Abstract
BACKGROUND:
The Service member Transitional Advanced Rehabilitation Program (STAR) is a novel model designed to promote interdisciplinary collaboration while delivering comprehensive physical, psychosocial and vocational rehabilitation services. The STAR program was established as a pilot program in 2011 and is now approved for permanent continuation. It is housed in the Polytrauma Transitional Rehabilitation Center at Hunter Holmes McGuire VA Medical Center in Richmond, Virginia.
OBJECTIVE:
The present study introduces the development, implementation, participant characteristics and five-year program outcomes of the STAR program.
METHODS:
A paired sample t-test was employed to compare pre and posttest functioning in the following areas: physical, mental and emotional, vocational, and community integration. Mixed analysis of variance was further employed to determine the impact of demographic variables such as age, gender, race/ethnicity and marriage status on recovery.
RESULTS:
Statistically significant improvements in physical, mental and emotional, and vocational functioning was found in participants after completion of the STAR program. Regarding demographic variables, only marital status appeared to impact PTSD recovery.
CONCLUSIONS:
The STAR program’s vocational rehabilitation and community reintegration services, as well as efforts to improve physical and psychological functioning, have demonstrated positive outcomes. Overall, the development and implementation of the program has been highly successful.
Introduction
Military conflicts in Iraq and Afghanistan since 2001 are resulting in a large number of U.S. Service members sustaining traumatic brain injuries (TBI) and posttraumatic stress disorder diagnoses (PTSD; Burke, Olney, & Degeneffe, 2009); the two most common factors that interfere with successfully transitioning back into the workforce (Gaiter, 2015). According to the Bureau of Labor Statistics, the unemployment rate for Iraq and Afghanistan War Veterans increased from 7.3 percent in 2008 to 11.5 percent in 2011. The Veterans Employment Challenges survey conducted by Prudential (2012), found a 30.2 percent unemployment rate in the Veteran population, compared to 16.1 percent for civilians in the same age range. Furthermore, 98 percent of survey participants reported at least one service-related challenge to entering or re-entering the work force; two-thirds reported three or more obstacles to employment. In 2013, the rate was 5.5% compared to nonveterans (6.4%), however the unemployment rate for OEF/OIF Veterans is still much higher (7.3%; Roberts, 2016, p. 121).
Issues that Service members and Veterans struggle with are broad-ranged and include a host of health-related, physical, and psychological disorders. For instance, the Institute of Medicine (IOM; 2012) cited three categories of conditions that frequently co-occur with PTSD in military populations: psychiatric (e.g., depression and substance use disorders); medical (e.g., chronic pain, TBI, and spinal cord injury); and psychosocial (e.g., relationship problems, difficulties in social settings, intimate partner violence, child maltreatment, unemployment or lack of employment, homelessness, and incarceration). This wide range of co-occurring conditions indicates the need for population-specific comprehensive rehabilitation and support services.
In October 2010, a joint Veterans Affairs/Department of Defense (VA/DoD) Task Force met at Walter Reed Army Medical Center (WRAMC) to address coordinated efforts on behalf of wounded warriors. The findings of this task force identified a need for improved comprehensive transitional rehabilitation support and recommended a program be developed to address all aspects of the physical, psychological and vocational needs for injured and ill Service members. The Department of Veterans Affairs responded to this recommendation by launching the Service member Transitional Advanced Rehabilitation (STAR) program at Hunter Holmes McGuire VA Medical Center in Richmond, Virginia.
The purpose of this study was two-fold. First, the present study provided a description of the development and implementation of the STAR program as a novel model for the provision of vocational rehabilitation services within a comprehensive, residential rehabilitation setting. The second purpose was to report the characteristics of program participants and a summary of STAR program outcomes including, sample characteristics that differentiate STAR program outcomes.
Program description
The STAR program was established as a pilot program designed specifically for injured Service members and Veterans who require a comprehensive residential rehabilitation program in order to successfully return to work and integrate into the community. Following the disability conceptualization model of the International Classification of Function, Disability and Health (ICF), the STAR program includes various services to optimize health, mental, and vocational functioning, as well as opportunities for enhanced transition in returning to work and community.
As part of the mission to provide a comprehensive transition program for injured Service members and Veterans that would facilitate successful return to military, federal or civilian employment, the STAR program was established in the Polytrauma Transitional Rehabilitation Center on the Hunter Holmes McGuire VA Medical Center campus in Richmond, Virginia. This ten-bed residential program is co-located with the Polytrauma Transitional Rehabilitation Program and the Polytrauma Assistive Technology Center of Excellence.
