Despite this general pattern, there is considerable variation within these phases��in particular, many individuals do not progress to regular use and continue as irregular nondependent smokers (sometimes called ��chippers��). By age 11 years, about one www.selleckchem.com/products/Bosutinib.html third of children in the United Kingdom have tried a cigarette, although only 1% smoke every week, and by age 15 years, about two thirds have tried at least one cigarette (Woodhouse, 2004). A recent U.K. survey showed that 20% of 11- to 16-year-olds smoked regularly (Action on Smoking and Health, 2007), with more girls now smoking than boys. This has obvious implications for the future. The majority of adults who are tobacco dependent started smoking as teenagers. In the United Kingdom alone, it is estimated that 3,000 teenagers a week start to smoke (Royal College of Physicians, 1992).

Those who start early are more likely to smoke as adults and are less likely to stop (DiFranza et al., 2002; Karp, O��Loughlin, Paradis, Hanley, & Difranza, 2005; Khuder, Dayal, & Mutgi, 1999; Wellman, DiFranza, Savageau, & Dussault, 2004). Despite the overall decline in smoking over the last three decades, in the United Kingdom, cigarette smoking is highest among 20- to 24-year-olds. It is estimated that 38% of males and 35% of females in this age group are smokers (Office for National Statistics, 2004). Experimentation usually commences between the ages of 11 and 13, and a complex mixture of factors may influence subsequent tobacco use behavior, including biological, attitudinal, interpersonal, and socioeconomic factors.

Mental health problems (e.g., depression and anxiety) may also increase the risk of smoking (Patton et al., 1998; Tyas & Pederson, 1998). Dependence often develops rapidly, and it has been suggested that the adolescent brain may be more sensitive to the effects of nicotine (Slotkin, 2002). It is therefore important to understand the patterns and predictors of smoking initiation in adolescence and young adulthood in order to inform the development of more effective prevention. Longitudinal cohort studies have become a popular source of information on changing patterns of smoking behavior through adolescence. In recent years, a substantial number of studies have modeled such longitudinal data either by using polynomial growth models (Brook, Zhang, Brook, & Finch, 2010; Simons-Morton, 2007; Windle & Windle, 2001) or in combination with a mixture component, for example, Growth Mixture Models (Brook et al.

, 2010; Colder, Flay, Segawa, & Hedeker, 2008; Orlando, Tucker, Ellickson, & Klein, 2004). Entinostat As dropout is a common problem with cohort studies, estimation using full-information maximum likelihood (FIML), which allows any participant who responds on one or more occasion to be included in the analysis, is becoming more and more popular.