Predictors of attrition in a smoking cessation trial conducted in the lung cancer screening setting

Abstract

Significance: Although it is a requirement that tobacco treatment is offered to cigarette smokers undergoing low-dose computed tomographic lung cancer screening (LCS), not all smokers engage in treatment. To understand the barriers to tobacco treatment in this setting, we evaluated predictors of attrition in a smoking cessation trial among individuals undergoing LCS.

Methods: Prior to LCS, 926 participants, 50-80 years old, completed the baseline (T0) phone assessment, including demographic, clinical, tobacco, and psychological characteristics. Following LCS and receipt of the results, participants completed the pre-randomization (T1) assessment.

Results: At the T1 assessment, 735 (79%) participants were retained and 191 (21%) dropped out. In multivariable analyses, attrition was higher among those who: smoked >1 pack per day (OR = 1.44, CI 1.01, 2.06) or had undergone their first (vs. annual) LCS scan (OR = 1.70, CI 1.20, 2.42). Attrition was lower among those with: more education (associates (OR = 0.67, CI = 0.46, 0.98) or bachelor's degree (OR = 0.56, CI 0.35, 0.91) vs. high school/GED), some (vs. none/a little) worry about lung cancer (OR = 0.60, CI 0.39, 0.92), or a screening result that was benign (OR = 0.57, CI 0.39, 0.82) or probably benign (OR = 0.38, CI 0.16, 0.90) vs. negative.

Conclusions: This study illuminated several LCS-related factors that contributed to trial attrition. Increasing tobacco treatment in this setting will require targeted strategies for those who report little lung cancer worry, are undergoing their first LCS exam, and/or who have a negative LCS result. Addressing attrition and reducing barriers to tobacco treatment will increase the likelihood of cessation, thereby reducing the risk of developing lung cancer.

Keywords: Clinical trial; Low-dose computed tomographic lung cancer screening; Smoking cessation; Trial attrition.