At Japan's low intake, saturated fat showed no link with death

In the mid-1990s, 90,807 Japanese adults aged 45 to 74, none with a history of cancer, stroke or ischaemic heart disease, filled in a detailed food questionnaire. Researchers from the Japan Public Health Center-based Prospective Study (JPHC) then followed them for an average of 18.9 years, during which 22,618 died. On 2 November the team reported in The Journal of Nutrition that people who ate more saturated or trans fat did not die more often than those who ate the least.

The cohort's "high" intake was low by Western standards

Fat was counted as a share of daily energy. Among the men, the top fifth for saturated fat got 8.9% of energy or more from it; among the women, 10.3% or more. The authors put the average at 6.9% for men and 8.3% for women, against about 12% in national survey data from the United States and the United Kingdom. The World Health Organization advises keeping saturated fat at or below 10% of energy, so most of this cohort sat under that line, and its "high" group sat around it.

Trans fat was rarer still. The top fifth started at 0.43% of energy in men and 0.53% in women; only 405 of the 90,807 participants went above the WHO ceiling of 1%.

The top and bottom fifths died at similar rates

After adjustment for age, study area, smoking, alcohol, body mass index (BMI), physical activity, total energy, share of energy from protein, vegetables, fruit, green tea, coffee, sleep, work and living alone (and, in women, menopausal status and hormone use), the risk of death from any cause in the top fifth of saturated fat versus the bottom fifth was 0.96 in men (95% CI 0.90–1.03) and 0.99 in women (0.92–1.07). For trans fat in men it was 0.99 (0.93–1.06). For saturated fat and total trans fat, these adjusted all-cause mortality comparisons showed no positive association; results for specific causes of death differed by fat subtype and sex.

Two signals in women stand out, and neither changes the main picture. Women who ate the most polyunsaturated fat had slightly lower all-cause mortality (hazard ratio 0.92, confidence interval touching 1.00), but the link did not survive the authors' correction for multiple comparisons and was absent among women with hypertension or diabetes. In women, more saturated fat went with more deaths from respiratory disease, and that one did survive the correction; the authors note that these women also ate fewer vegetables and fruit.

This range cannot settle the Western question

In Western populations, where saturated fat typically supplies around 12% of energy, earlier studies cited by the authors linked it to higher LDL cholesterol and cardiovascular risk, and a review of prospective studies prepared for the WHO found all-cause mortality about 14% higher for every 2 percentage points of energy from trans fat. A finding of "no signal" in a group that mostly never reached the WHO threshold says something narrower: within the range this cohort actually ate, extra saturated fat did not track with dying sooner. It says nothing about what happens at Western levels, because the cohort's mean intake was lower and the study did not report a separate comparison at those levels; trans fat in the studies of that review ranged up to 2.7% of energy, against a median of 0.53% in this cohort's top group of men.

The sources differ too. Part of the saturated fat in this cohort came from fish and cooking oil, not only meat and dairy, and the authors raise this as one possible reason for the gap with Western results.

One questionnaire from the 1990s carries the whole result

This is an observational cohort, so it shows associations, not effects. Diet was measured once, in 1995–1998, and not again over almost two decades, while the trans-fat content of Japanese foods has since fallen; the authors themselves caution against applying the result to today's population. A food questionnaire ranks people reasonably well but measures intake imprecisely, and with trans fat varying so little, a real effect could be hidden. The study was funded by Japan's National Cancer Center research fund.

For a reader eating a typical Western diet, the useful takeaway is the range: the study describes only the Japanese intake range observed here, and the study does not show that adding saturated fat is harmless. Where blood lipids are already raised, how much saturated fat to eat is a question to discuss with your doctor.

This material is for information only and does not replace a consultation with a doctor.

Sources

• Kito K, Ishihara J, Kotemori A, et al. Saturated and trans fatty acids were not associated with all-cause or cause-specific mortality in a Japanese population with relatively low intakes: the JPHC Study. J Nutr. 2026. PMID 42971358

• Reynolds AN, Hodson L, de Souza R, et al. Saturated fat and trans-fat intakes and their replacement with other macronutrients: a systematic review and meta-analysis of prospective observational studies. World Health Organization, 2022. WHO IRIS