The Diet-Heart Trial That Sat in a Basement for 40 Years
In April 2016, the BMJ published a study most people have never heard of, even though it ran for five years, randomized 9,423 adults, and asked one of the most consequential questions in modern nutrition: when you replace saturated fat with vegetable oil rich in linoleic acid, do people live longer?
The answer in this trial — uncovered after the original data sat unpublished for four decades — was no. In fact, the participants whose cholesterol dropped the most also died at higher rates than the participants whose cholesterol barely moved.
This is the Minnesota Coronary Experiment (MCE). One peer-reviewed trial. One careful re-analysis. And a result that sits awkwardly next to the dietary advice many of us grew up with.
What was the Minnesota Coronary Experiment?
The Minnesota Coronary Experiment was a double-blind, parallel-group, randomized controlled trial conducted between 1968 and 1973 in one nursing home and six Minnesota state mental hospitals. Researchers randomized 9,423 institutionalized adults aged 20 to 97 to either a diet replacing saturated fat with vegetable oil rich in linoleic acid (corn oil and corn-oil polyunsaturated margarine) or a control diet. The intervention reduced saturated fat from 18.5% to 9.2% of calories and increased linoleic acid from 3.4% to 13.2% of calories. The serum-cholesterol cohort included 2,355 participants with a mean baseline cholesterol of 208 mg/dL. The trial's primary outcomes were serum cholesterol response and mortality. Results from the full dataset were not published in the peer-reviewed literature for over 40 years, until Ramsden et al. published a re-analysis in BMJ in April 2016.
The intervention diet, in detail
The two diets were designed to be matched in calories, protein, and total fat — only the type of fat differed. The intervention used corn oil in cooking and corn-oil-based polyunsaturated margarine in place of butter, lard, beef tallow, and other animal fats. The control diet kept saturated-fat sources at typical mid-century American levels.
| Dietary component | Control diet | Intervention diet |
|---|---|---|
| Saturated fat (% of calories) | 18.5% | 9.2% |
| Linoleic acid (% of calories) | 3.4% | 13.2% |
| Primary fat source | Mixed animal + vegetable | Corn oil + corn-oil margarine |
The intervention reduced saturated fat by roughly half and increased linoleic acid by more than 280%. By design, this is the swap many public-health bodies recommended — and continue to recommend — to lower serum cholesterol.
What did the trial find?
In the Minnesota Coronary Experiment, the linoleic-acid intervention group's mean serum cholesterol fell by 31.2 mg/dL (a 13.8% decrease, P<0.001), while the control group's cholesterol fell by 5 mg/dL (1.0%, P<0.001). Despite this large cholesterol reduction, Ramsden et al. (2016) found that a 30 mg/dL reduction in serum cholesterol was associated with 22% higher risk of death from any cause
(hazard ratio 1.22, 95% CI 1.14 to 1.32). Among participants aged 65 and older, the relationship was even stronger: a 30 mg/dL cholesterol decrease was associated with 35% higher risk of death (hazard ratio 1.35, 95% CI 1.18 to 1.54). The authors noted, There was no evidence of benefit in the intervention group for coronary atherosclerosis or myocardial infarcts.
| Outcome | Result (Ramsden et al., 2016) |
|---|---|
| Cholesterol change — intervention | −31.2 mg/dL (−13.8%) |
| Cholesterol change — control | −5 mg/dL (−1.0%) |
| All-cause mortality, per 30 mg/dL cholesterol drop | HR 1.22 (95% CI 1.14 to 1.32) |
| All-cause mortality in age 65+, per 30 mg/dL drop | HR 1.35 (95% CI 1.18 to 1.54) |
| Coronary atherosclerosis benefit in intervention | None observed |
| Myocardial infarct benefit in intervention | None observed |
The authors summarized the paradox plainly: Paradoxically, MCE participants who had greater reduction in serum cholesterol had a higher rather than a lower risk of death.
Why didn't we hear about this for 40 years?
