Cause of Death Trends Among Persons With Spinal Cord Injury in the United States: 1960-2017.

OBJECTIVE: To identify trends in causes of death after spinal cord injury (SCI) that could enhance understanding of why life expectancy after SCI has not improved in the last 3 decades. DESIGN: Cohort study. SETTING: Twenty-nine SCI Model Systems and 3 Shriners Hospitals. PARTICIPANTS: Individuals with traumatic SCI (N=49,266) enrolled in the SCI Collaborative Survival Study Database between 1973 and 2017. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Age-standardized cause-specific SCI mortality rates and 95% confidence intervals were calculated for 5 time intervals (1960-1979, 1980-1989, 1990-1999, 2000-2009, and 2010-2017). RESULTS: A total of 17,249 deaths occurred in 797,226 person-years of follow-up. Since 2010, the highest mortality rate was for respiratory diseases, followed by heart disease, cancer, infective and parasitic diseases (primarily septicemia), and unintentional injuries. Mortality rates for respiratory diseases, cancer, stroke, urinary diseases, and digestive diseases, initially decreased significantly but remained relatively stable since 1980, whereas essentially no progress occurred for infective and parasitic diseases. Mortality rates for heart disease, pulmonary embolus, and suicide decreased significantly throughout the entire study period, but were offset by increases in mortality rates for endocrine (primarily diabetes), nutritional, and metabolic diseases, as well as unintentional injuries. From 2010 to 2017, the overall age-standardized mortality rate was 3 times higher for individuals with SCI than the general population, ranging from 27% higher for cancer to 9 times higher for infective and parasitic diseases. CONCLUSION: Improving life expectancy after SCI will require: (1) reducing mortality rates from respiratory diseases and septicemia that have remained high, (2) reversing current trends in diabetes and unintentional injury deaths, and (3) continuing to reduce mortality from heart disease and other leading causes.

Demographic and Health Profiles of People Living With Traumatic Spinal Cord Injury in the United States During 2015-2019: Findings from the Spinal Cord Injury Model Systems Database.

OBJECTIVE: To estimate the population profile of people living with traumatic spinal cord injury (TSCI) to help evaluate health care needs of this aging population. DESIGN: Cross-sectional study. SETTING: SCI Model Systems (SCIMS) centers in the United States. PARTICIPANTS: Individuals (N=20,437) who: (1) incurred a TSCI between 1972 and 2019, (2) were initially treated at one of the SCIMS centers, and (3) were alive during the period from 2015 to 2019. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Demographics, injury characteristics, health conditions, and social participation, as compared with previous estimates in 2008 and general population statistics in 2017. RESULTS: People living with TSCI during the period from 2015-2019 (mean years since injury, 18y; 79.4% male, and 62.5% White) were older (51.6 vs 45.0y) and had a higher percentage of C1-C4 (21.9% vs 17.0%) and American Spinal Injury Association Impairment Scale D injuries (31.5% vs 26.0%) compared with the 2008 TSCI population profile. Although the proportion of people with a bachelor's degree or higher was similar between the TSCI and general US populations (30.7% vs 32.0%), the employment rate was lower in the TSCI population (24.0% vs 59.5%). People are affected by various medical problems over time post TSCI. The prevalence of pain and urinary tract infection remained high over postinjury years, at 86.1% and 52.6%, respectively. Rehospitalization and depression were most common during the first year (34.9% and 22.3%, respectively), and pressure injury was more common among those 20 years or more postinjury (>30.0%). Health conditions declined with advanced age, including self-perceived health, diabetes, and institutional residence. People who survived TSCI for years, however, had relatively good degrees of independence and social participation. CONCLUSION: Study findings highlight the need for greater involvement of primary care providers and geriatricians in the continuity of care for SCI to promote healthy aging. Improvement in employment should also be the target in promoting social participation and quality of life.

Rehabilitation Length of Stay, Body Mass Index, and Functional Improvement Among Adults With Traumatic Spinal Cord Injury.

