Association Between Traumatic Brain Injury (TBI)–Related Clinical Burden and Posttraumatic Stress Disorder Symptom Severity Among Special Operations Forces (SOF): Cross-sectional and Longitudinal Inferences Authors: Jolie N Haun, Justin McDaniel, Risa Nakase-Richardson, Tali Schneider, Julie McMahon-Grenz, Sharon Barton, Roberto Sandoval, Karen M Skop, Clara Dismuke-Greer, Jerome Sabangan, Kimberly Samson, Daniel W Klyce, Yvonne Friedman, Leah R Gause, Shannon R Miles, Kevin Wells, Mary Jo Pugh Published: 2026 ABSTRACT AbstractObjectivesTo examine how traumatic brain injury (TBI)–related clinical burden (eg, neurobehavioral symptoms, TBI-related lifestyle/participation deficits) is associated with posttraumatic stress disorder (PTSD) symptom trajectories during inpatient rehabilitation.DesignCross-sectional and longitudinal cohort study.SettingVeterans Health Administration Intensive Evaluation and Treatment Program for TBI distributed across 5 sites in the United States.ParticipantsParticipants were male Special Operations Forces with an average age of 42 years (SD, 5.4) and a history of mild to moderate TBI (n=105).InterventionsA longitudinal cohort evaluation was conducted to analyze baseline, discharge, and 6-month follow-up data from Special Operations Forces enrolled in the Veterans Health Administration’s Intensive Evaluation and Treatment Program for TBI, which is a variable-length program with interdisciplinary rehabilitation opportunities.Main Outcome MeasuresTBI-related clinical burden was operationalized using neurobehavioral symptom burden and functional participation impairment, and PTSD symptom severity was assessed using the PTSD Checklist for the Diagnostic and Statistical Manual of Mental Disorders, fifth edition. Multivariable linear regression and linear mixed-effects models were used to analyze the data.ResultsGreater neurobehavioral symptom burden (b=1.00; P<.001) and participation-related impairment (b=2.60; P<.001) were independently associated with higher PTSD symptom severity at baseline. Longitudinal analyses indicated that functional participation (b=–0.35; P=.01), but not neurobehavioral symptoms (b=–0.06; P=.14), moderated PTSD symptom trajectories, with greater participation scores predicting steeper reductions in PTSD symptoms over time.ConclusionsThese findings suggest that functional participation may represent an important measure of TBI-related clinical burden that is associated with PTSD symptom trajectories in a TBI rehabilitation program.Keywords: PTSD, Rehabilitation, Special operations forces, Traumatic brain injury Traumatic brain injury (TBI)–related clinical burden (eg, which may also be referred to as postconcussive symptoms) and posttraumatic stress disorder (PTSD) frequently co-occur in U.S. service members who served in recent military operations because of widespread blast and combat exposures.1 Service members with concussion or TBI often report elevated psychological and somatic symptoms, with clinically significant PTSD symptoms commonly present.2,3 The aforementioned comorbidity presents a substantial challenge for military and veteran health systems, as it is associated with poorer overall health, reduced quality of life, and increased healthcare utilization.4 Special Operations Forces (SOF) represent a distinct subgroup of the military population with higher cumulative blast exposure, greater operational demands, and more persistent neurobehavioral symptoms than conventional forces.1 Although the U.S. Department of Veterans Affairs’ (VA) Intensive Evaluation and Treatment Program (IETP) has published analyses of TBI-related outcomes after an inpatient rehabilitation program,5 it is still unclear if TBI-related burden (eg, postconcussive symptoms) influences PTSD recovery trajectories during rehabilitation for SOF, specifically. Participation limitations, including impairments in social, occupational, and daily functioning, may represent more clinically meaningful indicators of recovery than symptom counts alone. A key complication of previous literature is the substantial overlap in symptom presentation between TBI-related burden (eg, postconcussive symptoms) and PTSD, including sleep disturbance, irritability, concentration difficulties, fatigue, and hyperarousal.6 This overlap complicates diagnosis, symptom attribution, and treatment planning. In addition, inconsistency in how TBI-related burden (ie, postconcussive symptoms) is defined, such as variability in measuring loss of consciousness, alteration of consciousness, posttraumatic amnesia, and cumulative exposure, has contributed to mixed findings across studies.6,7 Despite these challenges, multiple sources of evidence support a strong association between TBI-related clinical burden (eg, postconcussive symptoms) and PTSD severity. For instance, studies of soldiers returning from Iraq show that TBI-related clinical burden is strongly linked to PTSD and depression symptoms, although psychological factors may account for much of the relationship between TBI and physical symptom reporting.3 Epidemiologic studies further identify PTSD symptoms as a primary correlate of persistent postconcussive symptoms.8 Mechanistic and cognitive neuroscience perspectives support this relationship. TBI may disrupt prefrontal-limbic circuitry and neural systems involved in emotion regulation, threat processing, and executive functioning, all of which may increase vulnerability to stress-related psychopathology.9,10 From a stress-diathesis framework, TBI-related clinical burden (ie, postconcussive symptoms) may act as a vulnerability factor that amplifies the effects of trauma exposure on PTSD development and persistence.10 Overall, the literature suggests a bidirectional relationship between TBI-related clinical burden and PTSD symptoms that warrants further investigation.11,12 Longitudinal research examining these relationships remains limited. Existing studies demonstrated that TBI-related clinical burden (eg, postconcussive symptoms) and PTSD symptoms during deployment are associated with long-term psychosocial outcomes, and that PTSD may mediate the relationship between injury severity and functional outcomes among veterans.4,13 However, few studies have evaluated whether TBI-related clinical burden is associated with PTSD symptom trajectories during rehabilitation. This gap in the literature is particularly important in clinical contexts, where treatment guidelines focus on individualized care based on symptom profiles rather than injury history alone.14 Therefore, the present study investigates both cross-sectional and longitudinal associations between TBI-related clinical burden (postconcussive symptoms) and PTSD symptoms among SOF personnel undergoing inpatient rehabilitation. Specifically, we examine baseline associations, symptom changes across key rehabilitation time points (baseline, discharge, follow-up), and whether TBI-related clinical burden moderates PTSD symptom trajectories over time.1,3,12 Methods Design This study used a prospective longitudinal evaluation design to examine cross-sectional and longitudinal associations between TBI-related clinical burden (postconcussive symptoms) and PTSD symptom outcomes among SOF enrolled in the Veterans Health Administration’s IETP for TBI. The protocol was conducted as part of a larger enterprise-wide, participatory, concurrent mixed-methods quality improvement initiative focused on evaluating clinical outcomes in an intensive interdisciplinary inpatient rehabilitation setting. Regulatory oversight was provided by the James A. Haley Veterans’ Hospital Research and Development Committee, which determined that the project met criteria for non-research quality improvement activities. All data were deidentified before analysis. Setting and participants Participants in this study were SOF with a well-documented history of concussive events who were referred to the VA’s IETP, which includes 5 sites throughout the U.S. Military bases, medical treatment facilities, and VA medical providers referred the participants to the IETP.5,15,16 Program eligibility criteria were established at the site level but consistently required documentation of one or more TBI events and persistent postinjury symptoms extending beyond the acute phase of recovery. The program was designed to address the cumulative effects of multiple concussive injuries. Between July 1, 2022, and May 30, 2024, patients with mild-to-moderate TBI participated in the quality improvement evaluation if they were among the first 50 individuals admitted to the program at each site; 1 site (Minneapolis) enrolled 22 participants before the conclusion of the evaluation period. Of the 222 participants admitted to IETP during the evaluation window, 206 completed both baseline and discharge assessments. Among these, 105 participants completed follow-up assessments approximately 6 months postdischarge and were identified as SOF. As described earlier, SOF represent a clinically distinct subgroup of service members with exposure profiles and functional demands that differ substantially from conventional military populations. Thus, SOF are the target population of the IETP and the current study. Loss at follow-up was attributable to the inability to re-establish contact with participants. Previous analyses of participants in IETP showed that those who were lost at follow-up did not differ significantly from participants with follow-up data on the 3 primary outcome measures in this study.16 Measures Participants completed a battery of standardized self-report instruments at 3 time points: program intake (eg, at baseline before the initiation of treatment), discharge (eg, after the treatment was completed and the patient was prepared to return to their base), and ∼6 months postdischarge (eg, while the service member was out of treatment and back on their base). Data were collected using the Qualtrics electronic data capture