The standard way is to do an intervention: either tell meat eaters to stop eating meat or tell vegetarians to start. Then compare with a control group. In medic

The standard way is to do an intervention: either tell meat eaters to stop eating meat or tell vegetarians to start. Then compare with a control group. In medicine it's also advisable to use double-blinding. I'm not suggesting the authors made a mistake or anything. An association study is a good cheap starting point to see whether a closer look is warranted.