Associations Between Multiple Remote Mild TBIs and Objective Neuropsychological Functioning and Subjective Symptoms in Combat-Exposed Veterans

Abstract

Objective

The purpose of this study was to evaluate relationships between multiple mild traumatic brain injuries (mTBIs) and objective and subjective clinical outcomes in a sample of combat-exposed Veterans, adjusting for psychiatric distress and combat exposure.

Method

In this cross-sectional study, 73 combat-exposed Iraq/Afghanistan Veterans were divided into three groups based on mTBI history: 0 mTBIs (n = 31), 1–2 mTBIs (n = 21), and 3+ mTBIs (n = 21). Veterans with mTBI were assessed, on average, 7.78 years following their most recent mTBI. Participants underwent neuropsychological testing and completed self-report measures assessing neurobehavioral, sleep, and pain symptoms.

Results

MANCOVAs adjusting for psychiatric distress and combat exposure showed no group differences on objective measures of attention/working memory, executive functioning, memory, and processing speed (all p’s > .05; η_p² = .00–.06). In contrast, there were significant group differences on neurobehavioral symptoms (p’s = < .001–.036; η_p² = .09–.43), sleep difficulties (p = .037; η_p² = .09), and pain symptoms (p < .001; η_p² = .21). Pairwise comparisons generally showed that the 3+ mTBI group self-reported the most severe symptoms, followed by comparable symptom reporting between the 0 and 1–2 mTBI groups.

Conclusions

History of multiple, remote mTBIs is associated with elevated subjective symptoms but not objective neuropsychological functioning in combat-exposed Veterans. These results advance understanding of the long-term consequences of repetitive mTBI in this population and suggest that Veterans with 3+ mTBIs may especially benefit from tailored treatments aimed at ameliorating specific neurobehavioral, sleep, and pain symptoms.

Introduction

As a result of significant improvements in battlefield protection over the past several decades, more military service members have survived wounds in the Iraq and Afghanistan conflicts that would have otherwise been fatal in previous wars. A consequence of this higher survival rate has been an increase in the number of military personnel sustaining traumatic brain injuries (TBI). Estimates suggest that approximately one in four or five deployed service members have sustained a TBI—the majority of which are classified as mild, and this number may be even higher in combat-exposed Veterans (Defense and Veterans Brain Injury Center, 2018; O'Neil et al., 2013; Tanielian & Jaycox, 2008; Terrio et al., 2009). Acutely, mild TBI (mTBI) is associated with objective neuropsychological dysfunction as well as numerous neurobehavioral symptoms that broadly reflect somatic, vestibular, cognitive, and affective-related sequelae (Belanger, Curtiss, Demery, Lebowitz, & Vanderploeg, 2005; Dikmen, Machamer, & Temkin, 2017; Kelly, Coldren, Parish, Dretsch, & Russell, 2012; Terrio et al., 2009; Vanderploeg et al., 2015). Although cognitive deficits and associated symptoms are generally expected to resolve within several days to weeks following mTBI (Rohling et al., 2011; Schretlen & Shapiro, 2003), some studies have shown that symptoms continue to be endorsed beyond this projected recovery window, particularly among military Veterans (Hou et al., 2012; Schwab et al., 2017; Vanderploeg, Curtiss, Luis, & Salazar, 2007; Vincent, Roebuck-Spencer, & Cernich, 2014). Identifying the factors contributing to symptoms and sequelae reported in the chronic phase of injury has been an area of active investigation, yet our understanding of such variables remains incomplete.

Comorbid psychiatric distress is one of the factors often associated with chronic deficits in the Veteran mTBI population (Chapman & Diaz-Arrastia, 2014; Lippa, Pastorek, Benge, & Thornton, 2010; Mashima et al., 2019; Polusny et al., 2011; Vanderploeg, Belanger, Curtiss, Bowles, & Cooper, 2019). Specifically, there is mounting evidence to suggest that Veterans with a history of mTBI and comorbid mental health diagnoses such as posttraumatic stress disorder (PTSD) and depression often experience numerous neurobehavioral symptoms, including cognitive dysfunction, chronically following injury (Belanger, Kretzmer, Vanderploeg, & French, 2010a; Cooper et al., 2011; Garber, Rusu, & Zamorski, 2014). Additionally, investigators have also established that deployment itself affects neuropsychological and neurobehavioral functioning (Mitchell, 2017; Vasterling et al., 2006). Still, others have raised the possibility that TBI-related injury characteristics may influence the nature, presence, and duration of post-injury sequelae (Mashima et al., 2019). Notably, service members are often faced with multiple combat deployments and may be subjected to repetitive blast exposure while deployed, which frequently results in multiple mTBIs throughout the course of one’s military career (Boyle et al., 2014; Galarneau, Woodruff, Dye, Mohrle, & Wade, 2008; MacGregor, Dougherty, Morrison, Quinn, & Galarneau, 2011; O'Neil et al., 2014; Wilk, Herrell, Wynn, Riviere, & Hoge, 2012). Although investigators have begun examining the effects of multiple mTBIs on clinical outcome, available findings are rather heterogeneous, and more research is needed to better understand the cumulative effects of mTBIs on objective and subjective functioning in military personnel.

