Implementing a Continuum of Evidence-Based Psychosocial Interventions for People With Severe Mental Illness: Part 1—Review of Major Initiatives and Implementation Strategies

Abstract

Objective:

Providing comprehensive care to people with severe mental illness (SMI) involves moving beyond pharmacological treatment and ensuring access to a wide range of evidence-based psychosocial services. Numerous initiatives carried out in North America and internationally have promoted the widespread adoption of such services. Objectives of this rapid review were 3-fold: to identify these implementation initiatives, to describe the implementation strategies used to promote the uptake of psychosocial services, and to identify key issues related to the implementation of a broad range of services. Part 1 presents findings for objectives 1 and 2 of the review.

Method:

Searches were carried out in MEDLINE and PsycINFO for reports published between 1990 and 2012 using key words related to SMI, psychosocial practices, and implementation. Contacts with experts and reference list and reverse citation searches were also conducted.

Results:

Fifty-five articles were retained that identified more than a dozen major North American and international implementation initiatives. Initiative leaders employed diverse strategies at the planning, execution, and evaluation stages of the implementation process. Stakeholder meetings, training, ongoing consultation, and quality or fidelity monitoring were strategies consistently adopted across most initiatives, whereas theory-based approaches and organizational- and system-level strategies were less frequently described.

Conclusion:

Insights from the initiatives identified in this review can help guide future efforts to implement a broad range of psychosocial services for people with SMI. However, such efforts will also need to be informed by more rigorous, theory-based studies of implementation processes and outcomes.

Most mental health authorities and care providers acknowledge that providing comprehensive health care to people with SMI involves more than pharmacological treatment and that access to psychosocial supports are critical for promoting recovery and improving lives. There is also growing consensus on the broad range of evidence-based psychosocial supports that should be routinely available in mental health care (Table 1). Nevertheless, despite this agreement, across jurisdictions many such supports are either unevenly applied or completely inaccessible to most people with SMI.^1–4 For health system stakeholders, the challenge today has become less about identifying best practices but rather implementing a continuum of available, effective practices and ensuring that these are both accessible and sustainable.^5–7

Table 1.

Evidence-based psychosocial interventions

Intervention domain and types

Illness self-management, medication management, health promotion, and (or) psychological interventions Healthy living interventions (for example, weight management, nutritional counselling, smoking cessation, and exercise programs) Patient psychoeducation Family psychoeducation and (or) family supports CBT and (or) other psychotherapies Cognitive remediation Motivational interviewing Social interventions Life skills and (or) social skills training Supported education Supported employment Supported housing Clubhouses and (or) recreational interventions Service level interventions ACT and (or) community crisis interventions and (or) other outreach interventions Early intervention programs for psychosis Integrated care for people with dual disorders Peer supports and (or) consumer-run interventions Case management Recovery-oriented services

Since the early 1990s, several major initiatives have attempted to make a broader range of psychosocial EBPs available to people with SMI. Thus we conducted a rapid review with 3 objectives: identify these initiatives, describe the main strategies adopted throughout the process of implementing these EBPs, and identify critical issues for stakeholders to consider when seeking to implement a broader range of psychosocial EBPs for people with SMI within complex health systems. Part 1 reports on the findings from review objectives 1 and 2.

Clinical Implications

• Seventeen types of psychosocial interventions can be considered EBPs that should be made widely available within routine mental health and addictions services.

• Implementing these interventions involves a range of strategies related to the planning of the implementation, the engagement of partners, and the execution and evaluation of the implementation.

• Numerous initiatives have successfully implemented multiple evidence-based psychosocial practices, and there is evidence that these practices have led to improved outcomes for people with SMI.

Limitations

• Methods adopted to streamline the review process may have led to missing some references for initiatives. Through alternative search strategies and contacts with experts, we aimed to minimize this risk.

• Some relevant implementation initiatives were excluded from the review because they failed to meet the established inclusion criteria, notably the requirements to describe initiatives for people with SMI and implementation activities for multiple types of psychosocial interventions.

