Competing events influence estimated survival probability: when is Kaplan-Meier analysis appropriate?

Area of Science:

• Orthopaedic surgery
• Biostatistics
• Medical research methodology

Background:

• The Kaplan-Meier estimator is widely used in orthopaedics to estimate event probabilities over time.
• This method assumes events, like death, will eventually occur for all subjects, which is often not true for orthopaedic outcomes.
• Competing events, such as death before implant loosening or revision surgery due to infection, can bias results.

Purpose of the Study:

• To evaluate the appropriateness of the Kaplan-Meier estimator in the presence of competing events in orthopaedics.
• To demonstrate the overestimation of event probability by the Kaplan-Meier estimator when competing risks are present.
• To introduce the cumulative incidence estimator as a suitable alternative for analyzing competing risks in orthopaedic research.

Main Methods:

• Comparative analysis of statistical estimators.
• Simulation or data analysis demonstrating the impact of competing events on Kaplan-Meier estimates.
• Introduction and application of the cumulative incidence function for competing risks.

Main Results:

• The Kaplan-Meier estimator overestimates the probability of the event of interest when competing events are present.
• Competing events lead to inaccurate survival probabilities in orthopaedic outcome studies.
• The cumulative incidence estimator provides a more accurate estimation of event probability in the presence of competing risks.

Conclusions:

• The Kaplan-Meier estimator is inappropriate for orthopaedic outcomes with competing events.
• The cumulative incidence estimator should be considered for analyzing long-term orthopaedic outcomes where competing risks are likely.
• Accurate statistical methods are crucial for reliable interpretation of orthopaedic implant survival and failure rates.