With regard to objective neuropsychological functioning in the context of multiple mTBIs, prior research conducted on military samples has yielded disparate findings. Dretsch, Silverberg, and Iverson (2015) examined a large sample of active duty service members divided into the following four groups: 0 mTBIs, 1 mTBI, 2 mTBIs, and 3+ mTBIs. Using domain scores from the Central Nervous System Vital Signs (CNS VS) computerized neurocognitive test as primary outcomes of interest, study authors found no group differences on any of the cognitive variables across the four groups. In another study, Cooper and colleagues (2018) examined military service members divided into five groups: 1 mTBI, 2 mTBIs, 3+ mTBIs, no prior mTBIs but orthopedic injury, and no prior mTBIs but PTSD diagnosis. They evaluated cognitive performance using a composite score comprised of measures of executive functioning, processing speed, working memory, and verbal memory. Although they found no group differences across the three mTBI groups (1 mTBI, 2 mTBIs, and 3+ mTBIs), those with a history of 1 mTBI and 3+ mTBIs performed more poorly than orthopedic controls on the cognitive composite; otherwise, no other group differences were reported. Finally, Spira, Lathan, Bleiberg, & Tsao (2014) evaluated active duty military personnel using the Defense Automated Neurobehavioral Assessment (DANA), another computerized neurocognitive assessment tool, and reported reduced cognitive functioning on measures of attention and rapid discrimination in those with multiple lifetime mTBIs (defined as 3+ mTBIs) relative to those with two or fewer mTBIs. However, no associations were found between the number of lifetime mTBIs and performance on memory-related tasks in this sample.

As for the relationship between lifetime number of mTBIs and subjective neurobehavioral symptoms, some military-specific studies have demonstrated a positive association between these variables. To illustrate, Dretsch and colleagues (2015) found that service members with a history of 3+ mTBIs reported greater neurobehavioral symptoms relative to service members with no prior mTBIs, as well as those with a history of 1 and 2 mTBIs, concluding that 3+ mTBIs is a risk factor for “ongoing” symptomatology. Miller, Ivins, and Schwab (2013) and Spira and colleagues (2014) reported similar findings—that is, greater neurobehavioral symptom reporting was associated with multiple lifetime mTBIs; however, this relationship was attenuated by time since injury in Miller and colleagues’ (2013) study and was found to be dependent on, or better accounted for by, psychiatric distress and military-specific factors such as combat exposure and deployment history in the study by Spira and colleagues (2014). These findings suggest a complicated relationship between the number of mTBIs and neurobehavioral symptoms but provide evidence that the number of mTBI events may be at least one important factor associated with chronic neurobehavioral symptoms.

In a more recent investigation, Cooper and colleagues (2018), after dividing their military sample into five distinct groups including three mTBI groups (those with a history of 1, 2, and 3+ mTBIs) as well as two control groups (orthopedic and PTSD controls), found that the mTBI groups showed a nonsignificant pattern of increased symptom reporting with increasing number of mTBIs. Moreover, the PTSD group had the highest number of neurobehavioral complaints, whereas the orthopedic group had the fewest, with the mTBI groups falling between the two control groups. Finally, Bryan (2013a) evaluated military personnel with a history of 0, 1, and 2+ mTBIs and found that those with a history of 1 mTBI and 2+ mTBIs endorsed greater neurobehavioral symptoms than those without a history of mTBI, but the two mTBI groups (1 and 2+ mTBIs) did not significantly differ with respect to symptom reporting.

Beyond evaluating traditional neurobehavioral symptoms, it is well recognized that sleep disturbance and pain are also commonly experienced in Veterans with mTBI histories (Balba et al., 2018; Bosco, Murphy, & Clark, 2013; Stojanovic et al., 2016). However, relatively few studies have examined associations between multiple mTBIs and sleep and pain complaints. Whereas Dretsch and colleagues (2015) found no associations between the number of concussions and self-reported sleep symptoms in a large sample of active duty soldiers, Bryan (2013b) reported that multiple TBIs were associated with higher rates of clinical insomnia as well as more severe insomnia in active duty military personnel even when adjusting for psychiatric comorbidities and neurobehavioral symptoms. However, Bryan (2013b) included service members with all TBI severity levels, making the results difficult to generalize to Veterans with mTBI only. As for pain, Lindquist, Love, and Elbogen (2017) found that, relative to a single TBI, Veterans with a history of two or more TBIs endorsed significantly higher rates of pain. Similar to Bryan (2013b), however, Lindquist and colleagues (2017) evaluated pain across all TBI severities; thus, the relationship between multiple mTBIs and sleep and pain symptoms remains to be determined.

