A Randomized Controlled Trial Evaluating the Effectiveness of Supported Employment Integrated in Primary Care
Abstract
Objectives:
Mental health issues can cause serious problems in occupational functioning, including higher rates of unemployment. Individual placement and support (IPS) is an evidence-based supported employment intervention that is typically integrated within a mental health setting; however, many primary care patients view referral to a mental health clinic as stigmatizing. Thus, this study examined whether delivery of IPS in a primary care setting provides an effective treatment option and avoids unnecessary delays in obtaining competitive employment.
Methods:
U.S. military veterans (N=119) who had a diagnosis in a broad range of nonpsychotic psychiatric disorders and who were receiving care from Veterans Health Administration (VHA) patient-aligned care teams were prospectively randomly assigned to IPS (N=58) or standard VHA non-IPS vocational rehabilitation (VR) (N=61). The primary outcome was achievement of steady worker status, defined as holding a competitive job for ≥6 months of the 12-month follow-up.
Results:
As hypothesized, a significantly greater proportion of IPS participants achieved steady worker status (45%), compared with VR participants (25%) (p=0.02; odds ratio=2.49, 95% confidence interval=1.14–5.43). On average, the IPS participants worked significantly more weeks (p=0.003) and earned significantly more income (p=0.033) from competitive jobs, compared with VR participants.
Conclusions:
The results provide supporting evidence for offering IPS within primary care with the aim of restoring meaningful and sustained competitive employment for veterans living with a mental disorder. Such modifications could improve veterans’ vocational outcomes, moving a significantly greater number of disabled veterans back to full and productive lives in the community.
HIGHLIGHTS
•
For patients in a primary care setting with a broad range of nonpsychotic psychiatric disorders, the individual placement and support (IPS) intervention led to steady work for nearly twice as many individuals, compared with Veterans Health Administration standard vocational services.
•
Integration of IPS within a primary care patient-aligned care team was feasible and may provide more immediate and less stigmatizing access to needed employment services.
•
Compared with Veterans Health Administration standard vocational services, IPS led to earlier job acquisition and yielded more weeks worked and income earned in competitive jobs.
Many veterans returning home from deployment face new battles, including unemployment, mental illness, addiction, and problems readjusting to community life (1, 2). Mental illness can substantially impede a person’s ability to sustain gainful employment (3–5). Job loss and financial hardship are associated with high rates of depression and posttraumatic stress disorder (PTSD) (6), and prolonged unemployment has detrimental effects on physical (7) and mental (8) health outcomes. The Veterans Health Administration (VHA) offers a range of vocational rehabilitation (VR) services, including prevocational counseling, sheltered transitional work (TW), and supported employment.
Individual placement and support (IPS) supported employment is a personalized service integrated within a treatment program that assists people with disabilities in obtaining and maintaining meaningful and well-matched competitive employment in the community. IPS is typically reserved for people living with serious mental illnesses, defined as schizophrenia, schizoaffective disorder, bipolar I disorder, and major depression with psychotic features (9, 10). IPS has demonstrated remarkably robust and consistent employment outcomes (11, 12), including for persons with disorders other than serious mental illnesses (13, 14), that are durable over many years (15).
In 2004, VHA implemented IPS for veterans diagnosed as having serious mental illness; the program provided limited access to veterans with other diagnoses (16, 17). Subsequent controlled trials confirmed the efficacy of IPS for veterans living with PTSD (18–20). Unfortunately, VR or IPS services have been used by fewer than expected veterans with PTSD, traumatic brain injury (TBI), depression, or substance use disorders (21–23). Disabled veterans often fear losing disability income benefits or experiencing worsening symptoms in the competitive work environment, which dampens their interest in pursuing employment and engaging in VR services (24). Additionally, service members transitioning to veteran status often delay seeking mental health treatment, because they fear not being able to obtain employment if psychiatric needs are disclosed—thus potentially delaying much-needed services.
To address these issues, this study sought to evaluate IPS when integrated within a primary care patient-aligned care team (PACT) and offered to veterans with a broad range of nonpsychotic mental disorders. The aim was to offer IPS employment services in a less stigmatizing setting, and it was hypothesized that this arrangement would yield more efficacious outcomes, compared with standard VHA VR services.
Methods
We conducted a randomized, controlled study at the Tuscaloosa Veterans Affairs Medical Center to evaluate the efficacy of IPS when delivered within a PACT, compared with standard non-IPS VR services. The protocol was approved by the Tuscaloosa Veterans Affairs Institutional Review Board, conforming to the Helsinki Declaration. Participants provided written informed consent. Safeguards included oversight by a Data and Safety Monitoring Board and approval of a National Institute of Mental Health Certificate of Confidentiality. Data collection occurred from July 8, 2015, through September 9, 2019. The methods and baseline characteristics of the sample have been described previously (25).
