|
Activity Number:
|
515
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #310364 |
|
Title:
|
A Strategy To Evaluate Noncoverage in a Binational, Population-Based Maternal and Child Health Survey on the U.S.-Mexico Border
|
|
Author(s):
|
Ruben Smith*+ and Jill McDonald and Christopher Johnson
|
|
Companies:
|
Centers for Disease Control and Prevention and Centers for Disease Control and Prevention and Centers for Disease Control and Prevention
|
|
Address:
|
4770 Buford Highway NE, Atlanta, GA, 30341,
|
|
Keywords:
|
data linkage ; birth certificate ; maternal and child health data
|
|
Abstract:
|
A design for collecting standardized reproductive health surveillance data from mothers of live-born infants was tested in two sister communities on the US-Mexico border. Mothers were sampled from six Mexican and four U.S. hospitals. To assess the potential bias of excluding mothers who delivered outside study hospitals (7% in Mexico, 2% in the US), we compared birth certificate (BC) characteristics of study hospital infants to all infants born during the study period and found no differences. Using multi-variable linkage, we matched sample data to BCs to estimate non-coverage and assess potential bias in the sampling frame. Eighty-seven percent of Mexican and 97% of U.S. women sampled were successfully matched to a BC. Three percent of both Mexican and U.S. BCs could not be linked to sample data. Results indicate minimal survey noncoverage.
|
