Individual placement and support programmes for young adults: Where are they and whom do they serve?

Abstract

Aim:

Individual Placement and Support (IPS), an evidence-based supported employment model developed for adults with serious mental illness, has been recently targeted to young adults with mental health conditions, but little is known about its adoption in this age group in the United States.

Methods:

We recruited a volunteer sample of nine IPS programmes in five states serving young adults with mental health conditions aged 16 to 24. IPS team leaders reported programme and participant characteristics and rated barriers to employment and education.

Results:

Most IPS programmes were located in community mental health centres, served a small number of young adults, and received most referrals from external sources. The study sample of 111 participants included 53% female, 47% under 21 years old, 60% diagnosed with a depressive disorder; 92% had an employment goal, and 40% had an education goal. IPS specialists reported that managing mental health symptoms was the most common barrier to achieving employment and education goals.

Conclusion:

Future research should examine how IPS programmes could best provide services to young adults.

1 |. INTRODUCTION

Young adulthood is a developmental stage defined by stress and change. Half of all lifetime mental illnesses begin by the age of 14 and three quarters of them by the age of 24 (Kessler et al., 2005). In the United States in 2020, 17% of those aged 12–17 and 12% aged 18–25 reported a major depressive episode within the last 12 months (Substance Abuse and Mental Health Administration [SAMHSA], 2021). The literature varies widely in the age range considered young adulthood; in this paper, we use the term young adult to refer to those aged 16–24. Historically, in treating young adults, community mental health centres (CMHCs) have focused primarily on medication management and symptom reduction with little regard for client goals and aspirations (Costa et al., 2017), although most young adults have a strong desire to work and further their education (de Waal et al., 2018; Iyer et al., 2011; Ramsay et al., 2011).

Employment is a social determinant of health for adults of all ages, including young adults (Drake & Wallach, 2020). Unfortunately, employment rates amongst young adults are far lower for those with disabilities than those without (e.g., 18% vs. 31%, respectively, for young adults 16–19 years of age) (U.S. Bureau of Labor, 2017). Young adults with mental health conditions similarly have poor employment outcomes. Of the young adults aged 18 to 24 years who received services from the US public mental health system in 2020, 15% were employed, 10% were in education, and the majority (75%) were neither employed nor in education (SAMHSA, 2022).

Unemployment undermines the integration of young adults into their community and furthers the misconception that they do not have the ability or motivation to work (Kaye et al., 2011). A lack of work experience prior to reaching adulthood portends long-term negative outcomes, such as long periods of unemployment, lower lifetime earnings, and a lifelong dependency on Social Security disability benefits (Topor et al., 2019). Educational achievements during young adulthood similarly impact later adulthood, most notably concerning total lifetime earnings (Luciano & Meara, 2014). Young adults who are not in education, employment, or training (NEET) experience discouragement and social exclusion and often develop problems with mental health and substance use (Gariépy et al., 2022).

Individual Placement and Support (IPS), an evidence-based model of supported employment defined by eight principles (Drake et al., 2012), addresses these challenges. Although IPS was originally developed for adults with long-term mental health conditions and historically has served clients with long-term psychiatric disorders (typically enrolling clients who are at least 20 years of age) (Bond et al., 2016), IPS has continued to expand to other populations, including younger adults. To address the goals of young adults, many IPS programmes have augmented supported employment with supported education (Manthey et al., 2012; Nuechterlein et al., 2020).

The successful implementation and maintenance of any evidence-based practice lies in continuous monitoring and quality improvement. An aid in this practice is the IPS-25, a 25-item fidelity scale used by independent assessors to evaluate adherence to the IPS model (Becker et al., 2019). The IPS-Y, a modification of the IPS-25 fidelity scale, is intended to better assess IPS services for young adults by incorporating supported education and other youth-oriented items, such as outreach to families and career exploration (Bond et al., 2019).

The number of young adults with mental health conditions receiving IPS is unknown. One segment of the young adult population with access to IPS consists of young adults experiencing an early episode of psychosis. Coordinated specialty care (CSC) is a programme model designed specifically for this population (Read & Kohrt, 2021). This model includes the provision of supported employment and education consistent with IPS (Addington et al., 2020). The number of CSC services continues to expand throughout the United States because of congressionally mandated funding (George et al., 2022). Currently, over 360 CSC programmes are operating in all 50 states for first episode psychosis (Heinssen & Azrin, 2022).

Few studies have documented the extent to which young adults receive IPS or where they access these services. In this paper, we describe the characteristics of a small national sample of IPS programmes and the young adults enrolled in these programmes, in addition to examining barriers to employment and education.