OBJECTIVE: To examine the modifying effect of body mass index (BMI) on the association between rehabilitation length of stay (LOS), severity of injury, and motor FIM (mFIM) improvement in patients with traumatic spinal cord injury (TSCI). DESIGN: Retrospective cohort study. SETTING: Seventeen SCI Model Systems (SCIMS) centers in the United States. PARTICIPANTS: A total of 3413 patients (N=3413) who had a TSCI were admitted to an SCIMS between October 2011 and August 2018. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The mFIM (12 items) improvement by discharge. Four rehabilitation LOS groups split by quartiles were created for each neurologic severity of injury group (C1-C4 level with American Spinal Injury Association Impairment Scale [AIS] ABC, C5-C8 AIS ABC, T1-S3 AIS ABC, AIS DE). The interrelationship among rehabilitation LOS, BMI, and mFIM improvement were examined using multivariate linear regressions. A stratified analysis was performed to examine the association between rehabilitation LOS and mFIM improvement by BMI status (underweight or normal weight, overweight, obesity) and neurologic groups. RESULTS: A total of 1099 (32.2%) and 821 (24.1%) patients were overweight and had obesity, respectively. Patients with obesity had less improvement in mFIM than those who were underweight or normal weight (unit of mFIM improvement, -3.71). After stratifying by BMI status, among patients with obesity, those with the longest rehabilitation LOS showed greater improvement in mFIM than those in the shortest LOS (unit of mFIM improvement=4.78). CONCLUSIONS: Longer inpatient rehabilitation LOS may benefit patients with TSCI by increasing mFIM improvement by discharge. Obesity is negatively associated with mFIM improvement.

Mortality Beyond the First Year After Spinal Cord Injury: Does Body Mass Index Matter?

OBJECTIVE: To examine the association between body mass index (BMI [calculated as weight in kilograms divided by height in meters squared]) and mortality after the first year post spinal cord injury (SCI) overall and across demographic and injury characteristics. DESIGN: Cohort study. SETTING: Sixteen Spinal Cord Injury Model Systems (SCIMS) centers. PARTICIPANTS: SCIMS Database participants age 20 years or older and having a BMI assessment during the 2007-2011 wave of data collection. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: All-cause mortality rate. Life table method and log-rank test were used to estimate and compare mortality rates across BMI groups and other factors. Cox proportional hazard regression model was conducted to estimate hazard ratio (HR) and 95% confidence interval (CI). RESULTS: A total of 2346 participants (N=2346) with SCI were classified into 1 of the 8 BMI groups: <18.5 (6.9%), 18.5-19.9 (7.3%), 20.0-22.49 (15.0%), 22.5-24.9 (18.8%), 25.0-27.49 (17.5%), 27.5-29.9 (13.2%), 30.0-34.9 (13.5%), and ≥35.0 (7.8%). Compared with people with BMI of 22.5-29.9, a higher mortality risk was observed among people with BMI<18.5 (HR, 1.76; 95% CI, 1.25-2.49), 18.5-19.9 (HR, 1.51; 95% CI, 1.06-2.15), and ≥35.0 (HR, 1.51; 95% CI, 1.11-2.07) after adjusting for confounding factors (sex, age at the time of BMI assessment, marital status, neurologic status). The U-shape BMI-mortality relationship varied by age, sex, neurologic status, and years since injury. CONCLUSIONS: To improve life expectancy after SCI, health care professionals could focus on weight management among patients with relatively low and extremely high BMI, defined by demographic and injury-related characteristics. Future studies should explore factors that contribute to such a higher mortality after SCI, including pre-existing conditions, poor diet and/or nutrition, and cardiorespiratory fitness.

A Concordance Table to Convert FIM Basic Mobility and Self-Care Scale Scores to SCI-FI/AT Scores.

OBJECTIVE: To estimate Spinal Cord Injury Functional Index Assistive Technology (SCI-FI/AT) scores from FIM motor items. DESIGN: Secondary data analysis. SETTING: Fourteen Spinal Cord Injury Model Systems (SCIMS) programs. PARTICIPANTS: Persons with traumatic spinal cord injury (SCI) discharged from inpatient rehabilitation at 14 SCIMS programs (N=1237). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: FIM motor items were matched to SCI-FI/AT domains and summary scores for each measure were developed. The kernel-based method was employed to develop a concordance table to estimate SCI-FI/AT domain summary scores from content-matched FIM motor item summary scores. We conducted analyses to compare agreement between actual SCI-FI/AT summary scores (actual SCI-FI/AT_S) and estimated SCI-FI/AT summary scores (est-SCI-FI/AT_S) for the total sample and for participants with different SCI injury categories. RESULTS: Nine FIM items matched SCI-FI/AT basic mobility and self-care domain content. Pearson correlations for actual and est-SCI-FI/AT_S scores (0.79) were adequate for using concordance linking methods. Intraclass correlation coefficient values (0.79; 95% confidence interval, 0.77-0.81) indicated moderate reliability. t tests revealed no significant differences between actual and est-SCI-FI/AT_S scores in the total sample. For almost 60% of the sample, actual and est-SCI-FI/AT_S score differences were <5 points (half of a SD). Greater differences between actual and est-SCI-FI/AT_S scores were noted for persons with tetraplegia American Spinal Injury Association Impairment Scales (AISs) A, B, and C. CONCLUSIONS: Despite differences between the FIM and SCI-FI/AT assessments, we developed a concordance table to estimate self-care and basic mobility SCI-FI/AT scores from content-matched FIM motor item scores. This concordance table allows researchers to merge FIM data with SCI-FI/AT data to analyze SCI functional outcomes at the group level. However, owing to greater differences between actual and estimated scores, the concordance table should be used with caution to interpret scores for those with cervical-level injuries AISs A, B, C.

