Estimating model-adjusted prevalence ratios through predicted marginal ratios with complex sample survey data

• Activity Number: 254
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Zhen Zhao*+
• Companies: CDC
• Keywords: Model-adjusted Prevalence Ratio ; Predicted Marginal Ratio ; Conditional Marginal Ratio ; Complex Sample Survey

Abstract

A recent procedure for estimating model-adjusted prevalence ratios in complex survey was proposed, however the performance of this important enhancement has not been assessed. The ratio of predicted marginal estimates obtained from multivariable logistic regression for complex survey, between exposed and non-exposed groups, was defined as model-adjusted prevalence ratio. A series of stratified one-stage cluster sampling surveys based on predetermined prevalence ratio (true prevalence ratio) were simulated. The results comparing the performance of predicted marginal ratio to conditional marginal ratio, adjusted prevalence odds ratio, and true prevalence ratio indicated the superiority of predicted marginal ratio for estimating model-adjusted prevalence ratio. This method has been applied to a disparity study in vaccination coverage with National Immunization Survey 2005 data.