2 |. METHODS

This study examines baseline information from a prospective observational study assessing employment and education outcomes for young adults receiving IPS.

2.1 |. Study sites and participants

In 2019, we announced plans for a proposed study of IPS programmes for young adults and solicited participation with IPS leaders from 24 states participating in a national IPS learning community (Pogue et al., 2022). To increase geographic diversity, we targeted our recruitment to five states in three general regions of the United States (West, Midwest, and East). Participating states were California, Kentucky, Minnesota, South Carolina, and Wisconsin. Each state leader nominated several local programmes providing IPS services to young adults. The research team contacted local IPS team leaders until we had our requisite sample of two sites per state.

Initially, each site would prospectively enrol 15 participants during 2020, identifying them as they were newly admitted to the IPS programme. Starting in March 2020, however, the COVID pandemic resulted in massive restrictions on in-person access to public places, severely affecting referral rates to the participating study sites. Because of low enrollment rates, we extended the recruitment period until June 2021. In addition to the requirement that participants be newly enrolled in IPS services, eligibility criteria were: (1) between ages 16 and 24, (2) a documented psychiatric disability, (3) not competitively employed, (4) expressed interest in a competitive job and/or mainstream education, (5) no prior enrollment in IPS services, (6) no legal or other restriction preventing study participation.

2.2 |. Measures

2.2.1 |. Site descriptions

IPS team leaders completed an 18-item checklist created for this project describing programme characteristics (e.g., number of IPS specialists, referral sources) and site characteristics (e.g., type of agency, population of community).

2.2.2 |. Baseline characteristics of study participants

Drawing on agency records, IPS team leaders maintained a spreadsheet of participant baseline information they shared with the research project coordinator after each new enrollment. Demographic information consisted of close-ended questions assessing characteristics, such as age, gender, race/ethnicity, and disability benefits. IPS team leaders also obtained psychiatric diagnoses from the mental health agency’s medical records.

2.2.3 |. Employment and education history

At baseline, participants indicated whether they had an employment goal, an education goal, or both. IPS specialists interviewed participants on their employment and education history using questions from the Dartmouth Employment and Income Review (Drake et al., 1996).

2.2.4 |. Barriers checklists

IPS teams rated participants on the Barriers to Employment Scale, a 23-item checklist of mental health, job-related, and personal factors associated with employment outcomes (Johannesen et al., 2007). Based on a review of the literature (Johannesen et al., 2007; Kukla et al., 2015; Manthey et al., 2015), we then created a parallel checklist, the Barriers to Education Scale, consisting of 21 items specific to education, such as instructor empathy and accommodations on campus. Both checklists provide four response alternatives: a major barrier, somewhat of a barrier, not a barrier, and unknown or no answer (i.e., not applicable).

2.3 |. Data collection and analysis

Leaders at each site agreed to collect baseline information on 15 participants. Recognizing that some young adults never meaningfully engage in services after an initial referral, we stipulated a priori that we would exclude early withdrawals. Thus, the final sample excluded participants withdrawing within the first 60 days of enrollment. Study enrollment continued until each site reached the target sample size or until the end of the recruitment period.

Data analysis was mostly limited to descriptive statistics. We compiled site characteristics using the information provided by site leaders. We aggregated participant data, constructing tables of frequencies and percentages of demographic and other background measures, including barriers to employment and education. In a series of exploratory statistical analyses, we compared two groups defined by broad diagnostic classification (psychotic disorder vs. nonpsychotic disorder) on each of the 23 employment barrier items using a 2 × 2 chi square test.

3 |. RESULTS

3.1 |. Description of study sites

One of the original 10 participating sites withdrew from the study during the second year. Table 1 presents descriptive information on the 9 remaining sites.

TABLE 1.

Agencies with IPS programmes for young adults.

State	Site	Community	Type of agency	Type of referrals	Funding	IPS team size	IPS staff turnover	Eligibility	Clients
California	Site A	Metropolitan	Youth & family services	Internal	Medicaid & county funds	2	1	SMI	YA only (18–25)
	Site B	Metropolitan	Family services	Internal	Medicaid & county funds	2	0	Any MH diagnosis	YA only (17–25)
Minnesota	Site C	Metropolitan	Mental health centre	External	VR	7	0	SMI or SPMI	22 YA, 95 > age 26
	Site D	Metropolitan	Rehabilitation agency & family services	External	VR	2.5	1	SPMI	10 YA, 32 > age 26
Wisconsin	Site E	Metropolitan	Rehabilitation agency	External	Medicaid & VR	8	0	Any MH diagnosis	24 YA, 146 > age 26
Kentucky	Site F	Rural	Mental health centre	Internal & external	VR & state contract	7.5	0	MH diagnosis	Both YA and adults > age 26
	Site G	Rural	Mental health & Substance use treatment	Internal & external	VR	3	2	VR eligible	9 YA, 30 > age 26
South Carolina	Site H	Small city	Mental health centre	Internal & external	VR, Medicaid, & state contract	2	2	Client at CMHC	8 YA, 34 > age 26
	Site 1	Small city	Mental health centre	External	VR, Medicaid, & state contract	2	0	Client at CMHC	11 YA, 42 > age 26

