pmc870255220963J Head Trauma RehabilJ Head Trauma RehabilThe Journal of head trauma rehabilitation0885-97011550-509X399985751235308010.1097/HTR.0000000000001045HHSPA2074490ArticleAssessing the Relationship Between Chronic Pain and Cognition: A NIDILRR and VA TBI Model Systems Collaborative ProjectPeckhamMackenzieMPHBeaulieuCynthia LPhDHaysKaitlinDPTLundsternMarissaMPHMacIntyreBriaMSOsborneCandicePhDRabinowitzAmandaPhDServiceAllan LPhDSevignyMitchMSAbbasiKatherineMAWalkerWilliam CMDWelchAbigailBATefertillerCandaceDPT, PhDCraig Hospital Research Department, Englewood, Colorado (Ms Peckham, Dr Hays, Ms Lundstern, Ms MacIntyre, Dr Osborne, Dr Service, Mr Sevigny, Ms Welch, and Dr Tefertiller); Department of Physical Medicine and Rehabilitation, The Ohio State University College of Medicine, Columbus, Ohio (Dr Beaulieu); Department of Physical Medicine and Rehabilitation, Moss Rehabilitation Research Institute, Elkins Park, Pennsylvania (Dr Rabinowitz); and Department of Physical Medicine and Rehabilitation, School of Medicine, Virginia Commonwealth University, Richmond, Virginia (Ms Abbasi and Dr Walker).1842025Sep-Oct202503920251682025405356367Objective:

To investigate the associations between current chronic pain and cognition and current chronic head pain and cognition in individuals with traumatic brain injury (TBI).

Setting:

Community.

Participants:

A total of 1762 participants from the TBI Model Systems who endorsed experiencing current chronic pain and who completed the Current Chronic Pain survey.

Design:

Secondary analysis of a subset of data collected through a multi-site, cross-sectional observational cohort study.

Main Outcome Measures:

Cognition as measured by the Brief Test of Adult Cognition by Telephone (BTACT).

Results:

Individuals with TBI who reported current chronic pain exhibited lower cognitive performance compared to those who reported no pain. Among individuals who reported pain, greater pain intensity and pain interference were negatively associated with cognition, resulting in poorer cognitive performance. The negative association was even greater for individuals acknowledging chronic head pain compared to pain from other body locations.

Conclusion:

The negative association between current chronic pain and cognition for individuals with TBI indicates the need to consider pain intensity and pain interference as factors possibly influencing cognitive ability.

chronic paincognitiontraumatic brain injuryINTRODUCTION

Chronic pain, defined as any pain that lasts for more than 3 months, has been shown to negatively impact cognition and memory in various patient populations.1,2 Melzack and Casey’s early conceptual model of pain describes chronic pain as much broader than a sensory experience. Rather, pain is an experience characterized by a sensorially, physically, and emotionally disruptive phenomenon that influences individual perception, motivation, and affect. Pain models hypothesize that chronic pain negatively impacts attention, decision-making, and cognitive flexibility,3 and recent literature supports that, in otherwise healthy individuals, chronic pain significantly correlates with attention, memory, learning, and executive function deficits.3,4 In individuals with traumatic brain injury (TBI), studies indicate that those who report current chronic pain 1–30 years after injury also report significantly worse physical and cognitive function outcomes,5 which suggests that these individuals are impacted by pain for years after injury and likely endure the symptoms with limited medical support.6

TBI itself is associated with cognitive changes, motor deficits, and sensory abnormalities; however, there is evidence that over 50% of persons with moderate-to-severe TBI also develop debilitating chronic pain that may additively affect long-term daily function and health-related quality of life.7 Headache is one of the most common pain locations after TBI, with 71% of individuals reporting it 1 year post-injury and 33% reporting it 5 years post-injury.8 Further, migraine-like symptoms are often reported by individuals with TBI who experience chronic head pain,9 and migraine pain is associated with poorer verbal comprehension, complex attention, immediate memory, and spatial cognition.10 Musculoskeletal pain, neuropathic pain, and pain related to mobility impairment are also common after TBI and negatively impact quality of life.11

Thus, individuals with moderate-to-severe TBI experience high rates of chronic cognitive deficits and chronic pain, both of which negatively impact functional outcomes. Emerging evidence suggests a complex and multifaceted relationship between pain and cognitive function after TBI that, to date, is not well understood.12,13 Consequently, it is difficult to develop or identify existing interventions to treat pain and/or cognitive deficits after TBI. To address this knowledge gap in individuals with moderate-to-severe TBI, our study objective is twofold: to examine the associations between (1) cognition and features of chronic pain, including pain intensity and pain interference; and (2) cognition and chronic head pain.

METHODSDesign

This study is a secondary analysis of a project entitled, “The Characterization and Treatment of Chronic Pain after Moderate to Severe Traumatic Brain Injury” (referred to as the Chronic Pain Study or CPS), a multicenter, cross-sectional, observational cohort study of individuals with moderate-to-severe TBI participating in the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR) Traumatic Brain Injury Model Systems (TBIMS).14 TBIMS is a national longitudinal study that tracks outcomes after TBI across the lifespan to inform clinical practices and advance standards of care.

Participants

The inclusion criteria of the CPS were also used for this study. Participants were enrolled in the TBIMS at 1 of the 18 participating rehabilitation centers, including one Veterans Health Administration TBIMS center, and completed at least one routine follow-up interview at any of the post-injury time points occurring at years 1, 2, 5, or at 5-year intervals thereafter. TBIMS participants who required a proxy to complete the scheduled follow-up or who were non-English speaking were excluded. The additional CPS inclusion criteria have been previously published.5 The CPS sample is comprised of 3804 participants; 1762 with current chronic pain, 525 with past chronic pain, and 1517 with no pain. Small effect-size differences in sociodemographic (eg, age, sex, race, employment status, military status, living situation) and injury characteristics (injury cause and post-traumatic amnesia [PTA]) were previously reported, suggesting little clinical difference between these groups. These results can be found in Supplemental Digital Content (available at: http://links.lww.com/JHTR/A924).

The final sample for this analysis objective part (1) included 1762 CPS participants who reported having current chronic pain. Current chronic pain was defined as, “Persistent or recurring pain that lasts longer than 3 months. It includes headaches or pain anywhere in the body which occurs on more than half of the days over a 3-month period.”5 The sample for objective part (2) included 828 of the current chronic pain participants who also reported experiencing head pain, such as headache or migraine. CPS participants with head pain responded to a series of separate headache-related questions15 with the goal of gaining a heightened understanding of the relationship between head pain/migraines and functional disability.

