This study does support that conclusion, which is why it's interesting. The authors explicitly control for physical activity in their analysis, so you have to a

This study does support that conclusion, which is why it's interesting. The authors explicitly control for physical activity in their analysis, so you have to argue, as I do in another post, why they are doing that wrongly in order to continue to uphold the "it's all activity levels" view.