The STAR program provides a comprehensive treatment setting for the acquisition of new knowledge, enhancement of resiliency skills, improvement in physical functioning, and increasing the likelihood of transfer of training to the community setting. These services are provided at a single site with an interdisciplinary team of physicians, psychologists, physical and occupational therapists, speech and language pathologists, recreation therapists, social workers, and vocational rehabilitation counselors to enhance the efficiency and standardization of treatments.
The Vocational Rehabilitation component of the program includes a comprehensive vocational intake assessment following by a graduated program of worksite skills assessments, work readiness and work hardening training, work re-entry, and support for reintegrating the individual into an appropriate occupational role. The comprehensive functional capacity evaluation and work hardening activities facilitate successful return to work. Integration with the Disability Evaluation System allows for more rapid assessment and transition through the medical board process if required as part of the military transition process. Using an interdisciplinary team approach to provide high-quality care, the STAR Program provides state-of-the-art medical and vocational rehabilitation in a residential setting; Table 1 provides a brief summary of each service.
STAR program medical and vocational services
STAR program medical and vocational services
Participants
Injured or ill Service members and Veterans with associated functional limitations were recruited and referred from various sources (see Table 2). Among the total of 102 injured Service members and Veterans who were admitted to the program, the majority were males (95%). Regarding age, the majority were in their 20 s (43%), followed by 30 s (32%) and 40 s (20%) ranges. Most participants were either single (46%) or married (30%). Racial/ethnic groups of participants included: Caucasian (64%), African-American (26%), Hispanic/Latino (9%), and Asian (1%). At the time of the participants’ admission to the program, the majority were active duty Service members (72%), with 40% of participants reporting that their injuries were sustained during deployment related activities in either Iraq or Afghanistan. The majority of participants stayed two to four months (68%) with 100 days being the average length of stay (median = 93 days). Descriptive characteristics of participants are summarized in Table 2.
Characteristics of STAR program participants. NA – Not Applicable
Characteristics of STAR program participants. NA – Not Applicable
[1]Only for the 75 participants who were active-duty at the time of admission were included in this category; [2]Three participants who did not completed program were excluded from the total; [3]Participants who did not complete or are still in the program were not included in the total.
The STAR program employed various instruments to assess patient functioning and program outcomes in a variety of domains. Mobility functioning was measured using the 2 Minute Walk Test (2MWT), Dynamic Gait Index (DGI), Berg Balance Scale (BBS), and 10 Meter Walk Test (10MWT). To measure psychological functioning, the PTSD Checklist – Specific (PCL-S), Beck Depression Inventory II (BDI-II), and World Health Organization Quality of Life – Brief Version (WHOQOL-BREF) were used. Regarding vocational functioning, the newly developed Work Perception Scale (WPS) and Job Readiness Scale (JRS), were used. Finally, the Canadian Occupational Performance Measure (COMP) and Community Reintegration for Service members (CRIS) were used to measure community integration. At the time of program completion, participants also completed the Vocational Preparation Effectiveness Survey (VPES), a self-reported descriptive assessment on the effectiveness of vocational rehabilitation services provided in the STAR program. For WPS and JRS, preliminary analysis (i.e., item-total correlation) was conducted and deviated items were removed from statistical analyses to improve measurement reliability.
2 Minute Walk Test (2MWT)
The 2MWT assesses the distance walked in two minutes and has been validated for people with various disabilities including cardiac surgery, lower extremity amputation, multiple sclerosis, neurologic impairment such as stroke, head injury and tumor, older adults/geriatric, and spinal cord injury (Connelly, Thomas, Cliffe, Perry, & Smith, 2009). Test-retest reliability tested across various sample populations revealed coefficients over 0.80 (e.g., 0.83 for lower extremity amputation, 0.97 for neurologic impairment, 0.95 for older adult, 0.98 for stroke), indicating good reliability. There is no cut-off score established for 2MWT.
Dynamic Gait Index (DGI)
The DGI (Shumway-Cook & Woollacott, 2001) assesses gait, balance, and fall risk to measure not only steady walking, but also walking in more challenging situations. The DGI uses a four-point ordinal scale (0 = severe impairment to 3 = normal), and obtains scores ranging from 0 to 24; lower scores indicate greater impairment. Previous researches (Jonsdottir & Cattaneo, 2007; Herman, Inbar-Borovsky, Brozgol, Giladi, & Hausdorff, 2009) indicates that the DGI is a reliable and valid measure; it has been shown to yield ratios of subject variability to total variability with excellent interrater reliability (0.96) and test-retest reliability (0.98) when rated by physical therapists (Shumway-Cook, Gruber, Baldwin, & Liao, 1997).
