Establishing severity levels for patient-reported measures of functional communication, participation, and perceived cognitive function for adults with acquired cognitive and language disorders.

PURPOSE: To empirically assign severity levels (e.g., mild, moderate) to four relatively new patient-reported outcome measures (PROMs) for adults with acquired cognitive/language disorders. They include the Communicative Participation Item Bank, the Aphasia Communication Outcome Measure, and Neuro-QoL's item banks of Cognitive Function (v2.0) and Ability to Participate in Social Roles and Activities (v1.0). METHOD: We conducted 17 focus groups that comprised 22 adults with an acquired cognitive/language disorder from stroke, Parkinson's disease, or traumatic brain injury; 30 care partners of an adult with an acquired cognitive/language disorder; and 42 speech-language pathologists who had experience assessing/treating individuals with those and other cognitive/language disorders. In a small, moderated focus-group format, participants completed "PROM-bookmarking" procedures: They discussed hypothetical vignettes based on PROM item responses about people with cognitive/language disorders and had to reach consensus regarding whether their symptoms/function should be categorized as within normal limits or mild, moderate, or severe challenges. RESULTS: There was generally good agreement among the stakeholder groups about how to classify vignettes, particularly when they reflected very high or low functioning. People with aphasia described a larger range of functional communication challenges as "mild" compared to other stakeholder types. Based on a consensus across groups, we present severity levels for specific score ranges for each PROM. CONCLUSION: Standardized, stakeholder-informed severity levels that aid interpretation of PROM scores can help clinicians and researchers derive better clinical meaning from those scores, for example, by identifying important clinical windows of opportunity and assessing when symptoms have returned to a "normal" range.

Identifying Health-Related Quality of Life Domains After Upper Extremity Transplantation.

OBJECTIVE: To identify the most important health-related quality of life (HRQOL) domains and patient-reported outcomes after upper extremity transplantation (UET) in individuals with upper extremity amputation. DESIGN: Verbatim audio-recordings of individual interviews and focus groups were analyzed using qualitative, grounded theory-based methods to identify important domains of HRQOL and provide guidance for outcomes measurement after UET. SETTING: Individual interviews were conducted by phone. Focus groups were conducted at 5 upper extremity vascularized composite allotransplantation (VCA) centers in the US and at an international conference of VCA experts. PARTICIPANTS: Individual phone interviews were conducted with 5 individuals with lived experience of UET. Thirteen focus groups were conducted with a total of 59 clinical professionals involved in UET. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Not applicable. RESULTS: Twenty-eight key HRQOL domains were identified, including physical functioning and medical complications, positive and negative emotional functioning, and social participation, relations, and independence. We identified key constructs for use in evaluation of the potentially substantial physical, medical, social, and emotional effects of UET. CONCLUSIONS: This study provides an overview of the most important issues affecting HRQOL after UET, including several topics that are unique to individuals with UET. This information will be used to establish systematic, comprehensive, and longitudinal measurement of post-UET HRQOL outcomes.

Overview of the Spinal Cord Injury-Functional Index (SCI-FI): Structure and Recent Advances.

The Spinal Cord Injury-Functional Index (SCI-FI) is a system of patient-reported outcome measures of functional activities developed specifically with and for individuals with spinal cord injury (SCI). The SCI-FI was designed to overcome limitations in measurement of the full range of activities and breadth of content of physical functioning commonly used in SCI research. Generic measurement tools of physical function (ie, those focused on the general population) tend to overemphasize mobility and do not contain enough items at the lower end of the functional range (eg, items appropriate for individuals with tetraplegia). The SCI-FI consists of 9 item response theory-calibrated item banks that represent relevant and meaningful item content for individuals with SCI, span a wide range of functional abilities, and subdivide physical functioning into important subdomains, including basic mobility, self-care, and fine motor function. Since the original publication of the SCI-FI in 2012, there have been significant advances in and publications on the reliability and psychometric properties of the measures. The manuscripts presented in this special section clarify the SCI-FI structure and present new research on the SCI-FI measurement system.

Linking the Spinal Cord Injury-Functional Index (SCI-FI) to the PROMIS Physical Functioning Item Bank.