Abbreviations: CMHC, community mental health centre; MH, mental health; SMI, serious mental illness; SPMI, serious and persistent mental illness; VR, Vocational Rehabilitation; YA, young adult.

Five sites were located in large metropolitan areas (population exceeding 1 million), whilst two sites were in much smaller cities (population between 60 000 and 200 000), and another two were in rural areas (population less than 40 000). Five sites were CMHCs, two were child and family services, one was a community rehabilitation agency, and one was a family services/rehabilitation agency.

Four sites (two CMHCs, one rehabilitation agency, and one rehabilitation agency/family services) relied exclusively on external referrals to fill their young adult caseload, two sites (family services) had internal referrals only, whilst three sites (all CMHCs) relied on a combination of external and internal referrals. The client eligibility criteria were similar across the nine study sites. Two sites required enrollment in services at the CMHC where the IPS programme was located. Six sites required IPS clients to have a mental health diagnosis, though the type of diagnosis and its documentation varied (i.e., whether any mental health diagnosis would qualify, or if the prospective client had a serious mental illness or serious and persistent mental illness). The two family services sites were IPS programmes exclusively for young adults. The other 7 sites were conventional IPS programmes in that the IPS caseloads included both young adults and clients older than 24. The mean proportion of young adults in the overall IPS caseload was 20% in 6 sites providing this information.

3.2 |. Participant characteristics

The final sample consisted of 111 participants. Table 2 presents the overall sample characteristics. The study sample included 59 (53%) female, 53 (47%) aged 18–20 years old, 55 (50%) White, and 20 (18%) Hispanic/Latino participants. The three most common psychiatric diagnoses were depressive disorders (60%), anxiety disorders (39%), and attention deficit hyperactivity disorder (23%). Few received any government benefits: 15 (14%) received Supplemental Security Income (SSI) whilst 9 (8%) received Supplemental Nutrition Assistance Program (SNAP) benefits. About 23% of the sample had been arrested.

TABLE 2.

Background characteristics of the young adult sample.

Characteristic	Participant sample (N = 111)
	N (%)
Gender
Female	59 (53.2%)
Male	47 (42.3%)
Other	5 (4.5%)
Age	(M = 19.2, SD = 2.2)
16–17	26 (23.4%)
18–20	53 (47.2%)
21–24	32 (28.8%)
Race
White	55 (49.5%)
Black	29 (26.1%)
Hawaiian/Pacific Islander	3 (2.7%)
Native American	1 (0.9%)
Asian	0 (0.0%)
Biracial/Multiracial	6 (5.4%)
Not reported	17 (15.3%)
Ethnicity
Hispanic/Latino	20 (18.0%)
Marital status
Never married	109 (98.2%)
Married/Living as married	2 (1.8%)
Psychiatric diagnosis a
Depressive disorder	66 (59.5%)
Anxiety disorder	43 (38.7%)
ADHD	26 (23.4%)
PTSD	20 (18%)
Schizophrenia spectrum	16 (14.4%)
Substance use	14 (12.6%)
Bipolar	12 (10.8%)
Autism	10 (9.0%)
Housing situation b
Living with family/caregiver	71 (64.0%)
Own apartment or house living with/without someone	26 (23.4%)
Homeless	5 (4.5%)
Other residential setting	9 (8.1%)
VR client	14 (12.6%)
Benefits
SSI	15 (13.5%)
SSDI	1 (0.9%)
SNAP	9 (8.1%)
Other	4 (3.6%)
Legal involvement
Arrested	26 (23.4%)
Convicted	15 (13.5%)
Probation	11 (9.9%)
Work history
Never worked	72 (64.9%)
One prior job	22 (19.8%)
Two prior jobs	9 (8.1%)
>2 prior jobs	8 (7.2%)
Education
Completed elementary	1 (0.9%)
Some high school	57 (51.4%)
Completed high school/GED	41 (36.9%)
Some post-secondaryc	11 (9.9%)
Associate’s degree	1 (0.9%)
Goals to work on
Employment	67 (60.4%)
Education	9 (8.1%)
Both	35 (31.5%)

^aSome participants had multiple diagnoses.

