Competing Risks: Impact on Risk Estimation and Algorithmic Fairness
Dr. Vincent Jeanselme
Postdoctoral Researcher
Department of Biomedical Informatics
Columbia University
Accurate time-to-event prediction is integral to decision-making, informing medical guidelines, hiring decisions, and resource allocation. Survival analysis — the quantitative framework used to model time-to-event data — accounts for patients who do not experience the event of interest during the study period, known as censored patients. However, many patients experience events that prevent the observation of the outcome of interest. These competing risks are often treated as censoring, a practice frequently overlooked due to a limited understanding of its consequences. Our work theoretically demonstrates why treating competing risks as censoring introduces substantial bias in survival estimates, leading to systematic overestimation of risk and, critically, amplifying disparities. First, we formalize the problem of mishandling competing risks as censoring and quantify the resulting error in survival estimates. Specifically, we develop a framework to estimate this error and demonstrate the associated implications for predictive performance. Furthermore, we examine how differing risk profiles across demographic groups lead to group-specific errors, potentially exacerbating existing disparities. In our analysis of cardiovascular management, we demonstrate that the common practice of mishandling competing risks in existing risk scores leads to a systematic overestimation of risk, with implications for downstream decision-making. The resulting bias increases overtreatment by 1.8% in the population, with men disproportionately prioritized, ultimately misallocating millions of dollars. By quantifying the error and highlighting the fairness implications of the common practice of considering competing risks as censoring, our work provides a critical insight: accounting for competing risks is not only essential for improving accuracy but also for reducing disparities in risk assessment and, consequently, downstream decisions.
