FROM:  William Grant, [log in to unmask]

Congratulations to Art De Vany for spotting some of the reasons the
Alzheimer's Association issued a press release shooting down my paper - the
data in Figure 1.  Other reasons included "flawed methodology" and "confusing
strokes with Alzheimer's disease.  However, one should remember that the
Alzheimer's Association has the Reagan Foundation for Alzheimer's Disease
Research, and if the donors knew that an atmospheric scientist found the
major risk factor for AD in 300 hours, the funds they receive might dry up.

In response to their concern that the Alzheimer's disease rate vs. fat
supply did show two distinct groupings, and the concerns of others that the
European countries showed lots of rate variation for small variations in
fat, I did another analysis using only European and North American
countries.  The results are given in Table 4, but the figures are not
included because the editor thought that a multiple regression with fat and
fish was too complicated.  What the fat/fish multiple regression shows (
using data from Table 2) is that 1 calorie of fish counters 4.3 cal. of fat
.  The mechanism is likely that fish oil reduces inflammation through
prostaglandin 3s, while fat promotes inflammation through prostaglandin 2s.
In the unpublished figure, the European countries now lie in a straight
line, and r2 = 0.937.

Regarding genetic vs. environmental causes of Alzheimer's disease.  I got
interested in this topic because Lon White showed that Japanese-Americans
have 2.5 times the AD rate of native Japanese.  I later learned that
African-Americans have 4 times the rate of AD as Nigerians, in both cases
age-adjusted to the U.S. population distribution over 65 years of age.
This does not look like a genetic factor to me.

As far as experimental uncertainty goes, I included 4 values from Japan,
which range from 1.5 to 2.4%, and 4 from Taiwan (not used in the analysis),
which range from 0.6 to 2.6%.  The U.S. data range from 3.5 to 6.24%.  The
Alzheimer's Association also questioned the use of national dietary supply
data as a measure of local dietary consumption.  It is well known that
Japanese-Americans still eat much fish and rice, while African-Americans
eat more fat and salt than the average American.  I averaged the two
ethnic groups in determining an average U.S. AD rate.  So, yes, there is
experimental uncertainty in the data.  However, the statistical analysis
was able to spot the fact that data from the period prior to 1980 were not
comparable with data after 1980, with the reason being, I learned later,
that the clinical determination of AD changed in 1980.

The way my results should be viewed as a hypothesis which should then be
tested using case control studies.  Such studies are already underway.
I refer to the one by Kalmijn et al., which actually got me to look at the
multiple regression for fat and fish.  Also, Mark Smith, referenced for
other works, has participated in some case control work which shows that
those with AD maintained the same caloric intake in their 60s as they did
in their earlier years, while the controls reduced their caloric intake by
400 calories.  Another study showed that those with AD have reduced intakes
of vitamins and minerals compared to controls.  Dharma Singh Khalsa in
Tucson is treating people in the early stages of AD with a progam
consistent with my findings and is able to slow the progression or reverse
cogintive impairment or AD.

I think that if people look at the supporting documentation, such as the
use of NSAIDs to reduce the risk of AD, and the use of vitamin E to slow
the progress of AD (both referenced in my paper), the changes in the
Japanese-American and African-American populations compared with native
populations, the recent case control results, the changes in metal ion
concentrations in the brains of those with AD, the high cross-correlations
with diseases associated with dietary fat, the results of the Nuns Study
showing that AD is a vascular disease, etc., they would be hard-
pressed to find a better hypothesis.  If anyone does, please let me know.

The epidemilogic approach linking diet and disease was very useful in the
1970s and 1980s in reestablishing the link between dietary fat and cancer.
John Weisburger sent me some papers in 1991 which later provided the model
for my work.  Ancel Keys and John Yudkin had also used the approach to show
that animal fat and sugar were associated with heart disease.  I have
submitted manuscripts linking meat and sugar to rheumatoid arthritis and
animal fat and sugar as links to heart disease (animal fat for men, sugar
for women).  I also have a letter to the editor on fat and cancer which
should be in the NY Times for Oct. 3.