The effects of smoking on adolescent trauma patients: a propensity-score-matched analysis

Abstract

Purpose Cigarettes have been demonstrated to be toxic to the pulmonary connective tissue by impairing the lung's ability to clear debris, resulting in infection and acute respiratory distress syndrome (ARDS). Approximately 8% of adolescents are smokers. We hypothesized that adolescent trauma patients who smoke have a higher rate of ARDS and pneumonia when compared to non-smokers. Methods The Trauma Quality Improvement Program (2014-2016) was queried for adolescent trauma patients aged 13-17 years. Adolescent smokers were 1:2 propensity-score-matched to non-smokers based on age, comorbidities, and injury type. Data were analyzed using chi square for categorical data and Mann-Whitney U test for continuous data. Results From 32,610 adolescent patients, 997 (3.1%) were smokers. After matching, 459 smokers were compared to 918 non-smokers. There were no differences in matched characteristics. Compared to non-smokers, smokers had an increased rate of pneumonia (3.1% vs. 1.1%, p = 0.01) but not ARDS (0.2% vs. 0%, p = 0.16). Compared to the non-smoking group, the smokers had a longer median total hospital length-of-stay (3 vs. 2 days, p = 0.01) and no difference in overall mortality (1.5% vs. 2.4%, p = 0.29). Conclusion Smoking is associated with an increased rate of pneumonia in adolescent trauma patients. Future research should target smoking cessation and/or interventions to mitigate the deleterious effects of smoking in this population.

Facility Size and Risk Factors for Mortality in Pediatric Trauma

“…However, the heterogeneity of complications captured by the TQP database suggests that further work is needed to substantiate and expand on this finding. 11 Our results were also subject to potential confounding by rural vs urban facility locations, which could not be determined from the data. Because the database included bed size as a categorical variable, we could not identify the exact cutoff at which bed size might be most predictive of outcomes.…”

“…3 The primary outcome was in-hospital mortality, and a secondary outcome was occurrence of any hospital complications as defined by the TQP database. 11 Our primary independent variable was facility size, dichotomized based on prior research as large (>600 beds) vs small/medium (≤600 beds). 5,11 In a sensitivity analysis, we also considered categorizations of >400 vs ≤400 beds and >200 vs ≤200 beds, representing the 3 available bed size cutoffs.…”

“…11 Our primary independent variable was facility size, dichotomized based on prior research as large (>600 beds) vs small/medium (≤600 beds). 5,11 In a sensitivity analysis, we also considered categorizations of >400 vs ≤400 beds and >200 vs ≤200 beds, representing the 3 available bed size cutoffs. 3 Patient-level characteristics included demographics (age, sex, race/ethnicity), primary injury mechanism (blunt, penetrating, other), injury severity (based on the Injury Severity Score (ISS)), transfer status (directly from accident scene vs or interfacility transfer), and insurance coverage (private, public, other coverage, or self-pay).…”

“…The American Surgeon™ 89 (11) mitigate increased mortality associated with known patient-level risk factors in pediatric trauma. Whereas pediatric trauma center verification level is well known to be correlated with outcomes, 3 and the role of facility size is less certain.…”