The primary purpose of this study was to examine the relationship between multiple mTBIs and objective neuropsychological functioning and subjective symptom reporting in a sample of combat-exposed Iraq and Afghanistan Veterans. Importantly, prior studies examining the effects of multiple mTBIs on post-injury clinical outcome have largely been conducted on active duty service members with relatively acute injuries. Moreover, previous work has inconsistently controlled for factors such as psychiatric distress and level of combat exposure—variables known to influence neuropsychological performance and symptom burden following mTBI (Hoge et al., 2008; Vasterling et al., 2006). Therefore, in the present study, we specifically sought to evaluate Veterans with remote mTBI histories (>1 year post-injury) and were particularly interested in determining the extent to which history of multiple mTBIs is associated with objective and subjective functioning independent of psychiatric distress and level of combat exposure. We hypothesized that after adjusting for psychiatric symptoms and combat exposure, that lifetime number of mTBIs would not be associated with objective cognitive functioning. In contrast, we posited that there would be a positive relationship between lifetime number of mTBIs and subjective symptom reporting, such that Veterans with a history of multiple mTBIs (i.e., 3+ mTBIs) would endorse the greatest levels of neurobehavioral symptoms, sleep disturbance, and pain, followed by those with a history of 1–2 mTBIs and those with no prior mTBIs.

Method

Participants and Procedures

Participants included 73 combat-exposed Iraq/Afghanistan Veterans divided into three groups based on mTBI history: 0 mTBIs (n = 31), 1–2 mTBIs (n = 21), and 3+ mTBIs (n = 21). Veterans were recruited between 2014 and 2017 from outpatient clinics within a large Veterans Affairs hospital as part of a larger Chronic Effects of Neurotrauma Consortium (CENC) project (Cifu, Williams, Hinds, & Agyemang, 2018; Walker et al., 2016). As part of this cross-sectional study, participants were administered a structured clinical interview to assess TBI history; a self-report measure assessing combat exposure; a comprehensive neuropsychological assessment; and several self-report questionnaires assessing psychiatric distress, neurobehavioral symptoms, and sleep and pain-related symptoms. Both performance and symptom validity measures were also included. Trained research personnel administered all study procedures under the supervision of a licensed clinical neuropsychologist. The study was approved by the local Institutional Review Board, and all participants provided informed consent prior to participating in the study.

Study inclusion criteria included being a combat-exposed Veteran who served in Iraq or Afghanistan (i.e., Operation Enduring Freedom, Operation Iraqi Freedom, and/or Operation New Dawn). Exclusion criteria included having a mental health diagnosis of bipolar disorder or schizophrenia (or another psychotic disorder); active substance dependence; recent (within the previous month) suicidal intent or attempt; and history of a moderate or severe TBI. Additionally, Veterans who did not perform within expectations on performance and symptom validity tests were excluded from analyses (see “Measures” for details), and failure to complete all study procedures was exclusionary. Finally, for Veterans with a history of mTBI, sustaining a recent mTBI was exclusionary (defined as sustaining an mTBI within 1 year of study enrollment), as we were specifically interested in addressing long-term outcomes following mTBI.

In total, 104 Veterans consented to the study; of these, 12 participants were excluded due to psychiatric/medical reasons, 4 due to history of a moderate TBI, 5 due to performance validity test failure, 2 based on symptom validity test failure, 6 for failing to complete all study procedures, and 2 for having sustained a recent mTBI. Our final sample (N = 73) was comprised of predominantly male Veterans (91.8%). On average, the sample was 34.14 years of age (SD = 6.33) and had completed 15.00 years of education (SD = 1.65). Overall, about half (56.2%) of the sample identified as “Caucasian/Middle Eastern” and about one-third (35.6%) of the sample identified as “Hispanic or Latino.” Of the 42 Veterans with a history of at least one mTBI, the average time tested post-injury was 7.78 years (SD = 5.98), and 66.7% of the mTBI sample had a history of blast-related mTBI.

Measures

Structured clinical interview

The Virginia Commonwealth University Retrospective Concussion Diagnosis Interview (rCDI) was used to evaluate Veterans’ mTBI history (Walker et al., 2015). For the purpose of this study, both the “blast” and “general” versions of the interview were administered. The rCDI gathers detailed information about previous concussive events, including the date of injury, setting in which the injury occurred, mechanism of injury, presence and duration of loss or alteration of consciousness and/or posttraumatic amnesia, and any other associated symptoms occurring at the time of injury. These details are collected for each injury reported, and the VA/DoD Clinical Practice Guideline for The Management of Concussion-mTBI Working Group (2009, 2016) criteria were used to determine whether a reported event qualified as an mTBI. According to VA/DoD guidelines, an mTBI is defined as experiencing a loss of consciousness between 0 and 30 min, an alteration of consciousness lasting up to 24 hr, and/or posttraumatic amnesia lasting no longer than 24 hr. Injuries meeting this definition were classified as an mTBI, and the total number of injuries sustained by each Veteran was calculated.

Combat exposure

The “Combat Experiences” (Section D) scale of the Deployment Risk and Resiliency Inventory-2 (DRRI-2) was used to assess combat exposure (Vogt et al., 2013). This scale is a 17-item self-report measuring comprised of questions pertaining to specific events that may have taken place during deployment (e.g., “During deployment, I went on combat patrols or missions”; “During deployment, I was injured in a combat-related incident”; etc.). Participants were asked to rate how often they experienced each event on a scale from 1 to 6, where “1” = “never” and “6” = “daily or almost daily.” A total score was computed by adding participant’s responses across the 17 items (possible range = 17–102); higher scores indicate greater exposure to combat (Vogt et al., 2013). Veterans were considered “combat-exposed” if they had a total score of > 17 on Section D of the DRRI-2 (or a score of “2” or more on at least one item of Section D).

Neuropsychological assessment

The Word Reading subtest of the Wide Range Achievement Test—Fourth Edition (WRAT4) (Wilkinson & Robertson, 2006) was used to assess premorbid intellectual functioning. Measures of attention and working memory included Digit Span Forward, Digit Span Backward, and Digit Span Sequencing from the Wechsler Adult Intelligence Scale—Fourth Edition (WAIS-IV) (Wechsler, 2008). Measures of executive functioning included the Inhibition and Inhibition/Switching conditions from the Delis-Kaplan Executive Function System (D-KEFS) Color-Word Interference Test (CWIT) (Delis, Kaplan, & Kramer, 2001), the Number-Letter Switching condition from the D-KEFS Trail Making Test (TMT) (Delis et al., 2001), and Perseverative Errors and Total Errors from the Wisconsin Card Sorting Test-64 Card Version (WCST-64) (Kongs, Thompson, Iverson, & Heaton, 2000). Measures of memory included the total recall and delayed recall trials of the Brief Visuospatial Memory Test-Revised (BVMT-R) (Benedict, 1997) and the total recall, short delay free recall, and long delay free recall trials from the California Verbal Learning Test—Second Edition (CVLT-II) (Delis, Kramer, Kaplan, & Ober, 2000). Finally, measures of processing speed included the Color Naming and Word Reading conditions from the D-KEFS CWIT (Delis et al., 2001), the Visual Scanning, Number Sequencing, Letter Sequencing, and Motor Speed conditions of the D-KEFS TMT (Delis et al., 2001), and the Symbol Search and Coding subtests from the WAIS-IV (Wechsler, 2008).

Manual-derived normative scores were calculated for each neuropsychological variable of interest; this allowed for adjustments for age, sex, and/or education (based on manual-specific guidelines) and ensured that all variables could be interpreted in a uniform direction—that is, higher scores reflect better performance across all measures. After deriving the normed scores, all variables of interest were converted to standard scores (M = 100, SD = 15) so that each variable could be interpreted on the same metric. Altogether, 21 cognitive variables were assessed in the present study (3 attention/working memory variables, 5 executive functioning variables, 5 memory variables, and 8 processing speed variables).

Self-report measures

Psychiatric distress

The PTSD Checklist for DSM-5 (PCL-5) (Weathers et al., 2013) is a 20-item self-report measure designed to assess PTSD symptoms. Participants were asked to rate the extent to which they have been bothered by each symptom over the past month using a rating scale of 0–4, where “0” reflects “not at all” and “4” reflects “extremely.” A PCL-5 total score was calculated for each Veteran by summing the selected scores from each of the 20 items (possible range: 0–80); higher scores correspond with greater PTSD symptomatology. Additionally, symptom cluster scores representing symptoms of intrusion, avoidance, cognition/mood disturbance, and arousal/reactivity were also generated for the PCL-5.

The Patient Health Questionnaire-9 (PHQ-9) (Kroenke, Spitzer, & Williams, 2001) is a 9-item self-report measure designed to assess depression symptoms. Participants were asked to rate the extent to which they have been bothered by each symptom over the past 2 weeks using a rating scale of 0–3, where “0” indicates “not at all” and “3” indicates “nearly every day.” A PHQ-9 total score was calculated for each Veteran by summing the selected scores from each of the nine items (possible range: 0–27); higher scores correspond with greater depressive symptomatology.

Neurobehavioral symptoms

The Neurobehavioral Symptom Inventory (NSI) is a 22-item self-report measure designed to assess commonly reported neurobehavioral symptoms such as feeling dizzy, headaches, poor concentration, fatigue/loss of energy, etc. (Cicerone and Kalmar, 1995). Participants were asked to rate the extent to which they have been bothered by each symptom in the last 2 weeks using a five-point rating scale ranging from “none” to “very severe,” corresponding with the following values: 0 = none, 1 = mild, 2 = moderate, 3 = severe, and 4 = very severe. An NSI total score was calculated for each Veteran by summing the ratings for the 22 items (possible range: 0–88); higher scores indicate greater neurobehavioral symptomatology. An additional summary index was also calculated from the NSI—the positive symptom total (PST) score (Derogatis, 1994), which is a measure of symptom breadth (as opposed to symptom severity). The PST was calculated by counting the total number of positively endorsed symptoms (i.e., any symptoms marked as ≥1; possible range: 0–22) (Derogatis, 1994; Merritt, Meyer, & Arnett, 2015). Finally, symptom cluster scores were also generated for the NSI based on a previous factor analysis (Vanderploeg et al., 2015): vestibular symptoms (items 1–3; possible range: 0–12), somatic/sensory symptoms (items 4–7 and 9–11; possible range: 0–28), cognitive symptoms (items 13–16; possible range: 0–16), and affective symptoms (items 17–22; possible range: 0–24).