Eligibility Criteria and Randomization
Prospective participants were referred by their treatment team or were self-referred after seeing a flyer, letter, or brochure that advertised the study or after hearing about the study from another veteran or staff member. Eligible veterans were receiving primary care in a PACT clinic and were age 19 or older, currently diagnosed as having a nonpsychotic mental disorder (including substance use disorders and excluding psychotic, bipolar I, or cognitive disorders), otherwise eligible for VR services (i.e., medical clearance to return to work), unemployed or underemployed (i.e., working ≤20 hours per week in a job not in keeping with the veteran’s ability, aptitude, or skills), seeking competitive employment, not participating in another VR study, not actively suicidal or homicidal, and able to be followed up for 12 months. The initial criteria limited enrollment to unemployed veterans who served after September 11, 2001, but these criteria were relaxed to include the underemployed and veterans who served after 1990.
Eligible veterans were randomly assigned to either IPS or VR by using permuted varied-sized block design. Participants, providers, and outcomes assessors were openly aware of the postrandomization assignment.
Baseline Assessments
A clinically experienced research coordinator (C.M.B.) collected baseline demographic data and assessed the participant with the structured MINI International Neuropsychiatric Interview (MINI) for DSM-IV diagnoses (26) and the Ohio State University Traumatic Brain Injury (OSU-TBI) identification method (27, 28). The MINI and the OSU-TBI have previously well-established interrater reliability and predictive validity (MINI, all κ values ≥0.75 and excellent positive predictive value; OSU-TBI, all κ values ≥0.60; intraclass correlations >0.80; predictive validity, p<0.006).
Interventions
The IPS intervention (29) used a rapid job search to help veterans obtain a broad diversity of jobs, assisted participants to directly engage in a competitive job rather than in prevocational training or set-aside sheltered jobs, embraced the notion of “zero exclusion,” provided personalized information about government entitlements, utilized a network of employers developed by the IPS specialist on the basis of veterans’ interests, and provided individualized follow-along supports that were open ended. The IPS model was not deliberately changed; however, implementation in a new setting required some adaptations as to how the colocated IPS specialist connected with the providers. Instead of participating in regularly scheduled comprehensive treatment team meetings as would be typical of IPS for populations with serious mental illness, the IPS specialist participated in brief PACT huddles, gave updates at monthly staff meetings, and met individually with providers for brief discussions, as clinic schedule permitted.
Available as an existing service within the medical center on a consultation basis and not embedded in PACT, the standard VR (control group) offered compensated work therapy (CWT); TW assignments in a set-aside, minimum-wage, time-limited job; or CWT–community-based employment services (CBES), which involved limited community job search and placement. The TW work assignments were typically low-skill jobs within the local VHA setting and aimed to help participants gain work-hardening experiences before they obtained competitive work in the community. Standard VR specialists had minimal contact with PACT providers, and follow-up supports for those receiving VR services ended soon after the TW assignment was completed or their first job was acquired.
In keeping with the IPS practice, an IPS expert (R.T.) provided technical assistance to the IPS specialists via biweekly teleconferences and quarterly onsite meetings and conducted semiannual onsite IPS fidelity-monitoring visits with performance improvement feedback. Standard VR services were also evaluated with this scale to ensure that there was minimal overlap of VR and IPS. Day-to-day supervision of the IPS and VR specialists was provided by the onsite VHA vocational service manager.
Employment Outcomes
Postrandomization outcome assessments occurred monthly over 12 months. On the basis of a self-report employment calendar diary, pay and tax documents, and medical records, the research coordinator (C.M.B.) assessed each week for the following: work for pay (yes, no, or unknown), type of work (competitive, TW, or other), type of job(s), number of days and hours worked, gross income earned, and whether the job was new. “Competitive” was defined as a job for salary, wages, or commission in a setting that was not set aside (i.e., same job could be held by someone without a disability), not transient (i.e., yard work, babysitting, or day labor), and not military drill. “Steady worker” was defined as employed for ≥26 weeks of the 52-week follow-up, and “employed” was defined as working in a competitive job for ≥1 hour per day for ≥1 day per week. Weeks did not have to be consecutively worked to count toward the threshold of “steady worker.”
