TY - JOUR
T1 - Reconciling human smoking behavior and machine smoking patterns
T2 - Implications for understanding smoking behavior and the impact on laboratory studies
AU - Marian, Catalin
AU - O'Connor, Richard J.
AU - Djordjevic, Mirjana V.
AU - Rees, Vaughan W.
AU - Hatsukami, Dorothy K.
AU - Shields, Peter G.
PY - 2009/12
Y1 - 2009/12
N2 - Background: Recent Food and Drug Administration legislation enables the mandating of product performance standards for cigarette smoke and the evaluation of manufacturers' health claims for modified tobacco products. Laboratory studies used for these evaluations and also for understanding tobacco smoke toxicology use machines to generate smoke. The goal of this review is to critically evaluate methods to assess human smoking behavior and replicate this in the laboratory. Methods: Smoking behavior and smoking machine studies were identified using PubMed and publicly available databases for internal tobacco company documents. Results: The smoking machine was developed to generate smoke to allow for comparing cigarette tar and nicotine yields. The intent was to infer relative human disease risk, but this concept was flawed because humans tailor their smoking to the product, and chemical yields and toxicologic effects change with different smoking profiles. Although smoking machines also allow for mechanistic assessments of smoking-related diseases, the interpretations also are limited. However, available methods to assess how humans puff could be used to provide better laboratory assessments, but these need to be validated. Separately, the contribution of smoke mouth-holding and inhalation to dose need to be assessed, because these parts of smoking are not captured by the smoking machine. Better comparisons of cigarettes might be done by tailoring human puff profiles to the product based on human studies and comparing results across regimens. Conclusions: There are major research gaps that limit the use of smoking machine studies for informing tobacco control regulation and mechanistic studies.
AB - Background: Recent Food and Drug Administration legislation enables the mandating of product performance standards for cigarette smoke and the evaluation of manufacturers' health claims for modified tobacco products. Laboratory studies used for these evaluations and also for understanding tobacco smoke toxicology use machines to generate smoke. The goal of this review is to critically evaluate methods to assess human smoking behavior and replicate this in the laboratory. Methods: Smoking behavior and smoking machine studies were identified using PubMed and publicly available databases for internal tobacco company documents. Results: The smoking machine was developed to generate smoke to allow for comparing cigarette tar and nicotine yields. The intent was to infer relative human disease risk, but this concept was flawed because humans tailor their smoking to the product, and chemical yields and toxicologic effects change with different smoking profiles. Although smoking machines also allow for mechanistic assessments of smoking-related diseases, the interpretations also are limited. However, available methods to assess how humans puff could be used to provide better laboratory assessments, but these need to be validated. Separately, the contribution of smoke mouth-holding and inhalation to dose need to be assessed, because these parts of smoking are not captured by the smoking machine. Better comparisons of cigarettes might be done by tailoring human puff profiles to the product based on human studies and comparing results across regimens. Conclusions: There are major research gaps that limit the use of smoking machine studies for informing tobacco control regulation and mechanistic studies.
UR - http://www.scopus.com/inward/record.url?scp=73449127381&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=73449127381&partnerID=8YFLogxK
U2 - 10.1158/1055-9965.EPI-09-1014
DO - 10.1158/1055-9965.EPI-09-1014
M3 - Review article
C2 - 19959678
AN - SCOPUS:73449127381
SN - 1055-9965
VL - 18
SP - 3305
EP - 3320
JO - Cancer Epidemiology Biomarkers and Prevention
JF - Cancer Epidemiology Biomarkers and Prevention
IS - 12
ER -