Have you ever heard of competing risks in clinical research? It’s a relevant concept that allows to estimate the probability of an event happening in the presence of other, mutually exclusive events. Imagine patients in a clinical study at risk for multiple, potential outcomes:
- A deceased patient cannot be rehospitalized.
- Patients may have, in many case definitions, either an acute or chronic manifestation.
- After a death from myocardial infarction, one cannot die from cancer.
There are different methods in survival analysis to model competing events (e.g., sub-distribution and cause-specific hazard models). Still, only few studies account for competing events:
In a recent survey of 136 cardiovascular disease studies with composite endpoints, only 14 (10%) conducted a competing risk analysis and reported the corresponding results. Omission of competing risk analysis can reduce the precision and power, and lead to erroneous inferences.
-> DOI: https://doi.org/10.1016/j.jclinepi.2023.05.015
So, watch out😊