Long-Term Tobacco Use Outcomes Among PLCO Participants
at trial enrollment. We propose to identify predictors of current smoking status, including 1) participation in cancer
screening tests (screening vs. control arm), 2) receipt of abnormal screening results for the chest x-ray (screening arm),
and 3) receipt of abnormal results for any of the cancer screening tests. In addition, we will evaluate demographic,
nicotine dependence factors, and perceived risk for lung cancer as predictors of current smoking status. Notification of
abnormal test results has yielded smoking abstinence rates between 7-21%, and may be an important setting to pursue
for nicotine dependence interventions. There have been 5 observational studies of smokers enrolled in lung screening
trials that have assessed smoking cessation and the impact of the screening result on cessation. Based on these studies,
it appears that undergoing lung cancer screening can have an impact on smoking, and that there is some evidence that
the screening result may play a role in cessation.
The PLCO trial provides an ideal study design in which to test the impact of cancer screening, including lung cancer screening, and also the impact of abnormal results. Participants were randomized to receive screening or not. Previous studies have only assessed this question in a serendipitous fashion, in which a single group has undergone screening, but with no comparison group who has not experienced the same event. The proposed study will provide the first test of the impact of cancer screening in which there is a control group that has not experienced this potentially motivating health event. Although the analyses will be most complete and interesting if we are able to utilize lung cancer diagnosis and all cancer diagnoses in the analyses described below, if this is not acceptable, we will need to exclude all lung cancers, as well as all cancers, from these analyses. Hypotheses 1) Among participants who were current smokers at T0, screening arm participants (who received the chest X-rays) will be more likely to have stopped smoking compared to the control arm participants (excluding those who sought screening on their own). 2) Among T0 current smokers in the screening arm, participants who received an abnormal chest X-ray result at T0 will be more likely to have stopped smoking compared to those who received normal chest-X-ray results at T0. 3) Among T0 current smokers in the screening arm, participants who received > 1 abnormal chest x-ray across all exam years will be more likely to have quit compared to those who received all normal chest x-ray results across all exam years. 4) Among T0 current smokers in the screening arm, participants who received > 1 abnormal screening result (across all exam types, across all screening years) will be more likely to have quit compared to those who received all normal results (across all exam types, across all screening years). 5) Among former smokers who have recently quit (in previous 0-23 months), those who received a normal chest x-ray at T0 are more likely to relapse than those who had an abnormal (suspicious for cancer) chest x-ray at T0. We will also look at those who quit within last 11 months. 6) Among former smokers who have recently quit (in previous 0-11 months), those who received a normal chest x-ray across all exam years are more likely to relapse than those who had an abnormal (suspicious for cancer) chest x-ray during any of the exam years.
Lung cancer screening as a teachable moment for smoking cessation.
Taylor KL, Cox LS, Zincke N, Mehta L, McGuire C, Gelmann E
Lung Cancer. 2007 Apr; Volume 56 (Issue 1): Pages 125-34 PUBMED