That's actually a very good observation! There does seem to be a major discontinuity between those two study periods. Very strange that it seems to only affec

That's actually a very good observation! There does seem to be a major discontinuity between those two study periods. Very strange that it seems to only affect fats, and that the effect was so dramatic... They do mention "There has been some research that addressed the impact of changes in methods and/or databases between selected surveys"... it'd be worth digging that up to see what the effects were determined to be. Edit: Found it? http://naldc.nal.usda.gov/catalog/7349 http://naldc.nal.usda.gov/catalog/7349 Nothing in the abstract explains a major discontinuity in reported fat intake, though. They only report small but statistically significant effects on various other nutrients. Edit 2: They also have a FAQ on this topic. From http://www.ars.usda.gov/Services/docs.htm?docid=13793 http://www.ars.usda.gov/Services/docs.htm?docid=13793: "Is it possible to combine WWEIA data from different release cycles?" It goes on to point out that sample weights, number of days, and the nutrient database do change over time, and so you have to take into account these effects. The above study covers effects as a result of changes in the nutrient database. I have trouble believing that sampling method or the data cycle length would result in such a major discontinuity in the data.