|
Activity Number:
|
341
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #309618 |
|
Title:
|
Comparison of Methods to Adjust Further for Nonresponse Due to Missing Lab Data in NHANES
|
|
Author(s):
|
Maya Sternberg*+ and Alula Hadgu
|
|
Companies:
|
Centers for Disease Control and Prevention and Centers for Disease Control and Prevention
|
|
Address:
|
1600 Clifton Road NE, Atlanta, GA, 30333,
|
|
Keywords:
|
nonresponse ; complex survey ; NHANES ; logistic regression ; weighting
|
|
Abstract:
|
This paper compares the impact of additional nonresponse adjustments to the standard National Health and Nutrition Examination Survey (NHANES) weights using weighting class adjustments and logistic propensity models to estimate the prevalence of six distinct sexually transmitted diseases (STDs). Specimens to test for six STDs were collected on different subgroups of participants during different survey cycles, and using different specimen collection methods depending on the STD. We investigated the impact of missing lab test results on prevalence estimates of each of these STDs by comparing the results of weighting class adjustments, weighted and unweighted logistic propensity models, and further post-stratification and trimming adjustments. Based on these results we make recommendations about the impact of the weighting adjustments on descriptive inferences for STD prevalence.
|