Sleep symptoms

The Pittsburgh Sleep Quality Index (PSQI) is a 19-item self-report questionnaire designed to assess sleep quality and sleep disturbances over a 1-month interval (Buysse, Reynolds III, Monk, Berman, & Kupfer, 1989). The questionnaire gathers both qualitative and quantitative data, including information pertaining to usual bedtime and wake time, number of hours slept, number of minutes to fall asleep, and an assessment of issues that may affect or interfere with sleep. A PSQI global score was computed that reflects the number and severity of sleep-related problems (possible range: 0–21); higher scores reflect poorer sleep quality.

Pain symptoms

The Short-Form McGill Pain Questionnaire (SF-MPQ) is a 15-item self-report questionnaire designed to assess pain (Melzack, 1987). Participants are presented with pain descriptors (e.g., throbbing, sharp, aching, punishing-cruel) and are asked to rate the extent to which they have experienced each type of pain over the past month using a rating scale of 0–3, where “0” reflects “none” and “3” reflects “severe.” A SF-MPQ total score was computed by summing the responses for all 15 items (possible range: 0–45); higher scores indicate greater/more intense pain.

Performance and symptom validity tests

The Test of Memory Malingering (TOMM) (Tombaugh, 1996) and the Forced Choice Recognition subtest from the CVLT-II (Delis et al., 2000) were used to assess performance validity. Standard cutoffs (according to manual-specific guidelines) were used to define inadequate performance for each of the performance validity tests. For the purpose of this study, invalid performance was defined as failure on one or more of the following trials: TOMM Trial 2, TOMM Retention Trail, and CVLT-II Forced Choice Recognition.

The Validity-10 (Val-10), developed by Vanderploeg et al. (2014), is an embedded symptom validity scale within the NSI. The Val-10 is comprised of 10 items from the NSI that are considered to be infrequently endorsed items (Jurick et al., 2016; Vanderploeg et al., 2014); the 10 items are summed to create a total score (Vanderploeg et al., 2014). Notably, the Val-10 has been frequently utilized as a measure of symptom validity in military mTBI studies and has been validated in both active duty and Veteran samples (Armistead-Jehle et al., 2018; Ashendorf, 2019; Bodapati et al., 2019; Jurick et al., 2016; Lange, Brickell, & French, 2015a; Lange, Brickell, Lippa, & French, 2015b). Despite the Val-10 having strong empirical support, there is currently no consensus as to what the optimal cutoff should be for adequately detecting symptom exaggeration, with published cutoffs ranging from ≥13 (Lange, Brickell, & French, 2015a) to ≥33 (Armistead-Jehle et al., 2018). For the purpose of this study, we utilized a Val-10 score of > 22 as our criterion for exclusion, consistent with several prior studies using similar samples (Jurick et al., 2016; Lange, Brickell, Lippa, et al., 2015b; Lu et al., 2019; Vanderploeg et al., 2014).

Approach to Statistical Analyses

As indicated earlier, Veterans were divided into three groups based on mTBI history: 0 mTBIs, 1–2 mTBIs, and 3+ mTBIs. To evaluate whether groups differed on participant characteristics (i.e., basic demographic variables, TBI characteristics, and psychiatric measures), analyses of variance (ANOVAs), independent samples t-tests, and chi-square or Fisher’s exact tests were used, as appropriate. Next, multivariate analyses of covariance (MANCOVAs) were used to investigate group differences on neuropsychological functioning. MANCOVAs were utilized for the cognitive variables to reduce Type I error and to evaluate whether mTBI groups differed along various dimensions of cognition (e.g., executive functioning, memory, etc.) (Field, 2013). Analyses of covariance (ANCOVAs) were used to evaluate group differences on neurobehavioral symptoms, sleep disturbance, and pain. Covariates for the MANCOVAs and ANCOVAs included psychiatric distress¹ and combat exposure². Any significant omnibus test was followed up with post hoc comparisons using Tukey’s Honestly Significant Difference (HSD) tests to compare all pairs of means. Effect sizes are reported as eta-squared values for MANCOVAs and ANCOVAs (interpreted as .01 = small, .06 = medium, and .14 = large) and as Cohen’s d for all pairwise comparisons (interpreted as .20 = small, .50 = medium, and .80 = large). All analyses were conducted using the Statistical Package for the Social Sciences (SPSS), Version 26.

Results

Participant Characteristics

Table 1 displays participant characteristics—basic demographic variables, TBI characteristics, and psychiatric measures—by mTBI group. Whereas mTBI groups were generally comparable across most variables, there were significant group differences on combat exposure as represented by the DRRI-2 total score (p = .046). Post hoc comparisons revealed that Veterans with a history of 3+ mTBIs experienced more combat relative to those with a history of 0 mTBIs (p = .036), but the 3+ mTBI and 1–2 mTBI groups did not differ on combat exposure (p = .299). Additionally, although there was a nonsignificant trend of increasing PTSD symptoms with increasing number of mTBIs (p = .099) for the PCL-5 total score, the PCL-5 arousal symptom cluster did show significant group differences, such that Veterans with a history of 3+ mTBIs had greater arousal symptoms relative to those with a history of 0 mTBIs (p = .010). With regard to mTBI-specific variables, time since injury significantly differed between the 1–2 mTBI and 3+ mTBI groups (p = .013), such that Veterans with a history of 3+ mTBIs were evaluated approximately 5 years post-injury, whereas those with a history of 1–2 mTBIs were evaluated approximately 10.5 years post-injury. Finally, Veterans with a history of 3+ mTBIs endorsed experiencing a blast-related mTBI at a greater rate than those with a history of 1–2 mTBIs (p = .050).

