Oral cancer: The association between nation-based alcohol-drinking profiles and oral cancer mortality

The unclear association between different nation-based alcohol-drinking profiles and oral cancer mortality was investigated using, as observational units, 20 countries from Europe, Northern America, Far Eastern Asia, with cross-nationally comparable data. Stepwise multiple regression analyses were run with mate age-standardised, mortality rate (ASMR) as explanatory variable and annual adult alcohol consumption, adult smoking prevalence, life expectancy, as explanatory. Large between-country differences in ASMR (range, 0.88-6.87 per 100,000) were found, but the mean value was similar to the global estimate (3.31 vs. 3.09 per 100,000). Differences in alcohol consumption (2.06-21.03 annual litres per capita) and in distribution between beverages were reported. Wine was the most prevalent alcoholic beverage in 45% of cases. Significant increases in ASMR for every litre of pure ethanol (0.15 per 100,000; 95 Cl, 0.01-0.29) and spirits (0.26 per 100,000; 95 CI, 0.03-0.49), non-significant effects for beer and wine were estimated. The impact of alcohol on oral cancer deaths would be higher than expected and the drinking profile could affect cancer mortality, probably because of the different drinking pattern of spirit drinkers, usually consuming huge alcohol quantities on single occasions, and the different concentrations of ethanol and cancer-preventing compounds such as polyphenols, in the various beverages. (C) 2005 Elsevier Ltd. All rights reserved.