Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) Clinical Normative Data for Gulf War Veterans

Abstract

Objective

Heterogenous test batteries and methods applied in neurocognitive research on Gulf War Veterans (GWVs) limit the translation of findings to clinical practice. A clinical data set is necessary.

Methods

Neurocognitive screening data from treatment-seeking GWVs were collected from multiple sites and compiled, informed by consideration of performance validity.

Results

Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) scores revealed the cognitive profile for GWVs (n = 189) as poorer across multiple domains when compared to similarly educated, non-veteran peers. However, mean scores generally remained within normal clinical limits. Data tables are presented to establish a comparison group for use in clinical care.

Conclusions

When assessing cognitive symptoms in GWVs, attention to education level and interpretation of subtle deficits is warranted. Current results highlight the importance of nuanced translation of neurocognitive research findings into clinical practice with GWVs.

Introduction

Although it has been 30 years since Operations Desert Shield and Desert Storm (ODS/S), there is still a lack of consistent guidance regarding how to interpret Gulf War Veteran (GWV) cognitive profiles in a clinical setting. A preliminary study of neuropsychological data collected from a small sample of GWVs found that despite many cognitive complaints, the ODS/S group did not differ significantly from matched controls on standardized measures.¹ However, an impairment index (i.e., the number of tests performed at least one standard deviation below the mean of the control group) showed that GWVs performed more poorly than controls. Other early papers studying cognitive concerns in this population were published in the Journal of the American Medical Association by the Iowa Persian Gulf Study Group (1997).² In that issue, Haley, Kurt, and Hom reported the results of a factor analysis of self-reported symptoms that comprised a syndrome of “impaired cognition.”³ This syndrome was characterized by problems with attention, memory, reasoning, insomnia, depression, daytime sleepiness, and headaches. A separate syndrome of “confusion-ataxia,” characterized by problems with thinking, disorientation, balance disturbances, vertigo, and impotence and significant neuropsychological impairment in GWVs involving “abstraction, problem-solving, complex flexibility of thought, and intelligence” compared to controls was also detailed. They hypothesized that chronic neuropsychological impairment was a result of service-related neurotoxic injury.

Early reviews noted variability in the detection of cognitive dysfunction despite consistent self-reported cognitive complaints in this cohort. Significant methodological flaws and a lack of evidence for impairment were noted.^4–6 Some studies have suggested that when psychiatric variables (e.g., posttraumatic stress disorder [PTSD] and depression) or multiple comparisons are accounted for, most group differences in cognitive functioning disappear.^7,8 In contrast, other authors have presented evidence that PTSD does not influence cognitive outcomes in treatment seeking samples.⁹ Deployed GWVs have also demonstrated poorer scores compared to non-deployed GWVs in sustained attention and motor functioning overall, although the means for both groups were well within the normal range of functioning as defined by generally accepted clinical standards (i.e., 25^th-75^th percentile).¹⁰ The authors described depressive symptoms as a negative influence that affected sustained attention, but not motor function. A National Academy of Sciences report (2016) reviewed the neurocognitive and neurobehavioral research up to that point and concluded “there is inadequate/insufficient evidence to determine whether an association exists between deployment to the Gulf War and neurocognitive and neurobehavioral performance” (p. 138).¹¹

However, a more recent meta-analysis on published GWV studies utilizing cognitive measures and found significantly decreased performances in several domains including attention, executive functioning, visuospatial skills, learning, and memory.¹² A subsequent review expanded on these findings to suggest that deficits in neuropsychological and motor functioning may be exacerbated by chemical exposures and comorbid mood disorders.¹³ In addition, a higher-than-expected rate of mild cognitive impairment (MCI) and a related concern about high rates of dementia in later life for a sample of GWVs have been suggested.¹⁴

Given the somewhat equivocal findings highlighted in the extant literature above, clinicians treating GWVs with cognitive concerns are presented with a nebulous evidence base for clinical decision-making. Although cognitive deficits are thought by many to be a core feature of GWI, clinically derived neurocognitive data sets of widely used measures are unavailable. The objective of the current study is to provide useful clinical comparison group data of ODS/S deployed veterans presenting for care. Clinical group data is immediately translatable in applied settings and supports both individual case conceptualization and feedback to veterans about their cognitive performance compared to their treatment-seeking peers.