Objective Neuropsychological Functioning

A MANCOVA revealed a nonsignificant effect of mTBI group on measures of attention and working memory when adjusting for psychiatric distress and combat exposure, Wilks Λ = 0.95, F(6, 132) = 0.56, p = .763; η_p² = .025. Similar nonsignificant results were found for measures of executive functioning (Wilks Λ = 0.85, F(10, 128) = 1.07, p = .387; η_p² = .077), memory (Wilks Λ = 0.82, F(10, 128) = 1.30, p = .236; η_p² = .092), and processing speed (Wilks Λ = 0.91, F(16, 120) = 0.36, p = .988; η_p² = .046) when adjusting for the same covariates. Unadjusted means and standard deviations for the neuropsychological variables are reported in Table 2, and adjusted means, standard errors, and results of all univariate analyses are reported in Table 3.

Subjective Symptom Reporting

With regard to neurobehavioral symptoms, ANCOVAs adjusting for psychiatric distress and combat exposure revealed a significant effect of mTBI group across all NSI variables, including the NSI total score, NSI PST score, and each NSI symptom cluster (p’s ≤ .001–.036, η_p² = .09–.43). Unadjusted means and standard deviations for the NSI variables are reported in Table 2, and adjusted means, standard errors, and results of all univariate analyses are reported in Table 4. Given the significant omnibus tests, Tukey’s post hoc tests were used to evaluate all pairwise comparisons; results of pairwise comparisons and associated effect sizes for the NSI variables are reported in Table 5. As indicated in the table, Veterans with a history of 3+ mTBIs endorsed significantly greater symptoms (across all NSI variables, including the NSI total score, NSI PST score, and each NSI symptom cluster) than both the 0 mTBI and 1–2 mTBI groups, and effect sizes were large. The only exception was on the cognitive symptom cluster, where the 3+ mTBI and 1–2 mTBI groups did not significantly differ from each other. As for comparisons between Veterans with 0 mTBIs and those with 1–2 mTBIs, no significant differences were found on any of the NSI symptom scores.

As for sleep and pain symptoms, ANCOVAs adjusting for psychiatric distress and combat exposure revealed a significant effect of mTBI group on both the PSQI global score and SF-MPQ total score (p’s ≤  .001–.037; η_p² = .09–.21). Unadjusted means and standard deviations for the sleep and pain symptoms are reported in Table 2, and adjusted means, standard errors, and results of all univariate analyses are reported in Table 4. Given the significant omnibus tests, Tukey’s post hoc tests were used to evaluate all pairwise comparisons; results of pairwise comparisons and associated effect sizes for the sleep and pain symptoms are reported in Table 5. As shown in the table, Veterans with a history of 3+ mTBIs endorsed significantly greater sleep and pain symptoms relative to the 0 mTBI and 1–2 mTBI groups, and effect sizes were large. In contrast, no significant differences were found between Veterans with 0 mTBIs and those with 1–2 mTBIs on sleep and pain symptoms.

Secondary Analyses

Given that the 1–2 and 3+ mTBI groups differed on time since injury and history of blast-related mTBI (see Table 1), we conducted follow-up analyses controlling for these additional variables. Specifically, ANCOVAs adjusting for psychiatric distress, combat exposure, time since injury, and history of blast-related mTBI were used to compare the 1–2 and 3+ mTBI groups on neurobehavioral symptoms, sleep disturbance, and pain (i.e., those variables found to differ in the earlier analyses). As before, ANCOVAs revealed a significant effect of mTBI group across all NSI variables (NSI total score, NSI PST score, and NSI symptom clusters), sleep symptoms, and pain, such that the 3+ mTBI group endorsed significantly greater symptoms than the 1–2 mTBI groups (p’s ≤  .001–.036; η_p² = .12–.32). Consistent with the earlier findings, the only exception to this was on the NSI cognitive symptom cluster, where the 3+ mTBI and 1–2 mTBI groups did not significantly differ from each other (p = .662; η_p² = .05).

