skip to page content
Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

Health Systems Research

Go to the VA ORD website
Go to the QUERI website

IIR 15-342 – HSR Study

IIR 15-342
Health Care Utilization of Veterans Receiving Supported Employment
Lori Lynne Davis, MD AB
Tuscaloosa VA Medical Center, Tuscaloosa, AL
Tuscaloosa, AL
Funding Period: October 2017 - September 2020
Of nearly 2 million Iraq and Afghanistan war (OEF/OIF/OND) Veterans, 61% are enrolled in VA healthcare, 58% are treated for a mental disorder of whom 55% (378,993) have posttraumatic stress disorder (PTSD), one-third have a service-connect disability, and 8% live in poverty. The 4 million Veterans who served between 1990 and 2001 have similar characteristics. PTSD and other health conditions can negatively impact their post-deployment readjustment. Although overall unemployment rate is 7% for all Veterans, 21% those age 18-24 are unemployed and 50% of Veterans are not participating in the labor pool. Employment is a key indicator of successful reintegration, but OEF/OIF/OND Veterans are less likely to be employed than their civilian counterparts. PTSD is a significant indicator of unemployment. African American and female Veterans are significantly less likely to return to work compared to non-Latino White and male Veterans, respectively. Difficulties at work and job loss are barriers to readjusting to civilian life. Among the 5% most costly patients receiving VA care in fiscal year 2010, 17% had a PTSD diagnosis. Given their young age, the OEF/OIF/OND and Gulf War cohort will use VA services for the next four or more decades. Unemployment and disability in OEF/OIF/OND Veterans have the potential to increase VA service utilization and cost, imposing a long-term strain on VA resources.

Interventions that assist Veterans in recovery and return to employment are expanding, yet little is known about how these evidence-based interventions impact the health service utilization patterns among Veterans with PTSD. This proposal seeks to fill this gap in order to inform policy and practice in the delivery of the most effective vocational rehabilitation for Veterans with PTSD. Accurate measurement of costs and benefits of different treatment options, including improvements in functioning and productivity as it relates to work and income, are needed so that fiscally responsible investments in care can be made.

This study will evaluate the health service utilization and costs among unemployed Veterans with a diagnosis of PTSD who have been randomized to receive either Individual Placement and Support (IPS) or Transitional Work (TW). IPS is a person-centered, patient-driven evidence-based model of supported employment. VA Cooperative Studies Program (CSP #589) is conducted a multisite randomized controlled trial to compare the effectiveness of IPS to TW in a large, geographically diverse group of unemployed Veterans with PTSD (n=541). The CSP study entitled "Veterans Individual Placement and Support Towards Advancing Recovery (VIP-STAR) compared these two approaches in terms of employment, PTSD symptoms, and quality of life. Using VA archival data, this HS&D study provides a unique opportunity to learn more about the impact of IPS on health care utilization and costs.

Aim 1: Compare the VA health service utilization in Veterans with PTSD randomized to IPS versus TW, primarily focusing on inpatient usage.

Hypothesis 1:Veterans with PTSD randomized to IPS will use significantly fewer inpatient days compared to those randomized to TW during the first 18 months after randomization. Secondary analyses will look at an extended follow-up time (up to 3.5 years), and consider other types of care.

Aim 2. Compare the costs for high intensity (i.e. inpatient, emergency room, domiciliary, residential) services and outpatient services among Veterans with PTSD randomized to IPS versus TW during the first 18 months after randomization. Secondary analyses will look at an extended follow-up time (up to 3.5 years).

Hypothesis 2a. The cost of high intensity services will be lower for the IPS than the TW group.

Hypothesis 2b. The cost of outpatient services will be higher for the IPS than the TW group.

Aim 3. Assess the cost-effectiveness of IPS relative to TW among Veterans with PTSD during the first 18 months after randomization. Secondary analyses will consider an extended follow-up time (up to 3.5 years).

This proposed HSR&D IIR study will create a new health services utilization dataset from VA administrative databases to be integrated with the clinical data prospectively collected through VA CSP #589 VIP-STAR-- a multisite, prospective, randomized controlled clinical trial comparing employment outcomes of IPS versus TW in unemployed Veterans with PTSD. This proposed HSR&D study will leverage the prospectively collected clinical data from VIP-STAR by merging it with administratively-collected healthcare utilization and cost data, thus creating a detailed database on clinical outcomes, employment, and VA healthcare utilization and costs.

The main dependent variable of interest for Aim 1 is the number of inpatient days. For Aim 1, we will compare the number of inpatient days (all types) between the IPS and TW groups. The analyses will primarily focus on the period of 18 months from baseline, and secondary analyses will evaluate all available data from the point of randomization to the point of data-pull (i.e. varied time as much as 3.5 years, however randomization between groups was evenly distributed over the enrollment period). Inpatient stays will be identified using the Inpatient Encounters table within the CDW. For each inpatient admission the number of inpatient days will be calculated as the difference between the admission and discharge dates. For sensitivity analysis, descriptive purposes, and as a way to compare our results to the other studies, we will also collect and categorize number of inpatient stays and types of inpatient (medical, surgical, psychiatric) between groups.

