🤖 AI Summary
To address bias in causal inference arising from unmeasured confounding in observational survival studies, this paper proposes a semiparametric sensitivity analysis framework grounded in influence functions. We first derive the nonparametric influence function for time-to-event data and establish a mapping mechanism from uncensored to censored data, thereby circumventing reliance on specific parametric survival model assumptions. The method integrates competing-risks modeling with semiparametric statistical inference, requiring no specification of baseline hazard functions or confounder structures. Evaluated on real-world prostate cancer data comparing radical prostatectomy versus external-beam radiotherapy combined with androgen-deprivation therapy—and across multiple simulation scenarios—the approach demonstrates consistent estimation, effective bias correction, and substantially improved robustness and credibility of marginal causal effect estimates.
📝 Abstract
In this paper, we develop a semiparametric sensitivity analysis approach designed to address unmeasured confounding in observational studies with time-to-event outcomes. We target estimation of the marginal distributions of potential outcomes under competing exposures using influence function-based techniques. We derived the non-parametric influence function for uncensored data and mapped the uncensored data influence function to the observed data influence function. Our methodology is motivated by and applied to an observational study evaluating the effectiveness of radical prostatectomy (RP) versus external beam radiotherapy with androgen deprivation (EBRT+AD) for the treatment of prostate cancer. We also present a simulation study to evaluate the statistical properties of our methodology.