OBJECTIVE: To link 3 Spinal Cord Injury-Functional Index (SCI-FI) item banks (Basic Mobility, Fine Motor Function, Self-Care) to the Patient-Reported Outcome Measurement Information System (PROMIS) Physical Function (PF) metric. DESIGN: Observational study SETTING: Six SCI Model Systems rehabilitation hospitals in the United States. PARTICIPANTS: Adults with SCI (n=855) and healthy individuals (n=730) (N=1585). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Three SCI-FI item banks (Basic Mobility, Fine Motor Function, Self-Care), PROMIS PF v1.0 item bank. RESULTS: SCI-FI item banks (including 30 items from the PROMIS PF item bank) were administered to 855 adults with SCI as part of the original SCI-FI development study. The data were used to attempt to link 3 SCI-FI banks to the PROMIS PF metric via 2 item-response theory methods: fixed-parameter calibration and separate calibration. Sixteen items common to SCI-FI and PROMIS and verified as free of differential item functioning were used as anchor items to implement the methods. Of the 3 banks, only SCI-FI Basic Mobility could be linked with sufficient precision to PROMIS PF. Comparisons of actual vs linked PROMIS PF scores and test characteristic curves suggested the fixed-parameter method provided slightly more precision than the separate calibration method. CONCLUSIONS: The linkage between PROMIS PF and SCI-FI Basic Mobility was considered satisfactory for group-level usage. Score equivalents computed from SCI-FI Basic Mobility will be useful for researchers comparing functional levels in SCI to those observed in other clinical and nonclinical groups (eg, in comparative effectiveness research).

Applying a Bookmarking Approach to Setting Clinically Relevant Interpretive Standards for the Spinal Cord Injury-Functional Index/Capacity Basic Mobility and Self-Care Item Bank Scores.

OBJECTIVES: To develop clinically relevant interpretive standards for the Spinal Cord Injury-Functional Index/Capacity (SCI-FI/C) Basic Mobility and Self-Care item bank scores. DESIGN: Modified "bookmarking" standard-setting methodology, including 2 stakeholder consensus meetings with individuals with spinal cord injury (SCI) and SCI clinicians, respectively, and a final, combined (consumers and clinicians) "convergence" meeting. SETTING: Two SCI Model System centers in the United States. PARTICIPANTS: Fourteen adults who work with individuals with traumatic SCI and 14 clinicians who work with individuals with SCI. MAIN OUTCOME MEASURES: Placement of bookmarks between vignettes based on SCI-FI Basic Mobility and Self-Care T scores. Bookmarks were placed between vignettes representing "No Problems," "Mild Problems," "Moderate Problems," and "Severe Problems" for each item bank. RESULTS: Each consensus group resulted in a single set of scoring cut points for the SCI-FI/C Basic Mobility and Self-Care item banks. The cut points were similar but not identical between the consumer and clinician groups, necessitating a final convergence meeting. For SCI-FI/C Basic Mobility, the convergence group agreed on cut scores of 61.25 (no problems/mild problems), 51.25 (mild problems/moderate problems), and 41.25 (moderate problems/severe problems). For SCI-FI/C Self-Care, the convergence group agreed on cut scores of 56.25 (no/mild), 51.25 (mild/moderate), and 38.75 (moderate/severe). CONCLUSIONS: The results of this study provide straightforward interpretive guidelines for SCI researchers and clinicians using the SCI-FI/C Basic Mobility and Self-Care instruments. These results are appropriate for the full bank, computer adaptive test, and short-form versions of the SCI-FI/C Basic Mobility and Self-Care item banks.

Validation of the Spinal Cord Injury-Functional Index for Use in Community-Dwelling Individuals With SCI.