Procedure

During TBIMS follow-up interviews, participants were educated about the CPS, asked about their willingness to participate, and verbally consented if willing. Participants were then provided the standard definition of chronic pain as stated above. They were then asked, Are you experiencing chronic pain? If they endorsed current chronic pain, the participant completed the Current Chronic Pain survey. Participants had an 8-week window to complete their assigned survey. The survey platform used was based on personal preference: telephone, mail, or an emailed survey link.

Measures

Data for this analysis were collected from the TBIMS database and the CPS. This survey consists of a variety of previously validated pain measures, including measures for headache which are described in Pain characteristics below.

Demographics.

Sociodemographic variables included age, sex, race/ethnicity, marital status, level of education, employment status, body mass index (BMI), past mental health interventions, controlled substance use, and medication/opioid use. Controlled substance use is defined as a person having either taken illicit drugs in the past 12 months, consumed 5 or more alcoholic drinks in one sitting within the past month, or consumed greater than 14 drinks in a week for males or 7 drinks for females in the past month. Opioid use was defined as having ever used prescribed opioid medication in the past or having used prescribed opioid medication in the past year that was not prescribed. Injury characteristics included injury etiology, time since injury, and days of PTA, (injury severity). Functional outcomes like communication ability, motor response, cognitive ability to complete daily living activities, and employability were measured using the Disability Rating Scale.16

Cognition.

Cognition was assessed using the Brief Test of Adult Cognition by Telephone (BTACT).17 The BTACT is a 15- to 20-minute brief measure of global cognitive function validated for telephone administration. The BTACT was administered to participants within 8 weeks prior to completing the CPS. The subscales include episodic verbal memory, working memory, verbal fluency, inductive reasoning, processing speed. For analyses, the three Midlife in the United States (MIDUS)18,19 standardized BTACT composite scores20 were selected as the primary outcomes. The MIDUS BTACT scores include an overall composite score and two subscales for episodic verbal memory and executive functioning. All scores were standardized based on the participant’s age, sex, and education level.

Pain characteristics.

The Brief Pain Inventory (BPI) was used to collect data regarding pain location, intensity, and interference. The BPI assesses the severity of pain, impact of pain on daily function, location of pain, pain interference, and pain relief. Responses are made on a Likert scale of 0 (no pain or interference) to 10 (pain as bad as you can imagine or completely interferes).21

The Headache Impact Test (HIT-6) is a self-report measure used to assess the impact of headaches on daily function. The items cover severe pain, pain-related functional limitation, tiredness, feelings of irritation, and concentration difficulties. Responses are made on a 5-point Likert scale of never to always.22 The HIT-6 was only asked of those participants who reported head pain.

The Lucas and colleagues’ headache characterization strategy23 was used to classify headache type. Headaches were classified into migraine, probable migraine, tension-type headache, and cervicogenic headache (a symptom-defined secondary headache). If head pain did not fit into these categories, it was characterized as unclassifiable. For this analysis, a participant was considered to have “migraine features” if they were classified into migraine or probable migraine. Decisions for headache classification were made based on the responses to the headache questionnaire, International Classification of Headache Disorders, 2nd edition (ICHD-2) criteria, and expert review using a decision tree. A headache was classified as a migraine if it was reported as moderate-to-severe and met at least two of the following criteria: (1) The headache had a significant disabling impact; (2) The headache was unilateral; (3) The headache was throbbing/pulsating; (4) The headache was worsened by movement or physical activity.

Psychosocial outcomes.

Sleep data were collected using the Pittsburgh Sleep Quality Index (PSQI).24 The PSQI is a self-report measure designed to measure overall sleep quality that is a combination of open-ended and Likert scale questions. Respondents indicate how frequently they experience sleep difficulties and rate their overall sleep quality. The PSQI total score ranged from 0 to 21 with higher scores indicating worse sleep quality.

Post-Traumatic Stress Disorder (PTSD) symptomology was assessed using the PTSD Checklist for DSM-5 (PCL-5).25,26 The PCL-5 is a 20-item self-report measure that assesses 20 DSM-5 symptoms of PTSD. Responses are made on a Likert scale of 0 (not at all) to 4 (extremely). The PCL-5 ranged from 0 to 80 with higher scores indicating worse PTSD symptomology.

Depression severity was measured using the Patient Health Questionnaire (PHQ-9),27 a self-report measure of depression symptoms. Respondents indicate how often they experience symptoms of depression from 0 (not at all) to 3 (nearly every day). PHQ-9 scores range from 0 to 27 with higher scores indicating worse depression symptoms.

Anxiety severity was measured by the Generalized Anxiety Disorder-7 (GAD-7),28 a self-report measure of anxiety symptoms. Respondents indicate how often they experience symptoms of anxiety from 0 (not at all) to 3 (nearly every day). GAD-7 scores range from 0 to 21 with higher scores indicating worse anxiety.

Primary insurance.

The primary insurance provider was hierarchically ordered as Veterans Affairs (VA), private insurance, Medicare, and Medicaid. If participants reported two or more coverage sources, their primary insurance source was determined based on this order.

Analytic plan

All statistical analyses were performed using SAS v9.429 assuming a significance level of 0.05 unless otherwise stated. Sample characteristics included all demographic, injury-related, psychosocial, cognitive, and pain items. The continuous variables were summarized using means and standard deviations (SDs) and frequency counts and percentages were used for the categorical variables. Medians and interquartile ranges described any skewed variables. All characteristics were summarized for both the current chronic pain and current chronic head pain samples.

BTACT scores were compared between the current chronic pain sample and those participants without chronic pain using two-sided t-tests and Cohen’s D effect sizes (ESs). Associations between pain and cognition were assessed for the current chronic pain sample using correlations, ESs, and both unadjusted and adjusted linear regression models. All three BTACT scores were used as cognitive outcomes, and pain was represented by the combination of mean BPI pain intensity and mean BPI pain interference, and head pain location (Y/N).

All adjusted models controlled for age, level of education, GAD-7, PSQI, PTA duration, primary insurance provider, and opioid use. Due to the high correlations between the GAD-7, PHQ-9, and PCL-5, only one of these psychosocial variables could be included within the multivariate models to avoid multicollinearity. Preliminary analyses revealed that the GAD-7 had the strongest association with cognition within our sample and therefore was included in our models. For analytic reasons, PTA duration was normalized using a log-transformation and a Bonferroni correction was assessed when necessary. All regression model results were summarized using mean estimate change, a 95% confidence interval, and significance levels. ESs were calculated using model estimates divided by the root mean square error of the respective model.

For those with current chronic head pain, the associations between pain and cognition were analyzed using the same methods previously described. All BTACT scores were used as outcomes, and the HIT-6 represented head pain. Consistent covariates were controlled for in adjusted models with the addition of migraine classification.

