In "Eat, drink and be healthy" Walter Willett provides an interesting
critique of the USDA food pyramid in his recent book "Eat, drink and be
healthy" and sets out a radically different food pyramid. He writes: "At
best, the USDA pyramid offers wishy-washy, scientifically unfounded advice
on an absolutely vital topic - what to eat. At worst, the misinformation
contributes to overweight, poor health and unnecessary early deaths."

Willett is widely regarded is the founder of nutritional epidemiology and
his alternative pyramid reflects his well informed view of current
epidemiological evidence. His pyramid is aimed at omnivores, but is entirely
consistent with a vegan diet though to my knowledge he is not a vegetarian
himself.

The base of the pyramid is daily exercise and weight control. The second
layer is whole grain foods (at most meals) and plant oils, followed by
vegetables (in abundance) and fruit (2-3 times per day). The fourth layer is
nuts and legumes (1-3 times per day) with animal products appearing next as
fish, poultry, eggs (0-2 times per day). Note the zero. The sixth layer is
dairy or calcium supplement, 1-2 times per day. Willett makes very clear
that he regards the calcium supplement as the healthier option. Red meat,
butter, refined grains, potatoes and sweets appear at the top under "use
sparingly". He also suggests the use of a multivitamin and the consumption
of alcohol in moderation (unless contraindicated, e.g. by a family history
of drink problems).

As well as differing radically from the USDA pyramid which places meat and
dairy at the heart of its recommendations, Willett's pyramid is notably
different to PCRM's four food groups (grains, legumes, fruit and vegetables)
in that it is by no means a low-fat diet. Epidemiological evidence supports
a beneficial effect of plant oils provided weight is kept under control and
also supports a more specific benefit from nuts and seeds. The jury is not
entirely in on whole-food low-fat diets versus whole-food diets with plenty
of nuts, seeds and plant oils but the evidence is more in favour of the
latter (provided the fat content is mostly monounsaturated and weight
control is good). Large amounts of omega-6 rich oils are unlikely to be
healthful within a vegan diet as vegans rely on conversion of
alpha-linolenic acid to longer omega-3 fatty acids and omega-6 consumption
inhibits this. Also, saturated fat and cholesterol are already low in vegan
diets so the cholesterol reducing effect of most plant oils is not needed to
the same extent. However monounsaturated fats from olive oil, rapeseed
(canola) oil, cashews, almonds, hazelnuts (filberts) and avocadoes have a
useful beneficial effect on cholesterol and triglycerides without
interfering with omega-3 conversion and can be a healthy part of a vegan
diet. These fats can be included according to personal preference subject to
maintaining a healthy weight. For people struggling to maintain a healthy
weight (above a BMI of 18 for men and 17 for women) such fats can be a real
boon. For people struggling against weight gain they probably have a limited
role to play. Flax seed oil is the best source of plant omega-3s, though
rapeseed (canola) oil is also a useful source and green leafy vegetables can
make a significant contribution if eaten in large amounts.

Willett's pyramid is similar in most ways to the food pyramid in Becoming
Vegan. There is even more emphasis on vegetables and less emphasis on soy
products in Willett's pyramid and I consider this appropriate. Fortified soy
products are emphasised in the Becoming Vegan pyramid as a source of
calcium. I prefer Willett's approach of emphasising vegetables and
suggesting a calcium supplement as a top up for calcium if required. The
supplement option is certainly cheaper and may be healthier than heavy use
of soy, though moderate use of soy is likely to be beneficial. On the other
hand the Becoming Vegan pyramid is better for vegans in that omega-3 fatty
acids are specifically emphasised and high omega-6 oils such as sunflower
oil are not recommended. It is also appropriate that the Becoming Vegan
pyramid singles out B12 and vitamin D as of particular importance.

Willett pushes potatoes to the "use sparingly" category based on the view
that potatoes contribute to elevated glucose levels and the development of
diabetes as they have a high glycaemic index. This does not seem justified
as brown rice and wholemeal bread which are included in the whole grain (at
most meals) category have a comparable glycaemic index to potatoes. The
mineral content per calorie of potato compared with whole wheat is either
similar (e.g. magnesium) or better (e.g. 5 times as much potassium). Both
potatoes and whole grains have much better mineral content than refined
grains. I therefore consider this particular conclusion at best premature,
and think potatoes belong with the whole grains (this is not just because
I'm Irish - honest). The epidemiological evidence is much more positive for
whole grains than for potatoes, but as potatoes are often consumed with
animal fats or as fries it seems quite possible that their true benefit is
being masked.

http://www.mendosa.com/gilists.htm gives an extensive list of glycaemic
index values and is a useful reference on this topic. Of the commonly used
forms of grain wholewheat spaghetti has the lowest glycaemic index.

As Willett is not writing from a vegan perspective it is not surprising that
some of his detailed comments are not quite right for vegans. The different
emphasis appropriate for vegans on omega-6 versus omega-3 fats is discussed
above. Willett also makes some errors in terms of vegan calcium sources. He
suggests that spinach is a good source of calcium: it isn't as absorption of
its calcium is almost entirely blocked by its oxalate content. This is also
the case for rhubarb and for beet greens and probably for Swiss Chard. He
also suggests tofu is a good source of calcium, whereas this is true only if
a large amount of calcium salts have been used in preparing the tofu. Some
tofu brands have more sodium than calcium and will not promote bone health.
The best vegan sources of calcium are dark green leafy vegetables that are
low in oxalate such as spring greens, turnip greens and kale. For more
details on diet and bone health see
www.vegansociety.com/briefings/dietandbone/dietandbone.doc

Willett also presents a table that suggests there is B12 in whole wheat
flour. This is not so. Unfortified plant foods are not a useful source of
B12. For more detail on B12 see www.vegansociety.com/html/info/b12sheet.htm.
He does recommend 6 micrograms of B12 a day as a supplement which is roughly
in line with the recommendations in this information sheet.

I thoroughly recommend Willett's book. I don't agree with all of it but I
found it a stimulating and informative read. It was a pleasure to see such
an eminent scientist commenting on milk: "I don't recommend it as a beverage
for adults". A useful book to buy, learn from, and pass on to omnivorous
friends. It is also a useful book to show health visitors and other medical
professionals resistant to the idea that a vegan diet is a healthy option.

Stephen Walsh