Your morning coffee may be doing your liver a quiet favour. One of the largest studies yet links higher coffee consumption to markedly lower risks of serious liver disease — and finds decaf works about as well as regular.

Researchers at Cedars-Sinai Medical Center, writing in Clinical Gastroenterology and Hepatology, followed 354,957 UK Biobank participants for a median of 13 years, using medical records, blood tests and liver MRI scans.

What the numbers showed

Compared with non-drinkers, people drinking 5+ cups a day had:

32% lower risk of cirrhosis. 47% lower risk of liver cancer. 42% lower risk of liver-related death.

Benefits appeared at just 1–2 cups a day and were strongest around 3–4 cups. MRI scans showed less liver fat, iron, scarring and inflammation in coffee drinkers.

Why the imaging matters more than the risk figures

The MRI component distinguishes this from earlier work and strengthens it considerably.

Disease risk figures come from records of who was diagnosed, which is subject to differences in how often people see doctors and how thoroughly they are investigated — and those differ systematically between people with different habits.

Imaging measures the organ directly, in everyone, regardless of whether they sought care. Finding less fat, iron and scarring on scans means the association is present in measurable tissue rather than only in diagnosis rates.

The iron finding is the least expected. Iron accumulation in the liver contributes to oxidative damage, and coffee compounds are known to bind iron and reduce its absorption from the gut — a plausible mechanism entirely separate from caffeine.

What the decaf result rules out

Both caffeinated and decaf coffee showed similar benefits, which suggests caffeine is not the whole story.

This is the most informative detail in the study. Caffeine is the obvious candidate and the one most often invoked, and equivalent benefit without it effectively removes caffeine as the explanation.

What remains are the hundreds of other compounds coffee contains. Chlorogenic acids, present in substantial quantities and largely unaffected by decaffeination, have antioxidant and anti-inflammatory activity. Diterpenes and melanoidins formed during roasting have also been proposed.

The decaf finding also has practical consequence: it means any benefit is available to people who avoid caffeine for sleep, anxiety or cardiac reasons.

The dose pattern

Benefit appearing at 1–2 cups and peaking around 3–4 is a reasonable-looking dose-response, and it is worth reading carefully.

A relationship that rises then plateaus is generally more credible than one increasing indefinitely, because biological effects saturate.

That the 5+ group shows the headline figures while 3–4 is described as strongest suggests the curve flattens rather than continuing to climb — so the practical message is a few cups, not as many as possible.

The caveat

This is an observational study: it shows a strong association, not proof that coffee causes the protection.

Reverse causation deserves specific attention here. People with early liver disease often lose their taste for coffee, and those who drink heavily reduce it too — so some of the association may reflect that unwell people drink less coffee rather than that coffee keeps people well.

Thirteen years of follow-up mitigates that partly, since it is a long time for undiagnosed disease to explain the pattern, and it does not remove it.

Why liver disease is becoming more common

The interest in anything protective reflects a shift in what causes liver disease in most countries.

Alcohol and viral hepatitis were historically the dominant causes. Hepatitis C is now curable with short courses of antiviral therapy, and hepatitis B is vaccine-preventable, so the viral contribution is declining where treatment and vaccination reach people.

Metabolic liver disease has expanded to fill that space. Fat accumulation associated with obesity and insulin resistance now affects a substantial share of adults in many populations, and a minority of those progress through inflammation to fibrosis and cirrhosis.

The proportion progressing is small; the denominator is enormous, which makes the absolute numbers large. Liver cancer arising on that background has risen accordingly.

Treatment options remain limited. The first drugs for metabolic liver disease have only recently been approved, and the mainstay is weight loss, which is difficult to sustain. That gap is why observational findings about ordinary dietary factors attract more attention here than they might in a field with effective therapy.

What UK Biobank can and cannot represent

The cohort behind this finding shapes how far it generalises, and its characteristics are well documented.

UK Biobank recruited volunteers aged 40 to 69 who agreed to extensive assessment and long-term follow-up. Participants are healthier, wealthier, better educated and less ethnically diverse than the UK population, and smoke and drink less — a pattern common to volunteer cohorts, since people who join long research studies differ systematically from those who do not.

For estimating how common a disease is, that skew matters a great deal. For estimating a relationship between an exposure and an outcome, it matters less: the association between coffee and liver disease can hold within a healthier-than-average group even if disease rates there are lower overall.

Where it does bite is in interactions. If coffee protects mainly against liver damage from heavy alcohol use, a cohort that drinks less than average would understate the effect. If the benefit depends on something correlated with affluence, the cohort would overstate it. Neither can be resolved from within the study.

Coffee affects people differently, so this is not a prescription. General information, not medical advice.