RESULTSSample description

The current chronic pain sample consisted of 1762 participants with an average age of 46 years. The majority were male (73%), white (66%), had greater than a high school diploma (60%), and had private insurance (37%). The sample has a median of 5.4 years for time since injury and PTA duration ranged from 0 to 441 days with a median of 16 days. Chronic head pain was reported in 47% of the sample, and the majority reported the use of opioids (60%), having a vehicular etiology (53%), no use of controlled substances (70%), and no pre-injury history of mental health treatment (69%). The majority of participants with current chronic head pain reported having migraine features (81%).

Median scores for depression and anxiety were 7 and 5, respectively. Average sleep quality scores were 8.6 and median PTSD scores were 19. Average cognitive outcomes were below the mean standardized score of 0. The BTACT composite score had a mean of −0.5 and ranged from −3.4 to 1.6. The average BTACT executive function was −0.4 with a range of −3.6 to 2.3, and the average BTACT episodic verbal memory was −0.7 with a range of −3.0 to 3.4. Table 1 summarizes all characteristics for both the current chronic pain and the current chronic head pain samples.

Current chronic pain associations with cognition

Cognitive scores were significantly worse for those with current chronic pain compared to those with no pain. The unadjusted associations were statistically significant for all pain predictors with all cognitive outcomes and small ESs were identified in both analyses. BTACT scores decreased on average between 0.10 and 0.14 SDs for every one-point increase in a participant’s pain intensity. Increases in pain interference demonstrated a similar relationship with cognitive scores decreasing between 0.07 and 0.10 SDs. Those participants who reported current chronic head pain also showed worse cognitive outcomes compared to those who did not. See Tables 2 and 3 respectively for results.

After adjusting for relevant characteristics that could impact cognition, significant relationships remained between all pain predictors and the BTACT composite and executive functioning scores. Pain intensity was the only variable found to significantly impact BTACT episodic verbal memory. Pain intensity and interference negatively impacted each of the cognitive outcomes and those with current chronic head pain had worse BTACT scores. All adjusted associations between pain and cognition revealed small ESs and correlations.

Controlling for pain intensity and pain interference, significant relationships were also seen with injury severity, level of education, primary insurance provider, opioid use, age, and anxiety. Participants with longer durations in PTA, less education, and those who did not report opioid use demonstrated lower cognitive scores for all outcomes. Participants with either Medicare or Medicaid as their primary insurance provider also had worse cognitive scores for each outcome when compared to those with private insurance. Increased age was associated with better cognition (BTACT composite scores) while anxiety was negatively related to episodic verbal memory.

Similar associations were seen in the adjusted models controlling for head pain location. Injury severity, level of education, and insurance provider were significantly related to all three cognitive outcomes. Lack of opioid use and anxiety were negatively related to BTACT composite and episodic verbal memory scores and those who reported poorer sleep quality demonstrated significantly worse cognition (BTACT composite and executive function scores). The full results for all adjusted models with the current chronic pain sample are shown in Table 4.

Chronic head pain associations with cognition

Unadjusted model results revealed a negative association between chronic head pain and cognition. For every one-point increase in the HIT-6, all BTACT scores decreased by 0.02 SDs on average for those who reported current chronic head pain. The HIT-6 maintained a significant relationship with the BTACT composite and executive function scores after adjusting for additional model covariates. All unadjusted and adjusted associations between the HIT-6 and cognition revealed small ESs and correlations. Significant associations were also seen between all BTACT outcomes and insurance provider; injury severity was significantly associated with BTACT composite and episodic verbal memory; and level of education was significantly associated with BTACT composite and executive functioning while opioid use was significantly associated with episodic verbal memory. Post hoc analysis evaluating pain intensity, pain interference, and head pain revealed no confounding impact with the model results. The complete unadjusted and adjusted model results for the current chronic head pain sample are described in Tables 3 and 4.

DISCUSSION

This study examined the association between chronic pain and cognition within the largest prospective cohort of persons surviving moderate-to-severe TBI to date. Individuals who reported chronic pain had worse cognitive scores compared to individuals who denied pain (n = 1517). Greater pain severity (intensity and interference) was associated with poorer cognition. Additionally, in an unadjusted comparison, individuals with head pain showed worse cognitive outcomes compared to those who denied head pain. Significant associations were observed for PTA duration, primary insurance provider, and opioid use across all three BTACT outcomes. The association with PTA duration was expected given that is the most robust clinical measure of TBI severity and overall functional prognosis,29–31 and individuals in this sample who took longer to emerge from PTA had worse cognition. Worse cognition was also found for non-opioid users when compared to opioid users. Similar covariate associations were observed for our subgroup analyses on head location pain. Worse cognition for non-opioid users is difficult to reconcile with the literature, but may represent selection bias for those prescribed opioids, or an avoidance of opioids by individuals with limited cognitive ability, substance abuse histories, or mental health conditions. Data collection included medications prescribed regardless of long-term scheduled daily use or occasional as-needed use, so this does not provide insights into actual use patterns.

Individuals with private insurance scored higher on BTACT than those with Medicare or Medicaid which is consistent with prior studies of long-term cognitive and physical functional outcomes in the TBIMS cohort.32–34 Though increased age relating to better cognition was a surprising result, it may be associated with higher levels of education reported by this sample and warrants further investigation. Anxiety was also found to negatively impact BTACT composite and episodic verbal memory scores, but not scores of executive function.

All of the above relationships remained significant after adjusting for other factors in multivariable regression analyses. Although associations were statistically significant, standardized ESs were found to be small indicating these differences in cognition may be of limited clinical significance. However, there is no literature yet to guide clinical interpretation of BTACT scores. Overall, these findings provide preliminary evidence for the impact of comorbid pain on cognition following TBI.35–37 As expected, the mean and median BTACT scores for our sample with chronic pain were below those found in normative data as full cognitive recovery after moderate-to-severe TBI is atypical.35–37 Our findings that individuals without pain after TBI had better cognitive scores than those who reported pain, had greater pain intensity or greater perceived effects of pain, or had greater impact of headache on daily life is consistent with general population data.3,4 The study data further indicate that individuals with TBI and chronic pain whose pain is located in the head are at greatest risk for reduced cognition. This suggests that scores on cognitive testing derived from individuals with TBI may not accurately reflect true cognitive ability when chronic pain is present, especially with higher pain intensity and/or head pain. Our findings support a holistic approach to clinical care, with the inclusion of pain screening and treatment or referral pathways for individuals with TBI and cognitive difficulties. Individuals with TBI and chronic pain should be counseled, along with their families, that pain may be contributing to their day-to-day cognitive symptoms, difficulties, or challenges. Cognitive-behavioral therapy may be beneficial in this regard, along with other potential behavioral strategies for minimizing the pain experience, such as biofeedback, controlled breathing, or meditation.38

