So the takeaway is that more than zero but fewer than six cups is better than none or greater than six cups?

We need to know whether this means real cups for the bullshit "cups" commonly used to measure coffee were used in the study.