platform (Qualtrics).a The Neurobehavioral Symptom Inventory (NSI) is a 22-item self-report measure designed to assess postconcussive and neurobehavioral symptoms commonly associated with TBI.17 The instrument captures symptom burden across multiple domains relevant to TBI sequelae. Prior research has demonstrated adequate convergent validity through associations with established psychiatric symptom measures and indices of emotional distress.18 The NSI has also shown strong internal consistency (α=0.81-0.91) and good test-retest reliability (intraclass correlation coefficients=0.78-0.94).19 We used NSI total scores in this study as one measure of TBI-related clinical burden. TBI-related functional participation outcomes were assessed with the Mayo-Portland Adaptability Inventory-Participation Index (M2PI), an 8-item measure evaluating engagement and functioning across key life domains affected by brain injury.20 Data on the M2PI were collected via self-report using Qualtrics. These domains include initiation, social interaction, leisure activities, self-care, residential responsibilities, transportation, employment, and financial management. The abbreviated M2PI has been shown to correlate strongly with the full-length instrument, supporting its validity as a concise indicator of participation outcomes. Psychometric evaluations using Rasch Facet analysis indicate acceptable person (>0.80) and item (>0.90) reliability, and the scale demonstrates good internal consistency (α≈0.83).21 PTSD symptom severity was assessed using the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders, fifth edition (PCL-5), a 20-item self-report questionnaire that evaluates the extent to which individuals have been bothered by PTSD symptoms during the past month.22 Items are rated on a 5-point Likert-type scale ranging from 0 (not at all) to 4 (extremely), with higher total scores indicating greater symptom severity. The PCL-5 has demonstrated strong construct validity, including positive associations with other validated PTSD measures, and exhibits excellent internal consistency reliability for the total score (α≈0.91).22 Additional variables were extracted from each participant’s electronic health record to characterize and account for factors known to influence TBI-related clinical burden (postconcussive symptoms). These variables included age (in years), race (categorized as White vs Other, including Black, Asian, and Pacific Islander), Hispanic ethnicity (Hispanic or non-Hispanic), military rank (enlisted or officer), educational level (graduate degree or not), length of military service (years), and number of military deployments.16 Data analysis We used a nonparametric random forest imputation procedure to impute missing values for <5% of the sample’s demographic variables.23 We evaluated the plausibility of the missing at random (MAR) assumption by examining patterns of missingness among the observed demographic variables. Specifically, we compared participants with and without missing data on key covariates and found no systematic differences, supporting the assumption that data were plausibly MAR. Given the low proportion of missingness (<5%), the risk of substantial bias due to departures from MAR is likely minimal. Analyses addressed 2 primary questions examining associations between TBI-related clinical burden and PTSD symptom severity using cross-sectional and longitudinal approaches. Program intake data were analyzed to evaluate whether TBI-related clinical burden was associated with PTSD symptom severity at baseline. PTSD symptoms, assessed with the PCL-5, served as the dependent variable. Two separate multivariable linear regression models were estimated, each including one TBI-related clinical burden (postconcussive symptoms) predictor: total neurobehavioral symptom burden or functional participation limitations related to brain injury. All models were adjusted for age, race, ethnicity, military rank, education, service length, and number of deployments. To address the second question, longitudinal analyses examined whether TBI-related clinical burden moderated changes in PTSD symptoms across rehabilitation stages (baseline, discharge, and 6-month follow-up). Linear mixed-effects models were estimated using all available repeated measures, with PTSD symptom severity as the outcome and time modeled as a fixed effect. Two separate models were fit, each including one TBI-related clinical burden predictor (ie, NSI scores or M2PI scores) and a corresponding interaction term between time and that predictor to test moderation. Participant-level random intercepts were specified to account for within-person temporal correlation. The same covariates used in cross-sectional analyses were included as fixed effects. Model assumptions for linear and mixed-effects models were evaluated via diagnostics with the residuals, including assessment of normality, linearity, and homoscedasticity. Multicollinearity was examined using