Longitudinal Associations of Cigarette Smoking, Smoking Intensity, and Smoking TrajectoriesWith Emergency Room Utilization Among U.S. Adults

@misc{guo2026longitudinal,
  title = {{Longitudinal Associations of Cigarette Smoking, Smoking Intensity, and Smoking TrajectoriesWith Emergency Room Utilization Among U.S. Adults}},
  author = {W. Guo and X. Yan and C. Tatsuoka},
  howpublished = {medRxiv},
  year = {2026},
  month = oct,
  doi = {10.64898/2026.10.31.26366988},
  url = {https://www.medrxiv.org/content/10.64898/2026.10.31.26366988v1},
  note = {CC BY},
}

TY  - UNPB
TI  - Longitudinal Associations of Cigarette Smoking, Smoking Intensity, and Smoking TrajectoriesWith Emergency Room Utilization Among U.S. Adults
AU  - W. Guo
AU  - X. Yan
AU  - C. Tatsuoka
PY  - 2026
DA  - 2026/10/31
PB  - medRxiv
DO  - 10.64898/2026.10.31.26366988
UR  - https://www.medrxiv.org/content/10.64898/2026.10.31.26366988v1
AB  - Background: Tobacco smoking is a leading cause of morbidity and mortality in the United States, yet its longitudinal association with emergency room (ER) utilization remains understudied. We examined whether cigarette smoking, smoking intensity, and smoking trajectories were associated with subsequent ER use. Methods: We analyzed adult respondents from Waves 1-6 of the Population Assessment of Tobacco and Health Study. Wave 1-3 smoking status, intensity, and cigarettes/day were examined in relation to next-wave ER outcomes. Logistic generalized estimating equation models were used for any ER visit and frequent ER use (>=2 visits), and negative binomial models were used for ER visit count. Trajectory analyses examined Wave 1-5 smoking patterns and Wave 6 ER outcomes, with Wave 1-4 to Wave 5 as sensitivity analysis. Marginal structural models (with IPW) and propensity score matching were used to strengthen causal interpretation. Models adjusted for demographics, insurance, self-rated health, and baseline ER use. Survey weights with balanced repeated replication and Fay's adjustment were used. Results: In adjusted GEE models, current smoking was associated with higher subsequent ER utilization, including any ER visit (OR 1.128, 95% CI [1.068-1.191]), frequent ER use (OR 1.275, 95% CI [1.185-1.372]), and ER visit count (RR 1.264, 95% CI [1.160-1.377]; all p<0.001). Smoking intensity showed a dose-response pattern: smoking <20 cigarettes/day was associated with any ER visit (OR 1.130, 95% CI [1.060-1.204]), frequent ER use (OR 1.231, 95% CI [1.124-1.347]), and ER visit count (RR 1.262, 95% CI [1.110-1.435]), while smoking >=20 cigarettes/day showed stronger associations with any ER visit (OR 1.231, 95% CI [1.127-1.345]), frequent ER use (OR 1.447, 95% CI [1.293-1.618]), and ER visit count (RR 1.368, 95% CI [1.245-1.504]; all p<0.001). Each additional cigarette/day was also associated with higher ER utilization (OR/RR range, 1.007-1.011; all p<0.001). Findings were consistent in marginal structural models and propensity score matched analyses. In trajectory analyses, new/intermittent and persistent smokers had significantly higher ER utilization, while former/intermittent smokers and sustained quitters did not differ from non-smokers, suggesting that this harmful association was eliminated by smoking cessation. Conclusions: Smoking, heavier intensity, and persistent or emerging smoking patterns were associated with higher subsequent ER utilization, while smoking cessation reduces ER risk to the same level as non-smokers. This supports smoking prevention and cessation as strategies to reduce acute healthcare burden.
ER  -