Assessing traumatic brain injury outcome measures for long-term follow-up of community-based individuals

Objectives: To determine which outcome measures are best and least suited f or assessing long-term functional outcome of individuals with traumatic bra in injury (TBI) in the community. Design: Survey of participants in the community an average of 5 years after TBI. A battery of outcome measures was given. Setting: Community in northern California after inpatient rehabilitation. Participants: Forty-eight adult individuals with prior moderate to severe T BI. All subjects had received inpatient rehabilitation 2 to 9 years previou sly and could be reached for telephone interview. Main Outcome Measures: The Community Integration Questionnaire, Neurobehavi oral Functioning Inventory (NFI), Patient Competency Rating Scale (PCRS), L evel of Cognitive Functioning Scale (LCFS), FLM(TM) instrument, Functional Assessment Measure (FIM+FAM), Supervision Rating Scale (SRS), Disability Ra ting Scale (DRS), Revised Craig Handicap Assessment and Reporting Technique (R-CHART), and Glasgow Outcome Scale (GOS). The number of maximal scores o n each of the surveys was studied to determine which instruments continued to reveal deficits years after TBI. Results: Most individuals obtained maximum scores, ie, functional independe nce, on these scales: LCFS, FIM motor subscale and total score, R-CHART phy sical independence subscale, FIM+FAM, GOS, and the SRS. Measures with the f ewest maximum scores (<36%, measuring deficits still extant in the group) w ere the R-CHART cognition subscale and the NFI memory/attention and communi cation subscales, and employment subscales. Items, subscales, and total sco res that showed good variability and correlated most highly and frequently with other scales also demonstrating good variability were the PCRS, the DR S and FIM+FAM employment items, the R-CHART cognition subscale, and the NFI motor, memory/attention, communication, and depression subscales (the R-CH ART cognition subscale and NFI memory/attention subscale were highly correl ated with the PCRS;.84,.83). Conclusions: Measures that appeared to contribute little to assessing funct ional status of a TBI sample years postinjury were the FIM, FIM+FAM, SRS, G OS, and LCFS. Measures that showed a range of deficits across participants were DRS employability, the NFI, PCRS, and the R-CHART cognition subscale.