Statistical Analyses
The primary hypothesis was that IPS would result in a greater proportion of steady workers, compared with standard VR. The primary analysis used a logistic regression model to calculate an odds ratio (OR), controlling for baseline variables where relevant. The proportion of weeks worked in a competitive job unencumbered by a TW assignment was compared between groups, thus controlling for the time in a sheltered work setting. For purposes of outcome analyses, missing data for weeks worked were counted as “not worked.” However, missing data were minimal and did not have a statistically significant effect on our outcomes. The analysis adhered to the intent-to-treat principle, and participants were retained in the arm to which they were randomly assigned for the follow-up period, even if they discontinued the intervention or exited early. Secondary analyses of the employment outcomes included total time worked, income earned from competitive sources and all sources, and type of jobs held. These continuous employment outcomes were compared by using an independent, two-tailed t test, and categorical outcomes were analyzed by using the chi-square distribution or Fisher’s exact test when appropriate.
Results
Participants
Of 502 veterans considered for participation, 140 consented, of whom 19 were not eligible and two were randomly assigned in error, resulting in 119 who were randomly assigned (IPS, N=58; VR, N=61) and included in the analyses. (A CONSORT diagram listing reasons for exclusion or early exit is included in an online supplement to this article.) Seven (12%) VR participants exited early (weeks 0, 4, 17, 20 [two participants], 21, and 29), and 12 (21%) IPS participants exited early (weeks 0, 8 [two participants], 9, 12, 20 [3 participants], 21, 33, and 34 [two participants]), leaving 100 participants who completed the study (84%). Time in the study and amount of missing employment data did not statistically differ between groups and did not have a significant effect on outcomes.
No significant differences between groups were found in terms of baseline demographic or clinical characteristics (Table 1). Before the study, the mean±SD amount of time employed was 1.4±2.3 years, and 27% (N=32) of participants were unemployed for at least the past 3 years. In the revised eligibility criteria, the military service period was expanded beyond September 11, 2001; therefore, the service period was controlled for in a logistic regression, with steady worker as the dependent variable. Other covariates included intervention, gender, age, race-ethnicity, education, marital status, primary branch of service, combat zone experience, years of service, time in active duty, service-connected disability status, and total and medical percentage of service-connected disability at baseline. Study arm was significant (p<0.01) with or without controlling for these variables, and period of military service was not statistically significant.
| IPS (N=58) | VR (N=61) | Total (N=119) | ||||
|---|---|---|---|---|---|---|
| Characteristic | N | % | N | % | N | % |
| Gender | ||||||
| Male | 49 | 84 | 49 | 80 | 98 | 82 |
| Female | 9 | 16 | 12 | 20 | 21 | 18 |
| Race-ethnicity | ||||||
| White | 15 | 26 | 16 | 26 | 31 | 26 |
| Black | 42 | 72 | 45 | 74 | 87 | 73 |
| Other | 1 | 2 | 0 | — | 1 | 1 |
| Hispanic | 1 | 2 | 1 | 2 | 2 | 2 |
| Marital status | ||||||
| Married | 18 | 31 | 11 | 18 | 29 | 24 |
| Divorced | 19 | 33 | 19 | 31 | 38 | 32 |
| Never married | 15 | 26 | 17 | 28 | 32 | 27 |
| Separated, cohabitating, or widowed | 6 | 10 | 14 | 23 | 20 | 17 |
| Education | ||||||
| High school diploma or equivalent | 14 | 24 | 24 | 39 | 38 | 32 |
| Some college credit | 24 | 41 | 28 | 46 | 52 | 44 |
| Associate’s degree | 11 | 19 | 5 | 8 | 16 | 13 |
| Bachelor’s degree or higher | 9 | 16 | 4 | 7 | 13 | 11 |
| Branch of serviceb | ||||||
| Army | 34 | 59 | 54 | 89 | 88 | 74 |
| National Guard | 11 | 19 | 17 | 28 | 28 | 24 |
| Navy or Merchant Marines | 16 | 28 | 2 | 3 | 18 | 15 |
| Marine Corps | 6 | 10 | 1 | 2 | 7 | 6 |
| Air Force | 4 | 7 | 3 | 5 | 7 | 6 |
| Period of serviceb | ||||||
| Persian Gulf War (1990–1991) | 10 | 17 | 13 | 21 | 23 | 19 |
| 1991–2001 | 14 | 24 | 14 | 23 | 13 | 11 |
| After September 11, 2001 | 47 | 81 | 48 | 79 | 96 | 81 |
| Served in combat zone | 49 | 85 | 49 | 80 | 98 | 82 |
| Department of Veterans Affairs service-connected disability | 42 | 72 | 38 | 62 | 80 | 67 |
| Diagnosis on MINI International Neuropsychiatric Interviewc | ||||||
| Major depressive disorder (past) | 49 | 84 | 48 | 79 | 97 | 82 |
| Major depressive disorder (current) | 43 | 74 | 41 | 67 | 84 | 71 |