It is well known in the literature that GWVs report multiple comorbid conditions, many with potential to impact cognitive performance (e.g., pain, sleep apnea, insomnia, depression, PTSD, chronic fatigue).¹⁵ It is beyond the scope of the current study to investigate the contribution of multiple conditions on cognitive outcome in this population. Rather, the current aim is to provide a clinical profile of GWV patients as they are, enabling comparisons to other clinical groups (depression, dementia, Parkinson’s disease, etc.) as is often useful in clinical practice. The question of cognitive impairment can often be addressed by comparing individual performances on objective measures to data from a healthy normative population. However, it is also important to determine whether an individual veteran’s performance differs from what we might expect from the population of GWVs presenting for care. This information is currently unknown, is not available in existing meta-analytic research, and underscores the utility of a clinically-derived approach to data collection.

Methods

Participants

The current project was approved (#01719) by the Institutional Review Board at the Washington DC Veterans Affairs Medical Center and participants gave written informed consent to participate. Authors received no specific funding for this study. The current sample included 194 GWVs deployed to ODS/S. Participants were referred to the War Related Illness and Injury Study Center (WRIISC) between 2008–2018. Of note, more than 95% of the current sample includes participants assessed between 2013–2018. The WRIISC is a national program that has three locations at VA medical centers across the United States: Washington, DC, Palo Alto, CA, and East Orange, NJ. These centers provide comprehensive clinical assessments for unexplained symptoms, chronic multi-symptom illnesses, treatment-resistant illness, or injuries potentially related to environmental exposures during military deployment. Veterans from a catchment area encompassing all U.S. states and territories are referred to the appropriate WRIISC by VA healthcare providers. Veterans from all branches of the U.S. Armed Forces are evaluated at the WRIISC, contributing to a diverse sample (Table 1).

Table 1.

GWV Sample Characteristics

Demographics (N = 194)
Sex (Percentage)	23 F (12%), 171 M (88%)
Mean Age in Years (SD)	49.4 (6.06 years)
Age Range	34–68 years
Mean Years of Education (SD)	13.93 (2.35)
Education Years Completed (Percentage of Sample)	<HS: 14 (7%) HS: 50 (26%) 13–14: 51 (27%) 15–16: 43 (22%) 17–23: 31 (16%) No response: 4 (2%)
Ethnicity (Percentage of Sample)	Caucasian: 140 (72%) African American: 19 (10%) Native Americans: 15 (8%) Hispanic: 10 (5%) Native Americans: 6 (5%) Native Hawaiian or Pacific Islander: 2 (1%) Asian: 2 (1%)
Psychological Measures (N= 189) *
PHQ-8	14.64 (5.95)
PCL Total	36.83 (19.61)
NSI	41.34 (15.43)
CDS	57.40 (18.60)

Note: SD = Standard Deviation; SF-36 PCS = SF-36 Physical Component Score; SF-36 MCS = SF-36 Mental Component Score; PHQ-8 = Modified Patient Health Questionnaire; PCL = PTSD Checklist; NSI = Neurobehavioral Symptom Inventory.

^*Five cases excluded from analysis due to incomplete data.

Data Collection

Data was collected from clinical neuropsychology assessment reports written by neuropsychologists, post-doctoral psychology fellows supervised by neuropsychologists, or psychologists with post-doctoral training in neuropsychology across three sites in Washington, DC, East Orange, NJ, and Palo Alto, CA. All cognitive data were interpreted in the context of each GWVs clinical presentation, and all participants received a written report in their medical records and completed a clinical feedback appointment to discuss results. This information was then accessed retrospectively to build the current data set, which allows access to retrospective clinical interpretation decisions from reports, including behavioral observations, cognitive test performance including performance validity testing, and psychological/psychiatric data. The initial data set included all participants with complete Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) data.¹⁶ The RBANS Effort Index (EI) was then applied per the method described by Silverberg et al. and participants with poor EI scores were removed from the dataset.¹⁷ Cases were also excluded when neuropsychological reports described behavioral observation of suboptimal engagement, diminished effort, or poor scores on a variety of performance validity measures including the Test of Memory Malingering (TOMM),¹⁸ Victoria Symptom Validity Test (VSVT),¹⁹ Medical Symptom Validity Test (MSVT),²⁰ WAIS-IV Reliable Digit Span (RDS),²¹ and Rey 15 Item Test (FIT).²² All participant data included a combination of these validity measures, but specific application varied across the sample due to individual clinician preference and clinical decision-making based on participant performance and presentation. GWVs that did not complete the full battery of psychological questionnaires were also excluded.

