In answer to Pommie's specific question, no, low-fat diets do not make people fat. People become fat because they eat too much for the amount of energy they burn. The mix of foods they eat may affect things like the perception of hunger and nutritional well being, but that was not the question.
As you know, it is almost impossible to publish a negative result, until the positive has been accepted. In other words, you cannot publish a study showing that eating an apple a day keeps the doctor away, until someone else (usually several others) publish that eating an apple a day causes more frequent doctor visits. Such publication bias inevitably leads to "today's new medical discovery" and subsequent skepticism of all medical reports. I ,for one, am particularly skeptical of new medical discoveries reported in the newspapers.
The NY Times reports an interesting observation about cholesterol:
NY Times article said:
Our cholesterol levels have been declining, and we have been smoking less, and yet the incidence of heart disease has not declined as would be expected. ''That is very disconcerting,'' Willett says. ''It suggests that something else bad is happening.''
The "Framingham Study" (
https://en.wikipedia.org/wiki/Framingham_Heart_Study) is considered by many to be the seminal study on the effects of serum cholesterol on cardiovascular disease. What people don't pay much attention to is that of the two groups studied -- those who did not practice interventions to reduce their total serum cholesterol and those who did --
the death rates from all causes were not statistically different .* The study was initially set up as a two-tailed study (i.e, does cholesterol improve health or lead to earlier death), because such analysis is inherently more robust. The fact that no difference in death rates was observed was shocking to the original investigators, and they changed their analysis to focus only on cardiovascular death rates in order to prove their hypothesis. Such changes to a study after it is completed is a very slippery slope. Nobel Laureate Michael S. Brown (
https://en.wikipedia.org/wiki/Michael_Stuart_Brown) was one of the early critics of that change in analysis because it biased the results. Fortunately, his work on the actual metabolism of cholesterol led to quite useful discoveries, so now we recognize both "good" and "bad" cholesterol. Nevertheless, the public and politics are slow to change and we still have people preaching about the evils of "cholesterol." Somewhat ironically, there are proponents today of increasing "good" cholesterol by eating coconut oil to ward off the effects of Alzheimer's.
In sum, I remain skeptical of any report of a startling result, until I have seen the study design and the actual data. Even that may not be enough, as one never knows how much the published report has been distorted from what really was done to ensure continued funding for the investigators.
John
Edit: Just for clarification, that result was due to the fact that the death rate in the intervention group from other causes was higher. Tongue in cheek, if you get hit by a car while riding a bicycle 5 miles a day to reduce your cholesterol, your are still dead. Not so tongue in cheek, the people in the intervention group had a higher rate of suicide. Maybe cholesterol is an anti-depressant?