| Major depressive disorder (recurrent) | 25 | 43 | 22 | 36 | 47 | 39 |
| Agoraphobia (current) | 7 | 12 | 13 | 21 | 19 | 16 |
| Panic disorder (lifetime) | 24 | 41 | 27 | 44 | 51 | 43 |
| Social anxiety disorder (current) | 6 | 10 | 8 | 13 | 14 | 12 |
| Posttraumatic stress disorder (lifetime) | 46 | 79 | 44 | 72 | 90 | 76 |
| Alcohol use disorder (past year) | 29 | 50 | 26 | 43 | 55 | 46 |
| Nonalcohol substance abuse (past year) | 15 | 26 | 13 | 21 | 28 | 24 |
| Nonalcohol substance dependence (past year) | 14 | 24 | 14 | 23 | 28 | 24 |
| Length of past military service (M±SD years) | 9.6±7.4 | 9.7±7.3 | 9.7±7.3 | |||
| Length of active duty service (M±SD years) | 7.4±6.3 | 5.6±4.3 | 6.5±5.6 | |||
| Age | ||||||
| M±SD | 37.8±8.4 | 38.6±8.4 | 38.2±8.4 | |||
| Range | 23–65 | 24–58 | 23–65 | |||
a
No significant between-group difference was found for any variable.
b
Participants could have served in more than one branch and in more than one period.
c
More than one disorder could be present for one participant.
Employment
Although similar proportions of participants obtained a competitive job for ≥1 week (IPS, 62%, versus VR, 51%), a significantly greater proportion of IPS participants achieved steady worker status, compared with VR participants (45% versus 25%, p=0.02; OR=2.49, 95% confidence interval=1.14–5.43) (Table 2). Compared with the VR group, participants in the IPS group worked significantly more weeks (p=0.003), days (p=0.012), and hours (p=0.023), as did the subgroup of only those who held a competitive job during the 12-month follow-up. For weeks unencumbered by a time-limited TW assignment, the proportion of weeks worked competitively was significantly higher for the IPS group (40% of unencumbered weeks worked), compared with the VR group (25% of unencumbered weeks worked) (p=0.001). Although no between-group difference was found in number of weeks worked in a full-time job (≥30 hours per week), IPS participants were more likely than VR participants to hold a full-time job (p=0.006). The IPS participants gained competitive work in significantly less time (8.7 weeks), compared with the VR participants (17.1 weeks) (p=0.005) and were more likely to obtain a competitive job in the first 3 months (48% versus 28%, respectively, p=0.022). The IPS group was significantly more likely than the VR group to hold a competitive job at any point during the 12-month follow-up (p<0.001) (Figure 1).
| IPS (N=58) | VR (N=61) | ||||
|---|---|---|---|---|---|
| Variable | N | % | N | % | p |
| Steady workera | 26 | 45 | 15 | 25 | .020 |
| Income earned (M±SD $) | |||||
| Competitive sources | |||||
| All participants | 11,106±12,491 | 6,605±10,150 | .033 | ||
| All participants who held a competitive job | 18,945±10,792 | 13,813±10,809 | .062 | ||
| Steady workers | 21,948±10,757 | 21,816±8,416 | .967 | ||
| Transitional work sources (among those who held a transitional work assignment) | 0 | 4,295±2,967 | — | ||
| All sources (transitional work, competitive, and other) | 17,960±11,310 | 11,168±9,682 | .004 | ||
| Time worked in a competitive job for all participants (M±SD) | |||||
| Weeks | 23.00±22.09 | 12.00±17.24 | .003 | ||
| Days | 104.00±99.45 | 61.00±84.64 | .012 | ||
| Hours | 807.00±783.0 | 493.00±702.60 | .023 | ||
| Time worked for those who held a competitive job (M±SD) | |||||
| Weeks | 38.6±14.8 | 24.6±17.2 | .001 | ||
| Days | 177.0±61.9 | 123.3±82.8 | .004 | ||
| Hours | 1,000.0±704.7 | 1,374.0±508.9 | .019 | ||
| Weeks in full-time competitive employment | 41.9±15.2 | 39.0±16.0 | .596 | ||
| Proportion of weeks unencumbered by transitional work in which a competitive job was held | 40.0±13.7 | 25.0±16.7 | .001 | ||
| Job acquisition | |||||
| Time to first competitive job (M±SD weeks) | 8.7±8.8 | 17.1±14.2 | .005 | ||
| Obtained competitive job within | |||||
| 12 months | 36 | 62 | 31 | 51 | .261 |
| 9 months | 33 | 57 | 25 | 41 | .083 |
| 6 months | 32 | 55 | 23 | 38 | .056 |
| 3 months | 28 | 48 | 17 | 28 | .022 |
| Held full-time (≥ 30 hours per week) competitive job | 25 | 43 | 12 | 20 | .006 |
| Competitive jobs held per person employed | |||||
| 3 | 4 | 7 | 1 | 2 | .153 |
| 2 | 15 | 26 | 9 | 15 | .131 |
| 1 | 34 | 59 | 30 | 41 | .302 |
| Hollingshead job classificationb | |||||
| 1, 2, or 3 | 2 | 3 | 1 | 2 | .529 |
| 4 or 5 | 24 | 41 | 14 | 23 | .031 |
| 6 or 7 | 21 | 36 | 25 | 41 | .592 |
| Worked in a transitional work assignment | 0 | — | 31 | 51 | |
| Weeks in transitional work assignment (M±SD) | 0 | 13.4±10.6 | |||
a
Steady worker was defined as holding a competitive job for ≥6 months of the 12-month follow-up. Odds ratio=2.49, 95% confidence interval=1.14–5.43.