As a supplement to hypothesis 1, we will evaluate other services, including the number of days of VA residential and domiciliary care, the number of ED visits, and the number of outpatient visits. We will examine specific types of outpatient services based on clinic stop codes. We will determine which outpatient visits were primary care, mental health services, or other specialty services.

The dependent variables for Aim 2 are high intensity (i.e. inpatient, emergency room, domiciliary, residential) and outpatient service costs. The analyses will primarily focus on the period of 18 months from baseline, and as with Aim 1, secondary analyses will evaluate all available data at the point of data-pull (i.e. as much as 3.5 years). We will also examine total costs of care and specific components of total cost. Total costs will include the costs of the intervention (IPS or TW) and costs of other services (i.e., inpatient, residential, domiciliary care, ED, and outpatient services). We predict that cost for high intensity services (inpatient, domiciliary, residential, and ED) will be less for IPS participants than for TW participants, and cost for outpatient services will be higher for IPS due to the nature of IPS encouraging the appropriate use of PTSD, mental health, and addictions treatment.

The dependent variables for Aim 3 are measures of the intervention's cost-effectiveness derived from total costs, days employed, and change in Frisch Quality of Life. For Aim 3, we will examine the cost-effectiveness of IPS versus TW. To conduct the cost-effectiveness analysis, we will examine the total cost for each program divided by two measures of effectiveness. We will assess effectiveness as the change in the Frisch QOLI from baseline (at randomization) to end point (18 months). Additionally, we will assess effectiveness as the number of days employed. We will measure costs in our cost-effectiveness analysis as total costs at 18 months, including costs of the interventions and healthcare costs. The analyses will primarily focus on the period of 18 months from baseline, and secondary analyses will evaluate all available data from the point of randomization to the point of data-pull (i.e. varied time as much as 3.5 years, however randomization between groups was evenly distributed over the enrollment period).

The study is being initiated in 2018. There are no findings to date.

Reintegration for recently deployed Veterans requires finding mainstream competitive employment that provides the Veteran with identity, structure, income, daily activity, friends, and other benefits. Without employment, many Veterans become enveloped by preoccupation with symptoms, social isolation, economic instability, familial disintegration, substance abuse, legal problems, homelessness, and wayward lifestyles. Recovery includes employment and should be a primary treatment aim. Recovery is a multi-faceted process in which people with illnesses or disabilities move beyond preoccupation with illness, become hopeful about the future, and pursue their own journeys and goals. Functional and social recovery involves maintaining valued societal roles and responsibilities, including employment, education, stable housing, and meaningful relationships with family and community.

Reduced reliance on acute care health systems and more appropriate use of outpatient services are perhaps two of the most important potential impacts of IPS at a systems level. In previous studies focused primarily on patients with SMI, persons in IPS had reduced hospital length of stays and increased use of appropriate outpatient services. Individuals who gain steady employment report increased self-esteem, decreased psychiatric symptoms, reduced social disability, and overall greater quality of life. For those who become steady workers, mental health treatment costs decline dramatically over the long term after adjusting for morbidity/needs. In essence, employment helps people to escape from the disabled patient role, to experience a sense of purpose and accomplishment, and to establish a new identity as a working, contributing citizen.

In conjunction with VIP-STAR, this study advances HSR&D's priority area of improving mental and behavioral health interventions and contributes to the systematic investigation on how IPS impacts acute care health care service utilization within VHA for Veterans with PTSD. In collaboration with the VA Center for Innovation and VA Office of Mental Health Operations' Therapeutic Supported Employment Services (TSES), this proposal provides an unprecedented opportunity for CSP and HSR&D-linked studies of IPS for Veterans with PTSD and informs TSES' decision-making on future expansion and implementation of IPS to a broader Veteran population.

External Links for this Project

NIH Reporter

Grant Number: I01HX002082-01A1

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


Journal Articles

  1. Jordan N, Stroupe KT, Richman J, Pogoda TK, Cao L, Kertesz S, Kyriakides TC, Bond GR, Davis LL. Comparing Service Use and Costs of Individual Placement and Support With Usual Vocational Services for Veterans With PTSD. Psychiatric services (Washington, D.C.). 2022 Oct 1; 73(10):1109-1116. [view]
  2. Stroupe KT, Jordan N, Richman J, Bond GR, Pogoda TK, Cao L, Kertesz SG, Kyriakides TC, Davis LL. Cost-Effectiveness of Individual Placement and Support Compared to Transitional Work Program for Veterans with Post-traumatic Stress Disorder. Administration and policy in mental health. 2022 May 1; 49(3):429-439. [view]

DRA: Mental, Cognitive and Behavioral Disorders
DRE: Treatment - Observational, TRL - Applied/Translational
Keywords: PTSD, Reintegration Post-Deployment, Utilization
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.