|
Activity Number:
|
260
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #305075 |
|
Title:
|
Efficient Tests for Burden of Illness
|
|
Author(s):
|
Jing Cheng*+ and Dylan Small
|
|
Companies:
|
University of Florida and University of Pennsylvania
|
|
Address:
|
1329 SW 16th Street, Room 5130, Gainesville, FL, 32610,
|
|
Keywords:
|
Burden of illness ; empirical likelihood ; vaccine trial
|
|
Abstract:
|
The burden of illness (BOI) is an efficacy measure for prevention trials of interventions which may affect both disease incidence and disease severity. The BOI assigns a score of zero if a person does not have the disease and a positive severity score if the person does have the disease. One approach to testing for an effect of an intervention on the BOI is to combine a test of whether the intervention affects disease incidence with a test of whether the intervention affects disease severity among those who have the disease. We develop a new method of combining such tests using the empirical likelihood approach which produces a more powerful test than existing tests over a wide range of distribution for the data. We apply our approach to a malaria vaccine trial.
|