Berg Balance Scale (BBS)
The BBS (Berg, Wood-Dauphinee, Williams, & Gayton, 1989) is comprised of 14 balance-related tasks, ranging from standing up from a sitting position to standing on one foot. The BBS uses a five-point rating scale (0 = unable to 4 = independent) designed to measure static and dynamic standing balance. The total score range is 0 to 56, and the sum of all obtained scores on each task is used as the final measure, with higher total scores indicating better balance. The BBS is a popular measure used with traumatic brain injuries, community dwelling geriatric patients, multiple sclerosis, orthopedic surgery recovery, osteoarthritis, Parkinson’s disease, spinal cord injury, stroke and vestibular dysfunction. The internal consistency of the BBS reported in existing literature ranges from 0.85 to 0.98 (Berg, Wood-Dauphinee, & Williams, 1995; Brusse, Zimdars, Zalewski, & Steffen, 2005). Additionally, the BBS has demonstrated excellent test re-test reliability (0.94) and has been recommended for use in rehabilitation and wellness programs for determining whether change during or after intervention is clinically significant (Steffen & Seney, 2008).
Timed Ten-Meter Walk Test (10MWT)
The unique aspect of the 10MWT (Collen, Wade, & Bradshaw, 1991) compared to other mobility measures is that it assesses gait speed; an essential mobility function necessary for safely navigating within the community (e.g., crossing a street before the light changes). The 10MWT has shown excellent test-retest reliability, interrater reliability, and construct and predictive validity across various medical populations including: healthy adults (Bohannon, 1997; Watson, 2002) and patients with hip fractures (Hollman et al., 2008), spinal cord injuries (Bowden & Berman, 2007; Burns, Delparte, Patrick, Marino, & Ditunno, 2011; Lam, Noonan, Eng, & SCIRE Research Team, 2007), stroke (Collen, Wade, & Bradshaw, 1990; Flansbjer, Holmback, Downham, Pattern, & Lexell, 2005), multiple sclerosis (Paltamaa, Sarasoja, Leskinen, Wikstrom, & Malkia, 2007), Parkinson’s disease (Steffen & Seney, 2008), and TBI (van Loo, Moseley, Bosman, de Bie, & Hassett, 2004). Overall changes in total score between pre- and post-test assessment were used in the present study.
PTSD Checklist – Specific (PCL-S)
Using a five-point Likert scale (1 = not at all, 5 = extremely), the PCL-S (Conybeare, Behar, Solomon, Newman, & Borkovec, 2012) is a 17-item self-report measure designed to assess the extent and degree of DSM-IV symptoms of PTSD experienced in the past month. Generally, the cut-point scores of 36–44 are used for VA primary care or specialized medical clinics (e.g., TBI or pain population), 45–50 for VA mental health clinics, and 30–35 for civilian primary care. In terms of measuring pre and post interventions, 5–10 point changes are considered reliable (i.e., not due to chance) and a 10-point threshold is used to determine clinical meaningfulness (Monson et al., 2008). The PCL-S has demonstrated sufficient internal consistency (α= 0.94; Blanchard, Jones-Alexander, Buckley, & Forneris, 1996; Ruggiero, Del Ben, Scotti, & Rabalais, 2003) as well as test-retest reliability (r = 0.88; Ruggiero et al., 2003), and has been used in both general psychiatric (e.g., Sampson, Kinderman, Watt, & Sembi, 2003) and military (e.g., Barrett et al., 2002; Sutker, Corrigan, Sundgaard-Riise, Uddo, & Allain, 2002) research.
Beck Depression Inventory II (BDI-II)
The BDI-II (Beck, Steer, & Brown, 1996) includes 21 items measuring behavioral and affective symptoms of depression using a four-point rating scale (0 = symptom not present to 3 = symptom very intense). Scores range from 0 to 63, with higher scores indicating greater depressive symptom severity. The BDI-II has been widely used to assess depression in a variety of populations, including Veterans (e.g., Chard, Schumm, Owns, & Cottingham, 2010; Nelligan et al., 2008), and demonstrates adequate convergent validity, discriminant validity, test-retest reliability (r = 0.93), and good internal consistency (α< 0.92; Beck, Steer, & Brown, 1996).
