Although cancer mortality has declined in the US over the past three decades, a large observational trend analysis found the opposite overall trend for alcohol-attributable cancer mortality.

The research team estimated that, between 1990 and 2023, the number of cancer deaths attributed to alcohol more than doubled from 11,361 to 23,126. After accounting for population size and age, the alcohol-attributable cancer mortality rate increased by nearly 8%.

“Our study adds a long-term, population-level perspective on how alcohol-attributable cancer mortality has changed in the United States over more than three decades,” said lead author Chinmay Jani, MD, medical oncologist at Sylvester Comprehensive Cancer Center at the University of Miami in Miami.

One of the “most striking findings,” Jani said, was the fact that “overall cancer mortality has declined considerably because of improvements in prevention, screening, diagnosis, and treatment, yet alcohol-attributable cancer mortality did not consistently follow the same pattern.”

Jason Oke, DPhil, who wasn’t involved in the study, cautioned that the findings should be interpreted as estimates of population-level burden rather than direct measurements of cancer deaths caused by alcohol.

“The analysis is based on modeled or predicted alcohol-attributable burden/mortality and not directly observed data,” said Oke, of the Centre for Evidence-Based Medicine at the University of Oxford in Oxford, England. The analysis, published in The Lancet Regional Health - Americas, requires assumptions and approximations, residual confounding cannot be excluded, and the effects of alcohol can be difficult to disentangle from related factors such as smoking and obesity, he explained.

Alcohol’s Share of Cancer Mortality Grows

Alcohol has been classified as a group 1 carcinogen by the International Agency for Research on Cancer, with established causal links to several cancers, including those of the breast, colorectum, liver, esophagus, and upper aerodigestive tract.

Previous studies, however, have not comprehensively evaluated national cancer mortality trends attributable to alcohol across multiple cancer types, the researchers explained. Jani and colleagues addressed this gap by analyzing alcohol-attributable cancer deaths in the US overall and for 10 individual cancer types, with analyses stratified by sex, age, and state.

From 1990 to 2023, the age-standardized alcohol-attributable cancer mortality rate increased from 3.8 to 4.1 deaths per 100,000 population, rising 7.9% overall. At the same time, alcohol accounted for a larger share of all cancer mortality, increasing from 2.2% of cancer deaths in 1990 to 3.5% in 2023 — a roughly 59% relative increase in that proportion.

Alcohol-related cancer mortality trends differed by sex, age, cancer type, and location.

Alcohol-attributable cancer mortality rates increased 14.3% among men but fell 16% among women. At the same time, the proportion of all cancer mortality attributable to alcohol increased 92% among men and 16.7% among women.

Age patterns also differed. The alcohol-attributable cancer mortality rate remained essentially unchanged among adults aged 20-54 years but increased 21.5% among those aged 55 years or older.

Among men, the trends were particularly notable. In the younger cohort, the overall cancer mortality rate declined by 34%, whereas alcohol-attributable cancer mortality rate increased by 14%. In the older cohort, the corresponding changes were more prominent — a 37% decline and a 24% increase, respectively.

Liver cancer accounted for the largest number of alcohol-attributable cancer deaths in 2023 and had an age-standardized mortality rate of 1.3 deaths per 100,000 population, followed by esophageal cancer (0.9) and colorectal cancer (0.7). The alcohol-attributable liver cancer mortality rate increased 114% from 1990 to 2023 (0.6 to 1.3 deaths per 100,000 population), the largest increase in alcohol-attributable cancer mortality among the cancer types studied.

Meanwhile, the proportion of cancer mortality attributable to alcohol increased most for stomach cancer (57.2%), pancreatic cancer (40.5%), and colorectal cancer (39.7%); for liver cancer, the proportion declined (0.6%).

Among adults aged 55 years or older, liver cancer had the highest alcohol-attributable mortality rate in men, whereas breast cancer had the highest rate in women. Among adults aged 20-54 years, colorectal cancer ranked highest in men and breast cancer in women. Liver cancer mortality increased particularly sharply in this younger group, rising 89% in men and 148% in women.

The researchers highlighted geographic differences as well. Washington, DC, had the highest alcohol-attributable cancer mortality rates in 2023 whereas Utah had the lowest. Rates increased among men in 47 states and among women in 17 states, with some of the largest increases among men in Oklahoma, New Mexico, West Virginia, and Tennessee.

“So the [study’s] key contribution is not new evidence that alcohol causes cancer, but a clearer understanding of where the alcohol-attributable burden is increasing and which populations may benefit most from targeted prevention and education,” Jani said.

The investigators did note several limitations to the analysis. The Global Burden of Disease estimates rely on modeled exposure and mortality data and may be affected by incomplete or delayed reporting, differences among data sources, and uncertainty introduced by age- and sex-splitting algorithms. The database also lacked granular patient-level information on race, ethnicity, and social determinants of health. Additionally, this observational analysis cannot establish causality.

Oke highlighted that the analysis does not account for the long latency between alcohol exposure and cancer death. Failure to account for latency makes it difficult to directly link changes in alcohol exposure during the study period with changes in cancer mortality during the same period. “Someone diagnosed with cancer in 2015 may have developed their cancer partly because of alcohol consumption 20 or more years earlier,” he said.

Oke also cautioned against placing too much emphasis on the share of cancer deaths attributable to alcohol because that measure can rise as cancer deaths decline overall. The change in the alcohol-attributable cancer deaths in the US population, he said, may provide a more useful gauge of the underlying trend.

Overall, the study “reinforces that cancer prevention discussions should include alcohol alongside tobacco and other established risk factors,” Jani said. “These conversations should be factual rather than alarmist: physicians can explain that cancer risk generally increases with alcohol exposure and that alcohol is linked to multiple cancers, including breast, colorectal, liver, and upper aerodigestive cancers.”

The study received no funding. Disclosures for the authors are available with the original study publication. Oke had no relevant disclosures.