Dietary fat is necessary for mental and physical health. Excessively lean diets may be mentally and physically unhealthy. A balanced diet, with abundant fruits and vegetables, at least 30% of calories coming from fat, and with carbohydrates coming from foods with a lower glycemic index (e.g. reducing amounts of simple sugars), is probably a sound recommendation for good physical and mental health.
The type of fat is important, though: trans-fats are particularly harmful (these are from hydrogenated oils including hydrogenated margarines). It is probably true that omega-6 fatty acids (present in vegetable or soybean oils), while necessary in moderation, are over-abundant in western diets. Saturated fats (such as from red meat and dairy) have been associated with worse health outcomes.
Yet, as I review the literature, I see that this assumption about saturated fat may not be as clear as what most people assume. I intend to review this literature more thoroughly, and add to this post later. It may be that saturated fats from plant foods such as coconut are more benign. And it may be that health problems associated with eating a lot of red meat are due to factors aside from the saturated fat content.
*As I look into the coconut oil issue, I see there is a tremendous amount of hype and salesmanship going on--it seems to be touted as some kind of miracle food, also with a variety of scientific claims (e.g. about medium-chain triglyceride content) intended to strengthen the persuasion. When I look into what the research literature has to say, there really isn't a lot out there. What is out there at this point is not very consistent. It is true that there are groups of people, such as in Polynesia, who consume a lot of coconut oils, apparently without developing high rates of heart disease. In any case, I think it is fair to say that coconut or coconut oil in small quantities could be reasonably included in a healthy diet.
Clearly healthy sources of fat include fish, olive oil, nuts, avocados, and canola oil.
There are several types of cholesterol in the blood, the main subtypes being LDL and HDL. High LDL is a risk factor for cardiovascular disease (e.g. heart attacks and strokes). HDL is considered "the good cholesterol", and it is quite clear that higher HDL levels reduce the risk of developing cardiovascular disease. It is possible to increase HDL by exercising regularly, maintaining a healthy weight, stopping smoking, increasing dietary intake of monounsaturated fat (e.g. olive oil & canola oils), and increasing soluble fiber in the diet (e.g. oats, fruits, vegetables, legumes). 1-2 drinks per day (but no more) of alcohol may favourably impact HDL levels and overall health. It is important to note that the actual cholesterol present in certain foods, such as eggs, has an inconsistent relationship with serum cholesterol levels (perhaps a stronger relationship in some people than others), and an even less consistent effect on health variables--so cholesterol content of foods need not be a particularly important variable to assess.
In this 1998 study from the British Journal of Nutrition, subjects initially consumed a diet with 41% of calories coming from fat, then half of these subjects switched to a low-fat diet with only 25% of calories from fat. The group with the lower-fat diet developed higher levels of anger, hostility, and anxiety compared to the group continuing the higher-fat diet:
In this 2008 meta-analysis from Annals of Behavioural Medicine, an inverse association is found between serum cholesterol levels and depression. It is an interesting and surprising finding, given that we recognize lower cholesterol levels as beneficial for your heart:
In this 2008 study, a group with chronic depression was compared with a group with normal mood, and it was found that depression was associated with lower HDL levels (i.e. lower "good cholesterol"), even after controlling for several confounding factors. This type of study is unfortunately a bit weak. Here's a link to the abstract:
Here's a reference to a 2003 article from Encephale reviewing some of the evidence about low cholesterol being associated with depression and suicide. The authors also suggest that inadequate omega-3 fatty acids compared to omega-6 fatty acids in the diet may be a contributing factor to higher rates of depression.
This is a small but convincing 2008 study which showed significantly lower cholesterol levels in suicidal patients with schizoaffective disorder, compared to non-suicidal patients with schizoaffective disorder, and compared to healthy controls. HDL (the "good cholesterol") was higher in the non-suicidal patients and in the control group. The groups did not differ significantly with respect to BMI, so the association between cholesterol and symptoms would not have been due to weight.
Here's another 2007 study showing low cholesterol levels in an elderly group with cognitive impairment, and in an elderly group with depression, compared to a healthy elderly group.
Here's a 2007 study showing strong association between higher HDL cholesterol and better physical functioning among the oldest elderly (over 80 years old):
Here's a 2004 review describing the many findings about higher HDL being associated with better physical and mental functioning in the elderly, and in particular that people who live over 100 years have higher HDL levels:
In this strong, prospective 2009 study following 1,468 nurses with type II diabetes, higher dietary saturated and trans fat intake, and a lower ratio of polyunsaturated fat to saturated fat in the diet, was associated with worse cognitive decline (those in the highest third of saturated+trans fat intake effectively aged an extra 7 years with respect to cognitive decline, compared to those in the lowest third):
Here's a similar 2004 article from Neurology showing worse cognitive decline associated with higher saturated fat intake, lower monounsaturated fat intake, and a lower ratio of polyunsaturated to saturated fat intake:
In this strong, prospective, randomized 2007 study from JAMA, a diet with a low glycemic load (e.g. reducing simple sugars and increasing complex, slowly-digested carbs) and 35% of energy coming from fat, was compared with a low-fat diet (20% of energy from fat), with follow-up over 18 months. The higher-fat, low-glycemic load diet led to better improvement (increase) of HDL levels, and considerably better weight control: