cycling through the programs, inflating the numbers beyond the estimates produced by the NBP model.  To exacerbate the estimates even further, clients who were waiting on the list to access MMT were also required to go through withdrawal management; and often, the waiting list for MMT was so long, that clients would cycle through withdrawal management services several times.  Working with this pilot site confirmed that we should include substitution treatment (e.g., MMT) as a separate service category in the model. MMT or other substitution treatment do not necessarily fall into any of the nine service categories identified in the model, but clients accessing these services may make up a significant part of service utilization data. We also understood that the 2002 CCHS 1.2 survey data underestimated the prevalence of opiate misuse and is not well represented in the NBP model. We made further progress on this issue when piloting in Nova Scotia (discussed below).  When conducting the gap analysis exercise in PEI, project staff noted that there appeared to be a large number of substance use services and supports across the province. However, we were advised that there was a pressure to provide even more services, and that much of this pressure came from local communities and often through political representatives. Stakeholders were encouraged by results indicating that PEI had adequate service capacity to meet the needs of the population (with the exception of residential services), and indicated that the NBP model could be used as a form of evidence to counter community and political pressures for additional services.  One major gap identified in PEI was the lack of Residential Services available in the province. Health PEI opened a ten-bed residential treatment facility in April 2014.

111

5. Province of Nova Scotia is home to approximately 929,000 people, and offers a mixed urban and rural orientation on substance use services and planning. The province of Nova Scotia is divided into 10 District Health Authorities, and Addiction Services are operated by the District Health Authorities. Addiction Services provides a continuum of services, including assessment and referrals, family programs, community programs, and hospital-based programs for people involved with alcohol, other drugs, or gambling. Nova Scotia Gap Analysis Withdrawal Management Service

Required Capacity

Home-based/Mobile

1,173


1,218


408

Total


Current Capacity in Specialized Services # cases per year Program name Withdrawal Management Opioid 0 Stabilization ORT (Opioid Replacement Therapy) 112


Gap

112

1,061

548

548

670

4,842

4,842

-4,434

-

5,502

-2,703

Withdrawal Management Day Withdrawal Management Inpatient

2,799

Required Capacity

-

Current Capacity in Specialized Services Program name DWI AEP AIIP Relapse prevention RMV Driver Competency RMV Medical Referral CBS Opioid Stabilization CBS Adolescent CBS CHOICES CBS General CBS Women

# cases per year 251 474 375 170 21 25 71 619 187 3145 251


Gap

1,316

1,261

4,273

-842

Community Minimal

2,577

Community Moderate

3,431

Community Intensive

1,297

-

-

0

1,297

7,305

-

-

5,589

1,716

TOTAL

112


Current Capacity in Specialized Services

Required Capacity

Program name

# cases per year

757


793

Complexity-enhanced/ (medical/psychiatric) TOTAL

STP 21 Day

56

STP CBDHA

127

STP CDHA

86

STP CHOICES 24/7

55

STP CHOICES Day Program

9

STP Women's Life Enhanc

35

STP WRAP I

13


Gap

0

757

381

412

384

-

-

0

384

1,934

-

-

381

1,553

 One of the largest gaps across the province was in Opiate Replacement Treatment (ORT); there are long wait lists for this service, and clients often cycle through withdrawal management while waiting for access to ORT. After our experience in Prince Edward Island, we attempted to determine prevalence of prescription opiate misuse across Canada, so that we could better assess the needs for ORT services. However, when we presented the prevalence figures to the pilot site group in Nova Scotia, it was determined that the number of clients currently receiving ORT services in the province (not including those on the wait list) was three times the prevalence estimates from literature and surveys. This clearly demonstrated that population survey data are significantly underestimating the prevalence of opiate misuse, and that we should plan to work with a wider range of survey, literature, and key informant opinion to develop a unique category for ORT/MMT substitution services.  One issue that was raised in Nova Scotia was related to withdrawal management services. As per regulations, withdrawal management services are offered as an intensive service, and are provided in a hospital setting, with a physician present. This accounted for the unexpectedly large number of clients accessing hospital/complexity-enhanced withdrawal management services.  Staff from the Nova Scotia pilot site also encouraged the project team to expand on the inclusion of health promotion and prevention (HPP) activities in the NBP model, as HPP is a significant

113

component of Addictions Services across the province. Resource planning for ‘treatment’ could not be separated from resource planning for ‘prevention’ since often the same staff are involved.  Pilot site stakeholders noted major gaps in opiate replacement treatment and complexityenhanced residential services.

6. Sarnia-Lambton, Ontario is home to approximately 106,300 people, and is in the unique position of being located near a United States-Canada border. Sarnia-Lambton is located within the Erie St Clair LHIN; the LHIN approved funding for the development of a withdrawal management facility for the Sarnia-Lambton region in summer of 2013. Bluewater Health, a large healthcare facility located in Sarnia, was responsible for planning the region’s new withdrawal management facility. They approached the NBP team to request support for their planning process. As the scope of Bluewater Health’s project was focused only on the development of a withdrawal management facility, we only conducted the gap analysis for withdrawal management services. As there was previously no community/residential withdrawal management facility in the region, instead of populating the gap analysis with regional data, the Bluewater Health team used the model’s estimates to identify the gap and plan from there. Press releases regarding this project have been added in Appendix E. Sarnia-Lambton/Chatham-Kent Withdrawal Management Gap Analysis Withdrawal Management Service

Required Capacity


# cases per year

Gap

Home-based/Mobile

268

-

268


278

-

278


92

-

92

638

-

TOTAL

0

638

114

 We applied the NBP model to Sarnia-Lambton population data, taking regional factors (e.g., border city) into consideration. Working with the Bluewater Health team, we confronted the ‘economies of scale’ challenge (i.e., model projects X but may as well plan for Y). The estimates projected by the model predicted a need for 12 withdrawal management beds; however, taking local indicators of need into account, the Bluewater Health team determined to increase the estimates by four beds, for a total of 16 beds in the withdrawal management facility. The NeedsBased Planning model served as a starting point for the planning process, and the estimates produced by the model shaped the discussions and final decisions made by Bluewater Health.  The NBP project team was also invited to a planning meeting with stakeholders from a variety of sectors, including emergency services, addictions and mental health, social services, and law enforcement in the Erie St. Clair region. Participating in this planning meeting demonstrated the importance and value of using the NBP model as a component of a planning process; the results of this process were very encouraging to the project team.

7. North East LHIN, Ontario is Ontario’s second largest Local Health Integration Network, and is home to 550,000 people. Due to the large geographic area of the LHIN, and issues relating to planning for service provision in remote areas (e.g., James Bay), this pilot site focused on the Nipissing, Timiskiming, Parry Sound, and Muskoka regions within the LHIN. This sub-region comprises 23% of the LHIN’s total population.

115

Nipissing-Timiskiming-Parry Sound-Muskoka Gap Analysis

Withdrawal Management Service

Required Capacity


# cases per year

Gap

Home-based/Mobile

331

None available

0

331


295

NDSAP (30 beds)

358

-63


91

N/A

717

-

TOTAL


91 358

Current Capacity in Specialized Services Required Capacity

Program name Probation/mandated

Community Minimal

817

Community Moderate

992

Community Intensive

312 TOTAL


2,121

Gap

0 14 216

High school

50

NB Community Counselling

228

NB Recovery Home - ASH

27

NB Recovery Home - parole NB Recovery Home Continuing care, evening ACT West Nipissing Hospital Alliance Program Addiction Outreach Muskoka Parry Sound CCAC School Nurse

20

537

120 35 225

-139

364 12

Parry Sound First Nations CD Opiate support program and parenting Parry Sound First Nations opioid/early intervention psychosis

22

NDSAP Day Support

65

247

1476

645

-

60 18


Supported Recovery

218

N/A


193

TOTAL

# cases per year

Forensic outreach NB Community Counselling

Required Capacity


359

# cases per year

Gap 218

North Bay Recovery Home

135

NDSAP

60

-2

84

N/A

495

-

84 195

300

116

 Results from the North East LHIN gap analysis indicated that the capacity of services offered in the region was about equal to the capacity requirement projected by the NBP model. However, stakeholders in the region were able to provide evidence (e.g., wait lists and cases referred out of the region) demonstrating unmet needs. As a result, the project team made a change to the help-seeking parameters; introducing a high, medium, and low help-seeking range, following review and consultation of further research. This flexibility allows for planners to modify the proportion of clients seeking help (the 'demand' population), based on local indicators of need. At the suggestion of the planning committee, we used the medium help-seeking parameters for the gap analysis.  Despite the North Bay and surrounding areas being fairly well-served, there were still indicators of need such as waiting lists and referrals out of the area for more specialized services. This may be a result of the fairly high proportion of First Nations people living in the region, as well as an Armed Forces base in the city of North Bay – both of these populations were not included in the 2002 CCHS 1.2 survey and were thus underrepresented in our estimates of need.  After conducting the gap analysis with stakeholders from the region, we determined that substance use services formed a ‘service funnel’; they have many low-impact, non-intensive services to help capture clients early (e.g., substance use nurses in schools across the district), but do not have enough capacity in the more intensive services (e.g., residential services) to meet the demand, and as a result, there are wait lists for these services.

8. North Simcoe Muskoka LHIN, Ontario is home to 453,700 people living in mainly rural areas. The NSM LHIN is geographically unique in Ontario, as it is the only region between southern (e.g., Toronto) and northern Ontario (e.g., North Bay) that provides a fairly comprehensive continuum of substance use treatment services. As a result, the NSM LHIN receives a lot of referrals, particularly from the York region of Ontario. At the suggestion of the local planning group, we used the low help-seeking parameters for the gap analysis tables below.