LIMITATIONS

Several limitations may be considered for this analysis. First, the sample analyzed may not represent all individuals with TBI due to the exclusion criteria for the CPS.5 The results may not apply to those who experienced past chronic pain with resolution as the pain sample included only those who endorsed current chronic pain. Further, the use of self-report surveys is inherently biased as respondents may be prone to socially desirable responses, may not understand the questions, may be untruthful, may be influenced by the proctor, or may not be adept at using the survey delivery platform.39 Those who completed the survey over the phone (79% of respondents who reported current pain; 77% of respondents who reported current head pain) may have disclosed more or less information compared to those who completed the survey online or on paper. Analytically, all sample sizes for our study were quite large and may have resulted in inflated significance levels. All model results should be interpreted cautiously with an emphasis on the ESs as these are standardized measures of change. Clinical interpretation of BTACT scores has not yet been established which may limit clinical translation. Lastly, given the nature of survey-based data collection, the relationship between TBI, chronic pain, and cognition beyond the Likert-scale responses could not be explored, and further research in this area is warranted. A future study that employs a mixed-methods approach may further elucidate the complex relationship between cognitive deficits and chronic pain after TBI.

CONCLUSION

In conclusion, this study suggests that chronic pain negatively impacts cognition after TBI above and beyond other factors including TBI severity, especially among individuals who experience greater pain intensity and/or head located pain. Future studies of cognition after TBI should include pain intensity and location as possible mediating or moderating factors. Clinically, these findings suggest that pain management should be addressed as part of a holistic approach to cognitive rehabilitation after TBI.

Supplementary MaterialAcknowledgements

James A. Haley Veteran’s Administration: The views expressed in this manuscript are those of the authors and do not necessarily represent the official policy or position of the Defense Health Agency, Department of Defense, or any other U.S. government agency. This work was prepared under Contract HT0014-22-C-0016 with DHA Contracting Office (CO-NCR) HT0014 and, therefore, is defined as U.S. Government work under Title 17 U.S.C.§101. Per Title 17 U.S.C.§105, copyright protection is not available for any work of the U.S. Government. For more information, please contact dha.TBICOEinfo@health.mil.

Other Contributions:

The study authors would like to acknowledge staff at the following study sites for their efforts in recruitment, data collection, project management, and study design:

James A. Haley Veteran’s Administration: Amanda Tweed, BA, Bridget A. Cotner, PhD, George Rocek, MS, Jennifer L. Murphy, PhD, Erin Brennan, MS, Deveney Ching, MA, Jordan Moberg, MPH, Danielle R. O’Connor, MA, Curtis Takagishi, PhD, Georgia Kane, MD

Craig Hospital: Clare Morey, MA, CCC-SLP, Dave Mellick PhD, Stephanie Agtarap, PhD, William Williams, MS, Wendy Beukelman, BS, Selena Cruz, MS

University of Washington: Silas James, MPA

Indiana University School of Medicine: Amanda Melton, Christina Miller, Darby Dyar, Victoria Hammond, Grace Brackemyre

Spaulding Rehabilitation Hospital: Ross Zafonte, DO

Wayne State University School of Medicine: Carole Koviak, Renee Sun, and Robert Kotasek

Mayo Clinic College of Medicine and Science: Dmitry Esterov, DO

Rusk Rehabilitation: Tamara Bushnik, PhD, Michelle Smith, MPH, CHES, Alejandro Zarate, BS

Baylor Scott and White Institute for Rehabilitation: Simon Driver, PhD, Cynthia Dunklin, BS, Aimee Muir, MBA, Stephanie Calhoun, BS

Virginia Commonwealth University: Karen Brooke, MT, Laura Boylan, BS, Laura Albert Suarez, BA

Moss Rehabilitation Research Institute: Kelly McLaughlin, BA

University of Alabama at Birmingham: Mitchell Drew Mauldin, BS, Kay Canupp, MSN, CRNP

TIRR Memorial Herman: Jay Bogaards, MA

The Ohio State University: John D. Corrigan, PhD, Jennifer Bogner, PhD, Michael MaHaffey, BS, Shivangi Bhardwaj, BS, Ally Guiher, BS, Nathaniel Dusseau II, BS

Kessler Foundation: Nancy Chiaravalloti, PhD, Jean Lengenfelder, PhD

Carolinas Rehabilitation: Shanti Pinto, MD, Tami Pringnitz Guerrier, CBIST, CRA, Kimberly S. Welsh, BS, CBIS, Kelly Crawford, MD

JFK Johnson Rehabilitation Institute: Yelena Goldin, PhD, ABPP-CN, Monique Tremaine, PhD

