I appreciate the modeling work of Han and coworkers including the recent article: "The impact of overdiagnosis on the selection of efficient lung cancer screening strategies." 1 However, I think a few points need to be raised regarding this analysis so we don't make mistakes in drawing conclusions. First, there are aspects of how the models were set up that limits what we can say about overdiagnosis. Second, we have to be careful in prospectively applying what is fundamentally a retrospective concept.