OBJECTIVE: To evaluate the psychometric properties of the Spinal Cord Injury-Functional Index (SCI-FI) instruments in a community-dwelling sample. DESIGN: Cross-sectional study. SETTING: Community setting. PARTICIPANTS: Individuals (N=269) recruited from 6 SCI Model Systems sites. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Participants completed computer adaptive test and short form versions of 4 SCI-FI/Capacity (C) banks (ie, Ambulation, Basic Mobility, Fine Motor, Self-Care) and 1 SCI-FI/Assistive Technology (AT) bank (Wheelchair Mobility) at baseline and after 2 weeks. The Self-Report Functional Measure (SRFM) and the clinician-rated motor FIM were used to evaluate evidence of convergent validity. RESULTS: Pearson correlations, intraclass correlation coefficients, minimal detectable change, and Bland-Altman plots supported the test-retest reliability of the SCI-FI instruments. Correlations were large with the SRFM (.69-.89) and moderate-to-large for the FIM instrument (.44-.64), supporting convergent validity. Known-groups validity was demonstrated by a significant main effect of injury level on all instruments and a main effect of injury completeness on the SCI-FI/C instruments. A ceiling effect was detected for individuals with incomplete paraplegia on the Fine Motor/C and Self-Care/C Short Forms. CONCLUSION: Findings support the test-retest reliability, convergent validity, and known-groups validity of the SCI-FI/C instruments and the SCI-FI/AT Wheelchair Mobility instruments for use by community-dwelling individuals.

Physical Function Recovery Trajectories After Spinal Cord Injury.

OBJECTIVE: To explore trajectories of functional recovery that occur during the first 2 years after spinal cord injury (SCI). DESIGN: Observational cohort study. SETTING: Eight SCI Model System sites. PARTICIPANTS: A total of 479 adults with SCI completed 4 Spinal Cord Injury-Functional Index (SCI-FI) item banks within 4 months of injury and again at 2 weeks, 3, 6, 12, and 24 months after baseline assessment (N=479). INTERVENTION: None. MAIN OUTCOME MEASURES: SCI-FI Basic Mobility/Capacity (C), Fine Motor Function/C, Self-care/C, and Wheelchair Mobility/Assistive Technology (AT) item banks. RESULTS: Growth mixture modeling was used to identify groups with similar trajectory patterns. For the Basic Mobility/C and Wheelchair Mobility/AT domains, models specifying 2 trajectory groups were selected. For both domains, a majority class exhibited average functional levels and gradual improvement, primarily in the first 6 months. A smaller group of individuals made gradual improvements but had greater initial functional limitations. The Self Care/C domain exhibited a similar pattern; however, a third, small class emerged that exhibited substantial improvement in the first 6 months. Finally, for individuals with tetraplegia, trajectories of Fine Motor Function/C scores followed 2 patterns, with individuals reporting generally low initial scores and then making either modest or large improvements. In individual growth curve models, injury/demographic factors predicted initial functional levels but less so regarding rates of recovery. CONCLUSIONS: Trajectories of functional recovery followed a small number of change patterns, although variation around these patterns emerged. During the first 2 years after initial hospitalization, SCI-FI scores showed modest improvements; however, substantial improvements were noted for a small number of individuals with severe limitations in fine motor and self-care function. Future studies should further explore the personal, medical, and environmental characteristics that influence functional trajectories during these first 2 years and beyond.

Spinal Cord Injury-Functional Index/Capacity: Responsiveness to Change Over Time.

