Abstract
BACKGROUND: Studies of cigarette use and exposure often rely on either self-report or cotinine assay. In adolescence it is not clear how well assays and self-report correspond, or what effect estrogen exposure has on cotinine. OBJECTIVES: This study sought to identify optimal cut-points for salivary cotinine thresholds for girls with primary, secondary, and no smoke exposure, and whether menarche and hormone contraceptive use are important for interpreting salivary cotinine. METHODS: This longitudinal prospective study recruited 262 healthy adolescent girls who participated in three annual interviews across 24 months. Salivary cotinine assays and self-report of primary and secondary smoke exposure, menarcheal status, and hormone contraceptive use were collected. RESULTS: No adolescents reported primary smoke exposure without secondary exposure. Optimal cut-points for distinguishing primary smoke exposure from secondary-only and no smoke exposure were 1.05 and 3.01 ng/ml, respectively based on receiver operator curves (ROC); no reliable cut-point for secondary-only versus no smoke exposure was identified. The ideal salivary cotinine cut-point to distinguish primary smoke exposure varied by hormone contraceptive use and was 2.14 ng/ml for those using progesterone contraceptives, higher than that of girls using estrogen contraceptives and those not using hormone contraceptives. CONCLUSIONS: This study is the first to examine variance in salivary cotinine cut-points based on hormone exposure for adolescent girls, with findings indicating that hormone contraceptive use in particular may be a key consideration when identifying adolescent smoking. The use of previously recommended salivary cotinine cut-points of 3.85 ng/ml or higher may overestimate nonsmokers.