I question a lot of the studies referenced above due to the fact that none were double blinded placebo controlled trials which would control for factors other than the measured variable for the outcome of the study.
An association does not mean cause and effect.
The person is right about low cholesterol being an
associative risk factor for early death. (but it has not been proven as cause and effect.) However, they failed to subtype the cholesterol fractions since we know low HDL and LDL subtypes can have inverse effect on mortality. In addition, they did not control for things like smoking (which lowers total cholesterol by lowering HDL drastically), body weight (very low body weight confers higher mortality..look at anorexics) , nutritional status etc.
He does bring up a good point about familial hypercholesterolemia where up to 40% of them have normal lifespans while others seem to have greater cardiovascular risks. My interest is in genetics and there are studies that are getting close to figuring out the genetic phenotype that makes one person's high cholesterol dangerous but not the other with similar cholesterol numbers.
http://www.athero.org/newsletters/2004february/comments/pauciullo.pdf#search='familial%20hypercholesterolemia%20and%20mortality'
I think the point presented by AnthonyG is well taken in that heart disease is multifactorial and we cannot just blame one factor (ie: cholesterol) as a single cause. Most physicians know that only fifty percent of all heart disease can be explained by high cholesterol but the data on the other fifty percent is compelling enough to treat. Inflammmation does play a role, which is why asprin has a protective effect independant of cholesterol levels. We know high homocystine levels or low folic acid is a cardiac risk factor.(thus a nutritional deficiency can play a role in heart disesease.) Infection is still being looked at and the list goes on.
However, I don't think it is good science to say cholesterol has no effect on cardiovascular health. In some it definately does but in others it may not make much of a difference. It is medicine's job to continue to look further and see why some are more affected by high cholesterol than others. This same paradigm of thinking is true with cigarette smoke. Only 10% of all smokers ever get lung cancer. (Apparently only 10 % of all smokers are genetically susceptible to its carcinogenic properties. However, 90% of all lung cancers are attributed to smoking so we all know that avoiding smoking is a good idea if you want to diminish your chances of getting lung cancer!) By the way,we are getting close to isolating the lung cancer gene for those who are interested!