Certified Peer Specialist Roles and Activities: Results From a National Survey

A brief survey was developed with input from a CPS. Participants were asked to identify the type of program in which they work and the percentage of their work time (up to 100%) they spend in activities at the agency or program site or in the community. They were then asked to estimate the percentage of time that they generally spend supporting people in groups or individually and working with families and with community members, employers, or others. Finally, respondents were asked to indicate on a scale from 1 (never) to 5 (always) how often they supported their peers in a number of specific areas.

Inclusion for this study was limited to CPSs who were currently working in a paid CPS position in the United States. The survey instrument was entered into Survey Monkey, an online survey vendor, which enabled persons with Internet access to complete it. Five respondents printed the survey and submitted it by postal mail. A written announcement about the survey was displayed on numerous Web sites that target CPSs and was distributed on CPS and related electronic mailing lists. The survey was available online from June to November 2008; a total of 336 individuals responded in that period. Of these, 45 were excluded for any of the following reasons: they were not currently employed as a CPS, they were working as a CPS on only a voluntary basis, they were not living or working in the United States, or they did not answer questions about their work activities. The final study sample was 291. The survey was deemed to be exempt by the authors' institutional review board, and respondents read an informed consent statement and agreed that they understood it before completing the survey.

The final sample of 291 CPSs included individuals from 28 states. Of the 260 CPSs who responded to a question about their gender, 85 (33%) indicated that their gender was male, 172 (66%) responded that they were female, and three (1%) indicated that they were transgender. Of the 251 participants who disclosed their racial-ethnic identification, 198 (79%) indicated that they were white or Caucasian; 29 (12%) black or African American; eight (3%) Hispanic or Latino; three (1%) Asian, Pacific Islander, or Hawaiian Native; three (1%) Native American; and ten (4%) multiracial. The average respondent had been employed as a CPS for a mean±SD of 23.8±22.8 months and had worked 29.6±11.5 hours per week. Of the 106 respondents who indicated "other" for the type of program they worked in, 102 provided a brief description of their program. Twenty-four of these were recoded into similar program categories already provided, and 33 were coded into four new categories: residential, education and advocacy, crisis services, and psychiatric rehabilitation. An "other" category was developed for those whose responses could not be coded and for when there were fewer than five respondents in the category.

Overall, CPSs spent almost twice as much time on site as in the community (59% compared with 33%) ( Table 1 ). However, this varied greatly by program. Only those in case management programs spent more time in the community than at the program site. Those in more site-based programs, such as partial hospitalization programs, clubhouses, or drop-in centers, as well as those involved in education and advocacy, spent 65% or more of their time at the agency and less than 20% in the community. CPSs devoted almost twice as much time supporting peers individually as in groups (48% versus 25%), although there was some variability across programs. Those working in case management and residential programs provided a significant amount of individualized support, followed by CPSs in stand-alone peer support programs and employment-oriented programs (vocational rehabilitation and clubhouse). CPSs in the other programs provided nearly equal amounts of group and individual supports. CPSs across all programs spent relatively little time working with families or other community members compared with the time spent supporting their peers individually and in groups.

Peer support was by far the most prevalent activity provided by CPSs (4.48±.77 on a scale of 1, never, to 5, always), followed by encouragement of self-determination and personal responsibility (4.26±.88). Other prevalent supports provided were health and wellness (3.87±.93), addressing hopelessness (3.84±1.08), communication with providers (3.68±.99), illness management (3.62±1.13), addressing stigma in the community (3.56±1.13), developing friendships (3.51±1.05), leisure and recreation (3.25±1.14), education (3.16±1.12), transportation (3.06±1.28), and developing wellness recovery action plans (3.04±1.31). Less central supports, with means below 3 (indicating "sometimes"), included family relationships (2.95±1.11), employment (2.94±1.06), citizenship (2.83±1.15), spirituality and religion (2.74±1.12), developing psychiatric advance directives (2.27±1.18), parenting (2.14±1.15), and dating (1.74±.98). A factor analysis of these supports was conducted, which resulted in a stable five-factor solution (6). No consistent differences in activities were seen across programs.

Overall, our findings indicate that CPSs spent almost twice as much time at their agency site as in the community and almost 50% of their time working with peers individually. One concern was that agencies hiring CPSs would view them as capable of providing supports only in groups, given the strong self-help and mutual-aid group history of peer support. Relatively little time was spent working with family members or those in the community who could provide key "natural supports" outside the mental health system. We also identified great variability overall in CPS activities. More frequent activities included the provision of peer support, understood as a sharing of personal experiences and provision of mutual aid, encouragement of self-determination and personal responsibility; a focus on health and wellness; addressing hopelessness; assistance in communications with providers; education about illness management; and combating stigma in the community. These activities are all directly related to the recovery principles from the Substance Abuse and Mental Health Services Administration ( 7 ). Other activities were not addressed as often—employment, citizenship, spirituality and religion, developing psychiatric advance directives, parenting, and dating. These are areas in which CPSs are uniquely knowledgeable and qualified because of their own personal experience and understanding of the importance of these areas to recovery.

Although not definitive because small samples in certain programs prevented statistical analysis, our results nonetheless appear to support the influence of settings on CPS activities. Given that CPSs spent almost twice as much time in their agencies as in the community, this agency-based effort was even more extreme for those who work in partial hospitalization, inpatient, or crisis settings; vocational rehabilitation centers or clubhouses; drop-in centers; or education and advocacy programs. CPSs in case management programs were the only ones who spent more time working in the community than at their agency. Intuitively, these results are consistent with what might be expected for all providers who work in these programs. Moreover, although most CPSs' efforts are spent working with individuals, those working in residential programs—and case management programs to a certain extent—stand out in this regard. CPSs working in drop-in centers were the only ones to provide more supports in groups than individually. No patterns were detected in working with family members and those from the larger community because of the limited amount of activity with these groups. One clear conclusion is that CPS activities, like other disciplines in the behavioral health workforce, are assorted and vary greatly by the setting in which the CPS works. In other words, a CPS is not a CPS is not a CPS.

Study limitations include our inability to verify that respondents were indeed CPSs. We are also unable to claim that the sample is necessarily representative of CPSs as a group, and the study may include unknown biases. Furthermore, we reasonably presumed that most information submitted was accurate but still vulnerable to the threat of a respondent's mischievous behavior or deception. The list of activities was based on input from CPSs and good knowledge of CPS activities but was not exhaustive. For example, peers assisting other peers in finding housing is likely a major support activity that was not included. It is not known how the exclusion of this and other supports may affect the results. Finally, small samples in certain programs prevented us from conducting statistical analyses. This limited us to judging program differences on the basis of our impressions rather than on statistical analyses. Nonetheless, we attempted to be cautious in our judgments.

These findings have important implications for this emerging workforce. They demonstrate the great variability in CPS roles and activities that foundations and agencies should consider as they continue to move forward in bringing CPSs into the system. However, such range also presents a challenge for CPS training. Evidence shows that CPS training is effective ( 8, 9, 10 ), yet current training is 80 hours or less and may cover only the basics—usually the core activities that CPSs are found to universally provide across all settings. Greater attention should be paid to providing effective supervision and continuing education focused on the specific setting in which a CPS works and on additional supports that CPSs can provide. One potential area that is deserving of greater attention in training and continuing education is a focus on areas where CPSs might have the most insight into their importance for leading a meaningful life in the community, such as employment, spirituality and religion, citizenship, and dating. These are all areas that are not currently a central focus of CPS work.

This study did not address the extent to which the supports provided by CPSs are effective. However, the findings suggest that research aimed at advancing the evidence base about the effectiveness of CPS-delivered services must keep in mind that the CPS designation is akin to a professional discipline, like psychiatrist, psychologist, or social worker, and that CPSs engage in a wide variety of activities in different settings that may result in different effects. Any study of CPS effectiveness must take this range of activities into consideration when selecting outcomes of interest. Such research will go a long way in validating investments in CPSs as important members of the behavioral health workforce.

As with any other emerging workforce, CPSs will face challenges. Their work needs to be accepted as worthy of pay rather than as a volunteer effort. Moreover, there is pressure to develop national CPS training, professional standards, and ethical standards. Standards will be hotly debated because standardization raises dilemmas on many dimensions—professional versus layperson, professional or academic knowledge versus experiential knowledge, and traditional mental health service delivery approaches versus approaches based on the ethos of self-help and mutual aid. The National Association of Peer Specialists and statewide CPS organizations are taking action to address these issues and to recognize the varied CPS roles and uniqueness of peer support.

Support for the study was provided by a grant H133B031109 from the U.S. Department of Education and National Institute on Disability and Rehabilitation Research (Dr. Salzer, principal investigator). However, the contents of this brief report do not necessarily represent the policy of the U.S. Department of Education, and the reader should not assume endorsement by the federal government.

The authors report no competing interests.

Dr. Salzer and Mr. Brusilovskiy are affiliated with the Department of Psychiatry, University of Pennsylvania School of Medicine, 3535 Market St., 3rd Floor, Center for Mental Health Policy and Services Research, Philadelphia, PA 19104 (e-mail: [email protected]). Mr. Schwenk is with Creative Health Services, Pottstown, Pennsylvania.