Rehabilitation length of stay and functional improvement among patients with traumatic spinal cord injury.

STUDY DESIGN: Retrospective cohort study. OBJECTIVE: Investigate the association between rehabilitation length of stay (LOS) and motor FIM® (mFIM) between rehabilitation admission and discharge among patients with traumatic spinal cord injury (SCI). SETTING: Seventeen SCI Model Systems (SCIMS) centers in the United States. METHODS: A total of 3386 patients with traumatic SCI enrolled in the SCIMS Database from 2011 to 2018. The main outcome measure was the mean change in mFIM (12 items) between rehabilitation admission and discharge by twelve neurological categories (C1-C4 American Spinal Injury Association impairment scale (AIS) A-B, AIS C, AIS D, and C5-C8 AIS A-B, AIS C, AIS D, and T1-T10 AIS A-B, AIS C, AIS D, and T11-S3 AIS A-B, C, D). Linear regression models were applied to estimate changes across rehabilitation LOS groups (shortest LOS, quarter2, quarter3, and longest LOS) after adjusting for covariables for each neurological category. RESULTS: The mean age of study patients was 44.5 years. Patients were predominantly men (78.5%), non-Hispanic white (64.8%), and had private insurance (57.1%). The median LOS was 42 days across the entire sample. Longer LOS was associated with a higher mFIM score compared to the shortest LOS among patients with C1-C4 AIS D; C5-C8 AIS D; T1-T10 AIS A-B; and T11-S3 AIS A-B, C, and D after adjusting for demographics and clinical characteristics. CONCLUSION: Among patients with C1-C4 AIS D; C5-C8 AIS D; T1-T10 AIS A-B; and T11-S3 AIS A-B, C, and D injuries, those with longer rehabilitation stays tended to have more motor function improvement.

Factors Increasing Risk of Suicide after Traumatic Brain Injury: A State-of-the-Science Review of Military and Civilian Studies.

Primary Objective: Survey TBI literature to identify evidence of risk for post-injury suicide.Literature Selection: Search terms ((traumatic brain injury OR TBI) AND (suicidality OR suicidal behaviour OR suicidal ideation)) entered in PubMed, OVID Medline, PsychInfo, and Web of Science for papers published in print 01/01/1997 to 06/30/2019.Analysis of Literature: Authors screened abstracts, excluding duplicates and articles not meeting inclusion/exclusion criteria. Full papers were reviewed to make final exclusions. Data were extracted from 40 papers included co- and premorbid disorders, demographics, injury-related and psychological factors.Results: Persons with TBI have a higher risk for suicide than the general population. Reviewed articles reported comorbid depression and/or PTSD as risk factors for post-TBI suicide. Co- or premorbid substance misuse, sex, and sleep disturbance moderate risk. Quality of the literature was limited by sample size, the predominance of male participants, and inconsistency in reporting of findings.Conclusions: Comorbid depression and PTSD are significant post-TBI risk factors for suicide. Several variables combine to moderate or mediate TBI's connection with suicide. Civilian and military clinician cross-talk and consistent reporting of results from reproducible studies of post-TBI suicide risk factors could improve prevention and treatment efforts in veterans and civilians.

Racial and Ethnic Differences in Obesity in People With Spinal Cord Injury: The Effects of Disadvantaged Neighborhood.

OBJECTIVE: To examine the role of neighborhood in the relation between race and obesity in people with spinal cord injury (SCI). DESIGN: A cross-sectional analysis of survey data from National SCI Database linked with neighborhood data from American Community Survey by census tract. SETTING: A total of 17 SCI Model Systems centers. PARTICIPANTS: Individuals (N=3385; 2251 non-Hispanic whites, 760 non-Hispanic blacks, 374 Hispanics) who completed a follow-up assessment during 2006-2017 (mean duration of injury, 8.3±9.9y) and resided in 2934 census tracts. INTERVENTION: Not applicable. MAIN OUTCOME MEASURES: Body mass index (BMI) (kg/m2). RESULTS: The overall prevalence of obesity was 52.9% (BMI≥25.0) and 23.3% (BMI≥30.0). Hispanics were 67.0% more likely to be obese (BMI≥30.0 kg/m2) relative to non-Hispanic whites (odds ratio, 1.67; 95% confidence interval, 1.27-2.18), after controlling for demographic and injury-related characteristics. Most of the non-Hispanic blacks (66.8%) were living in neighborhoods with high concentrated disadvantaged index (CDI), compared to 35.0% of Hispanics and 9.2% of non-Hispanic whites living in this similar neighborhood status (P<.0001). After accounting for CDI, the odds of being obese in Hispanics decreased (odds ratio, 1.51; 95% confidence interval, 1.15-1.99). Regardless of race and ethnicity, people with SCI from disadvantaged neighborhoods were 42.0%-70.0% more likely to be obese than those from minimal CDI neighborhoods. CONCLUSIONS: Neighborhood characteristics partially diminish racial differences in obesity. Weight management for the SCI population should target those who are Hispanic and living in the disadvantaged neighborhoods.