Gulf War Illness Classification

The current cohort consists of GWVs referred to WRIISC sites for multiple health complaints, including self-report of cognitive symptoms, limited to cases where complete RBANS data was available. From a clinical perspective, a formal medical/psychiatric GWI diagnostic classification or associated code does not currently exist. It is also important to note that there is no current consensus on GWI definition in the research literature. Of note, VA clinical case definition projects are underway, primarily utilizing WRIISC-collected data. However, to be accepted for a WRIISC evaluation, veterans must present with multiple chronic symptoms following military deployment that can include fatigue, headaches, joint pain, gastrointestinal symptoms, and cognitive concerns. It is likely that GWVs in the current clinical sample would meet IOM research criteria.²³ However, the current cohort was derived from clinical cases, and no formal classification of Gulf War Illness (GWI) per Institute of Medicine (IOM) recommendations for research was conducted at the point of their care.

Measures of Cognitive and Emotional Functioning

Available neuropsychological data for GWVs were compiled from the three WRIISC sites. Only cases with complete data from the RBANS Form A¹⁶ were included in the current sample. The RBANS is a widely used screening battery that efficiently (20–30 minutes of administration time) assesses functioning with 12 subtests in five key cognitive domains: Attention, Immediate Memory, Delayed Memory, Visuospatial/Constructional, and Language. The RBANS normative sample was based on the 1995 USA Census percentage distributions for race/ethnicity, with roughly 80% Caucasian, 12% African American, and 6% Hispanic participants.²⁴ This measure has previously demonstrated adequate reliability and convergent validity in multiple clinical samples.^25,26 All RBANS measures were administered and scored using standardized procedures detailed in the most recent manual.²⁷

The Neurobehavioral Symptom Inventory (NSI) is a measure of post-concussive symptoms (PCS) in rehabilitation settings and is a reliable and valid instrument for self-reported cognitive symptoms.²⁸ The Cognitive Difficulties Scale (CDS) is a 38-item self-report measure of subjective complaints regarding immediate and delayed memory, attention, language, temporal orientation, and psychomotor abilities.²⁹

Level of posttraumatic stress was evaluated with the use of the 17-item Posttraumatic Stress Disorder Checklist – Civilian Version (PCL-C).³⁰ This Likert-type scale assesses symptoms of PTSD based on Diagnostic and Statistical Manual of Mental Disorders, 4^th Edition criteria.³¹ The scale has high internal consistency, good test-retest reliability, and convergent and discriminant validity.^32,33 Symptoms of depression were assessed with the Patient Health Questionnaire (PHQ-8), a widely used self-report measure which has demonstrated strong construct validity.³⁴ Although this version of the measure excludes an item about suicidal ideation, it was demonstrated to have a very strong correlation (.998) with the original PHQ-9 that includes this item in a study of 1044 veterans.³⁵ The potential for suicidal ideation was addressed in-person as part of routine clinical care with patients.

Statistical Analysis

Continuous variables in the current analyses were presented as means and standard deviations (SD). Categorical variables were expressed as percentages (%) and sample size (n). For comparative analyses, the index and subtest scores for the education-matched participants were compared via independent-samples t-tests with effect sizes (Cohen’s d) for each comparison. Pearson’s two-tailed correlation coefficients were calculated to study the relationships between RBANS scores, demographic variables, and clinical variables. Two-tailed p-values <0.05 were considered statistically significant.

Results

The RBANS EI excluded 14 of 212 (7%) cases. Four of the remaining 198 (2%) cases were removed from analyses following a review of neuropsychological reports that described poor performance on performance validity measures and/or suboptimal engagement per behavioral observations. Demographic data for the resulting GWV sample (n=194) are displayed in Table 1. An additional five cases were excluded from some analyses due to incomplete psychological questionnaires.

RBANS index scores were calculated using age-based norms. An overall RBANS GWV sample profile collapsed across age and education level was plotted (Figure 1). Compared to clinical samples with neurologic and psychiatric disorders described in the RBANS normative sample, the cognitive profile of the current sample is most similar to a subsample of 13 older adults (mean age 69.9) diagnosed with major depressive disorder (MDD).²⁴ Notably, significant levels of depression in the GWV sample are reflected by the sample mean PHQ-8 score in Table 1. The total index score for the GWV group was 90.75 compared to 91.4 for the MDD group in the original RBANS sample. Other more impacted groups described in the manual including groups with Alzheimer’s disease or schizophrenia reported total index score means of 58.4 and 72.8 respectively.

