Considering age at smoking initiation we found that Finnish Twin Cohort subjects born at later decades began smoking at younger age compared to subjects born at earlier decades, which is consistent with a previous Swiss study (Morabia et al. 2002), and that females started smoking later than males, which is also in accordance with previous findings (Okoli et al. 2013). This is also consistent with the evolution of tobacco use in Finland in the 20th century. Although the Q–Q plots (Figs S1 and S2) suggest that the family correlation structure was well controlled for, a mild overdispersion is observed. This implies that common environmental factors may also play a role in smoking initiation even after adjusting for birth year. DZ twins share more environmental factors than non‐twin members of a family, and this environmental correlation structure is not captured by the used kinship matrix, which may lead to slight P‐value inflation.