OBJECTIVE: To establish responsiveness of 3 Spinal Cord Injury-Functional Index/Capacity (SCI-FI/C) item banks in the first year after spinal cord injury (SCI). DESIGN: Longitudinal patient-reported outcomes assessment replicated through secondary analysis of an independent data set. SETTING: A total of 8 SCI Model Systems rehabilitation hospitals in the United States. PARTICIPANTS: Study 1 participants included 184 adults with recent (≤4 months) traumatic SCI and 221 community-dwelling adults (>1 year post injury) (N=405). Study 2 participants were 418 individuals with recent SCI (≤4 months) (N=418). INTERVENTIONS: In study 1, SCI-FI/C computer adaptive tests were presented in a standardized interview format either in person or by phone call at baseline and 6-month follow-up. Responsiveness was examined by comparing 6-month changes in SCI-FI scores within and across samples (recently injured vs community-dwelling) because only the recent injury sample was expected to exhibit change over time. Effect sizes were also computed. In study 2, the study 1 results were cross-validated in a second sample with recent SCI 1 year after baseline measurement. Study 2 also compared the SCI-FI/C measures' responsiveness to that of the Self-reported Functional Measure (SRFM) and stratified results by injury diagnosis and completeness. MAIN OUTCOME MEASURES: The SCI-FI Basic Mobility/C, Self-care/C and Fine Motor/C item banks (study 1 and study 2); Self-reported Functional Measure SRFM (study 2 only). RESULTS: In study 1, changes in SCI-FI/C scores between baseline and 6-month follow-up were statistically significant (P<.01) for recently injured individuals. SCI-FI Basic Mobility/C, Self-care/C, and Fine Motor/C item banks demonstrated small to medium effect sizes in the recently injured sample. In the community-dwelling sample, all SCI-FI/C effects were negligible (ie, effect size<0.08). Study 2 results were similar to study 1. As expected, SCI-FI Basic Mobility/C and Self-care/C were responsive to change for all individuals in study 2, whereas the SCI-FI Fine Motor/C was responsive only for individuals with tetraplegia and incomplete paraplegia. The SRFM demonstrated a medium effect size for responsiveness (effect size=0.65). CONCLUSIONS: The SCI-FI Basic Mobility/C and Self-care/C banks demonstrate adequate sensitivity to change at 6 months and 1 year for all individuals with SCI, while the SCI-FI/C Fine Motor item bank is sensitive to change in individuals with tetraplegia or incomplete paraplegia. All SCI-FI/C banks demonstrate stability in a sample not expected to change. Results provide support for the use of these measures for research or clinical use.

Responsiveness of the Traumatic Brain Injury-Quality of Life (TBI-QOL) Measurement System.

OBJECTIVE: To assess the responsiveness of the Traumatic Brain Injury-Quality of Life (TBI-QOL) measurement system. DESIGN: Participants completed the 20 TBI-QOL item banks and the Participation Assessment with Recombined Tools-Objective (PART-O) Productivity Subscale at baseline and 6-month follow-up assessments. Participants were categorized into 4 groups (increased productivity, unchanged productivity, and decreased productivity) based on PART-O Productivity scores. Paired sample t tests were used to compare TBI-QOL scores at baseline and 6 months, and standardized response means and Cohen's d were computed to estimate effect sizes. SETTING: Three traumatic brain injury (TBI) Model Systems rehabilitation centers in the United States. PARTICIPANTS: Two hundred one community-dwelling adults with TBI. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: 20 TBI-QOL item banks. RESULTS: As expected, given that there was no intervention, group mean TBI-QOL subdomain scores for the entire sample showed no change or small improvement over the 6-month study period. At the follow-up assessment, 72 participants reported increased productivity, 71 reported decreased productivity, and 58 reported the same level of productivity as they had 6 months prior. When compared with participants who reported unchanged or decreased productivity, participants who reported increased productivity on the PART-O subscale had clinically meaningful (d≥0.30) improvements on 7 TBI-QOL measures. The largest improvement was in the Independence subdomain (mean change, 7.06; df=0.84), with differences also observed in the Mobility, Positive Affect and Well-Being, Resilience, Grief/Loss, Ability to Participate, and Satisfaction with Participation subdomains. CONCLUSIONS: The 20 TBI-QOL item banks demonstrate responsiveness to change and measurement stability in a community-dwelling sample. Researchers may use the TBI-QOL to detect changes in HRQOL after a clinical intervention and clinicians may use it in their daily practices to monitor patient recovery.

Establishing the Factor Structure of a Health-Related Quality of Life Measurement System for Caregivers of Persons Living With Traumatic Brain Injury.

