Early initiation of tobacco use in adolescent girls: key sociostructural influences.

BACKGROUND: An important developmental task during adolescence is ego development. Millions of adolescents choose to initiate cigarette smoking at a young age as a result of transitional conflicts during this phase of normal developmental progression. Unfortunately, the decision to use tobacco compromises both the short- and long-term health status of smokers as well as the health of those around them.
OBJECTIVE: It was hypothesized that in early adolescence, girls choose to begin smoking cigarettes as a result of sociostructural influences, including media, peers, and family. The purpose of this study was to investigate what modifiable sociostructural variables will decrease the risk of initiating cigarette smoking before the age of 12 years among adolescent girls. The aim of the study was to develop a risk reduction model that increases the likelihood of healthy behavior choices in girls during early adolescence.
METHOD: Data collected for the New Hampshire Teen Assessment Project survey were used for this secondary data analysis. The original study included a multicommunity sample of 7,648 students from eight school districts enrolled in New Hampshire schools between January 2000 and October 2001. This current analysis focused only on the health behavior of the adolescent girls enrolled in the larger study (n = 3,775). With the use of a socioecological theoretical framework as a guide, this secondary data analysis first identified correlates of the early initiation of tobacco use using Pearson's correlations. Then, the data were explored for variables that decreased the relative risk for the early initiation of cigarette smoking among adolescent girls. The dependent variable of interest (i.e., early initiation of tobacco use) referred to girls who reported having smoked their first cigarette at or before the age of 12 years. Twenty-eight percent of the sample had initiated tobacco use before they were 12 years old. Multiple logistic regression was used to predict the final risk reduction model.
RESULTS: The first level of analysis confirmed previously reported evidence that there is a correlation between adolescent girls' initiation of smoking early (n = 1,047) and their engagement in other health risk behaviors such as daily use of alcohol (n = 859), daily use of marijuana (93%), and engaging in unprotected intercourse (15%). Next, logistic regression was used to predict a risk reduction model that demonstrated the importance of community, family, and school variables in decreasing the relative risk for the early initiation of tobacco use. Those sociostructural variables that decrease the relative risk for the initiation of tobacco use were noted in the following: (1) 71% of the girls who feel that it is important to contribute to their community (odds ratio [OR] = 1.71, 95% confidence interval [CI] = 1.31-2.23) are less likely to initiate tobacco use and (2) 54% of the girls are more likely to feel that their community is a good place to live in (OR = 1.54, 95% CI = 1.20-1.97), are more than twice as likely to have parents who think smoking is wrong (OR = 2.09, 95% CI = 1.77-2.48), are 9% more likely to have parents whom they can talk to when they have personal problems (OR = 1.09, 95% CI = 1.02-1.17), and are 38% more likely to enjoy school. DISCUSSION: The findings of this study may be an important portal for prevention intervention in the area of early adolescent tobacco use. A risk reduction model is presented based on the theory that both the environment and modeling play an important role in the development of health behavior.