Re: Boeckstyns MEH, Merser S, Cool P. Reporting implant survival. J Hand Surg Eur. 2019, 44: 761–3

Abstract

I was interested to read the recommendations of Boeckstyns et al. in their recent article. Unfortunately, they have made a number of important factual errors that need correcting. They have fallen into the trap of misinterpreting what competing risk models do, or do not do as the case may be. In the article ‘Are competing risks models appropriate to describe implant failure?’ (Sayers et al., 2018), we describe the differences between Kaplan–Meier curves and the cumulative incidence function as estimated by competing risk models. In the penultimate paragraph we state: