Employment Outcomes After Certification as a Behavioral Health Peer Specialist in Four U.S. States

Certified peer specialists (CPSs) use lived experience of a behavioral health disorder plus skills learned in formal training to deliver support services (1). Forty-five states offer peer specialist certification (2), and more than 25,000 CPSs have been certified nationwide (3). CPSs are increasingly employed in U.S. behavioral health service delivery systems (4, 5), and the Centers for Medicare and Medicaid Services recognize peer support as a billable service (5, 6). Peer support demonstrates good outcomes for service users (7) and also relieves some of the nationwide workforce shortage, which has led to increased investments in CPS training and certification (8).

Despite the growth of this workforce, little is known about the extent to which a CPS credential improves employment of workers with behavioral health disorders. CPSs are individuals with psychiatric disabilities (9), a group that faces ongoing challenges to employment (10, 11). Employing CPSs may alleviate low labor force participation by this group (8), but only 17% of state entities that certify CPSs collect outcome data (3).

Research on the impact of the CPS credential has focused on training and certification trends or on employment as a peer specialist, leading to a lack of knowledge about CPS employment in other social service fields or other industries. A recent study in Michigan of CPS graduates investigated professional activities, job satisfaction, wages, and personal finances but excluded nearly 20% of respondents because they were not employed as a CPS (12). We do not know how many CPSs remain in the peer specialist workforce, use the credential to obtain other positions, or do not work at all. CPSs face employment challenges even when trained and credentialed (13), including workplace discrimination, lack of clear job descriptions and role confusion, difficulty requesting accommodations, and lack of networking opportunities (14–16).

This article presents findings from a cross-sectional analysis of baseline data from the CPS Career Outcomes Study, a 3-year (2020–2022), national study of CPSs that gathers information on the work experiences and career perspectives of CPSs after receipt of state-sponsored certification. It includes both employed and unemployed respondents and those working in peer support versus other types of positions. We describe the CPS cohort’s demographic, clinical, work history, and geographic characteristics and the relationships of these domains to employment status and current employment in peer services jobs, as well as characteristics of peer services jobs and non–peer services jobs.

Methods

Sample

This study recruited a sample of 681 adults who completed a peer specialist certification in one of four participating states in the United States. The target population for recruitment was individuals who were recently certified, regardless of their current employment status. According to our inclusion criteria, respondents were age 18 or older and had a 2019 or 2020 certification date. This included recertifications and individuals “legacied” into their state’s new certification, and these respondents (38%, N=261) reported their original certification date.

Recruitment

The CPS Career Outcomes Study includes four geographic regions (northeast [Pennsylvania], southeast [North Carolina], southwest [Texas], and west coast [Oregon]). These states were chosen on the basis of having large cohorts of graduating CPSs in 2019, and they maintained a public certification database, were willing to share their master list of CPSs, or obtained consent from CPSs to reveal their identifying information to researchers.

This study was approved as exempt by the Temple University Institutional Review Board. First, respondents were recruited by a letter sent via the U.S. Postal Service introducing the study, which included a $5 unconditional cash incentive (17). Next, an e-mail was sent with a personalized survey link to an informed consent document, after which respondents completed the questionnaire. Recruitment efforts included multiple e-mail, telephone, and text message reminders. The study principal investigator (L.O.) and research assistants (among them M.P.) who conducted study development and recruitment were people with lived experience of the behavioral health system.

Measures

Survey data were collected online from March to October 2020 using Qualtrics. The survey included validated measures and items developed by the research team, which were reviewed by a five-person advisory group of experts in peer support employment and training and certification. The survey could be completed anywhere with an Internet connection and on multiple devices. If they requested, respondents completed the survey over the telephone.

Data were collected on domains of participant characteristics previously shown to be associated with employment or peer services work among people living with diagnoses of serious mental illness. This included disability benefits and veteran status (18, 19), physical health (18), mental health diagnosis and clinical services (18), and prior work history, negative work experiences, workplace stigma, and work motivation (10).

Respondents were asked about demographic characteristics and benefits, including age, gender, race, ethnicity, education, veteran status, and receipt of Social Security Disability Insurance (SSDI) or Supplemental Security Income (SSI) benefits. Physical health was assessed with a self-report item (20). From a provided list, participants selected psychiatric disorders for which they had received a formal clinical diagnosis. Self-reported diagnoses were coded in the following hierarchy: schizophrenia spectrum disorder, bipolar disorder without schizophrenia, depressive disorder without schizophrenia or bipolar disorder, and all other diagnoses. A diagnosis of a substance use disorder was coded as a separate yes-no variable. Use of mental health services in the past year was coded as outpatient mental health counseling or therapy, outpatient medication management by a psychiatrist or other prescriber, and psychiatric inpatient hospitalization.

To capture work history, respondents were asked about their employment history in the 5 years prior to the 2020 survey. Motivation for working in peer service positions was also reported, with options representing a series of internal and external reasons, which for this analysis was coded into “giving back and helping others” versus other reasons (e.g., being valued, recovery language, open disclosure, and making meaning of suffering), given prior research on the importance of the transformational role of peer services (21). Respondents were asked about patterns of disclosure of psychiatric history at work (“Do you openly disclose your personal experience of mental health challenges to work colleagues?”), and responses were coded as never willing to disclose (versus sometimes or always) at a peer services or other job. Although disclosure is part of peer services, some peer services jobs are in larger settings, such as hospitals or mixed service agencies, in which a peer specialist might not disclose to all work colleagues all the time. Number of negative experiences at any job used a modified version of a 14-item Kessler Foundation checklist (22). Job satisfaction was measured with the Brief Index of Affective Job Satisfaction (23). Years since original certification was included in our models.

Two external factors that affect employment were assessed. The first was the local labor market, which was assessed by examining local area unemployment rates. Respondent zip codes were used to identify the average county unemployment rate in 2020 (24) and rural or small-town residence versus larger residential area (25). Participating states were assured that results would not be reported publicly by state, although state was included as a covariate in our models to account for regulatory and practice differences.

Respondents reported the characteristics of paid positions held in the 12-month period prior to the survey, including average hours per week, hourly or annual salary, use of job benefits, job title, location, and responsibilities. They also reported whether the position was a peer specialist job, and if so, whether certification or training was required. On the basis of this information, jobs were coded by research staff by using the Bureau of Labor Statistics (BLS) standard occupational categories (26). All jobs were coded by one research staff, and a 10% random subsample (N=85 jobs) was then coded by a second researcher (J.B.-M.) to assess reliability. The two researchers then met to compare codes and reach a consensus on conflicts.

Statistical Analysis

Associations between participant characteristics and employment status and job type were examined by using chi-square and t tests. Variables were selected for inclusion in logistic regression models. Given the exploratory nature of the analysis and the relatively large number of predictor variables, the final model terms were identified using backward stepwise regression to exclude independent variables that did not add significantly to the model.

Results

Of 681 respondents who consented to the survey, 591 (87%) provided data on current employment status and were included in the analysis. Participant characteristics are shown in Table 1. Most identified as female (66%) and as White (69%), followed by Black or African American (24%); and 9% identified as Latinx. Only 16% reported receiving SSDI income, and even fewer reported SSI (4%). Of the sample, 11% were veterans. The mean age of participants was 47. Among participants, 15% reported an education level of high school or less, 39% reported some college, and 46% reported any degree (associate’s degree or higher).

Psychiatric diagnoses were reported as follows: schizophrenia, 9%; bipolar disorder, 25%; and depressive disorders, 41%. Almost a third (31%) reported a substance use diagnosis. About half (56%) used outpatient therapy or counseling in the past year, and a similar proportion (54%) reported use of medication management or psychiatry services. Only 5% reported inpatient psychiatric hospitalization in the past year. On a scale from 1, low, to 5, high, participants reported their physical health at an average of 3.

On average, it had been 2.5 years since participants were first certified as peer specialists. Most participants were employed at the time of the survey (N=448, 76%); 96% (N=567) had worked in the 5 years prior to the survey, and 41% had worked continuously during that time. Table 2 shows the relationship between participant characteristics and employment status (currently employed or not employed), as indicated by bivariate chi-square and t tests. Only 13% of those currently employed were receiving SSDI, compared with 26% of those who were unemployed (p<0.001), and only 3% of employed respondents reported SSI, compared with 8% of the unemployed group (p=0.005). Among veterans, the proportion currently employed was lower than the proportion unemployed (10% versus 17%, p=0.018). Among employed respondents, the proportions reporting past-year use of outpatient and inpatient mental health services were lower than among those who were unemployed, and mean self-reported health ratings were better among employed respondents than among unemployed respondents (3.0 versus 2.8, p=0.044).

Work history was significantly associated with employment status, and 47% of employed respondents had worked continuously in the past 5 years, compared with 22% of unemployed respondents (p<0.001). Only 7% of those who were currently employed reported never disclosing their psychiatric history to colleagues, compared with 15% of those who were unemployed (p=0.004). Employed respondents had been certified for significantly longer than those who were unemployed (2.6 years versus 2.0 years, p=0.015). Other characteristics—including state, rural residence, and county unemployment rate—were not associated with current employment status.

In the final multivariable logistic regression, SSDI and SSI beneficiary status and veteran status continued to be significantly associated with a lower likelihood of current employment, as did receipt of outpatient counseling in the past year and never disclosing one’s psychiatric history to colleagues. Longer time since certification remained significantly associated with greater likelihood of current employment.

Among those currently employed, we examined characteristics associated with having a peer services position, compared with a non–peer services job, using the same multivariable model (Table 3). Respondents working in peer services jobs had been certified longer than those in non–peer services jobs (2.9 years versus 2.0 years, p=0.002). A smaller proportion of those working in peer services jobs reported never disclosing their psychiatric history to colleagues, compared with those in non–peer services positions (3% versus 15%, p<0.001). Finally, the proportion of respondents working in peer services jobs differed significantly by state (p<0.001).

In the final step of the multivariable logistic regression model predicting likelihood of being employed in a peer services job, those reporting depressive disorders were significantly more likely than those with other diagnoses to have peer services jobs (odds ratio [OR]=1.93, p=0.028). In addition, those who did not disclose their mental health status to colleagues were significantly less likely to hold a peer services job, compared with those who did disclose (OR=0.24, p<0.001). Certification state was significantly associated with the likelihood of working in peer services jobs (p<0.001). A higher local unemployment rate also was associated with lower likelihood of working in peer services jobs (OR=0.78, p=0.003).

Finally, we compared characteristics of peer services jobs and non–peer services jobs among those who were employed (Table 4). Among those with peer services jobs, the proportion who had held the job for 1 or more years was higher, compared with the proportion among those who held non–peer services jobs (72% versus 57%, p=0.003). A larger proportion of those who held peer services jobs reported employee benefits, compared with those who held non–peer services jobs, including paid time off (73% versus 53%, p<0.001) and health insurance (55% versus 35%, p<0.001). Job satisfaction was significantly higher among those in peer services jobs, compared with those in other jobs (p<0.001). No significant differences were found between peer services jobs and other jobs in hourly wage, average hours worked per week, and job tenure measured in months.

In the BLS Standard Occupational Classification system, peer services are classified as “community and social service occupations.” As shown in Table 4, 93% of peer services jobs were coded as such, and 6% were coded as a “management occupation.” Among non–peer services jobs, the largest group was classified as community and social service occupations (30%), followed by other unrelated categories. The BLS uses job zones to characterize a group of occupations (from lower to higher) that are similar in educational requirements, experience needed, and on-the-job training needed. Compared with the peer services jobs, over half of the non–peer services jobs were in a lower job zone (59%, N=72), 34% (N=42) were in the same job zone, and only 7% (N=9) were in a higher job zone.

Discussion

The results of this first national study to examine the labor force participation of individuals after their certification as peer specialists revealed a high employment rate, exceeding that observed in other studies of adults with behavioral health conditions (10, 27). Predictors of unemployment were similar to those found in other studies of adults with psychiatric disabilities. These overarching unemployment predictors include poorer physical health and higher use of outpatient mental health services, reliance on public disability and other benefits, and stigma (as reflected in lower likelihood of disclosure of mental health status).

