|
Activity Number:
|
175
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology*
|
| Abstract - #300955 |
|
Title:
|
Comparison of Population Attributable Risk Estimates
|
|
Author(s):
|
Jayawant Mandrekar*+ and Melvin Moeschberger
|
|
Affiliation(s):
|
Ohio State University and Ohio State University
|
|
Address:
|
B107 Starling-Loving Hall, 320 W. 10th Avenue, Columbus, Ohio, 43210-1240, USA
|
|
Keywords:
|
population attributable risk ; epidemiologic methods ; simulations
|
|
Abstract:
|
One of the important questions of interest in the field of public health is to find out the amount of disease burden in a population that could be eliminated or reduced if risk factors were eliminated or reduced from a population. To address this question, epidemiologists calculate the population attributable risk (PAR), which helps determine the exposures that have the most relevance to the health of a community, and thus provides guidance to policy makers in planning public health interventions. The commonly used formula to calculate the PAR in the literature is the one that is proposed by Levin (1953). This formula is not specific to any type of study design and also not appropriate when there is confounding. We will compare the PAR estimates obtained using Levin's formula and using different formulas, under various scenarios using simulations. Recommendations regarding use of PAR formulas will be suggested. Applications to some large-scale national data sets will also be provided.
|
