The fragility of statistical findings in the intertrochanteric fracture fixation literature: a systematic review of randomized controlled trials

Abstract

Introduction

Intertrochanteric fractures are common and can lead to significant disability and morality, particularly in the elderly. Utilizing the fragility index (FI), reverse fragility index (rFI), and fragility quotient (FQ), this study evaluates the statistical fragility of outcomes reported in intertrochanteric fracture fixation randomized controlled trials (RCTs).

Materials and methods

Data sources: Pubmed, Embase, and MEDLINE were queried for RCTs published between 2010-present. Study selection: RCTs reporting 1:1 categorical, dichotomous outcomes were included. Articles were excluded if they were not RCTs, had over two treatment groups, included in vitro/animal/cadaveric data, and did not feature intertrochanteric fractures. Data extraction: Publication and individual outcome data were collected by three independent reviewers. Data synthesis: FI and rFI were calculated as the number of event reversals required to reverse the statistical significance for each outcome. The FQ was calculated by dividing FI by the study sample size. Subgroup analysis was performed based on outcome types.

Results

Two hundred thirty-two articles were screened, and 52 articles with a total of 370 outcomes were included for analysis. The median FI was 5 (IQR 4–6) with a FQ of 0.05 (IQR 0.032–0.078). 57/370 outcomes were statistically significant with a median FI of 3 (IQR 1–8). 313 outcomes were statistically nonsignificant with a median rFI of 5 (IQR 4–6). The number of patients lost to follow-up was greater than or equal to the FI in 127/370 outcomes (34.32%). Outcomes relating to malunion/nonunion were the most fragile, encompassing 11 outcomes with a median FI of 3 (IQR 2.5–5).

Conclusion

Outcomes in intertrochanteric fracture fixation RCTs are fragile as reversal of a few outcomes or maintaining follow-up may alter the significance of study findings. Thus, P-values are recommended to be routinely reported with FI and FQ metrics in order to provide a comprehensive understanding of the statistical robustness of outcomes in orthopedic trauma literature.

Level of evidence

I.