variance inflation factors, with no evidence of concern. Inspection of residual and Q-Q plots indicated that assumptions were adequately met, and no model adjustments were necessary. We used R softwareb for all analyses. Results The sample consisted of 105 male SOF (table 1) who were predominantly White (68.6%), with a smaller proportion identifying as Hispanic (7.6%). Approximately one-third were officers (36.2%), and a similar proportion reported a graduate degree (37.1%). Participants were mid-career service members with an average age of 42 years (SD, 5.4) with extensive military experience, averaging 18.5 years of service and over 7 deployments (M=7.38; SD, 3.11). Table 1Sociodemographic characteristics of the SOF included in this study (n=105)Table 1 dummy alt textVariablen%Race: White7268.57Ethnicity: Hispanic87.62Rank: officer3836.19Education: graduate degree3937.14MSDAge (y)42.005.40Years of military service18.542.87Number of deployments7.383.11 Sociodemographic characteristics of the SOF included in this study (n=105) Cross-sectional analyses examined associations between TBI-related clinical burden indicators and PTSD symptom severity at baseline. In the first model, baseline neurobehavioral symptom burden was significantly associated with baseline PTSD symptom severity (b=1.00; P<.001). Higher scores on the NSI were associated with higher scores on the PCL-5, with model estimates indicating approximately a 1-point increase in neurobehavioral symptoms for each 1-point increase in PTSD symptom severity (table 2; fig 1). This association was statistically significant even with adjustment for relevant demographic and military covariates. Table 2Adjusted cross-sectional associations between TBI-related clinical burden and PTSD symptom severity among SOFTable 2 dummy alt textPTSD and NSIPTSD and M2PIVariableb95% CIPb95% CIPNSI1.000.90-1.20<.0012.602.10-3.10<.001Age (y)–0.13–0.65 to 0.39.60–0.66–1.30 to –0.07.03Race: White–0.07–4.60 to 4.50.991.80–3.50 to 7.10.50Hispanic7.50–0.18 to 15.00.0610.001.30-19.00.03Rank1.60–3.40 to 6.70.505.10–0.69 to 11.00.08Education0.65–4.20 to 5.50.802.90–2.80 to 8.60.30Service length0.23–0.65 to 1.10.600.65–0.38 to 1.70.20Deployments–0.11–0.80 to 0.58.80–0.36–1.20 to 0.46.40Abbreviations: b, adjusted beta coefficient; CI, confidence interval. Adjusted cross-sectional associations between TBI-related clinical burden and PTSD symptom severity among SOF Abbreviations: b, adjusted beta coefficient; CI, confidence interval. Cross-sectional association between neurobehavioral symptom burden and PTSD symptom severity at baseline among special operations personnel. In a separate model evaluating functional participation limitations, baseline scores on the M2PI Participation Index were also significantly associated with baseline PTSD symptom severity (b=2.60; P<.001). Greater participation-related impairment was associated with higher PTSD symptom scores (fig 2). In this model, age was additionally associated with PTSD symptom severity, with younger participants reporting greater PTSD symptoms. Hispanic ethnicity was also significantly associated with PTSD symptom severity, indicating Hispanic participants had higher PTSD symptom scores relative to non-Hispanic participants. These associations were observed while adjusting for other covariates included in the model. Cross-sectional association between functional participation impairment and PTSD symptom severity at baseline among special operations personnel. Longitudinal mixed-effects models evaluated whether TBI-related clinical burden indicators moderated changes in PTSD symptom severity across IETP rehabilitation stages, including baseline, discharge, and 6-month follow-up (table 3). In the model examining neurobehavioral symptom burden as a moderator, the interaction between time and NSI score was not statistically significant (b=-0.06; P=.14). Neurobehavioral symptoms were associated with PTSD symptom severity cross-sectionally. Baseline neurobehavioral symptom burden did not significantly alter the trajectory of PTSD symptoms over the course of rehabilitation and follow-up (fig 3). Table 3Adjusted longitudinal associations between TBI-related clinical burden and PTSD symptom severity among SOF enrolled in TBI rehabilitationTable 3 dummy alt textPTSD and NSIPTSD and M2PIVariableb95% CIPb95% CIPNSI1.000.85-1.20<.0013.002.40-3.60<.001Age (y)–0.26–0.61 to 0.08.13–0.66–1.11 to –0.26.001Race: White–1.60–5.20 to 2.00.400.22–4.00 to 4.40.99Hispanic2.80–3.30 to 8.90.406.00–1.20 to 13.00.10Rank1.20–2.70 to 5.10.503.10–1.60 to 7.80.20Education1.90–1.90 to 5.70.302.60–2.00 to 7.10.30Service length–0.18–0.46 to 0.10.200.12–0.21 to 0.46.50Deployments0.06–0.46 to 0.59.800.09–0.54 to 0.71.80Time2.50–0.32 to 5.40.080.66–1.70 to 3.00.60NSI×time–0.06–0.13 to 0.02.14–0.35–0.61 to 0.09.01Abbreviations: b, adjusted fixed beta coefficient; CI, confidence interval. Adjusted longitudinal associations between TBI-related clinical burden and PTSD