Figure 1.

RBANS Mean Index Scores for GWV sample (N=194).

To facilitate clinical normative comparisons with the original RBANS sample, GWV data was split into individuals with a high school education and individuals that completed greater than a high school education. Index scores were calculated using age-band norms appropriate to each participant. No statistically significant differences in cognitive functioning were noted when GWV participants were compared by education group (Table 2). However, GWVs with more years of formal education achieved better cognitive performance. Index and Total scores (i.e., standard scores) were all well within 1 SD of the mean. Comparing standard scores, GWVs demonstrated statistically significantly worse performance on most RBANS indices and total score when compared to their education-matched normative sample peers (Table 3). RBANS scores across age ranges are presented in Table 4.

Table 2.

GWV RBANS Performance Split by Education Level

	HS or less (N=64)	>12 years (N=125)	Combined (N=189)
RBANS Index Score	Mean Score (SD)	Mean Score (SD)	Mean Score (SD)
Immediate Memory Index	87.66 (16.98)	92.08 (15.60)	90.55 (16.23)
Visuospatial Index	98.72 (16.53)	98.63 (17.37)	98.65 (17.08)
Language Index	93.31 (10.16)	94.48 (7.80)	94.07 (8.70)
Attention Index	89.99 (15.01)	90.57 (15.89)	90.37 (15.60)
Delayed Memory Index	91.31 (16.21)	92.63 (16.12)	92.17 (16.16)
Total Score Index	89.52 (12.91)	91.40 (13.68)	90.75 (13.45)

Note. RBANS Index scores are standardized, M = 100, SD = 15.

Table 3.

Comparison of RBANS Normative and GWV Samples on RBANS Index Scores

	RBANS Normative Sample Mean (SD)	Gulf War Normative Sample Mean (SD)***	Effect Size (d)
Education =Highschool
RBANS	(N=187)	(N=50)
Immediate Memory Index	99.50 (15.80)	88.12 (16.75)	0.71**
Visuospatial Index	99.00 (15.10)	96.70 (15.84)	0.15
Language Index	98.20 (13.50)	92.54 (10.89)	0.45**
Attention Index	98.40 (15.30)	89.12 (14.71)	0.61**
Delayed Memory Index	98.90 (15.20)	90.94 (15.59)	0.52**
Total Score Index	98.20 (14.50)	88.60 (12.56)	0.69**
Education >Highschool
RBANS	(N=244)	(N=125)
Immediate Memory Index	104.20 (13.40)	92.08 (15.60)	0.82**
Visuospatial Index	103.60 (14.10)	98.63 (17.37)	0.31**
Language Index	105.60 (14.70)	94.48 (7.80)	0.96**
Attention Index	104.50 (13.70)	90.57 (15.89)	0.92**
Delayed Memory Index	103.80 (14.20)	92.63 (16.12)	0.72**
Total Score Index	106.00 (13.50)	91.40 (13.68)	1.06**

Note:

^*= p < .05

^**= p < .01

^***14 cases that with less than a HS diploma not included in this analysis to allow for direct comparison to the RBANS sample.

Table 4.

GWV RBANS Performance Split by Age

	34–44 (N=40)	45–59 (N=136)	60–68 (N=18)
RBANS Index Score	Mean Score (SD)	Mean Score (SD)	Mean Score (SD)
Immediate Memory Index	94.05 (16.42)	89.67 (15.82)	90.33 (18.83)
Visuospatial Index	94.75 (18.03)	100.23 (16.36)	97.06 (20.23)
Language Index	93.70 (7.68)	93.63 (9.08)	97.06 (8.99)
Attention Index	90.08 (14.18)	90.38 (16.51)	90.17 (10.66)
Delayed Memory Index	91.13 (18.88)	91.35 (15.41)	101.56 (12.57)
Total Score Index	90.23 (13.84)	90.64 (13.51)	93.50 (12.52)

Note. RBANS Index scores are standardized, M = 100, SD = 15.

Post hoc analysis revealed that when all cases, including 18 participants that were found to have suboptimal effort per the RBANS EI and/or neuropsychology report review were included in analyses, the RBANS Total Index Score (i.e., Standard Score) for GWVs fell to 88.92 (SD = 14.45) (z-score −0.75). Isolating performance for just the group of 18 suspected sub-optimal effort cases was a Standard Score of 66.15, which is more than two standard deviations below the mean. There were no significant differences between education groups on the PCL, PHQ-8, the NSI, or CDS.

