A centralized navigation model with AI-assisted EHR alerts increased the system-wide lung cancer screening rate to 42.8% in 2025, outperforming the projected US national average of 19.5%.
Observational
Yes
Does a centralized navigator-led and AI-assisted platform improve lung cancer screening rates and early detection in eligible high-risk patients?
A centralized, AI-assisted navigation model effectively doubled the national benchmark for lung cancer screening uptake and improved early-stage detection.
Absolute Event Rate: 42.8% vs 19.5%
109 Background: Despite the known mortality benefits of low-dose computed tomography (LDCT) for lung cancer screening (LCS), national uptake remains stagnant (~15-20%). Significant barriers include primary care burden and geographic disparities in rural areas. This study evaluates the impact of a centralized navigation model combined with AI-assisted electronic health record (EHR) alerts on LCS rates and stage distribution within a large integrated health system. Methods: We conducted a retrospective analysis of LCS performance across a 16-hospital system (OSF HealthCare) from 2019 to 2025. The intervention included: 1) centralization of LCS navigators to manage registries and scheduling, and 2) implementation of AI-assisted EHR alerts to identify eligible high-risk patients. Outcome measures included absolute LDCT volume, system-wide screening rates, and AJCC staging at diagnosis. Benchmarks were derived from American Lung Association (ALA) national and state (Illinois) data. Results: The system-wide LCS screening rate increased from 18.2% in 2020 to 42.8% in 2025, significantly outperforming the projected 2025 US national average (19.5%) and the Illinois state average (20.8%). Absolute screening volume nearly doubled, rising from 2,257 scans in 2019 to 4,108 in 2025. Notably, the program demonstrated high resilience during the 2020 COVID-19 pandemic, maintaining 98.5% of prior-year volume compared to a 5% national decline. Early-stage detection (Stage I) showed the greatest improvement in rural facilities, with some centers seeing a +21% increase in Stage I diagnoses over the study period. Conclusions: A centralized, technology-enabled navigation model effectively doubles the national benchmark for lung cancer screening uptake. By removing administrative burdens from primary care and utilizing AI-assisted EHR alert system to close the "eligibility gap", this model provides a scalable blueprint for improving early cancer detection and addressing rural health inequities. OSF HealthCare lung cancer screening performance vs. national benchmarks (2020–2025). Year OSF Annual LDCT Volume OSF Screening Rate (%) National Screening Rate (%)* Performance Gap (Percentage Points) 2020 2,223 18.2% 14.5% +3.7% 2021 2,647 21.4% 14.8% +6.6% 2022 3,280 27.8% 15.5% +12.3% 2023 3,406 33.6% 16.0% +17.6% 2024 3,522 38.2% 18.2% +20.0% 2025 4,108 42.8% 19.5% +23.3% *National benchmarks based on American Lung Association "State of Lung Cancer" annual reports and CDC/ACS prevalence estimates.
Zhang et al. (Wed,) conducted a observational in Lung cancer screening. Centralized navigation model combined with AI-assisted EHR alerts vs. US national average was evaluated on System-wide lung cancer screening rate. A centralized navigation model with AI-assisted EHR alerts increased the system-wide lung cancer screening rate to 42.8% in 2025, outperforming the projected US national average of 19.5%.