symptom severity among SOF enrolled in TBI rehabilitation Abbreviations: b, adjusted fixed beta coefficient; CI, confidence interval. Longitudinal association between neurobehavioral symptom burden and PTSD symptom trajectories across rehabilitation stages. Analyses examining functional participation limitations demonstrated a significant interaction between time and M2PI Participation Index score (b=–0.35; P=.01). This interaction indicated that the course of PTSD symptom changes over time differed as a function of baseline participation-related impairment. Specifically, individuals with higher participation index scores exhibited a significantly steeper decline in PTSD symptom severity across time compared with those with moderate or lower participation index scores (fig 4). Longitudinal moderation of PTSD symptom trajectories by functional participation impairment across rehabilitation stages. Discussion Our findings are consistent with prior literature demonstrating that greater neurobehavioral symptom burden is associated with greater PTSD symptom severity at program intake. Furthermore, we found that participation-related functional impairment (M2PI) was independently associated with higher PTSD symptom severity at intake even with adjustment for demographic and military covariates. Longitudinal analyses showed that neurobehavioral symptom burden did not significantly influence changes in PTSD symptoms over time. In contrast, functional participation limitations did affect PTSD trajectories, with individuals who had more severe participation scores at baseline showing larger reductions in PTSD symptoms during rehabilitation. Participation domains measured by the M2PI, such as home independence, financial management, work-life balance, and social engagement, are modifiable targets within interdisciplinary TBI treatment programs that integrate mental health care. These findings suggest that neurobehavioral symptoms and functional participation reflect different aspects of clinical burden, with participation more closely related to changes in PTSD symptoms over time. Assessing participation may therefore provide additional insight beyond symptom severity when evaluating outcomes in TBI rehabilitation. The observed cross-sectional associations between TBI-related clinical burden (postconcussive symptoms) and PTSD symptom severity are consistent with a substantial body of literature documenting the frequent co-occurrence of these conditions in military populations. Prior epidemiologic and clinical studies have shown that service members with a history of concussion or mild TBI report significantly higher PTSD symptom severity than those without such injury histories, even when accounting for deployment-related exposures.3,8 Our findings extend this work by demonstrating that both neurobehavioral symptom burden and participation-related functional impairment are independently associated with PTSD symptom severity at rehabilitation intake among SOF. Our work aligns with previous research suggesting that PTSD symptoms may be central to how individuals experience and report postconcussive symptoms, rather than being merely a downstream consequence of neurological injury.6,7 The association between participation limitations and PTSD symptom severity aligns with ecological models of TBI outcomes, which suggest that functional disruption across daily life domains is closely associated with psychological distress following injury and recovery.13,20 Importantly, the longitudinal findings of this study add to our current understanding of how TBI-related clinical burden (postconcussive symptoms) relates to PTSD symptom trajectories over time. Although neurobehavioral symptom burden was strongly associated with PTSD severity cross-sectionally, it did not moderate changes in PTSD symptoms across rehabilitation and follow-up stages of IETP, suggesting that symptom overlap alone may not meaningfully shape recovery trajectories. In contrast, functional participation impairment significantly moderated PTSD symptom change, with individuals exhibiting more severe participation limitations demonstrating steeper declines in PTSD symptoms over time. This finding is consistent with rehabilitation-focused studies indicating that improvements in functional engagement and role participation may facilitate downstream improvements in psychological outcomes.4,14 From a mechanistic perspective, participation limitations may capture broader contextual and environmental constraints that are more amenable to interdisciplinary rehabilitation than neurobehavioral symptoms alone, leading to greater gains in PTSD symptom reduction.10 The present findings have several important implications. The strong cross-sectional associations between PTSD symptom severity and both neurobehavioral symptoms and participation-related impairment may indicate that it is important to routinely assess psychological distress alongside TBI-related symptoms at