Discussion

The purpose of this study was to evaluate relationships between multiple mTBIs and objective and subjective clinical outcomes in a sample of combat-exposed military Veterans, adjusting for psychiatric distress and level of combat exposure. We specifically focused on military Veterans with remote histories of mTBI (>1 year) to improve understanding of the long-term effects associated with repetitive injury. Consistent with our hypotheses, mTBI groups (0 mTBIs, 1–2 mTBIs, and 3+ mTBIs) demonstrated comparable performance on objective measures of attention/working memory, executive functioning, memory, and processing speed. In contrast, we found group differences with regard to symptom reporting, such that Veterans with a history of 3+ mTBIs endorsed significantly greater neurobehavioral symptoms, sleep difficulties, and pain relative to Veterans with a history of 0 mTBIs and 1–2 mTBIs; however, when comparing the 0 mTBI and 1–2 mTBI groups, no group differences were found on symptom reporting. These findings suggest that multiple (3+) remote mTBIs may be uniquely associated with enduring subjective symptomatology but not objective cognitive functioning when controlling for psychiatric distress and combat exposure. These results advance understanding of the long-term consequences of repetitive mTBI in this population and suggest that Veterans with three or more mTBIs may especially benefit from tailored treatments that are aimed at ameliorating specific neurobehavioral symptoms, sleep disturbance, and pain.

With regard to the relationship between lifetime number of mTBIs and objective neuropsychological functioning in the context of military mTBI, our findings suggest that when controlling for psychiatric distress and combat exposure, a history of multiple mTBIs does not confer risk for worse cognitive performance in the years following injury. Consistent with our findings, both Dretsch and colleagues (2015) and Nelson and colleagues (2019) reported a lack of association between recurrent mTBI and cognitive performance in their respective military samples. However, others have shown an inverse relationship between the number of mTBIs and cognitive functioning (Cooper et al., 2018; Spira et al., 2014). Although the earlier studies all examined military personnel and specifically evaluated cognitive outcomes following mild TBI, it is notable that studies varied with regard to mTBI chronicity (or timing from most recent injury to study participation), sample demographics (e.g., history of combat exposure/deployment history, degree of psychiatric comorbidities, etc.), control group characteristics, and neuropsychological variables assessed. Certainly, these methodological differences may account for the observed inconsistencies across studies.

To reduce methodological concerns in our study, we explicitly set out to evaluate Veterans with remote mTBIs and adjusted for psychiatric distress and military-specific factors such as combat exposure, which have been shown to influence clinical outcomes. Taking this approach, our findings add to the current literature by establishing that the number of lifetime mTBIs does not appear to be linked to objective cognitive functioning in combat-exposed Veterans in the chronic phase following mTBI (i.e., several years post-injury) and further corroborates findings from the broader mTBI literature showing relatively rapid resolution of cognitive dysfunction (Rohling et al., 2011; Schretlen & Shapiro, 2003). Moreover, our findings are consistent with a meta-analysis conducted on concussed athletes showing a minimal impact of multiple mTBIs on neuropsychological functioning (Belanger, Spiegel, & Vanderploeg, 2010b). Nevertheless, our findings will need to be replicated using larger samples, and future studies may consider using more novel approaches to evaluate cognitive functioning such as calculating cognitive impairment scores or examining patterns of intraindividual cognitive variability, as these approaches may be more sensitive to potentially subtle group differences in cognitive functioning (Merritt et al., 2018). Furthermore, given the controversy regarding the relationship between repetitive mTBIs and susceptibility to neurodegenerative disorders in late life (DeKosky & Asken, 2017; Fields, Didehbani, Hart, & Cullum, 2019; Stern et al., 2011; Zhang et al., 2019), longitudinal studies are needed to better understand associations between the number of mTBIs and cognitive functioning in the context of aging Veterans.

As for associations between lifetime number of mTBIs and neurobehavioral symptoms, our findings indicated a significant positive relationship between these variables, consistent with several prior military mTBI studies (Bryan, 2013a; Cooper et al., 2018; Dretsch et al., 2015; Miller et al., 2013; Spira et al., 2014). Specifically, when controlling for psychiatric distress and combat exposure, we found that Veterans with a history of 3+ mTBIs endorsed more severe neurobehavioral symptoms (as reflected by the NSI total score) and experienced a greater breadth of symptoms (as reflected by the NSI PST score) than Veterans with 1–2 or no prior mTBIs. Likewise, we showed significant associations between mTBI history and all NSI symptom clusters—vestibular, somatosensory, cognitive, and affective—such that Veterans with 3+ mTBIs endorsed greater symptoms relative to the 1–2 and 0 mTBI groups. The only exception to this was for the cognitive symptom cluster which did not significantly differ between the 3+ and 1–2 mTBI groups. However, the 3+ mTBI group reported more cognitive symptoms relative to the 0 mTBI group despite the absence of any differences in objective neuropsychological performance, adding further evidence to suggest that objective and subjective cognitive functioning are often incongruent with one another (French, Lange, and Brickell, 2014; Karr et al., 2019; Spencer, Drag, Walker, & Bieliauskas, 2010).

