Alcohol drives most cirrhosis; metabolic risk matters most at moderate intake

Design

Population-based cohort linked to national registers

Population

354,624 Danish adults without cirrhosis, median age 55

Primary outcome

10-year risk of cirrhosis

Effect

10-year risk 0.14% (1–7 drinks/week) to 4.6% (>42); attributable fraction alcohol 59.9%, hypertension 29.3%, smoking 21.2%

This Danish cohort linked 354,624 adults from national health surveys (2010–2017) to registers, with 3.3 million person-years of follow-up and 1211 cases of cirrhosis. It estimated 10-year risks by alcohol intake and cardiometabolic risk factors.

Risk rose steeply with drinking: 10-year cirrhosis risk was 0.14% at 1–7 drinks a week, 0.68% at 15–28, 2.0% at 29–42 and 4.6% above 42. Obesity, diabetes, hypertension and smoking were each associated with two to three times the risk. About 60% of cirrhosis was attributable to alcohol, against 29% for hypertension, 21% for smoking, 14% for obesity and 9% for diabetes. Metabolic factors added risk mainly among people drinking up to 28 drinks a week; above that, alcohol dominated. Smoking added risk at every level of drinking.

This is directly relevant to MetALD, the overlap of metabolic and alcohol-related liver disease that is common in Indian hepatology clinics. For moderate drinkers with metabolic risk, both need addressing. For heavy drinkers, reducing alcohol comes first, and weight or glucose control alone will not offset it.

• Quantify alcohol intake in drinks per week at every liver assessment
• In moderate drinkers with obesity, diabetes or hypertension, address alcohol and metabolic risk together
• In heavy drinkers, make alcohol reduction the priority
• Include smoking cessation as part of liver disease prevention
• Consider non-invasive fibrosis assessment for patients with combined risk factors

Why it matters

It shows how alcohol and metabolic risk combine, which matters for the growing group of patients with both.

Don't overread it

This was observational, with self-reported drinking in a Danish population; the fractions assume causation and may differ elsewhere.

The statistics, in plain English

Population attributable fraction estimates how much cirrhosis would not occur if a risk factor were removed, assuming the association is causal. The 60% for alcohol is much larger than for any single metabolic factor. Absolute 10-year risks stay low below 15 drinks a week, which helps when counselling individuals.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.