What 53 Trials and 68,128 People Actually Showed About Low-Fat Diets

Fifty-three randomised controlled trials. Sixty-eight thousand, one hundred twenty-eight participants. Every trial running a year or longer. Analysed by researchers at Harvard.

The conclusion: low-fat diets did not produce greater long-term weight loss than higher-fat diets of comparable intensity. In the lead author’s own summary, there is no good evidence for recommending low-fat diets.¹

This article walks through that evidence and the saturated fat literature that followed, and ends on the marker that predicts cardiovascular risk better than the one most people have measured.

Getting the citation right

That meta-analysis is **Tobias and colleagues, published in Lancet Diabetes & Endocrinology in 2015**.

It is worth naming precisely, because this study is misattributed constantly — frequently to Sackner-Bernstein and colleagues in PLoS One, also 2015. That is a genuinely different and much smaller analysis: 17 trials and 1,797 participants. Both papers exist, both are real, and they are not the same study.

If you want to read the 53-trial analysis yourself, Tobias is the one to search for.

The Women’s Health Initiative

If the low-fat hypothesis were going to work anywhere, it should have worked here.

The Women’s Health Initiative dietary modification trial enrolled 48,835 postmenopausal women and followed them for an average of 8.1 years on a low-fat, calorie-reduced diet with increased physical activity.²

The intervention was genuinely implemented. Total fat intake fell by 8.2%. Participants consumed 361 fewer calories per day. That was sustained for eight years — which is a remarkable adherence result in itself.

The outcome: no significant weight loss. No reduction in coronary heart disease. No reduction in stroke.

That is what a large, well-funded, well-adhered-to test of the low-fat hypothesis produced.

Why fat behaves differently from carbohydrate

The mechanism the guidance missed is straightforward.

Fat is satiating, and it has the least effect on blood glucose and insulin of the three macronutrients. Protein has more effect. Carbohydrate has by far the most.

Excessive insulin production, followed by insulin resistance, is the process that precedes type 2 diabetes. So advice to hold total fat below 30% of calories did something worse than fail to help — it shifted a population toward the macronutrient that drives the insulin response, and away from the one that reliably switches appetite off.

The saturated fat evidence

This is the part most people find hardest to accept, so it is worth citing carefully.

In 2014, a meta-analysis of 76 studies covering more than 600,000 participants across 18 countries concluded that the available evidence does not clearly support cardiovascular guidelines encouraging high consumption of polyunsaturated fats and low consumption of total saturated fats.³

The following year, a systematic review and meta-analysis in the BMJ found that saturated fat intake was not associated with all-cause mortality, cardiovascular mortality, total coronary heart disease, ischaemic stroke, or type 2 diabetes.⁴

And in a randomised comparison published in 2015, diets using cheese and meat as the primary sources of saturated fat produced higher HDL cholesterol and apolipoprotein A-1 than a low-fat, high-carbohydrate diet — a less atherogenic profile, not a worse one.⁵

None of that means saturated fat is a health food to be pursued. It means the specific claim that saturated fat drives cardiovascular disease has not held up in systematic review, and dietary guidance built on it needs revisiting.

Real food versus extracted oil

Dairy fat is the clearest illustration. In cohort data, plasma saturated fatty acids derived from dairy sources — cheese, yogurt, milk — are inversely associated with incident type 2 diabetes.⁶ More dairy fat, less diabetes.

A separate cross-sectional study found that people in the highest tertile of whole-fat dairy intake had significantly lower odds of obesity. That study design shows association only. Reverse causation cannot be excluded, and people who choose full-fat dairy may differ in other ways.

But notice the direction of every one of these findings. It is the opposite of what most patients were told.

The pattern underneath is that fat arriving inside real food — packaged with the vitamins, minerals and antioxidants that evolved alongside it — behaves differently from fat that has been industrially extracted, heated, deodorised and bottled.

The marker that actually predicts risk

If saturated fat is not the driver, what is?

Replacing saturated fat with omega-6 vegetable oils does reliably lower LDL cholesterol. What it does not reliably do is improve cardiovascular mortality. Lowering the marker did not deliver the outcome — which should tell you something about whether the marker was ever the mechanism.

Total and LDL cholesterol are poor stand-alone predictors of cardiovascular risk. Elevated triglycerides combined with low HDL predict better, and that combination is a recognised signature of insulin resistance. Worth knowing: the QRISK calculator used across the UK does not use LDL at all — it uses the total cholesterol to HDL ratio.

The most striking estimate comes from the Archimedes simulation model applied to national survey data. In young adults, preventing insulin resistance was estimated to prevent 42% of myocardial infarctions — compared with 36% for correcting hypertension, 31% for low HDL, 21% for body mass index, and 16% for LDL.⁷

That figure comes from a mathematical model rather than a randomised trial, and that distinction matters. But every arrow points the same direction: insulin resistance is the largest single modifiable driver, and it is the one that is tested for least.

Frequently asked questions

Key takeaways

• The 53-trial, 68,128-participant meta-analysis is Tobias et al. 2015 — not the smaller study it is often confused with.
• The Women’s Health Initiative implemented a low-fat diet successfully for 8.1 years and found no weight, coronary or stroke benefit.
• Systematic reviews have not found saturated fat intake associated with mortality, coronary disease, stroke or diabetes.
• Lowering LDL by replacing saturated fat with omega-6 oils did not deliver the mortality benefit predicted.
• Triglyceride-to-HDL ratio, a marker of insulin resistance, tracks risk better than LDL alone.

Medically reviewed by Gurpreet Singh Padda, MD — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed July 2026.

This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary. Do not start, stop, or change any medication without consulting your physician. To be evaluated, request an appointment or call (314) 481-5000.

References

1. Tobias DK, Chen M, Manson JE, Ludwig DS, Willett W, Hu FB. Effect of low-fat diet interventions versus other diet interventions on long-term weight change in adults: a systematic review and meta-analysis. Lancet Diabetes & Endocrinology. 2015;3(12):968–979.
2. Howard BV, Manson JE, Stefanick ML, et al. Low-fat dietary pattern and weight change over 7 years: the Women’s Health Initiative Dietary Modification Trial. JAMA. 2006;295(1):39–49.
3. Chowdhury R, Warnakula S, Kunutsor S, et al. Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis. Annals of Internal Medicine. 2014;160(6):398–406.
4. de Souza RJ, Mente A, Maroleanu A, et al. Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes. BMJ. 2015;351:h3978.
5. Thorning TK, Raziani F, Bendsen NT, et al. Diets with high-fat cheese, high-fat meat, or carbohydrate on cardiovascular risk markers in overweight postmenopausal women. American Journal of Clinical Nutrition. 2015;102(3):573–581.
6. Mozaffarian D. Saturated fatty acids and type 2 diabetes: more evidence to re-invent dietary guidelines. Lancet Diabetes & Endocrinology. 2014;2(10):770–772.
7. Eddy D, Schlessinger L, Kahn R, Peskin B, Schiebinger R. Relationship of insulin resistance and related metabolic variables to coronary artery disease: a mathematical analysis. Diabetes Care. 2009;32(2):361–366.

Dr. Gurpreet Singh Padda, MD, MBA, MHP