b
Hollingshead job classification: 1, higher executive, major professional, owner of large business; 2, manager of medium-sized business, lesser professions (i.e., nurse, optician, pharmacist, social worker, teacher); 3, administrative personnel, manager, minor professional, owner/proprietor of small business; 4, clerical, sales, technician, small business, bank teller, bookkeeper, clerk, timekeeper, secretary; 5, skilled labor (i.e., baker, barber, chef, electrician, fire fighter, machinist, mechanic, painter, repair person, tailor, welder, police, plumber); 6, semiskilled labor (i.e., hospital aide, bartender, bus driver, cutter, cook, garage guard, checker, waiter, spot welder, machine operator); 7, unskilled labor (i.e., attendant, janitor, construction helper, laborer, porter).
FIGURE 1. Percentage of employed participants over 12-month follow-up after random assignment to individual placement and support (IPS) or standard vocational rehabilitation (VR)a
aThe IPS group was significantly more likely than the VR group to hold a competitive job at any point during the 12-month follow-up (p<0.001).
When all participants were considered, the IPS group earned significantly more income, on average, from competitive sources than did the VR participants (p=0.033) (Table 2). When only participants who held a competitive job were considered, average earnings for the IPS group were higher than for the VR group, but the difference was not significant (p=0.062). No difference between groups was found in mean total income from competitive sources for steady workers. When all sources of income (competitive, TW, and other) were considered, the IPS group earned significantly more, on average, compared with the VR group ($17,960 versus $11,168, p=0.004). For those randomly assigned to VR, 51% (N=31) held a TW assignment for an average of 13.4 weeks and 23% (N=14) received CBES.
IPS Fidelity
Eight fidelity-monitoring visits were conducted with the 25-item Supported Employment Fidelity Monitoring Scale (30), in which a score of 125–115 is exemplary, 114–100 is good, 99–74 is fair, and ≤73 is not considered IPS. During the 4-year study, the IPS fidelity score averaged 90.0±8.7, which is at the high end of fair implementation. During the first year, fidelity achieved a score of 100 (good), but scores for fidelity fell to 98, 87, and 78 (low end of fair) during year 3, in part because of turnover in IPS specialists. After staffing stabilized in year 4, the IPS fidelity score improved to 98 (upper end of fair).
Discussion
This is the first study to evaluate IPS in a primary care setting for people with a broad range of nonpsychotic mental health conditions. IPS efficacy research has traditionally focused on individuals with serious mental illness within a mental health clinical setting, with recent expansion to other diagnoses, such as PTSD (19), TBI (31), and spinal cord injury (32). However, like civilians, veterans have a broad range of nonpsychotic mental conditions that can interfere with maintaining a job, and they often initially present for treatment in a primary care setting. To overcome stigma related to being referred to a mental health clinic and the delays in access due to the referral and intake process, we innovated delivery of IPS by providing services directly within primary care clinics and including veterans with a broad range of nonpsychotic mental disorders. This allowed for more efficient delivery of IPS services to an expanded population.
Compared with standard non-IPS VR services, IPS led to a substantially greater number of steady workers in a much shorter time frame, which resulted in more full-time jobs and higher income earned. Research has consistently shown that IPS is efficacious in helping participants return to work (11, 13) and improves psychosocial recovery (20). Our results are consistent with those of a multisite study of veterans with PTSD, which found that IPS participants were more rapidly able to obtain employment and were twice as likely to maintain steady work, compared with those receiving usual-care VR services (19). Meaningful employment embraced by strong clinical supports contributes to health and well-being and serves as an antidote to social exclusion (11).
