Obesity, Diabetes, Coffee, Tea, and Cannabis Use Alter Risk for Alcohol-Related Cirrhosis in 2 Large Cohorts of High-Risk Drinkers : Official journal of the American College of Gastroenterology | ACG

Secondary Logo

Journal Logo


Obesity, Diabetes, Coffee, Tea, and Cannabis Use Alter Risk for Alcohol-Related Cirrhosis in 2 Large Cohorts of High-Risk Drinkers

Whitfield, John B. PhD, FRCPath1; Masson, Steven FRCP2; Liangpunsakul, Suthat MD3; Mueller, Sebastian MD, PhD4; Aithal, Guruprasad P. PhD, FRCP5; Eyer, Florian MD6; Gleeson, Dermot MD, FRCP7; Thompson, Andrew PhD8; Stickel, Felix MD, PhD9; Soyka, Michael MD10,11; Muellhaupt, Beat MD9; Daly, Ann K. PhD2; Cordell, Heather J. DPhil12; Foroud, Tatiana PhD13; Lumeng, Lawrence MD3,14; Pirmohamed, Munir PhD, FRCP8; Nalpas, Bertrand MD, PhD15,16; Jacquet, Jean-Marc MD15; Moirand, Romain MD, PhD17; Nahon, Pierre MD, PhD18,19,20; Naveau, Sylvie MD21; Perney, Pascal MD, PhD22; Haber, Paul S. MD, PhD23,24; Seitz, Helmut K. MD4; Day, Christopher P. MD, PhD25; Mathurin, Philippe MD, PhD26; Morgan, Timothy R. MD27,28; Seth, Devanshi PhD23,24,29;  for the GenomALC Consortium

Author Information
The American Journal of Gastroenterology 116(1):p 106-115, January 2021. | DOI: 10.14309/ajg.0000000000000833



Sustained high alcohol intake is necessary but not sufficient to produce alcohol-related cirrhosis. Identification of risk factors, apart from lifetime alcohol exposure, would assist in discovery of mechanisms and prediction of risk.


We conducted a multicenter case-control study (GenomALC) comparing 1,293 cases (with alcohol-related cirrhosis, 75.6% male) and 754 controls (with equivalent alcohol exposure but no evidence of liver disease, 73.6% male). Information confirming or excluding cirrhosis, and on alcohol intake and other potential risk factors, was obtained from clinical records and by interview. Case-control differences in risk factors discovered in the GenomALC participants were validated using similar data from 407 cases and 6,573 controls from UK Biobank.


The GenomALC case and control groups reported similar lifetime alcohol intake (1,374 vs 1,412 kg). Cases had a higher prevalence of diabetes (20.5% (262/1,288) vs 6.5% (48/734), P = 2.27 × 10−18) and higher premorbid body mass index (26.37 ± 0.16 kg/m2) than controls (24.44 ± 0.18 kg/m2, P = 5.77 × 10−15). Controls were significantly more likely to have been wine drinkers, coffee drinkers, smokers, and cannabis users than cases. Cases reported a higher proportion of parents who died of liver disease than controls (odds ratio 2.25 95% confidence interval 1.55–3.26). Data from UK Biobank confirmed these findings for diabetes, body mass index, proportion of alcohol as wine, and coffee consumption.


If these relationships are causal, measures such as weight loss, intensive treatment of diabetes or prediabetic states, and coffee consumption should reduce the risk of alcohol-related cirrhosis.

© 2020 by The American College of Gastroenterology

Full Text Access for Subscribers:

You can read the full text of this article if you:

Access through Ovid