117

North Simcoe Muskoka LHIN Gap Analysis Withdrawal Management Service

Current Capacity in Specialized Services

Required Capacity

Program name

# cases per year

Gap

Home-based/Mobile

831

CMHA Community Withdrawal Management

100

731


745

RVH Withdrawal Management

700

45


231

-

0

231

1,807

-

800

1,007

TOTAL

Community Services and Supports Community Minimal

Current Capacity in Specialized Services Required Capacity 1,944

Community Moderate

Program name

# cases per year

Counselling walk in clinics (new in Midland)

400

Adult services (CMHA)

550

AOMPS

474

Adult services (CMHA)

574

South Georgian Bay CHC

38

Gap

994

2,486

1,400

Chigamik CHC

CMHA (ACT and EPI teams) Community Intensive

250

789

539 OSMH Addiction counselling

TOTAL


-

5,219

Current Capacity in Specialized Services Required Capacity

Program name

545


488


213 TOTAL

2286

1,246

# cases per year 0

2,933

Gap 545

RVH Addiction Services (6-8 beds, 21 days)

105

7 South Street

13

Referrals from NSM LHIN agencies outside the LHIN

59

Georgianwood CD program

65

148

242

1,004

-

311

118

 The data collection system in Ontario is fairly comprehensive, and there were no issues collecting data for the gap analysis. Stakeholders at the NSM LHIN pilot site were also able to collect data from other agencies, such as hospitals, leading to a more comprehensive gap analysis.  Pilot site staff indicated that the gap analysis exercise was very helpful, and that results could be used to support a case for funding additional treatment services.  Gaps were noted for services across the continuum of care; these results were triangulated with the opinion of site participants. Major gaps in services for different genders (particularly womenonly services) were also identified.  In discussion with pilot site stakeholders, we used the low range of help-seeking estimates for the gap analysis process. Using the low help-seeking estimates still resulted in identified gaps across the treatment system, and increasing the help-seeking estimates to the medium or high range would have created gaps that stakeholders felt would have been unrealistic to attempt to meet. Low help-seeking estimates still resulted in gaps in services, but offered realistic targets.

9. Chaudière-Appalaches, Québec is home to 418,704 people, and located very close to Quebec City. A fairly comprehensive continuum of services are available; however, the only significant residential service is approximately one hour outside of the major population centre of the region, which can lead to accessibility (e.g., transportation) challenges.

119

Chaudière-Appalaches Gap Analysis Withdrawal Management Service

Required Capacity

Current Capacity in Specialized

Low HelpSeeking

Med HelpSeeking

High HelpSeeking

Program name

Home-based/Mobile

699

1,093

1,753

Impossible


623

975

1,553

Désintox III.7D


193

302

481

Media-med?

1,515

2,370

3,787

TOTAL


Required Capacity Med HelpSeeking

High HelpSeeking

Program name

Community Minimal

1,662

2,677

4,554

I-CLSC

Community Moderate

2,092

3,270

5,254

Externe

Community Intensive

660

1,032

1,642

Externe intensif

4,414

6,979

11,450


Med HelpSeeking

High HelpSeeking

699

1,093

1,753

375

248

600

1,178

244

-51

58

237

619

896

1,751

3,168

Low HelpSeeking

Med HelpSeeking

High HelpSeeking

1,662

2,677

4,554

550

1,728

3,712

660

1,032

1,642

2,872

5,437

9,908

Required Capacity

Low HelpSeeking

Med HelpSeeking

High HelpSeeking

# cases per year

1,542

1542

Gap


Gap

Low HelpSeeking

Med HelpSeeking

High HelpSeeking

Program name

# cases per year

459

718

1,156

Lits multifonctionnels

183

276

535

973

575

-167

64

444

178

279

445

287

878

1,862


408

639

1,019

RIH (125) + Croisée (123) + Aube de la paix (70) + (108) + Domremy


178

279

445

RIH ?

1,045

1,636

2,620

TOTAL

Gap Low HelpSeeking


Low HelpSeeking

TOTAL

# cases per year

758

 This was the second pilot site that considered the new range of help-seeking parameters in the application of the model. When working to apply the ranges of high, medium, or low, stakeholders noted that it would be useful to apply the different ranges of help-seeking to different service categories. This suggestion was brought forward as a result of service availability in the region, as well as setting realistic targets based on health promotion, or increased screening and brief intervention efforts in place. For example, stakeholders believed that with increased screening and brief intervention efforts in the region, the need for services in

120

Chaudière-Appalaches could realistically come close to the ‘high’ help-seeking parameter target. However, for more intensive services, such as residential, targets were set at a ‘moderate’ level.  Quebec hospitals are currently doing a lot of proactive screening and early identification by colocating Liaison Nurses in Addictions in hospitals. Results from this program are positive, and there are now 24 hospitals with Liaison Nurses across the province, demonstrating a real push for more screening and early identification.  There are quite a few independent substance use agencies across the region; this made it challenging to track service utilization, and may have impacted the results of the gap analysis.

Summary

Overall, there was considerable synergy between the gaps identified quantitatively and those identified locally through key informant opinion or after systematic analysis. We also used feedback from each pilot site to inform model development; experiences in some sites led us to make changes to the NBP model (e.g., introducing a range of help-seeking parameters from our experience in the North East LHIN), and further piloting of the model helped to validate any modifications made to the model (e.g., pilot testing the range of help-seeking parameters in Ontario’s North Simcoe Muskoka LHIN and Quebec’s Chaudière-Applaches health region).

Although several modifications and enhancements should be made to the Needs-Based Planning model prior to system-wide implementation, some jurisdictions have started to use the model to inform planning processes, and have been using the framework of the model to help inform the broader treatment system. Through our evaluation and piloting activities, system planners have indicated that using the model would significantly increase decision-making capacity, and that the model applies research evidence and best-practices thereby making it relevant to meet their needs.

121

5.2 Lessons Learned from Piloting: Challenges in the application of the Needs-Based Planning Model This section of the report describes the several challenges associated with using the NBP model as a tool to help guide system planning, identified by the project team through pilot testing activities. We articulate the main limitations associated with the application of the model, and the implications of these challenges in the planning context. Many of these limitations are sitespecific; however all limitations are worth noting. The identification of these limitations also served to reinforce the idea that the NBP model is but one tool that can help guide a system planning process, and that additional insights and supports (i.e., understanding of community context, social indicators of need; use of other regional data) are required to conduct a comprehensive system planning and resource allocation exercise.

Challenge 1: Data collection systems: Data collection challenges were observed in several pilot sites and these challenges varied across pilot sites.

Several provinces use more than one system to collect data on utilization of substance use services and this proved to be challenging as multiple data collection systems were rarely linkable and able to share client information electronically or standardize definitions to a sufficient degree to allow data roll-up. In addition, data on substance use services were not well-linked with data for mental health, primary care, or acute care, meaning that clients accessing non-specialized services (i.e., emergency room) were not being captured and made available for planning a comprehensive substance use system. Further in many regions substance use treatment services are contracted out to other agencies. While these contracted agencies must follow the guidelines outlined by the Health Authority or provincial mandates, collecting data has proven to be challenging in some locations. At the time of piloting some contracted agencies were not collecting/reporting data to the Health Authority/province.

Lastly, across Canada the integration of mental health and addiction services has contributed significantly to the development of concurrent disorder-capable programs that benefit people experiencing this range of challenges. However, in one pilot site the integration of mental 122

health and addictions services led to an unexpected limitation in the application of the needsbased planning model. This site had so thoroughly integrated mental health and addictions services, including their client information systems, that it was impossible to distinguish clients with substance use challenges accessing either substance use services, mental health services or both.

Implications: To effectively conduct an accurate gap analysis, planners require comprehensive data available for all substance use treatment services and supports in the health planning region, otherwise there is a risk of misidentifying gaps in service capacity. While the challenge specific to mental health and substance service integration was specific to one pilot site, our experience raises an important red flag to be raised to be highlighted during comprehensive service integration processes that sufficient data should remain distinct but well linked for planning based on the NBP model.

Challenge 2: Clients accessing services from outside the local planning region: This issue is particularly relevant for planning at the level of the Health Authority level as not all categories of treatment services are necessarily available across all regions (particularly in more rural regions), and, clients may need to travel outside of their local health planning region to access the appropriate treatment services. In addition, having services such as correctional facilities in a health planning region may also cause an influx of clients into local services who are not native to the region.

Implications: The issue of clients from outside the health planning region accessing local services is a challenge that is already being confronted by planners must often consider the availability of the service across the region or province. However, when using the NBP, this issue may serve as an important rationale for using the medium-high range of the help-seeking parameters to compensate for clients not included in the population of the local region.

Challenge 3: Independent or for-profit treatment services: One significant challenge applicable to several of the pilot sites was the presence of independent or for-profit treatment services, 123

particularly for residential treatment. These services are not always bound by the same minimum standards as publicly-funded services and there is currently have no mandatory mechanism in place to accurately assess service utilization at these independent facilities. However, clients are accessing these services, and data for these services are not included in the gap analysis process. This may lead planners to overestimate the treatment gap if a portion of the ‘in need’ population identified in the model is receiving treatment at private for-profit organizations. A related challenge that came to light was the issue of unknown and perhaps entirely unregulated quality of care in independent for-profit organizations and this also impacts whether the services being offered should be “counted” as part of local capacity.

Implications: The Needs-Based Planning model only includes public-funded services, or services that are contracted by the public sector, and does not account for any services provided by for-profit or independent agencies. If many independent treatment services are offered in a specified health planning region, it will be continuing challenge to assess the number of ‘in need’ clients or to determine an accurate treatment service gap.

Challenge 4: Variability in treatment services: Treatment services (particularly outpatient programs) differ across the country in terms of intake processes, length of the program, and the functions delivered within the program. We found it challenging sometimes for a region to map its services onto each of the nine categories outlined in the NBP model, and challenging to ensure that this is done consistently across regions using the model. This challenged our ability to compare service gaps in one region to those of another.

