The interplay between PROM score distributions and treatment effect detection likelihood in randomized controlled...

Area of Science:

• Orthopedics
• Rheumatology
• Clinical Trials

Background:

• Patient-reported outcome measures (PROMs) are crucial in musculoskeletal randomized controlled trials (RCTs).
• Ceiling effects, where scores cluster at the maximum value, can limit the ability to detect treatment effects.
• Understanding score distributions is vital for interpreting trial results.

Purpose of the Study:

• To investigate how baseline and follow-up score distributions in musculoskeletal RCTs using PROMs affect the likelihood of detecting statistically significant between-group differences.
• To examine the impact of score variability and ceiling effects on statistical power.

Main Methods:

• A meta-epidemiologic study of 255 RCTs published between 2015 and 2024.
• Included RCTs compared treatment effects on musculoskeletal disorders using PROMs.
• PROM scores were rescaled to 0-100; likelihood of significant difference (p<0.05) was calculated.

Main Results:

• PROM scores improved from baseline to follow-up in 98% of studies (mean change +28 points).
• Higher baseline scores correlated with lower score changes (r=-0.66).
• The mean likelihood of detecting a difference was 65% at both baseline and follow-up, rarely reaching the 80% benchmark.

Conclusions:

• Concentration of PROM scores near the upper limit, especially with high baseline scores, reduces the likelihood of detecting significant between-group differences.
• This effect is pronounced at follow-up assessments in musculoskeletal studies.
• Critical evaluation of study conclusions is necessary due to potential attenuation of measurable differences.