OBJECTIVES: To understand the factor structure of health-related quality of life specific to caregivers of people living with traumatic brain injury (TBI). DESIGN: Prospective, cross-sectional data collection. SETTING: Three TBI Model Systems rehabilitation hospitals, an academic medical center, and a military medical treatment facility. PARTICIPANTS: Caregivers (N=558) of people who have sustained a TBI (344 caregivers of civilians and 214 caregivers of service members or veterans; 85% women; 58% spouses; mean age, 46.12±14.07y) who have provided care for an average of 5.82±5.40 years. INTERVENTIONS: Not applicable. MAIN OUTCOMES MEASURES: The Traumatic Brain Injury Caregiver Quality of Life (TBI-CareQOL) measurement system including 10 Patient-Reported Outcomes Measurement Information System item banks (anger, anxiety, depression, social isolation, sleep disturbance, fatigue, ability to participate in social roles and activities, satisfaction with social roles and activities, emotional support, informational support) and 5 TBI-CareQOL banks (feelings of loss-self, feelings of loss-person with TBI, feeling trapped, caregiver-specific anxiety, caregiver strain). RESULTS: Confirmatory factor analysis model fit indices were compared for 14 empirically derived and 5 theoretically derived models. Confirmatory factor analysis results indicated that the best model fit was for a 6-factor model with dimensions that included mental health, social support, social participation, social isolation, physical health, and caregiver emotion. CONCLUSIONS: Results indicated that a 6-factor model provided the best model fit for health-related quality of life in caregivers of individuals with TBI. These results have utility for both research and clinical applications. Establishing the TBI-CareQOL's factor structure provides preliminary evidence of the measurement system's construct validity, helps inform the selection of measures for specific research or clinical interventions, and informs the development of composite scores.

Development and Psychometric Characteristics of the TBI-QOL Independence Item Bank and Short Form and the TBI-QOL Asking for Help Scale.

OBJECTIVE: To develop an item response theory (IRT)-calibrated, patient-reported outcome measure of subjective independence for individuals with traumatic brain injury (TBI). DESIGN: Large-scale item calibration field testing; confirmatory factor analysis (CFA) and graded response model IRT analyses. SETTING: Five TBI Model System centers across the United States. PARTICIPANTS: Adults with complicated mild, moderate, or severe TBI (N=556). OUTCOME MEASURES: Traumatic Brain Injury-Quality of Life (TBI-QOL) Independence item bank and the TBI-QOL Asking for Help scale. RESULTS: A total of 556 individuals completed 44 items in the Independence item pool. Initial factor analyses indicated that items related to the idea of "asking for help" were measuring a different construct from other items in the pool. These 9 items were set aside. Twenty-two other items were removed because of bimodal distributions and/or low item-total correlations. CFA supported unidimensionality of the remaining Independence items. Graded response model IRT analysis was used to estimate slopes and thresholds for the final 13 Independence items. An 8-item fixed-length short form was also developed. The 9 Asking for Help items were analyzed separately. One misfitting item was deleted, and the final 8 items became a fixed-length IRT-calibrated scale. Reliability was high for both measures. CONCLUSIONS: The IRT-calibrated TBI-QOL Independence item bank and short form and TBI-QOL Asking for Help scale may be used to measure important issues for individuals with TBI in research and clinical applications.

Development and Calibration of the TBI-QOL Ability to Participate in Social Roles and Activities and TBI-QOL Satisfaction With Social Roles and Activities Item Banks and Short Forms.

OBJECTIVE: To develop traumatic brain injury (TBI)-optimized versions of the Quality of Life in Neurological Disorders (Neuro-QoL) Ability to Participate in Social Roles and Activities and Satisfaction with Social Roles and Activities item banks, evaluate the psychometric properties of the item banks developed for adults with TBI, develop short form and computer adaptive test (CAT) versions, and report information to facilitate research and clinical applications. DESIGN: We used a mixed methods design to develop and evaluate Ability to Participate in Social Roles and Activities and Satisfaction with Social Roles and Activities items. Focus groups defined the constructs, cognitive interviews guided item revisions, and confirmatory factor analysis and item response theory methods helped calibrate item banks and evaluate differential item functioning related to demographic and injury characteristics. SETTING: Five TBI Model Systems centers in the United States. PARTICIPANTS: Community-dwelling adults with TBI (N=556). INTERVENTIONS: None. OUTCOME MEASURES: Traumatic Brain Injury-Quality of Life (TBI-QOL) Ability to Participate in Social Roles and Activities and TBI-QOL Satisfaction with Social Roles and Activities item banks. RESULTS: Forty-five Ability to Participate in Social Roles and Activities and 41 Satisfaction with Social Roles and Activities items demonstrated good psychometric properties. Although some of the items are new, most were drawn from analogous banks in the Neuro-QoL measurement system. Consequently, the 2 TBI-QOL item banks were linked to the Neuro-QoL metric, and scores are comparable with the general population. All CAT and short forms correlated highly (>0.90) with the full item banks and demonstrate comparable construct coverage and measurement error. CONCLUSION: The TBI-QOL Ability to Participate in Social Roles and Activities and TBI-QOL Satisfaction with Social Roles and Activities item banks are TBI-optimized versions of the Neuro-QoL Ability to Participate in Social Roles and Activities and Satisfaction with Social Roles and Activities item banks and demonstrate excellent measurement properties in individuals with TBI. These measures, particularly in CAT or short form format, are suitable for efficient and precise measurement of social outcomes in clinical and research applications.

