We read the article by Tang et al1 with great interest. Owing to the ever-changing disease activity and case-by-case response to treatment, it is often difficult to comprehensively recognize the long-term clinical course of patients with immunoglobulin A nephropathy (IgAN).2 The lack of a methodology to assess the complex process of chronic-phase IgAN postponed the solution to this problem. Tang et al addressed this issue using a recently established statistical method (marginal structural models).