"...all three populations showed an inverse relationship between total calories and CHD incidence. The more people ate, the less likely they were to suffer a heart attack."
I had to check this out for myself. Below is table 5 from that article (click on tables for larger view).
In TGCC Colpo continues:
"Those who remained healthy also had the lowest average body weight, despite their higher energy intake. This strongly suggests that higher levels of physical activity, not starch foods, were the most important protective factor, for if all participants were equally active (or inactive), those with the highest caloric intakes would be expected to weigh more, not less."
Colpo's conclusion (as well as the authors of the article) was based on the body weight data given in tables 2-4 (shown below). What surprises me is that the mean difference in body weight between healthy non-CHD and CHD cases is only 4-5 pounds! Assuming that the distribution of height is similar between non-CHD and CHD cases in all three studies, then body weight alone is not a very strong predictor for CHD. Perhaps body fat percentage data from these three studies would have shown a stronger effect with CHD?
As seen in Table 5 above, Framingham's CHD death rate (about 6%) was about triple that of Puerto Rico and Honolulu (about 2%), so it's interesting to compare the three groups side by side, as was done in Table 15 (below). Both Puerto Rico and Honolulu consumed less saturated and monosaturated fat, more carbohydrates from starches and less from sugars and less alcohol. The Puerto Rico and Honolulu men also weighed considerably less than Framingham men, but there is no data in the report on body height or body fat percentage.
There seems to be contradictory evidence comparing between and within groups. For example, the Framingham data (table 2) by itself suggests consuming alcohol for coronary heart health, and to a lesser degree so does the Puerto Rico and Honolulu data (tables 3 & 4). But between groups (table 15) there appears to be an association between less alcohol consumption and coronary heart health. Another example is that the between group data (table 15) and within group data of Puerto Rico and Honolulu (tables 3 & 4) associates higher starch consumption with coronary heart health, but the within group data from Framingham (table 2) doesn't show this association. Another example is that fat within any of the groups doesn't have much effect on coronary heart health, but the lower CHD rates for men from Puerto Rico and Honolulu associates with less dietary fat.
There was also some interesting data about serum cholesterol. Table 13 shows that the dietary starch intake had very little effect on serum cholesterol. Similar cholesterol tables for fat and protein were not given. These mean numbers (some were lower, some were higher) come from men FREE of CHD, and yet, the cholesterol levels range from 193-225 mg/dL which is at the high end or above the mainstream upper limit (200 mg/dL) for serum cholesterol.
So, my conclusions on this report are that I prefer the coronary heart disease rates of men from Puerto Rico and Honolulu to men from Framingham, but I don't think enough factors were studied to thoroughly understand the differences and implement them. Having said that, nothing that I read will change (but perhaps some support) my current nutrition plant: be liberal with starches (and carbs from fruit and vegetables), limit alcohol (as in only the occasional drink for special occasions), be moderate with fats (35-40% of calories from these studies actually seems a little high to me) and consume only enough calories to lower my body weight (to a level that allows me to run very well). And my plan for physical activity is to be physically active. And I will not be concerned when my doctor tells me I should lower my "higher than recommended" cholesterol level. And I will definitely not take drugs to lower my cholesterol, but that is another whole can of worms that I prefer to keep the lid on for now. On to chapter 7 of Colpo's TGCC I go, but at this rate I'll be done by Christmas, of 2012.