abstract
-
The fragility index (FI) is a measure of how many events in one randomized controlled trial (RCT) group need to be changed from a non-event to an event for a statistically significant finding to become non-significant. Until recently, the FI could only be calculated for dichotomous outcomes. In 2021, Caldwell et al published a method for calculating continuous fragility index (CFI) for trials with continuous outcomes. A 2021 systematic review of statistical fragility of hip and knee arthroplasty trials found that the median FI in arthroplasty trials was 1, indicating very high fragility. This was a well-conducted systematic review, but like other research on the topic of statistical fragility, it included only trials with dichotomous outcomes. Now that there is a method available to calculate CFIs, we aim to expand on the previous review to include continuous outcomes to provide a more complete picture of the state of statistical fragility in arthroplasty RCTs. The primary objective of this systematic review is to describe the statistical fragility of RCTs in hip and knee arthroplasty in the past decade. Our secondary objective is to determine study characteristics that are associated with fragility. This is a protocol for a descriptive systematic review of statistical fragility of arthroplasty RCTs.