This is one of those posts that started with a conversation with friends then sort of spiraled in to way too much time with Google, then I realized there’s a stats tie in and a post was born. Bear with me.

Some background:  Ramadan started this week, so I’ve been thinking a lot about dietary traditions in different cultures. In the book Antifragile, there is a moment where author Nicholas Nassim Taleb takes a surprising detour in to the world of human health and nutrition. As an economist/statistician who is best known for making predictions about the stability of financial markets, this seems like an odd road to go down. His take on diet is, unsurprisingly, unique: every Wednesday and Friday, he is vegan. Apparently in the Greek Orthodox tradition on Wednesdays, Fridays, Lent (48 days) and in the lead up to Christmas (40 days), you give up all animal products and oil. I am not clear how widely this is followed, but the US Greek Orthodox website calendar confirms this is the general set up. Since the thesis of the book is that some things actually improve when subject to disorder/inconsistency, Taleb wonders if the much touted benefits of the Mediterranean diet are due to the overall consumption, or the inherent variability in the diet due to the religious practices in the area.

Research tie in: I was interested by this point, as I’d definitely heard about the Mediterranean diet and its health benefits, but I’d never heard that this tradition was so common in that area. When it came back up last week I decided to ask a few other people if they’d ever heard of it. It was hardly a scientific poll, but out of the dozen or so people I asked, everyone knew the Mediterranean diet was supposed to be very healthy but no one had heard of the Wednesday/Friday fasting tradition. I even asked a few vegetarian and vegan friends, and they were similarly surprised. Given that two days a week plus all of Lent works out to over a third of the year, this seemed relevant.

Of course I am not sure what this might prove, but it did strike me as an interesting example of a time an average might be lying to you. The Greek Orthodox adherents who spawned the interest in the Mediterranean diet didn’t have one way of eating…they really had 2: normal days and fasting days. (Note: It appears not many American Greek Orthodox still follow the fasting calendar, but since Crete got on the map 70 years ago with the 7 countries study, it’s likely those who kicked this whole Mediterranean craze off were following it). By hearing only the average recommendations, it seems like some information got lost. Given that food recall questionnaires and epidemiological reports tend to only come up with one set of recommendations, I decided to take a look around and see if I could find other examples of populations whose “average” consumption might be deceptive. While many religions have a tradition of fasting, I’m only including the ones where the duration is substantial according to my own arbitrary standards. I’m also not including traditions that prohibit or discourage certain foods all the time, as that’s not the type of variability I was interested in.

Greek Orthodox I was curious if Taleb’s question had been addressed by any research, and it actually has been. This group noticed the same gap he did, and decided to follow a bunch of people on the island of Crete for 1 year. They all were eating the same famous Mediterranean diet, but those who followed the fasting traditions had better health markers after the holy days. This gives some credibility to the idea that something about the fasting that effects the health outcomes, though it could be that those who follow the fasting traditions are different in some other way.

Muslims This paper shows some interesting effects of Ramadan (no eating during daylight hours for 28-30 days) on health outcomes, but reaches no direct conclusions. Many of the studies didn’t include things like smoking status, so it’s hard to tell if there’s any benefit. Still, changing your eating patterns dramatically for a full month every year is probably enough to throw your “average” consumption a bit.

Ethiopian Orthodox According to this NPR story, the Ethiopian Orthodox Church observes a 40 day vegan fast prior to Christmas, where they only eat one meal a day.

Vacations and Holidays On the flip side, there are also occasions where people seem to consistently overeat in such a way that may change their “average”. Vacations appear to be correlated with weight gain that doesn’t immediately disappear, as does the holiday season. Interestingly, neither of these gains are that much (a little less than a pound overall for each), but if those persist after each holiday season and vacation, you could eventually see a real increase. Regardless, few of us call our vacation or holiday eating “typical”, but since holiday eating and vacations actually can take up a lot of days (November, December, 2 week vacation or so), this very well might skew our perception on what’s “typical”.

I’d be interested to hear any other examples anyone has.