The Minnesota Coronary Experiment was conducted between 1968 and 1973, but the full survival analysis was not published in peer-reviewed literature until 2016. A limited paper appeared in 1989 (Frantz et al., Arteriosclerosis) without the full mortality and pathology analyses. A 1981 thesis by Steven Broste contained survival data but was never cited in the peer-reviewed literature. Ramsden et al. recovered the dataset between 2013 and 2015 from two 9-track magnetic tapes and extensive paper documents — green-bar files, autopsy folders, FORTRAN coding sheets, and grant proposals. They then ran the full pre-specified analyses and published the findings in BMJ in 2016. The authors concluded: Findings from the Minnesota Coronary Experiment add to growing evidence that incomplete publication has contributed to overestimation of the benefits of replacing saturated fat with vegetable oils rich in linoleic acid.
That sentence is doing a lot of work. The researchers aren't claiming linoleic acid causes mortality. They're claiming that a major randomized trial designed to test the hypothesis didn't find the protective effect that decades of dietary advice had been built around — and that incomplete publication of trials like this one has skewed the evidence base.
What the authors did not say
This is the part where it matters to be careful. Ramsden et al. did not conclude that linoleic acid causes heart disease, or that saturated fat is protective, or that you should change your diet based on this single trial. Their conclusion was narrower and more important:
Available evidence from randomized controlled trials shows that replacement of saturated fat with linoleic acid effectively lowers serum cholesterol but does not support the hypothesis that this translates to a lower risk of death from coronary heart disease or all causes.— Ramsden et al., BMJ 2016
Read that again. Lowering cholesterol happened. The trial confirmed that. Lowering mortality did not. The trial did not confirm that — and arguably suggested the opposite, at least in older adults.
The trial's limitations (because honesty matters)
No single study answers a complex question, and this one has limitations the authors disclosed in detail:
- Participants were followed only while hospitalized; only about 25% remained in the trial for one year or longer.
- The complete dataset was not recovered — notably, the full autopsy cohort (149 of 295 participants).
- The study population was institutionalized adults in mental hospitals and a nursing home, which limits generalizability to free-living adults.
- LDL subfractions were not measured.
- Trans-fat content of the diets was not available, so partial confounding from industrial trans fats in margarine is possible.
These caveats matter. They limit how confidently the result can be applied to a 35-year-old in 2026 cooking dinner at home. They do not, however, erase the directional finding — which is that in the largest randomized trial ever to test the saturated-fat-replacement hypothesis, lower cholesterol did not translate into longer life.
Where this leaves us, in our view
We're a chip company, not a clinical research lab. But our position on cooking fats was shaped by reading studies like this one. In our view, the evidence base for replacing saturated animal fats with linoleic-acid-rich vegetable oils is weaker than the dietary advice of the last 50 years suggested. Ramsden et al. (2016) is one of several reasons we think that.
We covered the chemistry of frying with these oils in what happens to seed oils when you fry — Grootveld and colleagues' aldehyde measurements at frying temperatures. We covered the longer historical arc in the history of seed oils in America. And we looked at the omega-6 to omega-3 ratio question, where credible researchers genuinely disagree, in why the omega-6 to omega-3 ratio matters. The Minnesota Coronary Experiment is one more piece of the picture. It's not the only piece, and we'll keep posting the other pieces as we read them.
Where beef tallow fits
Beef tallow is what was largely removed from the American diet during the same decades the cholesterol-lowering hypothesis was being tested. Tallow is mostly saturated and monounsaturated fat — stable at high temperatures, low in linoleic acid (around 3% by mass), and chemically much more like the fats human food culture used for thousands of years before industrial vegetable-oil refining became cheap.
| Cooking fat | Saturated | Monounsaturated | Linoleic acid (omega-6) |
|---|---|---|---|
| Beef tallow | ~50% | ~42% | ~3% |
| Corn oil | ~13% | ~28% | ~57% |
| Soybean oil | ~15% | ~23% | ~51% |
| Sunflower oil (high-linoleic) | ~10% | ~20% | ~66% |
That's why TIPS Sea Salt chips are cooked in 100% beef tallow. It's why our Jalapeño and Salt & Vinegar chips are too. And it's why our Variety Pack exists for people who want to try every flavor without committing to one. We made a chip that's cooked the way chips were cooked before the swap that the Minnesota Coronary Experiment tested didn't deliver the benefit it promised.
You can read more about the broader case for tallow on our seed-oil-free chips landing page, which is the home base for everything we write on this topic.