Method

Literature Search

Rapid reviews are literature reviews that aim to inform health system planning and policy development while employing methods to streamline the traditional systematic review process.⁸ For this review, comprehensive searches were carried out in MEDLINE and PsycINFO for English- and French-language literature using terms for SMI (schizophrenia, depression, BD, and substance use disorders), psychosocial EBPs (the interventions listed in Table 1), and implementation (implementation, dissemination, uptake, diffusion, innovation, knowledge transfer, and knowledge translation). Appropriate wild cards were used to account for plurals and spelling variations. Databases were searched from January 1990 to March 2012. Further relevant studies were identified through the French-language portal Érudit, reference list searches, and reverse citation searches of key articles, and consultations with experts in psychosocial EBPs.

Study Selection

The initial search yielded 2816 reports. One author screened the titles and abstracts to determine eligibility for full-text screening. Reports were excluded when they focused on nonadult participants (that is, youth below 18 years of age), participants with less SMI (for example, adjustment disorder and nicotine addiction) or chronic physical conditions, and interventions that were not psychosocial in nature. Also excluded were reports that did not discuss implementation initiatives, strategies, or issues. Finally, we excluded reports that addressed implementation strategies or issues for only one type of psychosocial EBP. This latter criterion stemmed from our interest in understanding the specific approaches and considerations surrounding the implementation of a broader continuum of psychosocial EBPs, defined as 2 or more of the intervention types presented in Table 1.

Based on this initial screening, 2682 reports were excluded. The full texts of the 134 remaining articles were independently screened for eligibility by both authors. Following this second screening, 27 articles were excluded because they did not address psychosocial interventions or a continuum of interventions (n = 11), they did not discuss implementation strategies or issues (n = 15), or because adults were not targeted (n = 1).

We extracted the following data from the 107 selected articles: implementation initiatives and context (for example, country, key policies, or organizations supporting the initiative), stakeholders involved and consultation process, leadership for initiatives, psychosocial interventions targeted and implemented, theories or ideas guiding the implementation, implementation strategies used, implementation outcomes, and implementation issues encountered. Constructs chosen for data extraction were consistent with the implementation framework developed by Damschroder et al⁹ and the implementation strategies compiled by Powell et al.¹⁰

Conceptual Framework

The CFIR developed by Damschroder et al⁹ was used to provide an overarching typology to identify critical stages of the implementation process as well as key constructs known to influence this process. According to the CFIR, the implementation process consists of 4 main types of activities: planning, engaging, executing, and reflecting and evaluating activities. Inspired by Powell’s list of implementation strategies,¹⁰ we matched the different implementation strategies identified through the review to these activity types (Table 2).

Table 2.

Implementation activities for major implementation initiatives

Implementation activity	Implementation initiatives
	NEBPP	VHA-EQUIP-2	VHA-QI	NIDA-SAMHSA	STEP-BD	MHTS	Housing First—United States	Housing First—Canada	OTP	UK DH	Israel
Planning and engagement
Assessment of local needs, readiness, barriers, and (or) facilitators	✓	✓	✓	✓				✓		✓
Select guiding theory	✓	✓
Stakeholder consultations and consensus building	✓	✓	✓	✓	✓			✓	✓	✓	✓
Flexibility and (or) tailoring of implementation	✓	✓		✓	✓	✓		✓	✓	✓
Identify champions	✓	✓	✓	✓	✓
Execution
Local and (or) regional support teams	✓	✓	✓	✓				✓		✓	✓
Educational materials	✓	✓	✓	✓	✓	✓	✓		✓	✓
Training or supervision	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Train-the-trainer approach			✓	✓	✓
Consultation or technical assistance	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Job shadowing	✓						✓
Create new clinical roles and (or) teams	✓	✓	✓			✓	✓	✓	✓	✓	✓
Revise professional and (or) service standards	✓		✓
Financial incentives for EBP adoption	✓									✓
New funding for implementation	✓		✓				✓			✓
Revise service contracts	✓
Modify billing procedures	✓
Quality or fidelity monitoring	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓	✓
Mechanisms for client feedback	✓		✓					✓	✓		✓
Encourage client advocacy	✓						✓
New government legislation and (or) policy	✓						✓			✓	✓
Reflection and evaluation
Implementation process monitoring	✓	✓		✓	✓	✓		✓		✓	✓
Share lessons and best practices	✓	✓	✓	✓						✓

The CFIR also identifies 5 domains that regroup constructs theorized to play a role in the implementation process, namely, the characteristics of the intervention, the characteristics of the individuals involved, the inner setting, the outer setting, and the implementation process. These domains are described further in part 2 of this review.¹¹ Taken together, our conceptual framework allowed us to gain a comprehensive understanding of the implementation experiences described in the literature.