Implications: As there is currently considerable variation in treatment services across Canada, project staff worked closely with pilot site staff to complete the service mapping exercise and resulting gap analysis. It is crucial that supports are available in this stage of model application, as incorrect mapping may skew the results of the gap analysis exercise, with the potential to over or underestimate service gaps.

124

Challenge 5: Political and funding pressures: Across Canada, there are political pressures to provide services based on identified investment areas or based on local community pressures. As a result, planners may not have the flexibility to modify services or reallocate resources to meet the needs identified by the NBP model.

Implications: It can be challenging to modify existing substance use treatment systems to meet evolving population needs and research evidence, when the province, health regions, or local political pressures are drawn to invest in different priority areas as highlighted by the results of applying the NBP model. We anticipate that the NBP model will serve as an evidence-based support to encourage planners and senior decision-makers to work to ensure that local needs are being addressed in the context of these complex contextual factors.

5.3 Key Lessons Learned 1. There is no one simple 'formula' to assist with treatment system planning/strengthening treatment systems, but rather a collection of tools that can be used together to inform treatment system planning. Feedback from stakeholders at pilot sites, and through the project evaluation, has indicated that people are ready for and interested in using the model, but we need to be careful that people see the model as but one piece of a larger planning process, and not solely as a stand-alone 'formula' to determine gaps. System planners should also use other indicators of need (e.g., wait lists) and regional indicators (e.g., accessibility to services based on distance and travel time). Ideally, the NBP model should be applied with the support of people trained in its application to ensure that it is being interpreted correctly and with the larger context in mind.

2. There is also no one standard formula that will work for the whole country; as health planning regions across Canada are so diverse, geographically, in terms of the types of services provided, and in terms of health care delivery structures (e.g., a centralized health care system or a number of health authorities), it can be challenging to implement a model with a set of firm parameters. To address this challenge, we have incorporated more flexibility into the model (e.g., ranges of

125

help-seeking proportions), to ensure that the model can be adapted to different health planning regions. This was widely supported by the pilot sites and the project Advisory Committee.

3. Finding the appropriate evidence to inform each parameter of the Needs-Based Planning model (i.e., projecting client care pathways, problem identification using different screening and assessment methods) can be challenging, as we found that research in this field is sometimes absent, or does not address solely substance use. For example many relevant surveys included questions on seeking help for either mental health or substance use problems, making it difficult to determine what the proportion for help-seeking for only substance use problems was). As a result, we worked very closely with the national project Advisory Committee and experts from the International Panel; members of these groups provided feedback opinions and linked us with research and expert opinion to assist in confirming/validating the final decisions surrounding the parameters of the model. Their guidance has been critical to achieving consensus on key aspects of Needs-Based Planning model challenged by limited published research evidence.

4. Our experiences to date have confirmed the importance of engaging with a diverse range of stakeholders, including senior planners, managers, and front line staff. When developing a planning model that has the potential to impact the entire substance use treatment system, it is crucial to include stakeholders from many different areas of the service delivery system to ensure that a wide range of feedback is reflected in the model. It was also important to include a diverse range of stakeholders in the piloting process, including different parts of Canada, as the different perspectives assisted with developing a more comprehensive planning tool. 6.0 Limitations, Special Considerations and Next Steps

This project was planned from the outset as one that would be implemented in stages, and certainly the planning model will undoubtedly require further iterations as our understanding of substance use treatment systems evolves, and as new research improves our model building and testing capability. We conclude this final report for work completed in project years 2010-2014

126

with a brief summary of the most salient limitations to date, special considerations and some of the next steps for the proposed work in 2014-2016. Modeling and Estimating: It is important to re-iterate our challenges modeling and estimating highly complex individual treatment trajectories. Our compromise position was to incorporate flow across the major categories of withdrawal management, and community and residential services and supports within a one year time frame, but not at a finer level within these broad categories. While this may be an area for future enhancement, it is a highly complex task challenged by lack of data on service use trajectories at a finer level of detail.

Internet and Mobile-Based Services: As important as Internet and mobile-based services and supports are becoming it is not possible to formally estimate the additional “capacity” they may be able to effectively treat and support in the context of the overall treatment system. However, while quantitative capacity estimates may not be possible at present, we recommend formal consideration of these alternatives in applications of the needs-based planning model as they hold considerable untapped potential for increasing reach, effectiveness, and efficiency. We will be exploring this in our pilot sites.

Other Populations: Substance use treatment systems must address the needs of several other populations such as youth, women and diverse populations defined by gender and/or sexual identity, and culture and ethnicity, for example. It is beyond the scope of the project at the present time to develop and test population-specific models. That said, developmental work for a youth-specific model is underway with Quebec-based funding for that province specifically and which will be adapted for the national level in the next phase of our work pending funding.

First Nations, Métis and Inuit Populations: With the support of additional funding and external project consultants, we undertook a process in 2011-12 to see how the key concepts, terminology, schematic representations and estimation procedures need to be adjusted for Canada’s First Nation, Métis and Inuit populations. The intention was not to develop a new version or versions of the model for these populations at the present time, but rather to begin to explore the need for,

127

and scope of, potential cultural adaptations (Dolan, 2012; Johnston, 2012). Many implications for adapting a needs-based planning model such as this for First Nations, Métis and Inuit populations were identified, including but not limited to, requirements for a broader community perspective on need and lack of available data to define categories of need/severity, needs for different types of services and current service utilization.

Services for Family Members and Significant Others: Many substance use treatment systems are organized to provide services for family members of those in substance use treatment, either as clients in their own right or as supporters of the client with substance use problems. Using data from the DATIS database in Ontario and perhaps other jurisdictions, future work will explore a process to develop estimates of the additional required capacity for family members/significant others given an estimated demand population for substance use services.

Gambling Services: Services for people with gambling problems are offered in many Canadian substance use services, but are not provided consistently across the country. To estimate the needs of this population is not within the scope of this project. However, the issue has been recognized for future consideration.

Non-Survey Populations: The CCHS 1.2 did not include certain populations in Canada, including people living on First Nations reserves; those incarcerated; people that were institutionalized during the survey timeframe (e.g., hospitals, long-term care); people living in the Canadian Territories; and people who are homeless (e.g., couch surfing, living in shelters, living outdoors). As a compliment to this project, we initiated a process to approximate the number of people from these non-surveyed populations in each of the Canadian health services planning areas. Another step will be to estimate the proportion of these populations that fall within each of the need categories in the five-tiered model. The initial work on homeless populations is shown in Appendix F. The First Nations populations living on reserves across Canada are shown in Appendix G. With these approximate population numbers and corresponding distributions across the need categories, local jurisdictions

128

may be able to apply these populations to the model as deemed necessary. Table 12 shows how these estimates may later be incorporated into regional populations for planning purposes. Table 12: In-Need Population Champlain LHIN, Ontario Region: Champlain LHIN Total Population: 1 044 481 In Need Populations

Tier 1

Tier 2

Tier 3

Tier 4

Tier 5

841 808

109 087

64 376

27 426

1784

Homeless (n= )

-

-

-

-

-

First Nations (n= )

-

-

-

-

-

Justice/Institutionalized (n= )

-

-

-

-

-

Long-term Psychiatric (n= )

-

-

-

-

-

841 808

109 087

64 376

27 426

1784

Survey Estimates (N=1 044 481)

Total In-Need Population

Estimating prevalence of opiate dependence and need for opiate substitution treatment: After conducting the gap analysis exercise at several pilot sites, we noted a trend emerging, whereby there was an ‘oversupply’ (current service utilization greater than the capacity estimated by the model) in Withdrawal Management and Community Moderate services. We conducted a detailed review of the gap analysis with pilot site staff, and determined that the inclusion of clients undergoing opiate substitution treatment was a likely cause for the high levels of service utilization being counted in these two treatment categories.

We further investigated this issue, reviewing data from the literature on the prevalence of prescription opiate misuse in Canada. We presented these data to one of the pilot sites, and discovered that the number of clients currently engaged in methadone maintenance treatment was approximately three times the prevalence estimates for that area as derived from the literature. Reviewing the current number of MMT clients in other pilot sites has confirmed that the numbers of people currently receiving methadone maintenance treatment far exceeds the estimates for opiate misuse prevalence gathered from the literature. These results have indicated that current Canadian surveys are considerably underestimating the prevalence of prescription opiate misuse and/or dependence, and that the estimates being produced by the NBP model are 129

not adequately capturing this in-need population, and are therefore too conservative, particularly for treatment service categories that health planning regions may include MMT clients in, such as Community Moderate. To address this issue we have determined that we need an additional category for substitution treatment services (methadone, suboxone) in the next phase of NBP model development. This has been proposed in the application for an additional two years of funding (2014-16).

Defining Local Treatment Systems on a Geographic Basis: For the purpose of this project the most relevant planning regions for model application were identified as the health regions/authorities/districts within each of the provinces/territories (P/Ts) across Canada. We have made the assumption that substance use service planning is done on a geographic basis, which puts boundaries on the populations being served. This assumption was validated by our baseline survey which noted that in the majority of the P/Ts that are divided into health regions, substance use planning is performed at the regional level; whereas the P/Ts that have centralized health care systems (e.g., PEI, Alberta) plan their substance use services at the P/T level (Health Systems and Health Equity Research Group, 2011). When applying the model to each individual health planning region, other needs assessment information that is available at this jurisdictional level should be brought to bear in the planning process. The 87 health planning regions used for the purposes of this project are shown in Appendix H. These planning regions may change in the future, requiring updating of the data being developed. Further Model Validation: Validation-related questions concern both the individual parameters being estimated within the treatment system flow diagram, as well as the conceptualization and results of the modeling process as a whole. Clearly we are building a largely “Made-in-Canada” planning model using mostly Canadian data and key informants. Thus, our primary interest in validation is in the Canadian context. That said, a role of our international expert panel has been to comment on the general approach we have followed (e.g., use of the tiered framework; the measurement model to operationalize need for services and need categories; definitions of system-level functions and service categories; development of the flow diagram and parameter 130

estimation procedures). The details and precise definitions, as well as availability of data, will no doubt vary across jurisdictions, but we have aimed to develop a needs-based planning process transferable to other jurisdictions, with the provision of the appropriate toolkits and supports. We have also offered limited but important consultation to similar work in Brazil and Australia and learned from work in these areas in return.

