An Assessment of Innovative Models of Peer Support Services in Behavioral Health to Reduce Preventable Acute Hospitalization and Readmissions. 1. Environmental Scan Report


In recent years, the peer support services (PSS) workforce has evolved to become an essential part of mental health and addiction treatment, family support, and primary care services. In 2003, the President's New Freedom Commission for Mental Health recommended expanded integration of PSS into the behavioral health treatment system. Interest and utilization of PSS was further bolstered by a 2007 letter to state Medicaid directors in which the U.S. Department of Health and Human Services (HHS), Centers for Medicare and Medicaid Services (CMS) identified PSS as evidence-based practice and approved coverage of PSS under Medicaid funding (CMS, 2007).

The implementation of the Affordable Care Act (ACA) and corresponding changes in the delivery of health care services generally provide additional opportunities to develop and integrate PSS across the health care system. Despite the growing number of PSS providers and the growing interest in these services, there is limited information about the impact of PSS on medical costs and, specifically, hospital admission and readmission rates.

The purpose of this study, funded by the HHS Office of the Assistant Secretary for Planning and Evaluation, is to examine innovative practices in PSS designed to bolster patient stability in the community and reduce preventable hospital admission, readmission, and emergency department (ED) utilization for behavioral health conditions. This project seeks to identify existing and emerging innovations; evidence supporting the use of these models; and options facilitating further adoption.

An environmental scan of PSS provides a review of the role of these services in health care and their evolving deployment to improve health outcomes. Information in this environmental scan was compiled through traditional literature searches, Internet searches, and key informant discussions with PSS leaders, researchers, and peer support specialists. This report is focused on answering the following research questions:

  1. What PSS approaches, models (and/or components of models), or methods of practice demonstrate the most promise toward reducing preventable psychiatric hospitalization, re-hospitalization, and ED use?

  2. To what extent are these models being utilized in the United States and at what level of the system (e.g., states, counties, cities, organizational networks)?

  3. What are the structural supports for these innovative practices -- including funding, training, and credentialing requirements -- offered through or outside the behavioral health care system?

A three-level framework for how PSS help address the research questions in this study is also presented.

Peer Support Services

Many definitions of PSS are found across published literature, state websites, and federal websites. The Pillars of Peer Support initiative has also defined the key principles and characteristics of PSS (Daniels, Bergeson, Fricks, Ashenden, and Powell, 2012). The HHS Substance Abuse and Mental Health Services Administration (SAMHSA) has been a leader within the Federal Government for promoting the use of PSS nationally. For the purposes of this study, we use the SAMHSA definition.

SAMHSA defines PSS as specialized assistance that is delivered by a person in recovery from an serious mental illnesses (SMI), substance use, or co-occurring mental and substance use condition, before, during, and after treatment to facilitate a recipient's long-term recovery in the community (Chinman, George, Dougherty, Daniels, Ghose, Swift, and Delphin-Rittmon, 2014). The goal of these services is to assist with the development of strategies to promote coping, problem-solving, and self-management of a person's behavioral health condition. This is accomplished by the peer support specialist drawing upon his/her own lived experiences and empathy to help others by promoting hope, developing skills and insights, fostering treatment engagement, accessing community supports, and building a satisfying life.

Key Definitions of Terms Used in This Report


Within the mental health and/or substance use field, this term is used to refer to someone who has experienced a behavioral health condition firsthand and is now in recovery from a mental health and/or substance use condition.

Peer Support

Peer support is a mutual form of shared interactions in which participants seek to use their personal experience to both help others and gain additional reinforcements for their own life circumstances. Peer support can occur in both individual and group settings. Usually, participants are not paid to participate in this process.

Peer Support Services

For the purpose of this report, the term "peer support services" is used to describe the intentional peer services that are delivered by people who have received training and certification to provide these services. They draw on their lived experiences in mental health, substance use, or co-occurring substance use problems. Their services are reimbursable through Medicaid or other payers, or they are employed to provide these services. In some cases these services may also be provided by a volunteer who has achieved the requisite training and certification. This review focuses on those services provided by peers in mental health, substance use, and/or co-occurring mental and substance use service settings who have received the necessary training and certification. The term "peer support specialist" is used to describe the individuals who are trained and certified and deliver these services. The training, certification, and scope of services are further described in this report. Settings in which PSS are delivered include inpatient and outpatient facilities, day-treatment programs, and community settings (Salzer, Schwenk, and Brusilovskiy, 2010).

It is important to note the key distinction between mutual peer support and PSS. PSS are intentional services that are based on the lived experience, training, and certification of the provider and are designed to promote engagement, facilitate recovery, and support resiliency. In PSS, the relationship is not reciprocal, and the skill and degree of recovery is not the same between the provider and recipient (Davidson, Chinman, Sells, and Rowe, 2006).

Peer Specialist

Peer specialists, sometimes referred to as peer support specialists, peer coaches, or peer support providers, are individuals in recovery who provide PSS to individuals seeking to achieve and/or maintain their recovery from mental or substance use disorder (SUD) or co-occurring mental and SUD. For the purpose of this study, the term "peer specialist" is used to describe this role.

Medicaid defines a qualified peer support provider as a self-identified consumer who is in recovery from mental health or substance abuse conditions and assists others with their recovery. Minimal requirements for training and certification, supervision, and care coordination have been established (CMS, 2007). Additionally, a number of states have established state criteria that must be met for peer support providers to receive reimbursement for their services.

The Role of Peer Support Services in Health Care

Peer specialists are increasingly being deployed to help those with mental health, substance use, and co-occurring substance use conditions develop and maintain recovery-based goals and resiliency. In this role they can help prevent unnecessary acute hospital admissions, avoid preventable readmissions, and lessen over-utilization of ED facilities. The evidence base for these services is emerging, and different service models are expanding.

Interest and utilization of PSS has greatly increased since the 2007 letter to Medicaid directors approving Medicaid reimbursement for PSS. Recent estimates indicate that there are at least 335 peer support organizations in the United States providing direct services (Lived Experience Research Network, 2013). Private insurers, the military health system, and the U.S. Department of Veterans Affairs (VA) have also expanded their interest in and use of PSS (Association for Behavioral Health and Wellness [ABHW], 2013; White House, 2012).

Eiken and Campbell (2008) have described three types of PSS for meeting the needs of individuals with behavioral health conditions:

  • Peers providing distinct services to support problem-solving and self-management strategies.

  • Peers with lived experiences of behavioral health conditions serving as part of a treatment team.

  • Persons with the lived experience providing services that may be other than peer support that are informed by and based in part on personal recovery experiences.

Salzer et al. (2010) conducted a national survey of peer specialists to assess their principal roles and activities. Respondents from 28 states described an average work week of about 75 percent full-time equivalency. The racial and ethnic representation from this review included 79 percent self-identified as White, 12 percent as Black/African American, and 3 percent as Latino. The majority (66 percent) were female. The most common work environments included independent peer support program (24 percent); case management (19 percent); partial hospitalization, day program, inpatient or crisis center (10 percent); vocational rehabilitation or clubhouse (8 percent); and drop-in center (7 percent). Peer support services reported as most common included peer support; encouragement of self-determination and personal responsibility; support for health and wellness; addressing hopelessness and stigma; communication with providers; illness management; and friendship and leisure activities. Less frequent activities included support services for spirituality and religion; parenting; and dating.

Chinman and colleagues (2006) examined how consumer providers address patient and treatment system factors that contribute to poor health outcomes. As described below (see Figure 1), consumer providers of PSS are able to address unmet needs within clinical systems of care.

FIGURE 1. Peer Services Providers Address Unmet Patient and Clinical Service System Factors
How Consumer Provider Services Address Patient and Treatment System Factors
Factors that contribute to poor outcomes for those with SMI
Patient Factors Treatment System Factors
Social isolation Disconnection with ongoing outpatient treatment Powerless & demoralization regarding illness Overburdened providers Fragmented services Lack emphasis on recovery-rehabilitation, empowerment
Consumer provider services address each of the factors:
Enhance social networks by:
  • role modeling
  • facilitating peer support activities
Engages patients; makes treatment more relevant through collaboration Activates patients; teach coping & street smarts; provide hope through role modeling Supplement existing treatment; increase access Provide case management/ system navigation to increase access Emphasize recovery:
  • liaison between consumer & system
  • advocate for community integration over symptom stabilization

The American Mental Health Counselors Association (ACMHA) Peer Services Toolkit (Hendry, Hill, and Rosenthal, 2014) reviews the range of roles and services provided by peer providers and how this workforce applies the principles of peer support in their work. The toolkit states:

"Before, during and beyond crisis points they (PSs) provide compassionate listening, and a positive vision of the future. Additionally peer providers can work with individuals in goal setting, and developing achievable action plans. They can play an important role in supporting people in self-managing and working towards whole health goals, and they are uniquely qualified to assist peers in connecting with their communities, building supportive relationships, accessing formal and informal resources, and working with cultural humility to support people across a wide range of cultural differences."

These functions are important across the full continuum of health services.

Peer Support Services as Part of the Larger Health Care Service Systems

Reimbursement for Peer Support Services

A major step in the evolution of PSS occurred in 2007 when Medicaid designated them as evidence-based and reimbursable by states that include them in their state plans (CMS, 2007; Daniels, Cate, Bergeson, Forquer, Niewenhous, and Epps, 2013). Currently more than 30 states are actively providing reimbursement for these services (Kaufman, Brooks, Bellinger, Steinley-Bumgarner, and Stevens Manser, 2014). The health home option established under the ACA in 2010 provides guidance on the core clinical features of organizations that provide PSS and the clinical service requirements for PSS (SAMHSA-HRSA 2012). In addition, the President's Executive Order (White House, 2012) has called for the training of 800 new peer specialists to serve the VA health systems.

In addition to the growing number of states that now allow Medicaid reimbursement for PSS, a substantial portion of PSS is reimbursed through state grants and contracts, federal grants, and service contracts with managed care organizations. Historically, these non-Medicaid funding sources have paid for peer specialists' involvement in services and programs such as psychiatric rehabilitation; drop-in centers; employment services; housing services; crisis and respite services; and young adult transition services (Hendry et al., 2014). Indeed, some of the pioneers of PSS are thriving in states where services are not Medicaid reimbursable and have always relied on non-Medicaid funding for reimbursement (e.g., California and New York).

A recent survey of peer support organizations providing direct services in the United States found that 77 percent of their funding come from governmental sources, with 33 percent reporting they receive federal funding, 8 percent are reimbursed by Medicaid, 61 percent receive state funding, and 45 percent receive funding from county/local governments (Lived Experience Research Network, 2013).

ABHW is an association representing the major behavioral health care management companies. ABHW notes that in specialty behavioral health care organizations, the use of PSS is most prevalent in Medicaid and other public sector care (ABHW, 2013). A variety of types of PSS are offered by ABHW member companies and include peer bridgers, whole health peer coaches, addiction recovery coaches, family peer navigators, family peer coaches, peer warm lines, and other community support programs. Core challenges noted for specialty behavioral health care organizations in developing these programs include the variable training and certification of peer specialists, the unstructured definitions of service types, and the capacity of peer service organizations to meet billing requirements. These organizations also struggle with their claims payment systems' capacity to adjudicate new PSS service codes.

ABHW has also identified a lack of national standards for PSS as a significant barrier to expanding their use in specialty behavioral health care organizations. ABHW notes that it is hard to assess core competencies among peer specialists, and national credentialing standards are difficult to establish due to variations across states. ABHW also identifies consistent billing and reimbursement policies as an impediment to broader use of this workforce (ABHW, 2013).

Coverage of Peer Support Services

Coverage for PSS has varied across different payer types. Medicaid has recognized PSS as a reimbursable service under state plans. Common billing codes used to bill for these services include Healthcare Common Procedure Coding System codes for peer services (H0038), psychiatric rehabilitation (H2017), community support (H2015), and Assertive Community Treatment (ACT) (H0039). Reimbursement is generally provided in 15-minute increments of service. Many states are building these services into their standard Medicaid coverage. However, they have not been included in the essential benefits that have been established for health exchange plans under the ACA. Nor are they generally included in the certificate of coverage for most commercial plans. Some specialty behavioral health care organizations have elected to cover these services as administrative costs for these plans.

To standardize the coverage of PSS in Medicaid, Medicare, and commercial plans, a set of level of care criteria for these services has been developed (Daniels et al., 2013). Standard practices for health insurers include the use of medical necessity and level of care criteria for determining the needed services at the appropriate level of intensity. These are evidence-based criteria that guide decision-making for necessity and payment.

Optum has developed four sets of level of care guidelines that address peer-to-peer services, peer bridger services, family-to-family support services, and family navigator services (Daniels et al., 2013). Through the establishment of these criteria it is possible to standardize the review of services and ensure they meet quality guidelines. Other specialty behavioral health care organizations have also developed level of care guidelines that are used in specific coverage contracts.

When contracting for peer-run services, managed care organizations consider three key factors: (1) ensuring high quality of services that are compliant with state and federal requirements; (2) ensuring services achieve positive, measurable results; and (3) supporting the principle of health care affordability, which calls for a cost-effective approach to services (Hendry et al., 2014).

Peer Support Services as an Evidence-Based Service

While the prevalence and availability of PSS continues to expand, the research literature yields few experimental trials of their effectiveness (Repper and Carter, 2011). This is due in part to the varying definitions of PSS (O'Hagan, Cyr, McKee, and Priest, 2010); a lack of established measures; variable range of training and certification of the workforce; unstructured service and intervention models; a lack of established outcome measures; and the lack of a uniform model or typology of peer-delivered services (Rogers, Kash-MacDonald, and Brucker, 2009). Past research has focused primarily on peer-delivered services as both individualized services models and team-based care approaches and not subjected to comparative analysis. Another limitation is that evaluation designs have typically featured pre-post measures without clearly established control groups or longitudinal follow-up.

The research literature includes four systematic reviews of PSS (Simpson and House, 2002; Doughty and Tse, 2005; Rogers et al., 2009; Repper and Carter, 2011). The reviews in these reports are constrained by the lack of consistent outcome measures and the fact that there is no widely accepted model or typology of peer-delivered services (Rogers et al., 2009), although most of the studies focus on PSS in mental health rather than addictions.

The Simpson and House (2002) review of PSS studies found that these services led to greater levels of satisfaction with personal circumstances among those receiving services and a decrease in hospitalizations. They concluded that "users [of behavioral health services] can be involved as employees, trainers, or researchers without detrimental effect." Further, they found that "involving users with severe mental disorders in the delivery and evaluation of services is feasible."

Doughty and Tse (2005) concluded that the research on peer-delivered services yielded positive outcomes for clients. They note that the research is limited by the settings in which services are delivered, and it is often difficult to differentiate outcomes between traditional and peer-delivered services.

In their review of the literature on PSS, Rogers et al. (2009) found that there were ten commonly used outcome measures to assess these services:

  1. Quality of life
  2. Recovery attitudes
  3. Perceptions of empowerment
  4. Self-confidence
  5. Self-esteem
  6. Hospitalization
  7. Relapse
  8. Psychiatric symptoms
  9. Criminal justice involvement
  10. Employment


They conclude that the results of research on peer services "remain equivocal." They note that there are promising results and that better research is needed to understand the unique contributions of peer services to the overall outcomes of care.

Repper and Carter (2011) reviewed studies that examined PSS for their effectiveness of peer support, benefits to consumers, empowerment, social support and social functioning, empathy and acceptance, reducing stigma, and hope. They concluded that "although scarce in the literature, the few experimental trials show that at the very least, peer support workers (PSWs) do not make any difference to mental health outcomes of people using services. When a broader range of studies are taken into account, the benefits of PSW become more apparent." They also examined the range of services provided by PSS and the impact of these on the recovery process. They found that "what PSWs appear to be able to do more successfully than professionally qualified staff is promote hope and belief in the possibility of recovery, empowerment and increased self-esteem, self-efficacy and self-management of difficulties and social inclusion, engagement and increased social networks. It is just these outcomes that people with lived experience have associated with their own recovery; indeed these have been proposed as the central tenets of recovery: hope, control/agency and opportunity." Indeed, many argue that these tenets are the most significant driving factor of PSS. Measurement of these factors is among the most common outcomes reported in the published literature. It could be argued that increasing hope, control/agency, and opportunity are the underlying mechanisms for clients' overall health.

Limited Research Findings

Despite the methodological challenges, there are significant examples of how innovative models of PSS are being used to help reduce inpatient admissions, limit readmissions, and control unnecessary ED utilization. A recent evidence-based review of PSS (Chinman et al., 2014) found that "[t]he level of evidence for each type of peer support service was moderate. Many studies had methodological shortcomings, and outcome measures varied. The effectiveness varied by service type. Across the range of methodological rigor, a majority of studies of two service types -- peers added and peers delivering curricula -- showed some improvement favoring peers. Compared with professional staff, peers were better able to reduce inpatient use and improve a range of recovery outcomes, although one study found a negative impact. Effectiveness of peers in existing clinical roles was mixed. PSS have demonstrated many notable outcomes. However, studies that better differentiate the contributions of the peer role and are conducted with greater specificity, consistency, and rigor would strengthen the evidence."

The evidence review by Chinman et al. (2014) summarizes the findings separately across each of the three methodological approaches that are directly relevant to the present study. These are randomized controlled trial (RCT) studies of PSS; quasi-experimental studies of PSS; and correlational or descriptive studies of PSS. An overview of the studies, interventions, outcomes measured, findings, and evidence rating review are described in Tables 1-3 below and inform the evidence base for the present study. Of note, since this evidence-based review was conducted on a set of inclusion criteria, some of the studies referenced in this report for the present study may not be included in this review.

TABLE 1. RCTs of PSS with Mental Health and Co-occurring Mental Health and Substance Abuse Service Recipients that have Measured Health Care Utilization Outcomes
Study and Sample Description Intervention Outcomes Measured Major Study Findings
Solomon & Draine, 1995 (Also see Solomon, Draine, & Delaney, 1995)
N=96 individuals with SMI in community mental health center at risk for hospitalization
Participants assigned to case management team of peers vs. case management team of non-peers. Therapeutic alliance, income, social network size, days hospitalized, psychiatric symptoms, attitudes toward medication compliance, quality of life, interpersonal contact, social functioning, treatment satisfaction. There were no significant differences between treatment and control group on measured outcomes 2 years after initiation of services.
Clarke et al., 2000
N=163 adults with chronic SMI
Participants assigned to usual care vs. ACT non-peer vs. ACT with peers. Percentage of participants hospitalized and number of days to hospitalization; time to first ED visit, arrest, homelessness. Time to first hospitalization was earlier for ACT non-peer group than the ACT with the peer group. There were no significant differences between these groups for the first instance of homelessness, first arrest, or first ED visits. More participants in the ACT non-peer group had hospitalizations and ED visits than those in the ACT with peer group.
Sells et al., 2006
N=137 adults with SMI; 70% had co-occurring SUD
Participants assigned to ACT alone vs. ACT plus peer-delivered case management. Therapeutic relationship, frequency and severity of substance use, utilization of various outpatient and day-treatment services, and treatment engagement. Participants with peers reported better therapeutic relationship than controls at the 6-month follow-up. Those who were least engaged with peers had more provider contact than the control group. The therapeutic relationship at 6 months predicted treatment engagement and service utilization at 12 months, but there were no between-group differences.
Druss et al., 2010
N=80 individuals with SMI and chronic medical illness
Peers delivering curricula: HARP program versus usual care. Patient activation, primary care visits, physical activity, medication adherence, and health-related quality of life Participants in HARP had higher patient activation and higher rates of primary care visits 6 months after intervention than those with usual care. Medication adherence, physical health quality of life, and physical activity did not differ between groups.
Sledge et al., 2011
N=74 patients with major mental illness hospitalized 3 or more times in past 18 months
Participants assigned to usual care vs. peer mentor plus usual care. Number of hospitalizations and hospital days Participants with peers had significantly fewer admissions and fewer hospital days than those in usual care at the 9-month follow-up.
NOTES: Articles are listed in chronological order. Various threats to both internal and external validity were considered in each study's rating of "limited" (study had several methodological limitations) or "adequate" (study had few or minor methodological limitations). Multiple publications based on the same RCT are described in the same row.


TABLE 2. Quasi-Experimental Studies of PSS Mental Health and Co-occurring Mental Health and Substance Abuse Service Recipients that have Measured Health Care Utilization Outcomes
Study and Sample Description Intervention Outcomes Measured Major Study Findings
Felton et al., 1995
N=104 participants with SMI
Case management teams vs. case management teams plus non-peer assistants vs. case management teams plus peer specialists. Self-image and outlook, treatment engagement, social support, quality of life, life problems, housing instability, income, and family contact. Clients on case management teams plus peer specialists reported gains in quality of life indicators, reductions in some major life problems, and more treatment engagement compared with those in the other two groups over the course of the 2-year study. There were no differences in outcomes between teams with non-peer assistants and those with standard care.
Klein et al., 1998
N=61 participants with co-occurring mental and SUD
Intensive case management teams with peers vs. case management teams without peers. Crisis events (e.g., ED visits), number of hospital days, social functioning, use of community resources and social integration, and quality of life. Participants with peers had fewer inpatient days, better social functioning, and some improvements in quality of life indicators at the end of the intervention.
Chinman et al., 2001
N=158 participants with SMI
PSS added to standard care vs. a matched control group in standard care. Number of hospitalizations and hospital days. There were no significant differences between groups in outcomes 6 months after the service start date.
Min et al., 2007
N=556 participants with SMI and SUD with history of hospitalization
Teams with case management vs. teams with case management plus peer worker. Days to first hospitalization and percentage hospitalized over 3-year period. Participants on teams with peers had more time in the community and less inpatient use.
Schmidt et al., 2008
N=142 participants with SMI and with a recent hospitalization
Case management team vs. case management team plus peer. Client contact, percentage with crisis center visits, percentage hospitalized, number of hospitalizations and hospital days, outpatient mental health service utilization, medication use, substance abuse, and housing stability. There were no significant differences between groups in outcomes measured at the 12-month follow-up.
van Vugt et al., 2012
N=530 participants with SMI in 20 ACT teams
ACT teams without peers vs. ACT teams with peers. Level of functioning, met and unmet needs, working alliance, number of hospital days, and number of homeless days. At 1-year and 2-year follow-ups, clients of teams with peers had better psychiatric and social functioning, improvements in met and unmet needs related to their personal recovery, and fewer homeless days than clients of teams without peers. Peer presence was associated with an increased number of hospital days.
NOTES: Articles are listed in chronological order. Various threats to both internal and external validity were considered in each study's rating of "limited" (study had several methodological limitations) or "adequate" (study had few or minor methodological limitations).


TABLE 3. Correlational or Descriptive Studies of PSS Mental Health and Co-occurring Mental Health and Substance Abuse Service Recipients that have Measured Health Care Utilization Outcomes
Study and Sample Description Intervention Outcomes Measured Major Study Findings
Chinman et al., 2000
N=1,203 homeless participants with SMI
Participants with homeless outreach teams alone vs. those with homeless outreach teams with peers. Quality of life, homelessness days, social support, symptoms and mental health problems, alcohol and drug problems, and days worked. There were no significant differences between groups on outcomes over a 12-month period.
Landers et al., 2011
N=35,668 participants with a reimbursed community mental health service
Participants without a PSS claim in past year vs. those with a PSS claim in past year. Percentage with a hospitalization or crisis stabilization. Compared with participants without peers, more participants with peers used crisis services, but fewer had a hospitalization.
NOTES: Articles are listed in chronological order. Various threats to both internal and external validity were considered in each study's rating of "limited" (study had several methodological limitations) or "adequate" (study had few or minor methodological limitations).

State-Level Training and Certification for Peer Specialists

As of 2014, 37 states have established training and certification program for PSS (Kaufman et al., 2014). There is a wide variation of training and certification standards. A range of educational models have been used by the organizations that train this workforce. Some principal training programs include Depression and Bipolar Support Alliance (DBSA), International Association of Peer Specialists (iNAPS), Recovery Innovations (RI), and Appalachian Consulting Group (ACG). The DBSA program has been used to train and certify almost 500 new peer specialists for the VA health system. Certification is generally established at the state level, is testing based, and administered by national training organizations, local academic institutions, and others.

The ACMHA Peer Services Toolkit (Hendry et al., 2014) identifies a core set of individual qualities required for the peer specialist workforce. These include the following: person has progressed in his/her own recovery or has 1 year of addiction recovery and is actively involved in recovery activities; willingness to self-identify; willingness to share knowledge and experience of recovery; exhibits signs of a spiritual awakening; can act as a role model; listens and learns from people served; creates environments that promote recovery; works in partnership with the individual; promotes trauma-informed care (e.g., asking "what happened," not "what's wrong"); helps to navigate the system; helps individuals to examine personal goals and define in achievable ways; motivates change desired by the individual; and may act as liaison or proxy for the individual if desired.

An analysis of the survey findings of Kaufman et al. (2014) provides a comparison of the types of training and certification programs offered by different states. It is important to note that not all states are included in this survey, as some states have yet to develop formal training and certification programs. This does not mean that those states are lacking in PSS programs, and findings from the informant interviews suggest that some of the states without established requirements still have robust and model programs (examples include New York and California).

