With the current prevailing wisdom being that dairy, salt, steak, etc are bad for you I worry that survey based (e.g. sampling behavior instead of experiment ba

With the current prevailing wisdom being that dairy, salt, steak, etc are bad for you I worry that survey based (e.g. sampling behavior instead of experiment based studies) exposes a huge bias: anyone with poor genes is likely to go to a doctor and modify their diet (well sort of) - the people who are naturally healthy are the least likely you need to modify their diet. So I'd take these results with a grain of salt.