Open Access

Relating Onset of Health Conditions to Changes in Tobacco/Nicotine Use — Analyses based on the U.S. PATH Study *



Evidence is limited on how the onset of health conditions affects tobacco/nicotine use. We report analyses from adults in the U.S. PATH study relating changes in established cigarette and/or e-cigarette use between Waves 1 and 4 to onset of conditions in Waves 2 or 3.


Nine conditions had sufficient onsets for analysis. The main analyses adjusted for sex and age. Additional analyses excluded other tobacco/nicotine product users, controlled for more variables, restricted changes in one product to never users of the other, or restricted attention to changes occurring after onset of the condition.


In the main analyses, onset of each condition predicted significantly increased overall e-cigarette initiation rates, with highest odds ratios (OR) for any major respiratory condition (2.86, 95% confidence interval (CI) 2.10–3.90) and cancer (2.51, 1.52–4.14). Each onset also predicted higher cigarette smoking initiation rates, but ORs were lower than for e-cigarette initiation, and only significant for any cardiovascular condition (1.65, 1.31–2.08), high blood pressure (1.34, 1.13–1.58), and any major smoking-related condition (1.35, 1.12–1.62). For quitting smoking only six conditions showed increased ORs, significant only for high cholesterol (1.29, 1.02–1.62). The additional analyses generally confirmed these patterns. Relationships with re-initiating cigarettes or becoming dual users of cigarettes and e-cigarettes were inconsistent. Onset of conditions also tended to reduce amount smoked by smokers.


While PATH is nationally representative, its sample size does not always allow precise estimates. There is concern about accuracy of statements by some older participants claiming to be Wave 1 never smokers but Wave 4 ever smokers. Nevertheless the previously unreported association between onset of the health conditions and e-cigarette initiation (which mainly occurs in Wave 1 cigarette smokers) appears real. [Contrib. Tob. Nicotine Res. 32 (2023) 1–10]

Publication timeframe:
4 times per year
Journal Subjects:
General Interest, Life Sciences, other, Physics