|
Activity Number:
|
289
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Health Policy Statistics
|
| Abstract - #300942 |
|
Title:
|
A Multiple-Step Approach to Investigate the Effectiveness of Follow-up Rehabilitation Therapy
|
|
Author(s):
|
Yong-Fang Kuo*+ and Glenn Ostir and Ling Zhang and Kenneth Ottenbacher
|
|
Companies:
|
University of Texas and University of Texas and University of Texas and University of Texas Medical Branch
|
|
Address:
|
Sealy Center on Aging, Galveston, TX, 77555-0460,
|
|
Keywords:
|
effectiveness ; cutpoint ; rehabilitation ; bootstrap
|
|
Abstract:
|
The benefit of follow-up rehabilitation therapy after hospital discharge is unknown. We proposed a multiple-step approach to investigate the benefit of follow-up therapy for patients with stroke. Data from the Uniform Data System for Medical Rehabilitation (UDSMR) were used in the analyses. The Functional Independence Measure (FIM) was administrated before hospital discharge and at follow-up (3-6 months) for each patient. We developed a graph of least square mean (and 95% CI) on FIM change score based on all possible cutpoints of discharge FIM for both therapy and nontherapy groups. The graph suggested follow-up therapy was effective among patients with a discharge FIM score <75. The graph of partial regression coefficient and t-test statistic on follow-up therapy from multiple regression models (based on all possible cutpoints of discharge FIM) indicated the degree of effectiveness declined for patients with a discharge FIM > 60. The maximum selected t-statistic was between 40 and 60 on discharge FIM. Further, sensitivity and bootstrap analyses on discharge FIM score (between 40 and 60) is being conducted to confirm the optimal cutpoint (and 95% CI).
|
