Background. Few studies have investigated the association between age at st
arting smoking and the average number of cigarettes smoked per day in adult
hood. To provide further evidence on this issue, we analyzed data from the
Catalan Health Interview Survey (CHIS).
Methods. The CHIS was conducted in 1994 on a randomly selected sample (N =
15,000) of the population of Catalonia, Spain. A total of 4,897 current or
exsmokers (3,276 males and 1,621 females) were included for analysis. Age-s
tandardized proportions of subjects smoking <15, 15-24, and greater than or
equal to 25 cigarettes/day, age-standardized mean number of cigarettes smo
ked per day, and multivariate odds ratios (OR) of being a heavy smoker (gre
ater than or equal to 25 cigarettes/day) according to age at starting smoki
ng (<15, 15-17, 18-19, greater than or equal to 20 years) were computed.
Results. Men who started smoking before the age of 15 smoked on average 5.5
cigarettes more than those who started at age 19 or over. Women who starte
d smoking early in life smoked, on average, 6.8 cigarettes/day more than wo
men who started later. The proportion of smokers of <15 cigarettes/day was
higher among subjects who started smoking later. Both for males and for fem
ales, the OR of being a heavy smoker significantly increased with decreasin
g age at starting smoking (OR = 2.4 for males and 4.5 for females who start
ed at age <15 versus greater than or equal to 20 years). The level of educa
tion did not modify the relationship in males, whereas the association with
age at starting was only apparent for more educated women.
Conclusions. This study confirms that age at starting smoking is inversely
and strongly associated to the number of cigarettes smoked per day. Thus, a
ctions aimed at the prevention or delay of smoking onset among adolescents
would have an important beneficial effect. (C) 1999 American Health Foundat
ion and Academic Press.