The error you are making is that the authors are calculating risk. That is, very broadly, the chance of getting cancer goes up if you drink. The authors do not

The error you are making is that the authors are calculating risk. That is, very broadly, the chance of getting cancer goes up if you drink. The authors do not claim that drinking == cancer. That is ridiculous and would rightly be ignored by cancer researchers. Lots of people drink and smoke and will not ever get cancer. The authors are not disputing this. They do however suggest that drinking might increase your risk of getting cancer.