Generalized Estimating Equations: an overview and application in IndiMed study

Abstract

Generalized Estimating Equations (GEE), developed by (Zeger & Liang 1986), is a method of estimation that accounts for correlations among repeated measurements and is widely used in longitudinal analysis. The purpose of this master’s thesis is to provide an overview of GEE and use this approach on a cluster-randomized study called IndiMed. The data are from the Estonian Genome Center of University of Tartu. The clusters are doctors who were randomized to intervention or control group and subjects are patients with high blood pressure. All subjects in one cluster are either in intervention (received genetic risk information on the 2nd visit) or control group (received risk information on the 4th visit). Systolic and diastolic blood pressures were measured on all 5 visits for all patients and compared for the two study groups, taking into account the correlation among the repeated measurements.