As of 2014, there is a range of different certification, training, and billing arrangements adopted by the states (Kaufman, 2014). As an example, characteristics of state requirements for PSS and the workforce that delivers them include the following:

  • Thirty-seven states have implemented statewide uniform PSS certification programs, and seven states are actively developing them.

  • In 19 states there is a state agency that is the program administrator or credentialing agency, and in six states this is provided by a combined state and external organization.

  • Sixteen states require up to 40 hours of training, 12 states require between 41 and 80 hours, and three states require more than 80 hours.

  • Thirty-five states have established requirements for the completion of a certification exam.

  • Twenty-nine states have reported that they have requirements for continuing education for recertification, and six states do not. Of those states reporting continuing education requirements, six have no requirements, seven require between 10 and 20 hours, and 14 require between 21 and 40 hours. Ten states did not list requirements. In general, states that require continuing education do so on a 2-year cycle.

  • Thirty of the states that have implemented a standard training and certification program for PSS also have Medicaid billable services.

  • Fifteen states cover the training costs for PSS, and an additional four share the cost.

  • Five competency areas are reported as common in the training programs for PSS. These include advocacy (15 states); professional responsibility (18 states); mentoring (11 states); recovery support (24 states); and cultural competency (13 states). Twelve states did not report cultural competency information.

The ACMHA Peer Services Toolkit (Hendry et al., 2014) cites 21 common elements in peer specialist training programs. These are as follows:

  1. The history of the peer movement;
  2. Insight into personal recovery;
  3. Five stages of recovery;
  4. Role of peer support;
  5. Creating program environments that promote recovery;
  6. Stages of change/the dynamics of change;
  7. Effective goal setting that promotes successful change;
  8. Facilitating support groups that promote recovery;
  9. Effective listening;
  10. Motivational interviewing;
  11. Facing one's fears;
  12. Combatting negative self talk;
  13. Problem-solving with individuals;
  14. Peer specialist ethics and boundaries;
  15. Power, conflict and integrity in the workplace;
  16. Creating the life one wants;
  17. Wellness recovery action plans (WRAP);
  18. Understanding the impact of trauma;
  19. Working towards shared responsibility;
  20. Looking at crisis as an opportunity; and
  21. Personal sharing and disclosure.


Cultural diversity and competency are also described as cross-cutting themes in many of these areas.

Peer Support Services and Health Care Utilization and Outcomes

The focus of this study is to investigate PSS that address the needs of individuals with mental health, substance use, or co-occurring mental and substance use conditions. As background for this study, a series of discussions was conducted with key informants who were identified as leading experts in the field (a complete list is included in Appendix A). The study typology of services outlined earlier in the report, including crisis and respite services, transitions in levels of care, and community-based services to promote recovery and resiliency, was used as a framework to discuss key activities for PSS and how these services help reduce unnecessary inpatient admissions and readmissions and utilization of emergency services. An overview of findings is presented in Figure 2 below.

In addition to the variables identified through our key informant interviews that impact ED and hospitalization utilization, there are a number of additional outcomes that could be considered. These outcomes will be determined, in part by the sites. For example, the ACMHA Peer Services Toolkit (Hendry et al., 2014) identifies a range of useful measurements to assess the successful individual-level outcomes of peer services. These include the following:

  • Personal Outcome Measures (Council on Quality and Leadership, 2012).

  • Recovery-Oriented Systems Indicators (ROSI) (Dumont, Ridgway, Onken, Dornan, & Ralph, 2005).

  • Community integration and measuring participation (Salzer and Baron, 2006).

  • Community Participation as a Predictor of Recovery-Oriented Outcomes Among Emerging and Mature Adults with Mental Illness (Kaplan, Salzer, and Brusilovskiy, 2012).

FIGURE 2. Overview of the Impact of PSS on Hospitalization and ED Utilization
Study Typology of PSS   Definition   How They Impact Hospitalization/ ED Rates   Key Factors in PSS That Mediate Outcomes   Desired Outcomes
Crisis and respite services Programs and services that provide an acute response to individuals who are experiencing a psychiatric emergency and need an urgent response. Provides alternatives to hospitalization and ED use. Fosters stability and community tenure.   Peers are employed to provide services.

Training and certification standards exist.

Services are reimbursable.

Services are covered under Medicaid state plans.

Peer specialists are integrated into the health care system.

Community has other supportive resources and services to support clients.

Peers are part of health care team and provide input into medical records.

Track record of success in the community.

Supported by other providers.

Peers focus on whole health.

  Reduction in ED, hospitalization, and inpatient use.

Meeting PSS program-specific goals.

Transition in levels of care Programs and services designed to provide assistance and support to individuals who are involved in changes to their treatment services that involve new providers or settings and levels of acuity. Helps reduce/ prevent crisis, crisis relapse, hospital readmission, ED use.  
Community-based services to promote recovery and resiliency Programs and services designed to provide ongoing engagement, support, and activation for those who have successfully established recovery and illness management plans. Keeps individuals healthy in the community and helps prevent hospitalization.  

