Modeling oral cancer screening in the United States population

Abstract

ObjectivesTo evaluate the impact of oral cancer screening if applied to the United States (US) population or various high-risk populations in the US. MethodsWe modeled the effects of applying an oral cancer screening program to the US population assuming a similar mortality reduction as seen in the randomized Kerala trial. We combined data on the incidence of oral cancer in the Surveillance, End Results, and Epidemiology database, data on the relative risk in various high-risk groups from the Prostate, Lung, Cervical, and Ovarian screening trial, and the National Lung Screening Trial and data on the prevalence of cigarette use from the National Health Interview Survey. ResultsWhen extrapolating to the US population we predict the number needed to screen to prevent one oral cancer death (NNS) = 9,845 in all individuals aged 35 + . Screening efficiency would increase if applied to higher-risk populations. If oral cancer screening were applied to male ≥ 60 pack-year current smokers or former smokers who have quit within 15 years aged 50–79 we predict a 4.6% reduction in oral cancer mortality with an NNS = 1,485. ConclusionsTargeted screening of individuals at high risk for oral cancer has the potential to maximize the efficiency of screening and meaningfully impact oral cancer mortality. We suggest a future screening trial in high-risk individuals be considered to clarify the role of oral cancer screening in the US.