Research reported in this article was funded through the National Institute on Disability, Independent Living, and Rehabilitation (NIDILRR), a Center within the Administration for Community Living (ACL), Department of Health and Human Services (HHS) Collaborative Grant Award (90DPTB0017) which leveraged the infrastructure of the NIDILRR and the Department of Veterans Affairs (VA) TBI Model Systems programs of research: James A. Haley Veterans Hospital TBI Model Systems, IRB PR00000094; see additional acknowledgement, Characterization and Treatment of Chronic Pain after TBI, 90DPTB0017, IRB PR00039496; Craig Hospital, Rocky Mountain Regional TBI Model System, 90DPTB007 (2017–2022) and 90DPTB0020 (2022–2027), IRB 231 579, Characterization and Treatment of Chronic Pain after TBI, 90DPTB0017, IRB 1 335 849; Craig Hospital TBI Model Systems National Data and Statistical Center, 90DP0084 (2016–2021) and 90DPTB0018 (2021–2026), IRB 231 626; University of Washington School of Medicine, University of Washington TBI Model System, 90DPTB0008 (2017–2022) and 90DPTB0024 (2022–2027), IRB STUDY00001788; Indiana University School of Medicine, Indiana TBI Model System, 90DPTB0002 (2017–2022) and 90DPTB0022 (2022–2027), IRB 1211010085R006; Spaulding Rehabilitation Hospital, Harvard Medical School, Spaulding-Harvard TBI Model System, 90DPTB0011 (2017–2022) and 90DPTB0027 (2022–2027), IRB 2012P002476; Wayne State University School of Medicine, Southeastern Michigan TBI System, 90DPTB006 (2017–2022) and 90DPTB0030 (2022–2027), IRB 102908B3E; Mayo Clinic College of Medicine and Science, Mayo Clinic TBI Model System, 90DPTB0012 (2017–2022) and 90DPTB0031 (2022–2027), IRB 69-03; NYU Langone Health, Rusk Rehabilitation TBI Model System at NYU and Bellevue, 90DPTB0010 (2017–2022) and 90DPTB0034 (2022–2027), IRB 13-00056; Baylor Scott and White Institute for Rehabilitation, North Texas TBI Model System, 90DPTB0013 (2017–2022) and 90DPTB0023 (2022–2027), IRB 002-212; Virginia Commonwealth University, The Virginia TBI Model System (VTBIMS), 90DP0033 (2017–2022) and 90DBTB0021 (2022–2027), IRB PR00039496; Thomas Jefferson University, Moss Rehabilitation Research Institute, Moss TBI Model Systems 90DPTB0004 (2017–2022) and 90DPTB001 (2022–2027), IRB; University of Alabama at Birmingham, UAB TBI Model System, 90DPTB0015 (2017–2022) and 90DPTB0029 (2022–2027), IRB 980 904 002; TIRR Memorial Hermann, Texas TBI Model System at TIRR, 90DPTB0016 (2017–2022) and 90DPTB0025 (2022–2027), IRB H-21 935; The Ohio State University College of Medicine, Ohio Regional TBI Model System, 90DPTB0001 (2017–2022) and 90DPTB0026 (2022–2027), IRB 1993H0142; Kessler Foundation, Northern New Jersey TBI System, 90DPTB0003 (2017–2022) and 90DPTB0032 (2022–2027), IRB R-597-07; Carolinas Rehabilitation, TBI Model System Follow-Up Center, TBI Model Systems National Data and Statistical Center, 90DP0084 (2017–2022) and 90DPTB0018 (2021–2026), IRB PR00022242; Icahn School of Medicine at Mount Sinai, The New York TBI Model System at Mount Sinai, 90DPTB0009 (2017–2022) and 90DPTB0028 (2022–2027), IRB 11-01799, IRB 22–0700; Hackensack Meridian JFK Johnson Rehabilitation, TBI Model System Projects at JFK Johnson Rehabilitation 90DPTB0014 (2017–2022), IRB CR00005522. The views expressed in this manuscript are those of the authors and do not necessarily represent the official policy or position of NIDILRR, ACL, HHS; Defense Health Agency, Department of Army/Navy/Air Force, Department of Defense (DOD); Veterans Health Administration (VHA), or any other U.S. government agency. No official endorsement should be inferred.

The authors declare no conflicts of interests.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (www.headtraumarehab.com).

