• Heritability;
  • growth mixture models;
  • liability threshold models;
  • smoking trajectories;
  • substance use;
  • twins.


Aims  To investigate the relative contribution of genetic and environmental factors on smoking trajectory membership and to test whether individual smoking trajectories represent phenotypical thresholds of increasing genetic risk along a common genetic liability dimension.

Design  Prospective study of a birth cohort of female like-sex twin pairs.

Setting  Participants completed diagnostic interview surveys four times from adolescence (average age 16) to young adulthood (average age 25).

Participants  Female twins who had smoked ≥100 cigarettes life-time (n = 1466 regular smokers).

Measurements  Number of cigarettes smoked per day during the heaviest period of smoking (two waves) or during the past 12 months (two waves).

Findings  A four-trajectory class solution provided the best fit to cigarette consumption data and was characterized by low (n = 564, 38.47%), moderate (n = 366, 24.97%) and high-level smokers (n = 197, 13.44%), and smokers who increased their smoking from adolescence to young adulthood (n = 339, 23.12%). The best genetic model fit was a three-category model that comprised the low, a combined increasing + moderate and high trajectories. This trajectory categorization was heritable (72.7%), with no evidence for significant contribution from shared environmental factors.

Conclusions  The way in which smoking patterns develop in adolescence has a high level of heritability.