Systems-level outcomes cited include the following:

  • Re-hospitalization rates compared to individuals not receiving peer support;
  • Changes in engagement rates for people in traditional services;
  • Number of outpatient services accessed;
  • Overall satisfaction with services;
  • Length of time people remain in traditional services; and
  • Improvement in quality of life and other wellness measures.


A Proposed Typology of Peer Support Services for the Reduction of Preventable Hospitalizations and Emergency Department Utilization

The focus of this study is to investigate and better understand how PSS address the needs of individuals with mental health or co-occurring mental and substance use conditions, and how these services help mitigate unnecessary psychiatric inpatient and ED use. While the principal funding support for PSS has historically occurred in publicly funded health systems, this study seeks to better understand the applicability across all payer systems. To better define this scope of services, we propose the following model, or typology, as displayed in Table 4.

TABLE 4. PSS Models of Care that are Likely to Impact Hospitalization and Health Care Costs
PSS Categories Definitions
Crisis and respite services These are programs and services that provide an acute response to individuals who are experiencing a psychiatric emergency and need an urgent response with the goal of fostering stability, thus reducing the need and use of psychiatric hospitalization or ED use, and supporting community tenure.
Transition in levels of care These programs and services are designed to provide assistance and support to individuals who are involved in changes in their treatment services that involve a shift from 1 provider or setting to another, as well as levels of the acuity of care.
Community-based to promote recovery and resiliency These programs are designed to provide ongoing engagement, support, and activation for those who are establishing and building community tenure, and who are in the process of recovery and illness management.

The programs across these categories may share common acute and routine services and recovery goals. While most PSS programs will be able to classify their program into one of these three categories, some may deliver a more comprehensive array that cuts across all three categories. For the current study, and from a health care systems perspective, we adopt this typology to highlight program characteristics, service models, and outcomes measures that are relevant to health care costs and specifically hospital admission and readmission rates.

Crisis Services and Alternatives to Acute Hospitalization


Being admitted to psychiatric hospital-based care is expensive and disruptive to both individuals with behavioral health conditions and their families. Having repeated admissions to psychiatric hospital care is a common and substantial problem. Preventing psychiatric readmissions requires the provision of short-term alternatives for individuals who are "not at significant risk of harm to self or others and ongoing community-based treatment services and supports" (Gaynes et al., 2015). Research on the challenges of psychiatric readmissions focuses on three levels of intervention: short-term alternatives to re-hospitalization; transition support services; and long-term approaches for reducing the needs for re-hospitalization (Gaynes et al., 2015).

Peer support specialists have actively worked to develop, operate, and provide services in a range of programs that provide alternatives to psychiatric hospitalization. Crisis respite programs provide a safe and homelike environment to support people through an episode of crisis (Ostrow and Fisher, 2011). Common principles include an environment that is safe and establishes acceptance through connections; hope is held by others when one may not be able to hold it for himself/herself; everyday language is used to describe experiences; self-care and personal responsibilities are a central focus; and gaining a sense of mastery and power over one's life is encouraged (Ostrow and Fisher, 2011). The goal of peer-run crisis respite is to encourage less dependence on formal mental health systems of care and the associated trauma that commonly occurs in EDs and inpatient psychiatric EDs. By bolstering an individual's stability in a time of crisis, it is possible to support resiliency and prevent unnecessary ED visits and hospital admissions.

One randomized controlled study of crisis respite care found that the average rate of symptom improvement was greater in this alternate care than in the hospital comparison. Recipients of care in these settings also demonstrated greater satisfaction. The average savings for respite care was more than $450 per day (Greenfield, Stoneking, Humphreys, Sundby, and Bond, 2008).

Respite care programs provide a spectrum of services and can accommodate a range of participants. A review of innovations in respite programs (Ostrow and Fisher, 2011) cites 12 examples in the United States. These programs have federal, state, and county grant funding and serve between two and eight people in their residential services at any given time. Their average length of stay is between 1 and 14 days. Many programs also include other services such as warm line crisis services, drop-in accommodations, and rehabilitation services such as housing and vocational care.

In Pierce County, Washington, the Regional Support Network (RSN) has helped develop a "living room crisis model." This approach is more welcoming than a traditional ED setting, and services are provided by both a peer support staff and consulting clinicians. This program has contributed to a reduction in admissions for psychiatric care of 32.3 percent and reduced readmissions of 26.5 percent over 3 years. It has also reduced the average number of inpatient days per thousand from 19.6 in 2009 to 13.7 in 2013 (Optum, 2014a).

A peer-run respite care program has also been developed by Project Hope Peer Support Network in Long Beach, California. The Hacienda of Hope is designed to provide crisis support services 24 hours/day, 7 days/week for those who do not require immediate on-site medical treatment (Project Return Peer Support Network, 2014). Most people who engage in this program stay for 1-3 days, and there is a maximum stay of 14 days. WRAP and the Eight Dimensions of Wellness programs are provided by the peer staff.

Peer Programs to Support Level of Care Transitions

Transitions between inpatient and outpatient levels of care can be difficult and require both careful care planning and ongoing support for an individual's re-integration into community settings. A number of service models have been developed to support these level of care transitions and are sometimes described as peer bridger programs. Arguably, the first such program was developed in 1994 when the New York Office of Mental Health (OMH) approved a pilot program for the New York Association for Psychiatric Rehabilitation Services (NYAPRS) to assist individuals with long or repeated psychiatric hospital stays make successful transitions to their home communities ( The primary role of the NYAPRS peer bridger program is to help individuals who are admitted to facility-based psychiatric care establish a trusting and engaged relationship that provides them a peer role model, mentor, teacher, advocate, and ally to facilitate successful transition to their home communities and promote long-term tenure.

