Editorial NexoScience
How much does cancer risk climb when someone both drinks regularly and smokes, compared with doing only one of the two? For years, most research treated alcohol and tobacco as separate risk factors, or lumped one in as a statistical adjustment for the other. A new analysis pooling decades of data suggests that treating them separately misses most of the story.
Both substances already carry a heavy individual burden. The International Agency for Research on Cancer classifies alcohol and tobacco as Group 1 carcinogens, its highest-certainty category. Alcohol alone was linked to roughly 741,300 new cancer cases worldwide in a single recent year. Smoking’s toll is larger still. Researchers behind the new work wanted to know what happens when both habits overlap in the same person, something few earlier studies had measured directly.
The risk does not simply add up, it multiplies. The team pooled 24 studies, four cohort studies and twenty case-control studies, screened down from an initial 4,452 candidates found across five literature databases. For head and neck cancers, light drinking combined with moderate smoking was associated with roughly four times the risk seen in people with neither habit. Push both habits to heavy levels and the combined risk jumped past thirty-five times the baseline, far beyond what simply adding the two individual risks would predict. Statisticians call that pattern a multiplicative interaction, and it showed up consistently across the data.
Oral and throat cancers showed the sharpest jump. Among heavy drinkers who also smoked heavily, oral cancer risk landed around thirty-six times higher than in people who did neither, and laryngeal cancer risk came in near thirty-nine times higher. Esophageal and pancreatic cancers followed a similar pattern, with combined exposure consistently outpacing what either habit alone would suggest.
Even modest drinking changed the picture once smoking entered the equation. People who drank only lightly or moderately but smoked heavily still showed meaningfully higher cancer risk than smoking alone would predict on its own. The interaction between the two habits held at every exposure level the researchers tested, not just among the heaviest drinkers and smokers.
None of this comes from a randomized experiment. The underlying studies were observational, mostly case-control designs relying on people’s own memory of how much they drank and smoked years earlier, and the authors flagged that about half of the included studies carried a high risk of bias tied to that kind of self-reported exposure. Statistical checks also turned up signs of publication bias, which the team adjusted for using a standard correction method. What survives that scrutiny is a pattern, not a proven mechanism. Still, the same dose-response gradient, more of either habit and more of both together tracking higher risk, showed up again and again across studies spanning several countries and decades, which is the kind of consistency epidemiologists take seriously even without a controlled trial to confirm it.
The findings come from Jun S, Park H, Kim UJ, et al., “The Combined Effects of Alcohol Consumption and Smoking on Cancer Risk by Exposure Level: A Systematic Review and Meta-Analysis,” published in the Journal of Korean Medical Science (2024), DOI: 10.3346/jkms.2024.39.e185.
