Habitual Diets of Real Americans

This will be a revolving blog post, that shall get bumped up every time I add a study where habitual intake -- aka baseline -- has been determined/reported prior to a dietary intervention.  It's time to collect this data in one place, as in study after study, I have noted that Americans are not eating a low fat diet, nor a particularly high carb diet, despite the very modest percent reductions in NHANES surveys.   I reserve the right to add in studies in industrialized/Westernized populations as honorary dietary Americans.   Newest additions will be on top.

In type 2 diabetes, randomisation to advice to follow a low-carbohydrate diet transiently improves glycaemic control compared with advice to follow a low-fat diet producing a similar weight loss  
H. Guldbrand, B. Dizdar, B. Bunjaku, T. Lindström, M. Bachrach-Lindström, M. Fredrikson, C. J. Östgren and F. H. Nystrom

Note:  Swedish study, 2012 publication
  • 61 Type 2 diabetics, mean BMI 32.7 ± 5.4 kg/m2, mean HbA1c 57.0 ± 9.2 mmol/mol
  • Broken into two study groups assigned LF or LC diet.
    • LF n=31, age 62.7 ± 11 yrs, 13M/18F, duration of known diabetes 8.8 ± 6.2 yrs. 
    • LC n=30, age 61.2 ± 9.5 yrs, 14M/16F, duration of known diabetes of 9.8 ± 5.5 yrs.
  • Combined age ~62, 27M/34F

Assessment:  Self-reported three day intake, including one weekend day, subjects provided scales and other measurement devices.  

I approximated an average for all 61 subjects at right since these were reported intakes at baseline and assigned group shouldn't have mattered. On average, approximately (alcohol missing from data):

Diet %  P~C~F     19~45~35

Notes:  Despite being given measurement devices, under-reporting seems likely.  

Forsythe CE, Phinney SD, Feinman RD, Volk BM, Freidenreich D, Quann E, Ballard K, Puglisi MJ, Maresh CM, Kraemer WJ, Bibus DM, Fernandez ML, Volek JS.

Subjects:  Eight men, aged 38–58 years old, with BMI of 25–35 kg/m2
Assessment:  Self-reported 3-day dietary record (logging presumed)

Diet   % P~C~F    25~34~41

Notes:  There appears to be about 20% under-reporting of baseline intake as test diet intakes were roughly 2500 cal/day.  Prior to the test phases, subjects consumed a "free living LC diet" for 3 weeks to determine weight stable intake.


Lerner said…
I'd assume that most people would under report fat and sugar calories, since those would be the guilty calories in their minds.

What about LCers and vegans? Would they suppress guilty transgressions, or obsess on them?
Susanne said…
Wow, that looks like a lot of protein to me -- 4 hamburger patties worth in a day. An egg is what, five grams protein each? I guess I always forget what meat servings look like in the states.

The carbs seem suspiciously low to me in comparison, maybe they're under-reporting like Lerner says. According to my LoseIt stats today, I got 172.3g carb out of a serving of steel cut oats and milk, a muffin and an apple, a salad with nuts and pears and one slice of bread, and half a beer. That doesn't sound like much for an average American man.

Or maybe they don't eat any fruit and veg and the 170 carb is all from bread.
CarbSane said…
This was not a usual study, but since it jumped out at me, I decided to start somewhere with logging these in one place. Most baseline diets tend to come out lower in protein and higher in carb, but not much different in fat. Usually fat is in the high 30's, so that was a tad high here as well.

It's funny this was a Volek study though, and they started out with a cohort that wasn't big carb eaters to begin with if their reporting was correct.
Puddleg said…
I wonder where they recruited the subjects. I also wonder whether any unreported carbs were mainly sugars; sodas and such.
Gabriella Kadar said…
Ha! These subjects should be sent to me. Unless they are OCD with dental hygiene, I probably could give a more accurate assesment of dietary composition in related to fermentable carbs..... All that gooey plaque! Arghhhh!

blogblog said…
To quote TV Doctor character Dr Gregory House "all people lie". Self-reported dietary intake is invariably woefully inaccurate. The likelihood that the study participants were actually eating only 1800Cal is close to zero.
CarbSane said…
I agree, that's not really the purpose of this post. As Lerner states, we're already in relatively high fat land here and folks are more likely to under-report fat (and sugar, but fat is likely worse). I'm intending on adding to this post often.
Lerner said…
Last night's House rerun was about a woman who could remember almost everything. I've heard of someone online like that, but can't remember who it is.
Anonymous said…
I do log my food every day (Weight Watchers habit):

I use fitday.com, which gives a report of percentages fat, protein, carbs, alcohol for various time periods. I don't eat low-carb, eat more than the average amount of fiber daily (I do follow the gut microbiota studies!) and go out for dinner once or twice a week (alcohol there). I have been actively trying to reduce saturated fat intake (I noticed that when I do that, my monounsat and polyunsat look much, much better!) The past two months: P-C-F is 18-48-31 with 3% alcohol.

I've been logging my food in fitday for years!
Susanne said…
Yes, I wonder if integrating tech like some of the easier to use food logging apps into these studies would result in more accurate results. I read about at least one study recently where they had people take pictures of their meals as part of the tracking. The difficulty with this is that certain demographics like older people who are not as technology-friendly, tend to be left out. (Perhaps you could incentivize it by giving them iPads? My mom loves hers.)

Also, even the act of live observation/logging will change the behavior, since people become more conscious of what they are eating.

Evelyn, I'm not sure I entirely agree on including all Western-industrial diet studies as honorary Americans. As Galina and I and others here have noted, many countries outside the US still have fairly strong food and eating traditions, not only in the types of food they eat but in customs like fixed mealtimes/established courses/limited snacking, which can have a marked effect on calorie consumption and even macro percentages. European colleagues and friends of mine have suggested that one of the problems with American foodways* is that we don't have a strong shared food and eating tradition, which is something that has allowed fast food etc advertisers to more easily make a case for eating anywhere, anytime.

*sorry, unattractive but useful anthropology term.