|
Activity Number:
|
113
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #310297 |
|
Title:
|
Using Administrative Data To Improve Sample Design and Estimation: Assaults Requiring Emergency Department Visits in New York City
|
|
Author(s):
|
Kevin Konty*+ and Catherine Stayton and Jingsong Lu
|
|
Companies:
|
New York City Department of Health and Mental Hygiene and New York City Department of Health and Mental Hygiene and New York City Department of Health and Mental Hygiene
|
|
Address:
|
133 concord Street, Brooklyn, NY, 11201,
|
|
Keywords:
|
Survey design ; Disease Surveillance ; Administrative Data ; Syndromic Surveillance ; Data quality
|
|
Abstract:
|
The NYC Department of Health's Injury Surveillance System (ISS) reviews hospital emergency department (ED) visits to estimate two quantities: the number of assaults requiring emergency room visits, and the proportion of those assaults perpetrated by a relative or partner of the victim. These estimates are used to monitor trends and evaluate and target public health initiatives. Currently, ISS reviews all ED visits for 4 weeks per year at 22 hospitals. Multipliers are used to estimate yearly assaults and the relationship of perpetrator to victim is determined by analyzing the assault data as a simple random sample. This paper describes an effort to recast the system into a probability sampling framework. We develop a novel use of syndromic surveillance ED data to improve estimation, including establishing measures of size, quantifying clustering effects, and correcting for poor coverage.
|