The NYAPRS peer bridger program offers a training curriculum that focuses on four key components, which are outreach and engagement, crisis stabilization, wellness and self-management skills, and community support. Initial data from this program demonstrated that for individuals served in 1998, there was a 71 percent reduction in re-hospitalization. Recent data supports continued positive outcomes for this program and increased contracting with managed care organizations for these services (Hendry et al., 2014).

Other consumer-led organizations are also actively developing and providing level of care and peer bridger transition services. However, these interventions are often not a fully structured or manualized service model. Therefore, there is no way to ensure the fidelity of program designs being implemented across different programs, the training requirements and core competencies of their peer providers, or the outcomes measured, if any.

A peer navigation model of intervention called The Bridge was tested against a treatment as usual group in a randomized trial (Kelly, Fulginiti, Pahwa, Tallen, Duan, and Brekke, 2013). The Bridge model of intervention is described as a comprehensive engagement and self-management model whereby participants are taught to access and manage their health care effectively. It is a manualized approach with four components: assessment and planning, coordinated linkages, consumer education, and cognitive-behavioral strategies to support health care utilization behavior change and maintenance. Findings of the study supported changes in seeking care from a primary care provider rather than the ED, and reduced physical health symptoms.

Optum has been a leader among managed behavioral health care organizations in the deployment and reporting of outcomes for their PSS programs. In Wisconsin and New York, Optum reports utilizing a peer bridger-based service model that supports community re-engagement after hospitalization. The Wisconsin program was delivered by the Grassroots Empowerment Project (GEP), a peer-run organization, and in New York the program was run by NYAPRS. Findings from this Optum program include reductions by 30 percent in inpatient days utilized and health costs savings of 24 percent in Wisconsin and New York. In New York, these programs resulted in a reduction of inpatient days by 63 percent and overall behavioral cost savings to the plan of 47 percent (Optum, 2014b). It is difficult to assess this self-report data, and it is also not clear if there is fidelity between the GEP and NYAPRS peer bridger programs. A rigorous evaluation of this type of program would require a structured evaluation of the components and outcome measures, and that was not done in this case.

Community-Based Peer Support Services Programs to Promote Recovery Supports

A number of community-based service programs have been developed that deploy peer specialists to deliver PSS in a variety of roles. Some are led by national organizations or by community-based consumer-run organizations, and others are built into formal clinical service systems. The goals of these services include ongoing community-based support for recovery, improved community engagement and tenure, and sustained resiliency.

Some of the major national organizations include the following:

  • Mental Health America ( provides a range of advocacy, screening and prevention, and community-based service supports. These are based on more than 100 years of service and offer a range of national community-based peer support programs.

  • DBSA ( offers both training and certification for peer support specialists and a national network of peer support groups and other wellness resources. DBSA recognizes an approach that says, "We've been there, we can help."

  • The National Alliance on Mental Illness ( offers programs for families and individuals that include screening, educational resources, advocacy, and community-based programs.

  • RI ( has developed a nationally recognized peer support specialist training program that is designed around the principles of hope, empowerment, wellness, personal responsibility, a community focus, and connectedness.

The SAMHSA-HRSA Center for Integrated Health Solutions has also developed a manualized PSS training program that supports the integration of both physical and behavioral health. Whole Health Action Management (WHAM) is a structured program that includes both advanced training for peer support specialists and a workbook for improved health management.

Community-based recovery support programs target a range of goals and promote services to foster resiliency and promote community tenure. This range of services focuses on an individual's need for housing, employment, and other psycho-social resources. Outcomes for these community-based services are generally service type and are program-specific. For example, the ACMHA Peer Services Toolkit (Hendry et al., 2014) reports that "Mental Health Peer Connection's Life Coaches helped 53 percent of individuals with employment goals to successfully return to work in the Buffalo, NY area, 2010 program evaluation data; Western NY's Housing Options Made Easy helped 70 percent of residents to successfully stay out of hospital in the following year, 2011 program evaluation data."

Conclusions -- Environmental Scan

An emerging evidence base has been developed that supports the findings that PSS can be effective in promoting recovery-based outcomes for those with behavioral health conditions. There are a number of inherent challenges for evaluating the impact of these services on health outcomes. These include the different training and certification standards for providers across states, varying levels of coverage and reimbursement for these services, and a lack of a consistent service models. However, both the research literature reviewed and discussions with key leaders in the field support the evolving role of this workforce.

There has been a less specific research focus on how these services are able to affect the health outcomes of inpatient hospital and ED utilization. For the purpose of this study, a three-tiered typology of services has been developed to specifically assess the different models of care. These services include crisis/respite care, level of care transitions, and ongoing community-based recovery supports.

The site visit protocols and site selection will be influenced by the findings of this environmental scan.

View full report


"PeerSupServ.pdf" (pdf, 1.24Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®