In multivariable analyses predicting likelihood of any type of employment, SSI and SSDI beneficiaries were less likely to be working. This finding is consistent with a large body of research reviewed by O’Day and colleagues (28) suggesting that participation in federal disability programs can act as a disincentive to find paid work, as well as with other research on the association between greater impairment and benefits eligibility (29). Veterans were also less likely to hold a paid job, in line with findings of other research (30, 31), which may be related to ongoing psychological stress (19). Respondents who never disclosed their psychiatric history to coworkers were also less likely to be currently employed, confirming the positive association between disclosure and employment status observed in other studies (32, 33). Finally, the association between use of outpatient mental health services and unemployment could be attributable to less need for services among the employed group because of the positive impacts of work on mental health (34, 35). However, many states require CPS applicants to demonstrate evidence of “recovery” (13), which may mean that certification deters them from seeking services.

The study compared respondents working in peer services positions versus other types of jobs and found that those who reported a lifetime depressive disorder diagnosis were more likely to hold peer services positions, compared with other types of jobs. This finding is consistent with the high prevalence of depressive disorders in prior studies of the peer specialist workforce (36, 37).

Compared with respondents working in non–peer services positions, those employed in peer services roles were more likely to disclose their psychiatric history to coworkers. This is not surprising, given that disclosure is a hallmark of peer-delivered recovery services. Employment in a peer services position was also more likely among those residing in areas with low unemployment, which may have reflected an association between low unemployment and expanded employment opportunities for disadvantaged workers (38). Finally, significant differences were found by state in the likelihood of holding a peer services position, and this finding may have reflected the availability of positions or the hiring requirements in certain states (e.g., a driver’s license or lack of criminal record), as described in the peer specialist database (2).

Job satisfaction was significantly higher among those holding peer services positions, compared with other jobs. This is not surprising given the extent to which respondents endorsed peer work as providing opportunities to help others and promote recovery ideals. High levels of job satisfaction have been found in other studies of those providing peer services (39) because of psychological benefits to working as a peer specialist (40). Although no significant differences in hourly wage or number of hours worked per week were noted, providers of peer services were significantly more likely to report a longer current job tenure. Other studies of peer services providers have also reported high levels of job stability, even during the COVID-19 pandemic (41, 42). These advantages of peer services employment are noteworthy, especially compared with the entry-level, low-wage, and low-skilled positions that are typically held by this group of workers (43).

This study had several limitations. Recruitment of newly certified CPSs proved difficult because of changes in certification processes. We collected the original date of certification from respondents, allowing us to compare new CPSs to individuals who had received certification several years ago. Most data came from respondent self-report, which could be subject to positive response bias or inaccurate recall. Our respondents were not a nationally representative sample of newly certified CPSs, although we recruited from four regionally diverse states, and the demographic characteristics of our sample were similar to those reported in a 2020 survey of 1,280 peer specialists in terms of gender, race, education, and age (41). In addition, the cross-sectional observational design of the study does not support causal inference. Finally, recruitment was conducted at the beginning of the COVID-19 pandemic, which may have altered responses and response rates. However, we found that numbers of surveys completed increased each month between April and August. Recent studies have found high levels of job stability in peer services and other types of employment in 2020 in the months after the pandemic began (41, 42).

Conclusions

This study adds to the sparse literature on employment trajectories after recent certification of peer specialists. Responses to this survey indicated that the CPS population, compared with all adults with psychiatric disabilities, has relatively high education, recent work history, lower current use of SSI and SSDI, high rates of disclosure of mental health status to coworkers, and a high current employment rate. Some of these patterns may stem from state requirements that CPS applicants have at least a high school diploma (or equivalent) and recent work experience (8). These policies may negate attempts to alleviate employment issues for people with psychiatric disabilities through paid peer services positions.

Research on this relatively new professional credential is hindered by the way in which CPS positions are tracked by the U.S. Department of Labor, which includes CPSs under the heading of “health educators and community health workers” (44). These job categories differ substantially in their educational requirements, competencies, and lived experience (14, 45). Creation of a designation of peer specialist as a Department of Labor standard occupational classification, as well as continued research on career trajectories, could advance our knowledge about the certification and employment outcomes of this essential workforce.