Racial Differences in Weight Gain: A 5-Year Longitudinal Study of Persons With Spinal Cord Injury.

OBJECTIVE: To assess racial differences in body mass index (BMI) change over 5 years among people with spinal cord injury (SCI). DESIGN: Multicenter longitudinal study. SETTING: Spinal Cord Injury Model Systems centers. PARTICIPANTS: Individuals (N=437; 313 non-Hispanic white, 81 non-Hispanic black, and 43 Hispanic; 335 men; mean age, 41.3±13.5y) who incurred an SCI from 1974 to 2010 and completed 2 follow-up assessments within 5 years between October 1, 2006 and September 18, 2015 (mean duration of injury, 9.1±9.6y at the start of the 5-year follow-up). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: BMI (in kilograms per meters squared). RESULTS: The mean BMI of 437 participants increased from 26.4±6.3 to 27.0±6.4kg/m2 over 5 years (P=.002). The greatest increase was noted for Hispanics (2.0±5.7kg/m2; P=.02), followed by non-Hispanic whites (0.6±3.9kg/m2; P=.01) and non-Hispanic blacks (0.01±3.7kg/m2; P>.99). The differences in BMI increase across racial groups were significant (P=.03) in those with paraplegia (American Spinal Injury Association Impairment Scale A, B or C), those who were underweight or of normal weight at baseline, and those within 10 years of their injury. Such racial differences remained significant after taking into account demographic and injury characteristics. CONCLUSIONS: Our study findings provide a foundation for future research to explore risk and protective factors that contribute to racial differences in weight gain after SCI, which help alert health care professionals to a high-risk group for obesity prevention and management.

Impact of Upper Limb Motor Recovery on Functional Independence After Traumatic Low Cervical Spinal Cord Injury.

Cervical spinal cord injury (SCI) causes devastating loss of upper limb function and independence. Restoration of upper limb function can have a profound impact on independence and quality of life. In low-cervical SCI (level C5-C8), upper limb function can be restored via reinnervation strategies such as nerve transfer surgery. The translation of recovered upper limb motor function into functional independence in activities of daily living (ADLs), however, remains unknown in low cervical SCI (i.e., tetraplegia). The objective of this study was to evaluate the association of patterns in upper limb motor recovery with functional independence in ADLs. This will then inform prioritization of reinnervation strategies focused to maximize function in patients with tetraplegia. This retrospective study performed a secondary analysis of patients with low cervical (C5-C8) enrolled in the SCI Model Systems (SCIMS) database. Baseline neurological examinations and their association with functional independence in major ADLs-i.e., eating, bladder management, and transfers (bed/wheelchair/chair)-were evaluated. Motor functional recovery was defined as achieving motor strength, in modified research council (MRC) grade, of ≥ 3 /5 at one year from ≤ 2/5 at baseline. The association of motor function recovery with functional independence at one-year follow-up was compared in patients with recovered elbow flexion (C5), wrist extension (C6), elbow extension (C7), and finger flexion (C8). A multi-variable logistic regression analysis, adjusting for known factors influencing recovery after SCI, was performed to evaluate the impact of motor function at one year on a composite outcome of functional independence in major ADLs. Composite outcome was defined as functional independence measure score of 6 or higher (complete independence) in at least two domains among eating, bladder management, and transfers. Between 1992 and 2016, 1090 patients with low cervical SCI and complete neurological/functional measures were included. At baseline, 67% of patients had complete SCI and 33% had incomplete SCI. The majority of patients were dependent in eating, bladder management, and transfers. At one-year follow-up, the largest proportion of patients who recovered motor function in finger flexion (C8) and elbow extension (C7) gained independence in eating, bladder management, and transfers. In multi-variable analysis, patients who had recovered finger flexion (C8) or elbow extension (C7) had higher odds of gaining independence in a composite of major ADLs (odds ratio [OR] = 3.13 and OR = 2.87, respectively, p < 0.001). Age 60 years (OR = 0.44, p = 0.01), and complete SCI (OR = 0.43, p = 0.002) were associated with reduced odds of gaining independence in ADLs. After cervical SCI, finger flexion (C8) and elbow extension (C7) recovery translate into greater independence in eating, bladder management, and transfers. These results can be used to design individualized reinnervation plans to reanimate upper limb function and maximize independence in patients with low cervical SCI.