Results

North American Initiatives

National Implementing Evidence-Based Practice Project

The NEBPP was a large, multi-site demonstration project launched in the late 1990s and led by researchers from the Dartmouth Psychiatric Research Center. The project’s aims were to facilitate and study the implementation of 5 EBPs in community mental health settings in 8 US states: ACT, SE, FPE, illness management and recovery skills, and integrated dual disorders treatment for substance abuse and mental illness.^5,12,13

Informed by decades of implementation efforts and theory,^14,15 project leaders’ overarching implementation strategy reflected 3 aims: to engage multiple stakeholders in an intensive, multi-faceted implementation process; to follow a staged approach designed to explicitly address motivations, enablers, and reinforcers of change; and to use social marketing concepts and account for the particular needs and barriers of stakeholders and implementation sites.¹² Consistent with this vision, leaders mobilized a wide variety of stakeholders and sponsors around the project and actively involved them in its realization and evaluation. Multi-stakeholder (researchers, clinicians, managers, service users, and family members) teams developed resources (tool kits) to support implementation of each of the 5 EBPs, and systematic training, consultation, and fidelity monitoring activities were carried out by the research team and trainer–consultants working closely with sites.^15–17 Technical assistance centres established in some states provided additional support,^15–17 and state-level organizations removed barriers through diverse social, organizational, financial, and policy activities.¹⁸

EBPs were introduced into more than 50 public-sector mental health agencies, with 2 of 5 EBPs being targeted in each participating state.¹⁷ All sites participated in a range of evaluation activities, including biannual fidelity assessments and monthly site visits and interviews with research assistants assigned to each site.¹⁷ Results from evaluations revealed that fidelity to EBP standards rose after 24 months (eTable 3), but that fidelity was harder to achieve with some EBPs, compared with others. No patient-level outcomes were measured.

Veterans Health Administration of the US Department of Veteran Affairs

Research on psychosocial interventions has been carried out in the VHA since the 1980s within the VA Office of Research and Development, including dissemination projects on ACT¹⁹ and SE.²⁰ In 1998, the Quality Enhancement Research Initiative program was created to improve the quality of VA health care services through the use of research-derived best practices.²¹ One demonstration project emerging from this program was the EQUIP project, which aimed to evaluate the effectiveness of psychosocial interventions for veterans with schizophrenia in 4 VHA service regions while studying the implementation process.²² The 3 EBPs targeted were FPE, a wellness (for example, weight management) program, and SE services. Implementation efforts were guided by 2 main theory-based models of practice change and were assessed through a multi-faceted formative evaluation.²² While a wide range of strategies were adopted to promote implementation of EBPs, initial findings have revealed limitations in the uptake of EBPs and few positive outcomes for service users.^23,24

Improving access to psychosocial services has also been a central focus of VHA service reforms during the past decade.^25,26 Quality improvement efforts (VHA-QI) have involved the introduction of new mental health service requirements and accreditation standards for VHA facilities, the hiring of Local Recovery Coordinators and Peer Support Technicians in local services, and massive training efforts through psychosocial rehabilitation training programs and so-called train-the-trainer approaches.^25,26 Reports suggests that these activities have significantly increased the amount of psychosocial services available (for example, ACT, CBT, SE, and social skills training), though whether new services meet quality standards and produce positive outcomes for service users remains unclear.

