The statistical stability of clavicle fracture management: A systematic review of randomized controlled trials with fragility analysis

Abstract

Background

Randomized controlled trials (RCTs) represent the highest level of evidence in guiding the management of clavicle fractures. They are conducted to ensure that evidence drives the basis of treatment algorithms. Relying solely on P values to assess orthopedic RCTs may be deceptive and challenge the validity of the studies.

Objective

To evaluate the degree of statistical fragility in the clavicle fracture literature using the fragility index (FI) and fragility quotient (FQ).

Methods

A systematic search strategy using the Medline, PubMed, and Embase databases was used to find all dichotomous data for randomized controlled trials (RCTs) in clavicle fracture research from 2000 to 2024. The FI of each outcome was calculated by reversing a single outcome event until significance was reversed. The FQ was calculated by dividing each fragility index by the study sample size. The interquartile range (IQR) was also calculated for the FI and FQ.

Results

Of the 3646 articles screened, 81 met the search criteria, with 30 RCTs evaluating clavicle fractures included for analysis. There were 250 total outcomes, where 62 significant and 188 nonsignificant outcomes were identified. The overall FI and FQ were 4 (IQR 3–5) and 0.045 (IQR 0.024–0.080). Statistically significant and nonsignificant outcomes had an FI of 3.5 (IQR 2–7) and 4 (IQR 3–5), respectively. Regarding loss to follow-up (LTF), 63.3% (19) reported LTF greater or equal to the overall FI of 4.

Conclusion

When scrutinizing management algorithms relying on statistical analysis, we recommend including the FI and FQ alongside the P value.