Abstract
BACKGROUND: Low-dose computed tomography (LDCT) significantly increases early detection rates of lung cancer and reduces lung cancer-related mortality by 20%. However, many significant screening barriers remain. This study conduct an initial feasibility and cost-effectiveness analysis of a community-based program that used a mobile low-dose computed tomography (LDCT) scan unit and discuss the operational challenges faced during its implementation. METHODS: This study was conducted in rural areas in Fujian Province, China from July 2022 to August 2022. Individuals aged 40 years and above who had not previously undergone LDCT and who were socioeconomically marginalized were included. Participants received a LDCT program from a multidisciplinary research team. Physicians analyzed the images with the assistance of artificial intelligence "InferRead CT Lung Research" and completed structured reports on their impressions. The primary evaluation indicators for mobile LDCT screening effectiveness were the lung cancer detection rate and diagnosis rate, while the main evaluation indicators for cost-effective analysis were the cost-effective ratio and early detection cost index. RESULTS: A total of 10,159 individuals participated in this study. The detection rates of suspected lung cancer cases and confirmed cases were 1.06% (n = 108) and 0.7% (n = 71), respectively. The cost of lung cancer screening (LCS) was ¥1,203,504 (US$188,847.71), the average cost per screening was ¥118.47 (US$18.65), and the cost effective ratios for the detection of suspected lung cancer and confirmed lung cancer were ¥11,143.56 (US$1,753.29) and ¥16,950.76 (US$2,669.94), respectively. The early detection cost indices for suspected lung cancer were 0.09 and 0.13 for confirmed lung cancer, respectively. CONCLUSION: This LDCT with artificial intelligence model for LCS holds economic promise for reducing health disparities in underserved areas and promote larger populations in similar low-income country.