Inference of Causal Effects from Complex Longitudinal Data Based on the New G-Formula
Arising from a sequence of treatments, two types of causal effects are of practical interest: the sequential causal effects and the blip effects of individual treatments on a specific outcome after the last treatment. Recently, a new G-formula was proposed (Wang & Yin, Annals of Statistics, 2020), which identifies these causal effects by point effects of individual treatments like treatment in single-point causal inference. It makes it possible to infer these causal effects by methods extended from single-point causal inference. Fortunately, methods in single-point causal inference are well developed and readily applicable in practice. It is possible to illustrate that the G-formula method leads to the more accurate and efficient point and interval estimates of these causal effects than available methods.
This project will develop a new efficient approach for sequential causal inference based on the new G-formula in the framework of single-point causal inference. Furthermore, the research team will apply the method to stomach cancer patients’ clinical data and find optimal individual treatments and treatment regimes for stomach cancer, which have not been seen in the literature to the best of our knowledge. Because treatments are assigned in the form of a sequence in most practices, the current project’s methods will have broad applicability in medical areas and economic areas.
2020-01-01 – 2022-12-31
The project is financed by the Swedish Research Council.
The project is collaborative. The researchers involved in the project are Associate Professor Xiaoqin Wang from the University of Gävle (project leader), Professor Fan Yang Wallentin from Uppsala University, Dr Li Yin and Dr Johannes Blom from Karolinska Institutet.