Effectiveness and Patient Selection in Post-Hospital Brain Injury Rehabilitation

Indiana University Indianapolis, Indianapolis IN

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): This study proposes retrospective analyses of a large database (3000+ unique cases) for post- hospital brain injury rehabilitation (PHBIR) programs from a number of diverse PHBIR programs offered by 11 rehabilitation provider organizations throughout the U.S. This database was developed with funding from the National Institute for Neurologic Diseases and Stroke (NINDS). Programs are classified in the database as intensive (aimed to improve function) or supportive (aimed to maintain functional gains) and may be offered either in residential (participants live at the facility) or nonresidential settings. The lng-term goal of the proposed research is to improve the evidence base for the practice of PHBIR through the analysis of real world programmatic methods and outcomes. The objectives are (1) to confirm that intensive PHBIR programs produce superior outcomes to supportive programs, and (2) to determine the characteristics of patients who benefit the most from various program types. Central hypotheses are that intensive programs are more effective and that the characteristics of individuals who benefit most from specific program types can be identified. The rationale for this line of research is that identifying the specific types of PHBIR programs that are of greatest benefit to specific types of patients with brain injury will improve the practce of PHBIR by providing a standardized evaluation method to guide PHBIR recommendations. The primary outcome variable of interest will be the Mayo-Portland Adaptability Inventory (MPAI-4), a widely used measure with sound, well- established psychometric properties. Data analyses will include (1) item-response theory analyses to develop parametric equivalent metrics for MPAI-4 variables for this sample, (2) statistical comparison of outcomes for intensive programs with supportive programs, (3) cluster analysis to identify distinct MPAI-4 profiles which signify pattern of disability, (4) a variety of converging methods to establish the minimal clinicaly important difference (MCID) for the MPAI- 4, and (5) linear and logistic regression analyses of demographic, injury-related, program- related and MPAI-4 severity and pattern data to determine which types of patients benefit most from which types of programs. The large sample will allow for cross-validation of appropriate analyses. The impact of accomplishing research goals using an externally valid, generalizable research methodology is to provide practical, evidence-based, and currently unavailable information that translates into a standardized evaluation to guide the practice of PHBIR.