It is also noteworthy that when evaluating the NSI variables in the three-group (omnibus) analyses, the vestibular and somatosensory symptom clusters demonstrated the greatest effect sizes, suggesting that these specific symptoms may be particularly prominent in the context of repetitive injury. Although speculative, it is possible that these symptoms are more “neurologically based” and that Veterans with repetitive mTBI may be particularly vulnerable to developing these types of symptoms in the chronic phase of injury. Finally, as for sleep and pain-related symptoms, we showed that Veterans with a history of 3+ mTBIs also endorsed greater difficulties with sleep and pain relative to those with a history of 1–2 and 0 prior mTBIs. Interestingly, though, neurobehavioral symptoms, sleep dysfunction, and pain symptoms did not significantly differ between Veterans with a history of 1–2 mTBIs and the group with no prior mTBIs. Taken together, these results add to the growing body of research showing that there appears to be a threshold at which increasing number of mTBIs may confer substantially greater risk for persisting symptoms and potentially poorer long-term outcomes in military service members and Veterans (Dretsch et al., 2015; Spira et al., 2014).

Similar to the neuropsychological literature, prior studies examining the relationship between mTBI history and symptom reporting have utilized varying methodologies and have inconsistently controlled for potentially important contributing variables such as psychiatric distress and combat exposure. Whereas our findings generally align with the results of prior military studies that have examined mTBI burden and symptom reporting, it is important to appreciate that our findings are independent of psychiatric distress and combat exposure and thus suggest that there may be an important link between greater mTBI burden and elevated subjective symptoms that is not accounted for by psychiatric overlay or combat experiences. Furthermore, we conducted follow-up analyses controlling for additional factors including time since injury and blast-related mTBI, and all results held. Unlike some prior studies, it is also worth mentioning that we excluded Veterans who likely engaged in symptom over-reporting (as reflected by the Val-10), thus underscoring the significance of the observed group differences. While additional research is needed to confirm these findings, our results indicate that Veterans with three or more prior mTBIs may be more likely to experience clinically significant sequelae and may especially benefit from individualized interventions aimed at targeting their specific constellation of symptoms. Recent work by Kontos et al. (2018) established that a precision medicine intervention called “Targeted Evaluation, Action, and Monitoring of TBI” (or “TEAM-TBI”) was effective in reducing post-concussive symptoms as well as other post-injury sequelae in a sample of Veterans and civilians with remote mTBI; though promising, future research is necessary to ensure that interventions such as TEAM-TBI and other multidisciplinary approaches are advantageous to specific patient populations such as those with multiple, remote mTBIs.

Strengths and Limitations

Notable strengths of the current study include a focus on combat-exposed Veterans with remote (>1 year) mTBI histories to improve understanding of the long-term effects associated with repetitive injury. Moreover, we controlled for both psychiatric distress and level of combat exposure—variables known to influence clinical outcome following mTBI—so that we could determine the independent effects of multiple mTBIs on objective and subjective functioning. We also excluded Veterans who obtained invalid scores on performance and symptom validity measures and utilized rigorous and well-validated neuropsychological and self-report measures. Nevertheless, our study is not without its limitations. First, consistent with other prior studies, mTBI history was based on participant self-report and therefore may be subject to recall bias or other inaccuracies. However, a well-validated, structured clinical interview administered by trained research personnel was utilized in the present study to reduce concerns related to recall bias and accuracy of TBI details. Second, although we statistically adjusted for psychiatric distress, some items on the PCL-5, for example, overlap with items on the NSI, a primary outcome of interest. Stated differently, there is inherent overlap between symptoms of psychiatric distress and neurobehavioral symptoms, and teasing apart the etiology of these overlapping symptoms remains an ongoing challenge. Third, caution should be taken when generalizing our findings to other populations such as sports-related concussion and non-sport civilian samples. Additionally, our sample included predominantly male Veterans; thus, our findings may not extrapolate to the female Veteran population. Moving forward, it will be essential for studies to specifically establish whether females are at greater risk for experiencing poor clinical outcomes with increasing number of mTBIs. Beyond issues of generalizability, our data are cross-sectional, and thus we are unable to make inferences regarding cause and effect relationships. Finally, our sample size was relatively small, and replication of findings in a larger sample is necessary.

Conclusions

In summary, the present study highlights that history of multiple, remote mTBIs is associated with elevated subjective neurobehavioral symptoms, sleep difficulties, and pain, but is not associated with objective neuropsychological functioning in military Veterans when controlling for psychiatric distress and combat exposure. These results advance understanding of the long-term consequences of multiple mTBIs in this population and suggest that Veterans with three or more mTBIs may be at elevated risk for experiencing persisting sequelae in the years following their most recent mTBI. Future research should replicate these findings and focus on targeted interventions that may benefit Veterans with remote histories of multiple mTBIs.

Funding

This work was supported by grant funding from the Department of Defense, Chronic Effects of Neurotrauma Consortium (CENC) Award [W81XWH-13-2-0095] and the Department of Veterans Affairs CENC Award [I01 CX001135]. Additionally, Victoria Merritt received salary support during this work from a Career Development Award [IK2 CX001952] from the VA Clinical Science Research & Development Service, and Laura Crocker received salary support during this work from a Career Development Award [IK2 RX002459] from the VA Rehabilitation Research & Development Service.

Conflict of Interest

None declared.

Acknowledgements

The views, opinions, and/or findings contained in this article are those of the authors and should not be construed as an official Veterans Affairs or Department of Defense position, policy, or decision unless so designated by other official documentation.

Footnotes

References