World Health Organization Quality of Life - Brief (WHOQOL-BREF)
The WHOQOL-BREF (WHOQOL Group, 1998) includes 26 items that measure quality of life (qol), which is defined by the WHO as “a person’s perception of his/her position in life within the context of the culture and value systems in which he/she lives” (WHOQOL Group, 1994, p. 28). Respondents rate the intensity, frequency, or evaluation of the selected attributes of QOL during the previous two weeks on a five-point Likert scale. The WHOQOL-BREF produces scores across four domains: physical, psychological, social relationship and environment. Each domain has demonstrated sufficient internal consistency (Cronbach’s alphas ranged from 0.66 to 0.84), test-retest reliability (ranging from 0.66 to 0.87), as well as discriminant and construct validity in both healthy populations and various patient groups, including patients with spinal cord injuries (Jang, Hsieh, Wang, & Wu, 2004; Skevington, Lotfy, & O’Connell, 2004; WHOQOL Group, 1998). Because the STAR program is designed to address overall quality of life rather than specific area of quality of life, the total score was used in this study. This is supported by the WHOQOL Group’s (1998) indication that all four domains should be considered when evaluating overall quality of life.
Work Perception Scale (WPS)
The WPS is composed of 10 items, measuring participants’ work-related motivation, engagement, and performance. The WPS uses a four-point rating scale (1 = strongly disagree to 4 = strongly agree) to examine the degree in which one values work, as well as one’s beliefs about injured Service members’ work ability and performance. Scores on the WPS range from 10 to 40, with higher scores indicating more positive perceptions of the value of working. WPS was developed for the present study; therefore, an item-total correlation analysis was completed first to refine items and improve reliability.
Job Readiness Scale (JRS)
The JRS is a 29-item measure that collects information regarding job seeking skills (14 items), employment attributes (8 items), and career preparation (7 items). The JRS uses a four-point rating scale (1 = strongly disagree to 4 = strongly agree) and yields scores ranging from 10 to 40, with higher scores indicating greater perceived job readiness. Items were adapted from existing validated measures such as the Job Search Knowledge Scale (Liptak, 2009), Job-Seeking Self-Efficacy scale (Barlow, Wright, & Cullen, 2002), and Self-efficacy of Job-seeking Skills scale (Hergenrather, Rhodes, Turner, & Barlow, 2008).
The WPS and JRS are self-report measures that were developed specifically for STAR program participants, and thus construct validity information for these instruments is not currently available; however, as stated with the WPS, preliminary item-total correlation analysis was first employed to refine items and improve overall reliability. In the present study, internal consistency alphas in a given sample examined in both pre and posttest were 0.860 and 0.963 for the WPS and JRS, respectively.
Vocational Preparation Effectiveness Survey (VPES)
Using 14 items and a five-point rating scale, the VPES is designed to examine the participant’s perceived effectiveness of the STAR program’s various vocational services (e.g., development of a resume, participation in an independent job search, etc.). As the VPES is not designed to measure a psychological construct, no reliability and validation information is available for this measure.
Canadian Occupational Performance Measure (COPM)
The COPM (Law et al., 1998) collects information on self-care (personal care, functional mobility, community management), productivity (paid/unpaid work, household management, play/school), and leisure (quiet recreation, active recreation, socialization) to identify the participants five most important problems. These problems are rated in terms of performance and satisfaction. The COPM has demonstrated strong evidence of divergent and convergent validity (Dedding, Cardol, Eyssen, & Beelen, 2004) as well as test-retest reliability (Law et al., 1998). It has been used with various general and specialty medical populations, including pain (Carpenter, Baker, & Tyldesley, 2001), stroke (Cup, op Reimer, Thijssen, & van Kuyk-Minis, 2003), and neurorehabilitation (Bodiam, 1999) patients.