REFERENCESKheraT, RangasamyV. Cognition and pain: a review. Front Psychol. 2021;12:673962. doi:10.3389/fpsyg.2021.67396234093370 MoriartyO, McGuireBE, FinnDP. The effect of pain on cognitive function: a review of clinical and preclinical research. Prog Neurobiol. 2011;93(3):385–404. doi:10.1016/j.pneurobio.2011.01.00221216272 PhelpsCE, NavratilovaE, PorrecaF. Cognition in the chronic pain experience: preclinical insights. Trends Cogn Sci. 2021;25(5):365–376. doi:10.1016/j.tics.2021.01.00133509733 SmithBH, ElliottAM, ChambersWA, SmithWC, HannafordPC, PennyK. The impact of chronic pain in the community. Fam Pract. 2001;18(3):292–299. doi:10.1093/fampra/18.3.29211356737 Harrison-FelixC, SevignyM, BeaulieuCL, Characterization and treatment of chronic pain after traumatic brain injury—comparison of characteristics between individuals with current pain, past pain, and no pain: a NIDILRR and VA TBI Model Systems Collaborative Project. J Head Trauma Rehabil. 2024;39(1):5–17. doi:10.1097/htr.000000000000091038167715 AndelicN, SobergHL, BerntsenS, SigurdardottirS, RoeC. Self-perceived health care needs and delivery of health care services 5 years after moderate-to-severe traumatic brain injury. Pm R. 2014;6(11):1013–1021. doi:10.1016/j.pmrj.2014.05.00524844444 WilliamsonML, ElliottTR, BerryJW, UnderhillAT, StavrinosD, FinePR. Predictors of health-related quality-of-life following traumatic brain injury. Brain Inj. 2013;27(9):992–999. doi:10.3109/02699052.2013.80151223781905 MoshourabRA, SchäferM, Al-ChaerED. Frontiers in Neuroengineering Chronic Pain in Neurotrauma: implications on Spinal Cord and Traumatic Brain Injury. KobeissyFH. ed.. Brain Neurotrauma: Molecular, Neuropsychological, and Rehabilitation Aspects. CRC Press/Taylor & Francis; 2015.ChenMH, SungYF, ChienWC, ChungCH, ChenJW. Risk of migraine after traumatic brain injury and effects of injury management levels and treatment modalities: a nationwide population-based cohort study in Taiwan. J Clin Med. 2023;12(4):1530.36836064 Begasse de DhaemO, RobbinsMS. Cognitive impairment in primary and secondary headache disorders. Curr Pain Headache Rep. 2022;26(5):391–404. doi:10.1007/s11916-022-01039-535239156 HoffmanJM, PagulayanKF, ZawaidehN, DikmenS, TemkinN, BellKR. Understanding pain after traumatic brain injury: impact on community participation. Am J Phys Med Rehabil. 2007;86(12):962–969. doi:10.1097/PHM.0b013e31815b5ee518090437 SanderAM, WilliamsM, LoyoK, Disparities in chronic pain experience and treatment history among persons with traumatic brain injury: a Traumatic Brain Injury Model Systems Study. J Head Trauma Rehabil. 2023;38(2):125–136.doi:10.1097/htr.000000000000087036883895 GorgoraptisN, Zaw-LinnJ, FeeneyC, Cognitive impairment and health-related quality of life following traumatic brain injury. Neuro Rehabilitation. 2019;44(3):321–331.doi:10.3233/nre-18261831177238 Traumatic Brain Injury Model Systems National Database. doi:10.17605/OSF.IO/A4XZB. http://www.tbindsc.orgRenjithV, PaiMS, CastelinoF, PaiA, GeorgeA. Clinical profile and functional disability of patients with migraine. J Neurosci Rural Pract. 2016;7(2):250–256. doi:10.4103/0976-3147.17618827114657 RappaportM, HallKM, HopkinsK, BellezaT, CopeDN. Disability rating scale for severe head trauma: coma to community. Arch Phys Med Rehabil. 1982;63(3):118–123.7073452 TunPA, LachmanME. Telephone assessment of cognitive function in adulthood: the Brief Test of Adult Cognition by Telephone. Age Ageing. 2006;35(6):629–632. doi:10.1093/ageing/afl09516943264 BrimOG, RyffCD, KesslerRC. The MIDUS National Survey: an overview. How healthy are we?: A national study of well-being at midlife. In: JohnD, and CatherineT, MacArthur Foundation Series on Mental Health and Development. Studies on Successful Midlife Development, The University of Chicago Press; 2004: 1–34.RyffC, LachmanM. Midlife in the United States (MIDUS 2): cognitive project, 2004–2006. Inter-university Consortium for Political and Social Research, Ann Arbor, MI; 2017.DiBlasioCA, SimaA, KumarRG, Research letter: performance of the Brief Test of Adult Cognition by telephone in a national sample. J Head Trauma Rehabil. 2021;36(4):E233–e239.doi:10.1097/htr.000000000000066433656486 TanG, JensenMP, ThornbyJI, ShantiBF. Validation of the Brief Pain Inventory for chronic nonmalignant pain. J Pain. 2004;5(2):133–137. doi:10.1016/j.jpain.2003.12.00515042521 YangM, Rendas-BaumR, VaronSF, KosinskiM. Validation of the Headache Impact Test (HIT-6™) across episodic and chronic migraine. Cephalalgia. 2011;31(3):357–367. doi:10.1177/033310241037989020819842 LucasS, HoffmanJM, BellKR, WalkerW, DikmenS. Characterization of headache after traumatic brain injury. Cephalalgia. 2012;32(8):600–606. doi:10.1177/033310241244522422623761 FictenbergNL, PutnamSH, MannNR, ZafonteRD, MillardAE. Insomnia screening in postacute traumatic brain injury: utility and validity of the Pittsburgh Sleep Quality Index. Am J Phys Med Rehabil. 2001;80(5):339–345. doi:10.1097/00002060-200105000-0000311327555 WeathersFW, LitzBT, KeaneTM, PalmieriPA, MarxBP, SchnurrPP. The PTSD Checklist for DSM-5 (PCL-5). Scale Available from the National Center for PTSD at www.ptsd.va.gov. 2013;10:206.Van PraagDLG, FardzadehHE, CovicA, MaasAIR, von SteinbüchelN. Preliminary validation of the Dutch version of the Posttraumatic stress disorder checklist for DSM-5 (PCL-5) after traumatic brain injury in a civilian population. PLoS One. 2020;15(4):e0231857. doi:10.1371/journal.pone.023185732310970 KroenkeK, SpitzerRL, WilliamsJB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. doi:10.1046/j.1525-1497.2001.016009606.x11556941 SpitzerRL, KroenkeK, WilliamsJB, LöweB. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. 2006;166(10):1092–1097. doi:10.1001/archinte.166.10.109216717171 ShererM, StruchenMA, YablonSA, WangY, NickTG. Comparison of indices of traumatic brain injury severity: Glasgow Coma Scale, length of coma and post-traumatic amnesia. J Neurol Neurosurg Psychiatry. 2008;79(6):678–685. doi:10.1136/jnnp.2006.11118717928328 WalkerWC, KetchumJM, MarwitzJH, A multicentre study on the clinical utility of post-traumatic amnesia duration in predicting global outcome after moderate-severe traumatic brain injury. J Neurol Neurosurg Psychiatry. 2010;81(1):87–89. doi:10.1136/jnnp.2008.16157020019222 WalkerWC, StrombergKA, MarwitzJH, Predicting long-term global outcome after traumatic brain injury: development of a practical prognostic tool using the Traumatic Brain Injury Model Systems National Database. J Neurotrauma. 2018;35(14):1587–1595.doi:10.1089/neu.2017.535929566600 Dams-O’ConnorK, KetchumJM, CuthbertJP, Functional outcome trajectories following inpatient rehabilitation for TBI in the United States: a NIDILRR TBIMS and CDC Interagency Collaboration. J Head Trauma Rehabil. 2020;35(2):127–139. doi:10.1097/htr.000000000000048431033744 HowreyBT, GrahamJE, PappadisMR, GrangerCV, OttenbacherKJ. Trajectories of functional change after inpatient rehabilitation for traumatic brain injury. Arch Phys Med Rehabil. 2017;98(8):1606–1613. doi:10.1016/j.apmr.2017.03.00928392325 LequericaAH, SanderAM, PappadisMR, The association between payer source and traumatic brain injury rehabilitation outcomes: a TBI Model Systems Study. J Head Trauma Rehabil. 2023;38(1):E10–e17.doi:10.1097/htr.000000000000078135452026 DikmenSS, CorriganJD, LevinHS, MachamerJ, StiersW, WeisskopfMG. Cognitive outcome following traumatic brain injury. J Head Trauma Rehabil. 2009;24(6):430–438. doi:10.1097/HTR.0b013e3181c133e919940676 PonsfordJ, DraperK, SchönbergerM. Functional outcome 10 years after traumatic brain injury: its relationship with demographic, injury severity, and cognitive and emotional status. J Int Neuropsychol Soc. 2008;14(2):233–242. doi:10.1017/s135561770808027218282321 RuttanL, MartinK, LiuA, ColellaB, GreenRE. Long-term cognitive outcome in moderate to severe traumatic brain injury: a meta-analysis examining timed and untimed tests at 1 and 4.5 or more years after injury. Arch Phys Med Rehabil. 2008;89(12 Suppl):S69–76. doi:10.1016/j.apmr.2008.07.00719081444 LiX, FengY, XiaJ, Effects of cognitive behavioral therapy on pain and sleep in adults with traumatic brain injury: a systematic review and meta-analysis. Neural Plast. 2021;2021(1):6552246. doi:10.1155/2021/655224634804154 PaulhusDL, VazireS. The self-report method. Handbook of Research Methods in Personality Psychology. 2007;1:224–239.SAS/ACESS 9.4 Interface to ADABAS SAS Institute Inc; 2013.

Summary of the demographic, injury-related, psychosocial, cognitive, and pain characteristics for the study samples