As noted earlier, we have pilot tested a draft model in nine diverse Canadian jurisdictions, and the piloting has informed us regarding the perceived usefulness of the model in local planning processes, and whether the capacity estimates being generated have sufficient face validity for local planners to incorporate them into discussions about funding and other resource allocation. The pilot testing was also seen as another step in the refinement of the model, and its parameters. This is consistent with a developmental evaluation process. Subsequent, multiple case studies can be implemented to continue that process of validation against other local data and key informant opinion and thereby continue the process of model refinement. Multiple applications may, for example, identify additional upper and lower limits of selected parameters that appear to be highly context-dependent. Ongoing substance use services and epidemiological research will also continue to inform the model parameters, for example, when we apply the new data from the 2012 CCHS 1.2 Statistics Canada survey (planned for the fall of 2014). In the United Kingdom, David Best and colleagues used treatment outcome data to calibrate their model parameters and similar work is being planned in Australia. Essentially this approach argues that the ultimate test of the validity of the model is whether people get to the services that optimize their outcomes. This may have been possible in the Birmingham situation, given a more restricted focus of the model on substance use services and the criminal justice system, and therefore a more circumscribed system for determining outcomes. The Australian outcome monitoring system is in the seminal stages of development. At present, there is no treatment outcome monitoring system in any Canadian jurisdiction capable of supporting this calibration function for our planning model. Outcome monitoring systems are at various stages of development in a small number of Canadian jurisdictions (most significantly in Ontario and Newfoundland through pilot work funding by provincial DTFP funds), and the

131

development/promotion of common outcome monitoring models and measures is on the future agenda of the National Treatment Indicators Project. Thus, testing and calibration of the needsbased planning model based on treatment outcomes may be possible in the future in some Canadian jurisdictions.

132

7.0 References Adair, C.E. (2009). Concurrent Substance Use and Mental Disorders in Adolescents: A Review of the Literature on Current Science and Practice. Submitted to The Alberta Centre for Child Family and Community Research. Accessed at: http://www.research4children.com/public/data/documents/ConcurrentSubstanceUseandMentalD isordersinAdolescentsAReviewoftheLiteratureonCurrentScienceandPracticepdf.pdf Afifi, T.O. (2005). Perceived need and help seeking for mental health problems among Canadian provinces and territories. Canadian Journal of Community Mental Health, 24(1), 51-61. Aldworth, J., Colpe, L., Gfroerer, J., Novak, S., Chromy, J., Barker, P., Barnett-Walker, K., Karg, R., Morton, K. & Spagnola, K. (2010). The National Survey on Drug Use and Health Mental Health Surveillance Study: Calibration analysis, 61–87. Published Online: May 31 2010. Alexander, M.J., Haugland, G., Lin, S.P., Bertollo, D.N. & McCorry, F.A. (2008). Mental health screening in addiction, corrections and social services settings: Validating the MMS. International Journal of Mental Health and Addiction, 6, 105–109. American Association of Community Psychiatrists. (2009). LOCUS: Level Care Utilization System for Psychiatric and Addiction Services. American Association of Community Psychiatrists. Anderson, P., Laurant, M., Kaner, E., Wensing, M. & Grol, R. (2004). Engaging general practitioners in the management of hazardous and harmful alcohol consumption: Results of a meta-analysis. Journal of Studies on Alcohol, 65(2), 191-199. Andrew, G., Cuijpers, P., Craske, M.G., McEvoy, P. & Titov, N. (2010). Computer therapy for the anxiety and depressive disorders is effective, acceptable and practical health care: A meta-analysis. PLoS ONE, 5(10), e13196. doi: 10.1371/journal.pone.0013196 Andrews, G., Henderson, G. & Hall, W. (2001). Prevalence, comorbidity, disability and service utilization: Overview of the Australian national mental health survey. British Journal of Psychiatry, 178, 145-153. Anglin, M. D., Hser, Y. I. & Grella, C. E. (1997). Drug addiction and treatment careers among clients in the Drug Abuse Treatment Outcome Study (DATOS). Psychology of Addictive Behaviors, 11(4), 308–323. Angst, J., Sellaro, R. & Merikangas, K. R. (2002). Multimorbidity of psychiatric disorders as an indicator of clinical severity. European Archives of Psychiatry and Clinical Neuroscience, 252, 147– 154. Aoun, S., Pennebaker, D. & Wood, C. (2004). Assessing population need for mental health care: A review of approaches and predictors. Mental Health Services Research, 6(1), 33-46.

133

Babor, T.F., McRee, B.G., Patricia A. Kassebaum, P.A., Grimaldi, P.L., Ahmed, K. & Bray, J. (2007). Screening, Brief Intervention, and Referral to Treatment (SBIRT). Substance Abuse, 28(3), 7-30. Babor, T.F., Stenius, K. & Romelsjo, A. (2008). Alcohol and drug treatment systems in public health perspective: mediators and moderators of population effects. International Journal of Methods in Psychiatric Research, 17(S1), S50–S59. Baldacchino, A. & Corkery, J. (2006). Comorbidity: Perspectives across Europe. London: European Collaborating Centres in Addiction Studies. Barry, K.L. (1993). Substance Abuse and Mental Health Services Administration/Centre for Substance Abuse Treatment Improvement Protocols: Treatment Improvement Protocol Series 34: Brief Interventions and Brief Therapies for Substance Abuse. Substance Abuse and Mental Health Services Administration, Maryland, USA. Bernstein, E., Bernstein, J. & Levenson, S. (1997). Project ASSERT: An ED-based intervention to increase access to primary care, preventive services, and the substance abuse treatment system. Annuals of Emergency Medicine, 30(2), 181-189. Bertholet, N., Daeppen, J-B., Wietlisbach, V., Fleming, M. & Burnand, B. (2005). Reduction of alcohol consumption by friend alcohol intervention in primary care. Systematic review and metaanalysis. Archives of Internal Medicine, 165(9), 986-995. Bewick, B.M., Trusler, K., Barkham, M., Hill, A.J., Cahill, J. & Mulhern, B. (2008). The effectiveness of web-based interventions designed to decrease alcohol consumption- a systematic review. Preventative Medicine, 47, 17-26. Bien, T. H., Miller, W. R. & Tonigan, J. S. (1993). Brief interventions for alcohol problems: A review. Addiction, 88(3), 315–336. Bjerke, T.N., Kummervold, P.E., Christiansen, E.K. & Hjortdahl, P. (2008). “It made me feel connected”- an exploratory study on the use of mobile SMS in follow-up care for substance abusers. Journal of Addictions Nursing, 19: 195-200. Blanchette-Martin, N., Ferland, F., Tremblay, J. & Garceau, P. (2011). Liaison nurses in addiction of the large region of the Capitale-Nationale and Chaudière-Appalaches: Portrait of services and users’ trajectories. Bibliothèque et Archives nationales du Québec and Bibliothèque nationale du Canada. Bodenheimer, T., Wagner, E. H. & Grumbach, K. (2002). Improving primary care for patients with chronic illness: The chronic care model, Part 2. Journal of the American Medical Association, 288(15), 1909–1914. Boulos, M.N.K., Wheeler, S., Tavares, C. & Jones, R. (2011). How smartphones are

134

changing the face of mobile and participatory healthcare: An overview, with example from eCAALYX. BioMedical Engineering OnLine, 10(1), 24. Bradshaw, J. (1994). The contextualisation and measurement of need: a social policy perspective. In J. Popay and W.G. London (Eds.), Researching the People's Health. London, Routledge. Breslin, C., Sobell, M., Sobell, L., Cunningham, J.A., Sdao, J. K. & Borsoi, D. (1998). Problem drinkers: evaluation of a stepped care approach. Journal of Substance Abuse, 10(3), 217–232. Brugha, T.S., Bebbington, P.E., Singleton, N., Melzer, D., Jenkins, R., Lewis, G., Farrell, M., Bhugra, D., Lee, A. & Meltzer, H. (2004). Trends in service use and treatment for mental disorders in adults throughout Great Britain. British Journal of Psychiatry, 185, 378-384. Budman, S.H. (2000). Behavioral health care dot-com and beyond: Computer-mediated communications in mental health and substance abuse treatment. American Psychology 55(11), 1290-1300. Burgess, P.M., Pirkis, J.E., Slade, T.N., Johnston, A.K., Meadows, G.N. & Gunn, J.M. (2009). Service use for mental heath problems: Findings from the 2007 National Survey of mental health and Wellbeing. Australian and New Zealand Journal of Psychiatry, 43, 615-623. California Department of Alcohol and Drug Programs (2008). California Drug and Alcohol Treatment Assessment (CALDATA), 1991-1993 [Computer File]. ICPSR02295-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2008-10-07. doi:10.3886/ICPSR02295.v1 Campbell, A.N.C., Nunes, E.V., Matthews, A.G., Stitzer, M., Miele, G.M., Polsky, D., Turrigiano, E., Walters, S., McClure, E.A., Kyle, T.L., Wahle, A., Van Veldhuisen, P., Goldman, B., Babcock, D., Stabile, P.Q., Winhusen, T. & Ghitza, U.E. (2014). Internet-delivered treatment for substance abuse: a multisite randomized controlled trial. American Journal of Psychiatry, 171, 683-690. Centre for Addiction and Mental Health. (2011). Development of Needs-Based Planning Models for Substance Use Service and Supports Across Canada: Current Practices. Accessed at: http://needsbasedplanningmodels.files.wordpress.com/2011/08/baseline-survey-report_final.pdf http://www.phac-aspc.gc.ca/ph-sp/docs/charter-chartre/pdf/charter.pdf Centre for Addiction and Mental Health. (2003). CAMH and Harm Reduction: A Background Paper on its Meaning and Application for Substance Use Issues. Accessed at: http://www.doctordeluca.com/Library/AbstinenceHR/CAMH&HR03.htm Cherpitel, C.J., Korcha, R.A., Moskalewicz, J., Swiatkiewicz, G., Ye, Y. & Bond, J. (2010). Screening, Brief Intervention and Referral to Treatment (SBIRT): 12-month outcomes of a randomized controlled clinical trial in a Polish emergency department. Alcoholism: Clinical and Experimental Research, 34(11), 1922–1928.