program intake. In addition, the longitudinal moderation findings reveal functional participation as a clinically meaningful target, as individuals with more severe participation-related scores demonstrated steeper improvements in PTSD symptoms over time. Limitations Several limitations should be considered when interpreting the findings of this study. First, PTSD symptom severity and TBI-related clinical burden (postconcussive symptoms) indicators were assessed using self-report measures, which may be influenced by reporting biases. Second, the analytic sample consisted exclusively of SOF enrolled in an intensive inpatient rehabilitation program, which enhances internal relevance for this high-risk group but limits generalizability to conventional forces, veterans, or individuals with TBI exposure that do not require rehabilitation. Third, attrition at the 6-month follow-up may have introduced selection bias, as participants who were unreachable may differ systematically from those retained. Fourth, although functional participation and neurobehavioral symptoms were examined as distinct indicators of TBI-related clinical burden, other relevant dimensions, such as objective neurocognitive performance, injury mechanism, or timing of injury relative to trauma exposure, were not directly assessed and may further clarify PTSD trajectories in future research. In addition, future work could incorporate domain-level analyses of NSI symptom clusters (eg, cognitive, affective, vestibular, and somatic) in combination with PTSD and participation measures. Lastly, the NSI and M2PI measure overlapping symptoms. Future research should build on this study by studying the mechanisms through which functional participation influences PTSD symptom improvement, including the potential roles of social reintegration, cognitive engagement, and the environment. Longitudinal studies incorporating objective neurocognitive measures, clinician-rated outcomes, and more diverse military and veteran populations may be needed to better understand causal pathways and improve generalizability. In addition, randomized or quasi-experimental designs testing participation-focused interventions within TBI rehabilitation programs could help determine whether targeting functional engagement leads to sustained improvements in PTSD. Conclusion To conclude, this study showed that TBI-related clinical burden (postconcussive symptoms) is associated with PTSD symptom severity among SOF seeking TBI rehabilitation. Although neurobehavioral symptoms and participation-related functional impairment were both associated with PTSD symptom severity at intake, only functional participation moderated PTSD symptom trajectories over time. Thus, it is important for TBI rehabilitation programs to promote integrated, participation-focused treatment approaches that address PTSD as a central component of recovery. Suppliers a.Qualtrics electronic data capture platform; Qualtrics. Qualtrics electronic data capture platform; Qualtrics. b.R software. R software. Disclosure The investigators have no financial or nonfinancial disclosures to make in relation to this project. Acknowledgments Acknowledgments We thank Joel Scholten, MD, the administrative staff, and clinical team members at the 5 IETPs for their contributions. Data statement Datasets will be provided by the corresponding author upon reasonable request. Human subjects research statement This project was reviewed and considered “not research” by the IRB at James A Haley Veterans’ Hospital. Instead, this project was considered a quality improvement. Footnotes Supported by the Veterans Health Administration Health Services Research and Development Partnered Evaluation Center (PEC 21-129) through the Quality Enhancement Research Initiative and Research Career Scientist Award (RCS 17-297). This work is a result of resources, support, and the use of facilities and personnel at the James A. Haley Veterans’ Hospital, Clinics, and Research Service and the VA HSR&D Informatics, Decision-Enhancement, and Analytic Sciences (IDEAS) Center of Innovation (CIN 13-414). Contents do not represent the views of the Department of Veterans Affairs or the US government. Members of the research team also had funding from the VA Office of Research and Development (HSR Career Award), the Quality Enhancement Research Initiative, the Congressionally Directed Medical Research Program, the VA Rehabilitation Research and Development program, Virginia Commonwealth University, and Partners Against Mortality in Epilepsy. List of abbreviations: IETP, Intensive Evaluation and Treatment Program; M2PI, Mayo-Portland Adaptability Inventory-Participation Index; MAR, missing at random; NSI, Neurobehavioral Symptom Inventory; PCL-5, PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders, fifth edition; PTSD, posttraumatic stress disorder; SOF, Special Operations Forces; TBI, traumatic brain injury; VA, U.S. Department of Veterans Affairs. References