Reliability of self-reported smoking history and its implications for lung cancer screening

Robert J Volk; Tito R Mendoza; Diana S Hoover; Shawn P E Nishi; Noah J Choi; Therese B Bevers

doi:10.1016/j.pmedr.2019.101037

Reliability of self-reported smoking history and its implications for lung cancer screening

Prev Med Rep. 2020 Jan 2:17:101037. doi: 10.1016/j.pmedr.2019.101037. eCollection 2020 Mar.

Authors

Robert J Volk¹, Tito R Mendoza², Diana S Hoover³, Shawn P E Nishi⁴, Noah J Choi^{1

5}, Therese B Bevers⁶

Affiliations

¹ Department of Health Services Research, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1444, Houston, TX 77230, USA.
² Department of Symptom Research, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1450, Houston, TX 77230, USA.
³ Department of Health Disparities Research, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1440, Houston, TX 77230, USA.
⁴ Department of Internal Medicine, UTMB Health Division of Pulmonary Critical Care and Sleep, 301 University Blvd, Galveston, TX 77555-0561, USA.
⁵ Rice University, 6100 Main St, Houston, TX 77005, USA.
⁶ Department of Clinical Cancer Prevention, The University of Texas MD Anderson Cancer Center, Unit 1360, Houston, TX 77230, USA.

Abstract

Clinical guidelines endorse either a 30 or 20 pack-year smoking history threshold when determining eligibility for lung cancer screening (LCS). However, self-reported smoking history is subject to recall bias that can affect patient eligibility. We examined the reliability of smokers' self-reported tobacco use and its impact on eligibility for LCS. Current or former smokers aged 55-77 years completed questionnaires requesting demographic information and smoking history. Data were collected between December 2014 and September 2015. Total pack-year smoking history was calculated for each participant based on their responses at baseline and one month later. One hundred and two participants completed the study (mean age = 63.6 years). The intraclass correlation coefficient for the pack-year estimate was 0.93. For the 30 pack-year threshold, eight (7.8%) participants were eligible at one but not both assessment periods. For the 20 pack-year threshold, twelve participants (11.8%) were eligible at one but not both assessment periods. Inconsistent reporting was higher among current compared to former smokers. Smokers' self-reported tobacco use appears highly reliable over short time periods. Nevertheless, there is some inconsistent reporting. We recommend that clinicians carefully assess smoking history, probe patients' recall of duration and quantity of smoking, and collect tobacco use information at every encounter.

Keywords: Diagnostic screening programs; Early detection of cancer; Mass screening; Patient reported outcomes; Self report; Smoking.