A0515
Title: Statistical methods for integrating longitudinal and cross sectional data: A real case study
Authors: Hui-Wen Lin - Soochow University (Taiwan) [presenting]
Abstract: Longitudinal data (including survival tracking data) is crucial in medical research as it enables researchers to detect subject changes over time. However, there are many challenges in analyzing longitudinal data, such as complicated probability functions and data gaps. A two-stage approach is proposed to collect cross-sectional and longitudinal data effectively while reducing costs. Statistical methods used for analyzing longitudinal data were conducted, and a statistical method that integrates longitudinal and cross-sectional data was proposed to correct estimation biases. This method was also used to investigate the association between chronic obstructive pulmonary disease (COPD) and herpes zoster (HZ). The results suggest that COPD may increase the risk of HZ, even after adjusting for potential confounders. The proposed method provides a way to account for missing confounders and reduce bias in observational studies.