Community Reintegration for Service members (CRIS)
Measuring an injured individual’s adjustment to life at home and in the community through the assessment of participation in life roles, the CRIS (Resnik, Plow, & Jette, 2009) yields three subscale domains: Extent of Participation (EP; 50 items); Perceived Limitations (PL; 54 items); and Satisfaction with Participation (SP; 47 items). The EP subscale assesses how often an individual experiences or participates in specific activities using seven-point scales indicating the number of times per week or other frequency of occurrence (not at all, very often, etc.). The PL subscale uses two different seven-point response scales. The first indicates the magnitude of perceived limitations and the second asks the degree to which one agrees or disagrees with specific statements about the amount of limitation that they have. The SP subscale asks about satisfaction with different aspects of participation using a seven-point response scale (1 = terrible to 7 = very happy). The CRIS has demonstrated strong concurrent and known-group validity, as well as excellent test-retest reliability, with intra-class correlation coefficients (ICC) of 0.91, 0.90, and 0.90, respectively (Resnik, Gray, & Borgia, 2011).
Exit satisfaction survey
The exit satisfactions survey used in this study utilizes a five-point Likert scale (1 = strongly disagree to 5 = strongly agree) and is comprised of 12 questions (e.g., I received accurate information about the program, I was involved in establishing my treatment goals, the team focus on goals important to me and my family). This survey was used to evaluate participant’s perceived satisfaction of the STAR program.
Results
Paired-sample t-tests were used in the present study to examine pre- and post-intervention differences. Concerning WPS and JRS, preliminary item-total correlation analyses were conducted as these were newly developed instruments specifically for STAR program participants and thus no prior reliability information exists. Results are summarized below.
Mobility functioning
Improvements were observed in the areas of overall walking distance and functional mobility (i.e., ability to navigate community safely). The results of paired-sample t-tests indicated that pre- and post-treatment differences across all measures were statistically significant (t (71) = –0.6.115, p < 0.0001 for 2MWT; t(80) = –6.279, p < 0.0001 for DGI; t(57) = –4.416, p < 0.0001 for BBS; and t(65) = –4.4108, p < 0.0001 for 10MWT). Results are summarized in Table 3.
Pre- and posttest score comparison
Pre- and posttest score comparison
[1]walking distance/second, *lower scores indicate improved symptoms, **p = 0.01. CRIS EP = CRIS Extent of Participation; CRIS PL = CRIS Perceived Limitation; CRIS SP = CRIS Satisfaction with Participation.
STAR program participants’ mental health functioning was assessed using the PCL-S, BDI-II and WHOQOL-BREF. Statistically, overall improvement was shown on all variables as indicated in Table 3 below. Particularly for the PCL-S and BDI-II, clinical improvement was further examined by using the interpretation manuals’ guidelines regarding interpreting symptom change. The PCL-S views a 5–10 point reduction in scores as reliable, and a 10-point threshold is used to determine clinically significant improvement. However, overall mean differences for the PCL-S were 4.26, indicating statistically significant reductions in PTSD symptom severity but not reaching clinically significant reductions. Regarding changes in BDI-II scores, as indicated in the frequency table, a large proportion of participants showed improvement in their depressive symptoms. Statistically significant improvement was also identified in WHOQOL after the STAR program. Results are summarized in Table 3.
Vocational functioning
Tools used to measure the impact and effectiveness of the STAR program’s vocational training included WPS, JRS and the VPES. While the WPS and JRS were designed to measure participants’ work perception and readiness to work, VPES is descriptive in nature and designed to document participants’ subjective rating on the effectiveness of various aspects of STAR’s vocational training such as portfolio and resume development. Therefore, paired sample t-tests were only used to compare pre and post-intervention differences in the WPS and JRS, while a descriptive analysis was employed to examine participants’ subjective rating on the effectiveness of various vocational training of VPES.
As indicated above, prior to the analyses for WPS and JRS, item-total correlation analyses were completed. Item-total correlation analysis identified items that significantly deviate from other items in terms of measuring the purported constructs. Concerning general decision guidelines, any item with a negative item-total correlation coefficient is removed unless a case for theoretical justification can be made. Because there were two data sets (i.e., pre- & posttest) and the numbers of each data set were not identical due to drop-out, the data set with the highest sample size and fewest missing cases was chosen for the analysis.
Regarding the WPS, there were no items with item-total correlations below 0.3, and thus no effect on the alpha coefficient (α= 0.908) was identified; however, seven items on the JRS had item-total correlations below 0.3. Once these items were removed (JRS items 2, 3, 6, 13, 14, 23 and 24), the original alpha coefficient (α= 0.860) was improved (α= 0.881). These seven items were also removed from further analyses to compare pre- and posttest outcomes. Both were designed as single-factor measures, thus total scores were used for the statistical analyses; results are summarized in Table 3.
