Misclassification of self-reported smoking in adult survivors of childhood cancer

I. Chan Huang, James L. Klosky, Chelsea M. Young, Sharon E. Murphy, Kevin K. Krull, Deo Kumar Srivastava, Melissa M. Hudson, Leslie L. Robison

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


We investigated misclassification rates, sensitivity, and specificity of self-reported cigarette smoking through serum cotinine concentration (liquid chromatography tandem mass spectrometry) among 287 adult survivors of childhood cancer. Overall, 2.5–6.7% and 19.7–36.9% of the self-reported never and past smokers had cotinine levels indicative of active smoking. Sensitivity and specificity of self-reported smoking were 57.5–67.1% and 96.6–99.2%. Misclassification was associated with younger age (OR = 3.2; 95% CI = 1.4–7.4), male (OR = 2.1; 95% CI = 1.1–4.0), and past (OR = 2.7; 95% CI = 1.3–5.8) or current (OR = 2.6; 95% CI = 1.0–6.6) marijuana use. After adjusting for tobacco-related variables, current marijuana use remained a significant risk for misclassification. Clinicians/researchers should consider bio-verification to measure smoking status among survivors.

Original languageEnglish (US)
Article numbere27240
JournalPediatric Blood and Cancer
Issue number9
StatePublished - Sep 2018

Bibliographical note

Publisher Copyright:
© 2018 Wiley Periodicals, Inc.


  • childhood cancer survivors, cigarette smoking, cotinine
  • misclassification, tobacco use


Dive into the research topics of 'Misclassification of self-reported smoking in adult survivors of childhood cancer'. Together they form a unique fingerprint.

Cite this