^bEach category except for ‘Homeless’ includes both independent and supervised participants.

^cIncludes one technical certificate.

Regarding education, 58 (52%) had not completed high school whilst 53 (48%) had completed high school or equivalent. At the time of IPS enrollment, 29 (26%) were in education. None were working at enrollment and nearly two-thirds had never worked in a paid job. Nearly all reported a goal of competitive employment, whilst 9 (8%) had a goal of education (but no employment goal), and 35 (32%) had both goals.

We also examined potential differences amongst participant characteristics by age group and by race/ethnicity. With a few exceptions, participants aged 16–17 years (n = 26), 18–20 years (n = 53), and 21 or older (n = 32) did not significantly vary on most background characteristics (See Appendix A, Table A1). Unsurprisingly, the youngest age group (16–17 years) was less likely to have work experience, have graduated high school, and somewhat more likely to live with their family. However, this age group was also significantly less likely to have a schizophrenia/bipolar diagnosis (4%) compared with 23% of those aged 18–20 and 41% of those 21 or older (χ² = 10.85, p < .01). Conversely, the youngest age group was significantly more likely to have an anxiety diagnosis compared with those aged 18–20 and those 21 or older (54% vs. 42% and 22%, respectively) (χ² = 6.51, p = .04).

When comparing participants based on their race/ethnicity, the White (n = 51), Black (n = 29), and Hispanic (n = 20) young adults also differed in some characteristics (See Appendix B, Table B1). Hispanic participants were more likely to live with family compared with White and Black participants (85% vs. 67% and 41%, respectively), and were least likely to have an education goal (15% vs. 41% and 52%, respectively). Black participants were more than twice as likely as White and Hispanic participants to be arrested (χ² = 6.82, p = .03). Whilst Black participants were significantly less likely to be diagnosed with depression (35%) compared with 71% of White participants and 65% of Hispanic participants (χ² = 10.34, p = .01), they were most likely to have a schizophrenia/bipolar disorder (48%) compared with White and Hispanic participants (16% and 5%, respectively) (χ² = 13.36, p < .01).

3.3 |. Barriers to employment and education

As shown in Table 3, the most common barriers to both employment and education were managing symptoms of their mental health condition, trouble concentrating, and managing time. Physical illness was a barrier to employment for only 19% and a barrier to education for only 17% of the study sample. Concerns about disclosing mental health issues was rated as more of a barrier to employment than to education (45% vs. 29%).

TABLE 3.

IPS team ratings of barriers to success in employment and education^a for the young adults in the study (N = 111).

Barriers to employment	N (%)	Barriers to education	N (%)
Managing symptoms of mental health condition	90 (81.1%)	Managing symptoms of mental health condition	78 (70.3%)
Trouble concentrating	80 (72.1%)	Trouble concentrating	78 (70.3%)
Managing time	78 (70.3%)	Managing time	76 (68.5%)
Difficulty finding a job	71 (64.0%)	Schoolwork is too stressful	66 (59.5%)
Lack of self-confidence	64 (57.7%)	Lack of self-confidence	57 (51.4%)
Personal problems	63 (56.8%)	Personal problems	56 (50.5%)
Transportation	62 (55.9%)	Fear of failure	51 (45.9%)
Difficulty keeping a job	60 (54.1%)	Being stigmatized because of mental health issues	46 (41.4%)
Being stigmatized because of mental health issues	59 (53.2%)	Academic record	40 (36.0%)
Work is too stressful	58 (52.3%)	Getting along with others	40 (36.0%)
Lack of education or skills	58 (52.3%)	Transportation	39 (35.1%)
Concerns about disclosing mental health issues	50 (45.0%)	Lack of support from family or friends	35 (31.5%)
Fear of failure	50 (45.0%)	Lack of instructor empathy	35 (31.5%)
Getting along with others	49 (44.1%)	Concerns about disclosing mental health issues	32 (28.8%)
Lack of support from family or friends	43 (38.7%)	Using alcohol or drugs	24 (21.6%)
Using alcohol or drugs	29 (26.1%)	Lack of access to campus services for people with mental health issues	22 (19.8%)
Side-effects of medications	29 (26.1%)	Side-effects of medications	21 (18.9%)
Possibility for advancement	28 (25.2%)	Lack of accommodations on campus	20 (18.0%)
Physical illness	21 (18.9%)	Racial discrimination	20 (18.0%)
Racial discrimination	20 (18.0%)	Physical illness	19 (17.1%)
Taking care of family members	18 (16.2%)	Lack of support from case managers	2 (1.8%)
Fired from a job	14 (12.6%)
Fear of losing benefits	12 (10.8%)
Lack of support from case managers	3 (2.7%)

^aItems on both sets of scales could be rated as unknown or no answer, not a barrier, somewhat of a barrier, and a major barrier. Responses for a major barrier and somewhat of a barrier were combined in this table.