National Institute on Drug Abuse and Substance Abuse and Mental Health Services Administration

In 2001, NIDA and SAMHSA partnered to establish a Blending Initiative to accelerate the implementation of research findings from NIDA-sponsored treatment studies into community-based practice.^27,28 A central feature of the Blending Initiative is a partnership between NIDA’s National Drug Abuse Treatment CTN, which conducts multi-site clinical trials on substance abuse treatments in community-based treatment settings, and SAMHSA’s ATTCs, which provide education, training, and technical assistance to substance abuse treatment providers across the United States.^27,29 Together with other partners, the CTN and ATTCs promote the uptake of specific treatments demonstrated to be effective in CTN trials.

To date, blending initiatives have been undertaken for 2 psychosocial EBPs, namely, motivational incentives treatment and motivational interviewing.^27,28,30 Initiatives typically feature several implementation strategies, including the creation of multi-stakeholder blending teams to create implementation support tools (blending products) based on the findings from CTN trials, Regional Blending Conferences to raise awareness of EBPs and blending products, train-the-trainer approaches, and finally strategic partnerships between state substance abuse authorities and community-based treatment providers to identify strategies for EBP implementation.^28,29 The initiatives for these 2 psychosocial EBPs have led to the training of several thousand clinicians in these practices, though no evaluation of patient-level outcomes was identified.

STEP-BD Project

The STEP-BD project was a national, longitudinal infrastructure for clinical trials that compared the effectiveness of 4 psychosocial practices (CBT, family focused therapy, interpersonal and social rhythm therapy, and patient education with illness self-management) for people with varying presentations of bipolar depression.³¹ The project was the largest federally funded treatment study ever conducted for BD, enrolling more than 4000 participants from specialty programs of 15 institutions across the United States.^32,33

A key feature of STEP-BD was its aim to transform critical care for BDs in the context of the project.³¹ Specifically, clinical teams at each site were responsible for administering 3 out of 4 psychosocial practices, with a team of national experts providing training and support in the form of training workshops, treatment manuals, journal articles, didactic videos for each practice, telephone supervision, and monthly conference calls.^32,34 The project was successful in promoting high-fidelity practices that translated into improved outcomes related to patient functioning and recovery.³³

Mental Health Treatment Study

The MHTS, conducted from 2006 to 2010, aimed to evaluate employment policies and services for people with SMI receiving Social Security Disability Insurance in the United States.^35,36 Study leaders hypothesized that access to SE and other psychosocial supports, coupled with the removal of some known programmatic disincentives, would improve work-related outcomes in insurance beneficiaries relative to usual services.³⁶

Twenty-three community mental health agencies across 19 US states participated in the study.³⁶ Implementation strategies resembled those used in the NEBPP, though nurse care coordinators were also hired to facilitate implementation of SE and other psychosocial services (for example, case management, social skills training, financial and housing assistance, and family counselling), and promote integration between treatment providers and SE staffs. The initiative was highly effective in helping study sites achieve high-fidelity SE, promoting case management services, and improving both employment- and health-related outcomes in insurance beneficiaries with SMI.³⁶

Housing First and At Home/Chez soi

In the early 1990s, the Housing First service model emerged in New York as an innovative approach to the provision of housing and treatment services for homeless people with SMI.^37,38 The program provided clients with immediate access to permanent housing and extensive treatment and psychosocial supports through multidisciplinary ACT teams.³⁸ In contrast to other available supported housing programs, service users’ participation in mental health or substance abuse treatment was not a prerequisite for access to housing.³⁷ Evidence supporting the model led to multiple replications within New York state and eventually across the United States and abroad.^38,39 To support implementation, Housing First originators provided replication sites with technical assistance and training in ACT and techniques, such as harm reduction and motivational interviewing.³⁸Training materials, fidelity assessments, and job shadowing activities are also now available.³⁹

In 2009, the largest pragmatic trial of the Housing First model was launched in 5 Canadian cities sponsored by the Mental Health Commission of Canada and supported by the model’s originators.^40,41 One site in each city received training and ongoing consultation from a centralized team of researchers to provide housing and ACT services. Further, extensive planning, both expert-driven and community-based in nature, and facilitated by site coordinators, helped stakeholders develop a common vision of the ongoing project and allow it to move forward at each site.⁴¹ No fidelity or service user outcomes have been reported to date.