Current chronic pain N = 1762Current chronic head pain N = 828
N Mean (SD)/Median (IQR) N Mean (SD)/Median (IQR)
BTACT composite score 1217−0.5 (0.8)526−0.6 (0.8)
BTACT executive functioning 1230−0.4 (0.8)530−0.5 (0.8)
BTACT episodic verbal memory 1269−0.7 (0.9)548−0.9 (0.9)
Mean BPI pain intensity 17414.8 (1.9)8225.1 (1.8)
Mean BPI pain interference 17274.4 (2.5)8145.0 (2.5)
HIT-6 total score --80562.3 (8.3)
Age (years)176146.4 (15.1)82843.3 (13.2)
BMI 146527.8 (5.9)62528.2 (6.1)
Time since injury (years)17625.4 (2.1–14.6)8285.5 (2.1–14.6)
PTA duration (days)170416 (3–31)79913 (1–29)
DRS 16702 (1–4)7883 (1–4)
PHQ-9 total score 16677 (3–12)7859 (5–14)
GAD-7 total score 16715 (2–10)7947 (3–13)
PCL-5 total score 168419 (8–34)78427 (14–41)
PSQI total score 16708.6 (4.6)77410.0 (4.7)
N % N %
Current chronic head pain
 No92252.7%--
 Yes82847.3%--
 (Missing)12-
Migraine classification
 Migraine features--65681.0%
 Migraine non-features--15419.0%
 (Missing)-18
Sex
 Female47827.1%23828.7%
 Male128472.9%59071.3%
Race/ethnicity
 White114965.7%49460.2%
 Black29917.1%16119.6%
 Hispanic20911.9%12214.9%
 Other925.3%445.4%
 (Missing)137
Marital status
 Single (never married)61935.2%28734.7%
 Married67638.4%33240.1%
 No longer married46426.4%20825.2%
 (Missing)31
Highest level of education
 Less than HS diploma28516.2%14217.2%
 HS diploma42224.0%19423.5%
 Some college65437.2%31638.3%
 Bachelor degree or greater39622.5%17421.1%
 (Missing)52
Employment status
 Employed/Student68739.1%34041.2%
 Unemployed25814.7%13516.4%
 Retired71940.9%30436.8%
 Other925.2%465.6%
 (Missing)63
Injury etiology
 Vehicular93953.4%38947.0%
 Fall35720.3%15719.0%
 Violence26315.0%18422.2%
 Other20011.4%9711.7%
 (Missing)31
Primary insurance provider
 Private64137.2%27433.8%
 Medicare51930.1%20725.5%
 Medicaid30917.9%16920.8%
 Veterans affairs25614.8%16119.9%
 (Missing)3717
Opioid use
 No69040.0%32039.4%
 Yes103760.0%49260.6%
 (Missing)3516
Use of controlled substances
 No114869.5%54970.7%
 Yes50430.5%22829.3%
 (Missing)11051
Pre-injury mental health treatment
 No90969.4%42667.1%
 Yes40130.6%20932.9%
 (Missing)452193

Abbreviations: BMI: body mass index; BPI: Brief Pain Inventory; BTACT: Brief Test of Adult Cognition by Telephone; DRS: Disability Rating Score; GAD-7: General Anxiety Disorder-7; HIT-6: Headache Impact Test-6; HS: High School; IQR: interquartile range; PCL-5: Post-Traumatic Stress Disorder (PTSD) Checklist for DSM-5; PHQ-9: Patient Health Questionnaire-9; PSQI: Pittsburgh Sleep Quality Index; PTA: Posttraumatic Amnesia; SD: Standard Deviation.

Comparison of MIDUS BTACT scores between participants with current chronic pain and participants with no chronic pain

Current chronic pain, N = 1762 Mean (SD)No chronic pain, N = 1517 Mean (SD)ESP-value
MIDUS BTACT composite score −0.50 (0.75)−0.38 (0.72)0.16<.001
MIDUS BTACT executive function −0.39 (0.82)−0.27 (0.78)0.15<.001
MIDUS BTACT episodic verbal memory −0.73 (0.93)−0.62 (0.95)0.12.003

Abbreviations: BTACT, Brief Test of Adult Cognition by Telephone; ES, effect size; SD, standard deviation.

Unadjusted linear regression model results

N Estimate95% CIP-valueESCorrelation
BTACT composite score
Mean BPI pain intensity1201−0.12(−0.15,−0.10)<.0010.18−0.33
Mean BPI pain interference1195−0.09(−0.11,−0.07)<.0010.13−0.31
Head pain (no vs yes)12080.21(0.13, 0.29)<.0010.28-
HIT-6a510−0.02(−0.03,−0.01)<.0010.03−0.22
BTACT executive functioning
Mean BPI pain intensity1214−0.14(−0.16,−0.12)<.0010.18−0.33
Mean BPI pain interference1208−0.10(−0.11,−0.08)<.0010.13−0.30
Head pain (No vs Yes)12210.22(0.12, 0.31)<.0010.27-
HIT-6a513−0.02(−0.03,−0.01)<.0010.03−0.22
BTACT episodic verbal memory
Mean BPI pain intensity1250−0.10(−0.13,−0.08)<.0010.11−0.21
Mean BPI Pain Interference1244−0.07(−0.09,−0.05)<.0010.08−0.20
Head pain (no vs yes)12580.21(0.11, 0.32)<.0010.23-
HIT-6a531−0.01(−0.02,−0.003).0090.01−0.11

Abbreviations: BTACT, Brief Test of Adult Cognition by Telephone; CI, confidence interval; ES, effect size; BPI: Brief Pain Inventory; HIT: Headache Impact Test.

Current chronic head pain participants only.

