To the Editor We read with great interest the important study in a recent issue of JAMA Neurology by Grande et al,¹ which described associations between long-term exposure to air pollution (particulate matter of ≤2.5 μm and nitrogen oxide) and dementia in a Swedish national cohort. This study makes a major contribution to scientific knowledge about the associations of air pollution with dementia risk.² The findings of mediation or effect measures modifying the roles of cardiovascular disease (CVD) not only provide mechanistic insights and identify more vulnerable populations, but will also have public health implications concerning the potential effectiveness of intervening on CVD to improve dementia risk among adults with high exposure. The study¹ was well conducted. However, we have some concerns regarding the validity of the claims that CVD is a mediator because of (1) the selection of confounders, (2) the nonlinear association between exposure and outcome, and (3) the presence of an interaction between the exposure and the mediator.