135

Chestnut Health Systems. (2002). Global Appraisal of Individual Needs: Initial. Accessed at: http://www.chestnut.org/li/gain/GAIN_Overview.pdf Chi, F.W. & Weisner, C.M. (2008). Nine-year psychiatric trajectories and substance use outcomes: An application of the group-based modeling approach. Evaluation Review, 31(1), 39-58. Chou, W.S., Hunt, Y.M., Beckjord, E.B., Moser, R.P. & Hesse, B.W. (2009). Social media use in the United States: Implications for health communication. Journal of Medical Internet Research, 11(4), e48. Cobain, K., Owens, L., Kolamunnage-Dona, R., Fitzgerald, R., Gilmore, I. & Pirmohamed, M. (2011). Brief interventions in dependent drinkers: A comparative prospective analysis in two hospitals. Alcohol and Alcoholism, 46(4), 434-440. Codony, M., Alonso, J., Almansa, J., Bernert, S., de Giorlamo, G., de Graaf, R., Hario, M.J., Kovess, V., Vilagut, G. & Kessler, R.C. (2009). Perceived need for mental health care and service use among adults in western Europe: Results of the ESEMeD Project. Psychiatric Services, 60(8), 1051 -1058. Cole-Lewis, H. & Kershaw, T. (2010). Text messaging as a tool for behavior change in disease prevention and management. Epidemiology Review, 32(1), 56-69. Corson, K., Gerrity, M. S. & Dobscha, S. K. (2004). Screening for depression and suicidality in a VA primary care setting: 2 items are better than 1 item. The American Journal of Managed Care, 10(11), 839 – 845. Cross, S. & Sibley-Bowers, L.B. (2002). The Standardized Tools and Criteria Manual: Helping Clients Navigate Addiction Treatment in Ontario. Toronto ON: Centre for Addiction and Mental Health. Cunningham, J.A., Wild, T.C., Cordingley, J., Van Mierlo, T. & Humphreys, K. (2010). Twelve-month follow up results from a randomized control trial of a brief personalized feedback intervention for problem drinkers. Alcohol and Alcoholism, 45(3), 258-262. Cunningham J.A. & Van Mierlo, T. (2009). Methadological issues in the evaluation of Internetbased interventions for problem drinking. Drug and Alcohol Review, 28, 12-17. Cunningham, J.A. & Blomqvist, J. (2006). Examining treatment use among alcohol-dependent individuals from a population perspective. Alcohol & Alcoholism, 41(6), 632-635. Cunningham, J.A. & Breslin, F.C. (2004). Only one in three people with alcohol abuse or dependence ever seek treatment. Addictive behaviors 29(1), 221-223. D’Onofrio, G. & Degutis, L.C. (2010). Integrating project ASSERT: A screening, intervention, and referral to treatment program for unhealthy alcohol and drug use into an urban emergency department. Academic Emergency Medicine, 17(8), 903-911.

136

Dennis, M., Chan, Y-F. & Funk, R. (2006). Development and validation of the GAIN Short Screener (GSS) for internalizing, externalizing and substance use disorders and crime/violence problems among adolescents and adults. The American Journal on Addictions, 15, 80-91. Dennis, M., Scott, C. K. & Funk, R. (2003). An experimental evaluation of Recovery Management Checkups (RMC) for people with chronic substance use disorders. Evaluation and Program Planning, 26, 339-52. Department of Indian Affairs and Northern Development. (2011). Registered Indian Population by Sex and Residence, 2010. Accessed at: http://www.aadncaandc.gc.ca/eng/1309357658533#chp9_6 Diaz-Granados, N., Georgiades, K. & Boyle, M.H. (2010). Regional and individual influences on use of mental health services in Canada. The Canadian Journal of Psychiatry, 55(1), 9-19. Dolan, W. (2012). An Inuit perspective on the needs-based planning model for substance use services and supports. Report prepared for the Centre for Addiction and Mental Health, Toronto, Canada. Drummond, D. C., Oyefeso, N., Phillips, T., Cheeta, S., Deluca P., Perryman, K., Winfield, H., Jenner, J., Cobain, K., Galea, S., Saunders, V., Fuller, T., Pappalardo, D. Baker, O. & Christoupoulos, A. (2005) Alcohol Needs Assessment Research Project (ANARP): The 2004 National Alcohol Needs Assessment for England. London: Department of Health. Drummond, C., Deluca, P., Oyefeso, A., Rome, A., Scrafton, S. & Rice, P. (2009). Scottish alcohol needs assessment. London: Institute of Psychiatry, King's College London. Druss, B.G., Wang, P.S., Sampson, N.A., Olfson, M., Pincus, H.A., Wells, K.B. & Kessler, R.C. (2007). Understanding mental health treatment in persons without medical diagnoses. Archives of General Psychiatry, 64(1), 1196-1203. Dunn, C.W. & Ries, R. (1997). Linking substance abuse services with general medical care: integrated, brief interventions with hospitalized patients. American Journal of Drug and Alcohol Abuse, 23(1), 1-13. Elahai, J.D. & Ford, J.D. (2007). Correlates of mental health service use intensity in the national comorbidity survey and national comorbidity survey replication. Psychiatric Services, 58(8), 11081115. Edwards, G. (2000). Editorial note: natural recovery is the only recovery. Addiction, 95(5), 747. Field, C.A., Baird, J., Saitz, R., Caetano, R. & Monti, P.M. (2010). The mixed evidence for brief intervention in emergency departments, trauma care centers and inpatient hospital settings: What should we do? Alcoholism, Clinical and Experimental Research, 34(12), 2004-2010.

137

Field, C.A. & Caetano, R. (2010). The effectiveness of brief intervention among injured patients with alcohol dependence: Who benefits from brief interventions? Drug and Alcohol Dependence, 111, 13-20. Fjeldsoe, B.S., Marshall, A.L. & Miller, Y.D. (2009). Behavior change interventions delivered by mobile telephone short-message service. American Journal of Preventative Medicine, 36(2): 165173. Ford, W. E. (1985). Alcoholism and drug abuse service forecasting models: comparative discussion. International Journal of the Addictions, 20, 233-252. Fraser, N. (Chair). (2009). Addiction and mental health integrated service delivery framework. Service Delivery Framework Working Group, Alberta Health Services, Addiction and Mental Health. Gastfriend, D.R. (2003). Addiction treatment matching: Research foundations of the American Society of Addiction Medicine (ASAM) criteria. Binghamton, NY, Haworth Medical Press. Glaser, F.B. (1974). The treatment of drug abuse in the rural south: Application of the Core-Shell Treatment System model. Southern Medical Journal, 67(5), 580–586. Gold, J., Aitken, C.K., Dixon, H.G., Lim., M.S.C., Gouillou, M. & Hellard, M.E. (2011). A randomized controlled trial using mobile advertising to promote safer sex and sun safety to young people. Health Education Research, 26(5): 782-794. Grothues, J.M., Bischof, G., Reinhardt, S., Meyer, C., John, U. & Rumpf, H-J. (2008). Differences in help seeking rates after brief intervention for alcohol use disorders in general practice patients with and without comorbid anxiety or depressive disorders. International Journal of Methods in Psychiatric Research, 17(S1), S74-S77. Hasin, D. & Hatzenbuehler, M. L. & Keyes, K. & Ogburn, E. (2006). Substance use disorders: Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) International Classification of Diseases, Tenth edition (ICD-10). Addiction, 101(S1), 59–75. Health Canada. (2001). (B. Rush - Principal Author). Best practice for concurrent mental health and substance use disorders. (Cat. No. H39-599/2001-2E/ISBN: 0-662-31388-7). Ottawa: Health Canada. Health Systems and Health Equity Research Group. (2011). Development of needs-based planning models for substance use services and supports in Canada – current practices. Toronto: Centre for Addiction and Mental Health. Available: http://needsbasedplanningmodels.wordpress.com/about/ Heather, N. (2012). Can screening and brief intervention lead to population-level reductions in alcohol-related harm? Addiction Science & Clinical Practice, 7(15), Available at: http://www.ascpjournal.org/content/7/1/15. 138

