It has 18000 results just for weight loss, and if you look at different diets for weight loss, they are conflicting, so I wouldn't trust it. I think PubMed is

It has 18000 results just for weight loss, and if you look at different diets for weight loss, they are conflicting, so I wouldn't trust it. I think PubMed is still the best thing we've got: read medical publications directly, even those are not clear, but at least closer to the data source.