Is the hair nicotine level a more accurate biomarker of environmental tobacco smoke exposure than urine cotinine?

STUDY OBJECTIVE The aim of this study was to compare the two biomarkers of exposure to environmental tobacco smoke (ETS); urine cotinine and hair nicotine, using questionnaires as the standard. DESIGN A cross sectional study of children consecutively admitted to hospital for lower respiratory illnesses during the period of the study. SETTINGS Three regional hospitals in the larger Wellington area, New Zealand. PARTICIPANTS Children aged 3-27 months and admitted to the above hospitals during August 1997 to October 1998. A total of 322 children provided 297 hair samples and 158 urine samples. MAIN RESULTS Hair nicotine levels were better able to discriminate the groups of children according to their household's smoking habits at home (no smokers, smoke only outside the home, smoke inside the house) than urine cotinine (Kruskall-Wallis; chi(2)=142.14, and chi(2)=49.5, respectively (p<0.0001)). Furthermore, hair nicotine levels were more strongly correlated with number of smokers in the house, and the number of cigarettes smoked by parents and other members of the child's households. Hair nicotine was better related to the questionnaire variables of smoking in a multivariate regression model (r(2)=0.55) than urine cotinine (r(2)=0.31). CONCLUSIONS In this group of young children, hair nicotine was a more precise biomarker of exposure to ETS than urine cotinine levels, using questionnaire reports as the reference. Both biomarkers indicate that smoking outside the house limits ETS exposure of children but does not eliminate it.