Cuthbert et al. published a study in CORR titled “What Is the Effect of Using a competing-risks Estimator when Predicting Survivorship After Joint Arthroplasty: A Comparison of Approaches to Survivorship Estimation in a Large Registry“.
They recommend using competing risk approach when assessing long-term survival of total joint replacements. This is of course an important aspect to consider.
But what I found more interesting was their examples of predicted survival probabilities. I think this approach should be utilized much more often in the TJR research. Usually we just end up reporting hazard ratios for certain baseline variables and not using the full potential of our statistical methods. In Cox regression we should more often reported estimated survival probabilities which better describe the effect of baseline variables. And more common utilization of nonlinear methods such as restricted cubic splines would also improve our TJR research.