A0231
Title: Sensitivity analysis with balancing weights estimators to address informative visit times in irregular longitudinal data
Authors: Sean Yiu - University of Cambridge (United Kingdom)
Li Su - University of Cambridge (United Kingdom) [presenting]
Abstract: Irregular longitudinal data with informative visit times arise when patients' visits are partly driven by concurrent disease outcomes. To address the selection bias from such data, existing methods rely on unverifiable assumptions and haven't adequately accommodated informative visit times for marginal regression analyses. Based on novel balancing weights estimators, a new sensitivity analysis approach is proposed to address informative visit times. The balancing weights are obtained by balancing observed covariate distributions and including a selection function with specified sensitivity parameters to characterize the additional influence of the concurrent outcome on the visit process. A calibration procedure is proposed to anchor the magnitude of the sensitivity parameters to the amount of variation in the visit process that could be additionally explained by the concurrent outcome given the observed history. Simulations demonstrate that the balancing weights estimators outperform existing weighted estimators for robustness and efficiency. The proposed methods are applied to analyze data from a clinic-based cohort of psoriatic arthritis.
