Zero tolerance for errors: An examination of the impact of high school suspension on young adults' smoking

The American Academy of Pediatrics condemned zero tolerance school disciplinary policies due to concerns about increasing delinquency. The theory of secondary deviance suggests youth punished for minor infractions will increase deviant behavior, but few studies have tested this hypothesis. This study examined whether youth suspended from high school are more likely to smoke than similar non-suspended youth using nationally representative Add Health data: 8638 participants never suspended at baseline (1995), 548 of whom were suspended in 1996. We compared daily smoking at ages 25-31 (2008). To minimize confounding on factors that cause both suspension and smoking, we identified 1653 non-suspended young adults using matched sampling on 20 baseline socioeconomic, educational, and health factors, and school-level disciplinary policy. We estimated relative risks using a multivariate Poisson working model in the full matched sample, and also separately for schools in the top quartile of harshest disciplinary policies and schools in the bottom 3 quartiles. Twelve years post-suspension, suspended students were more likely than non-suspended students to be daily smokers (26% versus 13%). After matching, suspended students were 36% more likely to smoke daily (IRR 1.36 (1.02, 1.81), p=0.03). Suspended students who attended schools with severe disciplinary policies were 2.3 times as likely as unsuspended students to be daily smokers (IRR 2.27 (1.22, 4.22), p=0.01), but there was no difference at schools with non-severe disciplinary policies (IRR 1.16 (0.82, 1.62), p=0.4). These results matched on pre-suspension smoking and suggest that suspension at schools with harsh policies may increase youth smoking.