Despite decades of public health education, the stigma associated with mental disabilities is particularly rampant in the workplace and causes delays in seeking mental health care, especially for members of the military (33). Therefore, the delivery of IPS within primary care has the potential to accelerate access to effective employment services in which people would otherwise not engage for fear of being stigmatized during the employment process. There is growing evidence that individuals with nonserious mental health issues are more likely to be treated in primary care than in a specialized mental health clinic (34). PACTs are already providing individualized, patient-centered, and evidence-based behavioral health services; thus, with more research to refine high-fidelity implementation, IPS could be added to the PACT armamentarium. This strategy is in line with the VHA Mission Act (35) and whole-health initiatives that aim to increase access to a full array of services for veterans both in the VHA and in community settings.
Implementation of IPS in a new setting required a steep curve of adaptation for our PACT providers, because some elements of IPS were not easy to implement, such as demonstrating an advanced understanding of the IPS principles by executive leadership and clinicians and integrating the IPS specialist into a fast-paced PACT. As reflected in the fair fidelity scores, the PACT team struggled with the core principle of IPS integration as it related to shared decision making, given the limited time that the primary care providers had available. Unlike other initiatives for persons with serious mental illness and for those with PTSD, where the IPS specialists work intensely with a mental health team, the PACT model did not allow for ease of access to the interdisciplinary team in which employment considerations were included within the overall treatment plan. This dynamic resulted in a greater burden on the IPS specialist to address case management and social work issues rather than focused employment services. Our IPS team was small, and the onsite supervisor was transitioning full retirement, which negatively affected the IPS fidelity score—i.e., higher points are given for one or more specialists and for a highly engaged VR supervisor. The IPS staff turnover required training and reintegration of new specialists, which likely affected implementation from the standpoint of sustained strategic job and employer development, consistency in establishing community connections, and maintenance of the network of employment resources. The IPS specialists had difficulty at times promoting a diversity of jobs within the community and leveraging the vocational assessment as a dynamic tool for veteran engagement and strategic job development. Ideally, study teams should allow time for IPS to reach a more mature state of implementation prior to starting randomization—although research budgets do not always allow for this luxury of time and readiness. However, within the constraints of any budget, the leadership and clinicians should be fully engaged and strongly advocate for high-fidelity IPS implementation in a new setting to achieve maximal outcomes.
The findings of this study are timely, in that the COVID-19 pandemic has put steady employment in jeopardy, as evidenced by a 12% increase in veteran unemployment (36, 37). Large-scale national crises that lead to economic recessions and increased unemployment are associated with adversities, such as increased risk of overdose and suicide (38, 39). Strengthening community partnerships and increasing services available to veterans outside traditional settings have the potential to improve timely delivery of efficacious services, mitigate negative health outcomes, and improve psychosocial functioning.
The study had strengths and limitations. The strengths included its randomized approach, 12-month follow-up period, broad eligibility criteria, and novel setting. The generalizability of the findings is limited by a single-site design in a sample of mostly male veterans; thus, results would require replication from a multisite, more diverse, and larger sample of participants to better understand the efficacy of IPS in PACT. Limitations of the study include the fair fidelity to the IPS model, which may have weakened employment outcomes in the intervention group. In addition, the TW assignments may have delayed entry into community-based competitive work, which may have weakened employment outcomes in the control group. However, we confirmed that, despite fluctuations in fidelity of IPS implementation and the fact that half the control group participated in a time-limited TW assignment, IPS resulted in a significantly greater proportion of weeks employed (40%), compared with VR (25%), when the analysis controlled for the number of weeks encumbered by a TW assignment.
Conclusions
This trial demonstrated significant efficacy of IPS supported employment when delivered in a primary care setting, compared with stepwise standard VHA VR, for veterans living with a broad range of nonpsychotic mental illnesses. The results support increasing access to IPS for veterans with common diagnoses of mental health conditions through the primary care environment to help restore them to well-matched and meaningful competitive employment and a recalibrated trajectory of recovery. When translated to a larger population, these improved employment outcomes will have beneficial economic impact on revenues and health costs to society at large. On the basis of these results, which add to the growth and sustainment of IPS (40), a multisite study is warranted as the next step for future research, as are studies to evaluate health service utilization and other health and wellness outcomes associated with the IPS intervention.
Supplementary Material
File (appi.ps.202000926.ds001.pdf)
- View/Download
- 90.26 KB
References
1.