Adjusted linear regression model results

BTACT composite score (N = 987)BTACT executive functioning (N = 996)BTACT episodic verbal memory (N = 1027)
Estimate95% CIEstimate95% CIEstimate95% CI
Pain intensity and interference (N = 984)(N = 993)(N = 1024)
Mean BPI pain intensity−0.06(−0.09,−0.03)a−0.07(−0.10,−0.04)a−0.04(−0.08,−0.01)a
Mean BPI pain interference−0.04(−0.06,−0.01)a−0.04(−0.07,−0.1)a−0.03(−0.06, 0.002)
PTA duration (days [log])−0.08(−0.12,−0.04)a−0.06(−0.10,−0.02)a−0.11(−0.16,−0.06)a
Age (10 years)0.03(0.001, 0.06)a0.03(−0.01, 0.06)0.04(−0.003, 0.08)
Highest level of education
 Bachelor or greater vs some college−0.16(−0.28,−0.03)b−0.24(−0.37,−0.11)b0.05(−0.11, 0.21)
 Bachelor or greater vs high school Diploma0.01(−0.12, 0.14)−0.03(−0.17, 0.11)0.10(−0.07, 0.28)
 Bachelor or greater vs less than high school0.29(0.14, 0.44)b0.31(0.15, 0.47)b0.27(0.07, 0.47)b
 Some college vs high school diploma0.17(0.06, 0.28)b0.21(0.10, 0.33)b0.05(−0.09, 0.19)
 Some college vs less than high school0.45(0.33, 0.57)b0.55(0.41, 0.68)b0.22(0.05, 0.38)b
 High School diploma vs less than high school0.28(0.16, 0.41)b0.34(0.20, 0.47)b0.17(−0.001, 0.33)
Insurance provider
 Private vs medicare0.28(0.17, 0.38)b0.27(0.16, 0.38)b0.28(0.15, 0.42)b
 Private vs medicaid0.26(0.15, 0.37)b0.24(0.12, 0.37)b0.31(0.16, 0.46)b
 Private vs VA0.06(−0.20, 0.31)0.08(−0.19, 0.36)−0.04(−0.37, 0.30)
 Medicare vs medicaid−0.02(−0.14, 0.10)−0.03(−0.16, 0.10)0.02(−0.13, 0.18)
 Medicare vs VA−0.22(−0.48, 0.03)−0.19(−0.46, 0.08)−0.32(−0.65, 0.01)
 Medicaid vs VA−0.20(−0.47, 0.06)−0.16(−0.44, 0.12)−0.35(−0.69,−0.004)
Opioid use (no vs yes)−0.12(−0.21,−0.04)a−0.11(−0.20,−0.01)a−0.16(−0.28,−0.05)a
GAD-7 total score−0.01(−0.02, 0.002)−0.002(−0.01, 0.01)−0.01(−0.02,−0.002)a
PSQI total score−0.001(−0.01, 0.01)−0.004(−0.02, 0.01)0.01(−0.01, 0.02)
Head pain (N = 987)(N = 996)(N = 1027)
Head pain (no vs yes)0.10(0.01, 0.19)a0.10(0.003, 0.20)a0.10(−0.02, 0.22)
PTA duration (days [log])−0.08(−0.12,−0.05)a−0.07(−0.11,−0.03)a−0.11(−0.16,−0.06)a
Age (10 years)0.02(−0.02, 0.05)0.01(−0.02, 0.05)0.03(−0.02, 0.07)
Highest level of education
 Bachelor or greater vs some college−0.09(−0.21, 0.03)−0.17(−0.30,−0.03)0.09(−0.07, 0.25)
 Bachelor or greater vs high school diploma0.14(0.004, 0.27)0.10(−0.04, 0.24)0.19(0.02, 0.36)
 Bachelor or greater vs less than high school0.45(0.31, 0.60)b0.48(0.32, 0.64)b0.38(0.19, 0.57)b
 Some college vs high school diploma0.23(0.12, 0.33)b0.27(0.15, 0.39)b0.10(−0.04, 0.24)
 Some college vs less than high school0.54(0.42, 0.67)b0.65(0.52, 0.78)b0.29(0.13, 0.45)b
 High school diploma vs less than high school0.32(0.19, 0.45)b0.38(0.24, 0.52)b0.19(0.02, 0.36)
Insurance provider
 Private vs medicare0.31(0.21, 0.42)b0.31(0.19, 0.43)b0.31(0.17, 0.45)b
 Private vs medicaid0.28(0.17, 0.40)b0.27(0.14, 0.40)b0.33(0.18, 0.48)b
 Private vs VA0.04(−0.22, 0.30)0.07(−0.21, 0.35)−0.05(−0.38, 0.29)
 Medicare vs medicaid−0.03(−0.15, 0.09)−0.04(−0.17, 0.10)0.02(−0.14, 0.18)
 Medicare vs VA−0.27(−0.53,−0.01)−0.24(−0.52, 0.04)−0.36(−0.69,−0.02)
Medicaid vs VA−0.24(−0.51, 0.02)−0.20(−0.49, 0.09)−0.37(−0.72,−0.03)
Opioid use (no vs yes)−0.09(−0.18,−0.01)a−0.07(−0.16, 0.02)−0.14(−0.25,−0.03)a
GAD-7 total score−0.01(−0.02,−0.001)a−0.01(−0.02, 0.002)−0.02(−0.03,−0.004)a
PSQI total score−0.01(−0.02,−0.001)a−0.02(−0.03,−0.01)a−0.002(−0.02, 0.01)
HIT-6 (N = 421)(N = 424)(N = 436)
 HIT-6−0.01(−0.02,−0.004)a−0.02(−0.03,−0.01)a−0.01(−0.02, 0.01)
 Migraine (features vs non-features)−0.01(−0.19, 0.17)0.01(−0.19, 021)−0.02(−0.25, 0.21)
 PTA duration (days [log])−0.06(−0.12,−0.001)a−0.04(−0.11, 0.03)−0.08(−0.16,−0.01)a
 Age−0.03(−0.08, 0.03)−0.04(−0.10, 0.02)−0.01(−0.08, 0.05)
Highest level of education
 Bachelor or Greater vs Some College−0.33(−0.54,−0.12)b−0.40(−0.63,−0.17)b−0.09(−0.36, 0.18)
 Bachelor or greater vs high school diploma−0.09(−0.32, 0.13)−0.12(−0.37, 0.12)−0.02(−0.30, 0.27)
 Bachelor or greater vs less than high school0.16(−0.09, 0.40)0.19(−0.08, 0.45)0.14(−0.17, 0.45)
 Some college vs high school diploma0.23(0.07, 0.40)b0.28(0.10, 0.46)b0.08(−0.13, 0.28)
 Some college vs less than high school0.48(0.31, 0.66)b0.59(0.39, 0.78)b0.23(0.01, 0.46)
 High school diploma vs less than high school0.25(0.07, 0.43)b0.31(0.11, 0.51)b0.15(−0.08, 0.39)
Insurance provider
 Private vs medicare0.39(0.22, 0.56)b0.38(0.19, 0.56)b0.40(0.19, 0.61)b
 Private vs medicaid0.34(0.17, 0.51)b0.32(0.14, 0.51)b0.35(0.14, 0.57)b
 Private vs va0.10(−0.27, 0.47)0.06(−0.33, 0.45)0.06(−0.42, 0.53)
 Medicare vs medicaid−0.06(−0.27, 0.47)−0.06(−0.25, 0.13)−0.05(−0.27, 0.17)
 Medicare vs va−0.30(−0.66, 0.07)−0.32(−0.71, 0.07)−0.34(−0.81, 0.13)
 Medicaid vs va−0.24(−0.61, 0.13)−0.27(−0.66, 0.13)−0.29(−0.77, 0.19)
Opioid use (no vs yes)−0.12(−0.25, 0.01)−0.06(−0.21, 0.08)−0.22(−0.38,−0.05)a
GAD-7 total score−0.01(−0.02, 0.01)−0.01(−0.02, 0.01)−0.01(−0.03, 0.01)
PSQI total score−0.01(−0.02, 0.01)−0.01(−0.03, 0.01)−0.004(−0.02, 0.02)

Abbreviations: BPI: Brief Pain Inventory; BTACT, Brief Test of Adult Cognition by Telephone; CI, confidence interval; GAD-7: General Anxiety Disorder-7; HIT: Headache Impact Test; PSQI: Pittsburgh Sleep Quality Index; PTA, post-traumatic amnesia; VA, Veterans Affairs.

Significant results (α = 0.05).

Bonferroni correction (α = 0.05/4 = 0.0125).