Other North American Initiatives

Our review identified several other past or current initiatives aiming to broaden access to evidence-based psychosocial services for people with SMI.^26,42,43 Training and consultation, among other activities, have been central to these smaller-scale implementation efforts.

International Initiatives

Optimal Treatment Project

The OTP was an international, multi-site pragmatic trial evaluating the effectiveness of evidence-based biomedical and psychosocial treatment strategies for psychotic disorders.⁴⁴ Launched in 1994 and led by Dr Ian Falloon, the OTP enrolled more than 1000 patients in over 80 centres across Europe and in Canada, Japan, Australia, and New Zealand.⁴⁵ Psychosocial practices promoted by the project included patient education and FPE, stress management training, assertive case management and outreach services, social living skills training, and CBT.^44–46

The OTP’s implementation strategy comprised 3 main activities: the creation of multidisciplinary clinical teams in each study centre that would receive training in optimal treatments, internal and external fidelity audits conducted several times a year, and technical assistance provided by OTP leaders, including Dr Falloon himself.^44,45 After 24 months, fidelity of implementation was found to be good to excellent (though no data were provided), and patients receiving optimal treatment had improved clinical and social outcomes relative to those in routine care.⁴⁵

UK Department of Health and National Institute for Health and Clinical Excellence

In the United Kingdom, several recent national mental health policy reforms have included the objective of promoting full-scale implementation of several psychosocial services, such as early intervention services, assertive outreach, and crisis services, in the NHS.^47,48 Reforms were initiated by a 10-year blueprint for mental health services issued by the DH⁴⁷ and were supported by local and national implementation teams, new funding programs, specified service targets, and guidance for management and operational procedures for practice implementation.^47–49 Evaluations suggest that these policy actions have been effective in embedding these EBPs within the NHS.^50,51

Implementation of EBPs has also been facilitated by the creation of the NICE, an arm’s-length body funded by the DH that develops guidance for the NHS about clinical- and cost-effectiveness.⁵² NICE produces comprehensive clinical guidelines for a range of mental disorders, many of which feature guidance on multiple psychosocial treatments. Public-sector NHS Trusts are obliged to consider how they will implement NICE guidelines and can receive additional funding for their implementation.⁵³ NICE also works to support guideline implementation.^52,54 To date, however, the few reports examining NICE guideline implementation have focused on implementation challenges rather than strategies or outcomes.^53,55,56

Psychiatric Rehabilitation in Israel

Psychiatric rehabilitation services have developed rapidly in Israel during the past decade, spurred largely by advocacy groups that have successfully pressured the government to improve access to psychosocial services.⁵⁷ New legislation was introduced that required providers to deliver a comprehensive basket of services, including SE, FPE, housing services, supported education, leisure activities, and case management services.⁵⁸ Implementation was supported by regional rehabilitation committees and a national rehabilitation council responsible for planning rehabilitation services in communities, improving quality and accessibility of services, and advising the government on rehabilitation policy.⁵⁸ A National Outcome Rehabilitation Monitoring Implementation Project was recently launched to perform routine evaluations of the services and impacts created by the rehabilitation legislation.⁵⁸

Discussion

The literature examining the implementation of psychosocial EBPs for people with SMI is vast and spans several decades. More recently, however, deliberate attempts have been made to broaden the range of supports available by facilitating the uptake of multiple EBPs in routine mental health care. This review identified more than a dozen such initiatives, nearly half of which attempted to implement EBPs across all 3 psychosocial intervention domains.