Read Next
- What Happens to Seed Oils When You Fry — The Food Chemistry Explained
- Omega-6 to Omega-3: Why the Ratio Matters (And Where Researchers Disagree)
- The History of Seed Oils in America — How Vegetable Oils Replaced Traditional Fats
Frequently Asked Questions
What was the Minnesota Coronary Experiment?
The Minnesota Coronary Experiment was a double-blind, randomized controlled trial conducted from 1968 to 1973 in one Minnesota nursing home and six state mental hospitals. It enrolled 9,423 institutionalized adults aged 20 to 97 and tested whether replacing saturated fat with vegetable oil rich in linoleic acid (corn oil and corn-oil margarine) reduced mortality and heart disease. The intervention reduced saturated fat from 18.5% to 9.2% of calories and increased linoleic acid from 3.4% to 13.2% of calories. The full survival analysis was not published in peer-reviewed literature until 2016, when Ramsden and colleagues recovered the original data from magnetic tapes and paper records and published their re-analysis in BMJ.
What did the Minnesota Coronary Experiment find?
The Minnesota Coronary Experiment found that replacing saturated fat with linoleic-acid-rich vegetable oil lowered serum cholesterol substantially (a mean drop of 31.2 mg/dL in the intervention group versus 5 mg/dL in controls) but did not lower mortality. Ramsden et al. (2016) reported that a 30 mg/dL cholesterol reduction was associated with a 22% higher risk of death from any cause (hazard ratio 1.22, 95% CI 1.14 to 1.32). Among participants aged 65 and older, the same cholesterol drop was associated with a 35% higher risk of death (hazard ratio 1.35, 95% CI 1.18 to 1.54). The authors found no evidence of benefit for coronary atherosclerosis or myocardial infarcts in the intervention group.
Why was the Minnesota Coronary Experiment data published 40 years late?
The Minnesota Coronary Experiment data was published 40 years late because the full survival and pathology analyses were never written up at the time. A 1989 paper by Frantz et al. in Arteriosclerosis reported limited findings without the full mortality analyses. A 1981 thesis by Steven Broste contained survival data but was never cited in peer-reviewed literature. Christopher Ramsden and colleagues spent 2013 to 2015 recovering the original dataset from two 9-track magnetic tapes, autopsy folders, FORTRAN coding sheets, and grant proposals — then ran the pre-specified analyses and published in BMJ in April 2016. The authors argued that incomplete publication of trials like this one has contributed to overestimation of the benefits of saturated-fat replacement.
Does the Minnesota Coronary Experiment prove vegetable oils are dangerous?
No, the Minnesota Coronary Experiment does not prove vegetable oils are dangerous, and Ramsden et al. (2016) did not make that claim. Their conclusion was narrower: that randomized trial evidence shows replacing saturated fat with linoleic acid lowers cholesterol but does not support the hypothesis that this swap reduces death from heart disease or all causes. The trial has limitations — institutionalized population, short individual follow-up, partial autopsy data, no LDL subfraction measurement, possible trans-fat confounding — that the authors disclose. The directional finding (lower cholesterol did not translate into lower mortality) is what matters most for evaluating the diet-heart hypothesis as it was originally formulated.
What does this study mean for cooking at home?
The Minnesota Coronary Experiment tested a specific dietary swap — saturated fat replaced with corn oil and corn-oil margarine — in institutionalized adults from 1968 to 1973. Direct application to a free-living adult cooking at home in 2026 has limits, and we don't think one trial should drive personal decisions. What the trial does do is challenge the confidence with which the saturated-fat-replacement hypothesis has been promoted. In our view, choosing cooking fats deserves more curiosity and less certainty than public-health messaging often allows. Reading the primary source (Ramsden et al., BMJ 2016, DOI 10.1136/bmj.i1246) is the best way to form your own view.
Sources
- Ramsden CE, Zamora D, Majchrzak-Hong S, Faurot KR, Broste SK, Frantz RP, Davis JM, Ringel A, Suchindran CM, Hibbeln JR. Re-evaluation of the traditional diet-heart hypothesis: analysis of recovered data from Minnesota Coronary Experiment (1968–73). BMJ 2016;353:i1246. DOI: 10.1136/bmj.i1246. PMID: 27071971.