Henderson, S., Andrews, G. & Hall, W. (2000). Australia’s mental health: An overview of the general population survey. Australian and New Zealand Journal of Psychiatry, 34, 197-205. Hermann, R., Rollins, C. & Chan, J. (2007). Risk-adjusting outcomes of mental health and substance-related care: A review of the literature. Harvard Review of Psychiatry, 15, 52–69. Hilton, T. F. (2011). The promise of PROMIS(R) for addiction. Drug and Alcohol Dependence, 119, 229–234. Hollander, M. & Prince, M. (2008). Organizing healthcare delivery systems for persons with ongoing care needs and their families: A best practices framework. Healthcare Quarterly, 11(1), 44-54. Huang, B., Dawson, D.A., Stinson, F.S., Hasin, D.S., Ruan, W.J., Saha, T.D., Smith, S.M., Goldstein, R.B. & Grant, B.F. (2006). Prevalence, correlates and comorbidity of non medical prescription drug use and drug use disorders in the United States: Results of the national epidemiological survey on alcohol and related conditions. Journal of Clinical Psychiatry, 67(7), 1062-1073. Humphreys, K., Wing, S., McCarty, D., Chappel, J., Gallant, L., Haberle, B., Horvath, A.T., Kaskutas, L.A., Kirk, T., Kivlahan, D., Laudet, A., McCrady, B.S., McLellan, A.T., Morgenstern, J., Townsend, M. & Weiss, R. (2004). Self-help organizations for alcohol and drug problems: Toward evidence-based practice and policy. Journal of Substance Abuse Treatment, 26, 51–158. Huyse, F. J. & Stiefel, F. C. & de Jonge, P. (2006). Identifiers, or “red flags,” of tiered frameworks for planning substance use service delivery systems complexity and need for integrated care. Medical Clinics of North America, 90(4), 703–712. Institute of Medicine. (1990). Broadening the base of treatment for alcohol problems. Washington: D.C: National Academy Press. Jacobi, F., Wittchen, H.-U., Hotling, C., Höfler, M., Pfister, H., Müller, N. & Lieb, R. (2004). Prevalence, co-morbidity and correlates of mental disorders in the general population: results from the German Health Interview and Examination Survey (GHS). Psychological Medicine, 34, 597-611. Johnson, M., Jackson, R., Guillaume, L., Meier, P. & Goyder, E. (2010). Barriers and facilitators to implementing screening and brief intervention for alcohol misuse: a systematic review of qualitative evidence. Journal of Public Health, 33(3), 412-421. Johnston, A. (2012). First Nations Perspective on the Development of a Needs-Based Planning Model for Substance Use Services and Supports in Canada. Report prepared for the Centre for Addiction and Mental Health, Toronto, Canada.

139

Kaner, E.F., Beyer, F., Dickinson, H.O., Pienaar, E., Campbell, F., Schlesinger, C., Heather, N., Saunders, J. & Burnand, B. (2009). Effectiveness of brief alcohol interventions in primary care populations. Drug and Alcohol Review 28(3), 301-323. Kathol, R. G., Kunkel, E.J.S., Weiner, J.S., McCarron, R.M., Worley, L.L.M., Yates, W.R., Summergrad, P. & Huyse, F.J. (2009). Psychiatrists for medically complex patients: Bringing value at the physical health and mental health/substance-use disorder interface. Psychosomatics, 50(2), 93–107. Kessler, R.C., Aguilar-Gaxiola, S., Berglund, P.A., Caraveo-Anduaga, J.J., DeWit, D.J., Greenfireld, S.F., Kolody, B., Olfson, M. & Vega, W.A. (2001). Patterns and predictors of treatment seeking after onset of a substance use disorder. Archives of General Psychiatry, 58, 1065-1071. Knight, J.R., Goodman, E., Pulerwitz, T. & DuRant, R.H. (2001). Reliability of the Problem Oriented Screening Instrument for Teenagers (POSIT) in adolescent medical practice. Journal of Adolescent Health, 29, 125–130. Knight, J.R., Sherritt, L., Harris, S.K., Gates, E.C. & Chang, G. (2003). Validity of brief alcohol screening tests among adolescents: A comparison of the AUDIT, POSIT, CAGE and CRAFFT. Alcoholism: Clinical & Experimental Research, 27, 67–73. Kohn, R., Saxena, S., Levav, I. & Sacareno, B. (2004). The treatment gap in mental health care. Bulletin of the World Health Organization, 82(11), 858-866. Koski-Jannes, A., Cunningham, J.A., Tolonen, K. & Bothas, H. (2007). Internet-based self assessment of drinking – 3 month follow-up data. Addictive Behaviors, 32: 533-542. Kovess-Masfety, V., Alonso, J., Brugha, .S., Angermeyer, M.C., Haro, J.M., Sevilla-Dedieu, C. & the ESEMeD/MHEDEA 2000 Investigators. (2007). Differences in lifetime use of services for mental health problems in six European countries. Psychiatric Services, 58(2), 213-220. Krupski, A., Sears, J.M., Joesch, J.M., Estee, S., He, L., Dunn, C., Huber, A., Roy-Byrne, P. & Ries, R. (2010). Impact of brief interventions and brief treatment on admissions to chemical dependency treatment. Drug and Alcohol Dependence, 110, 126-136. Lamb, S., Greenlick, M.R. & McCarty, D. (1998). Bridging the gap between practice and research: Forging partnerships with community-based drug and alcohol treatment. Washington, DC: National Academy Press. Lapham, G. T., Hawkins, E.J., Chavez, L. J., Achtmeyer, C. E., Williams, E. C., Thomas, R. M., Ludman, E.J., Kypri, K., Hunt, S. C. & Bradley, K. A. (2012). Feedback from recently returned veterans on an anonymous web-based brief alcohol intervention. Addiction Science & Clinical Practice, 7(17), Available at: http://www.ascpjournal.org/content/7/1/17.

140

Lev-Ran, S., Balchand, K., Lefebvre, L., Araki, K.F. & Le Foll, B. (2012). Pharmacotherapy of alcohol use disorders and concurrent psychiatric disorders: a review. Canadian Journal of Psychiatry, 57(6), 342-349. Lipsky, S. & Caetano, R. (2008). Is intimate partner violence associated with the use of alcohol treatment services? Results from the national survey on drug use health. Journal of Studies on Alcohol and Drugs, 69(1), 30-38. Madras, B.K., Compton, W.M., Avula, D., Stegbauer, T., Stein, J.B. & Clark, H. W. (2009). Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: Comparison at intake and 6 months later. Drug and Alcohol Dependence, 99(1), 280–295. Marshman, J. (Chairperson). (1978). The treatment of alcoholics: An Ontario perspective. The report of the Task Force of Treatment Services for Alcoholics. Toronto: Addiction Research Foundation. Martin, G. & Rehm, J. (2012). The Effectiveness of psychosocial modalities in the treatment of alcohol problems in adults: A review of the evidence. Canadian Journal of Psychiatry, 57(6), 350358. Martin, N. & Johnston, V. (2007). A Time for action: Tackling Stigma and Discrimination. Report to the Health Commission of Canada. Accessed at: http://www.mentalhealthcommission.ca/SiteCollectionDocuments/AntiStigma/TimeforAction_Eng.pdf Maulik, P.K., Eaton, W.W. & Bradshaw, C.P. (2010). Mediating effect of mental disorders in the pathway between life events and mental health service use: Results from the Baltimore Epidemiological Catchment Area study. The Journal of Nervous and Mental Disease, 198(3), 187193. McAlpine, D.D. & Mechanic, D. (2000). Utilization of specialty mental health care among persons with severe mental illness: The roles of demographics, need, insurance and risk. HSR: Health Services Research, 35(1), Part 2, 277-292. McGlynn, E.A., Asch, S.M., Adams, J., Keesey, J., Hicks, J., DeCristofaro, A. & Kerr, E.A. (2003). The quality of health care delivered to adults in the United States. New England Journal of Medicine, 348, 2635-2645. McGonigle, J. J., Krouk, M., Hindmarsh, D. & Campano-Small, C. (1992). Understanding partial hospitalization through a continuity-of-care model. International Journal of Partial Hospitalization 8(2), 135–140.

141

McGovern, M., Matzkin, A.L. & Giard, J. (2007). Assessing the dual diagnosis capability of addiction treatment services: The Dual Diagnosis Capability in Addiction Treatment (DDCAT) Index. Journal of Dual Diagnosis, 3(2), 111-123. McKay, J.R. (2005). Is there a case for extended interventions for alcohol and drug use disorders? Addiction, 100, 1594-1610. McLellan, A.T., Kushner, H., Metzger, D., Peters, R., Smith, I. & Grissom, G. (1992). The fifth edition of the addiction severity index. Journal of Substance Abuse Treatment, 9, 199-213. McQueen, J., Howe, T.E., Allan, L., Mains, D. & Hardy, V. (2011). Brief interventions for heavy alcohol users admitted to general hospital wards. Cochrane Database of Systematic Reviews, 8(2), doi: 10.1002/14651858.CD005191.pub3. Meadows, G., Harvey, C., Fossey, E. & Burgess, P. (2000). Assessing perceived need for mental health care in a community survey: Development of the perceived need for care questionnaire (PNCQ). Social Psychiatry and Psychiatric Epidemiology, 35, 427-435. Mechanic, D. (2003). Is the prevalence of mental disorders a good measure of the need for services? Health Affairs, 22(5) 8-20. Messias, E., Eaton, W., Nestadt, G., Bienvenu, O.J. & Samuels, J. (2007). Psychiatrists' ascertained treatment needs for mental disorders in a population-based sample. Psychiatric Services 58(3), 373-377. Miller, W.R. (2007). Bring addiction treatment out of the closet. Addiction, 102, 863-869. Mitchell, A.J., Meader, N., Bird, V. & Rizzo, M. (2012). Clinical recognition and recording of alcohol disorders by clinicians in primary and secondary care: meta-analysis. The British Journal of Psychiatry, 201, 93-100. Mojtabai, R., Olfson, M. & Mechanic, D. (2002). Perceived need and help-seeking in adults with mood, anxiety or substance use disorders. Archives of General Psychiatry, 59, 77-84. Mojtabai, R. (2009). Unmet need for treatment of major depression in the United States. Psychiatric Services, 60(3), 297-305. Moyer, A., Finney, J.W., Swearingen, C.E. & Vergun, P. (2002). Brief interventions for alcohol problem: a meta-analytic review of controlled investigations in treatment-seeking and nontreatment-seeking populations. Addiction, 97, 279-292. Mueser, K.T., Noordsy, D.L., Drake, R.E. & Fox, L. (2003). Integrated treatment for dual disorders: A guide to effective practice. New York, Guilford Press.

142

National Center on Addiction and Substance Abuse at Columbia University. (2012). Addiction medicine: Closing the gap between science and practice. New York, New York. National Mental Health Strategy. (2004). Responding to the mental health needs of young people in Australia. Australia: National Mental Health Strategy. National Treatment Agency (2002). Models of care for treatment of adult drug misusers. London, United Kingdom. National Treatment Agency for Substance Misuse. (2006). Models of care for treatment of adult drug misusers: Update 2006. London, United Kingdom. National Treatment Strategy Working Group. (2008). A Systems Approach to Substance Use in Canada: Recommendations for a National Treatment Strategy. Ottawa: National Treatment Strategy Working Group. Available: http://www.nationalframeworkcadrenational.ca/uploads/files/TWS_Treatment/nts-report-eng.pdf Nilsen, P., Aalto, M., Bendtsen, P. & Seppa, K. (2006). Effectiveness of strategies to implement brief alcohol intervention in primary healthcare. Scandinavian Journal of Primary Health Care, 24, 5-15. Proudfoot, H. & Teesson, M. (2002). Who seeks treatment for alcohol dependence? Findings from the Australian National Survey of Mental Health and Wellbeing. Social Psychiatry and Psychiatric Epidemiology, 37, 451-456. Quanbeck, A., Wheelock, A., Ford, J. H. 2nd, Pulvermacher, A., Capoccia, V. & Gustafson, D. (2013). Examining access to addiction treatment: scheduling processes and barriers. Journal of Substance Abuse, 44(3), 343-348. Ramchand, R., Marshall, G. N., Schell, T. L., Jaycox, L. H., Hambarsoomians, K., Shetty, V., Hinika, G. S., Cryer, H. G., Meade, P. & Belzberg, H. (2009). Alcohol abuse and illegal drug use among Los Angeles County trauma patients: prevalence and evaluation of single item screener. Journal of Trauma, 66(5), 1461-1467. Rehm, J. (2008). Key future research questions on mediators and moderators of behaviour change processes for substance abuse. International Journal of Methods in Psychiatric Research, 17(S1), S83–S87. Regier, D.A., Myers, J.K., Kramer, M., Robins, L.N., Blazer, D.G., Hough, R. L., Eaton, W. W. & Locke, B.Z. (1984). The NIMH Epidemiologic Catchment Area program. Historical context, major objectives, and study population characteristics. Archives of General Psychiatry. 41(10), 934-941. Reist, D. & Brown, D. (2008). Exploring complexity and severity within a framework for understanding risk related to substance use and harm. Paper presented at the International

143

Symposium on Needs Assessment and Needs-based Planning for Substance Use Services and Supports. Toronto, Ontario: Centre for Addiction and Mental Health. Ritter, A. (2012, June). A national planning model for drug and alcohol treatment in Australia: The brave new world? Symposium presentation at the annual meeting of the College on Problems of Drug Dependence, Palm Springs, USA. Ritter, A., Hailstone, S., Burgess, J., Woods, B. & McGrath, D. Developing a national planning tool for alcohol and other drug treatment services. (under review) Robinson, S., Perkins, S., Bauer, S., Hammond, N., Treasure, J. & Schmidit, U. (2006). Aftercare intervention through text messaging in the treatment of bulimia nervosa - feasibility pilot. International Journal of Eating Disorders, 39(8); 633- 638. Roche, A.M. & Freeman, T. (2004). Brief interventions: good in theory but weak in practice. Drug and Alcohol Review, 23, 11-18. Rockett, I.R.H., Putnam, S.L., Jia, H., Chang, C.F. & Smith, G.S. (2005). Unmet substance abuse treatment need, health services utilization, and cost: A population-based emergency department study. Annals of Emergency Medicine, 42(2), 118-127. Rush, B. R. (1990). A systems approach to estimating the required capacity of alcohol treatment services. British Journal of Addiction, 85, 49-59. Rush, B.R. (in press). Addiction assessment and treatment planning in developing countries. In N. El-Guebaly, M. Galanter, & Carra, G. (Eds.). Textbook of Addiction Treatment: International Perspectives. Springer Publishing. Rush, B. & Wild, T. C. (2003). Substance abuse treatment and pressures from the criminal justice system: Data from provincial client monitoring system. Addiction, 98, 1119-1128. Rush, B. (2012). A perspective on the effectiveness of interventions for alcohol and other substance use disorders. The Canadian Journal of Psychiatry, 57(6), 273–275. Rush, B.R. & Nadeau, L. (2011). On the integration of mental health and substance use services and systems. In D. Cooper (Ed.) Responding in Mental Health-Substance Use, Book 3, Chapter 13. Mental Health-Substance Use Book Series: Oxford, UK, Radcliffe Publishing Ltd. pp. 148-175 Rush, B. & Castel, S. (2011 September). Screening for co-occurring mental and substance use disorders: Telling the story of knowledge translation from research validation to clinical application. Paper presented at the World Psychiatric Congress, Buenos Aires, Argentina. Rush, B. (2010). Tiered frameworks for planning substance use service delivery systems: Origins and key principles. Nordic Studies on Alcohol and Drugs, 27, 617-636.

144

Rush, B.R., Scott, C. K., Dennis, M.L., Castel, S. & Funk, R. (2008). The interaction of co-occurring psychiatric problems and recovery management checkups. Evaluation Review, 32(1), 7-38. Rush, B. (2008). Symposium report: International symposium on needs assessment and needsbased planning for substance use services and supports. Toronto, ON: Centre for Addiction and Mental Health. Rush, B.R., Castel, S., Brands, B., Toneatto, T. & Veldhuizen, S. (2013). Validation and comparison of diagnostic accuracy of four screening tools for mental disorders in people with substance use disorders. Journal of Substance Abuse Treatment, 44(4), 375-383. Rush, B.R., Martin, G., Corea, L. & Khobzi, N. (2012). Engaging stakeholders in review and recommendations for models of outcome monitoring for substance abuse treatment. Substance Use and Misuse, 47(12), 1293-1302. Ryan, R. & Deci, E. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78. Saitz, R. (2012). Screening and brief intervention (SBI): Has is hit the tipping point? Addiction Science & Clinical Practice, 7(14). Available at: http://www.ascpjournal.org/content/7/1/14. Sareen, J., Jagdeo, A., Cox, B.J., Clara, I., ten Have, M., Belik, S.L., de Graaf, R. & Stein, MB. (2007). Perceived barriers to mental health service utilization in the United States, Ontario and the Netherlands. Psychiatric Services, 58(3), 357-364. Saunders, J.B, Aasland, O.G., Babor, T.F., De la Fuente, J.R. & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful alcohol Consumption II. Addiction, 88, 791-804. Schippers, G.M., Schramade, M. & Walburg, J.A. (2002). Reforming Dutch substance abuse treatment services. Addictive Behaviors, 27, 995-1007. Schmidt, L.A., Ye, Y., Greenfield, T.K. & Bond, J. (2007). Ethnic disparities in clinical severity and services for alcohol problems: Results from the national alcohol survey. Alcoholism: Clinical and Experimental Research, 31(1), 48-56. Scott, C.K., Dennis, M.L. (2009). Results from two randomized clinical trials evaluating the impact of quarterly recovery management checkups with adult chronic substance users. Addiction, 104, 959–971. Selby, P., Van Mierlo, T., Voci, S.C., Parent, D. & Cunningham, J.A. (2010). Online social and professional support for smokers trying to quit: An exploration of first time posts from 2562 members. Journal of Medical Internet Research, 12(3), e24. doi: 10.2196/jmir.1340 Shapiro, J.R., Bauer, S., Andrew, E., Pisetsky, E., Bulik-Sullivan, B., Hamer, R.M. & Bulik, 145

C.M. (2010). Mobile therapy: Use of text-messaging in the treatment of bulimia nervosa. International Journal of Eating Disorders, 43(6), 513-519. Shepard, D.S., Strickler, G.K., McAuliffe, W.E., Beaston-Blaakman, A., Rahman, M. & Anderson, T. (2005). Unmet need for substance abuse treatment of adults in Massachusetts. Administration and Policy in Mental Health, 32(4), 403-426. Slade, T., Johnston, A., Oakley Brown, M.A., Andrews, G. & Whiteford, H. (2009). 2007 National Survey of Mental Health and Wellbeing: methods and key findings. Australasian Psychiatry, 43(7), 594-605. Smith, P. C., Schmidt, S. M., Allensworth-Davies, D. & Saitz, R. (2010). A single-question screening test for drug use in primary care. Archives of Internal Medicine, 170(13), 1155-1160. Sobell, M.B. & Sobell, L.C. (2000). Stepped care as a heuristic approach to the treatment of alcohol problems. Journal of Consulting and Clinical Psychology, 68, 573–579. Statistics Canada. (2002). Canadian Community Health Survey (CCHS) - Mental Health and Wellbeing - Cycle 1.2. Available at: http://www.statcan.gc.ca/concepts/health-sante/cycle1_2/indexeng.htm Stockdale, S.E., Klap, R, Belin, T.R., Zhang, L. & Wells, K.B. (2006). Longitudinal patterns of alcohol, drug, and mental health need and care in a national sample of U.S. adults. Psychiatric Services, 57(1), 93-99. Tait, R.J. & Hulse, G. K. (2005). Adolescent substance use and hospital presentations: A record linkage assessment of 12-month outcomes. Drug and Alcohol Dependence, 79, 365-371. Tait, R.J., Hulse, G.K. & Robertson, S.I. (2004). Effectiveness of a brief intervention and continuity of care in enhancing attendance for treatment by adolescent substance users. Drug and Alcohol Dependence, 74, 289-296. Teesson, M., Baillie, A., Lynskey, M., Manor, B. & Degenhardt, L. (2006). Substance use, dependence and treatment seeking in the United States and Australia: A cross-national comparison. Drug and Alcohol Dependence, 81(2), 149–155. Tempier, R., Meadows, G.Hn, Vasiliadis, H.M., Mosier, K.E., Lesage, A., Stiller, A., Graham, A. & Lepnurm, M. (2009). Mental disorders and mental health care in Canada and Australia: Comparative epidemiological findings. Social Psychiatry and Psychiatric Epidemiology, 44, 63-72. ten Have, M., de Graaf, R., Ormel, J., Vilagut, G., Kovess, V. & Alonso, J. (2010). Are attitudes towards mental health help-seeking associated with service use? Results from the European study of epidemiology and mental disorders. Social Psychiatry and Psychiatric Epidemiology, 45, 153163.