Concerning the VPES, utilizing five-point rating scale with an additional column in which one may answer “not applicable,” participants were asked to rate the effectiveness of the STAR program in assisting them to accomplish thirteen aspects of job preparation, and one additional item asking overall perception on the effectiveness of STAR. As the VPES is not a psychological instrument to measure a certain construct, descriptive analysis was performed. Most participants rated the service effectiveness on each of the 13 areas positively and the results are summarized in Table 4 above.
Summary of VPES
Summary of VPES
The COMP and CRIS were used to examine functional improvement from a community integration perspective. COMP consists of two subscales (performance and satisfaction) and CRIS includes three subscales (extent of participation, perceived limitation, and satisfaction with participation). The results are summarized in Table 3.
Pre- and posttest difference depending on sample characteristics
Pre- and posttest differences were paired and compared based on four sample characteristics (i.e., gender, marital status, age, and race/ethnicity). Mixed analysis of variance was employed and no significant effects on gender, age, and race/ethnicity were found; however, marital status appeared to have an impact on recovery for both PTSD, F(3,61) = 5.539, p = 0.002, partial eta2 = 0.214, and depression, F(3,65) = 6.134, p = 0.001, partial eta2 = 0.221.
Exit survey
Information regarding participant satisfaction on various aspects of the STAR program was collected utilizing twelve questions. Results indicated a high level of satisfaction from majority of participants and are summarized in Table 5 below.
Exit survey
Exit survey
As previously stated, the purpose of this study was two-fold. First, the present study provided a description of an innovative pilot program that represents a unique model of care incorporating vocational rehabilitation services into a comprehensive residential rehabilitation program. The second purpose was to report the characteristics of program participants and a summary of STAR program outcomes including sample characteristics that differentiate STAR program outcomes.
Overall program
STAR has served patients with varying severities of TBI as well as patients with other brain conditions such as stroke, anoxia, and encephalitis. The program has also been able to successfully treat patients with a host of other neurologic and orthopedic conditions. Over time, it has evolved in order to meet the changing needs of the populations being served. This evolution in primary population served includes a transition from combat-related injuries to non-combat related injuries and illnesses. The program has also experienced a shift from injuries occurring overseas to more injuries occurring stateside. Despite this evolution, STAR has maintained high levels of success in terms of both patient satisfaction and achieving a 92% program completion rate.
It is important to emphasize that the STAR program represents a new model of care in several respects. The residential setting, where patients are actually living at the program in order to receive intensive vocational and rehabilitation services, is unique. Additionally, the program integrates vocational rehabilitation services directly with physical and cognitive rehabilitation services; Vocational Rehabilitation Counselors are part of the core interdisciplinary team. Another feature is that the program fully integrates mental health service providers as part of the treatment team. These services have also been enhanced over time in order to meet population needs. With successful implementation results, the STAR program, which was initially established as a pilot program, was approved for permanent continuation.
Physical functioning
Program participants had diverse and multiple injuries occurring both overseas, in conflict zones, as well as stateside. Of these conditions, TBI was the most frequently occurring primary injury (80.8%), and PTSD the most frequently occurring secondary injury (37.5%). Concerning mobility, improvement was identified in all mobility measures including 2MWT, DGI, BBS and 10MWT. The goal of rehabilitation is to facilitate optimal functioning so that the individual may return to his/her community and experience opportunities for participation and inclusion. Following an injury, the recovery process spans time and consists of not only medical support, but psychological support to process the injury and develop positive coping mechanisms towards adjustment and returning to work. This interplay between physical and psychological healing is paramount, as research suggests injured workers need to be equipped both physically and psychologically prior to entering a job search (Berglind & Gerner, 2002).
Psychosocial and mental health functioning
Program participants also experienced decreased levels of PTSD and depression symptom severity along with increased levels of QOL. This outcome is consistent with previous literature that demonstrated higher levels of QOL resulted in reduced mental health challenges such as depression (Hegelson, Reynolds, & Tomich, 2006). This result is not surprising as previous studies suggest that with increased self-efficacy and functioning, and decreased anxiety and pain, QOL also improves (Kerns, Otis, Rosenberg, & Reid, 2003; Lorig & Holman, 2003; Mackenzie & Bosse, 2006).
Although statistically significant reductions were found in participants’ PTSD symptoms from pre- to post-program participation, these reductions were not clinically significant (i.e., they did not reach over a 10-point reduction which has been indicated as necessary for clinically meaningful symptom reduction; Monson et al., 2008) but participants’ mean score remained within the suggested range for a probable diagnosis of PTSD. In light of this finding, several interpretations are offered below.
