Intended for healthcare professionals

CCBYNC Open access

Rapid response to:


Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies

BMJ 2015; 351 doi: (Published 12 August 2015) Cite this as: BMJ 2015;351:h3978

Rapid Response:

Saturated fats are not harmful, really?

In history research, where truth is often elusive, it’s legitimate to go back to data sources and make new interpretations. It’s different in natural science. After Galileo Galilei and colleagues established beyond reasonable doubt that the earth circles the sun, new generations of physicists do not open the case over and over again. This reminds me of the seemingly never ending set of meta-analyses on the health effects of saturated fats. Same sets of studies with their inherent weaknesses -- like adjusting factors in the pathogenetic pathway, or, not adjusting for eg. socioeconomic class or healthy user bias -- are put together to prove that saturated fats, after all, would not be not hazardous to our health.

Personally, I accepted the idea of harm from saturated fats through the following chain of reasoning: 1. saturated fat intake increases serum LDL cholesterol (proven repeatedly); 2. LDL cholesterol is a long-term risk factor for cardiovascular diseases (proven repeatedly); 3. lowering LDL cholesterol reduces cardiovascular events (also proven repeatedly). But like proponents of tobacco industry, some proponents of saturated fats are demanding more studies and direct proof in the same trial. This is practically impossible. Nowadays we cannot put people in control and diet groups for several years to directly prove that saturated fat intake increases cardiovascular events.

Once they could do it: in the 12-year, controlled Finnish mental hospital study (1) saturated fats were replaced with polyunsaturated fat and the results were compatible with the theory chain above: cholesterol was reduced with the experimental diet and cardiac events were reduced accordingly.

In the absence of new controlled trials, we are left with this endless series of meta-analyses from observational studies which cannot prove causality. And these studies also involve several potential biases: over- or underadjustment, inaccurate measurement of dietary intake moving results toward null (2) etc.

But why this research activity? An interesting explanation was recently proposed: In 2008 global dairy industry had a meeting, where it was decided that it is important to “neutralize the negative impact of milkfat by regulators and medical professionals” ( It is appropriate to say that the study of de Souza et al does not report funding from dairy industry.

1. Miettinen M, Turpeinen O, Karvonen MJ, Elosuo R, Paavilainen E. Effect of cholesterol-lowering diet on mortality from coronary heart-disease and other causes. A twelve-year clinical trial in men and women. Lancet 1972;ii(7782):835-8
2. Archer E, Pavela G, Lavie CJ. The inadmissibility of what we eat in America and NHANES dietary data in nutrition and obesity research and the scientific formulation of national dietary guidelines. Mayo Clin Proc 2015;90:911-926

Competing interests: I have made research on lipids and was one author of a book on cholesterol in 2011 (very minor royalties).

13 August 2015
Timo E Strandberg
MD, Professor
Universities of Helsinki and Oulu
PO Box 340, FIN-00029 Helsinki, Finland