Annual lung cancer screening saves lives — but only if people actually return for it each year. Despite a decade of U.S. Preventive Services Task Force guidance recommending yearly low-dose CT screening for high-risk adults, real-world adherence has remained stubbornly low. This pragmatic trial from Kaiser Permanente Washington directly confronts that gap with two scalable, system-level strategies tested head-to-head and in combination.
The trial enrolled patients who had completed an initial lung cancer screening scan with normal findings between late 2022 and early 2024, randomizing them across four arms: usual care, a health communication intervention alone, a Stepped Reminders intervention alone, or both combined. Health communication targeted knowledge barriers through print and video materials. The Stepped Reminders arm worked at the clinical infrastructure level, using an electronic health record registry to prompt primary care physicians with pre-pended scan orders and simultaneously sending patients outreach to schedule their next scan. A dedicated LCS coordinator facilitated both approaches. The primary outcome — completion of a follow-up LDCT or chest CT within a 9-to-15-month window — was assessed using a modified intent-to-treat framework, with appropriate censoring for lung cancer diagnosis, death, or disenrollment.
What makes this study particularly valuable is its pragmatic, real-world design within an integrated health system, which strengthens external validity compared to controlled efficacy trials. The 2×2 factorial structure also allows assessment of whether combining interventions produces additive or synergistic benefits. Adherence to repeat cancer screening is a well-documented weak link across multiple cancer types — colorectal, breast, and cervical — and system-level nudges have consistently outperformed patient-education-only approaches in prior work. If the Stepped Reminders arm demonstrates superiority, it suggests that embedding accountability into clinical workflow, rather than relying on patient initiative alone, is the more efficient lever. The main limitation is generalizability: Kaiser Permanente's integrated model may not translate readily to fragmented fee-for-service settings. Overall, this is a methodologically rigorous contribution to implementation science with meaningful population-level implications.