Alcohol-attributable cancer deaths in the United States doubled over three decades, rising from 11,361 in 1990 to 23,126 in 2023, according to a secondary analysis of Global Burden of Disease data published in The Lancet Regional Health Americas by researchers at Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine.
The finding that departs most from public assumption involves younger adults. Among people aged 20 to 54, colorectal cancer was the leading alcohol-attributable cause of cancer death in men, and breast cancer ranked first in women, followed by colorectal. Liver cancer, the association most people know, led only among adults 55 and older.
That matters because it lands on the age group where drinking is heaviest and where cancer screening has historically started latest. It also aligns with a broader pattern clinicians have been tracking: colorectal cancer rates rising in younger adults for reasons that remain incompletely explained.
Attributable Fractions Are Modeled, Not Counted
The method deserves to be stated plainly, because it shapes how much weight the numbers can carry. Nobody counted 23,126 death certificates naming alcohol. Researchers estimated the share of cancer deaths attributable to alcohol using population exposure data and established risk relationships, a standard epidemiological approach called an attributable fraction.
That approach produces estimates with real uncertainty ranges rather than exact counts. As summarized in CancerNetwork's account of the mortality analysis, age-standardized mortality rates were 3.8 per 100,000 in 1990, with a confidence interval from 1.7 to 6.9, rising to 4.1 per 100,000 in 2023, with an interval from 2.7 to 5.8. Those intervals overlap, which is why the rate change is modest even as the raw death count doubles.
Age-standardized rates are the figure that accounts for population growth and aging. They rose 14.3% in men and fell 16% in women across the period. The doubling of absolute deaths therefore reflects a larger, older population as well as changes in risk, and the study's own rate data make that distinction visible.
The share of cancer mortality attributable to alcohol did rise substantially in both sexes, by 92% in men and 16.7% in women. In 2023, most alcohol-attributable cancer deaths occurred in men, 17,477 of them, and in people 55 and older, 19,460. Liver cancer accounted for the most deaths that year at 7,686, followed by esophageal at 4,902 and colorectal at 3,818.
Corresponding author Chinmay Jani, chief fellow in hematology and oncology at Sylvester, said the striking part was the scope. The "most surprising finding was the breadth of alcohol's potential impact" across cancer types, he said in a University of Miami statement, noting that contributions to colorectal, esophageal, breast, pancreatic and prostate cancer are less widely understood than the liver connection. The findings were first presented at the 2025 ASCO annual meeting before publication.
Geography, and What the State Differences Suggest
The analysis found substantial variation by state. Washington, D.C., recorded the highest alcohol-attributable cancer mortality rates in 2023; Utah recorded the lowest. Those two endpoints track closely with drinking prevalence, which is the most straightforward reading, though state differences in population age structure and in cancer care access also contribute.
Liver cancer itself showed the steepest overall rise in age-standardized mortality across the study period, increasing 113.9%, even as the proportion of that mortality attributable to alcohol edged down slightly. Hepatitis C, obesity, and metabolic liver disease are competing drivers there, which is a useful reminder that alcohol is one contributor among several.
Public awareness has lagged the science by a wide margin. The International Agency for Research on Cancer classifies alcohol as a Group 1 carcinogen, the same category as tobacco smoke and asbestos. A U.S. Surgeon General advisory issued at the start of 2025 estimated that alcohol accounts for roughly 100,000 cancer cases and 20,000 cancer deaths a year and called for updated warning labels, as noted in coverage of the study and the Surgeon General advisory. The current label text dates to 1988 and does not mention cancer, and changing it would require an act of Congress.
MedicalDaily has previously reported on research finding no level of alcohol without risk. This analysis adds a different layer: which cancers the burden actually falls on, and at what ages.
Steps That Follow from the Finding
The practical implication is not abstinence as a mandate. It is that the relationship is dose-dependent, so reducing intake reduces risk, and that people weighing their own drinking now have clearer information about which cancers are involved.
Screening is the other concrete lever. Colorectal cancer screening is recommended beginning at age 45 for average-risk adults, and these findings reinforce that starting on time matters for people in their forties. Anyone with a family history, inflammatory bowel disease or symptoms such as rectal bleeding, persistent change in bowel habits or unexplained weight loss should be evaluated regardless of age.
For women, breast cancer being the leading alcohol-attributable cancer death under 55 is worth raising at an annual visit alongside mammography timing and family history. For heavy drinkers, discussing liver assessment and, where appropriate, hepatitis screening with a clinician is reasonable.
People who want to cut back and find it difficult should know that alcohol use disorder is treatable, with medications and counseling both available. Anyone drinking heavily and daily should not stop abruptly without medical guidance, because withdrawal can be dangerous.
What remains unknown is how much of the doubling reflects changes in drinking versus improved detection and coding of the underlying cancers, whether warning label legislation advances, and whether the under-55 colorectal pattern continues. The researchers describe alcohol as a modifiable risk factor, which is the part households can act on now.
Key Questions Answered
What did the study find? Alcohol-attributable U.S. cancer deaths rose from 11,361 in 1990 to 23,126 in 2023, with colorectal cancer leading in men under 55 and breast cancer in women under 55.
Does this prove alcohol caused those deaths? It estimates a share using modeled attributable fractions rather than counting individual cases. The approach is standard but produces estimates with uncertainty ranges.
Did the rate double too? No. Age-standardized rates rose modestly, up 14.3% in men and down 16% in women. The doubling of absolute deaths partly reflects a larger, older population.
Which cancers are involved? Liver, colorectal, breast, esophageal, pancreatic and others. Alcohol is a recognized risk factor for multiple cancer types, including several head and neck sites.
Which states had the highest rates? Washington, D.C., had the highest alcohol-attributable cancer mortality in 2023, and Utah the lowest.
Is there a safe amount? Risk is dose-dependent, and major reviews have found no level without added risk. Less consumption means lower risk.
What should someone do with this information? Start colorectal screening at 45 if at average risk, raise drinking with a clinician, and seek medical guidance before stopping heavy daily drinking.
