A-126 Clinical Utility of Risk Factors to Predict Self-reported Neurobehavioral Outcome Following Traumatic Brain Injury: PTSD, Sleep, Resilience, and Lifetime Blast Exposure

Objective To examine the clinical utility of PTSD, Sleep, Resilience, and Lifetime Blast Exposure as ‘Risk Factors’ for predicting poor neurobehavioral outcome following traumatic brain injury (TBI). Methods Participants were 993 service members/veterans evaluated following an uncomplicated mild TBI (MTBI), moderate–severe TBI (ModSevTBI), or injury without TBI (Injured Controls; IC); divided into three cohorts: (1) < 12 months post-injury, n = 237 [107 MTBI, 71 ModSevTBI, 59 IC]; (2) 3-years post-injury, n = 370 [162 MTBI, 80 ModSevTBI, 128 IC]; and (3) 10-years post-injury, n = 386 [182 MTBI, 85 ModSevTBI, 119 IC]. Participants completed a 2-hour neurobehavioral test battery. Odds Ratios (OR) were calculated to determine whether the ‘Risk Factors’ could predict ‘Poor Outcome’ in each cohort separately. Sixteen Risk Factors were examined using all possible combinations of the four risk factor variables. Poor Outcome was defined as three or more low scores (< 1SD) on five TBI-QOL scales (e.g., Fatigue, Depression). Results In all cohorts, the vast majority of risk factor combinations resulted in ORs that were ‘clinically meaningful’ (ORs > 3.00; range = 3.15 to 32.63, all p’s < .001). Risk factor combinations with the highest ORs in each cohort were PTSD (Cohort 1 & 2, ORs = 17.76 and 25.31), PTSD+Sleep (Cohort 1 & 2, ORs = 18.44 and 21.18), PTSD+Sleep+Resilience (Cohort 1, 2, & 3, ORs = 13.56, 14.04, and 20.08), Resilience (Cohort 3, OR = 32.63), and PTSD+Resilience (Cohort 3, OR = 24.74). Conclusions Singularly, or in combination, PTSD, Poor Sleep, and Low Resilience were strong predictors of poor outcome following TBI of all severities and injury without TBI. These variables may be valuable risk factors for targeted early interventions following injury.

A-117 Sleep Disruption has a Stronger Influence on Self-reported Neurobehavioral Status than Traumatic Brain Injury in U.S. Military Service Members and Veterans

Objective To examine the relation between sleep disruption and neurobehavioral outcome following mild, moderate, and severe traumatic brain injury (TBI). Methods Participants were 509 U.S. military service members/veterans divided into four groups: uncomplicated mild TBI (n = 183; MTBI); complicated mild TBI and moderate–severe TBI (n = 96; STBI); injured controls (n = 138; IC); and non-injured controls (n = 92; NIC). Participants completed a 2-hour neurobehavioral test battery 12 or more months post-injury that included the PTSD Checklist and 13 scales from the TBI-Quality of Life (TBI-QOL). Using the TBI-QOL Sleep Disturbance scale, participants were classified into two ‘Sleep’ subgroups: Poor Sleep (55 T or higher) or Good Sleep (50 T or lower). Results A higher proportion of the MTBI group was classified as having Poor Sleep (79.2%) compared to the IC (64.5%) and STBI (58.3%) groups; and all were higher than the NIC group (40.2%). In each group separately, participants with Poor Sleep had significantly worse scores on all TBI-QOL scales compared to those with Good Sleep (all p’s < .001, d = .68 to d = 1.98). Participants with Poor Sleep consistently had worse TBI-QOL scores regardless of TBI severity or the presence/absence of TBI (all p’s < .05). Additionally, there was a significant interaction between Sleep and PTSD. Both factors combined resulted in worse outcome than either factor alone (p’s < .05). Conclusions Poor Sleep had a very strong influence on self-reported neurobehavioral outcome, and a greater influence on outcome than TBI severity or the presence/absence of TBI. Poor Sleep may be a useful ‘risk factor’ that can be used clinically to identify individuals in need of early intervention.

B-57 Post-traumatic Stress Disorder is a Stronger Predictor of Long-term Neurobehavioral Outcome than Traumatic Brain Injury Severity in U.S. Military Service Members

Objective The purpose of this study was to examine long-term neurobehavioral outcome in SMVs with versus without PTSD following TBI of all severities. Methods Participants were 536 SMVs recruited into three experimental groups (TBI, injured controls [IC], non-injured controls [NIC]). Participants completed the PTSD Checklist and the TBI-Quality of Life (TBI-QOL). Participants were divided into six subgroups based on the three experimental categories, two PTSD categories (i.e., present/absent), and two broad TBI severity categories (i.e., ‘unMTBI’ [includes uncomplicated mild TBI]; and ‘smcTBI’ [includes severe TBI, moderate TBI, and complicated mild TBI): (1) NIC/PTSD-Absent, (2) IC/PTSD-Absent, (3) unMTBI/PTSD-Absent, (4) unMTBI/PTSD-Present, (5) smcTBI/PTSD-Absent, and (6) smcTBI/PTSD-Present. Results There were significant main effects across the six groups for all TBI-QOL measures (p < .001). Select pairwise comparisons revealed significantly worse scores (p < .001) on all TBI-QOL measures in all PTSD-Present groups compared to the PTSD-Absent groups (i.e., Group 3v4 and 5v6; d = 0.90 to 2.11). In contrast, when controlling for PTSD, there were no significant differences between the TBI severity groups for all TBI-QOL measures (i.e., Group 3v5 and 4v6). In the TBI sample, a series of step-wise regression analyses revealed that PTSD, but not TBI severity, was consistently a strong predictor of all TBI-QOL scales (all p’s < .001), accounting for up to 64% of the variance. Conclusions These results provide support for the very strong influence of PTSD, but not TBI severity, on long-term neurobehavioral outcome following TBI. Concurrent PTSD and TBI of all severities should be considered a risk factor for poor long-term neurobehavioral outcome that requires ongoing monitoring.