First, research indicates that between one third and one half of patients receiving empirically supported treatments for PTSD do not fully respond to treatment, at least on some measures (Schottenbauer, Glass, Arnkoff, Tendick, & Gray, 2008). As mentioned earlier, participants’ average length of stay in the residential program was between two to four months, which may not have allowed enough time for Service members and Veterans to fully engage in intensive PTSD treatment; it also may not have allowed enough time for clinical effects of pharmacological treatment to begin.
Second, the measure utilized to assess PTSD (i.e., PCL-S) asks respondents to report the severity of symptoms they have experienced within the past month, which may have also influenced the smaller reduction in PTSD symptom scores from pre- to post-program participation given their short duration of stay. Also, prior trauma was not assessed, which has been found to influence the severity of PTSD and responses to subsequent traumas (Ozer, Best, Lipsey, & Weiss, 2003), and therefore treatment outcomes. Likewise, research reviewing the relationships between trauma type and treatment outcomes indicates that treatment for combat-related PTSD has shown the lowest effect sizes (Bradley, Greene, Russ, Dutra, & Westen, 2005; Ford, Fisher, & Larson, 1997), which may possibly be attributed to the heightened severity of pathology in Veterans seeking care at VA hospitals or malingering due to secondary gain (i.e., VA disability benefits may depend on remaining symptomatic; Bradley et al., 2005). It is likely that this population has dealt with multiple traumas given the complexity of the acquired disabilities reported, and that longer-term, intensive trauma treatment would be most beneficial. Researchers have indicated that patients with TBI and co-occurring PTSD require a greater amount of time to benefit from PTSD interventions (Rettman, Sigford, & Friedman, 2009), which is particularly relevant for the present sample given the high rates of TBI.
Third, many of the patients who complete the STAR program still face the challenging processes of going through necessary medical board procedures and fully transitioning out of the military. This can be very stressful because as it will undoubtedly have a significant impact on future benefits and compensation that they will receive following military transition. Regardless, these findings suggest that the STAR program may benefit from placing a greater emphasis on PTSD treatment, considering expanding treatment opportunities to include coping skills classes and gold standard treatments in both individual and group therapy modalities (i.e., prolonged exposure and cognitive processing therapy). Mental health professionals working with the STAR program interdisciplinary team should monitor and ensure participants’ continuity of care with the VA’s PTSD clinic following their completion of the STAR program.
Vocational rehabilitation and community integration
The vast majority (95%) of participants were in their in their 20’s-50’s placing them within the working age range. The Vocational Rehabilitation Services provided in the STAR program have proven to be highly effective as indicated by the WPS, JRS, and VPES results. Work perceptions encompass an individual’s meaning of work as well as his/her motivations for doing so. Work perceptions influence an individual’s beliefs about the outcomes of the return to work process and contribute to whether an individual chooses to return or withdraw from the workforce following an injury or illness (Sampere et al., 2012). As such, participants’ work perceptions significantly improved throughout their participation in the STAR rehabilitation program, potentially increasing their motivation for returning to work and/or securing new employment.
In addition to enhancing work perceptions, the STAR program provides vocational rehabilitation to develop job readiness skills. Job readiness describes an individual’s skillset and preparation for engaging in the job search process. Edward, Li-Tsang, Lam, and Chan (2006) described the development of work readiness as the primary requirement for job placement. Furthermore, the participation in work readiness programs has been found to increase worker motivation and employment readiness. During their participation in the program, patients increased in where they were on the job readiness continuum. These findings are consistent with the participants’ responses to the VPES showing that individuals who participated in the STAR program felt better prepared to enter the workforce in various domains (see Table 4).
Providing Vocational Rehabilitation Services (VRS) in this new model presented a number of challenges as well as significant advantages. The program has found that VRS can be greatly enhanced through an interdisciplinary approach where informational interviews and worksite assessments are performed with a collaborative team approach. In the STAR program, physical therapists, occupational therapists, and speech-language pathologists often directly participate in these assessments with the vocational rehabilitation counselors. This approach has been able to more rapidly address essential worksite accommodations as well as facilitate more specific work hardening and training activities. This type of collaboration was essential in order to compress the entire vocational rehabilitation process into a 3-month time period.
