Journal of the Formosan Medical Association (Feb 2024)
Comparison of secondhand smoking exposure between self-report and creatinine-corrected urine cotinine: Result from Korean NHANES 2009–2018
Abstract
Background/purpose: Second-hand smoking (SHS) is usually examined by self-report (SR). However, there is a possibility that SR may not accurately measure SHS exposure. This study aimed to identify yearly trends and gender differences in SHS by SR and cotinine. Methods: We used data from the 2009–2018 Korea National Health and Nutrition Examination Survey (KNHANES) and included adults aged 19 years and older. We analyzed data of 47,907 respondents on SHS exposures in the past week and of 23,572 respondents who had both urine cotinine and creatinine measurements. SHS exposure was defined as those who answered ‘yes’ to whether or not they were exposed to smoking by SR. We performed descriptive analysis, Average annual percentage change (AAPC), and multiple regression analysis. All analyses were weighted reflecting the multi-stratified cluster sampling. Results: Exposure to SHS indoors at the work place (WSHS) (2009: 44.8%, 2018: 12.3%), indoors at home (HSHS) (2009: 14.1%, 2018: 3.9%), and indoors at public places (PSHS) (2013: 55.6%, 2018: 15.4%) decreased steadily over the years. WSHS and PSHS exposure was higher for males but HSHS was higher for females. However, the concentration of cotinine-to-creatinine ratio (Co/Cr) was higher among females regardless of SHS exposure status and environment. Multiple regression analysis showed that among males, the association between Co/Cr education level and WSHS was the strongest, and among females, the association was the strongest with HSHS. In addition, home exposure to SHS intensified the differences in Co/Cr levels between males and females. Conclusion: Our study confirmed that cotinine-induced SHS showed different results by year and gender compared to SR. In particular, Co/Cr level in females was higher than in males regardless of exposure by SR, suggesting that SR did not correctly evaluate SHS. To reflect reality more accurately, biomarkers should be monitored along with SR.
