ASPO Abstracts
Smoking Status at Time of Lung Cancer Diagnosis at an Academic Hospital
Category: Cancer Health Disparities
Conference Year: 2020
Abstract Body:
Purpose of the Study: The goal of this study was to examine changes in the prevalence of current smoking
among incident lung cancer (LC) patients.
Methods: We conducted a descriptive study using deidentified data collected from Thomas Jefferson University
Hospital cancer registry from 2005-2016. We described characteristics of patients and smoking patterns in
our cohort by gender and by the three largest racial groups (whites, blacks, Asians). We divided our time
periods into 2005-08, 2009-12, and 2013-16 and calculated the percent of current smokers at the time of
diagnosis by race and gender group for each time period. We used Chi-square tests and Fisher’s exact tests
to determine differences in current smoking prevalence at each time period. Logistic regression models were
used to examine multivariable associations and interactions.
Results: Our sample included 4251 LC patients.  The median age of our population was 68 years (range: 20-
96) and 53% were female.  Our demographics were as follows: 73.7% self-identified as white, 20.7% black,
4.3% Asian.  At the time of diagnosis, 1410 patients (33.2% of entire cohort) were current cigarette smokers
and 17 (0.4%) were cigar or pipe smokers. Interestingly, Asian LC patients were most likely to be never-
smokers in this cohort (69.9% Asian women, 20.6% Asian men). Black men had the highest prevalence of
current smoking (46.1%). The prevalence of current smoking decreased significantly over time in white women:
40% from 2005-2008, 33.6% from 2009-2012, and 28.5% from 2013-2016 (p<0.005). Multivariable analyses
demonstrated a significant interaction indicating that current smoking decreased over time in white women
(ORinteraction=0.95, p=0.031). However, in Asians, blacks, and white men, there was no statistically
significant interaction with time.
Conclusions: Despite a declining number of tobacco users in the US, the prevalence of current tobacco use in
our LC patient cohort remained unchanged from 2005-2016 in all groups except white females. Increased
efforts to augment tobacco cessation are warranted. Investigations to determine whether different
interventions might be more effective in different race and gender groups should be considered.
Keywords: smoking, lung cancer, disparities