Between 1992 and 1997, we conducted a case-control study of oesophageal can
cer in 3 areas of northern Italy. Cases were 275 men, ages 39-77 years (med
ian age 60), with a first incident squamous-cell carcinoma of the oesophagu
s. Controls were 593 men, ages 36-77 years (median age 60) admitted for acu
te illnesses, unrelated to tobacco and alcohol, to major hospitals of the a
reas under surveillance. Number of daily cigarettes was strongly associated
with risk [odds ratio (OR) for greater than or equal to 25 cigarettes/day
= 7.0]. Long-duration smoking showed particularly elevated ORs (OR = 6.4 fo
r greater than or equal to 35 years), and excess risk declined after smokin
g cessation (OR = 1.5 after greater than or equal to 10 years). Oesophageal
cancer risk steeply rose with increasing level of alcohol consumption, ORs
were 6.2 for 35-55 drinks and 24.5 for 84 drinks or more per week, No tren
d in risk emerged for duration of alcohol drinking or age at start of drink
ing. The risk in the highest joint level of alcohol drinking and current sm
oking was increased 130 folds (i.e., compatible with a multiplicative model
). Excess risk in drinkers chiefly derived from wine. In conclusion, alcoho
l drinking and cigarette smoking were both important, but: the roles of dos
e and duration of exposure differed. The association with alcohol was stron
ger than the one with smoking by exposure intensity, but apparently unaffec
ted by duration or other temporal variables. Int. J. Cancer 86:144-149, 200
0. (C) 2000 Wiley-Liss, Inc.