Sayer NA, Noorbaloochi S, Frazier P, et al: Reintegration problems and treatment interests among Iraq and Afghanistan combat veterans receiving VA medical care. Psychiatr Serv 2010; 61:589–597
2.
Kleykamp M: Unemployment, earnings and enrollment among post 9/11 veterans. Soc Sci Res 2013; 42:836–851
3.
Adler DA, Possemato K, Mavandadi S, et al: Psychiatric status and work performance of veterans of Operations Enduring Freedom and Iraqi Freedom. Psychiatr Serv 2011; 62:39–46
4.
Erbes CR, Kaler ME, Schult T, et al: Mental health diagnosis and occupational functioning in National Guard/Reserve veterans returning from Iraq. J Rehabil Res Dev 2011; 48:1159–1170
5.
Schnurr PP, Lunney CA: Work-related quality of life and posttraumatic stress disorder symptoms among female veterans. Womens Health Issues 2011; 21(suppl):S169–S175
6.
Riviere LA, Kendall-Robbins A, McGurk D, et al: Coming home may hurt: risk factors for mental ill health in US reservists after deployment in Iraq. Br J Psychiatry 2011; 198:136–142
7.
Tran TV, Canfield J, Chan K: The association between unemployment status and physical health among veterans and civilians in the United States. Soc Work Health Care 2016; 55:720–731
8.
Batic-Mujanovic O, Poric S, Pranjic N, et al: Influence of unemployment on mental health of the working age population. Mater Sociomed 2017; 29:92–96
9.
Drake RE, Bond GR, Goldman HH, et al. Individual Placement and Support services boost employment for people with serious mental illnesses, but funding is lacking. Health Aff 2016; 35:1098–1105
10.
Brinchmann B, Widding-Havneraas T, Modini M, et al: A meta-regression of the impact of policy on the efficacy of individual placement and support. Acta Psychiatr Scand 2020; 141:206–220
11.
Bond GR, Drake RE, Becker DR: An update on Individual Placement and Support. World Psychiatry 2020; 19:390–391
12.
Mueser KT, Drake RE, Bond GR: Recent advances in supported employment for people with serious mental illness. Curr Opin Psychiatry 2016; 29:196–201
13.
Bond GR, Drake RE, Pogue JA: Expanding individual placement and support to populations with conditions and disorders other than serious mental illness. Psychiatr Serv 2019; 70:488–498
14.
LePage JP, Crawford AM, Cipher DJ, et al: Blending traditional vocational services and individualized placement and support for formerly incarcerated veterans. Psychiatr Serv 2020; 71:816–823
15.
Bush PW, Drake RE, Xie H, et al: The long-term impact of employment on mental health service use and costs for persons with severe mental illness. Psychiatr Serv 2009; 60:1024–1031
16.
Resnick SG, Rosenheck R: Dissemination of supported employment in Department of Veterans Affairs. J Rehabil Res Dev 2007; 44:867–877
17.
Resnick SG, Rosenheck RA: Scaling up the dissemination of evidence-based mental health practice to large systems and long-term time frames. Psychiatr Serv 2009; 60:682–685
18.
Davis LL, Leon AC, Toscano R, et al: A randomized controlled trial of supported employment among veterans with posttraumatic stress disorder. Psychiatr Serv 2012; 63:464–470
19.
Davis LL, Kyriakides TC, Suris AM, et al: Effect of evidence-based supported employment vs transitional work on achieving steady work among veterans with posttraumatic stress disorder: a randomized clinical trial. JAMA Psychiatry 2018; 75:316–324
20.
Mueller L, Wolfe WR, Neylan TC, et al: Positive impact of IPS supported employment on PTSD-related occupational-psychosocial functional outcomes: results from a VA randomized-controlled trial. Psychiatr Rehabil J 2019; 42:246–256
21.
Twamley EW, Baker DG, Norman SB, et al: Veterans Health Administration vocational services for Operation Iraqi Freedom/Operation Enduring Freedom Veterans with mental health conditions. J Rehabil Res Dev 2013; 50:663–670
22.
Resnick SG, Rosenheck RA: Posttraumatic stress disorder and employment in veterans participating in Veterans Health Administration Compensated Work Therapy. J Rehabil Res Dev 2008; 45:427–435
23.
Abraham KM, Ganoczy D, Yosef M, et al: Receipt of employment services among Veterans Health Administration users with psychiatric diagnoses. J Rehabil Res Dev 2014; 51:401–414
24.
Drebing CE, Mueller L, Van Ormer EA, et al: Pathways to vocational services: factors affecting entry by veterans enrolled in Veterans Health Administration mental health services. Psychol Serv 2012; 9:49–63
25.
