A review of the available research on the association of saturated fat with mortality revealed that the food source of the saturated fat seems to determine whether or not saturated fat consumption has a negative impact on mortality. In the analysis of the 26 studies that met the research team’s criteria for inclusion in their review, the relative risk was calculated for the association of consuming meat, processed meat, milk, cheese, butter, and all dairy with mortality.
When looking at all causes of mortality combined, higher intakes of meat and processed meat were significantly associated with a greater risk of mortality. Looking at specific causes of mortality, both meat in general and processed meat specifically were statistically associated with greater cancer risk. Higher intakes of processed meats were also associated with a moderate increase in cardiovascular disease mortality. When looking at how the number of servings of meat impacts the relative risk of mortality from any cause, the increased risk begins when going from zero to one serving of meat a week, and that risk continues to rise in a linear fashion as weekly meat intake increases.
In contrast, milk, cheese, butter, and total dairy intake were not significantly associated with any increased risk of mortality. Looking more closely at the impact of dairy on relative risk, a few statistically significant findings emerge for some dairy products when examining the impact of the number of dairy servings per week. For milk consumption, compared to individuals consuming no milk, the relative risk of all causes of mortality actually decreases slightly at an intake level of four to 14 serving a week, with eight servings providing the greatest reduction in risk. At more than 18 servings of milk a week, there was a greater risk of mortality from all causes compared to individuals consuming no milk. In terms of cardiovascular disease, compared to no consumption, milk showed an even stronger positive association as servings per week of milk increased. Even 20 servings of milk per week showed a decreased relative risk for cardiovascular disease compared to not drinking milk, with the strongest association being at 12 servings per week of milk. For cheese, increased cardiovascular disease risk was associated with the consumption of more than eight servings of cheese per week. Below eight servings of cheese per week, no greater risk was observed.
Another interesting finding from this study was that for studies performed in Asian countries, there was a reduced risk of coronary heart disease mortality as meat intake increased—opposite the findings for non-Asian studies examined. The reasons for this remain unclear, but the authors suggest that this may have something to do with the lower overall consumption of saturated fats in Asian countries, or some other yet-to-be-identified difference in the overall dietary context in Asian countries.
Turning back to the main findings, this review suggests that with respect to mortality, the source of saturated fats in the diet may be more important than looking only at the overall saturated fat intake. The associations found between mortality and saturated fat differ considerably depending on the specific food sources of saturated fats studied.
Rather than focusing just on the saturated fat content in foods, which requires extra effort and research, this review suggests that with respect to concerns about the potential negative impacts of saturated fats, it may be better to give dietary advice based on specific foods themselves.
However, it should also be noted that saturated fats have also been associated with other negative health outcomes other than mortality alone. For example, other researchers have found that individuals with the highest levels of saturated fats had a greater relative risk of developing Alzheimer’s Disease, once potentially confounding factors were taken into account. Whether these effects are due to saturated fats from certain foods high in saturated fats but not others with a high saturated fat content remains unclear.