Patterns of adolescent smoking and later nicotine dependence in young adults: A 10-year prospective study

Summary

Objectives

There is considerable variability in progression from smoking initiation to established smoking. This paper addresses the extent to which different patterns of adolescent smoking, including periods of cessation, predict smoking status in young adults.

Study design

Ten-year, eight-wave prospective cohort study of a state-wide community sample in Victoria, Australia.

Methods

Participants were 1520 students from 44 secondary schools, initially aged 14 to 15 years. Adolescent smoking and quitting patterns were assessed during Waves 1–6 with self-reported frequency of use and a 7-day retrospective diary. The Fagerstrom Test for Nicotine Dependence (ND) was used to assess ND at the age of 24 years (Wave 8).

Results

The prevalence of ND in young adults was 16.9% for all adolescent smokers, with prevalence rates of 6.8% and 26.7% for adolescent non-daily and daily adolescent smokers, respectively. Maximum smoking levels, onset of daily smoking, duration of smoking, escalation time and duration of cessation during adolescence predicted later ND. Daily smokers who ceased smoking for at least two waves (≥12 months) had a level of risk similar to adolescents who had never smoked.

Conclusions

Quitting smoking as an adolescent substantially alters the risk for later ND. For adolescents who become daily smokers, quitting for 12 months should be the aim in tobacco control and clinical interventions.

Introduction

Smoking remains the most common cause of preventable deaths, with an estimated 5 million deaths attributed to tobacco-induced diseases per year worldwide.¹ Although smoking rates in adults have dropped,² smoking rates during adolescence and young adulthood remain high.³ Twenty-five percent of adolescent daily smokers have developed nicotine dependence (ND) by young adulthood.⁴ However, the natural history of smoking in adolescence is variable, with phases of cessation, reduced use and relapse common.⁵ Given this variability in the course of smoking, it is surprising that smoking patterns during adolescence have received little attention as predictors of later smoking and ND. There has been some speculation that the frequency and duration of adolescent smoking, rates of escalation of use, age of onset and quitting attempts might predict adult ND.6, 7, 8, 9, 10 However, to date, there is little empirical evidence to support this speculation. The aim of the present paper is: (1) to identify the patterns of adolescent smoking and cessation that predict young adult ND; and (2) to consider these predictors within adolescent smokers stratified by maximal adolescent smoking.