3.4 |. Exploratory analysis of barriers by mental health condition

We examined differences in barriers for 52 participants with a psychotic diagnosis (schizophrenia or bipolar disorder) to 59 participants with a non-psychotic diagnosis (depressive disorder or anxiety disorder) (See Appendix C, Table C1). On most items, the IPS teams identified the group with psychotic diagnoses as more likely to encounter barriers than the group with non-psychotic diagnoses. A significantly greater percentage of the young adult participants with a psychotic diagnosis were rated as having barriers in side-effects of medications (χ² = 7.71, p = .01), using alcohol or drugs (χ² = 5.50, p = .02), physical illness (χ² = 4.09, p = .04), and fear of losing benefits (χ² = 4.28, p = .04). There was also a trend towards significance in other barriers, such as managing symptoms of mental health condition, difficulty keeping a job, and work is too stressful.

4 |. DISCUSSION

4.1 |. IPS programmes for young adults

The majority of IPS programmes in this study served young adults 18 years of age or older, although nearly a quarter of the sample were 16 or 17. Although a national survey is needed, this study suggests the hypothesis that, excluding IPS services provided in CSC programmes, most IPS services for young adults are found in standard IPS programmes enrolling clients of all ages rather than IPS programmes limited to young adults.

In the United States, the public mental health system is limited mainly to CMHCs, and young adults have limited access to these mental health services, including IPS. Moreover, perceptions of public stigma (i.e., negative views and discrimination from the public) discourage young adults from seeking mental health treatment (Pedersen & Paves, 2014). Perhaps unsurprisingly, young adults in the United States drop out more rapidly from mental health services than older adults, increasing their morbidity throughout their life (Edlund et al., 2002; Giaconia et al., 1994).

Outside the United States, several countries have national policies aimed to address the alarming growth in their NEET population. This trend is associated with a surge of early applicants for disability benefits as well as poor outcomes in later adulthood (Arulampalam et al., 2001; Gioia, 2006; Gregg & Tominey, 2005; Ralston et al., 2016; Reine et al., 2004; Sveinsdottir et al., 2018; Topor et al., 2019). In Norway, the national vocational rehabilitation services have targeted IPS to this group of vulnerable young adults (Sveinsdottir et al., 2020). In Australia, a national network of integrated youth services called headspace centres were created through federal funding (Rickwood et al., 2019). Between 2006 and 2019, headspace centres grew to 110 centres serving around 500 000 adolescents and young adults (McGorry et al., 2022). In addition to mental health and general medical care, these centres provide a range of programmes, including employment services. As of 2021, 50 headspace centres in Australia offered IPS services (Watts & Watson, 2022). A recent programme evaluation of IPS services provided through headspace centres showed promising employment outcomes (Simmons et al., 2023). In the United States, few young adults have access to integrated youth services comparable to headspace. Recently, however, two allcove centres based on the headspace model have been developed in California (https://allcove.org/all-centers/).

All of the sites struggled to fill their caseloads, and whilst the COVID pandemic was one main reason for low enrollment rates, it was not the sole barrier. This study did not assess which were the most dependable referral sources: external, internal, or multiple sources. In a previous study, IPS teams found that better integration with the clinical staff provided a more streamlined referral process for internal referrals (Cohen et al., 2019). With external referrals, other processes might need to be put in place. IPS programmes serving high school students need to develop additional referral sources, approaches to engagement, and new strategies to ensure integration with mental health treatment (Ellison et al., 2022). Also noteworthy in this sample is the absence of any programme explicitly targeting young adults with early episode psychosis.

4.2 |. Young adult demographics

Depressive and anxiety disorders, commonly classified as common mental disorders (CMDs), were two of the most prevalent psychiatric diagnoses in this sample. Globally, 31% of adolescents aged 10 to 19 have CMDs (Silva et al., 2020). Diagnostically, the sample differed from older adults served by IPS programmes located in CMHCs, which usually include a large proportion of people with schizophrenia (Campbell et al., 2011; de Winter et al., 2022; Richter & Hoffmann, 2019). The differences we found in psychiatric diagnoses based on race/ethnicity, specifically that Black young adults were significantly more likely to have a psychotic disorder diagnosis than other groups, present potentially important findings for disparities research. The additional significant differences in psychiatric diagnosis by age group may also be informative to research on outreach strategies, though replication with larger samples is necessary.