146

Timko, C., Moos, R.H., Finney, J.W., Moos, B.S. & Kaplowitz, M.S. (1999). Long-term treatment careers and outcomes of previously untreated alcoholic. Journal of Studies on Alcohol, 60(4), 437447. Toche-Manley, L., Grissom, G., Dietzen, L. & Sangsland, S. (2011). Translating addictions research into evidence-based practice: The Polaris CD outcomes management system. Addictive Behaviors, 36(6), 601–607. Urbanoski, K.A., Rush, B.R., Wild, C., Bassani, D.G. & Castel, S. (2007). Use of mental health care services by Canadians with co-occurring substance dependence and mental disorders. Psychiatric Services 58(7), 962-969. Urbanoski, K.A. & Wild, T.C. (2012). Assessing self-determined motivation for treatment: Validation of the Treatment Entry Questionnaire. Journal of Substance Abuse Treatment, 43(1), 7079. Vasiliadis, H.-M., Tempier, R., Lesage, A. & Kates, N. (2009). General practice and mental health care: Determinants of outpatient service use. Canadian Journal of Psychiatry, 54(7), 468-476. Wagner, E. H. (1998). Chronic disease management: What will it take to improve care for chronic illness? Effective Clinical Practice, 1(1), 2–4. Wallace, P. J. (2005). Physician involvement in disease management as part of the CCM. Health Care Financing Review, 27(1), 19–31. Wang, P.S., Lane, M., Olfson, M., Pincus, H.A., Wells, K.B. & Kessler, R.C. (2005). Twelve-month use of mental health services in the United States: Results from the national comorbidity survey replication. Archives of General Psychiatry, 62, 629-640. Williams, E.C., Johnson, M.L., Lapham, G.T., Caldeiro, R.M., Chew, L., Fletcher, G.S., McCormick, K.A., Weppner, W.G. & Bradley, K.A. (2011). Strategies to implement alcohol screening and brief intervention in primary care settings: A structured literature review. Psychology of Addictive Behaviors, 25(2), 206-214. Winters, K.C., Stinchfield, R.D., Opland, E., Weller, C. & Latimer, W.W. (2000). The effectiveness of the Minnesota Model approach in the treatment of adolescent drug abusers. Addiction, 94(4), 601-612. Wood, R.T.A. & Wood, S.A. (2009). An evaluation of two United Kingdom online support forums designed to help people with gambling issues. Journal of Gambling Issues, 23, 5-30. World Health Organization. (1996). Ottawa Charter for Health Promotion First International Conference Ottawa (WHO/HPR/HEP 95.1), Geneva, World Health Organization. World Health Organization (1998) Health Promotion Glossary. WHO, Geneva. 147

Wu, L-T., Kouzis, A.C. & Schlenger, W.E. (2003). Substance use, dependence, and service utilization among the US uninsured nonelderly population. American Journal of Public Health 93(12), 20792085. Wu, L-T. & Ringwalt, C.L. (2004). Alcohol dependence and use of treatment services among women in the community. American Journal of Psychiatry, 161(10), 1790-1797. Wu, L-T. (2010). Substance abuse and rehabilitation: responding to the global burden of disease attributable to substance abuse. Substance Abuse and Rehabilitation, 1, 5-11. Yacoubian, G.S. (2003). Measuring alcohol and drug dependence with New York City ADAM data. Journal of Substance Abuse Treatment, 24, 341-345. Zimmerman, M. & Mattia, J.I. (2001). The psychiatric diagnostic screening questionnaire: Development, reliability and validity. Comprehensive Psychiatry, 42(3), 175-189.

148

Appendix A: Research Team and National Advisory Committee members National Advisory Committee: Carol Hopkins, National Native Addictions Partnership Foundation, Muskoday, Saskatchewan David Brown, Evaluation Consultant, Winnipeg, Manitoba Gina Rideout, Department of Health and Community Services, St. John’s, Newfoundland Heather Bullock, Centre for Addiction and Mental Health, Toronto, Ontario Jesse Jahrig, Alberta Health Services, Edmonton, Alberta John Topp, Pavillon Foster, St-Philippe-de-Laprairie, Québec Nadine Blanchette-Martin, Centre de réadaptation Ubald-Villeneuve (CRUV) and Centre de réadaptation en dépendance de Chaudière-Appalaches, Québec Rebecca Hansen, Alcohol and Drug Services, Whitehorse, Yukon Rebecca Jesseman, Canadian Centre on Substance Abuse, Ottawa, Ontario Sherry Mumford, Addiction Services and Langley and Maple Ridge Mental Health, British Columbia Terry Gudmundson, Ministry of Health, Saskatchewan Shauna Reddin, Health PEI Margaret Kennedy, Health PEI Wanda McDonald, Department of Health and Wellness, Nova Scotia Project Research Team: Project Lead: Brian Rush, Centre for Addiction and Mental Health, Toronto, Ontario Quebec Lead: Joël Tremblay, Université du Québec à Trois-Rivières, Québec Project Coordinator: Chantal Fougere, Centre for Addiction and Mental Health, Toronto, Ontario Research Analyst: Anna-Marie Danielson, Centre for Addiction and Mental Health, Toronto, Ontario

149

Project Coordinator: Renée Behrooz, Centre for Addiction and Mental Health, Toronto, Ontario Research Analyst: Wendi Perez, Centre for Addiction and Mental Health, Toronto, Ontario Research Coordinator: Julia Fineczko, Centre for Addiction and Mental Health, Toronto, Ontario

150

Appendix B: Logic Model – Development of Needs-Based Planning Models for Services and Supports in Canada Model Development

Model Expansion

Develop a needs-based planning model for

specialized substance use services and Article I. supports.

(i) C o m p o n e n t s (iii) A c ti v it i e s

(ii) Outp uts

Expand the needs-based planning model to estimate the needs of special populations.

Pilot Testing and Model Enhancements

Evaluation, Project Management and Dissemination

Pilot and evaluate the needs-based model to refine & share with engaged jurisdictions across Canada.

Manage and evaluate the project and share lessons learned.

 Undertake literature review and access reviews from other DTFP projects  Apply current data on use of services to begin to populate the initial model and supplement with secondary data sources  Define and model components, pathways & parameters of service use in the ideal treatment system  Analyze available hospital & emergency services data in Quebec

 Expand literature review to those populations missed in the national survey data  Develop a plan to estimate the needs among Aboriginal, homeless, armed forces, and institutionalized populations

 Develop criteria for selection of pilot sites  Engage up to four settings/regions to pilot the model  Develop protocols and guidelines for pilot testing  Implement model at pilot sites and collect data  Refine model based on pilot results

 Incorporate into pilot sites the qualitative evaluation data collection component  Progress monitoring  Coordinate with DTFP projects  Develop recommendations for application of model beyond present project life cycle  Use knowledge exchange networks to disseminate information

 Literature review  Secondary data report  Mapping report: Specialized sector, mental health, and primary care  Planning model and toolkit for application

   

 Protocol for pilot testing  Revised planning model and toolkit  Recommendations

   

Immediate Outcomes

Literature review - expanded Data analysis reports Secondary data report Expanded planning model and toolkit  Recommendations

Evaluation plan Evaluation report Dissemination toolkits Communication updates and websites

Increased access to models and toolkits for needs-based planning and allocating resources for substance use services and supports.

Increased awareness and engagement of decision-makers concerning needs-based planning models and their relative advantage over existing approaches.

Intermediate Outcomes

Enhanced PT commitments to affect system change in DTFP treatment systems’ investment areas [with needs-based planning models].

Increased PT capacity to plan and evaluate substance abuse treatment systems’ capacity and performance. Longer-Term Outcomes

  

Across Canadian jurisdictions, increased use of needs-based planning models for substance use services. Increased ability to systematically allocate resources to better meet needs of individuals accessing services in all relevant health sectors. Improved decisions for resource allocation for substance use services and systems.

Strengthened evidence-informed substance abuse treatment systems including:

Development a Needs-Based Planning Model Substance Services in Canada  Clientoflevel outcomes such as reduced harmsfor associated withUse substance useand and Supports improved health and quality of life outcomes; 

System level outcomes such as better balance in continuum of care, improved continuity of care, increased penetration to in-need populations, and improved population health outcomes.

151

Appendix C: Category definitions and CCHS 1.2 variables

Definition

Code

Corresponding Value

Maximum Alcohol use is 5+ drinks in one occasion

Development of a Needs-Based Planning Model for

Development of a Needs-Based Planning Model for

Suggest Documents

Development of a Monte Carlo model for treatment planning dose ...

Cost Model for Planning, Development and Operation of a ...

MODEL OF SCENARIO DEVELOPMENT IN DEFENSE PLANNING

Development of an interactive model for planning the care workforce ...

Development of an Assortment-Planning Model for Fashion ...

The development of a sketch planning model to ...

Population and Development Planning (PDP) Model

A Hierarchy Planning Model for TFT-LCD

A stochastic model for operating rooms planning

A Model for Master Planning in Semiconductor

A simple clinical model for planning transfusion

proposal for development of a system for planning radiotherapy of ...

Improved Geo-ICT model GramyaVikas for rural development planning.

A Model-Driven Technique for Development of

DEVELOPMENT OF A MODEL FOR LAND SURVEY

Development of a neural network model for

Development of a Prediction Model for Colorectal

development of a radioecological model for ...

Development of a statistical model for discrimination

A Sysems-Approach Model for Conservation Planning of a Hilly

Scenario planning for innovation development

Transportation Planning for Sustainable Development

Cost Model for Planning, Development and Operation of a Data Center

Development of urban planning guidelines for ...