Wednesday, April 27, 2011

Why Do The Japanese Have Less Heart Disease?

Dr. Davis at the Heart Scan Blog is discussing the lower rates of heart disease in Japan compared to the US:
Why Do The Japanese Have Less Heart Disease?, Heart Scan Blog

They do have less heart disease. Below is cardiovascular mortality from 1990 to 1998 in Japanese and US men.

Source: Coronary Heart Disease Risk In Japan – An East/West Divide?, European Heart Journal, 2004.

He offered some dietary reasons, including the Japanese higher intake of seaweed, seafood, green tea, soy foods and lower intake of animal products (except seafood), sweet foods, and "the lack of a 'eat more healthy whole grain' mentality."

I don't know with great certainty how the Japanese are eating that's different from us so I'll take his word. I did think they ate a lot of rice. I know the Okinawans eat, or ate, a low-fat diet that included a lot of sweet potatoes.

He cited this study:
Dietary Patterns And Cardiovascular Disease Mortality In Japan: A Prospective Cohort Study, International Journal of Epidemiology, 2007

Which analyzed dietary patterns from 40, 547 Japanese men and women. They identified 3 dominant eating patterns:

1. A "Japanese" pattern
2. An "animal food" pattern
3. A dairy/fruit/vegetable pattern.

Pattern 1. (Japanese) differed from pattern 3. (dairy/fruit/vegetable) in that it included higher intakes of sodium, fish, soybean, seaweed and green tea. The two groups were similar in fruit and vegetable intake.

The median intake for rice in the highest quartile of these patterns was:

1. 567 grams
2. 744 grams
3. 524 grams

I thought that was a lot of rice, regardless of group. It was even more at lower quartiles. So I looked at the caloric intake for each group in corresponding quartiles:

1. Japanese pattern: 6861 kj = 1639 calories
2. Animal food pattern: 7458 kj = 1781 calories
3. Dairy/fruit/veg pattern: 5466 kj = 1305 calories

These seem low, unless I'm miscalculating. Anyway, I tried to figure what portion of those calories rice contributed. I used NutritionData's entry for Rice, White, Steamed, Chinese Restaurant:

Portion of calories contributed by rice:

1. 567 grams of rice contributed 855 calories (567x199cal/132g = 855), 52% of calories
2. 744 grams of rice contributed 1122 calories (744x199/132 = 1122), 42% of calories
3. 524 grams of rice contributed 790 calories (524x199/132 = 790), 61% of calories

Of course, these are estimates. And I don't know more about the type of rice they ate. But it does seem like a lot of rice.

Here's their fat, protein, and carb intake (in parentheses is the percentage that nutrient contributed):

Median total fat intake, highest quartile (g):

1. 41 (41x9 = 369 kcal. 369/1639 = 23%)
2. 41 (41x9 = 369 kcal. 369/1781 = 21%)
3. 41 (41x9 = 369 kcal. 369/1305 = 28%)

Median protein intake, highest quartile (g):

1. 74 (74x4 = 296 kcal. 296/1639 = 18%)
2. 72 (72x4 = 288 kcal. 288/1781 = 16%)
3. 69 (69x4 = 276 kcal. 276/1305 = 21%)

Carb intake, highest quartile (by default):

1. (100% - 23 + 18 = 59%)
2. (100% - 21 + 16 = 63%)
3. (100% - 28 + 21 = 51%)

The study found the Japanese dietary pattern (1.) was associated with a decreased risk of cardiovascular disease (CVD) mortality, the "animal food" dietary pattern (2.) was associated with an increased risk of CVD mortality, but the dairy/fruit/vegetable pattern (3.) was not.

Groups 1. and 2. were similar in macronutrient makeup. Although, the sources of those nutrients were different. That boosts Dr. Davis' argument that seafood (seaweed and fish), soy, and green tea may have a beneficial effect (at least among Japanese).

I do see a difference in macronutrients between Japanese and Americans. Japanese eat less fat (although, not as low as they were eating in the 1950s when fat contributed a mere 8% of calories), more carbohydrate, and possibly fewer calories overall. American men get about 33% of their calories from fat and 47% of their calories from carbohydrate. (NHANES 1999-2000.)
Photo of rice from Rice Wisdom.


Kieran O'Neill said...

The daily caloric intake seemed low to me, too, and I was concocting all kinds of explanations like the inclusion of children and the elderly, and Japanese people on average being smaller.

But then I realised it's just a misleading table. The medians shown are for each quartile, so the left-most figures (which could be read as the overall sample median), are instead the first, or sometimes last, quartile median.

So the 814 grams of rice per day is the median of what the upper 25% of the Japanese eaters are eating. Meanwhile, the upper 25% have a median caloric intake of 6861 kJ (which still feels low, but may be accounted for by smaller average size).

Anyway, I hope this helps to make more sense of the paper!

Bix said...

To be fair, Kieran's comment refers to a first draft where I used the lowest quartile figures. So that 814 grams of rice is the median amount in the lowest quartile of those eating the Japanese pattern. (You can see this in Table 3.)

Kieran O'Neill said...

Arg - that paper has had me quite confused, but I think I get it now.

Anyway, it should be noted that although fat intake was similar between group 1 and 2, the type of fat was quite different. I would hazard a guess from the weightings in table 1 that group 2 gets a *lot* more cholesterol and saturated fat from all that animal fat, whereas a lot more of group 1's fat intake would be vegetable oil with no cholesterol and less saturated fat.

The other big difference in the weightings is that group 1 got a lot more vegetables (seaweed included), suggesting better vitamin and fibre intake.

And that really makes a lot of sense. There is plenty of research associating low fibre intake, high cholesterol intake, and high saturated fat intake with CVD. I'm not so sure that the green tea and seaweed really play that much of a role (beyond the seaweed being a good source of vitamins and minerals).