New 2025 Meta-Analysis: The Evidence Does Not Support Saturated Fat Restriction

A new meta-analysis of RCTs reports, “the evidence available from RCTs does not support saturated fat restriction for the prevention of cardiovascular disease.” Let's elaborate and elucidate.

A new meta-analysis of randomized controlled trials (RCTs) reports, “the evidence available from RCTs does not support saturated fat restriction for the prevention of cardiovascular disease.”

This directly contradicts Dietary Guidelines that place a limit on saturated fat at 10% of energy intake, a recommendation that has been consistent since the first edition of the Dietary Guidelines published in 1980.

Now the questions I’m sure you’re asking yourself are:

• Why is this still a matter of confusion and debate?

• Why haven’t nutrition scientists and health authorities come to a clear consensus on what seems to be a simple topic?

I want to name these because I know they’re front of mind. And I promise I will answer them. But first, I want to review the findings of this new 2025 meta-analysis.

What is a Meta-Analysis?

For those who don’t know, a meta-analysis is a statistical method that combines data from multiple independent studies and, thereby, aims to arrive at a more robust conclusion. While meta-analyses are not infallible, they can help provide clarity on topics that lack consensus.

Furthermore, it is worth noting the quality and rigor of a meta-analysis is dictated, in part, by the criteria used to determine the scope of the meta-analysis. In this case, the authors chose to restrict the meta-analysis to human RCTs in which the intervention was saturated fat restriction and the primary outcome was cardiovascular disease. In so doing, they necessarily filtered out observational studies which are at higher risk for bias.

New Meta-Analysis: No Benefit of Saturated Fat Restriction on Cardiovascular Disease

In the end, they included nine RCTs with a total of 13,352 participants and found no benefits for saturated fat reduction on cardiovascular mortality, all-cause mortality, heart attacks, or all coronary events.

But it’s always better to show than tell. So below are the graphs, with some further commentary. (If you prefer, you can skip down to the section: “No Correlation Between Degree of Saturated Fat Reduction and Outcomes.”)

How to Interpret a Forest Plot:

These below graphs are called Forest Plots: graphical representations often used in meta-analyses.

• Each row represents a different study, which you can see named, e.g. “Ramsden CE 2013.” This is the name of the first author and year of publication.

• There are many columns, but let’s focus on the most important: the Odds Ratio (OR), which expresses the odds of an event occurring in one group compared to the odds of that event occurring in another group. In this case, the odds of dying from cardiovascular disease in saturated fat-restricted groups versus control groups.

• If an OR is >1, it suggests the odds of dying are higher if you were in the saturated fat-restricted group; if an OR is <1, it suggests the odds of dying are lower if you were in the saturated fat-restricted group, i.e. the intervention group.

• However, we can’t just look at the OR alone. We need to look at the confidence interval (CI) associated with a given OR, which tells us about the statistical variability. In general, the larger the range of CI, the more the statistical noise. The CI is represented as two numbers between the brackets [## - ##] as well as by the horizontal line on the graph.

• Importantly, if the CI includes 1 and the horizontal line crosses the vertical line that represents 1, it means the result is not statistically significant.

• As an example, see Ramsden CE 2013. The OR is 1.48, suggesting – maybe? – that saturated fat reduction increases risk of death from cardiovascular disease. However, the CI is large and crosses 1 [0.85 – 2.55]. Therefore, the results are not statistically significant, and we cannot conclude that there is any difference between groups.

• Finally, the little diamond at the bottom represents the totality of the trials. The OR is 0.94 and the CI [0.81 – 1.11] crosses 1.

Conclusion: the evidence available from RCTs does not support saturated fat restriction for the prevention of cardiovascular disease [-related death].

*Forest Plot of saturated fat reduction on cardiovascular mortality, i.e. death from heart disease. Again, the OR is 0.94 and the CI is basically centered on 1. Thus, the evidence does not support saturated fat restriction for the prevention of cardiovascular disease-related death.

*Forest Plot of saturated fat reduction on all-cause mortality, i.e. death from any cause. The OR is 1.01 and the CI is centered on 1. Thus, the evidence does not support saturated fat restriction for the prevention of death from any cause.

*Forest Plot of saturated fat reduction on heart attacks. Again, the CI includes 1 and, thus, we cannot conclude saturated fat restriction prevents heart attacks. However, the OR of 0.85 is further away from 1, and the CI only just doesn’t include 1. This is what we could call a “trend,” potentially suggestive of a signal in the available RCTs that saturated fat restriction reduces risk of non-fatal heart attacks. It’s fair to conclude more data are needed.

*Forest Plot of saturated fat reduction on any coronary event. The interpretation here is like that above on saturated fat restriction and heart attacks.

No Correlation Between Degree of Saturated Fat Reduction and Outcomes

Now, as any statistician could tell you, just because a result is not statistically significant doesn’t mean a relationship doesn’t exist. It could be that a study (including a meta-analysis) is simply underpowered to detect a relationship. This is what’s called a “false negative” result. And while we cannot rule out this possibility, there are complementary ways to approach the data to strengthen the conclusions.

One way is to look for any “dose-response” relationship, either one that is statistically significant or trending.

Simply put: if we rank the studies by degree of saturated fat restriction, do we start to see a pattern?

It could be that the difference in saturated fat intake between groups in some studies was too small to see an effect. Were this the case, we might expect a pattern to emerge where there was more benefit in studies with a greater relative degree of saturated fat restriction.

Was this the case?

No.

Quoting from the paper, “No association was observed between the difference in saturated fat intake and relative risk reduction or absolute risk reduction for any outcomes.”

In the rest of this letter, we will discuss:

• How this Meta-Analysis is different from (Better Than?) Prior Meta-Analyses

• Limitations of the Study and Existing Data

• Why there Isn’t Consensus on Saturated fat Reduction

• Momentum of the Status Quo

• Financial Conflicts of Interests

• “Saturated Fat Stereotyped”: SCFAs, MCTs, Odd-Chain Saturated Fat, Stearic Acid

• Table of Dietary Sources of “Special” Saturated Fats

• Individual Variation in Response

• How to get your cholesterol levels at home, on-demand

• Why this Matters: Dietary Guidelines and Beyond