Integrated data analysis of determinants of low-dose CT screening for lung cancer

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ASTRACT The leading cause of cancer death in the U.S. is lung cancer. Results from the National Lung Screening Trial in 2011 revealed the potential to detect lung cancer at earlier stages using annual low-dose CT (LDCT) screening, thereby decreasing mortality by up to 20% in high-risk patients. Despite support from professional organizations and insurers, early evidence suggests that LDCT screening utilization is unacceptably low, with <4% of eligible persons being screened in the past 12 months. Little is known, however, about whether LDCT screening utilization varies across the U.S., and what personal- and broader macro-level factors enable or impede individuals from obtaining LDCT screening. This study will employ pooled data from the 2015 National Health Interview Survey (NHIS) and 2017 Behavioral Risk Factor Surveillance System (BRFSS), in addition to a variety of linked area-level datasets (e.g., American Community Survey), to estimate state- and county-level estimates of LDCT screening eligibility and utilization, and its associated individual- and area-level determinants. First, using an integrative data analysis framework, we will create census tract-, county- and state-level estimates of LDCT screening uptake in South Carolina using an innovative multilevel post-stratification modelling approach. The resulting rates will then be mapped, and we will perform geospatial and statistical comparisons between areas in the highest vs. lowest quartile of LDCT screening uptake. Finally, we will identify individual- and area-level determinants of LDCT screening using the aforementioned multilevel modelling approach. The findings from the proposed study will identify geographic disparities in LDCT screening uptake and related modifiable factors, which will help detect geographic areas and populations that would benefit from targeted outreach efforts and policy change.