Davis LL, Blansett CM, Mumba MN, et al: The methods and baseline characteristics of a VA randomized controlled study evaluating supported employment provided in primary care patient aligned care teams. BMC Med Res Methodol 2020; 20:33
26.
Sheehan DV, Lecrubier Y, Sheehan KH, et al: The Mini-International Neuropsychiatric Interview (MINI): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59(suppl 20):22–33
27.
Bogner J, Corrigan JD: Reliability and predictive validity of the Ohio State University TBI identification method with prisoners. J Head Trauma Rehabil 2009; 24:279–291
28.
Corrigan JD, Bogner J: Initial reliability and validity of the Ohio State University TBI identification method. J Head Trauma Rehabil 2007; 22:318–329
29.
Swanson SJ, Becker DR: Supported Employment: Applying the Individual Placement and Support (IPS) Model to Help Clients Compete in the Workforce. Center City, MN, Dartmouth PRC–Hazelden, 2015
30.
Bond GR, Peterson AE, Becker DR, et al: Validation of the Revised Individual Placement and Support Fidelity Scale (IPS-25). Psychiatr Serv 2012; 63:758–763
31.
Wehman P, Kregel J, Keyser-Marcus L, et al: Supported employment for persons with traumatic brain injury: a preliminary investigation of long-term follow-up costs and program efficiency. Arch Phys Med Rehabil 2003; 84:192–196
32.
Ottomanelli L, Goetz LL, Barnett SD, et al: Individual Placement and Support in spinal cord injury: a longitudinal observation study of employment outcomes. Arch Phys Med Rehabil 2017; 98:1567–1575.e1
33.
Iversen AC, van Staden L, Hughes JH, et al: The stigma of mental health problems and other barriers to care in the UK Armed Forces. BMC Health Serv Res 2011; 11:31
34.
Mental Health in Primary Care: Illusion or Inclusion? Geneva, World Health Organization, 2018. https://www.who.int/docs/default-source/primary-health-care-conference/mental-health.pdf?sfvrsn=8c4621d2_2
35.
VA MISSION Act Strengthens VA care. Washington, DC, Veterans Health Administration, 2020. https://missionact.va.gov/
36.
Labor Force Statistics From the Current Population Survey. Washington, DC, US Bureau of Labor Statistics, 2020. https://data.bls.gov/timeseries/LNS14000000
37.
Shane L: Veteran unemployment up to nearly 12 percent amid coronavirus crisis. Mil Times (Epub May 8, 2020). https://www.militarytimes.com/education-transition/2020/05/08/veteran-unemployment-up-to-nearly-12-percent-amid-coronavirus-crisis/
38.
Parmar D, Stavropoulou C, Ioannidis JP: Health outcomes during the 2008 financial crisis in Europe: systematic literature review. BMJ 2016; 354:i4588
39.
Kerr K, Romaniuk M, McLeay S, et al: Increased risk of attempted suicide in Australian veterans is associated with total and permanent incapacitation, unemployment and posttraumatic stress disorder severity. Aust N Z J Psychiatry 2018; 52:552–560
40.
Drake RE, Becker DR, Bond GR: Growth and sustainment of individual placement and support. Psychiatr Serv 2020; 71:1075–1077
Information & Authors
Information
Published In
History
Received: 21 December 2020
Revision received: 29 June 2021
Revision received: 3 August 2021
Accepted: 13 August 2021
Published online: 15 September 2021
Published in print: June 2022
Keywords
Authors
Competing Interests
Dr. Davis reports receipt of consulting fees from Janssen, Lundbeck, Otsuka, and Signant Health and research funding or materials from Alkermes, Aptinyx, Social Finance, and Tonix. Dr. Pilkinton reports receipt of research funding or materials from Alkermes, Aptinyx, Avanir, and Tonix. The other authors report no financial relationships with commercial interests.
Competing Interests
The authors are grateful for technical support and assistance from J. Douglas Bell, M. Jerome Ennis, M.A., V.R.S., Michael A. Lavender, M.A., V.R.S., Jessie McAlpine, M.Div., Amy J. Mitchell, L.M.S.W., Andrew Oakland, Ph.D., Retired LTC Kenneth Phillips, Allen Riley, M.S., C.R.C., Katherine Thompson, M.S., Mary A. Watson, Ed.D., and John Merkle, F.A.C.H.E., VHA-C.M.
Funding Information
This work was supported by grant 1I01RX001489-01A2 from the VA Rehabilitation Research and Development Service (VA project ID, D1489-I). This trial is registered at www.clinicaltrials.gov (NCT02400736).
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBLogin options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).