Correlation analyses were applied to assess the relationship between RBANS performance and both demographic variables and psychological health indicators. Years of education was not associated with RBANS Total Index score. Depressive symptoms (PHQ-8) were negatively associated with RBANS Total Index scores (r = −.218, p < .01), PCL Total scores (r = −.263, p < .01), the Cognitive Difficulties Scale (CDS) (r = −.235, p < .01), and scores on the Neurobehavioral Symptom Inventory (NSI) (r = −.311, p < .01).

Discussion

The current study informs and refines clinical care of GWVs by providing tailored clinical normative comparison data (Tables 2–3) for healthcare providers working with this unique population. The current results demonstrate that subtle but statistically significant cognitive differences across multiple cognitive indices exist regardless of education and GWVs perform significantly poorer than the similarly educated RBANS normative sample. However, GWV performance on cognitive measures generally remained within the Average/expected range of clinical classification.

It is possible that subtle cognitive changes in GWVs are emphasized when matching to a higher education civilian normative group, where improved cognitive performance is generally expected with increased education. When interpreting results in a clinical setting, a college-educated GWV scaled score that may initially appear normal (e.g., a standard score of 91.04) may reflect a significant discrepancy compared to education-matched healthy norms where an average standard score over 100 is common (Table 3), potentially reflecting a clinically significant decline in cognitive function. This 10+ point standard score difference may reflect cognitive symptoms commonly reported by GWVs.¹ It is possible that GWVs are perceiving a subtle reduction in cognitive ability, however their performance on objective testing still appears within the “Average Range” and can be clinically considered “intact.” However, it is important to note when comparing treatment-seeking to normative samples, slightly poorer performance, which can be due to multiple variables associated with ill populations seeking help (i.e., coping with chronic disease, depression, pain, fatigue) is expected.

Results partially support the findings of Janulewicz et al., whose meta-analytic effort was an important step towards ameliorating the significant heterogeneity in tests and methods that preceded their study.¹² The advantage of current study is homogeneity of methods across the entire study sample and age-correction of raw scores that allowed for direct comparison to the original normative test sample. Meta-analytic studies have inherent bias that cannot be avoided (i.e., publication bias in that studies that tend to show significant findings tend get published). Our findings revealed statistically significant but subtle differences which may otherwise go unnoticed in some clinical settings, given that results generally fall within one standard deviation from the mean, which is commonly considered “normal” performance. While our results may reflect important deviations, they do not support multiple studies that suggest pronounced and clinically significant cognitive impairment across multiple domains of functioning for GWVs.¹³ Our findings do not indicate cognitive impairment in GWVVs. Some previous studies that included data from neurocognitive measures in the GWV population did not describe efforts to assess and exclude cases based on performance validity. As recommended in a recent review of the GWV neuropsychological literature, the current study included effort measures to allow for the exclusion of cases with less-than-optimal effort, increasing the validity of our data. Our post-hoc analysis suggests that identifying and excluding profiles with validity concerns could influence conclusions about cognitive deficits that appear in the published literature.

Although the scope of the current study was not intended to determine a specific etiology for any cognitive differences in this population, it is notable that the overall RBANS neurocognitive profile for the GWV sample is most consistent with a small subgroup of older adults (mean age 70 years) diagnosed with major depressive disorder selected from the RBANS normative sample (Figure 1). Although the small size (13 participants) of this subsample should be underscored as a major limitation and there are other important differences between this subsample and the present sample that limit direct comparison (e.g., mean age difference and differing levels of comorbidity and diagnostic complexity between groups), recent studies have found that the RBANS is sensitive to the neuropsychological deficits found in current and remitted depression.^36,37 In addition, a comparison between depressed deployed and non-deployed GWVs found deployed veterans had significantly higher rates of comorbid cognitive dysfunction than non-deployed controls.³⁸ Although the potential for depressive symptoms to go underreported or misattributed to other deployment-related concerns in this population has previously been discussed,³⁹ the possible contribution of specific GW exposures (e.g., organophosphate pesticides) to cognitive and emotional functioning (e.g., mood complaints should also be considered.⁴⁰

