A longitudinal study of the reciprocal relationship between ever smoking and urgency in early adolescence

Abstract

Background: Among early adolescents in the United States (U.S.), the prevalence of cigarette smoking is at its lowest level in recent decades. Nonetheless, given the risks of smoking in early development, it remains critically important to study both risk factors for smoking and risks from smoking. This longitudinal study with U.S. early adolescents examines smoking initiation and tests a model of reciprocal prediction between ever smoking and the personality trait of urgency (i.e., mood-based impulsivity), a trait that increases risk for multiple forms of dysfunction.

Methods: Participants (n=1906; 90% 10-11 years old, 50% female, 39% racial minorities at baseline) completed questionnaires 1-2 times per year starting in 5th grade and ending in 9th grade. Structural equation modeling allowed tests of bidirectional relationships between ever smoking and urgency controlling for pubertal status and negative affect at each wave.

Results: Incidence of ever smoking increased from 5% to 27% over time, with current smoking around 5% at the last wave. Urgency at each wave predicted ever smoking at the next wave above and beyond covariates and prior smoking (all p<0.01). Likewise, with one exception, ever smoking predicted an increase in urgency at the subsequent wave above and beyond covariates and prior urgency (all p<0.05).

Conclusion: Results show that risk for smoking increases with higher levels of urgency and urgency increases secondary to engagement in smoking. Future work should therefore explore urgency as a point of prevention for smoking and smoking cessation as a means to mitigate mood-based impulsivity.

Keywords: Adolescence; Impulsivity; Longitudinal design; Smoking; Urgency.

Figure 1. Reciprocal model between smoking and urgency at waves 4 to 8 Note

Note. Horizontal arrows connecting urgency at each wave with urgency at the subsequent waves and ever smoking at each wave with ever smoking at the subsequent waves represent Step 1 of the model, the autoregressive pathways; text above these arrows represents the estimate of the autoregressive effects. Diagonal arrows connecting urgency at each wave with ever smoking at the subsequent wave represent the pathways added at Step 2 of the model; text above these arrows represents the estimate of the pathway from urgency to ever smoking. Diagonal arrows connecting ever smoking at each wave with urgency at the subsequent wave represent the pathways added at Step 3 of the model; text above these arrows represent the estimate of the pathway from ever smoking to urgency. Solid lines indicate significant effects while dashed lines indicate non-significant effect. * p < .05; ** p < .01.