Measuring Pain in TBI: Development of the TBI-QOL Pain Interference Item Bank and Short Form.

OBJECTIVE: To develop a pain interference item bank, computer adaptive test (CAT), and short form for use by individuals with traumatic brain injury (TBI). DESIGN: Cross-sectional survey study. SETTING: Five TBI Model Systems rehabilitation hospitals. PARTICIPANTS: Individuals with TBI (N=590). INTERVENTIONS: Not applicable. OUTCOME MEASURES: Traumatic Brain Injury-Quality of Life (TBI-QOL) Pain Interference item bank. RESULTS: Confirmatory factor analysis provided evidence of a single underlying trait (χ2 [740]=3254.030; P<.001; Comparative Fix Index=0.988; Tucker-Lewis Index=0.980; Root Mean Square Error of Approximation=0.076) and a graded response model (GRM) supported item fit of 40 Pain Interference items. Items did not exhibit differential item functioning or local item dependence. GRM calibration data were used to inform the selection of a 10-item static short form and to program a TBI-QOL Pain Interference CAT. Comparative analyses indicated excellent comparability and reliability across test administration formats. CONCLUSION: The 40-item TBI-QOL Pain Interference item bank demonstrated strong psychometric properties. End users can administer this measure as either a 10-item short form or CAT.

Determining a transitional scoring link between PROMIS® pediatric and adult physical health measures.

PURPOSE: Having independent versions of the PROMIS® scales (for Pediatric and Adults) is problematic as scores cannot be evaluated longitudinally as individuals move from childhood into adulthood. The primary aim of this research project is to use item response theory (IRT) to develop a transitional scoring link (or "crosswalk") between the PROMIS adult and pediatric physical health measures. SETTING: Sample 1 was collected at 6 rehabilitation hospitals in the U.S., and participants in Sample 2 were recruited from public health insurance programs and an online research panel. METHODS: PROMIS pediatric and adult physical function, mobility, upper extremity, fatigue, and pain measures were administered to a sample of 874 individuals aged 14-20 years old with special health needs and a sample of 641 individuals aged 14-25 years with a disability. IRT-based scores were linked using a linear approximation to calibrated projection. RESULTS: Estimated latent variable correlations ranged between 0.84 and 0.95 for the PROMIS pediatric and adult scores. Root Expected Mean Square Difference values were below the 0.08 threshold in all cases except when comparing genders on the Mobility (0.097) and Pain (0.10) scales in the special health care needs sample. Sum score conversion tables for the pediatric and adult PROMIS measures are presented. CONCLUSIONS: The linking coefficients can be used to calculate scale scores on PROMIS adult measures from pediatric measure scores and vice versa. This may lead to more accurate measurement in cross-sectional studies spanning multiple age groups or longitudinal studies that require comparable measurement across distinct developmental stages.

Measuring Fatigue in TBI: Development of the TBI-QOL Fatigue Item Bank and Short Form.