The rate of young adult participants receiving SSI and Social Security Disability Insurance (SSDI) was low, especially when compared to young adult samples in other IPS/supported employment studies, such as one with an older age range (21–29) in which 46% were either SSI or SSDI beneficiaries (Bond et al., 2016). This rate is also low compared to an early psychosis sample that received supported employment and education in which 34% were SSI/SSDI beneficiaries (Rosenheck et al., 2017). The low rates in the current sample might have been due to their younger age range, as the number of beneficiary recipients typically increases with age (Social Security Administration [SSA], 2022).

Most young adults in this study expressed an interest in pursuing employment, including one-third who wished to pursue both employment and education. The small proportion who wanted to focus on education was unexpected, given the age group. The transitional nature of young adulthood denotes frequent shifts in goals. Self-worth and self-esteem are important to young adults with mental health conditions generally, and having the ability to choose a career path elevates not only young adults’ feelings about themselves but also facilitates their obtaining employment (Gmitroski et al., 2018). Interventions allowing for career development help to promote feelings of self-efficacy, as found in a qualitative study of young adult males with co-occurring disorders and substance abuse (Luciano & Carpenter-Song, 2014).

4.3 |. Barriers to employment and education

The IPS specialists perceived that managing symptoms of mental health conditions was the most common barrier to both employment and education that their young adult clients experienced. This is in line with the findings of a previous study in which poor control of psychiatric symptoms was also a barrier to employment and education amongst adolescents (Manthey et al., 2015; Noel et al., 2017; Sabella, 2021). Situational barriers, such as stress-inducing situations and interpersonal conflicts with teachers and peers, are also common barriers to education (Sabella, 2021).

Studies have also noted subgroup differences in barriers to employment. For example, economically disadvantaged young women face additional barriers that cause their low rates of employment compared with similarly disadvantaged young males, especially amongst young female high school dropouts with children (Miller & Porter, 2007). Transgender young adults face many barriers to employment, such as overt discrimination, and those who are employed have higher levels of internalized transphobia and greater fear of disclosing their mental health due to being more exposed to stigma at the workplace (Budge et al., 2010; Mizock & Mueser, 2014).

Nearly a quarter of the sample had some current or past legal issues, such as arrests. Involvement in the criminal justice system is another factor that may pose a barrier to employment for young adults, as it leads to difficulties in engaging with services and disengagement from services after they have begun (Sveinsdottir & Bond, 2020).

The IPS programmes served young adults with a variety of diagnoses. Given that IPS was first developed for people with serious mental illness, we compared two subgroups defined by whether or not they were similar to the target group for which IPS was first developed (psychotic vs. non-psychotic diagnosis). The staff on the IPS teams perceived young adults with psychotic disorders as having more barriers to employment, including management of work stresses that might lead to job loss, medication issues, and alcohol and drug use. However, larger samples are needed to assess whether these exploratory findings replicate.

The wide range of barriers to employment and education experienced by young adults suggests that solutions to overcoming these barriers are also context-dependent and usually require flexibility on the part of case managers, workplaces, and educational institutions (Corliss et al., 2007). Thus, continuing priorities for service providers are increasing access of these services to young adults and sustaining engagement with these services. Although a meta-analysis of 7 randomized controlled trials evaluating the effectiveness of IPS for young adults has found positive outcomes (Bond et al., 2023), additional rigorous studies are needed.

4.4 |. Limitations

The study sample consisted of an opportunity sample of sites located in states participating in a 24-state learning community, and the biases in this non-random sample are unknown. The COVID-19 pandemic hindered the data collection process of this study. During this time, mental health centres and other agencies providing IPS moved to provide services remotely, making it difficult to enrol new clients. Additionally, whilst IPS specialist ratings provide one perspective of barriers to employment and education, the perspectives of young adults would further our understanding of these barriers. The exploratory analysis of barriers to employment between young adults with and without a psychotic diagnosis that involved testing each of the 23 items raises the issue of a type I error (alpha inflation). The small sample size further hinders the generalizability of our exploratory findings, though the results warrant replicating this analysis with a larger sample size and an appropriate post-hoc test.

