Citation

Jun HJ, Rich-Edwards JW, Boynton-Jarrett R, Austin SB, Frazier AL, Wright RJ. J. Adolesc. Health 2008; 43(1): 55-63.

Affiliation

Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts 02115, USA. nhhjj@channing.harvard.edu

Copyright

(Copyright © 2008, Elsevier Publishing)

DOI

10.1016/j.jadohealth.2007.12.003

PMID

18565438

Abstract

PURPOSE: We examined the association between severity, accumulation, and timing of abuse in childhood and adolescence and smoking status among young women. METHODS: Retrospective self-reported childhood abuse was ascertained with the modified Conflict Tactics Scale from 91,286 Nurses Health Study II participants in 2001 (68,505 returned; 75.0% response rate). Childhood abuse was categorized by severity (mile/moderate/severe), type (physical/sexual), and timing (childhood/adolescence). Smoking status during adolescence was reported at baseline (1989). Logistic regression was used to predict smoking initiation by age 14 and smoking status between the ages of 15 and 19. RESULTS: A graded association between severity of abuse and early initiation of smoking (by age 14 years) was demonstrated (odds ratio [OR] = 1.9, 95% confidence interval [CI] = 1.7-2.1 for severe physical violence). Young women with both physical and sexual abuse were two times more likely to start smoking by age 14 than were those reporting no abuse (OR = 2.0, 95% CI = 1.8-2.3). Although abuse during childhood increased risk for adolescent smoking (OR = 1.7, 95% CI = 1.8-2.1) for those with childhood physical and sexual abuse, inclusion of adolescent physical and sexual abuse (OR = 2.2, 95% CI 2.1-2.4) diminished the impact of childhood abuse (OR = 1.1, 95% CI 1.1-1.2). The degree of familial emotional support was protective against smoking, and reduced the impact of abuse by 40% among those with high emotional support versus those without (p<.0001). CONCLUSIONS: A strong and graded association was observed between both severity and accumulation of abuse and the risk of early initiation of smoking among girls. Smoking status during late adolescence was more strongly associated with adolescent abuse than childhood abuse. Early smoking onset is associated with both heightened risk for disease in adolescence but also increased morbidity and mortality in adulthood. Identifying and intervening in potentially modifiable risk factors for smoking onset in young women, such as early-life physical and sexual abuse, and building familial strengths, such as emotional support, may have significant public health implications.

Language: en