OBJECTIVE: To develop a traumatic brain injury (TBI)-specific, item response theory (IRT)-calibrated Fatigue item bank, short form, and computer adaptive test (CAT) as part of the Traumatic Brain Injury-Quality of Life (TBI-QOL) measurement system. SETTING: Five TBI Model Systems rehabilitation centers in the US PARTICIPANTS:: Adults with complicated mild, moderate, or severe TBI confirmed by medical record review. DESIGN: Cross-sectional field testing via phone or in-person interview. MAIN MEASURES: TBI-QOL Fatigue item bank, short form, and CAT. RESULTS: A total of 590 adults with TBI completed 95 preliminary fatigue items, including 86 items from the Patient-Reported Outcomes Measurement Information System (PROMIS) and 9 items from the Quality of Life in Neurological Disorders (Neuro-QOL) system. Through 4 iterations of factor analysis, 22 items were deleted for reasons such as local item dependence, misfit, and low item-total correlations. Graded response model IRT analyses were conducted on the 73-item set, and Stocking-Lord equating was used to transform the item parameters to the PROMIS (general population) metric. A short form and CAT, which demonstrate similar reliability to the full item bank, were developed. Test-retest reliability of the CAT was established in an independent sample (Pearson's r and intraclass correlation coefficient = 0.82 [95% confidence interval: 0.72-0.88]). CONCLUSIONS: The TBI-QOL Fatigue item bank, short form, and CAT provide rehabilitation researchers and clinicians with TBI-optimized tools for assessment of the patient-reported experience and impact of fatigue on individuals with TBI.

Development and Psychometric Characteristics of the TBI-QOL Communication Item Bank.

OBJECTIVE: To develop an item response theory (IRT)-based patient-reported outcome measure of functional communication for adults with traumatic brain injury (TBI). SETTING: Five medical centers that were TBI Model Systems sites. PARTICIPANTS: A total of 569 adults with TBI (28% complicated-mild; 13% moderate; and 58% severe). DESIGN: Grounded theory-based qualitative item development, large-scale item calibration testing, confirmatory factor analyses, psychometric analyses with graded response model IRT. MAIN MEASURE: Traumatic Brain Injury-Quality of Life (TBI-QOL) Communication Item Bank, version 1.0. RESULTS: From an initial pool of 48 items, 31 items were retained in the final instrument based on adequate fit to a unidimensional model and absence of bias across several demographic and clinical subgroupings. The TBI-QOL Communication Item Bank demonstrated excellent score precision (reliability ≥ 0.95) across a wide range of communication impairment levels, particularly for individuals with more severe difficulties. The TBI-QOL Communication Item Bank is available as a full item bank, fixed-length short form, and as a computerized adaptive test. CONCLUSIONS: The TBI-QOL Communication Item Bank permits precise measurement of patient-reported functional communication after TBI. Future development will validate the instrument against performance-based, clinician-reported, and surrogate-reported assessments.

Measuring Self-Reported Cognitive Function Following TBI: Development of the TBI-QOL Executive Function and Cognition-General Concerns Item Banks.

OBJECTIVE: To develop and calibrate new patient-reported outcome measures of cognitive concerns for individuals with traumatic brain injury (TBI). SETTING: Five TBI model systems rehabilitation centers in the United States. PARTICIPANTS: Adults with medically confirmed history of TBI. DESIGN: Cross-sectional survey in interview format. MAIN MEASURES: Traumatic Brain Injury-Quality of Life (TBI-QOL) Executive Function and TBI-QOL Cognition-General Concerns item banks. RESULTS: A total of 569 adults with complicated-mild, moderate, or severe TBI completed preliminary item pools, which included 65 Executive Function items and 56 Cognition-General Concerns items. Confirmatory factor analysis supported the retention of 37 Executive Function and 39 Cognition-General Concerns items. Samejima's graded response model was used to estimate item parameters for associated computer adaptive test administrations, and informed the selection of corresponding static short forms. Data from an independent sample of 77 adults with complicated-mild, moderate, or severe TBI supported the test-retest reliability of these newly developed measures. CONCLUSION: The TBI-QOL Executive Function and Cognition-General Concerns item banks provide researchers and clinicians with reliable tools for assessing patient-reported post-TBI cognitive difficulties as part of the comprehensive TBI-QOL measurement system.

Development of the TBI-QOL Headache Pain Item Bank and Short Form.

