![IconGems-Print](images/IconGems-Print.png)
313 – Missing Data Methods for Epidemiologic Studies
Statistical Modeling of Subject and Proxy Observations Using Weighted GEE
Mina Hosseini
University of Maryland Baltimore County
Nagaraj K. Neerchal
University of Maryland
Ann L. Gruber-Baldini
University of Maryland School of Medicine
In epidemiological studies when the patients become unable to provide responses by themselves due to advancing severity of their conditions, proxy responses by a relative or a caregiver "proxy" are used. The resulting dataset contains a monotonically decreasing missing pattern for subject observations and a monotonically increasing missing pattern for proxy observations. Some statistical models are being investigated that can analyze subject and proxy observations together so that relevant parameters and their standard errors can be estimated in a single framework. The method of weighted Generalized Estimating Equations (GEE), which is commonly used for handling missing data, is applied to the combined proxy and subject dataset.