Statistical Fragility of Randomized Control Trials Comparing Surgical and Conservative Management of Distal Radius Fractures in Elderly Patients

Abstract

Randomized controlled trials (RCTs) comparing surgical fixation versus conservative management of distal radius fractures (DRFs) among the elderly often rely on P values for outcomes despite small sample sizes, attrition, and nonsignificant findings. This study uses the reverse fragility index (rFI) and reverse fragility quotient (rFQ) to assess the statistical stability of re-intervention and complication outcomes in RCTs. PubMed and Embase were searched for RCTs between January 1, 2000, and May 2024 for RCTs that evaluated surgical management versus conservative management of DRFs in the elderly population were included. RCTs were included if they contained 2 treatment arms reporting categorical dichotomous outcomes and had an equal or longer than 1-year follow-up. Non-RCT studies, RCTs with more than 2 treatment arms, RCTs without 1 year or longer follow-up, and RCTs without DRFs in elderly cohorts were excluded. rFI were calculated as the number of outcome even reversals needed to change statistical significance for nonsignificant (P ≥ .05) outcomes. rFQ was calculated by dividing the rFI by the sample size of the study. One thousand three hundred sixty-one articles were screened, with 12 studies with 1495 patients included for final analysis. The median rFI for re-intervention was 4 (range: 1-18), with a median rFQ of 0.035 (3.5%) within all RCTs evaluated. In every trial, the number of patients lost to follow-up was higher than the rFI, suggesting fragile statistical conclusions. For fracture healing complications, the median rFI was 4, with an rFQ of 0.02 (2%). Minor and major complications had rFI of 3, with rFQ of 0.02 (2%) and 0.04 (4%), respectively. Randomized controlled trials (RCTs) comparing surgical and conservative management of DRFs among elderly patients are statistically fragile, with a median of only 4 event reversals needed to alter significance. Caution is warranted when interpreting RCT results to determine fracture management among this patient population. Authors recommend larger, well-powered trials with standardized rFI analyses to better interpret outcomes in DRF RCTs among elderly patients.

Level of Evidence: Therapeutic Level III.