OBJECTIVE: To develop, calibrate, and evaluate the test-retest reliability of a new patient-reported outcome measure of headache pain relevant for individuals with traumatic brain injury (TBI). SETTING: Six TBI Model Systems rehabilitation centers in the United States. PARTICIPANTS: Adults with medically confirmed documentation of TBI. DESIGN: Cross-sectional calibration field testing and test-retest reliability analyses. MAIN MEASURES: Traumatic Brain Injury-Quality of Life Headache Pain item bank. RESULTS: Thirteen headache pain items were calibrated as a unidimensional measure using data from 590 participants. The new measure was reliable (α = .98; item-total correlation range: 0.71-0.91). Item parameter estimates were estimated using Samejima's Graded Response Model and a 10-item calibrated short form was created. Simulation testing confirmed that both the computer-adaptive test and the short-form administrations were equivalent to the full item bank. One- to-2-week test-retest reliability of the computer-adaptive test was high (Pearson r and intraclass correlation coefficients = 0.81). Approximately two-thirds of the sample reported at least 1 headache symptom. CONCLUSION: The Traumatic Brain Injury-Quality of Life Headache Pain item bank and short form provide researchers and clinicians with reliable measures of the subjective experience of headache symptoms for individuals with a history of TBI.

Measuring Self-Reported Physical Function in Individuals With TBI: Development of the TBI-QOL Mobility and Upper Extremity Item Banks and Short Forms.

OBJECTIVES: To describe the development and field testing of the patient-reported outcome measures of Mobility and Upper Extremity function from the Traumatic Brain Injury Quality of Life (TBI-QOL) measurement system, and to evaluate the use of computer adaptive testing. SETTING: Five rehabilitation facilities funded as part of the TBI Model Systems network. PARTICIPANTS: Individuals with complicated mild, moderate, or severe traumatic brain injury (n = 590). INTERVENTIONS: Not available. OUTCOME MEASURES: TBI-QOL Mobility and Upper Extremity item banks. RESULTS: Item response theory and factor analyses supported the unidimensionality of the Mobility and Upper Extremity banks. Descriptive statistics showed a ceiling effect for both measures. Simulated computer adaptive tests (CATs) showed that measurement precision was maintained across administration formats for both measures. The Upper Extremity CAT showed a loss of precision for individuals without impairment and that a higher number of items were required to achieve sufficiently precise measurement, compared to the Mobility CAT. CONCLUSIONS: The TBI-QOL Upper Extremity and Mobility item banks achieved good breadth of coverage, particularly among those individuals who have experienced some degree of functional limitation. The use of CAT administration minimizes respondent burden, while allowing for the comprehensive assessment of adults with TBI. The combined use of the TBI-QOL with performance-based measures could guide the development of targeted rehabilitation treatments.

Linking the GAD-7 and PHQ-9 to the TBI-QOL Anxiety and Depression Item Banks.

OBJECTIVE: To link scores on commonly used measures of anxiety (7-item Generalized Anxiety Disorder Scale; GAD-7) and depression (9-item Patient Health Questionnaire; PHQ-9) to the Traumatic Brain Injury Quality of Life (TBI-QOL) measurement system. SETTING: 5 Traumatic Brain Injury Model Systems. PARTICIPANTS: A total of 385 individuals with traumatic brain injury (TBI) (31% complicated mild; 14% moderate; and 54% severe). DESIGN: Observational cohort. MAIN MEASURES: GAD-7, PHQ-9, TBI-QOL Anxiety v1.0 and TBI-QOL Depression v1.0. RESULTS: Item response theory-based linking methods were used to create crosswalk tables that convert scores on the GAD-7 to the TBI-QOL Anxiety metric and scores on the PHQ-9 to the TBI-QOL Depression metric. Comparisons between actual and crosswalked scores suggest that the linkages were successful and are appropriate for group-level analysis. Linking functions closely mirror crosswalks between the GAD-7/PHQ-9 and the Patient-Reported Outcomes Measurement Information System (PROMIS), suggesting that general population linkages are similar to those from a TBI sample. CONCLUSION: Researchers and clinicians can use the crosswalk tables to transform scores on the GAD-7 and the PHQ-9 to the TBI-QOL metric for group-level analyses.