Community integration was also assessed on a continuum of vocational rehabilitation as fostering work perception, job readiness and vocational preparation to facilitate patient employability upon their return to the community. However, the results were mixed in that participants showed improvement on both of COMP Performance and Satisfaction indicating improvement in occupational performance and participants’ level of satisfaction on their performance. Concerning CRIS, only the Satisfaction with Participation domain showed improvement, and no improvement was observed in the areas of Extent of Participation and Perceived Limitation. This result indicates that the STAR program improves participants’ ability to re-join the community, but is not indicative of ensuring that it is actually happening (i.e., no improvement in the area of extent of participation and perceived limitation). This can be explained in terms of sample characteristics of STAR program participants. In the STAR program, most of the individuals completing the program are Active Duty Service member (74%) who had limited prior work experience and education outside of the military. Thus, they do not transition directly into employment. This is typically because they are either still active duty at the time of program discharge and/or their transition out of the military is delayed by the medical board. Because of this, the desired focus and primary outcome for the program is vocational preparation rather than employment. However, community integration services can be further tailored to better address the needs of participants with the future growth of the program.
Marriage factor in mental health functioning
Relationship status was also found to influence indices of mental health. When compared to single or divorced participants, individuals who were married had significantly fewer PTSD and depressive symptoms. Social support is paramount for psychological adaptation to disability and is associated with perceptions of growth (Livneh & Martz, 2016). Furthermore, research indicates that social support following trauma exposure is a significant predictor of PTSD (Ozer et al., 2003) as well as posttraumatic growth in Veteran populations (Tsai, El-Gabalawy, Sledge, Southwick, & Pietrzak, 2015).
Although we do not know how supportive these relationships are, it is presumed participants of this study have probably benefitted from a primary relationship in which psychological and physical support may have been available during the recovery process. Family members, particularly spouses, are often involved in the processing of a traumatic event and deeper interpersonal relationships could be a result of these efforts (Barskova & Oesterreich, 2009; Hawley et al., 2017), which may further mitigate distress and augment QOL. Particularly considering that most Veterans have expressed interest in greater family involvement in their treatment (Batten et al., 2009), findings in the present study highlight the utility of involving Service members’ identified support networks in the rehabilitation process for improving mental health outcomes.
Limitation of study
There are several limitations to consider in reviewing the research findings. First, although the predominance of males in the present study is characteristic of the military population, findings related to outcomes are limited in their generalizability to the female Veteran population. For instance, the prevalence of combat-related PTSD is greater in males than females, although female Veterans are significantly more likely to be diagnosed with depression (Maguen et al., 2010) and are much more likely to experience sexual assault during their military service (IOM, 2012).
Second, beyond injury type and severity, psychological factors can either facilitate or inhibit the return to work process (Gustafsson et al., 2013). For instance, individual perceptions of work and perceived readiness to return are among the key factors associated with employment following an injury or illness (Reiso et al., 2003; Shultz et al., 2004). Both of these aspects significantly increased upon patients’ completion of the STAR program.
Of note, however, findings related to participants’ improved work perceptions and perceived job readiness may be limited in the present study; this is primarily due to the instruments used to assess these constructs (i.e., the WPS and JRS), as they were developed specifically for the STAR program and therefore relatively less psychometric information of those measures exists. Nonetheless, STAR program outcomes suggest that the services were beneficial to Service members and Veterans with regards to vocational outcomes. Work perception and job readiness are complex constructs to measure, thus not many validated tools exist; however, strong results of item-total correlation analyses of WPS and JRS indicated that further development and validation of these tools would deliver an important contribution to the field.
Third, this is a study of the outcomes of a pilot rehabilitation program. Future studies would benefit from the inclusion of a comparison group to assess group differences and overall effectiveness of the STAR program, beyond that of treatment-as-usual, across a multitude of outcomes. The current study also has not collected follow-up data on participants’ job obtainment and retention following discharge from the program. As employment stability is a key factor in determining the long-term efficacy of vocational interventions (Kreutzer et al., 2003), it would be beneficial to conduct further research on job acquisition and the job stability of those exiting the program into paid employment.
Conclusion
The STAR program represents a novel model for the provision of vocational rehabilitation services within a comprehensive, residential rehabilitation setting for injured Service members and Veterans. The development and implementation of the STAR program has been highly successful and has demonstrated positive outcomes in relation to the program’s vocational rehabilitation and community reintegration services as well as in the domains of physical and psychological functioning. The program has evolved over time to meet the changing needs of the population served and it is likely that this continued evolution will be required in the future.
Conflict of interest
None to report.