Given that myalgic encephalomyelitis/chronic fatigue syndrome, fibromyalgia, and irritable bowel syndrome are medical conditions that fall under the umbrella of Gulf War Illness and are each independently associated with cognitive dysfunction, it is important to note that the present findings are also somewhat consistent with the literature on cognitive functioning in these conditions. Chronic fatigue syndrome is known to be associated with difficulties in attention, memory, and reaction time, including decreased cognitive capacity following physical exercise, and some of these difficulties may be more apparent subjectively to the patient than on formal neurocognitive testing.^41–43 Fibromyalgia is known to be associated with difficulties in concentration, pain attention, wakefulness, effort, working memory, episodic memory, semantic memory, verbal fluency, and executive function.⁴⁴ Irritable bowel syndrome has been conceptualized as a microbiome-gut-brain axis disorder that can include neuropsychiatric manifestations such as mood disorders, anxiety, and cognitive symptoms.⁴⁵ These findings highlight the importance of addressing cognitive concerns as part of routine clinical care for veterans with chronic multisymptom illnesses.

Current results cannot be interpreted as a complete neurocognitive profile for GWVs, or an adequate substitution for data from a more comprehensive evaluation, especially given that the RBANS is intended to be a screening measure and normative data for some RBANS measures (e.g., Line Orientation and List Recall) is non-normally distributed. In addition, direct measurement of executive functioning, an especially sensitive domain of cognitive functioning that requires careful assessment and nuanced interpretation, is difficult with current RBANS subtests. A method of analyzing errors on the RBANS to assess for executive dysfunction has been developed, but this approach is not yet established as a replacement for more direct measures of this complex construct.⁴⁶ In addition, as the RBANS can be completed in less than an hour, certain concerns common in this population fluctuate and may not present during the short window during which the GWV is being assessed (e.g., cognitive fatigue may require longer than 30 minutes to manifest). A common data elements (CDE) project for research measures has recently been completed, which may further address current heterogeneity in cognitive testing that may have confounded previous efforts to study neuropsychological differences in this population.⁴⁷ Although the RBANS was not included in recommended research CDEs, the current data can refine RBANS utility for neuropsychological assessment of GWVs given its widespread use in neuropsychology clinics. It is also our opinion that future GWV research CDEs include the RBANS to support translational study of GWI where bedside findings also influence research measurement. An additional limitation of the current study is the lack of a GWV comparison group. Although control participant inclusion was limited by the clinical nature of the current setting, a comparison to GWVs that denied cognitive or emotional symptoms as opposed to published norms could have significantly strengthened current conclusions about subtle differences in cognitive functioning. Despite these limitations, results suggest that the RBANS may be sensitive enough to detect subtle cognitive differences in the GWV population, especially in individuals with greater than high school education, where better performance is expected. Current findings are in line with research describing the utility of neurocognitive screening in some difficult to diagnose populations such as older adults with mild cognitive impairment and support previous work suggesting that the RBANS is an effective screening tool with a range of populations.⁴⁸ Similar normative data tables from these studies have also been presented for clinical utility.^49,50 The clinical profile presented in the current study may aid interpretation of impaired performance beyond what would be expected from GWVs. Current results from a widely-used screening instrument suggest some evidence against expectations of severe impairment of cognitive functioning in GWVs with multiple concerns, at least in domains excluding executive functioning. This may be helpful in reducing concerns about and misinterpretations of dementia as a direct manifestation of common GWV exposures. Clinical data is an educational tool that can help explain significant cognitive decline to patients and their families and can assist with false attribution errors (e.g., Gulf War Illness vs. early-onset dementia).

Possible relationships between exposures sustained during the Gulf War, differences in brain morphology, emotional distress, and MCI as GWVs age have been explored in recent GWV studies¹⁴. As GWVs represent a closed cohort from a specific point in history, the cognitive age-corrected profile presented in the current study may serve as a useful yardstick comparison for future studies continuing this line of investigation into potential decline beyond what might be expected with normal aging (e.g., further exploration of factors related to a suggested trend towards earlier onset of MCI in this cohort). Additionally, post-9/11 veterans present with high rates of environmental exposure concerns and a chronic multisymptom illnesses that include cognitive concerns.⁵¹ Thus, findings from GWVs may be beneficial for post-9/11 veterans as well. Further inquiry into how particular comorbidities (e.g., fatigue, sleep disorder diagnoses, chronic pain) contribute to specific aspects of cognitive outcome could help further refine clinical decision-making and feedback at a more granular level.