5 |. CONCLUSION

IPS is a promising approach to helping young adults achieve competitive employment and pursue education as a way of changing their life trajectories. The current study identified a group of young adults with mental health conditions who primarily accessed employment services in standard IPS programmes in CMHCs. Future work should include a national survey assessing the number of young adults receiving IPS services, their demographic and clinical characteristics, and where they receive IPS.

APPENDIX A

TABLE A1.

Young adult characteristics by age group.

		Age
		16–17 (N = 26)	18–20 (N = 53)	21 or older (N = 32)	Significance
Gender	Female	11 (42.3%)	24 (45.3%)	12 (37.5%)	χ2 = 0.67, p = .72a
	Male	14 (53.8%)	26 (49.1%)	19 (59.4%)
	Other	1 (3.8%)	3 (5.7%)	1 (3.1%)
Race/ethnicity	White	13 (50.0%)	23 (43.4%)	15 (46.9%)	χ2 = 3.04, p = .55a
	Black	3 (11.5%)	17 (32.1%)	9 (28.1%)
	Hispanic	5 (19.2%)	10 (18.9%)	5 (15.6%)
	Other	5 (19.2%)	3 (5.7%)	3 (9.4%)
Education	Less than high school	25 (96.2%)	24 (45.3%)	9 (28.1%)	χ2 = 28.58, p < .001
	High school or higher	1 (3.8%)	29 (54.7%)	23 (71.9%)
Depression	Yes	15 (57.7%)	35 (66.0%)	16 (50.0%)	χ2 = 2.17, p = .34
	No	11 (42.3%)	18 (34.0%)	16 (50.0%)
Schizophrenia/bipolar	Yes	1 (3.8%)	12 (22.6%)	13 (40.6%)	χ2 = 10.85, p < .01
	No	25 (96.3%)	41 (77.4%)	19 (59.4%)
Anxiety	Yes	14 (53.8%)	22 (41.5%)	7 (21.9%)	χ2 = 6.51, p = .04
	No	12 (46.2%)	31 (58.5%)	25 (78.1%)
Housing	Living with family	19 (73.1%)	35 (66.0%)	17 (53.1%)	χ2 = 2.67, p = .26
	Living elsewhere	7 (26.9%)	18 (34.0%)	15 (46.9%)
Work history	Worked previously	5 (19.2%)	20 (37.7%)	13 (40.6%)	χ2 = 3.47, p = .18
	Never worked	21 (80.8%)	33 (62.3%)	19 (59.4%)
Arrested	Yes	5 (19.2%)	11 (20.8%)	10 (31.3%)	χ2 = 1.56, p = .46
	No	21 (80.8%)	42 (79.2%)	22 (68.8%)
Employment goal	Yes	23 (88.5%)	50 (94.3%)	29 (90.6%)	χ2 = 0.91, p = .64
	No	3 (11.5%)	3 (5.7%)	3 (9.4%)
Education goal	Yes	10 (38.5%)	23 (43.4%)	11 (34.4%)	χ2 = 0.70, p = .71
	No	16 (61.5%)	30 (56.6%)	21 (65.6%)

^aDoes not include ‘Other’.

APPENDIX B

TABLE B1.

Young adult characteristics by race/ethnicity.

