Assessment of potentially reduced risk tobacco products (PRRPs) can be facilitated with availability of a single baseline population estimate for biomarkers of exposure (BOEs) for select constituents in cigarette smoke. The purpose of this analysis is to establish such a population estimate for BOEs to carbon monoxide (carboxyhemoglobin - COHb), nicotine (Nicotine Equivalents - NEQ), and nicotine-derived nitrosamine ketone (NNK; total urinary 4-[methylnitrosamino]-1-[3-pyridyl]-1-butanol [NNAL] - NNAL) based on pooled weighted average from published literature. 

Four databases - PubMed^®, ScienceDirect^®, TOXNET^®, and Google Scholar^™ – were systematically searched for published literature between 2008-2020 based on a protocol for literature assessment to identify and select studies. A total of 217 scientific articles were identified and reviewed for potential inclusion, 87 studies met the pre-set criteria (reported clinical studies and/or observational studies with reportable original values), and 53 studies met the inclusion criteria for each BOE. We determined the pooled weighted average based on Epanechnikov kernel density curves. The NNAL levels were normalized to a unit (pmol/mg creatinine) commonly used in research studies and reported as geometric means.

Smokers had significantly (p<0.05) higher levels of NNAL (1.112 pmol/mg creatinine; 95% confidence interval [CI] 0.161, 2.047), COHb (5.21 % saturation; 95% CI 3.91, 6.00), and NEQ (13.81 mg/24 h; 95% CI 9.08, 21.19)compared to non-smokers (0.008 pmol/mg creatinine, 95% CI 0.001, 0.017 for NNAL; 1.05% saturation, 95% CI 0.04, 2.07 for COHb; and 0.058 mg/24 h, 95% CI 0.016, 0.112 for NEQ) This research addresses the existing gap in lack of population level estimates for BOEs by establishing population level estimates for COHb, NEQ, and NNAL that can be used to determine changes in exposure for smokers switching to PRRPs.