Initiatives differed from one another regarding the leaders that drove the change process and the types of initiative pursued. Several initiatives, such as the NEBPP, EQUIP, STEP-BD, MHTS, At Home/Chez soi, and OTP projects, were large demonstration projects and effectiveness trials conceived mainly by researchers interested in both increasing access to EBPs and studying the implementation process. In other cases, initiatives involved broad service reforms resulting from new government policies and legislation, as observed in the United Kingdom and Israel. National agencies also led initiatives (NIDA-SAMHSA), as did health system administrators (VHA), and community-based service providers (Housing First). This finding highlights the wide range of stakeholders concerned by psychiatric rehabilitation services and demonstrates that change can arise from various sources and take many forms. Interestingly, while leaders employed a diversity of approaches and strategies to enhance psychosocial services, several patterns emerged across initiatives. At the planning stage, stakeholder engagement and flexible implementation approaches were commonly emphasized. Adopting conceptual or theoretical frameworks to guide implementation was not common, however, despite repeated calls for such approaches in the literature.^59,60 Models, concepts, or theories should inform future implementation efforts as they can help frame objectives, plan activities, identify mechanisms of action, and increase the likelihood of producing generalizable knowledge.⁶¹ At the execution stage, educational strategies, such as training activities or technical assistance, were core features of all initiatives. Similarly, monitoring of service quality or fidelity was also consistently pursued, though the extent to which authors reported actual data from these assessments varied considerably. In contrast, there were much fewer reports on organization- and system-level activities to support implementation of psychosocial EBPs.

Most initiatives also evaluated their implementation efforts. Evaluations were often limited to simple assessments of the number of activities carried out and participants involved. Only one-third of initiatives, all researcher-driven trials, produced data related to the fidelity of implementation and (or) impacts on service users. This is striking, as most initiatives were costly, large-scale projects. By their nature, the EBPs targeted within these initiatives can be considered complex,⁶² and thus require more complex, multi-faceted evaluations. Indeed, implementation research has repeatedly shown that leaders cannot assume that their selected interventions will produce desired results, even when these are supported by strong empirical evidence. Misfires can occur at any point during the implementation process and different outcomes are likely when EBPs are applied in different contexts.⁶³ Researchers have begun to describe the components and principles of some EBPs and monitor fidelity to practice standards, but our review was unable to provide much insight into when and why interventions were (or were not) effective in different circumstances and settings. Similarly, it remains unclear which elements of the multi-faceted implementation strategies adopted in each initiative were critical to producing positive or negative implementation outcomes and how these may link to different service user outcomes. These findings suggest that an urgent need for more rigorous, context-sensitive evaluations, perhaps drawing on realist or other theory-based approaches,^64,65 to advance implementation science and ensure that future initiatives, especially large-scale ones,⁶⁶ actually enhance the health and recovery of people with SMI.

Finally, several limitations of our review should be noted. First, as a rapid review, methods adopted to streamline the review process may have led to missing some references for initiatives.⁸ Through alternative search strategies and contacts with experts, we aimed to minimize this risk. Second, many initiatives were not included in our review because they failed to meet our inclusion criteria, notably the requirements to describe initiatives for people with SMI or implementation activities or issues for multiple intervention types. Many lessons can be drawn from these other initiatives that remain relevant for stakeholders seeking to broaden access to psychosocial supports for people with SMI. More focused syntheses of these literatures would be highly valuable to future implementation efforts.

Conclusion

There is a growing literature on the implementation of psychosocial EBPs for people with SMI. Many insights and lessons can be drawn from this literature to guide future implementation initiatives, avoid fruitless effort and wasted resources, and promote the delivery of comprehensive, recovery-oriented mental health care.

Abbreviations

ACT

assertive community treatment

ATTC

Addiction Technology Transfer Center

BD

bipolar disorder

CBT

cognitive-behavioural therapy

CFIR

Consolidated Framework for Implementation Research

CTN

Clinical Trials Network

DH

Department of Health

EBP

evidence-based practice

EQUIP

Enhancing Quality-of-care In Psychosis

FPE

family psychoeducation

MHTS

Mental Health Treatment Study

NEBPP

National Implementing Evidence-Based Practice Project

NHS

National Health Service

NICE

National Institute for Health and Clinical Excellence

NIDA

National Institute on Drug Abuse

OTP

Optimal Treatment Project

SAMHSA

Substance Abuse and Mental Health Services Administration

SE

supported employment

SMI

severe mental illness

STEP-BD

Systematic Treatment Enhancement Program for Bipolar Disorder

VA

Veteran Affairs

VHA

Veterans Health Administration