Up years ago when people were thinking that dietary fat was predictive of breast cancer. And you could see wide variability across cultures which is confounded. With many other things and then within a given cohort. You had less power because you just didn't have examples of people. Eating within a single culture didn't have examples appealing Leading one very low fat in other people eating mary high fat so the The ability in epidemiology to discover asssociation in part depends on the range of the variable in if the range is more narrow than you can't see as much dia patterns are also restricted but the correlation. And i'm talking about its consistency within person. So epidemiology worked better. If you have a single characteristic which characterizes the person for their whole life if you think about the framingham score and what happened in the early days cardiovascular epidemiology with cholesterol and blood pressure smoking. They had measured those things once between age. I think thirty five and sixty four and found ten years later that they predicted disease. So why would that. One measurement predict you know lie witnessed. Just be some random number. If i came back and had measured you tomorrow or next year for years ago it would be completely different. So that kind of correlation with in-person says that people follow along some kind of consistent track if it's less and less consistent than to get the average exposure said the average dietary pattern in takeover many years he would have to have by a few measures and average them so the variability in epidemiologists a complex If there's not too much variability within your sample you may miss an important effect but if people are too variable within themselves than you don't characterize him very well and you will not see any outcomes as well One of the things that this is is useful for as you've highlighted is epidemiology this concept of by nature. It's looking at things from top down approach but what are do you think the kind of challenges of you know for example to look at a dietary pattern. It's going to be a a score based using some form of index on. So what kind of challenges do you see for you. Know for the top down approach epidemiology rather than say. Oh we've got this range of Saturated fats the only thing we're looking at you're looking at is for example. A plant based diet score or a healthy eating index type score challenges are kind of there from an epidemiological perspective in deriving at achieving. This kind of relative estimate of consistency in a person over time when you talk about complexity one of the reasons that people went to a nutrients is.