		White (N = 51)	Black (N = 29)	Hispanic (N = 20)	Other (N = 11)	Significancea
Gender	Female	25 (49.0%)	10 (34.5%)	7 (35.0%)	5 (45.5%)	χ2 = 2.16, p = .34b
	Male	24 (47.1%)	18 (62.1%)	12 (60.0%)	5 (45.5%)
	Other	2 (3.9%)	1 (3.4%)	1 (5.0%)	1 (9.1%)
Age	16–17	13 (25.5%)	3 (10.3%)	5 (25.0%)	5 (45.5%)	χ2 = 3.04, p = .55
	18–20	23 (45.1%)	17 (58.6%)	10 (50.0%)	3 (27.3%)
	21 or older	15 (29.4%)	9 (31.0%)	5 (25.0%)	3 (27.3%)
Education	Less than high school	24 (27.1%)	17 (58.6%)	12 (60.0%)	5 (45.5%)	χ2 = 1.48, p = .48
	High school or higher	27 (52.9%)	12 (41.4%)	8 (40.0%)	6 (54.5%)
Depression	Yes	36 (70.6%)	10 (34.5%)	13 (65.0%)	7 (63.6%)	χ2 = 10.34, p = .01
	No	15 (29.4%)	19 (65.5%)	7 (35.0%)	4 (36.4%)
Schizophrenia/ Bipolar	Yes	8(15.7%)	13 (44.8%)	1 (5.0%)	4 (36.4%)	χ2 = 13.36, p < .01
	No	43 (84.3%)	16 (55.2%)	19 (95.0%)	7 (63.6%)
Anxiety	Yes	24 (27.1%)	7 (24.1%)	8 (40.0%)	4 (36.4%)	χ2 = 4.09, p = .13
	No	27 (52.9%)	22 (75.9%)	12 (60.0%)	7 (63.6%)
Housing	Living with family	34 (66.7%)	12 (41.4%)	17 (85.0%)	8 (72.7%)	χ2 = 10.26, p = .01
	Living elsewhere	17 (33.3%)	17 (58.6%)	3 (15.0%)	3 (27.3%)
Work history	Worked previously	19 (37.3%)	12 (41.4%)	4 (20.0%)	3 (27.3%)	χ2 = 2.61, p = .27
	Never worked	32 (62.7%)	17 (58.6%)	16 (80.0%)	8 (72.7%)
Arrested	Yes	9 (17.6%)	12 (41.4%)	3 (15.0%)	2 (18.2%)	χ2 = 6.82, p = .03
	No	42 (82.4%)	17 (58.6%)	17 (85.0%)	9 (81.8%)
Employment goal	Yes	48 (94.1%)	27 (93.1%)	19 (95.0%)	8 (72.7%)	χ2 = 0.08, p = .96
	No	3 (5.9%)	2 (6.9%)	1 (5.0%)	3 (27.3%)
Education goal	Yes	21 (41.2%)	15 (51.7%)	3 (15.0%)	5 (45.5%)	χ2 = 6.92, p = .03
	No	30 (58.8%)	14 (48.3%)	17 (85.0%)	6 (54.5%)

^aSignificance tests compare White, Black, and Hispanic groups.

^bDoes not include ‘Other’.

APPENDIX C

TABLE C1.

Perceived barriers to employment by psychiatric diagnosis.

Barriers to employment	Psychotic disorder (N = 52)	Non-psychotic disorder (N = 59)	Significance
Managing symptoms of mental health condition	46 (88.5%)	44 (74.6%)	χ2 = 3.47, p = .06
Trouble concentrating	39 (75.0%)	41 (69.5%)	χ2 = 0.42, p = .52
Managing time	38 (73.1%)	40 (67.8%)	χ2 = 0.37, p = .54
Difficulty finding a job	33 (63.5%)	38 (64.4%)	χ2 = 0.01, p = .92
Lack of self-confidence	26 (50.0%)	38 (64.4%)	χ2 = 2.35, p = .13
Personal problems	27(51.9%)	36 (61.0%)	χ2 = 0.93, p = .33
Transportation	29 (55.8%)	33 (55.9%)	χ2 = 0.00, p = .99
Difficulty keeping a job	33 (63.5%)	27 (45.8%)	χ2 = 3.49, p = .06
Being stigmatized because of mental health issues	31 (59.6%)	28 (47.5%)	χ2 = 1.64, p = .20
Work is too stressful	32 (61.5%)	26 (44.1%)	χ2 = 3.38, p = .07
Lack of education or skills	29 (55.8%)	29 (49.2%)	χ2 = 0.40, p = .49
Concerns about disclosing mental health issues	24 (46.2%)	26 (44.1%)	χ2 = 0.05, p = .83
Fear of failure	20 (38.5%)	30 (50.8%)	χ2 = 1.71, p = .19
Getting along with others	26 (50.0%)	23 (39.0%)	χ2 = 1.36, p = .24
Lack of support from family or friends	18 (34.6%)	25 (42.4%)	χ2 = 0.70, p = .40
Using alcohol or drugs	19 (36.5%)	10 (16.9%)	χ2 = 5.50, p = .02
Side-effects of medications	20 (38.5%)	9 (15.3%)	χ2 = 7.71, p = .01
Possibility for advancement	13 (25.0%)	15 (25.4%)	χ2 = 0.00, p = .96
Physical illness	14 (26.9%)	7(11.9%)	χ2 = 4.09, p = .04
Racial discrimination	10 (19.2%)	10 (16.9%)	χ2 = 0.10, p = .75
Taking care of family members	8 (15.4%)	10 (16.9%)	χ2 = 0.05, p = .82
Fired from a job	8 (15.4%)	6 (10.2%)	χ2 = 0.68, p = .41
Fear of losing benefits	9 (17.3%)	3 (5.1%)	χ2 = 4.28, p = .04
Lack of support from case managers	1 (1.9%)	2 (3.4%)	χ2 = 0.23, p = .63

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are currently unavailable publicly.