Primary Submission Category: Applications in Health and Biology
The Clone-Censor-Weight Method Applied to Continued Colorectal Cancer Screening Participation: A Target Trial Emulation
Authors: Buket Öztürk Esen, Lars Pedersen,
Presenting Author: Buket Öztürk Esen*
Background: The effectiveness of the fecal immunochemical test (FIT) in reducing colorectal cancer (CRC) mortality is well-known. However, studies evaluating the effect of attending a second FIT screening remain limited.
Objective: By utilizing the target trial framework and the clone-censor-weight (CCW) method, we aim to evaluate the effect of attending a second FIT screening while mitigating immortal time bias and selection bias.
Method: We applied the CCW method in three steps: 1) Cloning: We created two copies of each eligible individual (N=622,265) and assigned each copy to one of two screening strategies: participating within three years of an initial negative result vs. not participating. 2) Censoring: At monthly intervals, we assessed adherence. If a copy deviated from the assigned strategy, it was censored. 3) Weighting: To adjust for selection bias introduced by censoring, we estimated stabilized inverse probability weights. This created a pseudo-population in which censoring was independent of baseline factors.
Result: At 6 years, the risk of CRC mortality was 0.07% (95%CI: 0.06%; 0.08%) for continued participation vs. 0.17% (95%CI: 0.14%; 0.21%) for non-participation, resulting in a relative risk of 0.39 (95%CI: 0.30; 0.52) and a risk difference of -0.10% (95%CI: -0.14%; -0.07%).
Conclusion: CCW is a useful method for evaluating the effectiveness of attending a second FIT screening; however, selection bias due to unmeasured factors cannot be ruled out.