Sample size calculations in orthopaedics randomised controlled trials: revisiting research practices

Area of Science:

• Orthopaedic Surgery
• Clinical Trials Methodology
• Biostatistics

Background:

• Sample size calculations are fundamental for ensuring adequate statistical power in randomized controlled trials (RCTs).
• Previous assessments indicate suboptimal reporting of sample size calculations in orthopaedic research.
• Type II errors, failing to detect a true treatment effect, are a concern in underpowered studies.

Purpose of the Study:

• To assess the frequency of sample size calculation reporting in recent orthopaedic RCTs.
• To determine the proportion of studies with non-significant findings that were underpowered (at risk of type II error).

Main Methods:

• A systematic search identified RCTs published in 2012 in top-ranked orthopaedic journals.
• Post hoc power analyses were conducted to evaluate study power for small, medium, and large effect sizes.
• Standard criteria for sufficient power (80%), alpha (0.05), and two-tailed tests were applied.

Main Results:

• Of 120 identified RCTs, only 73 (60.8%) reported appropriate sample size calculations.
• Among studies with non-significant primary outcomes, 93.15% were at risk of type II error for small effects.
• Only 5.48% of negative studies were at risk of type II error for medium effect sizes.

Conclusions:

• While reporting of sample size calculations has improved in orthopaedic surgery over the past decade, significant deficiencies persist.
• A high proportion of orthopaedic RCTs with negative results are underpowered, risking type II errors.
• Ensuring adherence to methodological guidelines, including robust sample size calculations, is a shared responsibility among researchers, reviewers, and editors.