Date of Award

1-1-2008

Embargo Period

4-22-2024

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Biostatistics, Bioinformatics, and Epidemiology

College

College of Graduate Studies

First Advisor

Kit Simpson

Abstract

Background and Purpose: Stroke is a devastating event that drastically changes the lives of many people worldwide. Stroke statistics vary across the United States with the highest rates in the southeast. Physical therapy is often required to return to normal function. Appropriateness for physical therapy is based on individual impairments, thus access to and amount of physical therapy should not differ between races. Methods: Using a South Carolina UB-92 data subset for 2002, univariate analysis is performed against physical therapy and race for ischemic stroke victims. Logistic regression is used to test for differences in access to physical therapy. Multiple linear regression is then used to assess differences between racial groups in the amount of physical therapy received. Goodness of fit is tested and results are interpreted and presented. Results: After controlling for all covariates, race is not associated with access to physical therapy care. Variables associated with access to physical therapy are insurance, intensive care unit, length of stay, age, and Charlson score. Hosmer and Lemeshow test is significant (p<0.01) with 75% concordance. Multiple linear regression results indicate that blacks receive $60 (p<0.01) less physical therapy than whites. Significant variables include length of stay, age, gender, and hospital size. The adjusted R-square is 0.35. When individuals that did not survive are removed from the analysis the results remain similar with an R-square of 0.45. Although there is no significant difference in access to physical therapy, the whites receive more physical therapy than blacks. The under-utilization of physical therapy for African American patients may be IV substantial given their high stroke burden and the resulting greater rehabilitation need for this group. Further studies are indicated to examine physical therapy treatment rates in surrounding states for stroke and other common cardiovascular conditions.

Rights

Copyright is held by the author. All rights reserved.

Download

Share

COinS